Delgado Francisco - Derivados Financieros (2010)

Derivados Avanzados Diciembre 2010 Página: 1/358
FUTURES AND OPTIONS MARKETS

PART 1 - INSTITUTIONAL BACKGROUND
1 Introduction to Futures and Options

2 Functioning of Futures and Option Markets
PART 2 - FUTURES
3 Structure of Futures Prices

4 Hedging with Futures Contracts
5 Risk and Return in Futures Markets
6 Futures Contracts on Physical Commodities
7 Stock Index Futures Contracts
8 Interest Rate Futures Contracts
9 Currency Futures Contracts
PART 3 - OPTIONS
10 Structure of Option Prices: Arbitrage Relations

11 Valuation of Options
12 Option Trading Strategies
13 Common Stock Option Contracts
14 Stock Index and Stock Index Futures Option Contracts
15 Interest Rate and Interest Rate Futures Option Contracts
16 Other Option Contracts
Francisco A. Delgado: Francisco_Delgado@Hotmail.Com Versión Junio 2010 Impresión January 19, 2010
FUTURES AND OPTIONS MARKETS

PART 1 - INSTITUTIONAL BACKGROUND
1 Introduction to Futures and Options

What Are Futures Contracts?
What Are Options?
Economic Purpose of Futures and Options
Allocation of Risk
Price Information
Transaction Costs
Summary
2 Functioning of Futures and Option Markets

Exchange Markets Versus OTC Markets
Contract Design
Maturity Months
Contract Size
Method of Contract Settlement at Maturity
Grade of Underlying Commodity
Point of Delivery
Time of Contract Settlement at Maturity
Exercise Price
Trading Procedures
Types of Orders
Types of Markets
Types of Traders
Trading Costs
Price Reporting
The Clearing House
Margin
Trading and Settlement Example
Summary
PART 2 - FUTURES
3 Structure of Futures Prices

The Basis
Basis Arbitrage in Detail
Carrying Costs
Term Structure of Futures Prices
Forward and Futures Contracts
Summary
Appendix 3.1: The Storage Cost Fund
Appendix 3.2: Forward and Futures Contracts
The Difference Between Forward and Futures Positions
Forward and Futures Prices Are Equal if the Interest Rate Is Known
PART 2 – FUTURES. Continuation
4 Hedging with Futures Contracts

A Traditional Hedge
Basis Risk
Commodity Price Risk
Carrying Cost Risk
Hedging Non-Deliverable Commodities
Optimal Hedging-Mean/Variance Approach
Notation
Hedge Portfolio Profit
Optimal Hedge Ratio
Optimal Hedging-OLS Regression Approach
Expected Return
Optimal Hedge Ratio
Hedging Effectiveness
Hedging with Several Futures Contracts
Estimating the Hedge Ratio
Price Changes, Price Change Intervals, and Other Issues
Hedging with a Single Futures Contract
Hedging with Two Futures Contracts
Summary
5 Risk and Return in Futures Markets

Role of Speculators
Is the Futures Price an Unbiased Estimate of the Expected Spot Price?
The Capital Asset Pricing Model for Futures Contracts
Equilibrium of Hedgers and Speculators
Summary
6 Futures Contracts on Physical Commodities

Futures Contracts on Physical Commodities
Why Do Futures Markets in Physical Commodities Arise?
Uncertainty 77 Large and Competitive Market
Storability and Deliverability
Inventory and Price Patterns
Commodities in Seasonal Supply
Non-Seasonal Commodities
Hedging Commodity Price Risk
Storer's Short Hedge
Merchandiser's Long Hedge
Producer Hedge
Natural Hedge Versus Futures Market Hedge
Estimating the Hedge Ratio
Supply of Storage and Demand for Storage
Returns to Speculators
International Price Links: The Law of One Price
Comers and Short Squeezes
Summary
7 Stock Index Futures Contracts

Stock Index Futures Markets
Composition of Stock Indexes
Value-Weighted Arithmetic Indexes
Price-Weighted Arithmetic Indexes
Equal-Weighted Geometric Indexes
Stock Index Simulations
Correlation Among Index Returns
Index Arbitrage and Program Trading
Treasury Bill Substitute
Stock Replacement
Practical Considerations in Index Arbitrage
Intraday Behavior of Returns
Hedging Market Risk
Forming the Hedge Portfolio
Assessing Hedging Effectiveness
Decomposing the Hedge Portfolio Return
Summary
8 Interest Rate Futures Contracts

Interest Rate Futures Markets
Underlying Bonds
Zero-Coupon or Discount Bonds
Coupon Bonds
Interest Rate Risk
Duration and Convexity
Term Structure of Interest Rates and Yield Curve
Spot Rates and Forward Rates
Short-Term Interest Rate Futures Contracts
Long-Term Interest Rate Futures Contracts
Conversion Factor and Invoice Price
Cheapest to Deliver
Cost-of-Carry Relation
Hedging with Interest Rate Futures Contracts
Short-Term, Long Hedge
Long-Term, Short Hedge
Asset Allocation
Summary
9 Currency Futures Contracts

Markets for Currencies
Futures Market
Interbank Spot and Forward Market
Currency Quotations and Triangular Arbitrage
Structure of Currency Futures Prices
Factors Affecting Exchange Rates
Purchasing Power Parity
The Monetarist Approach to Exchange Rate Determination
Balance of Payments Approach to Exchange Rate Determination
The Asset Market Approach to Exchange Rate Determination
Returns to Speculators in Currencies
Spot Speculation and Futures Speculation
The International Fisher Effect
Forecasting Exchange Rates
Hedging Currency Risk
Transactions Risk
Balance Sheet Risk
Summary
PART 3 - OPTIONS
10 Structure of Option Prices: Arbitrage Relations

Option Nomenclature, Notation, and Definitions
Profit Diagrams and Vector Notation
General Discussion of Lower Price Bounds for Options
European Options
American Options
Arbitrage Proofs of Lower Price Bounds
European Call Option
American Call Option
European Put Option
American Put Option
Futures Options
Early Exercise
American Call Option
American Put Option
American Futures Options
General Discussion of Put-Call Parity
Arbitrage Proofs of Put-Call Parity Relations
European Options
American Options
Commodity Options Versus Futures Options
European Options
American Options
Summary
PART 3 – OPTIONS. Continuation
11 Valuation of Options
Risk-Neutral Valuation
Riskless Hedge Portfolio Using a Simple Binomial Model
Risk-Neutral Valuation Using the Binomial Model
Risk-Averse Valuation Using the Binomial Model
Commodity Price and Return Distributions
Risk-Neutral Valuation of European Call Option
Evaluation of X Pr ob(ST  X) ; of E ( ST | ST  X ) and of E (cT )
Current Value of Call
Risk-Neutral Valuation of European Put Option
Properties of the European Call and Put Option Pricing Formulas
Change in Commodity Price
Percentage Change in the Commodity Price
Change in Delta
Change in the Exercise Price
Change in the Cost-of-Carry Rate
Change in the Interest Rate
Change in the Volatility
Change in the Time to Expiration
European Exchange Option
Valuation of American Options
Estimation of the Option Pricing Parameters
Historical Volatility and Implied Volatility Estimation
Summary
Appendix 11.1
Appendix 11.2: Approximations for the Cumulative Normal Density Function
Appendix 11.3: Cumulative Normal Probability Tables
Appendix 11.4: Partial Derivatives of European Commodity Option Valuation
Equations
12 Option Trading Strategies

Trading Strategies and Profit Diagrams
Assumptions and Six Basic Positions
Riskless Arbitrage Strategies
Synthetic Positions
Multiple Option/Commodity Positions
Spread Strategies
Writing/Speculative Strategies
Computing Breakeven Probabilities and Expected Profits
Breakeven Probabilities and Expected Terminal Profit
Replicating Long-Term Options
Dynamic Portfolio Risk Management
Expected Return and Risk
Managing Unexpected Changes in the Commodity Price
Managing Unexpected Changes in Volatility
Managing Time Decay
Summary
13 Common Stock Option Contracts

Common Stock Option Markets
Price Bounds and Arbitrage Relations
Non-Dividend-Paying Stocks
Dividend-Paying Stocks
Valuation of Call Options on Stocks
European Call Option on Non-Dividend-Paying Stocks
European Call Option on Dividend-Paying Stocks
American Call Option on Non-Dividend-Paying Stocks
American Call Option on Dividend-Paying Stocks
Dividend Spreads
Valuation of Put Options on Stocks
European Put Option on Non-Dividend-Paying Stocks
European Put Option on Dividend-Paying Stocks
American Put Option on Non-Dividend-Paying Stocks
American Put Option on Dividend-Paying Stocks
Rights and Warrants
Summary
Appendix 13.1: Approximation for the Cumulative Bivariate Normal Density
Function
14 Stock Index and Stock Index Futures Option Contracts

Stock Index and Stock Index Futures Option Markets
Stock Index Options 345 Stock Index Futures Options
Lower Price Bounds and Put-Call Parity
Stock Index Options
Stock Index Futures Options
Index Options Versus Index Futures Options
Valuation Equations
European and American Stock Index Options
European Stock Index Futures Options
American Stock Index Futures Options
Discrete Dividends and the S&P 100 Index Option
S&P 100 Index Option Valuation
The Wildcard Option
Portfolio Insurance
Static Portfolio Insurance
Dynamic Portfolio Insurance
Summary
15 Interest Rate and Interest Rate Futures Option Contracts

Interest Rate Option Markets
Short-Term Interest Rate Option Pricing
Relation Between Price Volatility and Yield Volatility
Treasury Bond Option Pricing-Price-Based Valuation
Treasury Bond Option Pricing-Yield-Based Valuation
Managing Duration and Convexity
Summary
16 Other Option Contracts

Currency and Currency Futures Options
Valuation of Currency Options
Uses of Currency Options
Physical Commodity Futures Options
Exotic Options
Options on Options
Options on the Maximum and the Minimum
Lookback Options
Barrier Options
Other Exotic Options
Summary
1 INTRODUCTION TO FUTURES AND OPTIONS
1.1 WHAT ARE FUTURES CONTRACTS?
1.2 WHAT ARE OPTIONS?
1.3 ECONOMIC PURPOSE OF FUTURES AND OPTIONS

 Allocation of Risk
 Price Information
 Transaction Costs
1.4 SUMMARY
A look at the business pages of the newspaper reveals a bewildering array of price quotations for futures and
options. While futures contracts on agricultural commodities have been with us since the mid-1800s, futures
trading in financial assets-such as bonds, currencies, and stock indexes were introduced as recently as 1975
and have grown at an explosive rate since that time. Likewise, trading of options on financial and agricultural
commodities is a relatively recent event, dating to the founding of the Chicago Board Options Exchange in
April 1973. Today, call options trade on five U.S. exchanges in nearly 400 common stocks. Moreover,
options also trade on bonds, foreign currencies, stock indexes, and traditional agricultural commodities.
While futures and options contracts on a variety of underlying commodities1 have been developed,
certain principles of valuation and price behavior are common across all commodities. For example, the
essence of the price relation between the futures contract and its underlying commodity is captured by a
simple arbitrage argument, even though the types of commodities range from agricultural to purely financial.
In this book, we emphasize the principles that determine the value of a futures contract, an option contract,
and a futures option contract relative to the value of its underlying commodity. For example, consider the
S&P 500 stock index to be the underlying commodity. The index is a value-weighted stock portfolio
consisting of 500 large common stocks that trade predominantly on the New York Stock Exchange. The
Chicago Mercantile Exchange lists a futures contract on the S&P 500, as well as an option contract on the
futures, while the Chicago Board Options Exchange lists an option contract on the index itself. In other
words, for this particular commodity –the S&P 500 stock index portfolio– the four markets depicted in
Figure 1.1 trade simultaneously. Inextricable linkages exist among prices in these four markets, and, in this
book, we identify the nature of these price linkages and the implications they have for expected return/risk
management. In this chapter, we begin by defining futures, options, and futures options.
1.1 WHAT ARE FUTURES CONTRACTS?

A futures contract is a contract to buy or sell an underlying commodity at a future time, at a price-the futures
price-specified today. Payment for the underlying commodity is not made unless, and until, delivery of the
underlying commodity is taken. In organized futures markets, contracts can be reversed before expiration by
taking a position of opposite sign but equal magnitude in the same futures contract. Someone who buys
futures takes a long position and gains to the extent that the futures price at which that position is reversed
(the terminal futures price) is above the initial futures price. Someone who sells futures takes a short position
and gains if the terminal futures price is below the initial futures price.
1
In this book, the term “commodity” is defined as being something of value. The commodity may be a foodstuff such as wheat, a
currency such as the Japanese yen, or a stock index such as the S&P 500.
The profit from a long futures position initiated at price F0 is plotted in Figure 1.2a. For every dollar price
rise above F0 the investor makes one dollar. For every dollar decline below F0 , the investor loses one dollar.
The profit from a short position initiated at the futures price, F0 , is shown in Figure 1.2b.
As an example, suppose someone buys the July 1992 wheat futures contract listed in Table 1.1. At
the close of trading on November 13, 1991, the futures price ( F0 ) is reported to be $3.21 per bushel. The
denomination of this wheat futures contract is 5,000 bushels. Suppose further that the individual reverses his
position on the following February 6 by entering into a contract to sell July wheat. If the price of the July
wheat futures ( FT ) is $4.00 on February 6, the individual earns a profit of ($4.00 - $3.21) x 5,000 =
$3,950.00. Alternatively, if the futures price on February 6 is $3.00, the individual earns a profit of ($3.00 -
$3.21) x 5,000 = $1,050.00. Most futures contracts are, in fact, reversed in this manner prior to expiration.
Futures contracts are a means for reducing risk or assuming risk in the hope of profit, not a means of taking
possession of the underlying commodity. Users of the underlying commodity generally prefer a grade and
delivery location of the underlying commodity that are different from the grades and locations allowed under
the terms of the futures contract.
As a second example, consider the March 1992 S&P 500 index futures reported in Table 1.2.
Suppose someone buys this contract at the close of trading on November 13, 1991, at the reported price ( F0 )
of $400.35. The contract size for the S&P 500 futures is 500 times the price or $400.35 x 500 = $200,175.
Suppose the position is reversed on February 6 when the futures price ( FT ) is, say, $410.00. In this case, the
individual earns a profit of ($410.00 - $400.35) x 500 = $4,825.00. If, instead, the index falls to, say,
$390.00, the individual has a loss of ($400.35 - $390.00) X 500 = $5,175.00.
Every futures contract entered into has two sides: a willing buyer and a willing seller. If one side of
the contract makes a profit, the other side must make a loss. All futures markets participants taken together
can neither lose nor gain-the futures market is a zero-sum game.
FIGURE 1.1 Interrelations Between Commodity Market and Markets for the Commodity's Derivative
Instruments
Option
Commodity
Market
Market
Futures Futures
Market Options
Market
FIGURE 1.2 Profit Diagrams for Futures Position Held to Expiration
Profit
(a) Long Futures Position
Terminal
Futures
Initial Futures Price, F0 Price, FT
F0
Profit (b) Short Futures Position
F0
Terminal
Futures
Initial Futures Price, F0 Price, FT
TABLE 1.1 Prices of wheat futures contracts at the close of trading on Wednesday, November 13, 1991.
Lifetime
Open High Low Settle Change High Low Interest Open
---GRAINS AND OILSEEDS---
WHEAT (CST) 5,000 bu.; cents per bu.
Dec 347 ½ 352 ¾ 347 ½ 352 ½ +5¼ 369 ¼ 272 ½ 19,480
Mr92 349 ½ 353 ¾ 348 353 ½ +5¾ 367 279 23,394
May 334 338 ½ 334 338 +5¼ 352 ½ 280 ½ 5,174
July 317 ½ 321 316 ¾ 321 +4¾ 337 ½ 279 8,067
Sept 326 326 326 326 +5 341 292 690
Dec 333 ½ 335 332 335 +4¾ 351 329 ½ 703
Est vol 15,500; vol Tues 11,502; open lnt 57,508, -149.
Source: Reprinted by permission of Wall Street Journal, (November 14, 1991) Dow Jones & Company, Inc.
All. Rights Reserved Worldwide.
FUTURES
TABLE 1.2 Prices of S&P 500 stock index futures contracts at the close of trading on
Wednesday, November 13, 1991.
S&P 500 INDEX (CME) 500 times index`
Open
Open High Low Settle Chg High Low
Interest
Dec 395.00 398.50 394-30 398.30 +1.00 401.50 316.50 139,341
Mr92 396.80 400.50 396.50 400.35 +1.00 404.00 374.70 7,544
June 398-30 402.35 398.30 402.20 +1.10 407.00 379.00 1,102
Est vol 42,125; vol Tues 41,413; open lnt 148,048, +916.
Indx prelim High 397.42; Low 394.01; Close 397.42 +.68
Source: Reprinted by permission of Wall Street Journal, © (November 14, 1991) Dow Jones & Company,
Inc. All Rights Reserved Worldwide.
1.2 WHAT ARE OPTIONS?

An option contract conveys the right to buy or sell an underlying commodity at a specified price within a
specified period of time. The right to buy is referred to as a call option; the right to sell is a put option.
Options are generally described by the nature of the underlying commodity. An option on a common stock is
said to be a stock option; an option on a bond, a bond option; an option on a foreign currency, a currency
option; an option on a futures contract, a futures option; and so on. The specified price at which the
underlying commodity may be bought (in the case of a call) or sold (in the case of a put) is called the
exercise price or the striking price of the option. To buy or sell the underlying commodity pursuant to the
option contract is to exercise the option. Most options may be exercised at any, time, up to and including the
expiration date. These are called American options. If an option can only be exercised at expiration, it is
termed a European option.
The buyer of an option pays the option writer (seller) an amount of money called the option premium
or option price. In return, the buyer receives the privilege, but not the obligation, of buying (in the case of a
call) or selling (in the case of a put) the underlying commodity for the exercise price. In the case of a call
option, if the price of the commodity exceeds the exercise price, the call option is said to be in-the-money;
and the call option buyer could exercise the option, thereby earning the difference between the two prices-the
exercise value or intrinsic value. On the other hand, if the price of the commodity is below the exercise price,
the call option is out-of-the-money and will not be exercised. Its intrinsic value is zero. In the case of a put
option, if the price of the commodity is below the exercise price, the put option is said to be in-the-money.
The put option buyer could exercise the option to earn the difference between the exercise price and the price
of the commodity. A put option is said to be out-of-the-money when the commodity price exceeds the
exercise price.
The profits from various option positions held to expiration are plotted in relation to the price of the
underlying commodity in Figures 1.3 and 1.4. The illustrations assume that options are held to the expiration
date, T . The price of the underlying commodity at the option's expiration is denoted as ST .
The position of a call buyer –a long call position– is profitable if the price of the underlying
commodity, ST , exceeds the exercise price, X , by more than the initial price of the call option, C0 . This is
depicted in Figure 1.3a. On the other hand, if ST is below X at expiration, the call option is not exercised.
The maximum gain to the call buyer is unlimited because the exercise value of the option increases directly
with increases in the value of the underlying commodity, which is unlimited in principle. The maximum loss
to the option buyer is C0 .
The position of a call seller or call writer –a short call position– is depicted in Figure 1.3b. The
position is the reverse image of the long call position. A call seller faces the possibility of large losses if the
price of the underlying commodity increases, because, in that case, the call will be exercised, and the call
seller will be requested to purchase the underlying commodity at ST and deliver it to the call buyer at X .
The maximum gain to the call seller is C0 . It is evident from Figures 1.3a and 1.3b that the sum of the profits
of the call buyer and call seller at any terminal price, ST is zero. As in the case of futures markets, the option
market is also a zero-sum game.
The position of a put buyer –a long put position– depicted in Figure 1.4a is profitable if the price of
the underlying commodity falls below the exercise price by more than P0 , the initial price of the put. If the
price exceeds the exercise price at expiration ( ST  X ), the put is not exercised. The maximum loss to the
put buyer is P0 and the maximum profit is X  P0 . The position of the put seller-the short put position-
depicted in Figure 1.4b is the reverse image of the put buyer's position. The put seller has a maximum gain of
P0 and a maximum loss of X  P0 .
Option buyers may choose to realize profits by exercising their options as we have just discussed.
More frequently, however, option positions are closed out by selling the option. At expiration, an option may
be sold for its exercise value. Before expiration, options usually sell for more than their exercise value. As a
result, it is usually, but not always, preferable to close out an option position prior to expiration by selling the
option rather than by exercising it. The gain or loss on the option is then just the change in the price.
An example using S&P 100 index options will, perhaps, make this discussion more concrete. Table 1.3
contains the closing prices of S&P 100 index options on November 13, 1991. The S&P 100 option contract
size is 100 times the index value. The December call with an exercise price of $370 has a reported price of
$7.375. This means that a call option buyer would pay a premium of $7.375 X 100 = $737.50 for the right to
“buy” the S&P 100 stock index at $370 x 100 = $37,000 any time before the expiration date2. If the index
level rises from $371.21 on November 13 to, say, $400.00 on November 30, the call option buyer can
exercise her option to earn a profit of ($400.00 - $370.00 - $7.375) x 100 = $2,262.50. Alternatively, the
option buyer can reverse her position by selling an option contract with the same terms. Suppose that on
November 30 the price of the call is $32.00. If the buyer of the option on January 16 chooses to sell the same
option, she realizes a profit of ($32.00 - $7.375) x 100 = $2,462.50. Note that closing the position by selling
the option as opposed to exercising it yields an additional profit of $2,262.50 - $2,462.50 = $200.00.
The fact that the option can be sold for more in the marketplace than the intrinsic value reflects the
time value of the option. On November 13, the 370 call can be sold for $7.375 and yet, if that same option is
exercised on November 13, its value is $371.21 - $370.00 = $1.21. The difference between the two values is
called the time value of the option and reflects the probability that the stock index will rise significantly from
its current level by the third Friday in February. The factors affecting the level of option premiums for
different striking prices and different maturities are described later in the book. The price of the 370
December put option in Table 1.3 is reported to be $5.25. This option is out-of-the-money since the current
value of the stock index exceeds 370. Someone who buys this put option on November 13 earns a positive
profit if the index level falls below $370.00 - $5.25 = $364.75 before the third Friday in December 1992.
It is important to recognize that the option writer (seller) faces payoffs exactly opposite those of the
buyer. If, for example, a call option is in-the-money at expiration, the option writer must deliver a
commodity worth more than the exercise price received by the writer. In terms of the call option example
above, when the buyer of the option chooses to exercise, the option seller in effect has to purchase the index
at $400.00 and deliver it to the option buyer at $370.00. This loss is offset in part by the premium, $7.375,
that the writer collected at the outset. The net loss to the writer equals the option buyer's net profit. On the
other hand, if the index level stays below the exercise price of the call until expiration, the option will not be
exercised, and the option writer keeps the premium collected from the buyer when the option contract was
written. This time the writer makes a positive profit, but, again, it is equal to the buyer's loss.
2
S&P 100 index options are American-style and-expire on the Saturday following the third Friday of the contract month.
FIGURE 1.3 Profit Diagrams for Call Option Positions Held to Expiration
Profit
(a) Long Call
Exercise Price, X
Terminal Price of
Underlying
-C0 X+C0 Commodity, ST
Profit (b) Short Call
C0 X+C0
Terminal Price of
Underlying
Exercise Price, X Commodity, ST
FIGURE 1.4 Profit Diagrams for Put Option Positions Held to Expiration
Profit
(a) Long Put
Exercise Price, X
Terminal Price of
X-P0 Underlying
-P0 Commodity, ST
Profit (b) Short Put
P0
X-P0
Terminal Price of
Underlying
Commodity, ST
Exercise Price, X
TABLE 1.3 Prices of S&P 100 stock index option contracts at the close of
trading on Wednesday, November 13, 1991.
OPTIONS CHICAGO BOARD
S&P 100 INDEX- $100 times Index
Strike
Calls-Last Puts - Last
Price
Nov Dec Jan Nov Dec Jan
335 … … … 1/16 … …
340 30 31 ½ 31 ¼ 1/16 5/8 1½
345 23 27 28 1/16 7/8 2 1/8
350 21 ½ 21 1/8 … 1/16 13/16 3
355 16 3/8 19 195/8 1/16 1¾ 37/8
360 11 ¼ 14 ¾ 17 1/8 21/2 4½
365 6 3/8 10 5/8 13 ½ 5/16 3 5/8 6
370 2 5/16 7 3/8 10 ½ 1 1/16 5¼ 7¾
375 3/8 4 3/8 7 3/8 4 1/8 7 5/8 10 7/8
380 1/16 2 7/16 5¼ 9¼ 10 7/8 13 3/4
385 1/16 1¼ 3 1/8 … 151/4 …
390 1/16 5/8 1 7/8 18 5/8 20 20 ¾
Total call volume 142,982 Total call open Int. 387,114
Total put volume 128,390 Total put open Int. 421,693
The Index: Vol 371.22; Low 367.54; Close 371.21, +0.92
1.3 ECONOMIC PURPOSE OF FUTURES AND OPTIONS

Traditional financial instruments such as stocks and bonds are mechanisms for channeling funds from savers
to borrowers as well as a means for spreading risks. (Of course, most trading in stock and bond markets is
trading in the secondary market which does not directly allocate new funds to new investment opportunities.)
Futures and options facilitate the channeling of funds but are not savings devices. They are primarily a
means for dealing with uncertainty. At the same time, they serve to maintain the liquidity and the reliability
of underlying commodity prices, which are important for the proper allocation of new investment funds.
The economic benefits of futures and options arise along three dimensions.3 First, futures and options are a
means for allocating risk more efficiently. Second, futures and options provide price information that is
useful in allocating resources in the economy to their best uses. Third, futures and options may lower
transaction costs of trading in financial markets below the costs associated with trading in existing financial
instruments.
Allocation of Risk
Futures and options provide an efficient mechanism for allocating risk from those who wish to avoid risk to
those who are interested in bearing the risk. Futures contracts tend to arise when the underlying commodity
is costly or cumbersome to trade. For example, futures on agricultural commodities allow an investor to bear
the risk associated with holding an agricultural commodity without the troublesome details of trading in the
3
For an extended discussion of the economic purposes of futures and options, see Carlton (1984), Jaffe (1984), Peck (1985),
Silber (1985), and Stoll and Whaley (1988).
actual commodity. In this manner, processors of the underlying commodity can pass on price risk to
investors and retain the economic function of processing the underlying commodity.
Options provide an additional benefit in allocating risk because the profit function for options is
different from the profit function for the underlying commodity or for a futures contract. As is evident from
Figures 1.3 and 1.4, the profit from options positions are asymmetric. Such an asymmetric payoff pattern is
useful for example, in dealing with situations that involve both quantity and price risk. Consider a farmer
who is interested in avoiding the risk associated with a drop in the price of the commodity that he grows.
Before the harvest, the farmer does not know the size of the crop or the price. Selling futures against the crop
would hedge the farmer against a price decline if the harvest were known, but a futures hedge would expose
him to risk if the harvest failed and prices increased because then the farmer, would not have the wheat that
he had committed to sell in the futures market. The farmer would take a loss in acquiring the wheat to
delivery against the futures contract. Buying a put option on the underlying commodity provides a more
effective hedge against price and quantity risk than selling futures. If prices fall, the put is exercised (or
liquidated at a profit). If prices rise, the put option expires, worthless, and the farmer realizes the revenues
from his crop, regardless of the size of the harvest. The cost of this one-sided protection for the farmer is the
put option premium. Similar examples exist for other underlying commodities. In addition to this hedging
use, options are also a useful portfolio management tool. For example, index put options can be used to limit
the downside risk of stock portfolios while retaining part of the upside potential.
Price Information
Some trading in futures and options markets, as in other financial markets, arises not because individuals
have a desire to shift risk, but because they have different information and disagree about the correct price of
the underlying commodity. This kind of informational trading is termed speculative trading. Society benefits
from speculative trading because the analysis and search for information on which it is based cause the prices
of futures and options and underlying commodities to correspond more closely to their correct values. Even
if an underlying commodity is traded, futures and options on that commodity are likely to increase the
interest and the number of judgments bearing on the underlying commodity's price. Because futures and
options prices are related to the price of the underlying commodity by an arbitrage relation, factors affecting
futures and option prices tend to be conveyed to the price of the underlying commodity; conversely, factors
affecting the price of the underlying commodity tend to be conveyed to option and futures prices. Thus, to
the extent that futures and option trading increase the total interest in an underlying commodity, the
commodity's price will be more broadly based and less likely to be influenced by only a few judgments.
Futures and option markets encourage increased research and analysis. If financial markets are like
other markets, this leads to greater efficiency in the production of information. Insofar as increased analysis
and increased interest improves the quality of prices, resources will be allocated more efficiently. The price
signaling benefit of futures and options is most evident in the case of agricultural futures. For example, the
daily newspaper prints futures prices for wheat deliverable many months in the future. That price is the price
someone has paid to receive delivery of a particular grade of wheat in the future. Producers or storers of
wheat can use this price as a signal for production and storage decisions. If the price is high relative to their
costs, more wheat will be produced and stored. If the price is low relative to their costs, less wheat will
produced and stored. In this way, the proper amount of wheat will be allocated for future consumption. Of
course, if the futures price is wrong-because it is manipulated or for other reasons resources will be
misallocated. Futures markets depend on the presence of many knowledgeable participants to avoid this. One
of the benefits of the introduction of futures is the fact that futures trading increases competition. Producers
interested in entering into contracts for delayed delivery are no longer compelled to deal with relatively few
users of a commodity.
As we shall see later, option prices depend primarily on the projected volatility of the commodity
underlying the option. As a result, option prices quoted in the newspaper provide information on the future
price uncertainty of an underlying commodity. Processors and users of the underlying commodity therefore
have information not only on the expected price of the underlying commodity but also on the price
uncertainty in the future.
Transaction Costs
It is sometimes argued that futures and options are redundant securities because any futures or option
position is achievable by trading the underlying commodity. For example, a long position in wheat futures
has the same profit or loss potential as a long position in the wheat itself. A long position in stock index
futures can also be achieved by buying a diversified stock portfolio. The payoff to a Treasury bond futures
contract can be replicated by a position in the underlying T-bonds. Similarly, option positions can be
replicated by appropriate trading strategies in the underlying commodity.
An important benefit of futures and options, however, is that they reduce transaction costs of
achieving certain risk return positions compared with the cost of trading the underlying commodity. It is
certainly much less costly to trade wheat futures than to trade wheat itself. But the lower trading cost in
futures markets also exists with respect to financial instruments. For example, the transaction costs
associated with trading index futures contracts are estimated to be 1/15 of the costs associated with trading
the corresponding underlying stocks. Index futures are therefore a less expensive means of trading claims on
a portfolio of stocks.
1.4 SUMMARY
In this chapter, payoffs to long and short positions in futures and options are illustrated. The economic
purposes of futures and options-risk transfer, price discovery, and reduced transaction costs-are discussed.
2 FUNCTIONING OF FUTURES AND OPTION MARKETS
2.1 EXCHANGE MARKETS VERSUS OTC MARKETS
2.2 CONTRACT DESIGN

 Maturity Months
 Contract Size
 Method of Contract Settlement at Maturity
 Grade of Underlying Commodity
 Point of Delivery
 Time of Contract Settlement at Maturity
 Exercise Price
2.3 TRADING PROCEDURES

 Types of Orders
 Types of Markets
 Types of Traders
 Trading Costs
 Price Reporting
2.4 THE CLEARING HOUSE

 Margin or Performance Bond
 Trading and Settlement Example
2.5 SUMMARY
2 FUNCTIONING OF FUTURES AND OPTION MARKETS

In the last chapter, we defined the nature of futures and option contracts and the nature of the markets in
which they trade. In section one of this chapter, the two types of futures and option markets are described
exchange-traded markets and over-the-counter markets. Section two contains a discussion of the elements of
futures and option contract design. Trading procedures in futures and option markets are described in section
three, and the role of the clearinghouse in exchange-traded markets is described in section four.
2.1 EXCHANGE MARKETS VERSUS OTC MARKETS

This book deals primarily with organized exchange markets in futures and option contracts, which are to be
distinguished from over-the-counter (OTC) markets. In exchange markets, contracts are standardized, and a
clearinghouse stands between the buyer and seller and guarantees contracts. The clearinghouse is the buyer
to every seller of a contract and the seller to every buyer. Under this arrangement, secondary market trading
is possible because a buyer of a contract who wishes to liquidate the position need not find the original seller
of the contract but may negotiate a transaction with any individual. Organized exchange markets thus
facilitate trading between strangers.
In OTC markets, contracts are tailored to the needs of the transacting parties, and no clearinghouse
exists. As a result, a futures contract or an option contract is a contract between the two parties whose names
are on the contract and who make their own arrangements for guaranteeing the contract’s financial integrity.
Secondary market trading is very inefficient in OTC markets because a buyer who wishes to liquidate
a position must seek an agreement with the original seller of the contract. The term “forward contract” is
frequently used to designate OTC contracts that have the form of futures contracts except that they may lack
standardization or the presence of a liquid secondary market.
OTC markets have usually preceded organized exchange markets in futures and options. For
example, prior to the start of wheat futures trading in the, late 1800s, forward contracts were used by grain
elevator operators to purchase wheat from farmers. Prior to the inception of the Chicago Board Options
Exchange in 1973, over-the-counter options on common stocks were arranged by put and call dealers.
Organized futures and option markets began in order to provide a means for temporarily hedging a position
in the underlying commodity without giving up its control. In OTC markets, delivery is usually implied. The
secondary market trading made possible by organized exchange markets allows participants to temporarily
hedge their positions without making a commitment to relinquish control of the underlying commodity.
Optimal transactions in the underlying commodity can then be arranged either for spot (immediate) delivery
or for forward (delayed) delivery. Once the transaction in the underlying commodity is complete, the
position in the futures or option market can be liquidated. The recent growth in organized futures and option
markets does not mean that OTC markets have declined in importance, however.
Forward contracts in most tangible underlying commodities (such as wheat or oil) are frequently
used. Such contracts are the means by which specific grades of the underlying commodity are sold to
particular buyers of the commodity. In recent years, OTC market trading in derivative financial instruments
has grown dramatically alongside organized markets in these instruments. This reflects the fact that OTC
forward and option contracts can be tailored directly to the needs of a customer. Institutions create OTC
forward or option contracts tailored to the needs of their retail customers and then use organized markets to
offset their OTC market positions. For example, suppose a corporation needs Japanese yen on a particular
future date, and the date does not correspond to the maturity of any available Japanese yen futures contract.
In this case, the corporation is likely to go to a bank that is willing to design a forward contract to meet its
needs.
2.2 CONTRACT DESIGN

Futures and options are contracts to do something at a later date. The obligations of the buyer and seller are
defined in the contract. Only the price and number of contracts are negotiated at the time of transaction. Two
opposing forces influence contract design: standardization versus market depth. Market participants would
prefer the commodity underlying a futures or option contract to be clearly and narrowly defined. However, a
narrowly defined contract, while useful to certain participants, may fail to attract sufficient participants to
provide a deep and liquid market. A deep and liquid market is desirable for two reasons: first, to permit
secondary market trading “in size” to be carried out with relatively little impact on price; and, second, to
limit the possibility of corners or short squeezes.4
A corner or short squeeze arises when sellers of a futures or option contract cannot acquire the
underlying commodity for delivery. If the commodity is narrowly defined, it is possible for someone to
monopolize the supply of the commodity. If that individual also purchases futures or option contracts on the
commodity, a corner results –the individual owns or has a claim on more of the commodity than is available.
If the contract is broadly defined to allow delivery of various related underlying commodities, control of the
deliverable supply is much more difficult, and corners and squeezes are much less likely. If a narrowly
defined underlying commodity is in large supply, the futures and option contracts can be narrowly defined.
However, if the supply of a single grade is not large enough or if a single delivery location is not convenient
enough, futures and option contracts are broadly defined to allow delivery of several grades at several
locations.
Most futures and option contracts must make provisions for the following features: (a) maturity
months; (b) contract size; (c) method of contract settlement; (d) grade of deliverable commodity; (e) point of
delivery; (f) time of settlement at maturity; and, in the case of options, (g) the number of exercise prices.
Maturity Months
Too many maturity months reduce the depth and liquidity in any one month. Too few maturity months
reduce the usefulness of a contract. How these conflicting objectives are balanced depends on the underlying
commodity. For example, in wheat, five maturity months (March, May, July, September, December) are
traded at any time, reflecting the harvesting and marketing cycles for wheat. In silver, sufficient volume
exists to warrant the trading of ten different maturity months extending one and a half years into the future.
In Eurodollar futures, forty maturity months extending ten years into the future are traded.
It is worth noting that the number of days until maturity of a particular futures contract is changing as
the maturity date is approached. This is in contrast to various forward contracts, the prices of which are
quoted in the newspaper. For example, the newspaper may quote the three-month forward price of silver or
the three-month forward price of a currency. These quotes are for new contracts originated on that day. The
secondary market for these contracts after they are originated is not very active, so price quotes of existing
contracts do not appear in the newspaper.
Contract Size
Contract sizes vary considerably and are chosen to meet the needs of the users of the contract. In many of the
grains, the standard contract size is 5,000 bushels, or approximately $15,000 at the current price of the
commodities. Contract sizes are considerably larger in some of the financial futures. For example, futures on
Treasury bills and Eurodollars have contract sizes of $1 million.
Method of Contract Settlement at Maturity

Most futures and option contracts are settled by delivery at maturity. Should a contract be carried into the
delivery month, certain rules and procedures govern delivery. With futures contracts, the seller of futures
(the short) may make delivery of the underlying commodity during a time in the maturity month specified by
4
For more analysis of contract design and the success and failure of futures contracts, see Black (1986), Carlton (1984), Johnston
and McConnell (1989), and Stoll and Whaley (1985).
exchange regulation. Delivery of tangible commodities may be made at any time during an extended period,
such as two weeks, and usually takes the form of warehouse receipts giving claim to the commodity, which
is stored at an approved location. Delivery of financial commodities is usually more narrowly defined, with
delivery taking place through an approved bank. The buyer of futures (the long) is obligated to take delivery
if called upon to do so. The assignment of delivery notices by the exchange takes various forms. In some
markets, the oldest long is assigned the delivery notice. In other markets, delivery notices are assigned
randomly. Futures contracts differ as to the flexibility remaining to the long after the receipt of a delivery
notice. In some cases, usually in the tangible commodities, the long has the opportunity to pass the notice on
to someone else and to liquidate the futures contract. The shorts usually have the greatest flexibility because
they can choose the particular grade of underlying commodity that will be delivered as well as the exact
timing of delivery.
With American style call options, the buyer may request delivery –exercise the options– at any time
during the option’s life. The seller of the call is then obligated to make delivery. With European-style call
options, delivery may be requested only on the expiration date. Options tend not to be written on underlying
tangible commodities that may be difficult and cumbersome to deliver. Instead, options on Commodities are
written on the futures contract on those commodities. Exercising futures-option on corn, for example, is a
request for the option seller to deliver the long futures position in corn. When the long futures position is
received, the position may be held to maturity if delivery of the corn is desired. The exercise of a put option
on corn is a request for the option seller to deliver a short futures position in corn. Options on many financial
instruments, such as options on individual common stocks, call for delivery of the underlying financial
instrument.
Certain futures and option contracts call for cash settlement rather than delivery at maturity. The
buyer of a cash settled futures contract, holding the position until expiration, receives the difference between
the final settlement price of the futures contract and the price at which the contract was purchased.5 The final
settlement price of the futures contract is the cash price of the underlying commodity. The seller of the
futures contract receives a profit exactly opposite that received by the buyer. In the case of cash settlement
call options, the exercise of a call option results not in the delivery of an underlying commodity, but rather in
a profit equal to the difference between the price of the underlying commodity and the exercise price of the
option.
Cash settlement is particularly useful when the underlying commodity is difficult to deliver. In U.S.
markets, stock index futures and options are cash settled because it is difficult to deliver a large portfolio of
many different common stocks. Municipal bond futures and futures on the consumer price index are also
cash settled, because, in these cases, the underlying commodity is impossible to deliver. In the case of
municipal bond futures, the price of the underlying commodity is actually an average of dealer quotations in
municipal bonds.
Grade of Underlying Commodity

When the underlying commodity has differing characteristics, the futures or option contract specifies the
standard grade of the deliverable commodity as well as the other grades that may be delivered. For example,
the Chicago Board of Trade (CBT) wheat contract calls for the delivery of No. 2 soft red wheat, however
certain other grades are also deliverable. Another example is the CBT’s T-bond futures contract. Nominally,
the contract calls for the delivery of an eight-percent coupon bond with a maturity of at least fifteen years.
However, T-bonds with other coupons and with maturities in excess of fifteen years are also acceptable for
delivery.
The choice for a particular grade of an underlying commodity to be delivered against a futures
contract is left to the short, and the short naturally chooses the “cheapest to deliver.” For example, if both
5
Technically, this statement applies to forward contracts only. The futures contract holder has accumulated over the life of the
contract an amount equal to the difference between the final settlement price and the price at which the futures was purchased. In
the interest of clarity, we defer detailed discussion of the distinction between forward and futures contracts to Chapter 3.
soft red wheat and hard red winter wheat are deliverable against the wheat futures contract, but hard red
winter wheat is selling in the marketplace at a lower price than soft red wheat, the person with a short
position will choose to deliver hard red winter wheat because it is the cheapest to deliver. Indeed, the futures
price at maturity reflects the price of the cheapest to deliver grade of wheat, not necessarily the grade
specified as standard in the futures contract. In T-bond futures, it is sometimes desirable to deliver a low
coupon, long-maturity bond and at other times desirable to deliver a high-coupon, short-maturity bond. The
eight-percent coupon bond specified as standard in the contract is usually not the bond being priced by the
futures contract.
The cheapest to deliver commodity may change during the futures contract life. The exchange
specifies the price relation between the deliverable grades prior to the start of the contract. As market
conditions change, some grades go to a premium or discount relative to the standard grade.
Point of Delivery
An important feature of futures contracts on tangible commodities is the number and the location of delivery
points. Transportation of tangible commodities to the delivery location may be costly. As a result, an
increase in the number of delivery locations benefits the shorts who are obligated to deliver. To see this,
suppose a wheat futures contract calls for delivery only in approved warehouses in Chicago. If wheat is in
relatively short supply in Chicago, it is possible for someone to buy up most of the remaining supply while at
the same time buying wheat futures contracts. Such an individual would have engineered a corner if the
market did not have sufficient time to ship wheat to the Chicago delivery location. As a result, it is
sometimes desirable to specify several delivery locations in a contract, thereby making it difficult to corner
the available supplies at all the delivery points.6
Time of Contract Settlement at Maturity

Most futures contracts on tangible commodities and certain financial futures and option contracts allow a
period of time in the maturity month during which delivery may be made. In the case of wheat futures, for
example, delivery may be made at any time in the maturity month, at the option of the seller. In the case of
futures on tangible commodities, the seller usually has the option of when to deliver, what grade to deliver,
and where to deliver it. These features provide protection for the seller against the danger that someone may
corner the available supply of the underlying commodity which must be delivered. When delivery is easy
and the danger of a corner of the underlying deliverable supply is small, as in many financial instruments, the
time of delivery and other features, such as grade and location, are more narrowly prescribed.
Exercise Price
In the case of options, a feature is required that is not required in futures contracts namely, the number of
exercise prices which should be available. It would be possible, for example, to have only one exercise price
for each option maturity. This is not done because the usefulness of an option is greatest when the exercise
price is close to the price of the underlying commodity. As a result, additional options with new exercise
prices are created whenever the underlying commodity’s price moves by a prespecified amount. For
example, a stock selling at $40 might have an option with an exercise price of $40. If the stock price
increases to $45, a new option with an exercise price of $45 would be initiated. The $5 increment at which a
new option with a new exercise price is initiated is determined by the exchange.7 All the options on a
particular underlying commodity with the same maturity are called an option series. The number of options
in an option series is determined by the price volatility of the underlying commodity and the price increment
at which additional exercise prices will be set. If the price increment is small, many options, each with
relatively little liquidity, will be created. If the price increment is large, few options will be created.
6
A recent book by Pirrong, Haddock, and Kormendi (1991) provides a detailed analysis of delivery terms for agricultural
commodities.
7
Stock option exercise prices at $25 or above have $5 increments. Below $25, the increments are $2.50.
2.3 TRADING PROCEDURES

Public customers wishing to trade futures or options open an account with a brokerage firm. In the futures
industry, a brokerage firm is sometimes called a futures commission merchant (FCM). Futures and futures
option contracts are traded in a commodity account, and options written directly on financial instruments are
traded in securities accounts. These two types of accounts are subject to Commodity Futures Trading
Commission (CFTC) oversight and Securities and Exchange Commission (SEC) oversight, respectively.8
Types of Orders
In trading futures and options, as with other securities, investors can place a variety of orders. A market order
instructs the broker to trade at the best price currently available. A limit order instructs the broker to buy or
sell at a specific price. Naturally, the price given for a limit order to buy is below the current market price,
and the price given for a limit order to sell is above the current market price. A stoploss order is an order to
sell below the market or to buy above the market. A spread order instructs the broker to buy one contract and
sell a related contract. In a maturity or calendar spread, for example, the trader buys a contract in one
maturity month and sells a contract in the same commodity for a different maturity month. In an
intercommodity spread, the trader purchases a contract in one commodity and sells a contract in a different
commodity. An almost unlimited array of spread transactions is available in futures and option markets, and
these will be discussed in later chapters of this book. An important point to remember is that the trader in a
spread transaction is interested in favorable changes in the price differential between two contracts.
Types of Markets
Orders placed by customers with their brokers are transmitted through the brokerage firms’ back offices to
the floor of the appropriate exchange for execution. The mechanics by which such orders are executed differ
between options on securities and futures. Futures and futures option contracts are traded in a pit in an “open
outcry” format. Generally, one pit or ring is assigned to each commodity traded on an exchange. Traders
stand on the steps around the pit and trade pairwise with each other. Certain actively traded futures contracts,
such as T-bond futures or S&P 500 index futures, attract in excess of 400 traders in the pit. Orders are
received on the trading floor by telephone and transmitted to the appropriate trading pit by messengers.
Unlike the NYSE stock market where trading in a particular stock occurs sequentially in time at a particular
location on the floor, many transactions can occur simultaneously in an active futures contract. Futures
markets do not therefore guarantee the same degree of price and time priority that NYSE stock markets
guarantee because of the possibility that two simultaneous transactions might occur in different parts of the
trading pit at different prices or because limit orders held by a particular broker may for some reason not be
exposed to all other brokers in the crowd. Such price differences within the ring are, however, infrequent and
small because many traders on each side of the market are each searching for the best price. Competition
among floor traders thereby reduces any price deviations within the pit and also provides tremendous
liquidity for orders flowing in from the public.
Options on financial instruments are traded according to two different procedures. Options on
securities exchanges such as the American Stock Exchange and the Philadelphia Stock Exchange are traded
using the specialist system. In a specialist system, market orders are usually traded at the bid or ask price
quoted by the specialist on his own behalf or on the behalf of limit orders previously left with the specialist,
although there is an opportunity for other traders in the crowd to better the specialist’s price. Limit orders are
left with the specialist to be executed when the market price reaches the limit price. (In futures markets, each
floor broker has his or her own “deck” of customer limit orders.) The specialist system has been criticized
because only a single specialist makes a market in each option. As a result, investors do not have an
opportunity to shop for better prices from other marketmakers.
The Chicago Board Options Exchange (CBOE) system combines elements of futures markets and the
specialist system of stock markets. An Order Book Official (OBO) maintains the book of limit orders but
8
The Chicago Board of Trade (1989) provides useful information on trading procedures and other aspects of futures markets.
does not trade for his own account. At the same time, many professional floor traders are prepared to trade
for their own accounts to absorb temporary imbalances and maintain market liquidity. More so than either
the futures market system or the specialist system, the CBOE floor trading system represents what many
have called for in the stock market: a system that combines competition among marketmakers with full
exposure of all limit orders through the open book of the OBO.
Types of Traders
Floor traders in futures and option markets can be divided into two general classifications: floor brokers and
professional traders. Floor brokers are agents who execute transactions for public customers such as
processors of tangible commodities or portfolio managers of financial instruments. Professional traders buy
and sell for their own accounts. Professional traders are sometimes called speculators because they take on
varying amounts of risk. Professional traders in futures markets are often classified into position traders, day
traders, and scalpers. Position traders take on risks and positions that are held for longer periods of time –
days or weeks. Day traders have a short horizon and take on positions that are usually liquidated at the end of
the day. Scalpers have a very short horizon and make their income primarily from short-term, minute-by-
minute transactions. Scalpers provide liquidity to other investors by buying at the bid price when public
customers desire to sell and by selling at the ask price when public customers desire to buy.
In stock exchanges, the major type of professional trader is the specialist. The specialist’s role in the
stock market is similar to the scalper’s role in the futures market. Both are responsible for maintaining
market liquidity, and profit from the spread between bid and ask prices. In a specialist system, however,
there is only one specialist for each security. In futures markets, many scalpers compete in each contract. In
securities markets, there are relatively few professional traders other than the specialist. Most position traders
and day traders in options and stocks submit orders from off the exchange through brokerage firms. By
contrast, in futures markets, a great deal of volume is the result of trading by professional traders on the
floor.
Trading Costs
The costs of trading options and futures consist of two components: the commission charges of the broker
and the price concession that may be necessary to execute the transaction. The price concession reflects the
fact that sales are made at the bid price of professional traders on the floor (scalpers or the specialist) and
purchases are made at the higher ask price of professional traders on the floor. In addition, the broker
carrying out a customer transaction is compensated by a commission. Commissions on futures and option
exchanges are competitively determined and vary from broker to broker. Commissions cover the back-office
services of the broker as well as the charges for floor brokerage and the clearing of transactions.
Price Reporting
On stock exchanges, the price and size of each transaction is reported on the ticker tape immediately after the
transaction occurs. In futures markets, not every transaction is reported because many transactions occur
simultaneously. Instead, price reporters in the trading pit report each different transaction price and each
different bid or ask price in the pit. Systems for price reporting differ among the futures exchanges. Some
record prices manually on a price board above the exchange floor. The prices are then entered into computer
terminals for transmission worldwide. Others enter price information directly into computer terminals, and
then the information is automatically displayed on the exchange floor and is transmitted worldwide. In
futures markets, statistics on the volume of trading are not available on a real time basis. Instead, such
statistics are compiled at the end of the day on the basis of transactions clearing data.
2.4 THE CLEARING HOUSE

Futures and option contracts are “created” instruments. When a buyer and a seller meet, a contractual
agreement containing the specific rights and obligations of each party is created. The number of such
contracts that are created is not limited, unlike stock markets where the supply of stock is limited to the
number of shares that the firm has issued. Open interest is the number of futures or option contracts
outstanding at any one time.
The clearinghouse is critical to the trading of futures and option contracts because it settles and
guarantees the contracts. After a contract is agreed to, the clearinghouse interposes itself between buyer and
seller and, in effect, becomes the party to whom delivery is made and from whom delivery is taken. Since the
number of buyers always equals the number of sellers, the clearinghouse always has a zero net position.
Secondary markets in exchange-traded futures and options are made possible by two considerations:
(a) the clearinghouse and (b) the standardized contract design. A buyer who does not wish to hold a position
until maturity enters into another contract of identical terms but on the opposite side prior to maturity. Since
the individual is now buyer and seller of the same contract, the clearinghouse nets out the position. Most
futures and options positions are not held to maturity but are offset in this manner prior to maturity.
In OTC markets, secondary market trading is not possible for two reasons. First, there is no
clearinghouse. This means that the buyer must negotiate with the particular seller with whom the contract
was first arranged in order to undo the contract before maturity. This is cumbersome and also puts the party
that seeks to reverse its position at a competitive disadvantage. Second, contracts in the OTC market are not
standardized. Thus, even if a clearinghouse existed, it would be difficult to find traders on the other side
willing to trade in a very specific instrument. OTC futures and options exist because the tailored contracts
are sufficiently attractive to particular investors to offset the disadvantages of an absence of a secondary
market.
Each of the futures exchanges has its own clearing organization. Options traded on SEC-regulated
securities exchanges (NYSE, AMEX, Philadelphia Exchange, Pacific Exchange and the NASD) are issued
and guaranteed by the Options Clearing Corporation (OCC).
Margin or Performance Bond

Aside from its clerical role of accounting for all contracts and overseeing delivery, the clearinghouse
maintains the financial integrity of markets as guarantor of all contracts. Clearing members post margin
deposits to guarantee the transactions carried out through that clearing member. Not all members of a futures
or option exchange are members of the clearinghouse. Firms that are not members of the clearinghouse clear
through member firms. The ultimate guarantor of the contract is the clearinghouse and its members. Most
exchanges establish a guarantee fund that protects clearing members and therefore the customers of those
clearing members. In futures markets, the clearinghouse imposes margin requirements on its clearing
members. Customer margins are established by individual brokerage firms and are at least as high as the
margins imposed by the clearinghouse. Margins on futures and futures option contracts are not set by any
regulatory authority. Margins on stocks and stock options are regulated by the Federal Reserve Board and the
SEC.
It is important to distinguish margins on futures and options from margins on common stocks. For
this reason the futures exchanges prefer now the term Performance Bond. The margin on a common stock
represents the percentage of the total purchase price paid by the investor. The remainder is borrowed.
Currently, the minimum initial margin on common stocks, set by the Federal Reserve Board, is 50 percent.
Margins on futures contracts represent a performance guarantee. When a futures contract is entered into, no
credit is extended, no asset changes hands, and no payment is made by the buyer to the seller. Full payment
(part of which could be borrowed at that time) is required only if delivery is made. Both the buyer and seller
of a futures contract deposit margin, which may be in the form of interest earning securities rather than cash.
Positions in futures contracts are settled daily as the futures price changes. Traders are required to make up
losses, if any, or are permitted to withdraw profits, if any, each day. These payments from losers to gainers
are called variation margin and must be in cash.
Options have some features of common stocks and some features of futures contracts. As in the case
of common stocks, a payment is made when the option contract is entered into. The buyer of an option pays
a premium for an insurance service rendered by the seller of an option. For example, the buyer of a call
option has the right to purchase the underlying asset at a known exercise price and is insured against any
losses should the underlying asset price fall below the exercise price. Although no asset changes hands when
an option contract is entered into, payment for the “insurance service” is made, and money changes hands.
Under current margin procedures for futures options as well as securities options, the buyer of the call or put
pays 100 percent of the premium. The seller of the option is required to post margin as a performance
guarantee, and the margin must be at least as great as the current market value of the seller’s obligation. In
this way, the clearinghouse is assured that the seller’s obligations will be carried out.
Trading and Settlement Example

Secondary market trading in futures and the role of the clearinghouse is perhaps best illustrated with the help
of a simple example presented in Table 2.1. Suppose that A and B agree to trade one wheat futures contract
(5,000 bushels at $3.00 per bushel) on day 1. The volume of trading in day 1 is one contract, and open
interest at the end of the day –the number of contracts outstanding– is also one. But, A does not settle profits
and losses directly with B. The clearinghouse becomes buyer to B and seller to A. No money is paid by A to
B. Instead, both A and B post margin with the clearinghouse. In the case of wheat, the margin is $750 per
contract for each party to the transaction and may be pledged in T-bills.
On day 2, A decides to sell. The clearinghouse becomes buyer to A and seller to a new entrant into
the market, C. Since A is now a buyer and seller of the same contract on the books of the clearing house, A’s
position is closed out by the clearing house. In effect, C replaces A as the offset long to B’s short position
(without B’s knowledge). At the end of day 2, cumulative volume of trading is two contracts, and open
interest remains at one.
The price of the wheat futures rose from $3.00 to $3.03 from day 1 to day 2. The value of A’s
position therefore rose by $150, so A makes $150. Since B was a seller and prices rose, the value of B’s
position falls by $150, and, even though B did not enter into any transaction, B is required to pay $150 in
variation margin to the clearinghouse. Futures markets require the daily settlement of all positions whether or
not a transaction is entered into. In forward contracts, profits and losses are realized only when an offsetting
transaction is entered into or at maturity. Forward contracts therefore rely much more heavily on the
adequacy of margins or other forms of guaranteeing the contract. In futures markets, the margin deposit is
needed to protect only against the delay in collecting the variation margin from losers.
On day 3, C buys a second contract from D. The cumulative volume of trading through day 3 is three
contracts, and open interest increases to two. The price of the wheat futures has fallen from $3.03 to $2.96,
so B gains $350 and C loses $350. On day 4, B buys one contract and C sells one. Since these transactions
offset previous positions established by B and C, open interest declines by one contract. Since the price did
not change, no profits or losses are realized by any remaining participants in the market. On day 5, C sells
one more contract, and D buys one. Since these transactions offset previously established positions, open
interest declines to zero.
TABLE 2.1 Example of trading in futures and associated cash flows.

Daily Profit or Loss Margin Cash Flow9
Cummul
Open Futures Contract
Time Buyer Seller Contract A B C D A B C D
Interest Price Value
Volume
1 A B 1 1 3.00 15000 -750 -750
2 C A 2 1 3.03 15150 150 -150 750 -750
3 C D 3 2 2.96 14800 350 -350 -750 -750
4 B C 4 1 2.96 14800 750 750
5 D C 5 0 3.00 15000 200 -200 750 750
Total 150 200 -150 -200 0 0 0 0
TABLE 2.2 Example of trading in options and associated cash flows.

Premium Cash Flow Margin Cash Flow10
Cum.
Open Contract Option
Time Buyer Seller Contract A B C D A B C D
Interest Value Premium
Volume
1 A B 1 1 3000 300 -300 300 -1300
2 C A 2 1 3100 310 310 -310 -10
3 C D 3 2 2800 280 -280 280 30 -1280
4 B C 4 1 2800 280 -280 280 1280
5 D C 5 0 3000 300 300 -300 1280
Total 10 20 -10 -20 0 0 0 0
9
Margin deposits are required from both the buyer and the seller. We assume a constant margin deposit of $750 (5%).
10
Margin is required of the seller only. Margin requirements are complex and vary across contracts and exchanges. We assume a margin deposit of $1000 plus the current
value of the option
Derivados Avanzados enero de 2010 Página: 29/358
Futures trading is a zero-sum activity, as is reflected in the example by the fact that the sum of profits of the
four traders is zero each day. On balance, A and B gain; C and D lose. To the extent that margin is pledged
in cash, there is a net loss to traders in the form of foregone interest on the funds pledged as margins.
However, if the margin is pledged in the form of interest earning assets such as U.S. Government Securities,
this loss is avoided.
Table 2.2 presents a corresponding example of trading and settlement in call options on a common
stock. The underlying contract is for 100 shares of the stock which on day 1 is selling for $30. The option
premium is assumed to be 10 percent of the contract value. The structure of the example is similar to the
futures market example but there are some important differences. First, an option transaction requires the
payment of a premium by the option buyer to the option seller. Thus, on day 1, A pays $300 to B. Second,
only sellers of options are required to post margin. Buyers of options pay the full premium and meet all their
obligations at the time the premium is paid. Third, profits and losses arising from changes in the option price
are not settled daily. The buyer of an option can realize gains only by entering into a transaction. Such gains
are not transferred by a daily settlement procedure. Sellers of options do, in effect, realize gains and losses
daily because they are required to adjust their margin positions by the amount of the gain or loss in the
option. That is, their margin is marked-to-market daily. Thus, individual B must post an additional $10 of
margin on day 2 to guarantee the ability to purchase the underlying stock and deliver it.
2.5 SUMMARY
In this chapter, we first discussed the difference between over-the-counter (OTC) option and futures markets
and organized exchange markets. In OTC markets contracts can be tailored to the particular needs of
customers, but secondary market trading is difficult. In organized markets, contracts are standardized so that
secondary market trading is facilitated. Another key distinction between organized futures and option
exchanges and OTC markets is the existence of a clearinghouse. The clearinghouse is central to the operation
of organized futures markets. It interposes itself between buyer and seller and guarantees contracts. It sets
margins for clearing members and settles profits and losses daily. In contrast, in OTC markets, no
clearinghouse exists. The parties make their own arrangements for guaranteeing the contract’s financial
integrity.
Next, the factors that must be considered in designing futures and option contracts are discussed. The
design of contracts balances the benefits of a narrowly and clearly defined contract against the benefits of a
broadly defined contract that is less susceptible to manipulation and is more liquid.
Trading procedures are examined next. The basic types of orders that may be used are similar across
markets, but trading procedures differ quite markedly and range from the open outcry procedures of futures
markets to the specialist system used on certain option exchanges. Types of traders, trading costs, and price
reporting procedures are also discussed.
4. COBERTURA CON CONTRATOS DE FUTUROS
4.1 UNA COBERTURA TRADICIONAL
4.2 RIESGO DE LA BASE

 Riego del Precio de la Mercadería
 Riesgo del Costo de Acarreo
4.3 COBERTURA DE MERCADERÍAS NO-DISPONIBLES (NON-DELIVARABLE)
4.4 COBERTURA ÓPTIMA – ENFOQUE MEDIA/VARIACIÓN

 Utilidad de Cartera de Cobertura
 Ratio de Cobertura Óptimo
4.5 COBERTURA ÓPTIMA – ENFOQUE DE REGRESIÓN DE MÍNIMOS CUADRADOS

 Retorno Esperado
 Ratio de Cobertura Óptimo
 Efectividad de la Cobertura
 Cobertura con Varios Contratos de Futuros
4.6 CALCULANDO EL RATIO DE COBERTURA

 Cambios de Precios, Intervalos de Cambio de Precio y Otros Puntos
 Cobertura con un Solo Contrato de Futuros
 Cobertura Con Dos Contratos
4.7 RESUMEN
4. COBERTURA CON CONTRATOS DE FUTUROS

Un individuo que enfrenta riesgos (protector) utiliza los mercados de futuros para reducir o eliminar el riesgo
de fluctuaciones de precios adversas en la mercadería que compra o vende. El protector en corto (short
hedger) vende corto en los mercados de futuros contra una posición larga en la mercadería base. Un protector
en corto típico es una persona que, en el curso normal de un negocio, mantiene inventario de una mercadería
–por ejemplo, el gerente de una cartera de acciones. El protector en largo (long hedger) compra futuros y
tiene una posición corta en la mercadería base. Una posición corta en una mercadería significa que él tiene
un compromiso futuro a precio fijo para entregar la mercadería base o algo altamente correlacionado con el
precio de la mercadería. Un ejemplo de un protector largo es un exportador que ha prometido entregar trigo a
un precio fijo pero todavía no posee el trigo. El exportador compra un contrato de futuros de trigo para
asegurar el precio en el momento de la entrega y luego va adquiriendo el trigo para entrega. En capítulos
posteriores veremos ejemplos de cobertura cuando se examinan los mercados al contado particulares. En este
capítulo, se discuten los principios básicos de la cobertura.
4.1 UNA COBERTURA TRADICIONAL

El protector se preocupa de los cambios inesperados en el precio de una mercadería base en la cual él tiene
una posición. La Figura 4.1 muestra estos cambios inesperados en los precios. La Figura 4.1 es similar a la
Figura 3.1 excepto que se traza una ruta particular de contratos de futuros, al contado y precios del efectivo.
La ruta se desvía del camino esperado ilustrado en la Figura 3.1 porque el precio de los futuros cae. Como
resultado, los precios al contado y del efectivo también caen. Sin embargo, si los comerciantes hubiesen
anticipado una caída en los precios de los futuros en el momento t , los precios de los futuros hubiesen sido
menores en ese momento. Los protectores negocian en el mercado de futuros para protegerse de las pérdidas
por movimientos de precios inesperados para así poder concentrarse en su actividad comercial principal, que
por ejemplo puede ser, procesar una mercadería o manejar una cartera de valores.
Figura 4.1 Caminos de Precios Al Contado y Futuros

Figure 4.1 Paths of Spot and Futures Prices
500
400
Ft
300
Prices
200
St
100
0 Sit
-100
-200
Time
Un protector con una posición en una mercadería quiere tomar una posición en el mercado de futuros que lo
proteja contra una disminución en el precio del tipo que se muestra en la Figura 4.1. La cobertura es efectiva
si el precio al contado de la mercadería base se mueve de la misma manera que los contratos de futuros de la
mercadería. Otra manera de decir lo mismo es que no hay cambios inesperados en la base, Ft  St . La Tabla
4.1 contiene una ilustración numérica simple de una cobertura corta para alguien en el negocio de almacenar
trigo. En la Tabla 4.1, nosotros asumimos que el que almacena el trigo compra una tonelada de trigo el 1º de
setiembre y se protege vendiéndolo mediante un contrato de futuros para el 1º de diciembre. El 1º de
noviembre, el inventario es vendido y la cobertura es levantada mediante una compra compensadora de

futuros. La Tabla 4.1 considera el valor de esta cartera de cobertura en los dos puntos en el tiempo ilustrados
por dos las líneas verticales en la Figura 4.1.
El 1º de setiembre, la tonelada de trigo se adquiere a $3.00, y se vende un contrato de futuros para
diciembre sobre la tonelada de trigo a $3.09. Nosotros asumimos que el precio de los futuros obedece la
relación base (3.2) desarrollada en el Capítulo 3. También asumimos que los costos de acarreo (incluyendo
una utilidad normal para el dueño del almacén) son de $0.03 por tonelada por mes al momento en que se
establece la cobertura, o $0.09 por un período de tres meses.
Tabla 4.1 Resultados de la utilidad de una cobertura corta, asumiendo una base constante por mes
Mercado en Efectivo (Spot) Futuros a Diciembre
Fecha Transacción Precio Transacción Precio
1º de setiembre Comprar 1 tonelada a 3.00 Vender futuros a 3.09
1º de noviembre Vender 1 tonelada a 2.70 Vender futuros a 2.73
Ganancia -0.30 0.36
Ganancia neta 0.06
Ganancia neta menos costos de almacenaje de $0.03 por mes 0.00
Después de dos meses, el 1º de noviembre, el precio del trigo cayó drásticamente y el trigo se vendió
a $2.70, causando una pérdida de $0.30 en el mercado financiero. Los precios de los futuros, sin embargo,
también cayeron a $2.73, resultando en una ganancia de $0.36. La diferencia entre los precios de los futuros
y el precio en efectivo se ha acortado de $0.09 a $0.03, pero la base por mes sigue en $0.03 ya que sólo falta
un mes para el vencimiento del contrato de futuros. Dado que el protector estaba corto en el mercado de
futuros, la ganancia allí de $0.36 compensa de más la pérdida de $0.30 en el mercado en efectivo o al
contado. Esta ganancia refleja el hecho que la base total, Ft  St , se acorta conforme se va acercando el
vencimiento para reflejar el tiempo reducido a lo largo del cual se debe mantener la mercadería. Sin
embargo, el protector mantuvo la mercadería básica durante dos meses e incurrió en gastos de almacenaje,
los cuales hemos asumido son $0.06 por tonelada. Como resultado la ganancia neta de $0.06 por los cambios
en los precios de los contratos futuros y los precios al contado se compensan con el costo de $0.06 de
mantener la mercadería. La Tabla 4.1 ilustra el caso en el cual la cobertura es completamente efectiva en el
sentido que el protector no sufre pérdidas ni obtiene ganancias.
4.2 RIESGO DE LA BASE

La cobertura no es completamente efectiva si la diferencia entre el precio de los futuros y el precio de la
mercadería base no convergen con la tasa del costo de acarreo durante la vida de los futuros. El riesgo de
base surge si la diferencia, Ft  St , se desvía de la base constante por mes. En general, si la base se amplía
inesperadamente (o se “debilita”), el protector en corto pierde. Si la base se reduce inesperadamente (o se “se
refuerza”), el protector en corto obtienen una ganancia. La ganancia total depende de los costos de acarreo
pagados por el protector. Si los costos de acarreo no son fijos, las ganancias de la cobertura se pueden
compensar con los mayores costos de almacenamiento, y las pérdidas pueden ser compensadas por costos de
almacenamiento menores.
Estos puntos se ilustran en la Tabla 4.2 al asumir precios futuros alternativos el 1º de noviembre.
Supongamos que todas las cosas se mantienen idénticas que en la Tabla 4.1, excepto que el precio de los
futuros cae sólo a $2.75. En otras palabras, el 1º de noviembre, la base se ha debilitado a un nivel de $0.05
por mes. En ese caso el protector sólo gana $0.04, lo cual no es suficiente para cubrir sus costos de
almacenamiento para los siguientes dos meses. Sin embargo, si el protector mantiene una mercadería
disponible contra un contrato de futuros, él podrá elegir no vender la mercadería el 1º de noviembre. Más
bien, el protector puede continuar manteniendo la mercadería hasta el 1º de diciembre y entregarla contra el
contrato de futuros. En el caso de mercaderías disponibles, la convergencia de los precios de los futuros y al
contado garantizan una ganancia de $0.09, lo cual es igual a la base cuando se estableció la cobertura.
Si los costos de almacenamiento para el mes restante se mantienen en $0.03 por tonelada, se obtendrá
una utilidad en exceso de $0.02 en el último mes, la cual es suficiente para compensar la pérdida de $0.02 en
los primeros dos meses. Por otro lado, si los costos de almacenamiento se elevan a $0.05, como se implica
por la base el 1º de noviembre, la pérdida el 1º de noviembre es inevitable. La pérdida no se origina por la
falta de efectividad de la cobertura, sino por la falla de la cobertura de no asegurar los costos de
almacenamiento durante todo el período de tres meses.
Tabla 4.2 Resultado de las utilidades de una cobertura corta con riesgo de base
Mercado al Contado Futuros a Diciembre
Fecha Transacción Precio Transacción Precios Alternativos
1º setiembre Comprar 1 tonelada a 3.00 Vender futuros a 3.09 3.09
1º noviembre Vender 1 tonelada a 2.70 Comprar futuros a 2.75 2.71
Ganancia 0.34 0.38
Ganancia neta 0.04 0.08
Ganancia neta menos costos de almacenaje de $0.03 por mes -0.02 0.02
Esta es sólo otra manera de decir que el tenedor del inventario enfrenta dos fuentes de riesgo: (a)
fluctuaciones en el precio de la mercadería y (b) fluctuaciones en el costo de mantener la mercadería. Los
mercados de futuros pueden ser perfectamente efectivos en la cobertura del riesgo de precio, pero no son
efectivos para proteger el segundo riesgo.
El segundo caso ilustrado en la Tabla 4.2 es una disminución en el precio de los futuros a $2.71, lo
cual implica un estrechamiento o reforzamiento de la base. En este caso, existe una ganancia neta de $0.08,
que cubre de más los costos de almacenaje de $0.06 en los primeros dos meses. En este ejemplo la base ha
cambiado en una dirección favorable. Siempre que el protector no haya contratado un almacén en el último
mes, esta ganancia es real. Sin embargo, si el protector ha pedido prestados fondos y ha contratado un silo de
granos para todo el período de tres meses, se incurre en costo de almacenaje en el tercer mes, ya sea que se
almacene el trigo o no. En ese caso la ganancia no es real. Si se va a incurrir de todas maneras en el costo de
almacenaje de $0.03, es mejor mantener el trigo un mes más para obtener la base de $0.01, la cual compensa
en parte el costo de almacenamiento de $0.03.
El ejemplo anterior y las implicancias enumeradas a continuación se basan en que el protector tiene
una posición en la mercadería base del contrato de futuros. La cobertura de mercaderías no-entregables se
discute en la siguiente sección. A continuación veremos un resumen de las implicancias de los ejemplos en
esta sección (esto es, ejemplos donde el protector mantiene una mercadería disponible y protege utilizando
los futuros) bajo dos encabezados:
Riego del Precio de la Mercadería

La cobertura con contratos futuros puede eliminar el riesgo del precio de la mercadería.
Riesgo del Costo de Acarreo

a. La falla de no asegurar los costos de acarreo de la mercadería durante la vida del contrato de futuros
causa que el protector en corto incurra en el riesgo que el costo se incremente por encima de la cantidad
planeada o esperada. Por el contrario, el protector en corto obtendrá ganancia si los costos de acarreo
caen por debajo de la cantidad esperada.
b. El asegurar los costos de acarreo elimina el riesgo de base de la cobertura. En términos de las Tablas 4.1
y 4.2, si el protector puede garantizar los costos de almacenaje de $0.03 por mes, y si el trigo base se
puede entregar el 1º de diciembre a un precio de futuros de $3.09, se obtendrá una ganancia neta de
$0.00. No existe riesgo en este ejemplo (irreal) porque los precios son pre-establecidos el 1º de
setiembre.
c. Puede no ser posible fijar los costos de acarreo de una mercadería por adelantado. Como resultado, los
tenedores de una mercadería están sujetos a cierta cantidad de riesgo base que se origina por costos de
acarreo inciertos.
La única fuente de riesgo de base cuando una mercadería se puede entregar contra un contrato de futuros es
el riesgo del costo de acarreo.
4.3 COBERTURA DE MERCADERÍAS NO-DISPONIBLES (NON-DELIVARABLE)

Además del riesgo del costo de acarreo, el protector también enfrenta el riesgo del precio de la mercadería
cuando la mercadería que se está protegiendo no se puede entregar para cumplir del contrato de futuros. Si la
mercadería que se está protegiendo no se puede entregar, no es posible eliminar completamente el riesgo del
precio de la mercadería porque el precio al contado y el precio de los futuros no necesitan converger al
vencimiento del contrato de futuros. La Figuras 4.1 y 3.1 ilustran este punto mediante la brecha entre Si y F
al vencimiento. En efecto, el 1º de setiembre, el protector compra la mercadería base y toma un contrato de
futuros para vender otra mercadería (relacionada). El 1º de setiembre se sabe que no existe la garantía que la
tonelada de trigo se puede vender por $3.09 si se mantiene hasta el vencimiento.
Se utiliza el término cobertura cruzada para describir situaciones en las cuales los futuros son
utilizados para proteger mercaderías que no se pueden entregar contra ese futuro. En la práctica, la mayor
parte de las protecciones son cruzadas. Un abastecedor de trigo en Oklahoma posee un nivel de trigo
entregable, pero los costos de transporte para llevar el trigo a un lugar entregable son lo suficientemente altos
para que el trigo no sea entregable a la presentación del contrato de futuros. Esto significa que el precio del
trigo en Oklahoma y en Chicago no convergen. Es más, la diferencia entre el precio del trigo en Oklahoma y
Chicago puede ser incierta debido a las condiciones de demanda en los dos mercados. Otros ejemplos de
cobertura cruzada incluyen (a) futuros de plata para proteger una posición en platino, (b) futuros de la bolsa
de valores para proteger una posición en una sola acción, (c) futuros de bonos del tesoro para proteger una
posición en bonos corporativos y (d) futuros de trigo de Chicago para proteger trigo almacenado en Francia.
Cuando se realiza una cobertura cruzada, es poco probable que los cambios inesperados en el valor de
la posición base se igualen en magnitud a los cambios en el precio de los futuros. Por ejemplo, un cambio de
1 porciento en los precios del platino podría ir acompañado por una cambio de 1.5 porciento en los precios
de la plata. Hasta el grado en que dichas relaciones sean estables a lo largo del tiempo, uno puede ajustar el
tamaño de la cobertura para proporcionar una mejor cobertura cruzada. Ahora veremos un método para
determinar la cantidad óptima a protegerse y para evaluar la efectividad de la cobertura.
4.4 COBERTURA ÓPTIMA – ENFOQUE MEDIA/VARIACIÓN

Una persona que quiere reducir el riesgo del precio de una posición en una mercadería base debe elegir el
tamaño de la posición para aceptar el contrato de futuros. En los ejemplos anteriores, asumimos que la
posición del mercado de futuros es del mismo tamaño que la posición en el mercado al contado, pero esto no
es siempre óptimo. En particular, cuando el precio de los futuros cambia menos que el precio al contado, una
posición de futuros mayor que la posición al contado es lo óptimo. Por otro lado, si el precio de los futuros
cambia más que el precio al contado, lo óptimo son futuros menores que la posición al contado. Para
determinar el número óptimo de contratos futuros a utilizarse, desarrollamos un modelo formal.
Notación
La notación utilizada en el desarrollo del ratio de cobertura óptimo es la siguiente:
S0  precio al contado inicial. (Esta variable corresponde al primer ingreso en la columna de mercado al
contado de la Tabla 4.1; esto es, el valor $3.00.) Estamos omitiendo el subíndice i indicando el precio al
contado.
F0  precio inicial de futuros. (Este valor corresponde al primer ingreso en la columna del mercado de
futuros de la Tabla 4.1, eso es, el valor $3.09.)
ST  precio al contado incierto al momento futuro T . (Esta variable corresponde al valor del 1º de
noviembre de la columna al contado de la Tabla 4.1, el cual es uno de los muchos resultados posibles en ese
momento.)
FT  precios de futuros en T . (Esta variable corresponde al valor el 1º de noviembre en la columna de
futuros de la Tabla 4.1, el cual es uno de los muchos resultados posibles en ese momento.)
nS  número de unidades de mercadería al contado en cartera. ( nS es positivo para las posiciones largas y es
negativo para las posiciones cortas.)
nF  número de contratos de futuros en cartera. ( nF es positivo para las posiciones largas y es negativo para
las posiciones cortas.)
Utilidad de Cartera de Cobertura

La ganancia incierta, ~h , del protector que mantiene nS unidades de la mercadería y protege utilizando
nF contratos futuros es
~ ~
~h  ( ST  S 0 )nS  ( FT  F0 )nF (4.1)
~
El término ( ST  S 0 ) es el cambio de precio (aleatorio) por unidad de la mercadería a lo largo de la vida de la
~
cobertura, y el término ( FT  F0 ) es el cambio de precio (aleatorio) del contrato de futuros a lo largo de la
vida de la cobertura. La ganancia esperada para el protector a lo largo de la vida de la cobertura es
~ ~
E (~h )  [ E ( ST )  S 0 ]nS  [ E ( FT )  F0 ]nF (4.2)
La expresión (4.2) proporciona una manera conveniente para ilustrar dos ideas. Primero, si asumimos que el
precio actual de los futuros es un mecanismo de predicción del precio de los futuros en el momento T como
se implica en la Figura 4.1, esto es, si
~
F0  E ( FT ) (4.3)
el último término de la ecuación (4.2) se elimina. La utilidad esperada sobre la cartera de cobertura depende
solamente del cambio esperado en el precio al contado. En un mercado de costos de acarreo, la diferencia
entre el precio al contado actual y el precio al contado futuro esperado representa los costos de almacenaje.
Segundo, si la mercadería está disponible y se mantiene hasta el vencimiento, la convergencia asegura que
~ ~
E ( ST )  E ( FT ) (4.4)
Combinando (4.2), (4.3) y (4.4) obtenemos
E (~h )  ( F0  S 0 )nS (4.5)
En otras palabras, la utilidad esperada sobre la cobertura es directamente proporcional a la base inicial como
se indicó en la Tabla 4.1. El resultado general dependerá de los costos de acarreo incurridos por el protector
que no son incorporados en la ecuación (4.1). Incluso si la mercadería es no-entregable, la ganancia esperada
es igual a la base, siempre que la diferencia entre el precio de entrega inmediata y el precio al contado en la
Figura 3.1 permanezca constante.
Los precios futuros realizados y los precios al contado se desvían de los precios esperados; por lo
tanto, las ganancias realizadas son mayores o menores que las ganancias esperadas. Esto es cierto, en
particular, para la cobertura cruzada en cuyo caso no hay oportunidad para entregar la mercadería. La
cobertura cruzada óptima es la cobertura que minimiza la desviación a partir de la ganancia esperada. Ahora
veremos una derivación de la cantidad protegida óptima basada en este criterio.
Ratio de Cobertura Óptimo

Para derivar el número óptimo de contratos de futuros a venderse, primero escribimos la ecuación (4.1) en
términos de cambios en el precio por unidad de la mercadería base, esto es,
~h ~ ~ n
 ( ST  S 0 )  ( FT  F0 ) F . (4.6)
nS nS
~ ~ ~ ~
Ahora definamos los términos del cambio de precio,  S  ( ST  S 0 ) y  F  ( FT  F0 ) , y sustituimos estos
términos en (4.6),
~h ~ ~
  S  h F . (4.7)
nS
El ratio de cobertura, h  nF / nS , , es el número de contratos de futuros por unidad de la mercadería base.
Como el protector está preocupado por minimizar el riesgo11, utilicen (4.6) para escribir la variación de la
utilidad de la cartera de cobertura.
 h2   S2  h 2 F2  2h SF , (4.8)
donde  h2 es la variación de la utilidad de la cartera de cobertura por unidad de mercadería;  S2 es la

variación del cambio del precio al contado;  F2 es la variación del cambio del precio de los futuros; y  SF es
la covariación entre los cambios del precio al contado y el precio de los futuros. El valor de h que minimiza
 h2 se encuentra tomando la derivada de (4.8) con respecto a h y haciéndola igual a cero:
Despejando h* , el ratio de cobertura óptimo es
 SF
h*  . (4.9)
 F2
11
El marco referencial del ratio de cobertura media-variación desarrollado aquí es similar al desarrollado en Ederington (1979). En
este marco referencial, se asume que el protector sólo está interesado en minimizar su exposición al riesgo. En un marco
referencial más general, se permitiría al protector considerar no sólo la variación en el precio sino también el cambio de precio
esperado para la determinación del ratio de cobertura óptimo.
Por lo tanto, el ratio de cobertura óptimo, h* , depende de la covariación entre el cambio en el precio al
contado y el precio de los futuros con relación a la variación del cambio en el precio de los futuros. Es
interesante notar que la expresión para el ratio de cobertura óptimo,  SF /  F2 , es la pendiente de la regresión
~
normal de mínimos cuadrados (OLS) del cambio del precio al contado,  S , sobre el cambio en el precio de
~
los futuros,  F . El enfoque de la regresión de los mínimos cuadrados para calcular el ratio de cobertura
óptimo se describe a continuación.
4.5 COBERTURA ÓPTIMA – ENFOQUE DE REGRESIÓN DE MÍNIMOS

CUADRADOS
La regresión normal de mínimo cuadrados12 proporciona una manera alternativa para derivar el ratio de
cobertura óptimo. Consideren la siguiente ecuación de regresión:
~ ~
 S   0  1 F  ~, (4.10)
donde  0 y 1 son parámetros de la regresión, ~ es un término aleatorio con E (~ )  0 . La ecuación (4.10),
la cual es trazada en la Figura 4.2, muestra que el cambio en el precio al contado tiene tres componentes: (a)
un componente constante no-aleatorio de cambio en el precio  0 , la intersección en la figura; (b) un
~
componente aleatorio que es relacionado sistemáticamente con el cambio en el precio de los futuros, 1 F , y
(c) un componente aleatorio singular, ~ , que no está correlacionado con el cambio del precio de los futuros
y es representado por las distancias verticales entre la línea y los puntos fuera de la línea en la Figura 4.2.
Retorno Esperado
El término de intersección,  0 , captura cualquier cambio esperado en el precio al contado acompañado por
~
un cambio esperado en el precio de los futuros. De la Figura 3.1, sabemos que, E ( F )  0 , el cambio del
~ ~
precio al contado esperado es igual a la base, esto es, E ( S )  E ( ST  S 0 )  F0  S 0 . El modelo de regresión
~
establece que el cambio en el precio al contado esperado es igual a  0 bajo la premisa que E ( F )  0 (y que
E (~ )  0 ), por lo tanto el término de intersección en (4.10) representa la base. La base, a su vez, refleja los
costos de almacenaje que el abastecedor de los activos debe recuperar mediante el aumento del precio. El
~
término, 1 F , refleja el hecho que los cambios aleatorios en el precio de los futuros se reflejará en el precio
al contado de acuerdo con el coeficiente de inclinación, 1 . El término ~ refleja el riesgo de base, el cual se
~
origina del hecho que ciertos cambios aleatorios en  S , , son únicos para la mercadería al contado y no están
correlacionados con el cambio del precio de los futuros.
12
En Pyndyck y Rubinfeld (1981, Capítulos 1-6) se proporciona una excelente explicación de la regresión de mínimos cuadrados.
Figura 4.2 Cambio del Precio al Contado Versus Cambio del Precio de los Futuros
Cambio Precio
Spot,  S



 

 


1
 1

Cambio Precio de

Futuros,  F
Ratio de Cobertura Óptimo

La expresión original para la utilidad de la cartera de cobertura (4.7) se puede reformular sustituyendo la
~
ecuación de la regresión (4.10) por  S , eso es
~h ~ ~ ~
  0  1 F    h F   0  (1  h) F   (4.11)
nS
~
La Ecuación (4.11) muestra claramente que la utilidad de la cartera de cobertura,  F , puede ser
independiente de los movimientos de los precios al contado y futuros al establecer h  1 . Si 1  1 , un
cambio de un dólar en el precio al contado es equiparado con un cambio de un dólar en los precios de los
futuros. Esto implica que los precios al contado y de los futuros se mueven a la par. En este caso, la
cobertura óptima es h  1 , o una cobertura de 100 porciento. Si 1  0.0 , los precios de los futuros y el
precio al contado no están relacionados y no tiene sentido proteger y el ratio de cobertura óptimo es cero
(h  0) .
Efectividad de la Cobertura
Una cobertura es completamente efectiva sólo si los cambios en el precio de los futuros y el precio al
contado están perfectamente correlacionados. Esto significa que el término de error aleatorio, ~ , en (4.19)
siempre es cero. Aunque podría ser óptimo proteger 100 porciento, la cobertura puede no ser totalmente
efectiva porque pueden originarse desviaciones de la relación promedio, representada por los puntos fuera de
la línea en la Figura 4.1.
Para poder medir la efectividad de una cobertura, primero se mide la falta de efectividad de una
cobertura. La falta de efectividad de una cobertura es medida por el ratio de variación del cambio en el
precio de la cartera protegida (  h2 ) sobre la variación del cambio del precio de la cartera sin cobertura
(  h2 cuando h  0 ). De la ecuación (4.8), el cambio del precio de la cartera sin cobertura sobre la variación
del cambio del precio sin cobertura es  S2 . De (4.11), la variación del cambio del precio de la cartera
protegida óptimamente ( h  1 ) es  2 , esto es, la variación del término residual, ~ , es el modelo de
regresión. El ratio que mide la falta de efectividad de la cobertura es por lo tanto  2 /  S2 ; este ratio puede
variar entre 0 y 1. Dado que la efectividad de la cobertura es simplemente el complemento de la falta de
efectividad, la efectividad de la cobertura se puede medir como
 2
 1 . (4.12)
 S2
Coincidentemente, la expresión (4.12) es la definición de los R-cuadrados ajustados de la regresión

~
normal de los mínimos cuadrados (4.10). Si la variable independiente,  F , explica 100 porciento de la
~
variación en  S ,  2  0 y  2  1 . En este caso, la cobertura es perfectamente efectiva. Si la variable
~ ~
independiente  F , no explica nada de la variación en  S ,  2   s2 y  2  0 y la cobertura es tiene una
completa falta de efectividad.
Si 0   2  1 , la cobertura sólo es parcialmente efectiva en el sentido que algunos de los
movimientos en el precio al contado son reflejados en los precios de los futuros, pero algunos movimientos
son únicos. La cobertura es efectiva al eliminar los movimientos sistemáticos que también se reflejan en el
precio de los futuros. La cobertura no es efectiva al eliminar el riesgo de precio único que aparece en el
término aleatorio ~ .
Cobertura con Varios Contratos de Futuros

El enfoque de la regresión normal de mínimos cuadrados se puede generalizar muy fácilmente para manejar
casos donde se utilizan dos o más contratos diferentes para proteger una posición al contado. El valor de una
cartera de bonos corporativos, por ejemplo, es sensible tanto al riesgo de tasa de interés como al riesgo de la
bolsa de valores. Para proteger el valor de este tipo de cartera, se requieren protecciones contra los
movimientos tanto en las tasas de interés como en el nivel de la bolsa de valores. Los contratos de futuros de
bonos del tesoro y los contratos de futuros del índice S&P 500 son probablemente los más adecuados para
proteger simultáneamente los dos tipos de exposición al riesgo de esta cartera.
Para protegerse contra múltiples fuentes de riesgo de precio, se realiza una regresión del cambio en el
precio al contado sobre los cambios de precios de varios contratos futuros en la forma,
~ ~ ~ ~
 S   0  1 F ,1  1 F , 2    n  F ,n  ~, (4.13)
Los coeficientes de regresión estimados 1 a  n , son los respectivos ratios de cobertura para cada uno de los
n contratos de futuros. Para protegerse contra todas las fuentes de riesgo, establezcan hi  ̂ i para todos
los contratos futuros. Para proteger solamente exposiciones de riesgo seleccionadas, establezcan hi  ̂ i
para los contratos de futuros correspondientes a los riesgos seleccionados.
4.6 CÁLCULO DEL RATIO DE COBERTURA

En esta sección, se brindan dos ejemplos del marco referencial del ratio de cobertura óptimo. En el primero,
un gerente de cartera de acciones utiliza futuros del índice de acciones para protegerse del riesgo del precio,
y en el segundo, el valor de un bono corporativo es protegido utilizando futuros del índice de acciones y de
bonos del Tesoro. Antes de describir las dos aplicaciones, sin embargo, se deben discutir algunos temas
relacionados con la medición del cambio del precio y el cálculo de la regresión.
Cambios de Precios, Intervalos de Cambio de Precio y Otros Puntos

Una decisión importante que enfrenta el analista que está tratando de calcular el ratio de cobertura óptimo es
la decisión sobre la duración del intervalo de tiempo sobre el cual se deben medir los cambios en el precio.
El modelo de regresión de series de tiempo
~ ~
 S ,t   0  1 F ,t  ~t , (4.14)
da cierta guía sobre este tema. El término de error, ~t , en (4.14) es gobernado por las siguientes premisas:
E ( )  0, E (  ), E ( 2 )   2 y E ( ,  )  0 . El intervalo de cambio en el precio debe ser elegido de
t t F ,t t  t t 1
manera que ninguna de estas premisas de regresión sea violada.

Los intervalos de cambio en el precio utilizados más frecuentemente son diarios, semanales y
quincenales. Manteniendo la duración del período de cálculo constante (digamos, un año de datos), uno
podría imaginar que mientras más frecuentes son las observaciones del precio, más información se está
recolectando sobre la covariabilidad de los cambios en el precio al contado y el precio de los futuros. Pero,
las observaciones demasiado frecuentes también originan otros problemas. Por ejemplo, los precios de la
transacción son generalmente en los niveles de oferta o demanda. Mientras más corto es el intervalo, el
mayor movimiento aleatorio entre los precios de oferta y demanda contribuye a la variabilidad del cambio en
el precio y los resultados de la regresión se vuelven menos confiables. El efecto del precio de oferta/demanda
introduce una correlación en serie negativa en la serie de cambios de precio del título valor y un problema de
errores en las variables en el cálculo (4.14).
Un segundo problema con los cambios en el precio medidos durante intervalos de tiempo cortos es
que si el título valor al contado y el contrato de futuros no son negociados con igual frecuencia, los cambios
en el precio de los dos instrumentos pueden no reflejar el mismo conjunto de información del mercado. Este
problema se manifiesta a sí mismo mediante la autocorrelación no-cero en el término de error, esto es,
E ( t ,  t 1 )  0 .
Un tercer problema tiene que ver esencialmente con la estacionalidad en los cambios en el precio del
título valor. French (1980) y Gibbons y Hess (1981), por ejemplo, han documentado un efecto día-de-la-
semama en los retornos de las acciones y los retornos de ciertos instrumentos del Tesoro. La estacionalidad
día-de-la-semana causará que se viole la premisa del término de error homoscedástico, E ( t2 )   2 cuando
se utilicen en el cálculo los cambios diarios en el precio.
Para ilustrar los efectos de las diferentes premisas de la medición del intervalo de cambio de precio,
calculamos la regresión (4.14) utilizando cambios de precios diarios, semanales y quincenales para el índice
S&P 500 y contratos de futuros del índice S&P 500 durante el año calendario 1989. En 1989, hubo 252 días
de negociación, de manera que se calcularon 251 cambios diarios. Los 251 días produjeron 51 cambios de
precios semanales y 25 cambios de precios quincenales. Los cambios de precios semanales y quincenales son
medidos de miércoles a miércoles porque hay menos miércoles feriados que feriados en otros días de la
semana. No se utilizan las semanas parciales al comienzo y final de la serie de cambios de precios diarios.
Los cambios de precios de los futuros son para los contratos de futuros próximos. Al vencimiento del
contrato más próximo, la serie de cambios de precios de los futuros se empalma de ese contrato al siguiente
contrato más próximo13. Los resultados de la regresión se reportan en la Tabla 4.3.
13
Para 1989,se utilizan los precios de los contratos de futuros del índice S&P 500 de marzo de 1989, junio de 1989, setiembre de
1989, diciembre de 1989 y marzo de 1990 para generar la serie de cambios de precios de los futuros. El contrato de marzo de 1989
es utilizado hasta el último día de negociación el 16 de marzo de 1989. En ese día, se registran tanto los precios los contratos de
marzo de 1989 como de junio de 1989. El precio de los contratos de marzo se utiliza en combinación con su precio el día anterior
para calcular el cambio de precio de los futuros del 16 de marzo. El precio de los contratos de junio es utilizado en combinación
con su precio el día siguiente para calcular el cambio del precio de los futuros para el 17 de marzo. Este procedimiento de
“empalme” permite que la serie de cambios del precio de los futuros sea continua a lo largo del año.
Tabla 4.3. Resumen de los cálculos del coeficiente de ratio de cobertura utilizando los cambios14
diarios, semanales y quincenales del precio del índice S&P 500 y los futuros del índice S&P 500
durante el año calendario 1989
 S ,t   0  1 F ,t  t . Intervalo de Confianza de 95 Porciento.
Intervalo n ̂1 S (ˆ1 ) 15 Inferior Superior Rango R2
Diario 251 0.8034 0.0226 0.7589 0.8478 0.0889 0.8352
Semanal 51 0.9914 0.0163 0.9586 1.0241 0.0673 0.9867
Quincenal 25 1.0013 0.0170 0.9662 1.0365 0.0703 0.9928
Un número de resultados interesantes surgen de la Tabla 4.3. El primero es que pendiente calculada
utilizando cambios de precios diarios, 0.8034, es drásticamente diferente de aquellos utilizados para los
cambios de precios semanales y quincenales, 0.9914 y 1.0013, respectivamente. Esta diferencia resulta de los
problemas arriba mencionados. Por ejemplo, el efecto de precio oferta/demanda introduce el problema de
errores en las variables en la regresión de precios diarios, la cual tiende a sesgar los coeficientes hacia abajo.
Otra comparación interesante en la Tabla 4.3 es la de los intervalos de confianza. Manteniendo otros
factores constantes, el rango del intervalo de confianza (el error estándar) en 1 , debería aumentar conforme
el intervalo de cambio de precio se incrementa de diario a semanal y quincenal porque más y más
información se está perdiendo. Sin embargo, ese no es el patrón que aparece en la Tabla 4.3. El error
estándar y el rango del intervalo de confianza son menores para las regresiones de cambio de precios
semanales y quincenales que en la regresión de cambio de precio diaria. Una vez más, este resultado refleja
los problemas asociados con el uso de cambios de precios diarios.
La comparación de los resultados semanales y quincenales también favorece el uso de los datos
semanales en el procedimiento de cálculo. Las magnitudes del coeficiente están muy cercanas, y aún así el
error estándar y el rango del intervalo de confianza son menores para la regresión semanal. El error estándar
inferior refleja el hecho que en la regresión semanal y la quincenal se introduce casi el doble de información
sobre cambios de precios.
Relacionado a la selección del intervalo apropiado de cambio de precio se encuentra el horizonte de
tiempo de la cobertura. El marco referencial de la cobertura desarrollado en este capítulo es para un solo
período. En principio, cuando el ratio de cobertura es calculado mediante el análisis de regresión, el cálculo
del ratio de cobertura es independiente de la distancia en las observaciones del precio en la regresión. Sin
embrago, el término de intersección aumentará conforme se incremente el tiempo entre las observaciones de
los precios porque es un estimado de la base entre los precios de futuros y al contado durante la duración del
intervalo de observación.
Finalmente, para terminar la discusión, también es importante notar que el parámetro estimado 1 , se
produce de una regresión de datos históricos y que la cobertura que estamos elaborando es para un período
futuro (es decir, los estimados de a son ex-post, pero las decisiones de cobertura ex ante). Al aplicar este
procedimiento, nosotros estamos invocando implícitamente una premisa de inmovilidad en la relación entre
los cambios de precios al contado y de los futuros. A priori, debemos está cómodos con que esa premisa sea
razonable.
14
Los cambios de precios del índice S&P 500 y los contratos de futuros del índice S&P 500 se calculan como  S ,t  ( St  St 1 ) 6+, y
 F ,t  ( Ft  Ft 1 ) , respectivamente.
15
s () es el error estándar del coeficiente de regresión.
Cobertura con un Solo Contrato de Futuros

Para ilustrar cómo se puede determinar la cobertura óptima para una cartera de acciones, consideren a un
administrador de cartera que tiene $50 millones en una cartera de acciones similar en composición al índice
S&P 500 a fines de 1989. Temiendo que el mercado de valores caiga en los próximos dos meses, el
administrador quiere reducir su riesgo de precio vendiendo contratos de futuros del índice S&P 500.
¿Cuántos contratos debe vender?
Los resultados de la regresión reportados en la Tabla 4.3 son útiles para calcular el número de
contratos a venderse. La pendiente estimada (ratio de cobertura) de la regresión de cambios en los precios
S&P 500 semanales sobre los cambios en los precios de los contratos de futuros del índice S&P 500 es
0.9914. En otras palabras, un cambio de un dólar en el precio de los futuros produce un cambio de 0.9914 en
el precio de la cartera del índice de acciones, o alternativamente, el número óptimo de contratos futuros a
venderse por cada unidad invertida en la cartera de acciones es 0.9914. Para determinar el número de
unidades invertidas en la cartera de acciones, se divide el valor de la cartera entre el nivel del índice S&P
500. Para fines de 1989 el índice S&P 500 era 353.40. El número de unidades de la cartera de acciones por lo
tanto es 141,482.74. Además, el contrato de futuros S&P 500 es denominado en 500 veces el nivel del
índice, de manera que el número de unidades de la cartera de acciones expresado en la denominación del
contrato de futuros es 141,482.74/500 = 282.97. Si el ratio de cobertura óptimo fuese uno, deberíamos
vender 282.97 contratos de futuros para proteger la cartera de acciones de $50 millones. Sin embargo, dado
que el ratio de cobertura óptimo es 0.9914, el número óptimo de contratos a venderse es (0.9914)(282.97) =
280.54.
Vale la pena enfatizar que la regresión produce sólo un estimado del ratio de cobertura óptimo. En
este caso, el coeficiente de determinación relativamente alto de 0.9867 indica que el estimado y la
efectividad de la cobertura parecen ser bastante buenos. No obstante, aún en el mejor estimado de 0.9914
existe un potencial de error. El intervalo de confianza de 95 porciento dice que el ratio verdadero estará entre
0.9586  1  1.0241 .
Este rango implica que tenemos 95 porciento de confianza que el número óptimo de contratos de futuros a
venderse esté entre 271.26 y 289.79.
Cobertura Con Dos Contratos

Para ilustrar la cobertura con dos contratos de futuros, consideren la cobertura de $10,000,000 en bonos
cupón de 8½ de Mobil Oil que vencen en el año 2001. Los bonos corporativos tienen una exposición a la tasa
de interés a largo plazo y al mercado de valores, de manera que se realiza una regresión de los cambios de
precios semanales del bono de Mobil sobre los cambios de precio del contrato de futuros de bonos de Tesoro
de CBT y del contrato de futuros del índice S&P 500 para el año 1989. Una vez más, asumimos que la
cobertura está siendo formada a fines de diciembre de 1989. Los resultados de la regresión son reportados en
la Tabla 4.4. También se reportan los resultados cuando los cambios en los precios de los contratos de
futuros de bonos del Tesoro y del índice S&P 500 se utilizan separadamente como variables independientes.
Los resultados en la Tabla 4.4 indican que el contrato de futuros del bono del Tesoro explica más la
variación en los cambios en el precio del bono de Mobil que el contrato de futuros S&P 500. Cuando se
consideran los resultados de la regresión de una sola variable, la efectividad de la cobertura del contrato de
futuros del bono del Tesoro es 0.7533 y la efectividad de la cobertura de los futuros S&P 500 es 0.1920.
Pero, si el bono de Mobil tiene exposiciones de tasa de interés y del mercado de valores, se deben utilizar los
resultados de la regresión múltiple al establecer la cobertura. De hecho, cuando se consideran los resultados
de la regresión múltiple, la efectividad de la regresión, 0.7651, es mayor que cualquier contrato utilizado por
sí solo. El número óptimo de contratos de futuros de bonos del Tesoro a venderse es 0.4473 y el número
óptimo de contratos de futuros S&P 500 a venderse es 0.0159 por cada dólar invertido en los bonos Mobil.
Una vez más, se deben tomar en cuenta las denominaciones del contrato. El contrato de futuros del bono del
Tesoro tiene una denominación de $100,000, de manera que el número de unidades de la mercadería al
contado en términos de futuros de bonos del Tesoro es 100. El nivel del índice S&P 500 era 353.40 a fines de
1989 y la denominación del contrato de futuros S&P 500 es 500 veces el índice, de manera que el número de
unidades de la mercadería al contado en términos de los futuros S&P 500 es 56.59. Utilizando los estimados
de coeficiente reportados en la Tabla 4.4, el número óptimo de futuros de bonos del Tesoro a venderse es 100
x 0.4473 = 44.73, y el número óptimo de futuros S&P 500 a venderse es 56.59 x 0.0159 = 0.90.
Tabla 4.4. Resumen de los estimados del coeficiente del ratio de cobertura utilizando cambios16 de
precios semanales para bonos de Mobil Oil al 8½ porciento con vencimiento en el año 2001 (MO),
contrato de futuros del bono del Tesoro CBT (TBF), y contrato de futuros S&P 500 de CME (SPF)
~ ~ ~
durante el año calendario 1989:  MO ,t   0  1 T  B ,F ,t   2  SP ,F ,t  ~t ,
n ̂1 s (ˆ1 ) 17 t (ˆ1 ) 18 ̂ 2 s (ˆ 2 ) t (ˆ 2 ) R2
Futuros de Bono del Tesoro y S&P 500:
51 0.4473 0.0407 10.98 0.0159 0.0091 1.75 0.7651
Solo Futuros de Bono del Tesoro:
51 0.4756 0.0382 12.46 0.7553
Solo Futuros de S&P 500:
51 0.0554 0.0154 3.59 0.1920
4.7 RESUMEN
este capítulo comienza con una explicación de la cobertura corta tradicional en la cual el propietario de una
mercadería asume una posición corta en futuros para asegurarse contra una disminución en el precio de la
mercadería. Un protector en corto asegura la base, Ft  S t . Los protectores en corto enfrentan el riesgo que la
base pueda cambiar a lo largo del tiempo. En el caso de los protectores de mercaderías entregables o
disponibles, el riesgo de base representa el riesgo de un incremento en los costos de mantener la mercadería.
En el caso de los protectores de mercaderías no disponibles contra un contrato de futuros (cobertura
cruzada), el riesgo de base también representa el riesgo de cambio de los precios relativos de la mercadería y
el contrato de futuros.
En este capítulo es desarrollado un marco referencial para la cobertura según el análisis de cartera
media/variación y también dentro de un marco referencial de regresión de mínimos cuadrados. Se define el
ratio de cobertura óptimo y se discute la medición de la efectividad de la cobertura. El marco referencial de
cobertura óptima es entonces aplicado en el contexto del manejo del riesgo de una cartera de acciones y de
una cartera de bonos, dando cuidadosa consideración a la naturaleza de los datos del precio del título valor y
el cálculo de la regresión.
16
Los cambios en los precios son calculados como  i ,t  I t  I t 1 donde I representa el bono de Mobil Oil, el contrato de futuros del
bono del Tesoro o el contrato de futuros S&P 500.
17
s() es el error estándar del coeficiente de regresión.
18
t () es el valor- t del coeficiente de regresión bajo la hipótesis nula que   0 .
5 RISK AND RETURN IN FUTURES MARKETS
5.1 ROLE OF SPECULATORS

 Many different types of speculators and speculative strategies are possible.
5.2 IS THE FUTURES PRICE AN UNBIASED ESTIMATE OF THE EXPECTED SPOT PRICE?
5.3 THE CAPITAL ASSET PRICING MODEL FOR FUTURES CONTRACTS
5.4 EQUILIBRIUM OF HEDGERS AND SPECULATORS
5.5 SUMMARY
5 RISK AND RETURN IN FUTURES MARKETS
Hedgers enter the futures market to reduce or eliminate the risk of a commodity position. Conceivably, the
futures market could consist solely of hedgers. For example, farmers may want to sell wheat futures, while
processors of wheat may want to buy wheat futures. Both these parties are hedgers and could have futures
transactions that exactly offset each other. But usually the transactions of the long and short hedgers are not
exactly offsetting. Liquid and active futures markets typically require the participation of speculators.
Speculators analyze information concerning futures contracts and their underlying commodities in the hope
of identifying, and profiting from, futures contract mispricings.
5.1 ROLE OF SPECULATORS

Speculators are willing to bear risk that others –hedgers– wish to avoid. Consequently, society benefits.
Speculators also help determine futures prices that more accurately reflect underlying economic conditions.
This is sometimes called the price discovery function of futures markets. Speculators help “discover” the
right price by analyzing underlying economic conditions and trading based upon their analyses. If the futures
price is too low, speculators buy futures, and, if the futures price is too high, they sell. In equilibrium, the
futures price is a consensus estimate of what speculators think the future price of the underlying commodity
ought to be. If speculators are wrong, they lose money. If they are right, they make money. As in any
financial market, the profit motive causes prices to be good estimates of true economic values.
Many different types of speculators and speculative strategies are possible.

Sometimes a speculator concentrates on a particular commodity and, based upon his analysis, concludes that
the price will go up (or down). In this case, a naked long (or naked short) position is appropriate. At other
times, a speculator may conclude that a particular futures contract will increase in price relative to another
futures contract. In this case, the speculator buys one contract and sells the other, thereby becoming a
“spreader.” A spreader benefits only from a favorable change in the difference between two futures prices,
whereas a naked speculator benefits from a change in a particular futures price.
5.2 IS THE FUTURES PRICE AN UNBIASED ESTIMATE OF THE EXPECTED SPOT PRICE?
~
The expected spot price, Et ( ST ) , is the market’s expectation at time t of the spot price when the futures
contract expires (time T ). At any given time t prior to contract maturity, the expectation is likely to be
wrong. Unexpected events after time t cause the actual spot price to be different from what was expected.
Over a long period of time and many futures contract cycles, however, expectations should, on average, be
~
realized, that is, Et ( ST )  avg ST , in other words, the average spot price at maturity ought to reflect the
expectation at some prior time.
In order for speculators to make money, futures prices must trend upward toward the expected spot
price when speculators are long futures, and futures prices must trend downward when speculators are short
~
futures. A market in which futures prices trend upward [i.e., where Ft  Et ( ST ) ] is said to exhibit normal
backwardation. This situation is illustrated as the upward sloping line in Figure 5.1. A market in which
~
futures prices trend downward [i.e., where Ft  Et ( ST ) ] is said to exhibit contango. This situation is depicted
in Figure 5.2 as the downward sloping line. If there is no trend in the futures price, the futures price is an
~
unbiased estimate of the expected spot price, that is, Ft  Et ( ST ) .
FIGURE 5.1 Normal Backwardation
Price ~
Et (ST )
Convergence
Ft
St
Time
FIGURE 5.2 Contango
Price
Ft
Convergence
~
Et (S T )
St
Time
Students of futures markets have long discussed whether or not speculators as a group make money. Some,
like Keynes (1930), Hicks (1939), and Cootner (1960a, 1960b) argue that speculators make money because
they bear risk and must be compensated for their risk-bearing services. They usually argue that speculators I
0 tend to be long futures because hedgers tend to be short.19 Sales by hedgers force the futures price below
the expected spot price and lead to the situation of normal backwardation shown in Figure 5.1. Speculators
make money on the upward trend in futures prices.
Others, particularly Telser (1958, 1960), argue that speculators as a group are not risk-averse and do
not require compensation for risk. This is possible if there are different categories of speculators.
Professional speculators have to make money. Otherwise, they would be unable to support themselves. But
19
That is, on balance, there are more short hedgers than long hedgers.
amateur speculators could lose money to professional speculators, so speculators as a group just break even.
Amateur speculators consist of two categories-gamblers and fools.
Gamblers enjoy the risks of small futures positions. They know the risks and the fact that there is a
house-take, but they enjoy the game. Fools believe they have a successful strategy. They think they know
how to make money in futures, but do not. Fools tend to lose money and then withdraw from the market. The
supply of fools is replenished by Barum’s Law. (There’s a sucker born every minute!) Telser thus argues that
speculators as a group do not make money even though they bear risk. If Telser’s argument is true, hedgers
are better off because they are provided insurance at no cost.
Finally, some argue that the amount of risk actually borne by speculators is small, if risk is properly
measured. Dusak (1973) takes this position. In modem finance theory, the appropriate measure of risk is the
amount of risk that cannot be diversified away. In other words, risk is measured in a portfolio context. Dusak
argues that commodity risk can be diversified away so that the systematic risk of a commodity is zero. That
means that speculators do not require a risk premium. Competition among speculators for futures contracts
will then drive the futures price to a point such that the futures price equals the expected spot price. To the
extent that the systematic risk of futures contracts is negative, speculators might be willing to accept losses.
For example, suppose futures were a good inflation hedge. In such a situation, speculators would be willing
to lose money in futures as a way to reduce the risk in other parts of their portfolio.
5.3 THE CAPITAL ASSET PRICING MODEL FOR FUTURES CONTRACTS

The most familiar form of the Sharpe (1964)-Lintner (1965) capital asset pricing model (CAPM) is the
security market line relation,
~ ~
E ( R )  r  [ E ( RM )  r ] , (5.1)
~
where r is the riskless rate of interest, RM is the rate of return on the market portfolio, and  is the asset’s
beta or relative systematic risk coefficient. Equation (5.1) says that the expected rate of return on a risky
~
asset, E ( R ) , equals the riskless rate of return, r , plus a market risk premium equal to the market price of
~
risk, [ E ( RM )  r ] , times the asset’s relative systematic risk level,  . Assume that the entire asset return is
~ ~
derived from price appreciation (depreciation), that is, R  (VT  V0 ) / V0 , where V is the asset price. Then
equation (5.1) may be rearranged to derive an expression for the current asset price:
E (VT )   Cov(VT , R M ) /  R2M 

V0  (5.2)
1 r
~
where   [ E ( RM )  r ] . Equation (5.2) says that the value of an asset today is the difference between
expected terminal value of the asset and an appropriate risk adjustment discounted to the present at the
riskless rate of interest.
To see how the capital asset pricing model applies to futures contracts, recognize that the terminal value of a
long futures position is
~ ~
VT  FT  F0 , (5.3)
~
where F0 is the futures price when the contract is negotiated and FT , is the uncertain futures price when the
contract expires.20 Substituting (5.3) into (5.2) yields
E ( FT )  F0   Cov( FT  F0 , R M ) /  R2M 

V0  (5.4)
1 r
However, the value of a futures position when the contract is first entered into is zero (i.e., the futures
position involves no initial investment outlay). Setting V0  0 in (5.4) and rearranging yields
E ( FT )  F0   Cov( FT  F0 , R M ) /  R2M  (5.5)
Dividing both sides of (5.5) by the initial futures price F0 ,

E ( R F )   Cov( R F , R M ) /  R2M  (5.6)
~ ~ ~
where RF , is the futures return, that is, RF  ( FT  F0 ) / F0 . Simplifying, equation (5.6) can be written as
~ ~
E ( RF )  [ E ( RM )  r ] F , (5.7)
where  F is the futures contract beta.
Figure 5.3 contrasts the expected return-risk relation for assets with the expected return/risk relation for
futures contracts. The two lines depicted in Figure 5.3 have the same slope, however the line corresponding
to assets has an intercept equal to the riskless rate of interest while the line corresponding to futures contracts
goes through the origin. Since asset positions require an investment outlay, an expected rate of return of at
least the riskless rate is earned on the asset position. Since futures positions require no investment outlay, the
minimum expected return for a futures contract is zero. In equilibrium, the expected rate of return on futures
equals only a market risk premium equal to the market price of risk times the futures contract beta.
In the preceding section, we discussed three alternative views of the expected return to holding futures
contracts. Under the Keynes/Hicks/Cootner view, speculators earn a positive risk premium. That view is
reflected in Figure 5.3 as a position along the futures pricing line with a positive level of systematic risk. The
positive return is earned from the upward drift in futures prices when speculators are long. An alternative
view is that of Dusak, who argues that futures contracts have no systematic risk. In other words, the beta of a
futures contract is zero. In that case, futures contracts are also on the futures contract pricing line but at the
point where the line crosses the origin and where beta equals zero. Under the Dusak view, there is no upward
drift in futures prices because no compensation for risk is necessary. Under the third view, that of Telser,
speculators in the aggregate earn no risk premium, even though risk exists. The Telser position in Figure 5.3
is an average of the position of professionals and amateurs in the figure. Professionals earn a normal risk
20
In (5.3), the daily settlement feature of futures is ignored. The interest earned or paid on daily settlements ought to be a part of
the profit or loss on the futures position. Usually the interest amount is very small. In addition, we have shown in Chapter 3 that, if
interest rates are known and transaction costs are zero, futures market positions can be adjusted so that expression (5.3) is an
appropriate measure of profit on a futures contract.
premium and are on the futures pricing line. However, gamblers and fools earn negative returns as shown in
the figure. The average of these two points is the Telser position-positive risk but no return.
FIGURE 5.2 The Capital Asset Pricing Model and Alternative Views of the
Returns to Speculators
~
Et ( RF )
Assets
Futures

r Keynes/Hicks/Cootner
Telser Professional
Beta
Dusak Telser
 
 Telser Amateurs
The empirical evidence on risk and return in futures markets is ambiguous and makes it difficult to
distinguish among these three alternative views. Telser and Cootner debated vehemently in the 1960’s as to
the meaning of the data for corn, wheat, and cotton. Cootner maintained that an upward drift in futures prices
was observable, while Telser argued it was not.
In a comprehensive investigation using semi-monthly price data for corn, wheat, and soybean futures
during the period May 1952 through November 1967, Dusak (1973) concludes that the expected futures
returns equal zero. To support her conclusions, she estimates (a) the mean realized futures return and (b) the
systematic risk coefficient for each of the futures contract months of the three underlying commodities. The
systematic risk coefficients are estimated using the OLS regression,
~ ~
RF ,t   F   F ( RM ,t  rt )   t , (5.8)
~
where the proxy for the market return, RM ,t , is the price appreciation on the S&P 500 stock portfolio, and the
proxy for the riskless rate, rt , is the return on a T-bill with fifteen days to maturity. The results are reported
in Table 5.1. In Table 5.1, note first that in only two of the sixteen cases reported is the mean realized return
significantly different from zero (i.e., twice its standard error), and in both of those cases the realized return
is negative. These results are further corroborated by the estimates of the systematic risk coefficients. In only
one of the sixteen cases is the beta of the futures contract significantly different from zero. The lack of
covariation of the futures returns with the market return is also reflected through the low R 2 values reported
in the table.
All in all, the Dusak results appear to support the position that futures prices are unbiased predictors
of expected spot prices, however, the generality of the results is not known. Other investigators have found
different results as far as agricultural and metal futures markets are concerned.21 A broader range of
21
Bodie and Rosansky (1980), for example, analyze futures prices for the period 1949 and 1976 and conclude that futures
contracts on average have a positive return. Unfortunately, the Bodie/Rosansky results are suspect given that they use annual data
(and, hence, very few time-series observations) in their statistical analyses.
underlying commodities and longer time series of daily or weekly prices are among the experimental
improvements necessary to determine which of the competing theories is best supported.
5.4 EQUILIBRIUM OF HEDGERS AND SPECULATORS

The CAPM indicates that speculators could receive a risk premium for holding futures contracts. Obviously,
some other group would have to pay a risk premium, since futures are a zero-sum game. Hedgers may be
willing to pay a risk premium to eliminate the risk of holding the commodity. The situation is more
complicated than that described in Chapter 4 because a hedger would consider not only the
TABLE 5.1 Summary of regression test results from Dusak (1973).

Contract Number of Obs. RF s ( RF ) 22 ̂ F s ( ˆ F ) R2
Wheat:
Jul. 302 -0.00164 0.00126 0.048 0.051 0.003
Mar. 302 0.00060 0.00139 0.098 0.049 0.013
May 302 0.00096 0.00142 0.028 0.051 0.001
Sep. 319 -0.00194 0.00127 0.068 0.051 0.006
Dec. 319 0.00044 0.00134 0.059 0.048 0.005
Corn:
Jul. 301 -0.00158 0.00116 0.038 0.046 0.002
Mar. 301 -0.00381 0.00138 -0.009 0.050 0.000
May 301 -0.00268 0.00120 -0.027 0.048 0.001
Sep. 320 -0.00243 0.00128 0.032 0.048 0.001
Dec. 320 -0.00212 0.00147 0.007 0.047 0.000
Soybeans:
Jan. 287 -0.00025 0.00146 0.019 0.058 0.000
Mar. 287 -0.00029 0.00152 0.100 0.065 0.008
May 287 0.00038 0.00148 0.119 0.068 0.011
Jul. 287 0.00006 0.00158 0.080 0.076 0.004
Sep. 287 -0.00105 0.00157 0.077 0.065 0.005
Nov. 287 -0.00071 0.00137 0.043 0.058 0.002
correlation between the futures price and the price of the underlying commodity but also the relation between
the futures and the hedger’s entire portfolio of all assets.23
Figure 5.4 shows how hedgers and speculators interact to determine a futures price in relation to the
expected spot price and the current spot price. The figure assumes homogeneous expectations on the part of
hedgers and speculators. Hedgers are distinguished from speculators because they have a position in the
underlying commodity. The HH schedule in Figure 5.4 depicts the futures market position that hedgers as a
group would like to hold for alternative futures prices. Note that the HH schedule crosses the vertical axis
below the expected spot price. That means that hedgers would sell futures at prices below the expected spot
price because of the attractiveness of risk transfer. The position of the HH schedule depends on the nature
and size of the underlying commitment. The slope of the schedule depends on the amount of price risk of the
underlying commodity and the degree of risk aversion of hedgers.
22
s () is the estimated standard error of the regression coefficient.
23
See Stoll (1979) for a more detailed discussion of this point.
FIGURE 5.4 Equilibrium Futures Prices

S Futures Price, F0
~
H E0 ( S T )
S
F0*
Short H
Futures
Position
Long
Speculators Short Speculators Long
Position Position
The SS schedule depicts the futures market positions that speculators would accept for alternative futures
prices. The SS schedule is downward sloping and intersects the vertical axis at the expected spot price. When
the futures price equals the expected spot price, the speculator has no incentive to take a futures position
either long or short. When the futures price falls below the expected spot price, speculators earn a risk
premium by taking a long position; and when the futures price is above the expected spot price, speculators
earn a risk premium by taking a short position. The downward sloping SS schedule implies that a larger risk
premium is required to induce speculators to take a larger position.24
In Figure 5.4, the equilibrium futures price, F0* , is determined such that the short position taken by hedgers
~
equals the long position taken by speculators. Speculators expect to receive a risk premium of E0 ( ST )  F0* ,
and hedgers expect to pay that risk premium. Hedgers hold real assets (like wheat or common stocks) and
sell futures to avoid risk. Speculators accept the risk; and, in return, earn a risk premium. Figure 5.4 is
consistent with the Keynes/Hicks/Cootner view and a capital asset pricing model in which the underlying
commodity has systematic risk.
Under the Telser and Dusak views of speculators, the SS schedule would be perfectly horizontal and would
~
cross the vertical axis at E0 ( ST ) . In such a case, hedgers would receive insurance at no cost, and speculators
would, as a group, not earn a risk premium.
It is possible that the risk premium is time-varying, particularly in agricultural commodities, which have a
seasonal harvest cycle. In the case of a commodity like wheat, for example, hedgers might be long wheat and
short wheat futures in the fall after the harvest has come in, and they might be short wheat and long wheat
futures in the spring when handlers of wheat make commitments to deliver wheat that they do not yet have in
hand. In terms of Figure 5.4, such a seasonal pattern would imply an HH schedule below the SS schedule in
the fall and an HH schedule above the SS schedule in the spring. In the fall, speculators are long futures and
~ ~
F0  E0 ( ST ); and in the spring, speculators are short futures and F0  E0 ( ST ). Futures prices would display
normal backwardation in the fall when speculators indirectly bear the risk of the long positions in the
commodity that has been harvested. In the spring, futures prices would display contango when speculators
indirectly bear the risk of the short positions in the underlying commodity assumed by hedgers.
24
Such an increase in risk premium as a function of position size is not modeled in the standard one-period capital asset pricing
model.
5.5 SUMMARY
In this chapter, the role of speculators and the returns speculators can expect in an efficient market are
discussed. Some researchers argue that futures prices are unbiased estimates of subsequent spot prices, which
means that futures prices do not trend up or down over time. In that case, speculators as a group do not make
profits. The absence of speculative profits is possible if amateur speculators lose to professional speculators
so that speculators as a group do not make profits. The absence of profits is also possible if the risk of
holding futures contracts is fully diversifiable. If the risk can be diversified away, no risk premium is
required to induce speculators to hold futures contracts.
On the other hand, some researchers argue that futures prices trend up (normal backwardation) or
trend down (contango). If futures prices have a trend, speculators as a group make profits commensurate
with the level of risk that they assume from hedgers.
This chapter also derives the expected return to speculators under the Sharpe-Lintner capital asset
pricing model. Since futures contracts require no investment, the expected return of a futures contract equals
only the futures contract’s market risk premium (no riskless return is earned). The last section of the chapter
contains a model in which hedgers and speculators interact and determine an equilibrium futures price.
6 CONTRATOS DE FUTUROS SOBRE MERCANCÍAS FÍSICAS
6.1 CONTRATOS DE FUTUROS SOBRE MERCANCÍAS FÍSICAS
6.2 ¿POR QUÉ SURGEN LOS MERCADOS DE FUTUROS EN MERCANCÍAS FÍSICAS?

 Incertidumbre
 Almacenamiento y Entrega
6.3 INVENTARIO Y PATRONES DE PRECIO

 Mercancías en Oferta Estacional
 Mercancías No-Estacionales
6.4 CUBRIÉNDOSE DEL RIESGO DEL PRECIO DE LA MERCANCÍA

 Cobertura en Corto del Almacenador
 Cobertura Larga del Comerciante
 Cobertura del Productor
 Cobertura Corta del Productor
 Cobertura Larga del Productor
 Cobertura Natural Versus Cobertura del Mercado de Futuros
 Calculando el Ratio de Cobertura
6.5 OFERTA Y DEMANDA DE ALMACENAMIENTO
6.6 RETORNOS PARA LOS ESPECULADORES
6.7 VÍNCULOS DE PRECIOS INTERNACIONALES: LA LEY DE UN SOLO PRECIO
6.8 MONOPOLIOS Y ACOSO CORTO (CORNERS AND SHORT SQUEEZES)
6.9 RESUMEN
6 CONTRATOS DE FUTUROS SOBRE MERCANCÍAS FÍSICAS

Los mercados de futuros organizados se iniciaron en respuesta a los riesgos asociados con la
comercialización de mercancías estacionales como el trigo y el maíz. Los agricultores tomaban contratos
forward con usuarios para deshacerse de sus cosechas, y los mercados de futuros surgieron para proporcionar
a los usuarios y agricultores un instrumento financiero para proteger sus riesgos mientras se negociaban los
contratos forward. Los principales mercados de futuros en Estados Unidos están en Chicago porque, por en
ese momento, Chicago tenia una localización ideal para los silos de cereales y el punto de transporte para los
embarques al Este. Hoy en día, los contratos de futuros no se negocian solamente para productos agrícolas
estacionales, sino también para mercancías reales como metales y varios materiales industriales. La mayor
expansión en la negociación de futuros se dio en los años 70 y 80 con el surgimiento de los futuros
financieros.
Este capítulo se enfoca en las mercancías físicas. Comenzamos nuestra discusión sobre por qué
surgen los mercados de futuros para ciertas mercancías físicas y no en otras. Luego nos enfocamos en los
patrones estacionales de inventarios y precios, la cobertura que realizan los productores y abastecedores, la
conducta de la base, los retornos para los especuladores y otros temas.
6.1 CONTRATOS DE FUTUROS SOBRE MERCANCÍAS FÍSICAS

Los contratos de futuros negociados más activamente en mercancías físicas se enumeran en la Tabla 6.1 por
categorías: cereales y semillas oleaginosas, ganado, alimentos y fibra, metales y petróleo. Los contratos de
futuros más antiguos son los de cereales, que se comenzaron a negociar en 1859. Los contratos de ganado
datan de los años 60. Las más recientes adiciones a la lista son los productos del petróleo. El petróleo crudo y
refinado se comenzaron a negociar en los años 70 y los futuros de gasolina se iniciaron en 1981.
TABLA 6.1 Especificaciones de los Contratos de Futuros para Mercancías Físicas

(contratos más activos en mercados estadounidenses)
Mercancía Horas de Meses de los Unidades/
Último Día de Negociación
(Bolsa de Valores) Negociación Contratos25 Cambio Mínimo en Precio
Cereales y Semillas Oleaginosas
7 días antes del último día
Maíz (CBT) 9:30-1:15 (CST) 1,2,3,5,7,9 5,000 toneladas/ ¼ ($12.50)
útil del mes
Avena (CBT) 9:30-1:15 (CST) 1,2,3,5,7,9 5,000 toneladas/ ¼($12.50) Igual al anterior
Soya (CBT) 9:30-1:15 (CST) 9,11,1,3,5,7,8 5,000 toneladas/ ¼($12.50) Igual al anterior
Harina de Soya 9:30-1:15 (CST) 1,3,5,7,8,9,10,12 100 toneladas 10 ($10) Igual al anterior
Aceite de Soya (CBT) 9:30-1:15 (CST) 1,3,5,7,8,9, 10, 12 60,000 libras $0.0001($6) Igual al anterior
Trigo (CBT) 9:30-1:15 (CST) 7,9,12,3,5 5,000 toneladas/ ¼ ($12.50) Igual al anterior
Ganado
40,000 libras 20avo día calendario del
Ganado vivo (CME) 9:05-1:00 (CST) 2,4,6,8,9,10,12
$0.00025($10) mes del contrato
30,00 libras 20avo día calendario del
Cerdo (CME) 9:05-1:00 (CST) 2,4,6,7,8,10,12
$0.00025($7.50) mes del contrato
Día útil antes de los últimos
Lomo de Cerdo 40,000 libras
9:05-1:00 (CST) 2,3,5,7,8 5 días útiles del mes del
(CME) $0.00025($10)
contrato
25
La notación en esta columna corresponde con el mes del año calendario (por ejemplo: 1 es enero, 2 es febrero y así
sucesivamente).
TABLA 6.1 Cont. Especificaciones de los Contratos de Futuros para Mercancías Físicas
Mercancía Meses de los Unidades/
Horas de Negociación 26 Último Día de Negociación
(Bolsa de Valores) Contratos Cambio Mínimo en Precio
Alimentos y Fibras
10 toneladas métricas
Cacao (CSCE) 9:30-2:15 (EST) 12,3,5,7,9
$1(10)
37,500 libras
Café (CSCE) 9:15-1:58 (EST) 3,5,7,9,12
$0.0001($11.20)
Azúcar 112,000 libras Última día útil del mes
10:00-1.43(EST) 1,3,5,7,10
(mundial)(CSCE) $0.0001($11.20) anterior al mes de la entrega
Actual + 17 50,000 libras
Algodón (CTN) 10:30-3:00 (EST)
sucesivos $0.0001($5)
Jugo de Naranja 15,000 libras
10:15-2:45 (EST) 1,3,5,7,9,11
(CTN) $0.0005($7.50)
Ganada vacuno vivo 40,000 libras 20avo día calendario del mes
9:05-1:00 (CST) 2,4,6,8,9,10,12
(CME) $0.00025($10) del contrato
Del tercer al último día útil
9:25-2:00 1,3,5,7,9,12,1,3,5,7 25,000 libras
Cobre (CMX) del mes en que vence la
(EST) ,9,12 + act +3 $0.0005($12.50)
entrega
8:20-2:30 2,4,6,8,10,12, act 100 onzas troy
Oro (CMX)
(EST) +2 10 centavos($10)
8:20-2:30 50 onzas troy
Platino (NYM) 1,4,7,10, incl. act 3
(EST) 10 centavos ($5)
8:25-2:25 1,3,5,7,9,12 5,000 onzas
Plata (NYM)
(EST) act + 2 1/10 centavos ($5)
Petróleo
18 meses
Petróleo Crudo 9:45-3:10 consecutivos 1,000 barriles
(NYM) (EST) comenzando el 1 centavo ($10)
mes actual
15 meses
Petróleo Refinado 9:50-3:10 consecutivos 42,000 US galones
(NYM) (EST) comenzando el $0.0001
mes actual
15 meses
Gasolina 9:50-3.10 consecutivos 42,000 US galones
(NYM) (EST) comenzando el $0.0001
mes actual
Los contratos de futuros sobre mercancías físicas a ser entregadas a la opción del corto en algún momento en
el mes de la entrega. El primer día en que se puede realizar la entrega es el primer día de notificación. En
muchas mercancías, la posición larga que recibe una notificación de entrega tiene la oportunidad de vender
sus contratos de futuros y re-entregar la notificación; pero una vez que se termina la negociación de los
futuros, todas las posiciones largas circulantes no tienen otra opción que aceptar entrega. En la práctica, sólo
una pequeña parte de los contratos futuros ingresados alguna vez resultan en entrega.
La actividad de negociación en varios contratos de futuros varía según cambian las necesidades del
cliente y conforme surgen contratos competitivos. A fines de 1989, la CFTC aprobó 121 contratos de futuros
de mercancías físicas27. Algunos de estos nunca tuvieron éxito. Algunos que antes fueron exitosos ahora son
dominantes, Otros se negocian pero son muy inactivos. Ahora veremos los factores que dan origen a los
mercados de futuros de mercancías físicas y que determinan su éxito o su fracaso.
26
La notación en esta columna corresponde con el mes del año calendario (por ejemplo: 1 es enero, 2 es febrero y así
sucesivamente).
27
Informe 1989 de la Comisión de Negociación de Futuros de Mercancías.
6.2 ¿POR QUÉ SURGEN LOS MERCADOS DE FUTUROS EN MERCANCÍAS

FÍSICAS?
Incertidumbre
Dado que el propósito principal de los mercados de futuros es cubrir los riesgos, los mercados de futuros no
surgen si el precio de la mercancía no es incierto. Si el sostén del precio agrícola determina el precio del
trigo, no surgirán mercados de futuros de trigo. Los futuros de monedas no existirían en un sistema de tasas
de cambio fijos. La negociación de los contratos de futuros de café muere cuando el cartel internacional del
café “estabiliza” el precio del café en un nivel fijo. La incertidumbre en los precios se origina por la
incertidumbre sobre la oferta de mercancías y la incertidumbre sobre la demanda de las mercancías. La
incertidumbre sobre la cantidad relativa en el lado de la oferta y el lado de la demanda varía según el tipo de
mercancía.
Aún cuando las mercancías agrícolas producidas más estacionariamente ha aumentado durante cierta
parte del año en todo el mundo, la oferta es incierta debido a que el tamaño de la cosecha se ve afectado por
condiciones climatológicas. Por otro lado, la demanda global de la mayoría de los alimentos agrícolas y
petróleos es razonablemente estable dado que los patrones de consumo no cambiaron drásticamente de un
período a otro. Sin embargo, la incertidumbre de la demanda surge incluso después que un cultivo ha sido
cosechado porque la oferta de una mercancía sustituta puede ser incierta. Una cosecha adicional de maíz, por
ejemplo, puede afectar adversamente el precio del trigo.
Las mercancías en producción continua –como el petróleo, gas, madera y ciertos metales– enfrentan
incertidumbre tanto en el lado de la oferta como de la demanda. Las huelgas y los incrementos inesperados
en los costos afectan la oferta. Al mismo tiempo, la incertidumbre en la demanda es mayor que en las
mercancías agrícolas porque las mercancías de producción continua tienden a ser materiales industriales que
están sujetos al ciclo comercial.
Algunas mercancías físicas, específicamente oro y plata, tienen una oferta casi fija en el sentido que
las existencias circulantes de estas mercancías son mayores con relación a la producción anual. Como
resultado, la incertidumbre en el precio surge principalmente en el lado de la demanda. El oro es como
muchos instrumentos financieros que también tienen una oferta casi fija. Los precios del oro y los
instrumentos financieros dependen de las tasas de interés, inflación y otros factores macroeconómicos.
Mercado Grande y Competitivo

Los mercados de futuros no tienen éxito si el mercado para la mercancía subyacente es pequeño, porque
existe una negociación de futuros insuficiente para mantener la liquidez del mercado. Es más, Telser y
Hoggonbotham (1977, p.998) señalan que “un mercado de futuros organizado facilita la negociación entre
extraños”. En los mercados pequeños, los productores y los usuarios de la mercancía encuentran preferible
tratar directamente entre ellos en lugar de incurrir en el gasto de establecer un mercado de futuros.
Un mercado de futuros líquidos surge sólo si el mercado de la mercancía subyacente es uno en el cual
está disponible un gran número de unidades de una mercancía estándar. El mercado automotor es grande,
pero las unidades no son estándar. Los mercados de futuros de mercancías surgen sólo para las mercancías
que pueden ser estandarizadas y por lo tanto se negocian fácilmente. Para algunas mercancías, como el oro,
es fácil lograr la estandarización. Para otros, la estandarización es más difícil. Por ejemplo, la estandarización
de ganado vacuno vivo (negociado en CME) requiere de una especificación compleja del contrato. Inclusive
las mercancías como el maíz y el trigo tienen un rango considerable de grados. Como vimos en el Capítulo 2,
los contratos de futuros son diseñados para permitir la entrega de una variedad de grados en una variedad de
puntos de entrega de manera que el peligro de un interesado es limitado, pero el ampliar la definición de la
mercancía de esta manera reduce su estandarización. El éxito del mercado depende del grado en el que estén
correlacionados los diversos grados de la mercancía.
Un mercado es grande si los precios de un gran número de unidades están estrechamente relacionadas
con el precio del contrato de futuros. Es poco probable que los contratos de futuros tengan éxito en mercados
no-competitivos cuando la producción de la mercancía subyacente es monopolizada o cuando son pocos los
compradores de la mercancía. En dichos mercados, el peligro es tan grande que el precio cash puede ser
manipulado para producir ganancias artificiales sobre el contrato de futuros (como a un interesado recién
llegado).
Al mismo tiempo, los mercados de futuros pueden mejorar la competencia en un mercado que no es
completamente competitivo. Por ejemplo, la negociación de los futuros de petróleo crudo le da a los usuarios
la oportunidad de asegurar el precio del futuro de petróleo sin negociar contratos a largo plazo con los
productores. El mercado de futuros proporciona una alternativa para negociar directamente con el productor.
Por esta razón, los productores con frecuencia se oponen a la introducción de los mercados de futuros. El
ejemplo más famoso de oposición de los productores a la introducción de mercados futuros es la oposición
de los agricultores de cebollas a la introducción de los futuros de cebollas, lo cual resultó en una prohibición
del congreso a la negociación de los futuros de cebollas en 1958.
Almacenamiento y Entrega
Las mercancías físicas sobre las cuales se basan los contratos de futuros pueden ser almacenadas directa o
indirectamente. El almacén usualmente se considera una necesidad para un contrato de futuros exitoso al
basarse en las opciones de compra del contrato para la entrega de la mercancía en una fecha posterior, y la
entrega se puede realizar solamente si la mercancía puede ser mantenida hasta la fecha de la entrega. Sin
embargo, si se puede producir la mercancía para la entrega, el almacén no sería necesario. Por lo tanto, un
contrato de futuros de huevos frescos existe, sin embargo, claramente los huevos no se pueden almacenar en
el sentido usual. La entrega no es el problema porque la disponibilidad futura de los huevos se puede
asegurar al tener las gallinas. Los huevos son almacenados indirectamente, como si estuviesen en las
gallinas. Similarmente, el ganado vacuno vivo no se puede almacenar en el sentido usual; no obstante,
claramente se puede almacenar (y alimentar) para su entrega posterior. Por lo tanto, en este sentido más
amplio, el requerimiento de almacenamiento y entrega se satisfacen para todas las mercancías físicas.
6.3 INVENTARIO Y PATRONES DE PRECIO
Mercancías en Oferta Estacional

El patrón de inventarios de las mercancías en oferta estacional puede representarse por el patrón de diente de
serrucho en la Figura 6.1. Los puntos altos reflejan las cosechas cuando los inventarios son reabastecidos, y
los puntos bajos representan el hecho que el cultivo anterior se acaba justo antes de la cosecha. La línea con
pendiente descendente representa un consumo gradual de la mercancía de nuestro inventario. Por supuesto,
el patrón simple y regular de la Figura 6.1 no es realista por un número de razones. Primero, las cosechas no
se dan todas a la vez sino a lo largo de un período de tiempo. Por ejemplo, la cosecha de trigo de Estados
Unidos se inicia en mayo en el Sudeste, cuando se cosecha el trigo de invierno plantado en el otoño anterior,
y continua en los estados del norte hasta setiembre, cuando se cosecha el trigo de primavera plantado a
comienzos de primavera. Los inventarios pico en Estados Unidos ocurren usualmente en setiembre.
La cosecha gradual nivela los picos y los valles en la figura. Segundo, las cantidades cosechadas no son las
mismas cada año como lo implica la regularidad del patrón en la Figura 6.1. Algunos años son mejores que
otros, lo cual da como resultado que algunos picos sean mayores que otros. En algunos años, la cosecha
nueva es tan reducida que la cosecha antigua es llevada hasta el año siguiente.
En la Figura 6.2 se muestra el patrón del precio spot, S , que corresponde al patrón de inventario.
Como se podría esperar, simplemente los precios son bajos cuando el inventario es alto y los precios altos
cuando el inventario es bajo. Los precios al contado reales no siguen este simple estereotipo dado que las
cosechas ocurren a lo largo de un período de tiempo. Los picos y las depresiones en los precios spot se
atenuarán así como se atenúan los picos y las depresiones en el inventario.
FIGURA 6.1 Patrón de Inventario Estacional en

Mercancías Agrícolas
Inventario
Tiempo
En el momento antes de la cosecha, como t en la Figura 6.2, normalmente existe un precio de futuros
para un contrato que vence en el momento H  1 , justo antes de la cosecha y para un contrato que vence en
un momento H  1 , justo después de la cosecha. Estos precios de futuros son pronósticos de los precios al
contado en los respectivos vencimientos de los contratos de futuros. La base en el momento t para el
contrato de futuros que vence antes del final de la cosecha del año pasado, Ft ( H  1)  S t , representa un
mercado de cargo de acarreo y normalmente es positivo para reflejar los costos de almacenamiento. La base
en el momento t para el contrato de futuros que vence inmediatamente después de la cosecha,
Ft ( H  1)  S t , representa un mercado invertido porque se espera que la cosecha nueva disminuya los
precios por debajo del nivel en el momento t .
Las figuras son útiles para identificar las tres principales fuentes de riesgo en mercancías estacionales.
Para un agricultor individual, el riesgo más importante es el riesgo de cantidad, esto es, el riesgo que se
relaciona con el tamaño de la cosecha en la cosecha estacional. Dado el clima y otros factores, el agricultor
no sabe la cantidad de cultivo que será cosechado.
El segundo riesgo es el riesgo de precio, el cual no sólo está presenta en el punto de cosecha sino
también durante el resto del año. Alrededor de la cosecha, la variabilidad de los precios refleja incertidumbre
sobre la cosecha agregada. El riesgo de precio y el riesgo de cantidad están relacionados. Si la cosecha
agregada es deficiente, el precio será mayor de lo normal. Si la cosecha agregada es buena, el precio será
menor que el normal. El agricultor cuya cosecha es representativa del conjunto agregado encontrará que el
precio y la cantidad están negativamente correlacionados –una cosecha deficiente se asocia con precios más
altos de lo normal, y una cosecha buena se asocia con precios menores de lo normal. Esta asociación
negativa mitiga colectivamente los riesgos de precio y cantidad. La cosecha de cada agricultor, sin embargo,
no está perfectamente correlacionada con la cosecha agregada dado que las condiciones climatológicas en las
diferentes partes del país varían. El agricultor que tiene una cosecha deficiente cuando otros agricultores han
tenido una buena cosecha estará particularmente mal porque el precio de la cosecha será bajo y al mismo
tiempo tendrá poco para vender. Correspondientemente, un agricultor con una cosecha excelente obtiene
grandes utilidades cuando otros agricultores tienen una mala cosecha. Después de la cosecha, el riesgo de
precio aún se mantiene porque la demanda por la mercancía es incierta. La demanda por la mercancía es
incierta porque la oferta de sustitutos es incierta y porque la demanda del consumidor final puede fluctuar.
FIGURA 6.2 Patrón de Precio Estacional en Mercancías

Agrícolas
Precio
Ft(H - 1)
St
Ft(H + 1)
H
H
Tiempo
t H-1 H+1
Una vez que ya se ha cosechado, se saca el cultivo del almacén. Como ya vimos, la mayor parte del
riesgo durante esta parte del ciclo viene del lado de la demanda. Sin embargo, también esta presenta otra
fuente de riesgo, el riesgo del almacenamiento. Los costos de almacenamiento incluyen alquiler de almacén,
el costo del interés de los fondos destinados a la mercancía, seguro, mano de obra y manipuleo, y
desperdicio. Las fluctuaciones en estos costos afectan la rentabilidad del almacenamiento y el precio de la
mercancía en el período entre cosechas.
Se pueden utilizar los mercados de futuros para cubrir el riesgo del precio de la mercancía, pero los
mercados de futuros son menos adecuados para cubrir el riesgo de cantidad o el riesgo del costo de
almacenamiento. La discusión de la cobertura a continuación, por lo tanto, enfatiza el riesgo de precio de la
mercancía.
Mercancías No-Estacionales
Las mercancías no-agrícolas como los metales y petróleo son producidas de manera continua. Para todas
excepto los metales preciosos, los inventarios son relativamente pequeños con relación al consumo y no
fluctúan estacionalmente como los inventarios de mercancías agrícolas. Sin embargo, los inventarios pueden
agotarse rápidamente si se interrumpe la producción. Por ejemplo, una huelga en una mina de cobre puede
detener la producción de cobre y causar que los inventarios de cobre se agoten. Similarmente, las
restricciones de la OPEP sobre la producción de petróleo crudo pueden agotar los inventarios de petróleo. En
esos casos, el precio spot puede verse drásticamente afectado. En el caso de los metales preciosos como el
oro y la plata, los inventarios son grandes con relación a la producción, y los precios son determinados
principalmente por la demanda. La interrupción de la producción de metales preciosos toma más tiempo en
afectar los inventarios.
El patrón de precios de los recursos naturales fue analizado primero por Hotelling (1931), y más
recientemente, por Miller y Upton (1985). En un mundo de certeza, el Principio Hotelling establece que el
margen de utilidad de la minería de un recurso natural, el precio del recurso extraído, ST , y los costos de
producción marginales netos por unidad, CT , –se incrementan a la tasa de interés, r * :
( ST  CT )  ( S 0  C0 )(1  r * ) (6.1)
donde T se refiere al período futuro y 0 se refiere al período actual.28 Intuitivamente, la idea es que el
margen de utilidad debe incrementarse para cubrir los costos de los fondos dedicados a las reservas del
recurso natural. Si el margen de utilidad en el período futuro, T , es menor que la cantidad especificada por
(6.1), los productores incrementarán la producción en el momento 0 en lugar de esperar a producir en T . El
margen de utilidad podría invertirse a r * para obtener más en T que si la producción hubiese sido demorada
hasta mañana. La acción disminuye S 0 y disminuye el margen de utilidad en el momento 0 hasta que se
satisfaga la condición (6.1). A la inversa, si el margen de utilidad en T es mayor que la cantidad especificada
en (6.1), el valor actual de lo que el productor ganaría mañana excedería el margen de utilidad de hoy. El
productor estaría mejor restringiendo la producción hoy y produciendo más mañana. Podría pedir prestado
contra la utilidad de mañana para dar una cantidad que exceda el margen de utilidad en el momento 0. El
proceso de producir más mañana disminuye ST hasta que se satisfaga la condición (6.1). En un mundo con
incertidumbre, se mantiene una versión modificada del Principio Hotelling en la que se utilizan los valores
~ ~
esperados de ST y CT y la tasa de retorno esperada ajustada al riesgo es r * .
~ ~
Si existe un mercado de futuros, (6.1) se mantiene con FT sustituyendo por ST . Con esa sustitución y
cierta manipulación, (6.1) se puede escribir de la siguiente manera
~ ~
FT  S 0 (1  r * )  CT  C0 (1  r * ). (6.2)
Si no existen costos de producción, ya sea en el presente o en el futuro, el precio de los futuros está por
encima del precio spot por el costo del interés de los fondos destinados a la mercancía. Esto es lo mismo que
el modelo del costo-de-acarreo discutido en el Capítulo 3. El modelo de costo-de-acarreo dice que el precio
de los futuros de un contrato que vence en T excede el precio spot por el costo de mantener la mercancía. En
~
la ausencia de costos de almacenamiento aparte del costo del interés, se tiene que FT  S 0 (1  r * ) . Este
equilibrio simple es apropiado para una mercancía como el oro que existe en grandes cantidades en forma
refinada.
La Ecuación (6.2) también muestra por qué el precio de los futuros de recursos naturales puede estar
temporalmente por debajo del precio spot. Esto puede suceder si el costo de producción actual, C0 , es
anormalmente alto, como en el caso de una huelga o una interrupción del suministro. En la crisis de Kuwait
de 1990-91, el precio spot del petróleo se elevó drásticamente con relación al precio de los futuros,
~
reflejando una elevación en C0 con relación a CT . A lo largo del tiempo, conforme los costos de producción
regresaron a la normalidad, el precio de los futuros de petróleo regresó a su prima normal sobre el precio
cash del petróleo.
6.4 CUBRIÉNDO EL RIESGO DEL PRECIO DE LA MERCANCÍA
Cobertura en Corto del Almacenador

Con frecuencia las mercancías son almacenadas para su reventa a los usuarios. La decisión de cobertura del
que almacena fue analizada en el Capítulo 4, y el ejemplo del Capítulo 4 se repite en la Tabla 6.2. El 1º de
setiembre, el comerciante decidió almacenar cereal durante tres meses y vender futuros para diciembre
porque la base de tres meses de nueve centavos por tonelada cubre los costos de almacenamiento de tres
centavos al mes. El éxito de la cobertura depende en parte de si se puede elegir la mercancía para entregarse
a la presentación de contratos de futuros. Asuman primero que la mercancía es entregable. Si se mantiene la
posición hasta el vencimiento, el comerciante puede simplemente entregar el cereal contra el contrato de
futuros y obtener una ganancia de nueve centavos que cubre los costos de almacenamiento. El único riesgo
28
Una vez más, adoptamos el acuerdo de utilizar un asterisco para denotar una tasa aplicada a lo largo de la vida del contrato de
futuros o el período de cobertura.
es que los costos de almacenamiento resulten mayores de lo esperado. Por ejemplo, si los costos de
almacenamiento resultaron ser 3.5 centavos al mes, el comerciante pierde 1.5 centavos al comprar la
mercancía a $3.00 el 1º de setiembre y entregándola contra el contrato de futuros a $3.09 el 1º de diciembre.
La única manera de cubrirse contra este riesgo es realizar un contrato forward por los costos de
almacenamiento, algo que no siempre es posible. Si todos los costos de almacenamiento son asegurados de
antemano, el comerciante está en posición de tener un retorno libre de riesgo (asumiendo que no existe el
riesgo de incumplimiento). En la práctica, es poco probable que los comerciantes aseguren todos los costos.
La Tabla 6.2 analiza el resultado de la cobertura del comerciante si la mercancía es vendida a un cliente el 1º
de noviembre, un mes antes del vencimiento del contrato de futuros. El precio cash de la mercancía el 1º de
noviembre se asume en $2.70, una disminución de 30 centavos desde el 1º de setiembre. Al vender futuros,
el comerciante ha eliminado el riesgo de tal movimiento adverso en el precio de la mercancía. Sin embargo,
se mantiene el riesgo base. Como vimos en el Capítulo 4, el riesgo base es el mismo que el riesgo de costo de
almacenamiento si la mercancía es entregable. El efecto sobre el riesgo base se muestra al asumir tres precios
alternativos de futuros el 1º de noviembre –$2.73, $2.75 y $2.71– cada uno implicando una base diferente. Se
da un incremento en la base (debilitamiento) de cinco centavos cuando el precio de los futuros cae a $2.75 en
lugar de $2.73. Esto produce una pérdida neta de dos centavos. El comerciante tiene la opción de volver a
almacenar el 1º de noviembre para obtener una base de cinco centavos por mes, pero sólo es rentable si los
costos de almacenamiento son asegurados en un nivel menor. Si los costos de almacenamiento no son
asegurados, el incremento en la base refleja un incremento en todo el mercado de los costos de
almacenamiento, lo cual probablemente afectará al comerciante analizado en la Tabla 6.2. Por otro lado, una
reducción (reforzamiento) de la base a un centavo da como resultado una ganancia neta de dos centavos para
el comerciante, asumiendo que no tiene la obligación de pagar costos de almacenamiento para el mes de
noviembre.
Asuman ahora que la mercancía no se puede entregar a la presentación de un contrato de futuros. En
este caso, el riesgo base refleja el riesgo del precio de la mercancía así como el riesgo del costo de
almacenamiento. En la ausencia de entrega, los precios cash y los precios de los futuros no necesitan
converger al vencimiento. Por ejemplo, el comerciante de trigo en Oklahoma no puede entregar su trigo a un
costo razonable a la presentación del contrato de futuros en Chicago. Como resultado el precio del trigo en
Oklahoma cae con relación al precio del contrato de futuros, aunque la caída diferencial no puede ser mayor
que los costos de transporte. Por lo tanto, una cobertura en corto no es completamente efectiva. Mientras más
distante esté la mercancía cash en grado y espacio, mayor será la posibilidad que el precio cash y el precio de
los futuros se muevan en diferentes direcciones entre el tiempo de la cobertura y el vencimiento del contrato
de futuros. Como vimos en el Capítulo 4, la efectividad de la cobertura dependerá del grado de correlación
entre el precio cash y el precio de los futuros.
Tabla 6.2 Utilidades en una Cobertura Corta

Mercado al Contado Futuros a Diciembre
Fecha Transacción Precio Transacción Precios alternativos
1º set Comprar tonelada a 3.00 Vender futuros a 3.09 3.09 3.09
1º nov Vender tonelada a 2.70 Comprar futuros a 2.73 2.75 2.71
Ganancia -0.30 0.36 0.34 0.38
Ganancia neta 0.0 0.04 0.08
Ganancia neta menos costos de almacenamiento de $0.03 al mes 0.0 -0.02 0.02
Cobertura Larga del Comerciante

Una cobertura larga involucra la compra de futuros para cubrirse contra un incremento en el precio de una
mercancía. Una situación típica es la de un exportador que el 1º de setiembre ingresa en un contrato forward
para vender 500,000 toneladas de maíz para entregarse en Nueva Orleans en tres meses. El exportador no
posee el maíz y hace los arreglos para adquirirlo y embarcarlo a Nueva Orleans. Para cubrirse contra los
incrementos en el precio del maíz mientras que realiza estos arreglos, el exportador compra 100 contratos de
futuros para diciembre. Si el precio del maíz sube, la ganancia del contrato de futuros compensa la pérdida
en el contrato de exportación. Conforme se adquiere el grado de maíz especificado y es embarcado a Nueva
Orleans, el exportador cierra su posición de futuros.
Cobertura del Productor

Un productor transforma un input, como trigo, en output, como harina, que es vendido a sus clientes. El
productor mantiene un inventario de materias primas (trigo) y un inventario de productos terminados
(harina). También debe adquirir materias primas y comercializar el producto final a los clientes. Para el
productor, la cobertura óptima en el mercado de futuros depende de si desea cubrirse contra el riesgo de
inventario; como el comerciante que almacena, o si quiere cubrirse contra un incremento en el costo de la
mercancía, como el comerciante; o ambas cosas.
Cobertura Corta del Productor

El productor está en la misma posición que el comerciante que almacena con relación al inventario de
productos terminados y materias primas, si el precio del producto final y el precio de la mercancía están
positivamente correlacionados. Debido a la correlación positiva, una caída en el precio del producto final es
como una caída en el precio de la mercancía, dando como resultado una pérdida con relación al costo de
ambos inventarios. Una posición corta en el mercado de futuros cubre del riesgo de inventario de este tipo, y
la efectividad para cubrir depende del riesgo base para el inventario de materias primas o producto
terminado. Si el precio del producto final no tiene correlación con el precio de los futuros, el productor
todavía puede utilizar una cobertura de futuros corta para limitar el riesgo del inventario de materias primas,
pero el mercado de futuros no proporciona ninguna cobertura contra cambios en el precio del producto final.
A diferencia de los que almacenan, los productores pueden estar dispuestos a mantener la mercancía
inclusive si la base es cero o negativa. Por ejemplo, supongamos que el 1º de febrero, el precio de los futuros
para mayo y el precio spot del trigo son ambos $3.00, mientras que los costos marginales del
almacenamiento son nueve centavos. Una operador de un silo de cereales no almacenará trigo porque el
ingreso por almacenamiento es cero, mientras que los costos son positivos. Sin embargo, el productor puede
almacenar el trigo porque la conveniencia de tener el trigo y mantener la producción compensa de más el
hecho de haber asegurado una pérdida neta. La cantidad por la cual los costos marginales de almacenamiento
exceden la base se llama rendimiento de conveniencia –nueve centavos en el ejemplo anterior. El productor
puede elegir asegurar una pérdida neta de nueve centavos para cubrirse contra la posibilidad de una pérdida
aún mayor si el precio de la mercancía cayera por debajo de $3.00.
Cobertura Larga del Productor

Un productor ingresa en una cobertura larga para poder asegurar el precio de los suministros futuros de una
mercancía. Por ejemplo, un molinero de harina puede desear cubrirse contra incrementos en el costo del
trigo. Una cobertura larga es óptima si el productor ha negociado un precio por el producto final y desea fijar
los costos.
Sin embargo, si el precio del producto final es incierto, no es una cobertura para asegurar un precio
fijo para los inputs. Una cobertura previsora no es una cobertura si el precio del producto final no es incierto.
Una cobertura previsora también puede ser un error porque el precio de la mercancía puede ser aún más bajo
más adelante. Por otro lado, un agente de compras conocedor puede ser capaz de predecir el costo futuro
probable de la mercancía.
Cobertura Natural Versus Cobertura del Mercado de Futuros

Los productores incurren en costos por los inputs y reciben ingresos por los outputs en la venta del producto
final. Si el costo de los inputs está correlacionado con el precio del producto final, existe una cobertura
natural. Por ejemplo, un molinero de harina puede encontrar que los incrementos en el precio del trigo
generalmente van acompañados por incrementos en el precio de la harina. Si ese fuera el caso, se mantiene el
margen de utilidad y la cobertura no reduce el riesgo. Por ejemplo, si el molinero asegura el precio del trigo
con un contrato de futuros y los precios del trigo y la harina caen, él sufre una pérdida debido a la
disminución en el precio del output mientras que el precio del input está fijo en el nivel original más alto
dado por la cobertura. Si él no hubiese protegido el costo del input, el trigo, estaría mejor porque la
disminución en los precios de la harina sería compensada por la disminución en el precio del input. Por otro
lado, si se elevan los precios del trigo y la harina, la cobertura del precio del trigo produce una ganancia
global porque el precio del output se eleva mientras que el precio del input está fijo por la cobertura. Este
ejemplo, el cual asume que los precios del input y el output están correlacionados, se resume en la Tabla 6.3.
6.3. Efecto de Cubrir las Utilidades si el Precio del

Input y el Output están Correlacionados
Cobertura No Cobertura
Precios caen Pérdida Ningún Efecto
Precios suben Ganancia Ningún Efecto
Es evidente que en este caso la “cobertura” –por la cual queremos decir: asegurar el precio del input– en
realidad incrementa la variabilidad de las utilidades. Sin embargo, si los precios del input y el output no están
correlacionados, la cobertura tendería a reducir el riesgo.
Muchos productores enfrentan este tipo de problema. Los fabricantes de dulces deben determinar si
van a proteger el precio del azúcar, cacao y otras materias primas. Los productores de cereales podrían
desear dar cobertura a los costos de los cereales. Los productores de cables eléctricos podrían desear dar
cobertura al costo del cobre. Los criadores de ganado podrían desear dar cobertura al precio de su alimento.
En cada caso, se debe determinar el deseo de asegurar el precio del input. Es más, si es deseable asegurar el
precio de los inputs, el fabricante podría elegir contratos forward con los proveedores o la cobertura del
mercado de futuros.
Ahora presentamos una formulación más precisa de la cobertura óptima cuando los precios del output
y el input son inciertos. La formulación es una modificación de la discusión de cobertura óptima presentada
en el Capítulo 4 para permitir la incertidumbre relacionada con el precio del output del productor así como el
precio de los inputs de las mercancías.
La notación utilizada en esta sección es la siguiente:
~
PT  precio incierto de una unidad de output en el momento futuro T .
QP  Número de unidades del producto a venderse
~
ST  precio cash incierto del input en el momento futuro T . Nosotros asumimos sólo un input, aunque
fácilmente se podrían incluir inputs adicionales.
QS  Número de unidades del input requeridas para producir QP unidades del producto.
F0  precio de futuros en el momento 0.
~
FT  precio incierto de los futuros en el momento futuro T .
nF  número de contratos futuros en cartera. ( nF es positivo para una posición larga y negativo para una
posición corta).
K  costos fijos incurridos en el proceso de producción.
La utilidad incierta, ~h , de un productor que vende QP unidades, utiliza QS unidades como inputs y protege
nF unidades en el mercado de futuros es
~ ~ ~
~  P Q  S Q  ( F  F )n  K
h T P T S T 0 F (6.3)
al dividirla entre QP , la ecuación puede re-enunciarse en términos de la utilidad por unidad de output:
~h ~ ~ QS ~ n K
 PT  ST  ( FT  F0 ) F  (6.4)
QP QP Q P QP
Ahora definimos una unidad del input como el tamaño del contrato de futuros. En otras palabras, si el input
es azúcar, ahora definimos una unidad de azúcar como 112,000 libras, que es un contrato de futuros.
También definimos una unidad del output como la cantidad producida. Por el número de unidades contenidas
en el contrato de futuros. Por lo tanto, si se utilizan 112,000 libras de azúcar para producir 1,000,000
chocolates, una unidad de output es 1,000,000 chocolates. Esto significa que QS / QP  1.0 .29 Con estas
convenciones la ecuación se puede escribir como
~h ~ ~ ~ K
 PT  ST  ( FT  F0 )h 
QP QP
donde h  nF / QP es el ratio de cobertura. Para ser consistente con el Capitulo 4, escribimos la ecuación con
cambios en precios agregando y restando P0  S 0 al término de la mano derecha:
~h ~ ~ ~
 c   P   S  h F (6.5)
QP
donde c  P0  S 0  K / QP
Siguiendo el procedimiento del Capítulo 4, la variación de la utilidad por unidad puede ser calculada como
 h2   P2   S2  h 2 F2  2 PS  2h PF  2h SF . (6.6)
Si el productor no ha cubierto con futuros de manera que h  0 , el riesgo es

 h2   P2   S2  2 PS . (6.7)
Noten que  PS   PS  P S , donde  es el coeficiente de correlación. Si el cambio en el precio del input y el

output están perfectamente correlacionados,  PS  1 , lo cual implica que  h2  0.0 , asumiendo que las
variaciones de los cambios en los precios del input y el output sean las mismas. Este es el caso de una
cobertura natural perfecta. Sin embargo, si el precio del output y el input no están perfectamente
correlacionados, puede ser deseable la cobertura con futuros. En ese caso, el problema es encontrar el ratio
de cobertura que minimice la variación de la utilidad por unidad del productor.
El valor de h que minimiza  h2 en (6.6) se encuentra tomando la derivada de  h2 con respecto a h e
d h2
igualándola a cero:  2h* F2  2 PF  2 SF  0.
dh
*
Despejando h , el ratio de cobertura óptima es
 SF   PF
h*  . (6.8)
 F2
Noten que  SF /  F2  bS es la pendiente de una regresión de  S sobre  F y que  PF /  F2  bP es la

pendiente de una regresión  sobre  . Por lo tanto, otra manera de escribir la cobertura óptima es
P F
h  bS  bP . (6.9)
*
Si tanto el input como el output reaccionan de la misma manera al cambio en el precio de los futuros, esto es,
si los coeficientes de regresión son iguales, h*  0.0 , es el caso de una cobertura natural perfecta.
Para algunos productores, los precios del output pueden ser muy estables o pueden fijarse mediante contratos
a largo plazo de manera que bP  0.0 . En este caso, la cobertura óptima se determina por bS , la sensibilidad
del precio del input ante el precio de los futuros. El productor comprará futuros para asegurar el costo de los
inputs. Para algunos productores, el precio del input podría ser estable o puede fijarse mediante la compra de
futuros de manera que bS  0.0 . En este caso, el productor venderá futuros para cubrirse contra una
disminución en el precio del output.
Como vimos en el Capítulo 4, la cobertura óptima también puede desarrollarse comenzando con
ecuaciones de regresión que relacionen los precios del output y el input con el precio de los futuros:
~ ~
 P  a P  bP  F  e~P , (6.10)
~ ~
  a b  ~
S S S e (6.11)
F S
donde a y b son los términos de la intersección y la pendiente, respectivamente; y e es un término de

alteración aleatoria.
Sustituyendo estas ecuaciones en la ecuación de la utilidad por unidad (6.5) se obtiene
 h
 c  aP  aS  (bP  bS  h) F  eP  eS (6.12)
QP
es claro que la utilidad por unidad puede ser independiente de los movimientos en los precios del input y el
output al establecer h*  bS  bP , el resultado derivado arriba minimizando la variación.
La efectividad de la cobertura depende del grado en que los errores residuales e~P y e~S son
eliminados. La variación de la utilidad por unidad restante después de la cobertura es la variación de (6.12).
Cuando h  bS  bP , esto es igual a Var (e~P  e~S ) .
Calculo del Ratio de Cobertura

La cobertura óptima requiere de un calculo de bS y bP en las ecuaciones (6.10) y (6.11). Como vimos en el
Capítulo 4, el procedimiento usual para calcular estos parámetros es utilizando datos de series de tiempo
históricas. Para ilustrar la cantidad de cobertura cuando el precio de un input y un output están
correlacionados, nosotros consideramos el caso de un refinador de petróleo que produce gasolina a partir del
petróleo crudo. Supongamos que un barril de petróleo crudo produce treinta galones de gasolina. Para
~
propósitos de la ecuación (6.4), el precio del output, PT , se refiere a treinta galones de gasolina; y el precio
~
del input, ST , se refiere a un barril de petróleo crudo. El vehículo de cobertura es un contrato de futuros de
petróleo crudo negociado en la Bolsa Mercantil de Nueva York, cuyo precio es cotizado en dólares-por-
barril. Se recolectaron los datos de los precios semanales (Martes) durante 53 semanas en el período entre
noviembre de 1988 y noviembre de 1989 para la gasolina sin plomo vendida en Nueva York, para el petróleo
crudo “amargo” –“sour”– de West Texas y el contrato de futuros “ligeramente dulce” –“ light sweet”– de
petróleo crudo.
Se realizaron las siguientes regresiones:
~ ~
 P  a P  bP  F  ~
eP , (6.13) y
~ ~ ~
  a  b   e (6.14)
S S S F S
29
Se debe notar que la empresa tiene un ratio fijo input/output y que planea producir un número fijo de unidades, todas las cuales
serán vendidas. En realidad, la empresa tiene cierta flexibilidad en la manera que combina los factores de producción, y puede no
En el análisis de regresión, las premisas que dictan el término de error son E (e )  0, E (e F )  0 . También
asumimos que los términos de error sucesivos en el tiempo son independientes. Los resultados de la
regresión se proporcionan en la Tabla 6.4. El calculo del ratio de cobertura óptima es
hˆ*  bˆS  bˆP = 0.9738 - 0.6487 = 0.3251 .
En otras palabras, por cada treinta galones de gasolina que la refinería produce, debe comprar
contratos de futuros por 0.3251 barriles de petróleo aún cuando se necesite un barril para producir treinta
galones de gasolina. La razón por la que sólo se compran 0.3251 barriles de futuros petróleo es que la
refinería tiene una cobertura natural parcial que se da por hecho que el precio recibido por la gasolina
compensa en parte cualquier cambio en el precio del petróleo utilizado para producir la gasolina.
Los valores pequeños de R 2 quieren decir simplemente que la efectividad de la cobertura no es
mucha. Los cambios en los precios de los futuros de petróleo sólo explican el 19.56 porciento de la variación
en los precios de la gasolina y sólo el 59.44 porciento de la variación de los precios del petróleo crudo de
Texas. Por lo tanto, el ratio de cobertura de 0.3251 puede resultar incorrecto después de todo. Por ejemplo,
supongamos que el estimado b̂S , está un error estándar por debajo de su verdadero valor y que el estimado
b̂P está un error estándar por encima de su verdadero valor. Entonces
h* = (0.9738 + 0.1098) - (0.6487 - 0.1741) 0.6090,
lo cual es casi el doble del ratio de cobertura estimado.
TABLA 6.4 Resumen de la regresión cambios de los futuros de petróleo crudo “amargo”, y
futuros de petróleo “ligeramente dulce” y cambios de precio de futuros de gasolina sin plomo de
noviembre de 1988 a noviembre de 1989.
Número de observaciones = 5330
 P = -0.0312 dólares ˆ P = 0.9000 dólares
 S = 0.1161 dólares ˆ S = 0.7994 dólares
 F = 0.2580 dólares ˆ F = 0.6370 dólares
â P = -0.1985 s (aˆ P ) = 0.1187
b̂P = 0.6487 s (bˆ ) = 0.1741
P
2
R = 0.1956 ˆ e = 0.6516
P
âS = -0.1351 s (aˆ S ) = 0.0748

b̂S = 0.9738 s (bˆ ) = 0.1098
S
2
R = 0.5944 ˆ e = 0.2592
S
6.5 OFERTA Y DEMANDA DE ALMACENAMIENTO

En el Capítulo 3 mostramos que el precio de los futuros no puede exceder el precio spot en más del costo del
almacenamiento de la mercancía:
Ft  St  Bt . (3.1)
poder vender todo lo que produce.

30
 denota la media del cambio de precio semanal y ˆ es el estimado de la desviación estándar. s() es el error estándar del
estimado del coeficiente, ˆ e es el error estándar de la regresión.
Otra manera de expresar esta relación es decir que la base no puede exceder el costo del almacenamiento:
Ft  S t  Bt . Los comerciantes que almacenan mantendrán la mercancía sólo si la base, que puede ser vista
como el precio de almacenamiento, cubre los costos de almacenamiento. Se debe notar que el costo de
almacenamiento, Bt , es el costo marginal del almacenamiento por el período de tiempo en cuestión. Otra
tonelada de trigo será almacenada si el costo marginal, Bt , de almacenar esa tonelada es menor que el precio
recibido, Ft  S t , por almacenar esa tonelada.
El comerciante que almacena toma el precio del almacenamiento como dado, pero en conjunto el
precio del almacenamiento depende de la interacción de la oferta y demanda del almacenamiento. La línea
sólida en la Figura 6.3 trata la curva de oferta agregada del almacenamiento para mercancías agrícolas como
trigo, maíz y soya. La curva se basa en datos reales que muestran cuánto está almacenado a cada valor de la
base, F  S . A la izquierda del punto A, el costo marginal out-of-pocket del almacenamiento, B ,
representado por la línea punteada, excede la base.31 A la derecha del punto A, la curva de oferta del
almacenamiento coincide con el costo marginal del almacenamiento, B . El segmento horizontal extendido
de la curva de la oferta significa que los costos de almacenamiento son constantes por encima de este rango.
A la izquierda del punto A, las cantidades positivas son almacenadas inclusive cuando la base es menor que
el costo marginal del almacenamiento, B , dado que los productores derivan una utilidad de conveniencia por
tener la mercancía. Mientras menor sea la cantidad de mercancía en existencia, mayor deberá ser la utilidad
de conveniencia para compensar el hecho que el precio recibido por el almacenamiento es menor que el
costo marginal del almacenamiento. Presumiblemente el productor traspasa el costo del almacenamiento en
el precio final del producto.
B, F-S Figura 6.3 Curva de Oferta del Almacenamiento
Utilidad de Conveniencia
Q
QL A QM QH
La demanda del almacenamiento varía a lo largo del año de cosecha. Es mayor inmediatamente después de la
cosecha y disminuye conforme se va utilizando el cultivo. La línea vertical QH representa la demanda de
almacenamiento después de una cosecha excelente. Se interseca con la curva de oferta donde los costos
marginales de almacenamiento están subiendo, lo cual refleja el hecho que se deben estar utilizando medios
de almacenamiento deficientes y de costo alto para manejar la cosecha. Más adelante en la estación de
cosecha, la curva de demanda se traslada hacia la izquierda a un punto como QM , y la base cae.
31
La línea punteada se inclina hacia arriba sobre la premisa que los costos marginales del almacenamiento para los productores
que almacenan son mayor cuando se almacenan pequeñas cantidades; en otras palabras, existen economías de escala hasta el punto
A.
Finalmente, cerca al final de la estación de almacenamiento y justo antes de la fecha de la nueva

cosecha, la curva de demanda está en un punto como QL . En este punto, la base es pequeña o negativa en
espera de la nueva cosecha, y los costos marginales del almacenamiento exceden la base. Por lo tanto,
durante la estación de la cosecha, la base de un mes normalmente disminuirá conforme los medios de
almacenamiento se usen cada vez menos.
Al comienzo de la estación de almacenamiento, los que almacenan y los productores pueden utilizar
los precios de los futuros para decidir cuánto almacenar y por cuánto tiempo. La estructura de plazos de los
precios de los futuros proporciona un estimado de la base para diferentes períodos futuros. Por ejemplo, la
Tabla 6.5 enumera los precios de los futuros de soya en octubre de 1989, y calcula una base mensual a partir
de una diferencia en los precios de futuros sucesivos. Por lo tanto, la base implícita entre noviembre y enero
es de 5.88 centavos. La demanda pico del almacenamiento parece darse entre noviembre y marzo, y luego
disminuye. Los costos de almacenamiento parecen ser menores entre marzo y abril y entre mayo y julio.
Después de julio, la base se vuelve negativa en espera de la cosecha del nuevo cultivo. Hasta el punto en que
se puedan utilizar los medios de almacenamiento para varias cosechas, el patrón de la base esperada para la
soya no dependerá solamente de la oferta y la demanda de almacenamiento de soya sino también de la oferta
y demanda del almacenamiento de otros cultivos. Esto significa que no es perfectamente exacto relacionar el
patrón de la base de la soya con el ciclo de cosecha de la soya. En realidad esto es más complicado. Pero
para el que almacena diferentes mercancías, la base implícita derivada de la estructura de plazo de los
precios de los futuros proporciona una guía útil para planear qué mercancías almacenar, cuándo almacenarlas
y durante cuánto tiempo. En general, los comerciantes que almacena a un costo alto almacenarán cuando la
demanda de almacenamiento sea alta, mientras que los que almacenan a costo bajo almacenarán por un
período de tiempo más largo.
TABLA 6.5 Precios de futuros de soya y la base mensual implícita en octubre de 1989
Vencimiento del Contrato Precio de Soya en centavos Base mensual32
Noviembre ‘89 552.5
Enero ‘90 564.25 5.88
Marzo ‘90 576.00 5.88
Mayo ‘90 586.00 5.00
Julio ‘90 592.25 3.18
Agosto ‘90 590.00 -2.25
Setiembre ‘90 573.00 -17.00
Noviembre ‘90 572.00 -0.50
La oferta del almacenamiento de mercancías no-estacionales como el petróleo, cobre y otros tendrá una
forma similar. Sin embargo, la demanda de almacenamiento no tendrá el patrón estacional predecible que es
típico en las mercancías agrícolas.
La sobre producción de petróleo colocaría una presión de demanda en los medios de almacenamiento
(un movimiento hacia la derecha sobre la función de la oferta de almacenamiento) y causaría un incremento
sobre la base.
Una huelga laboral en la producción de cobre reduciría la demanda de almacenamiento y causaría un
movimiento hacia la izquierda sobre la función de oferta de almacenamiento, tal vez inclusive hasta un punto
donde la base sea negativa. Así como en una mercancía agrícola, la base negativa significa que se anticipa un
incremento en la producción, en este caso, cuando acabe la huelga.
6.6 RETORNOS PARA LOS ESPECULADORES
32
La base se calcula como la diferencia entre los precios de contratos de futuros adyacentes dividida entre el número de meses que
separan el vencimiento de los contratos.
A diferencia de los que tratan de evitar el riesgo del precio de la mercancía, los especuladores asumen el
riesgo con la esperanza de obtener utilidad. La posición más riesgosa que puede asumir un especulador es
simplemente estar largo o corto en una mercancía. Una posición menos riesgosa es realizar un spread –
comprar(vender) un contrato de futuros y vender(comprar) otro contrato relacionado. Por ejemplo, un
meteorólogo que anticipa una sequía en Kansas podría decidir obtener una buena utilidad comprando futuros
de maíz. Si se equivoca, puede perder bastante. Un spread es menos riesgoso porque los precios de los
futuros tienden a moverse juntos. Los márgenes típicos se conforman de un vencimiento equiparado con otro
vencimiento en la misma mercancía o una mercancía con una mercancía relacionada para la misma fecha.
Por ejemplo, la sequía podría afectar los precios del maíz de manera diferente que los precios del trigo. En
este caso, podría ser deseable comprar futuros de maíz y vender futuros de trigo. Si los precios de los futuros
de maíz y trigo bajan juntos, no hay pérdida. Sólo si el maíz disminuye más que el trigo hay pérdida para el
especulador.
Los especuladores en futuros de mercancías, como en otros vehículos de inversión, deciden qué
comprar o vender sobre la base de un análisis fundamental o técnico. Los analistas fundamentales examinan
la oferta y la demanda de una mercancía y tratan de predecir la oferta y la demanda futura y así los cambios
futuros en el precio. En mercancías agrícolas, el clima es un factor importante en la oferta. En otras
mercancías, pueden ser importantes los factores políticos o la probabilidad de huelgas laborales. El análisis
técnico se enfoca en el patrón de los precios anteriores con la esperanza de predecir los cambios futuros en el
precio. Los técnicos preparan un diagrama de la conducta de los precios y el volumen de negociación y
buscan patrones que predigan los cambios futuros en los precios.
En un mercado eficiente, ni el analista técnico ni el fundamental pueden esperar realizar utilidades

anormales. Como en el mercado de valores, la evidencia es que los mercados de futuros de mercancías son
eficientes. Los estudios de la eficiencia de los mercados de futuros de mercancías físicas han tomado
diferentes enfoques. Como vimos en el Capítulo 5, un enfoque ha sido preguntar si los especuladores como
grupo obtienen una prima de riesgo. Como también vimos allí, existe muy poca evidencia que muestre que
los especuladores como grupo obtienen utilidades. Esto podría ser porque el riesgo asumido por los
especuladores es completamente diversificable, lo cual, según el modelo de precios del activos de capital
CAPM, significa que no necesita pagar una prima de riesgo. O podría significar que una cierta clase de
especuladores –apostadores o tontos– pierden dinero antes especuladores profesionales, de manera que los
especuladores como grupo no obtienen retornos anormales.
Un segundo enfoque ha sido analizar el patrón de series de tiempo de los precios de los futuros para
ver si existe alguna dependencia que se puede explotar para obtener utilidad. En el mercado de valores, Fama
(1970) ha denotado a estas pruebas como pruebas de “forma débil” de la eficiencia de mercado ya que
buscan determinar si un conjunto de información “débil” –la secuencia pasada de los precios– puede predecir
los cambios futuros en los precios. Si el mercado es eficiente, el precio de los futuros en t refleja toda la
información disponible en este punto, incluyendo la historia anterior de los precios. Por lo tanto, no se puede
utilizar la historia anterior de los precios para generar una utilidad positiva en el período. Una implicancia
empírica simple de los mercados eficientes es que el retorno actual de los futuros no debe estar
correlacionado con el retorno de los futuros mañana, esto es,  ( Rt , Rt 1 )  0.0 . Varios investigadores han
examinado la correlación de los retornos de futuros sucesivos. Vean, por ejemplo, a Smidt (1965) y
Stevenson y Bear (1970). Ellos encuentran que la correlación en serie no es económicamente significativa.
Cuando se observa una dependencia en serie, no es lo suficientemente grande para superar los costos de
transacción incurridos al tratar de obtener utilidad.
Una forma alternativa de investigación de las series de tiempo es simular una regla de inversión
basada en la secuencia de precios pasada. Por ejemplo, una regla podría ser comprar después que el precio de
los futuros se ha incrementado en tres porciento, mantener hasta que el precio disminuya en tres porciento,
momento en el cual la posición es vendida y se asume una posición corta, y así sucesivamente. Si los
mercados son eficientes y el retorno normal es cero, esta regla de inversión técnica no debe ser rentable. Las
pruebas empíricas concluyen que estas reglas no son rentables. Al llevar a cabo estas pruebas, se debe tener
cuidado en especificar la regla por adelantado antes de ver los datos, ya que uno siempre puede encontrar
alguna regla que hará dinero si se aplica a una secuencia particular de precios históricos.
Otra implicancia de los mercados eficientes es que el análisis fundamental tampoco puede proporcionar
retornos anormales si el análisis se basa solamente en la información pública disponible para todos los
analistas. En los mercados eficientes, toda la información pública se refleja en el precio actual. En otras
palabras, la información pública disponible en t no se puede utilizar para predecir el precio en t  1 .
Presumiblemente, los recursos son gastados en recolectar información con la esperanza de descubrir
información que no sea de conocimiento general, de manera que periódicamente se puedan obtener retornos
anormales. En los mercados eficientes, estos retornos anormales no deben, en promedio, exceder el costo de
adquirir la información especial que proporcionan estos retornos anormales.
Un tercer enfoque para probar la eficiencia de los mercados de futuros de mercancías es examinar
subgrupos de inversionistas, como comerciantes profesionales y consultores de inversión, para ver si ellos
pueden obtener retornos anormales. Qué es un retorno anormal requiere de una mayor discusión. Es poco
probable que el retorno anormal de los comerciantes y consultores profesionales sea cero; por que si lo fuese,
¿cómo alimentarían a sus familias? Uno esperaría que los inversionistas profesionales, aquellos que invierten
tiempo y dinero en el análisis, generen utilidades de negociación positivas o cobren comisiones. En la bolsa
de valores, un subgrupo muy popular para examinar son los fondos mutuos. Allí se ve que un fondo mutuo
típico no supere el desempeño del mercado, aunque cobran comisiones que permiten a los gerentes de cartera
alimentar a sus familias (algunas veces regiamente). Los descubrimientos en los mercados de futuros físicos
son consistentes con los mercados eficientes. Los tratados de Rockwell (1967) y Houthakker (1957)
examinan los retornos de los grandes protectores, los grandes especuladores y los pequeños comerciantes en
el mercado de futuros físicos. Estos estudios concluyen que los grandes especuladores sí obtienen utilidades,
lo cual es consistente con la idea que los especuladores profesionales deben tener una utilidad. Los estudios
no se ponen de acuerdo en si otros especuladores hacen o pierden dinero. Según la hipótesis nula de retornos
esperados cero, las ganancias de un grupo de especuladores deben compensar las pérdidas de los
especuladores restantes. Rockwell argumenta que este es el caso. Houthakker argumenta que los
especuladores pequeños también hacen dinero, lo cual significa que él rechaza la hipótesis nula de retorno
cero para los especuladores.
Elton, Gruber y Rentzler (1987, 1989) han examinado recientemente el desempeño de inversión de
los fondos de mercancías para el período 1979 a 1985. Los fondos de mercancías son análogos a los fondos
mutuos en la bolsa de valores en que son manejados profesionalmente. Durante los seis años analizados, el
retorno promedio anual del período de tenencia es –0.0007. Este retorno no refleja todos los costos de
transacción que se requiere paguen los inversionistas. Por lo tanto, los fondos de mercancías tienen un
desempeño inferior a otros instrumentos de inversión menos riesgosos como los títulos valores del gobierno.
El desempeño de los fondos de mercancías no apoyan la idea que los gerentes profesionales pueden obtener
utilidades positivas. De hecho, el desempeño es tan malo que Elton, Gruber y Rentzler (1989) cuestionan el
raciocinio de los inversionistas de fondos.
6.7 VÍNCULOS DE PRECIOS INTERNACIONALES: LA LEY DE UN SOLO

PRECIO
La mayoría de las mercancías son activamente negociadas internacionalmente. Estados Unidos es un
exportador importante de productos agrícolas. El petróleo crudo producido en el Medio Oriente es importado
por Europa y Estados Unidos. El cacao y el café son producidos en Sudamérica y Africa y son importados
por otros países. Por lo tanto, es importante especificar la relación de los precios de la misma mercancía
denominados en monedas diferentes y determinar si es posible utilizar contratos de futuros negociados en
Estados Unidos y denominados en dólares para cubrir posiciones de una mercancía en un país diferente.
En la ausencia de costos de transporte y de transacción, el precio en dólares, S d , de una mercancía

debe ser igual al precio extranjero, S f , de la misma mercancía ajustada por el tipo de cambio, X :
S d  XS f . (6.15)
La tasa de cambio se define como el precio en dólares de la moneda extranjera. La relación anterior,
conocida como la Ley de un Solo Precio (LOP), se mantiene debido al arbitraje de mercancías. Si el precio
en dólares de una mercancía excediese el costo de compra la mercancía en un país extranjero, alguien
dedicado al arbitraje compraría la mercancía en el país extranjero y la importaría a Estados Unidos,
depreciando así los precios en Estados Unidos y elevando los precios en el país extranjero. Supongamos, por
ejemplo, que el precio de una tonelada de trigo en Gran Bretaña cuesta 2.00 libras; la tasa de cambio es 1.60
dólares por libra; y el precio de una tonelada de trigo en Estados Unidos es 3.10 dólares. Alguien dedicado al
arbitraje podría comprar una tonelada por 3.10 en Estados Unidos y venderla en Gran Bretaña por 2.00 libras
y convertir las libras en (1.6)(2) = 3.20 dólares, lo cual proporciona una utilidad de 10 centavos por tonelada.
Este arbitraje eleva el precio del trigo en Estados Unidos y lo disminuye en Gran Bretaña hasta que el LOP
se re-establece. En la práctica, el LOP no se mantiene exactamente porque el arbitraje de mercancías es
costoso. Los costos de transporte y otros costos de transacción llevan a diferencias espaciales en el precio
entre los países. Así como el precio del trigo es diferente en Kansas y la ciudad de Nueva York, así el precio
del trigo establecido en dólares puede varias entre los diferentes países. Además, las comparaciones de
precios, inclusive de mercancías estrechamente definidas, usualmente no toman en cuenta totalmente las
diferencias de grado en las mercancías.
También se puede definir el LOP para los precios de los futuros:
Fd ,t (T )  Ft X (T ) F f ,t (T ) (6.16)
donde Fd ,t (T ) y Ff ,t (T ) son respectivamente los precios de futuros domésticos y extranjeros para los
contratos de mercancías que vencen en T , y Ft X (T ) es el precio de los futuros del contrato en moneda
extranjera con vencimiento en T . El LOP se debería mantener más estrechamente para los precios de los
futuros o los forward porque se pueden lograr acuerdos óptimos para el transporte a lo largo del tiempo hasta
el vencimiento.
TABLA 6.6 Desviaciones de la Ley de un Solo Precio para los precios de mercancías
estadounidenses y británicas33
Mercancía Media Desviación Estándar
Plata al contado 0.005 0.0200
Plata - forward de tres meses -0.0050 0.0146
Cobre al contado 0.0038 0.1959
Cobre - forward de tres meses 0.0116 0.0258
Futuros de café 0.1991 0.0941
Futuros de cacao -0.0386 0.0795
Futuros de trigo -0.2305 0.2023
La Tabla 6.6 presenta evidencia, tomada de Protopapadakis y Stoll (1983), sobre desviaciones del
LOP para ciertos precios inmediatos, forward y de futuros. Según el LOP, la media y la desviación estándar
del LOP serán cero porque el LOP es una relación de arbitraje no-aleatoria. La desviación promedio es
pequeña para la mayoría de las mercancías, particularmente para las mercancías como la plata que son
33
Las observaciones del precio de la mercancía y la tasa de cambio son semanales. El período observado es 1972-1980 para la
mayoría de las mercancías. Los datos del trigo son para el período 1976-1979.
definidas precisamente. Las grandes desviaciones de la media en café y trigo se pueden atribuir a factores
especiales. En 1977-80, el precio de futuros de café estadounidense era relativamente alto con relación al
precio de futuros británicos debido a la manipulación de un contrato de futuros de café en Nueva York
realizado por un cartel de café sudamericano. El precio de los futuros de trigo estadounidense era bajo en
comparación con el precio de futuros británicos debido a subsidios en los precios agrícolas en Gran Bretaña
y la CEE mantuvo artificialmente un precio alto.
6.8 MONOPOLIOS Y ACOSO CORTO (CORNERS AND SHORT SQUEEZES)

Un monopolio o un acoso corto se origina si alguien logra el control del suministro entregable de una
mercancía y actualmente mantiene una posición larga en futuros. Al vencimiento, las posiciones cortas
pueden liquidar su posición en futuros negociando con una larga; o pueden entregar la mercancía. En un
acoso corto, ambas opciones son canceladas. La posición corta en el mercado futuros no puede liquidarse
porque la larga se rehusa a vender, y la mercancía no puede ser entregada dado que la larga controla la el
suministro entregable y se rehusa a vender. Las posiciones cortas usualmente tratan de cubrir su posición
corta en el mercado de futuros comprando futuros; elevando así el precio de los futuros. La subida del precio
de los futuros en el mes de la entrega con relación al precio de otros futuros y el precio cash usualmente
señala la presencia de un acoso corto. En la ausencia de la intervención de bolsas o reguladores, la posición
larga se rehusará a vender hasta que los precios se eleven substancialmente. Hoy, las bolsas y la CFRC
usualmente intervienen cuando se sospecha un acoso corto y se requiere que una persona o empresa asume la
responsabilidad del acoso corto para liquidar los contratos de futuros.
El ejemplo más reciente de un acoso corto que casi se llega a dar es el intento de una empresa italiana
negociadora en cereales, Ferruzzi Finanziaria S.p.A., para monopolizar el contrato de soya de julio de 1989.
El contrato de julio vencía el 19 de julio. A comienzos de julio, Ferruzzi mantenía más de la mitad de las
posiciones largas netas de los futuros de julio, que era el doble del suministro entregable, y poseía el 85
porciento de la posición entregable de soya. Las posiciones cortas tendrían que mover cantidades masivas de
soya a los puntos de entrega aprobados (Chicago y Toledo) para poder realizar la entrega de sus contratos de
futuros, lo cual era imposible faltando tan poco tiempo para el vencimiento. En reacción a este monopolio o
acoso potencial, la Junta de Negociaciones de Chicago ordenó que aquellos que mantuvieran posiciones de
futuros por encima de tres millones de toneladas tenían que liquidar. Esto significaba que Ferruzzi tenía que
vender gran parte de su posición larga a los de posición corta, evitando así un acoso corto. Los precio de los
futuros de soya de julio, que se habían elevado en reacción al desarrollo del acoso, cayeron a sus niveles
normales.34
Los mercados de futuros tratan de proteger contra un monopolio o acoso al ampliar los plazos de
entrega a varios grados y localizaciones (Chicago y Toledo, en el caso de la soya); pero en el caso que los
suministros de una mercancía aún estén monopolizados, es muy apropiado que las bolsas tomen acciones
para forzar la liquidación para romper el acoso.
6.9 RESUMEN
En este capítulo, se identifican los principales contratos de futuros de mercancías físicas y se discuten los
factores que dan origen a los mercados de futuros de mercancías físicas. Se analizan los patrones de
inventario y precio para las mercancías con suministro estacional y las mercancías no-estacionales.
Se explica y modela el uso de contratos de futuros como una herramienta de cobertura para los
almacenan mercancías, los comerciantes de mercancías, y para los productores. Con frecuencia un productor
tiene precios inciertos en sus inputs y outputs, algo que hace que el problema de la cobertura sea más difícil
que si sólo existiese incertidumbre en el precio de una mercancía. En esta situación se presenta un marco
referencial de cobertura óptima y se ilustra el cálculo de la cobertura óptima.
34
Los detalles de este intento de acoso están en la Junta de Negociaciones de Chicago (1990).
La base en mercancías físicas depende de la cantidad de la mercancía que debe almacenarse. Después
de la cosecha, la demanda de almacenamiento es alta, lo cual causa un movimiento a lo largo de la curva de
oferta de almacenamiento a un costo mayor de almacenamiento y a una base mayor. Cuando la demanda de
almacenamiento disminuye, la base también disminuye.
Los especuladores en futuros de mercancías físicas tienen dificultades para obtener utilidades, igual
que en otros mercados. La evidencia implica que los mercados de futuros son eficientes en el sentido que no
se logran utilidades anormales de manera consistente. Las reglas de negociación técnica no son vistas como
rentables (después de los costos de negociación). Los fondos de mercancías administrados profesionalmente
se desempeñan por debajo del mercado. El hecho que parezca que los especuladores no obtienen utilidades
implica que se provee a los que deseen una cobertura el servicio de asumir el riesgo a un costo muy bajo.
Los contratos de futuros de mercancías son utilizados por los que desean una cobertura y los
especuladores en muchos países. Este capítulo muestra cómo los precios de la misma mercancía están
vinculados en términos de monedas diferentes. El capítulo termina con una discusión sobre monopolios y
acoso corto.
7 STOCK INDEX FUTURES CONTRACTS
7.1 STOCK INDEX FUTURES MARKETS
7.2 COMPOSITION OF STOCK INDEXES

 Value-Weighted Arithmetic Indexes
 Price-Weighted Arithmetic Indexes
 Equal-Weighted Geometric Indexes
 Stock Index Simulations
7.3 INDEX ARBITRAGE AND PROGRAM TRADING

 Stock Replacement
 Practical Considerations in Index Arbitrage
7.4 INTRADAY BEHAVIOR OF RETURNS
7.5 HEDGING MARKET RISK

 Forming the Hedge Portfolio
 Assessing Hedging Effectiveness
 Decomposing the Hedge Portfolio Return
7.6 SUMMARY
7 STOCK INDEX FUTURES CONTRACTS
Arguably, the most exciting financial innovation of the 1980s has been the introduction of stock index
futures contracts. These contracts, written on the value of various stock index portfolios, provide important
benefits to stock portfolio managers. The uses and benefits of these contracts are described in this chapter.
We begin with a description of the history of stock index futures contracts in the U.S. and an explanation of
current contract designs. The second section details the composition of the stock indexes that underlie
currently traded index futures contracts.
Section 3 describes the index arbitrage that holds the cost of carry relation in alignment and explains
the concept of “program trading.” In section 4, the intraday price behavior of the index and its futures
contracts is investigated to see how well the price movements in the two markets are synchronized. The
chapter concludes with an illustration of hedging with stock index futures contracts.
7.1 STOCK INDEX FUTURES MARKETS

The first stock index futures contract was introduced in February 1982 by the Kansas City Board of Trade.
This contract, the Value Line futures contract, is written on the Value Line Composite Index, a stock index
that consists of approximately 1700 stocks from the New York, American, and OTC stock markets.35 The
Chicago Mercantile Exchange quickly followed suit in April 1982 with a futures contract on the S&P 500
stock index, and then the Chicago Board of Trade in July 1984 followed with a futures contract on the Major
Market Index. Other stock index futures on over-the-counter stocks have been introduced, but most have
failed. Table 7.1 contains the contract specifications of the five stock index futures contracts currently active
in the U.S.
By far the most active stock index futures contract is that on the S&P 500 index. Table 7.1 shows that
this contract trades at the Chicago Mercantile Exchange from 8:30 AM to 3:15 PM (CST). On a given day,
S&P 500 futures contracts extending out four different maturities may trade. The contract maturities will be
the following March, June, September, and December. The last trading day of the S&P 500 futures contract
is the third Thursday of the contract month. Cash settlement of the contract takes place at the opening prices
of the index stocks on Friday.36
The contract denomination is 500 times the futures price. On November 13, 1991, for example, the
December 1991 futures price was $398.30, so the stock equivalent of the futures is $398.30 x 500 or
$199,150. The minimum price increment for changes in the futures price is $0.05 x 500 or $25. As of April
1991, the initial speculative margin for the S&P 500 contracts was $22,000,37 and the maintenance margin
was $9,000.
The specifications of the other index contracts are also shown in Table 7.1. Next to the S&P 500, the
most active markets are for the futures contracts on the NYSE Composite Index and the Major Market Index.
The Value Line futures contracts have never been particularly active relative to their counterparts on the
other futures exchanges, probably because of the way in which the index level is computed.38 The only
difference between the Value Line and Mini Value Line index futures contracts is that the latter contract is
one-fifth the size of the former.
Table 7.2 contains a clipping from the Wall Street Journal showing prices for the various index
futures contracts as of the close of trading on Wednesday, November 13, 1991. Only the three nearby S&P
35
The composition of the various stock indexes is discussed in the next section.
36
In June 1987, the Chicago Mercantile Exchange and the New York Futures Exchange changed the settlement of their S&P 500
and NYSE index futures contracts from the close of trading to the open in an attempt to mitigate concern about occasional
abnormal stock price movements in the “triple witching hour.” The futures contracts on the Major Market and Value Line indexes
continue to settle at the close. For an analysis of the effects of this change, see Stoll and Whaley (1991).
37
Margins are adjusted when the risk of the underlying index changes perceptibly. Prior to the October 19, 1987, stock market
crash, speculative margin on the S&P 500 futures contract was $6,000. Immediately following the crash, speculative margins were
set as high as $20,000.
38
The index composition is described later in this chapter.
500 futures contracts were active on November 13 –the December 1991 and the March and June 1992
contracts. The estimated trading volume on that day was 42,125 contracts. The implied dollar stock
equivalent of this volume of trading is at least $398.30 x 500 x 42,125, or $8.39 billion. As is usually the
case, the nearby futures contract is the most active, as is reflected through the higher open interest figure for
the December contract. The underlying S&P 500 index level, 397.42, is also reported in the table, just below
the futures price summary.
7.2 COMPOSITION OF STOCK INDEXES

The indexes underlying the futures contracts contained in Tables 7.1 and 7.2 fall into one of three general
categories: (a) value-weighted arithmetic stock indexes; (b) price-weighted arithmetic indexes; and (c) equal-
weighted geometric indexes. The term arithmetic refers to the fact that the market values or returns of the
individual stocks are “added up.” The term geometric refers to the case where the values or returns are
“multiplied.” The S&P 500 and NYSE Composite indexes are in the first category; the Major Market Index
falls in the second; and the Value Line Index falls in the third.
TABLE 7.1 Contract specifications of stock index futures contracts trading in the U.S.
Index Trading Contract Units/Minimum
Last Day of Trading40
(Exchange) Hours Months39 Price Fluctuation
S&P 500 8:30-3:15 500 x index/0.05
3,6,9,12 Third Thursday
(CME) (CST) ($25)
NYSE Index 9:30-4:15 500 x index/0.05 Thursday Preceding third
376,9,12
(NYFE) (EST) ($25) Friday
Major Market 8:30-3:15 3 current months 250 x index/0.05 First business day prior to
Index (CBOT) (CST) plus 3,6,9,12 ($12.50) Saturday following third Friday
Value Line 8:30-3:15 500 x index/0.05
3,6,9,12 Third Friday
Index (KC) (CST) ($25)
Mini Value Line 8:30-3:15 100 x index/0.05
3,6,9,12 Third Friday
Index (KC) (CST) ($5)
Value-Weighted Arithmetic Indexes

The “value” of the common stocks in a value-weighted index refers to the total market capitalization of the
firm’s outstanding shares, that is, the number of shares outstanding ( ni ,t ) times the current price per share
( pi ,t ). The total market value of the index at time t is therefore
Total market value of index t  i 1 ni ,t pi ,t
N
(7.1)
TABLE 7.2 Stock index futures contract prices at the close of trading on Wednesday,
November 13, 1991.
FUTURES
S&P 500 INDEX (CME) 500 times index
Open High Low Settle Chg High Low Open Interest
Dec 395.00 398.50 394.30 398.30 +1.00 401.50 316.50 139,341
Mr92 396.80 400.50 396.50 400.35 +1.00 404.00 374.70 7,544
June 398.30 402.35 398.30 402.20 +1.10 407.00 379.00 1,102
Est vol 42,125; vol Tues 41,413; Open Int 148,048, + 916.
Index prelim High 397.42; Low 394.01; Close 397.42 +0.68
39
The notation used in this column corresponds to the month of the calendar year (e.g., 1 is January, 2 is February, and so on).
40
All stock index futures contracts are cash settled.
TABLE 7.2 Cont. Stock index futures contract prices

NIKKEI 225 Stock Average (CME)-100 times NSA
Dec 24690 24700 24600 24700 -340 28900 22380 10,969
Mr92 25250 25250 25170 25230 -340 26725 23000 2,423
Est vol 1,107; vol Tues 1,132; open Int 13,292, +467.
The index- High 24814.35; Low 24416.23; Close 24416.23 -251.50
NYSE COMPOSITE INDEX (NYFE) 500 times Index
Dec 218.00 220.10 217.75 220.05 +.70 220.10 175.50 5,026
Mr92 218.80 221.00 218.90 221.00 +.80 221.00 207.60 746
June ... ... ... 222.00 +.80 220.10 208.90 172
Sept ... ... ... 223.00 +.80 221.00 217.50 123
Est vol 5,0571 –vol Tues 5,996; open Int 6,067, +344.
The Index High 219.37; Low 217.64; Close 219.37 +0.37
MAJOR MKT INDEX (CBT) 500 times Index
Nov 323.70 327-40 323.25 327.25 +1.55 327.40 315.20 2,819
Dec 323-50 327.70 323-50 327.70 +1.50 327-70 315.75 746
Est vol 2,500 –vol Tues 1,163 open Int 3,598, +172.
MGMI BASE METAL INDEX (FOX) 100 times Index
Nov ... ... ... 134.50 ... 140.50 132.50 2,246
Dec ... ... ... 134.90 ... 182.70 133.00 8,662
Ja92 ... ... ... 134.90 ... 137.10 132.30 120
Mar ... ... ... 135.30 ... 160.20 132.60 2,643
June ... ... ... 136.00 ... 155.90 134.50 962
Sept ... ... ... 136.80 ... 146.60 134.60 149
Est vol 0; vol Tues 0; open Int 14,712, .
The Index: High 134.52; Low 133.58; Close 134.03 +0.81
OTHERFUTURES
Settlement price of selected contract. Volume and open Interest of all contract months.
KC Mini Value Line (KC) –100 times Index
Dec 328.60 +0.85; Est. vol. 100; Open Int. 254
KC Value Line Index (KC) –500 times Index
Dec 328.30 +0.70; Est. vol. 250; Open Int. 1,722
CRB Index (NYFE) –500 times Index
The Index: High 214.43; Low 213.94; Close 214.20 +.26
CBT –Chicago Board of Trade. CME –Chicago Mercantile Exchange. KCBT –Kansas City Board of
Trade. NYFE –New York Futures Exchange, a unit of the New York Stock Exchange.
Source: Reprinted by permission of Wall Street Journal, © (November 14, 1991) Dow Jones &
Company, Inc. All Rights Reserved Worldwide.
where N is the number of stocks in the index. This market value is then scaled by a divisor so that the index
in period t is

N
n pi ,t
i 1 i ,t
St  (7.2)
Divisort
The divisor represents what the stocks currently in the index would have been worth in a base period. In the
base period, the divisor is the market value of the stocks in the index,
Divisor0  i1 ni , 0 pi ,0
N
(7.3)
Over time, the numerator of (7.2) changes because stocks enter or leave the index or because shares are
issued or repurchased by companies. Because such changes do not reflect a change in the value of the stocks,
an adjustment to the divisor is made on the day that a change in the index composition occurs. The new
divisor on day t is just the old divisor on day t adjusted by the ratio of the market value of the new index
composition on day t divided by the market value of the old index composition on day t ,
 market value new t 

new divisort    old divisort . (7.4)
 market value old t 
Both the S&P 500 and NYSE Composite indexes are value-weighted. The S&P 500 consists of 500 common
stocks, the majority of which trade on the NYSE, although about fifty stocks trade on the American
Exchange and in the OTC market. The index was designed by Standard & Poors’ to contain stocks from a
broad variety of industry groupings. The market value for the base period of the S&P 500 is based on the
average market values of the component stocks during the years 40; 1941 through 1943. At that time, the
index was set equal to 10. The NYSE Composite contains all common stocks traded on the NYSE, slightly
more than 1,500 in number. The base period for the NYSE index is December 31, 1965, at which
time the index was set equal to 50. As Table 7.2 shows, the values of the S&P 500 and NYSE Composite
stocks indexes were 397.42 and 219.37, respectively, at the close of trading on November 13, 1991,
reflecting percentage gains of 3,874 percent and 339 percent, respectively, from their base periods.
Price-Weighted Arithmetic Indexes

A price-weighted arithmetic index is like a value-weighted arithmetic index, except that the number of shares
outstanding does not play a role. The price-weighted arithmetic index is computed as

N
pi ,t
St  i 1
(7.5)
Divisort
In a price-weighted index, the divisor in the base period equals the sum of the prices of the stocks in the base
period, that is,
Divisor0  i 1 pi , 0
N
(7.6)
Like a value-weighted index, the divisor of a price-weighted index is adjusted to reflect stock splits and stock
dividends so that the index level remains unchanged during the stock split/stock dividend process [i.e., in the
manner of (7.4)]. Unlike the value-weighted index, however, the divisor of the price-weighted index is
unaffected by new stock issues or share repurchases.
The best known price-weighted arithmetic index is the Dow Jones Industrial Average (DJIA), which
consists of thirty “blue-chip” stocks. In an attempt to create an index that mimics the price movements of the
DJIA, the American Exchange created the Major Market Index (MMI). This price-weighted index contains
twenty stocks, seventeen of which are also members of the DJIA. Table 7.2 shows that the value of the MMI
at the close of trading on November 13, 1991, was 327.25.
Equal-Weighted Geometric Indexes

An equal-weighted geometric index is somewhat peculiar. To compute it, a geometric average of the rates of
return of the individual stocks within the index over a period ( Ri ,t ) is taken, that is,
RS ,t  N i 1 (1  Ri ,t )  1 .
N
(7.7)
This return is used to update the index from the previous period,
St  St 1 (1  RS ,t ) . (7.8)
Currently, the only equal-weighted geometric index is the Value Line Index. It consists of approximately
1,700 stocks. Approximately ninety percent of the Value Line index capitalization is from shares traded on
the NYSE, one percent from AMEX, and nine percent OTC. The Value Line index and its futures contracts
are of limited interest for two reasons. First, the index weights all stocks equally so small stocks have as
much impact on the index movements as large stocks. For an index to track the behavior of the “market,”
much greater weight should be placed on large capitalization issues. Second, geometric averaging causes the
rate of return on the index to be less than the rate of return that would be earned by an equal-weighted
investment in each of the 1,700 stocks. As a result, price movements (returns) of the Value Line index are
not as strongly correlated with most stock portfolios as are other indexes, which makes the Value Line
futures contract less useful for hedging purposes. Table 7.2 shows that the open interest of the Value Line
futures is much lower than the futures contracts on the other indexes. The Value Line index closed at 326.47
on November 13, 1991.
Stock Index Simulations

The arithmetic versus geometric averaging of the various stock indexes warrants further discussion, and the
discussion is best facilitated through a numerical example. Assume that there are two stocks, A and B, in the
marketplace. Both are priced at $20 per share, and both have 100 shares outstanding. Neither stock pays
dividends. Table 7.3 shows sample paths for the prices of each stock over a twelvemonth period. Alongside
of the stock prices are: (a) a value-weighted arithmetic index, (b) a price-weighted arithmetic index, and (c)
an equal-weighted geometric index corresponding to these two stocks. All the indexes are created to have a
base value of 100 at time 0. The index values are computed using equations (7.2), (7.5), and (7.8),
respectively.
In Table 7.3, note two things. First, the value-weighted and price-weighted arithmetic indexes have
identical values. This is because the simulation begins with equal investments in both stocks (the stocks’
market capitalizations and prices per share are equal). The price movements of these indexes are perfectly
positively correlated with any equal-weighted portfolio of these two common stocks formed at time 0.
Second, the equal-weighted geometric index has a terminal value considerably below the terminal values of
the other two indexes, 134.16 versus 140.00. This is the downward bias discussed earlier. The price
movements of a geometric index in general do not correspond to price movements in a stock portfolio, so
futures contracts on a geometric index are of less value for hedging purposes than are futures contracts on an
arithmetic index.
TABLE 7.3 Simulation of value-weighted arithmetic, price-weighted arithmetic, and equal-weighted

geometric stock index values created using two stocks.
Value-Weighted Price-Weighted Equal-Weighted
Time t Stock A Stock B
Arithmetic Index41 Arithmetic Index42 Geometric Index43
0 20 20 100.00 100.00 100.00
1 25 16 102.50 102.50 100.00
2 30 20 125.00 125.00 122.47
3 33 22 137.50 137.50 134.72
4 27 20 117.50 117.50 116.19
5 36 15 127.50 127.50 116.19
6 40 16 140.00 140.00 126.49
7 36 18 135.00 135.00 127.28
8 38 21 147.50 147-50 141.24
9 40 18 145.00 145.00 134.16
10 38 21 147.50 147.50 141.24
11 40 22 155.00 155.00 148.32
12 36 20 140.00 140.00 134.16
Correlation Among Index Returns

Still more intuition about the different stock indexes can be gathered by examining actual weekly rates of
price appreciation in selected U.S. stock indexes. Table 7.4 contains the means and standard deviations of the
percentage rates of price appreciation of six different stock indexes. Also included in the table are estimated
contemporaneous correlation coefficients between each pair of return series. Weekly returns are computed
using closing index levels each Wednesday during the calendar year 1989. Several interesting results appear
in the table.
First, note that the standard deviation of the rate of return for the arithmetic indexes is highest for
MMI –1.7453 percent per week. This result is not surprising considering that the MMI has the fewest stocks
of any of the indexes examined.
The reduction in standard deviation from the MMI to the DJIA to the S&P 500, and, finally, to the
NYSE reflects increasingly higher degrees of diversification. The DJIA has 30 stocks, the S&P 500 has 500,
and the NYSE has more than 1,500. The standard deviation of the return of the Value Line index reflects
both diversification and a downward bias due to the way in which the index is computed. (Recall the
geometric averaging discussed earlier in this section.)
Second, note that the correlation between pairs of return series is highest for the S&P 500 and the
NYSE indexes –0.9972. Both of these indexes are value-weighted and are highly diversified. The rates of
return of the two stock indexes are virtually perfectly positively correlated. Third, the returns of the MMI and
the DJIA are also strongly positively correlated-0.9779. One would expect this to be the case given that
seventeen of the stocks in the MMI are also in the DJIA. The fact that these indexes are not well-diversified,
however, attenuates to a small degree the correlation between the returns of these two indexes.
Finally, while the correlation among the returns of any pair of arithmetic indexes is very high (at least
0.93), the correlation between the returns of the Value Line index and any of the other indexes is much
lower. The geometric averaging of the returns of the stocks in the Value Line index portfolio and the
inclusion of many small companies undermines the index’s comovements with other indexes.
41
The value-weighted arithmetic index consists of 100 shares of Stock A and 100 shares of Stock B. At time 0, the market
capitalization is 4,000, which is adjusted to an index level of 100.
42
The price-weighted arithmetic index at time 0 equals the sum of the share prices of Stock A and Stock B divided by the divisor.
43
The equal-weighted geometric index equals 100 in the base period. The value at time 1 equals the time 0 index value times the
square root of the product of one plus the rate of return on Stock A and one plus the rate of return on Stock B.
TABLE 7.4 Summary statistics of weekly percentage rates of price appreciation in five U.S. stock
indexes during the calendar year 1989.44
Means and Standard Deviations of Index Returns
Index Mean Return Standard Deviation
DJIA 0.4537 1.6640
MMI 0.5072 1.7453
S&P 500 0.4481 1.5825
VL 0.1809 1.3180
NYSE 0.4127 1.4916
Contemporaneous Correlations Between Pairs of Index Returns
Index MMI S&P 500 VL NYSE
DJIA 0.9779 0.9774 0.8880 0.9750
MMI 0.9497 0.8104 0.9403
S&P 500 0.9137 0.9972
VL 0.9337
7.3 INDEX ARBITRAGE AND PROGRAM TRADING

The cost of carry relation (3.6) from Chapter 3 applies to the relation between the stock index futures price
and the price of the underlying index under the assumption that the dividend yield rate d is a constant,
continuous proportion of the index price level. Active stock index arbitrage ensures that
Ft  St e ( r d )(T t ) , (7.9)
where Ft and St , are the time t prices of the futures contract and the underlying stock index, respectively.
Note that the derivation of this relation in Chapter 3, as it applies to stock index arbitrage, implies that the
cash dividends, as they accrue through time, are being reinvested in the stock index portfolio.
Assuming that cash dividends are a constant, continuous proportion of the index level may be
inappropriate, particularly for a narrow-based index like the MMI, where the small number of stocks in the
index portfolio implies an obvious discreteness and seasonality of cash dividend payments.45 In such a case,
an assumption that the amount Di and the timing ti of the discrete cash dividends paid during the futures
contract life (i.e., between time t and time T ) are known is usually used. Furthermore, rather than assuming
that the dividends are being reinvested in the stock index portfolio, dividends are assumed to be reinvested at
the riskless rate of interest until the futures contract expires.
Under these assumptions, stock index arbitrage involves the transactions shown in Table 7.5a. The
~
long position in the index portfolio provides a terminal value equal to the uncertain index price ST plus a

n
known aggregate dividend income (plus accrued interest) i 1
Di e r (T ti ) . The stock portfolio position is
financed completely with riskless borrowings, which are repaid at time T at a cost St e r (T t ) ). The short futures
~
position has a terminal value  ( ST  Ft ) . Since the arbitrage portfolio involves a zero investment outlay and
has no risk, the net terminal value of the portfolio must equal zero for the market to be in equilibrium. Thus,
under the assumption of known discrete dividends, the cost-of-carry relation is
44
Rates of price appreciation axe computed on the basis of the closing index levels each Wednesday during 1989. Cash dividends
paid on index stocks are not considered.
45
Harvey and Whaley (1992) show pronounced seasonality in the cash dividends of the S&P 100 index, which contains
approximately forty percent of the market value of the S&P 500 index. In particular, during the period 1983 through 1989,
dividends tend to be highest in the months of February, May, August and November.
Ft  S t e r (T t )   i 1 Di e r (T ti )
n
(7.10)
TABLE 7.5a Index arbitrage transactions for establishing the relation between index futures and
underlying index prices, assuming known discrete cash dividends. Ft  St e r (T t )  i 1 Di e r (T ti )
n
Position Initial Value Terminal Value

~
ST  i1 Di e r (T ti )
n
Buy index portfolio  St
Borrow St St  St e r (T t )
~
Sell futures contract 0  ( ST  Ft )
Ft  St e r (T t )  i 1 Di e r (T ti )
n
Net portfolio value 0
A simple version of the cost-of-carry relation arises if one assumes dividends and interest are paid at the end
of the period corresponding to the life of the futures contract. Table 7.5b presents arbitrage transactions for
this case and shows that this simple cost-of-carry relation is
Ft  St (1  r *  d * ) (7.11)
where r * is the rate of interest and d * is the dividend yield over the remaining life of the futures contract.
Violations of the cost-of-carry relation (7.9), (7.10), or (7.11) signal profitable index arbitrage opportunities.
If, for example, the observed futures price is above the theoretical futures price as implied by the right-hand
side of (7.9), (7.10), or (7.11), arbitrageurs sell futures and buy the underlying stocks, driving the price of the
futures down and the prices of stocks up. The arbitrage becomes unprofitable (3 when the futures price
reflects the cost of carrying the underlying stocks, that is, the interest cost less the cash dividends.
TABLE 7.5b Index arbitrage transactions for establishing the relation between index futures and
underlying index prices, assuming dividends and interest are paid at maturity. Ft  St (1  r *  d * )
Position Initial Value Terminal Value
~
Buy index portfolio  St ST  d * S t
Borrow St St  St (1  r * )
~
Sell futures contract 0  ( ST  Ft )
Net portfolio value 0 Ft  St (1  r *  d * )
Unlike typical basis arbitrage, the underlying commodity is a precisely weighted portfolio of
common stocks, rather than a single asset. For example, engaging in index arbitrage with the S&P 500 index
requires a mechanism for buying or selling quickly and simultaneously all 500 stocks in the S&P 500 index
portfolio. Since the simultaneous purchase or sale of the stocks in a precisely weighted and timely fashion is
beyond human capability, computers and computer programs are usually used to place transaction orders as
well as to assist in the execution of those orders. For this reason, trading of portfolios of stocks is called
program trading, although program trades can also be done by manually preparing order tickets for each
stock. NYSE statistics define a program trade as any order for a portfolio of 15 or more stocks.
Treasury Bill Substitute

Technically speaking, one thinks of “index arbitrage” as being conducted by professional index arbitrageurs
who establish offsetting positions in the manner shown in Table 7.5a. However, deviations from the cost of
carry relation also offer opportunities for investors, such as pension funds, to structure an investment with
index futures that offers a higher return than an investment of equivalent risk in another market. For
example, if the futures price is high relative to the cost-of-carry equilibrium, fund managers can generate a
riskless investment with a rate of return higher than the return on a Treasury bill of maturity comparable to
the index futures by selling index futures and buying the index portfolio. Such a strategy is called a Treasury
bill substitute.
To understand how this strategy works, suppose that the current S&P 500 index level is 348.60 and
that the nearby S&P 500 futures contract has a price of 354.50 and a time to expiration of 73 days. Suppose
also that the future value of the S&P 500 dividends over the next 73 days is $2.79 and that a 73-day Treasury
bill will provide a rate of return of 1.6 percent over its life. Using (7.11), the implied riskless rate of interest, ,
r * on a 73-day investment involving selling the index futures and buying the stock index portfolio is
determined by solving
354.50  348.60(l  r * ) - 2.79 .
The interest rate from the Treasury bill substitute strategy, r * , is 2.5 percent. In other words, a pension fund
that might ordinarily invest $3,486,000 in T-bills to earn 1.6 percent over 73 days could invest the same
amount of money in a Treasury bill substitute to earn 2.5 percent over 73 days. To do so, the $3,486,000 is
invested in the index portfolio (i.e., 10,000 units of the index are purchased) and twenty index futures
contracts are sold (recall each index futures is 500 times the index value). Over the 73-day period, the index
portfolio will generate $27,900 in cash dividends and the index level will appreciate by 5.90 relative to the
futures (because the futures price and index level converge at the end of 73 days), for a total price
appreciation of $59,000. The overall rate of return on the Treasury bill substitute position is (27,900 +
59,000)/3,486,000 or 2.5 percent.
Stock Replacement
A second example of how index futures may be used to generate a higher return than an investment with
equivalent risk is a stock replacement strategy. When the actual futures price is below the theoretical futures
price, an arbitrageur enacts a short arbitrage –the short sale of stocks and the purchase of futures contracts.
But stock portfolio managers, too, can profit from such an opportunity by selling their stock portfolios and
using the proceeds to buy index futures and Treasury bills, that is, by engaging in stock replacement.
To illustrate a stock replacement strategy, consider the previous example in which the current S&P
500 index level is 348.60, the time to expiration of the nearby S&P 500 futures contract is 73 days, the rate
of return on a 73-day T-bill over the next 73 days is 1.60 percent, and the future value of the cash dividends
on the S&P 500 over the next 73 days is $2.79. However, this time, assume the nearby S&P 500 futures price
is $350.25. On the basis of these figures, the theoretical futures price is
F  348.60(l.0160) - 2.79  351.39 .

Since the observed futures price, $350.25, is less than its theoretical value, a stock replacement strategy can
be used to generate a rate of return that will exceed the rate of return on a direct investment in the S&P 500
index portfolio without assuming more risk. A portfolio manager with $50,000,000 in the S&P 500 index
portfolio will have a portfolio value of
~  50,000,000  ~ ~
VS &P 500,T   ( ST  2.79)  143,430.87 ST  400,172
 348.60 
in 73 days. On the other hand, if he liquidates his S&P 500 stock portfolio and buys T-bills and the nearby
S&P 500 futures contract, the portfolio value for the stock replacement strategy (SRS) will be
~  50,000,000  ~
VSRS ,T   ( ST - 350.25)  50,000,000(l.016)
 348.60 
~ ~
 143,430.87 ST - 50,236,662  50,800,000  143,430.87 ST  563,338 .
Note that the stock replacement strategy is certain to have a terminal value $163,166 higher than the stock
portfolio strategy. The fact that this incremental value is certain reflects the fact that, while each strategy’s
terminal value is uncertain, both strategies have equal risk. If the observed futures price is below its
theoretical level, however, the stock replacement strategy will dominate.
Practical Considerations in Index Arbitrage

In practice, there are several reasons why deviations from the cost of carry relation do not ensure that
arbitrage profits can be earned. First, and most important, are the transaction costs involved in trading the
underlying index stocks. These include the commissions and the market impact costs of buying stocks at the
ask price or selling stocks at the bid rice. Procedures for trading portfolios of stock have improved
dramatically in recent years and frequently involve the use of the NYSE computer entry system, DOT
(Designated Order Turnaround). Nevertheless, these costs can be substantial, particularly if a number of
portfolio transactions are hitting the market at the same time. Stock index arbitrageurs estimate the total
round-trip transaction costs to be on the order of 0.5 to 0.75 percent of the underlying portfolio value.46
Second, the dividends in the cost-of-carry relation are assumed to be known with certainty. In general, this
assumption is reasonable since firms tend to pay regular, constant, or constantly increasing quarterly
dividends. Any uncertainty about the anticipated dividend payments on the underlying stocks, however,
introduces uncertainty about the return of the index arbitrage and can therefore limit arbitrage somewhat.
Third, certain types of arbitrage may involve risk. In some cases, arbitragers do not trade all the underlying
stocks in the index. Instead, they buy or sell a representative basket of stocks because of the difficulty and
the cost associated with transacting, say, all 500 of the stocks in the S&P 500 index. If the representative
basket fails to move exactly like the underlying index, the arbitrage is risky.
Fourth, certain rules and regulations can impede arbitrage. For example, “circuit breakers” are now
used to suspend index futures trading when the DJIA moves by more than a pre-specified amount in a given
trading day. On such days, apparent arbitrage opportunities may be only illusory in the sense that the futures
leg of the arbitrage may not be executable. Another example of an instance where a rule impedes arbitrage is
when the arbitrage requires stocks to be sold and futures to be purchased. Since the index portfolio must be
sold short, the short-sale rule comes into play. Under the short-sale rule, a stock is required to uptick before it
may be sold short. When an entire portfolio of stocks must be sold, the time delay in waiting for an uptick in
each stock makes the short arbitrage difficult to implement, so the futures price may tend to be less than or
equal to its theoretical value. It is worthwhile to note that stock sales conducted by portfolio managers using
stock replacement strategies, however, are not subject to the uptick rule, and this will tend to limit the
amount by which the futures price will fall below its theoretical price.47
Fifth, arbitrage is sometimes limited by the lack of capital. Brokerage firms may be limited by net
capital requirement rules and the availability of higher yielding alternative fund uses. Moreover, many
institutional investors may not be authorized to engage in index arbitrage.
46
See Stoll and Whaley (1987, p. 18).
47
In August 1990, the NYSE implemented a rule requiring a downtick on each stock in an index arbitrage program purchase if the
DJIA rose by 50 points or more and an uptick on each stock in an index arbitrage program sale (short or from a long position) if
the DJIA declined by 50 points. This rule is counter productive because it impedes index arbitrage.
The efficacy of the index arbitrage process has been examined in a number of theoretical and
empirical papers.48 In general, these papers find that observed futures prices can deviate from the theoretical
futures price specified by arbitrage conditions by more than normal transaction costs. This is particularly the
case for deviations of the futures price below the theoretical price. Such deviations may be difficult to
arbitrage though because of the short sale restrictions and because of the lack of a sufficient number of
institutions willing to engage in stock replacement strategies.
7.4 INTRADAY BEHAVIOR OF RETURNS

In perfectly efficient and continuous futures and stock markets absent transaction costs, riskless arbitrage
profit opportunities should not appear so the cost-of-carry relation (7.9),
Ft  St e ( r d )(T t ) ,
should be satisfied at every instant t during the futures contract life. If such is the case, the instantaneous
rate of price appreciation in the stock index equals the net cost-of-carry of the stock portfolio plus the
instantaneous relative price change of the futures’ contract. To see this, take the natural logarithm of (7.9) at
time t and at time t  1 :
ln St  (r  d )(T  t )  Ft (7.12)
and ln St 1  (r  d )(T  t  1)  Ft 1 (7.13)
and then subtract (7.13) from (7.12),

RS ,t  (r  d )  RF ,t (7.14)
where RS,t  ln( St /St-1 ) and RF,t  ln( Ft /Ft-1 ) .
Several implications follow from (7.14) under the assumptions that the short-term interest rate and the
dividend yield rate of the stock index are constant and that the index futures and stock markets are efficient
and continuous:
~
a. The expected rate of price appreciation on the stock index portfolio E ( RS,t ) equals the net cost of
~
carry ( r  d ) plus the expected rate of return on the futures contract E ( RF,t )
b. The standard deviation of the rate of return on the futures contract equals the standard deviation of
the rate of return of the underlying stock index.
c. The contemporaneous rates of return of the futures contract and the underlying stock portfolio are
perfectly positively correlated.
d. The rates of return of the futures contract and of the underlying stock index portfolio are serially
uncorrelated.49
e. The non-contemporaneous rates of return of the futures contract and the underlying stock portfolio
are uncorrelated.
48
Cornell and French (1983), Figlewski (1984a), Gastineau and Madansky (1983), Modest and Sundaresan (1983), Peters (1985),
Stoll and Whaley (1986b), MacKinlay and Ramaswamy (1988), Kleidon (1991), Kleidon and Whaley (1991), and Miller,
Muthuswamy, and Whaley (1991) examine the arbitrage process and consider possible explanations for observed deviations from
theoretical prices. Other papers, notably Garcia and Gould (1987), Gould (1988), and Brennan and Schwartz (1990), analyze
strategies for trading on mispricing.
49
Technically speaking, more than an assumption of market efficiency is needed to ensure serially uncorrelated rates of return. It
must also be the case that the expected rates of return of the futures and stock index are constant. [See Fama (1976, pp. 149-151).]
Such an assumption is reasonable since the rate of return series that we will examine below are intraday.
Naturally, all of the above implications are based on the assumption that the cost-of-carry relation (7.9) holds
at all points in time. It has been shown, however, that (7.9) does not hold exactly; indeed, one of the puzzles
in stock index futures is the frequency with which deviations from the cost-of-carry relation are observed.
Stoll and Whaley (1986b, Table 23A), for example, report frequent violations of the cost-of-carry relation in
excess of transaction costs using hourly S&P 500 index and index futures data during the period April 1982
through December 1985. The frequency of violation is nearly eighty percent for the June 1982 futures
contract. For more recent contract maturities, however, the frequency falls below fifteen percent. MacKinlay
and Ramaswamy (1988, Table 6) report similar results for the S&P 500 futures contracts expiring in
September 1983 through June 1987. Using fifteen-minute price data, they find that the cost-of-carry relation
is violated 14.4 percent of the time on average.
Violations of the cost-of-carry relation may appear for a variety of reasons. Some, like transaction
costs, were discussed in the last section. The presence of transaction costs tends to introduce noise in the rate
of return relation (7.14). An important reason not mentioned in the last section is the infrequent trading of
stocks within the index. Markets for individual stocks are not perfectly continuous. Consequently, stock
index prices, which are averages of the last transaction prices of component stocks, lag actual developments
in the stock market. Fisher (1966) describes this phenomenon. Cohen, et (1986, Ch. 6) give a more general
discussion of serial correlation of stock index returns in terms pf delays in the price adjustment of securities.
Lo and MacKinlay (1988) model the effects of infrequent trading on index returns under certain restrictive
assumptions. Assuming that the index futures prices instantaneously reflect new information, observed
futures returns should be expected to lead observed stock index returns because of infrequent trading, even
though there is no economic significance to this behavior whatsoever.
Stoll and Whaley (1990b) use five-minute, intraday rate of return data for the S&P 500 index and the
nearby S&P 500 futures contracts to (a) model and purge the effects of infrequent trading in the stock index
portfolio, and (b) assess the degree of simultaneity between returns in the index futures and stock markets.
The effects of infrequent trading are shown in Table 7.6. Note that, while the S&P 500 futures contract
returns have virtually no serial correlation, the returns of the S&P 500 index portfolio are strongly positively
serially correlated. The first-order serial correlation in the S&P 500 index returns exceeds 0.5. Because not
all stocks within the S&P 500 index portfolio trade in every five-minute interval, a market movement within
this interval may not be recorded in the price of less actively traded stocks until some time later when the
stock finally trades. The effect of this phenomenon is positive serial correlation in the portfolio return series.
The serial correlation does not disappear until lag 4 or 5 using five-minute returns.
The effects of infrequent trading on observed stock index returns are modeled theoretically and
estimated empirically in Stoll and Whaley (1990b). The residuals (return innovations) from the estimated
model are examined to assess the degree of any remaining positive serial correlation. The last pair of
columns in Table 7.6 show these results. With the effects of infrequent trading modeled and purged, the
return innovations of the S&P 500 index are virtually white noise. None of the estimated serial correlation
coefficients exceed 0.02 in absolute magnitude.
Finally, to assess the degree of simultaneity between the S&P 500 index futures and stock market returns, the
return innovations of the S&P 500 index are regressed on lag, contemporaneous, and lead futures returns,
 S ,t  k 3  k RF ,t k  ut .
3
(7.15)
The regression results are shown in Table 7.7. In addition, for purposes of comparison, the regression results
of observed S&P 500 index returns regressed on lag, contemporaneous, and lead futures returns are also
reported.
The return innovation regression results in Table 7.7 indicate that the dominant relation between the
two markets is contemporaneous. The estimated coefficient of the contemporaneous futures return, ˆ0 , in
the return innovation regression is 0.1338, higher than any of the leading or lagged coefficients. The
estimated coefficient of the lag one futures return, ̂1 , is 0.1015, showing that there is a tendency for the
futures market to lead the stock market. All other coefficients in the return innovation regression are
indistinguishably different from zero in an economic sense. When stock index returns are used as the
dependent variable, the leading effect of the futures market appears considerably longer, but most of this is
illusion attributable to infrequent trading in the stock market. Overall, the evidence supports the notion that
futures markets tend to play a price discovery role in the marketplace.
TABLE 7.6 Estimated serial correlation coefficients of observed returns of the S&P 500 index ( RSo,t )
and the S&P 500 index futures contract ( RFo ,t ) for the 1249-day period April 21, 1982, through March
31, 1987.50
 k ( RSo,t , RSo,t k )  k ( RFo ,t , RFo ,t k )  k ( S ,t ,  S ,t k )
Lag k No. of ̂ k 52 t ( ˆ k ) 53 ̂ k 18 t ( ˆ k ) 19 No. of ̂ k 18 t ( ˆ k ) 19
Obs.51 Obs.54
1 86,952 0.5117 175-61 0.0229 6.77 84,454 0.0071 2.06
2 85,703 0.2654 80.60 0.0265 7.76 83,205 0.0053 1.52
3 84,454 0.1312 38.46 0.0015 0.45 81,956 0.0068 1.95
4 83,205 0.0759 21.96 -0.0137 -3.96 80,707 0.0050 1.41
5 81,956 0.0460 13.17 -0.0222 -6.36 79,458 0.0052 1.48
6 80,707 0.0199 5.64 -0.0108 -3-06 78,209 -0.0042 -1.18
7 79,458 0.0077 2.18 -0.0087 -2.46 76,960 -0.0119 -3.30
8 78,209 0.0154 4.32 -0.0015 -0.42 75,711 0.0017 0.46
9 76,960 0.0195 5.42 0.0039 1.07 74,462 -0.0005 -0.15
10 75,711 0.0110 3.04 -0.0030 -0.83 737213 -0.0082 -2.22
11 74,462 0.0018 0.49 0.0047 1.29 71,964 -0.0163 -4.37
12 73,213 0.0019 0.51 0.0002 0.07 70,715 -0.0067 -1.77
7.5 HEDGING MARKET RISK

Stock index futures contracts are useful in a variety of risk management situations. In this section, we
examine an important one-hedging market risk. Assume you are responsible for managing a $50,000,000
stock portfolio. This portfolio has a systematic risk coefficient (  p ) of 1.20 relative to the S&P 500 index
and a total risk (  p ) of forty percent on an annualized basis. The future value of the promised dividends on
this stock portfolio over the next three months is $400,000, or 0.8 percent of the current portfolio value. At
the same time, the S&P 500 stock index portfolio has a total risk level (  S ) of twenty-five percent annually
and promises cash dividends over the next three months amounting to one percent of the current index value.
The current S&P 500 index value is 373.63 and the price of the nearby, three-month S&P 500 futures
contract is 375.50. A three-month T-bill promises a 1.5 percent rate of return. This illustration assumes the
cost of carry relation (7.11) holds, that is, 375.50 = 373.63(l + 0.015 - 0.01).
50
The numbers in this table are taken from Stoll and Whaley (1990, Tables 1 and 3).
51
The number of observations used in the computation of the serial correlation coefficient. Note that as the lag k is incremented
by one, the number of observations lost equals the number of days in the sample period. This reflects the loss of one return each
day of the sample. The serial correlation coefficient estimates are, therefore, not contaminated by using returns from adjacent days.
52
The estimated lag k serial correlation coefficient across all five-minute returns in all days of the period, excluding overnight
returns and the first two returns each trading day.
53
The t - ratio corresponding to the null hypothesis that  k equals zero.
54
The number of observations drops by 2,498 as a result of fitting an ARMA(2,3) regression model to observed returns.
TABLE 7.7 Parameter estimates from regressions of S&P 500 index returns/return innovations on lag,
contemporaneous, and lead nearby S&P 500 futures returns for the 1249-day period April 21, 1982
through March 31, 1987.55
Returns : RS ,t    k 3  k RF ,t k  ut and Return Innovations  S ,t     k 3  k RF ,t  k  ut
3 3
Returns Returns Innovations

No. of Obs. 78,209 78,209
R2 0.4730 0.2132
Parameter
Parameter Estimate56 t - ratio 57 t - ratio
Estimate
̂ -0.0001 -1-08 -0.0002 -1.73
ˆ3 -0.0077 -6-57 -0.0094 -8.04
ˆ 2 -0-0158 -13.48 -0.0153 -13.04
ˆ1 0.0213 18.10 0.0194 16.54
ˆ0 0.1690 142.93 0.1338 113.50
̂1 0.2032 171.14 0.1015 85.72
̂ 2 0.1330 111.45 0.0153 12.87
̂ 3 0.0798 66.-50 0.0059 4.92
Suppose that your research director has informed you that the market (as reflected by the S&P 500)
will drop by sixteen percent over the next three months. You have a great deal of confidence in his prediction
so you decide to hedge the market risk of your portfolio. One option that you have is to liquidate the stock
portfolio and buy T-bills, however this strategy would not allow you to capture the non-market returns that
your portfolio of “winners” is expected to earn over the next three months. Selling S&P 500 futures
contracts, on the other hand, allows you to hedge the market risk of the stock portfolio without selling your
stocks.
Forming the Hedge Portfolio

The optimal number of futures contracts to sell in this instance can be obtained indirectly using the stock
portfolio beta  p  1.20 implies that the stock portfolio is expected to earn 1.2 times the gain/loss of the S&P
~ ~
Cov( RP , RS )
500 index per dollar invested. The stock portfolio beta is defined as  p  ~
Var ( RS )
~ ~
where RP and RS are the random rates of return on the stock portfolio and the market index (in this case, the
S&P 500), respectively. To understand the relation between the stock portfolio beta and the optimal hedge
ratio, we need to establish the relation between the futures and stock index returns over the hedge period,
which is equal to the futures contract life in this illustration. Over the hedge period, the stock
index return is
~ ~
~ ST  S 0 ~ FT  F0
RS  and the futures return is RF  .
S0 F0
55
The numbers in this table are taken from Stoll and Whaley (1990b, Table 5).
56
Parameter estimates obtained from times series regression across all five-minute returns in all days of the period, excluding
overnight returns and the first two returns each trading day.
57
The t - ratio corresponding to the null hypothesis that the respective coefficient equals zero.
Using (7.11) to substitute for Ft
~ ~
~ S  S 0 (1  r *  d * ) RS r*  d *
RF  T   .
S 0 (1  r *  d * ) (1  r *  d * ) (1  r *  d * )
~ ~ ~
Rearranging to isolate RS , we get: RS  RF (1  r *  d * )  r *  d * .
Substituting for the stock index return, the expression for the stock portfolio beta becomes
~ ~ ~ ~
Cov[ R p , RF (1  r *  d * )  r *  d * ] Cov[ R p , RF ]
p  ~  ~ .
Var[ RF (1  r *  d * )  r *  d * ] Var[ RF ](1  r *  d * )
The remaining step in showing the relation between the stock portfolio beta and the hedge ratio involves
~ ~
substituting the relations between returns and price changes. These relations are R p   p / p0
~ ~
and RF   F / F0 . Hence, the stock’s rate of return beta  p is.
1 ~ ~
~ ~ Cov( p ,  F ) ~ ~
Cov( p / p0 ,  F / F0 ) p0 F0 F0 Cov( p ,  F )
p  ~   ~
Var ( F / F0 )(1  r *  d * ) 1 ~ p Var (  F )(1  r  d )
* *
2
Var ( F )(1  r  d )
* * 0
F0
Using the definition of the optimal hedge ratio given in Chapter 4 and assuming the cost-of-carry relation,
S
(4.9),  p can be written as  p   h* 0 .
p0
Finally, the initial investment in the stock portfolio and the cash index with respect to which S 0 is calculated
are the same, so S 0  p0 . This implies that  p  h*
In other words, the optimal hedge ratio is the negative of the stock portfolio beta. In the case of the example,
the optimal hedge ratio is
h*  1.2
The optimal number of futures contracts to sell is therefore the stock portfolio beta times the number of units
 50,000,000 
of the stock portfolio, 1.2    321.17
 373.63(500) 
Assessing Hedging Effectiveness

The information indicates that the variance of the unhedged stock portfolio return is 0.40 2  0.16 . If the
stock portfolio investment is one dollar, the price change variance of the unhedged portfolio is also 0.16. In
Chapter 4, we learned that hedging effectiveness is measured by the adjusted R-squared of the regression of
cash price changes on futures price changes. The R-squared, in turn, is closely related to the correlation of
cash price changes with futures price changes ( i.e., R 2   p2 ,F ). To find the effectiveness of the S&P 500
hedge in our illustration, therefore, we focus on the correlation coefficient between the stock portfolio price
changes and the futures price changes. Over the life of the futures contract, the futures and stock index price
changes are perfectly correlated and the standard deviation of the futures price change equals the standard
deviation of the stock index price change, so the correlation coefficient may be written
 p ,S
 p , F   p ,S  .
 p S
Also, we know that  p   p ,S /  2p , so
On the basis of the given values (  p  1.20,  S  0.25 and  p  0.40 ); the correlation coefficient,
ρ p,F , is 0.75 . The R-squared is thus 0.5625 and the proportion of the stock portfolio return variance that is
unrelated to the return variance of index futures is 1 - 0.5625 = 0.4375. The remaining variance of the rate of
price change on the hedged portfolio is therefore
 h2  0.4375(0.16)  0.07
Decomposing the Hedge Portfolio Return

Suppose that the S&P 500 index drops by twenty percent over the three-month period after the hedge
portfolio is formed. Over the same time, your stock portfolio drops to a value of $40,000,000, excluding
dividends. Find the overall rate of return on your hedged portfolio, and decompose the overall return into its
riskless rate and abnormal return components. Table 7.8 provides such a decomposition for a hedge
established on September 15 and liquidated on December 15, when the futures contract is assumed to expire.
TABLE 7.8 Hedging market risk of a stock portfolio that has a 1.2 and a three-month dividend yield of
0.8 percent.
Cash Market December Futures
Value of
Value of Stock Value of Hedged
Index Level Futures Price Future’s
Portfolio Portfolio
Position58
Sept 15 373.63 50,000,000 375.50 -60,300,000
Dec 15 298.90 401400,00059 298.90 -48,000,000
Gain -9,600,000 12,300,000 2,700,000
60 26
Return (%) -20.00 -19.20 -20.40 -24.60 5.4026
The overall rate of return on the unhedged portfolio can be measured easily by focusing on the price
appreciation and dividend yield components of total return, that is,
 40,000,000   400,000 
Rp    1     19.20%
 50,000,000   50,000,000 
To find the hedged portfolio return, we must also compute the rate of return on the futures. At the outset, the
S&P 500 index level was 373.63 and the three-month S&P 500 futures price was 375.50. If the S&P 500
58
The optimal hedge involves selling 321.17 futures contracts, with each contract valued at 500 times the index futures price.
59
Includes dividends of $400,000.
60
Dollar gain divided by the initial stock portfolio value, $50,000,000.
index level fell by twenty percent over the three-month period, the new index level and futures price (recall
that futures had three months to expiration when they were sold) are 373.63(0.80) or 298.90. The rate of
return on the index futures over the period was therefore
298.90
RF   1  20.40%
375.50
Thus, the total return of the hedged portfolio over the three-month period is
F   375.50 
Rh  R p  h 0  RF  - 0.1920 - 1.2 ( - 0.2040)  5.40%
S
 0  373.63 
The alternative to hedging in this example is to liquidate the stock portfolio and buy three-month T-bills.
Such an action would have produced a 1.5 percent return, given our assumption that the T-bill rate is 1.5
percent. The riskless rate and abnormal performance components of the hedge portfolio return in this
illustration are therefore 1.5 percent and 3.9 percent, respectively. In other words, the 3.9 percent return was
the abnormal or extra-market rate of return arising from the fact that the portfolio of “winners” outperformed
the market on a risk-adjusted basis.
The hedged portfolio would also have earned 1.5 percent if the stock portfolio had declined exactly
according to its beta of 1.20, without an abnormal return. In that case, the return would have been
R p  r *  ( Rm  r * )  p  0.015  (-0.19 - 0.015)1.2  - 0.231 ,
where Rm is the return on the stock index including the dividends, or -0.19. That implies a value for the
stock portfolio, including dividends, of $38,450,000, instead of the value of $40,400,000 shown in Table 7.8.
The values in Table 7.8 for the cash index and the futures market would remain the same. The dollar gain on
the hedged portfolio becomes $750,000, and the hedged return becomes 1.5 percent, exactly the same as the
riskless rate.
It is worth noting that the hedged stock portfolio has basis risk because the portfolio’s return is not
perfectly correlated with the index futures return. If, for example, the stock portfolio had a negative abnormal
return, the hedged portfolio would have earned less than the riskless rate.
7.6 SUMMARY
In this chapter, stock index futures contract index and the composition of stock indexes underlying futures
contracts are described. The cost-of-carry relation for stock indexes is derived, and the role of index arbitrage
in maintaining the link between stock index futures and cash prices is explained. Evidence on the short-run
behavior of the returns of index futures and of the cash index is presented. Finally, the use of stock index
futures to hedge the market risk in a stock or a portfolio of stocks is illustrated in detail.
8 INTEREST RATE FUTURES
8.1 INTEREST RATE FUTURES MARKETS
8.2 UNDERLYING BONDS

 Zero-Coupon or Discount Bonds
 Coupon Bonds
8.3 INTEREST RATE RISK

 Duration
 Convexity
8.4 TERM STRUCTURE OF INTEREST RATES

 Yield Curve
 Spot Rates and Forward Rates
8.5 SHORT-TERM INTEREST RATE FUTURES CONTRACTS
8.6 LONG-TERM INTEREST RATE FUTURES CONTRACTS

 Conversion Factor and Invoice Price
 Cheapest to Deliver
8.7 COST-OF-CARRY RELATION
8.8 HEDGING WITH INTEREST RATE FUTURES CONTRACTS

 Short-Term, Long Hedge
 Long-Term, Short Hedge
 Asset Allocation
8.9 SUMMARY
8 INTEREST RATE FUTURES
Like stock index futures contracts, interest rate futures have been extremely successful financial futures
contracts. Interest rate futures provide a means of trading future loan commitments and are important and
useful risk management tools for fixed-income portfolio managers. This chapter begins with a description of
interest rate futures markets and the specifications of the contracts traded in those markets. Section 2
analyzes the instruments underlying interest rate futures contracts. In section 3, a key measure of interest rate
risk, duration, is explained. In section 4, the term structure of interest rates is examined. We focus on the
effect of the term structure on the value of coupon and non-coupon bonds. Spot rates, forward rates, and
yield to maturity are defined. Examples illustrate most of the important concepts. Sections 5 and 6 describe,
in detail, the interest rate futures contracts corresponding to short-term and long-term interest rates. The most
active short-term interest rate futures contracts in the U.S. are the Chicago Mercantile Exchange’s Treasury
bill and Eurodollar contracts, while the most active long-ten-n interest rate futures contract is the Chicago
Board of Trade’s Treasury bond contract. Contracts in the intermediate-term range, the CBT’s five-year and
ten-year Treasury note futures, are also discussed. Section 7 describes the cost-of-carry equilibrium in the T-
bond futures market. Section 8 contains three illustrations of the use of interest rate futures contracts, and
section 9 contains a summary.
8.1 INTEREST RATE FUTURES MARKETS

Table 8.1 contains the contract specifications of the most active interest rate futures contracts traded in the
U.S. Of the contracts listed in Table 8.1, the CME’s T-bill futures contract is the oldest, with trading
beginning in January 1976. The CBT had introduced a futures contract on GNMA pass-through certificates
in the fall of 1975, but it was later delisted.
TABLE 8.1 Interest rate futures contracts specifications (most active contracts in U.S. markets).
Units/Minimum
Security Trading Contract Last Day of
Price Deliverable Grade62
(Exchange) Hours Months61 Trading
Fluctuation
Nominal 8% coupon
Seven business days
T-Bond 8:00-2:00 $100,000/ with 15 years to
3,6,9,12 prior to last business
(CBT) (CST) 1/32 ($31.25) maturity or first call
day of month
date
10-yr. Seven days prior to Nominal 8% coupon
8:00-2:00 $100,000/
T-Note 3,6,9,12 last business day of with 6.5 to 10 years to
(CST) 1/32 ($31.25)
(CBT) month business maturity
Seven business days Any of the four most
5-yr. T-Note 8:00-2:00 $100,000/
3,6,9,12 prior to last business recently auctioned 5-
(CBT) (CST) 1/32 ($31.25)
day of month year Treasury notes
Business day
91-day T-Bill 7:20-2:00 $1,000,000/ Any of the three T-bills
3,6,9,12 preceding issue date
(CME) (CST) 0.01 ($25.00) 91-days from maturity
of new 91-day T-bill
3,6,9,12
Eurodollar Second London
7:20-2:00 plus $1,000,000/
Time Deposit business day before Cash settled
(CST) current 0.01 ($25.00)
(CME) third Wednesday
month
61
62
All interest rate futures contracts other than the Eurodollar contract call for delivery of the underlying security.
Unlike stock index futures contracts, most interest rate futures contracts call for the delivery of the
underlying interest rate instrument. The CBT’s T-bond futures contract, for example, requires the delivery of
a nominal eight-percent coupon, $100,000 face value, U.S. Treasury bond. Delivery may take place at any
time during the delivery month, at the discretion of the short. The last day of trading of the futures contract is
the eighth-to-last business day of the contract month.
Table 8.2 shows prices of these contracts as of the close of trading on November 13, 1991. The open
interest figures in Table 8.2 show that the T-bill, Eurodollar, and T-bond futures contracts are the most active
interest rate futures contracts presently trading. These contracts are the focus of the dominant part of this
chapter.
TABLE 8.2 Interest rate futures contract prices.

INTEREST RATE INSTRUMENTS
FUTURES
Yield Open
Open High Low Settle Chg Chg
Settle Interest
TREASURY BONDS (CBT)-$100,000; Pts. 32nds of 100%
Dec 101-04 101-05 99-15 100-03 -29 7.991 +0.091 280,772
Mr92 101-04 100-10 98-20 99-07 -30 8.079 +0.095 31,420
June 99-12 99-12 97-30 98-10 -30 8.173 +0.097 10,197
Sept 98-13 98-13 97-06 97-14 -31 8.264 +0.101 2,717
Dec 97-19 97-19 96-10 96-20 -31 8.350 +0.102 4,487
Mr93 96-00 96-05 95-28 95-28 -32 8.430 +0.107 511
Est vol 370,000; Vol Tues 245,19 1 ; op Int 330,091, +8,975.
TREASURY BONDS (MCE)-$100,000; Pts. 32nds of 100%
Dec 101-02 101-02 99-15 100-04 -32 7.987 +0.100 13,542
Est Vol 6,600; Vol Tues 6,396; open lnt 13,641, -132.
T-BONDS (LIFFE) U.S. $100,000; pts of 100%
Dec 101-02 101-03 99-21 100-05 0-23 101-03 96-24 5,443
Est Vol 2,273; Vol Tues 4,066; open Int 5,480, +721.
GERMAN GOV'T. BOND (LIFFE) 250,000 marks; $ per mark (0.01)
Dec 86.73 86.25 86.02 86.19 +0.02 86.44 83.73 75,176
Mr n.a. n.a. n.a. n.a. n.a. 86.70 85.39 6,879
Est Vol 47,392; Vol Tues 51,218; open lnt 82,055, -2,487.
TREASURY NOTES (CBT)-$100,000; pts. 32nds of 100%
Dec 103-23 103-73 102-18 103-06 -14 7.540 +0.061 86,289
Mr92 102-29 102-29 101-26 102-13 -14 7.651 +0.062 12,194
June … … … 101-18 -13 7.772 +0.058 418
Est Vol 30,000; Vol Tues 25,215; open Int 98,902, +4,714.
5 YR TREAS NOTES (CBT)-$100,000; pts. 32nds of 100%
Dec 04-275 104-28 104-12 04-215 -5.5 6.880 +0.040 91,919
Mr92 04-015 04-015 103-19 03-275 -6.0 7.071 +0.045 10,105
2 YR TREAS NOTES (CBT)-$20,000, pts. 32nds of 100%
Dec 103-26 103-26 103-17 03-255 -1/4 … … 13,800
Mr92 103-11 103-11 03-057 03-105 -1/2 … … 3,553
Est Vol 1,500; Vol Tues 895; open lnt 17,353, +153.
TABLE 8.2 Cont. Interest rate futures contract prices.

FUTURES
Yield Open
Settle Interest
30-DAY INTEREST RATE (CBT)-$5 million; pts. of 100%
Nov 95.14 95.14 95.13 95.14 -0.01 4.86 +0.01 1,254
Dec 95.12 95.12 95.08 95.09 -0.05 4.91 +0.05 1,232
Ja92 95.16 95.17 95.15 95.17 -0.05 4.83 +0.05 1,100
Feb 95.23 95.27 95.23 95.26 -0.04 4.74 +0.04 962
Mar 95.18 95.21 95.18 95.21 -0.05 4.79 +0.05 570
Apr 95.20 95.20 95.20 95.20 -0.05 4.80 +0.05 107
June 95.10 95.11 95.10 95.11 -0.04 4.89 +0.04 189
Est Vol 725; Vol Tues 529; open lnt 5,464, +185.
TREASURY BILLS (IMM)-$l mil.; pts. of 100%
Dec 95.38 95.38 95.31 95.34 -0.03 4.66 +0.03 21,996
Mr92 95.52 95.52 95.42 95.50 -0.02 4.50 +0.02 79,371
June 95.23 95.34 95.23 95.32 -0.03 4.68 +0.03 3,8912
Sept 95.09 95.09 95.05 95.06 -0.02 4.92 +0.02 338
Dec 94.64 94.64 94.64 94.64 … 5.36 … 156
Est vol 6,378; Vol Tues 5,830; open lnt 55,771, + 261.
LIBOR-1 MO. (IMM)-3,000,000; points of 100%
Nov 95.05 95.05 94.98 95.02 -0.04 4.918 +0.04 6,950
Dec 94.60 94.60 94.50 94.55 -0.10 5.45 +0.10 7,336
Ja92 95.11 95.11 95.02 95.06 -0.05 4.912 +0.05 9,651
Feb 95.01 95.08 94.99 95.06 -0.05 4.94 +0.05 2,515
Mar 94.96 95.01 94.94 95.00 -0.05 5.00 +0.05 1,390
Apr … … … 95.01 -0.04 4.99 +0.04 163
Est vol 1,577; vol TL@es 2,06,3; open lnt 78,005, + 300.
Open
Interest
MUNI BOND INDEX (CBT)-$1,000; times Bond Buyer MBI
Dec 95-23 95-23 95-01 95-06 15 95-25 88-16 12,542
Mr92 95-04 95-04 94-07 94-13 19 95-04 88-00 827
The Index: Close 95-09; Yield 6.82.

EURODOLLAR (IMM)-$1 million; Pts of 100%
Yield Open
Settle Interest
Dec 94.85 94.8.6 94.75 94.80 -0.07 5.20 +0.07 242,049
Apr92 94.99 94.99 94.84 94.94 -0.05 5.06 +0.05 252,314
June 94.77 94.78 94.62 94.74 -0.04 5.26 +0.04 145,943
Sept 94.48 94.50 94.34 94.46 -0.03 5.54 +0.03 100,739
Dec 93.94 93.96 93.93 93.93 -0.03 6.07 +0.03 71,656
Mr93 93.76 93.17 93.65 93.74 -0.03 6.26 +0.03 55,473
June 93.46 93.46 93.36 93.43 -0.02 6.57 +0.02 44,243
Sept 93.17 93.19 93.09 93.17 -0.01 6.83 +0.01 31,658
Dec 92.79 92.82 92.72 92.82 +0.02 7.18 -0.02 21,132
Mr94 92.76 92.81 92.71 92.80 +.03 7.20 -0.03 27,888
June 92.50 92.58 92.46 92.57 +0.05 7.43 -0.05 17,956
Sept 92.28 92.37 92.23 92.35 +0.06 7.65 -0.06 11,828
Dec 91.97 92.06 91.92 92.05 +0.07 7.95 -0.07 9,816
Mr95 91.96 92.05 91.93 92.04 +0.07 7.96 -0.07 7,178
June 91.94 91.93 91.82 91.92 +0.07 8.08 -0.07 6,868
Sept 91.69 91.78 91.67 91.77 +0.07 8.23 -0.07 6,419
Est vol 283,796; vol Tues 130,709; open Int 1,053,277; + 4,629.
EURODOLLAR (LIFFE)-$l million; Pt of 100%
Lifetime
Open
Open High Low Settle Change High Low
Interest
Dec 94.86 94.87 94.75 94.93 -0.02 94.94 90.58 17,546
Mr92 94.97 94.99 94.95 94.96 … 95.06 90.60 10,098
June 94.77 94.78 94.66 94.75 … 94.83 90.97 5,226
Sept 94.49 94.49 94.47 94.46 … 94.53 90.97 2,724
Dec 93.97 93.97 93.95 93.93 … 94.00 91.54 614
Mr93 93.80 913.80 93.78 93.74 +0.02 93.80 91.55 545
June … … … 93.43 -0.23 93.44 9,2.60 405
Sept … … … 93.17 +0.08 93.09 92.82 137
STERLING (LIFFE) - GBP500,000; pts of 100%
Dec 89.79 89.82 89.78 89.81 +0.02 90.35 86.52 52,369
Mr92 90.25 90.29 90.24 90.28 +0.04 90.49 86.68 45,90
June 90.34 90.37 90.33 90.36 +0.03 90.46 87.45 34,626
Sept 90-31 90-33 90-29 90.32 +0.02 90.41 87.46 10,000
Dec 90.22 90.23 90.20 90.23 +0.02 90.32 87.55 6,664
Mr93 90.07 90.10 90.07 90.09 +0.02 90.16 87.50 4,548
June 89.97 89.97 89.97 89.98 +0.02 90.09 87.58 2,095
Sept 89.93 89.95 89.93 89.95 +0.02 90.08 88.20 1,746
Dec 89.88 89-90 89-88 89-92 +0.04 90.02 98.95 1,641
Est vol 20,529; vol 20,507; open lnt 160,391, - 1,238.

LONG GILT (LIFFE) –GBP 50,000; 32nds of 100%
Yield Open
Settle Interest
Dec 95-17 95-23 95-09 95-17 +0-03 97-17
89-10 43,299
Mr92 95-24 95-24 95-18 94-22 +0-05 96-06
94-18 2,524
Est vol 24,368; vol Tues 23,953; open Int 45,823, +1,160.
OTHER INTEREST RATE FUTURES
Settlement Prices of selected contracts. Volume and open Interest of all contract months.
Mortgage-Backed (CBT)-$100,000, pts. & 64ths of 100%
Nov Cpn 8.5 102-04 -6; Est. vol. 0; Open Int. 90
5-Yr. Int. Rate Swap (CBT)-$25 per 1/2 b.p.; Pts of 100%
Dec 92.770 -0.010; Est. vol. 50; Open I nt. 707
3-Yr. Int. Rate Swap (CBT)-$25 per 1/2 b.p.; pts of 100%
Dec 93.490 -0.010; Est. vol. 0; Open Int. 456
Treas. Auction 5 Yr (FINEX)-$250,000, 100 minus yield
Dec 93.22 - 4.0; Est. vol. 100; Open Int. 4
CBT –Chicago Board of Trade. FINEX –Financial Instrument Exchange, a division of the New
York Cotton Exchange. IMM –International Monetary Market at Chicago Mercantile Exchange.
LIFFE –London International Financial Futures Exchange.
MCE –MidAmerica Commodity Exchange
Source: Reprinted by permission of Wall Street Journal ©, (November 14, 1991) Dow Jones &
The T-bill and Eurodollar futures contracts are on short-term debt instruments and are discussed first. The T-
bond (and T-note) futures are written on long-term interest rates and are discussed second. Prior to beginning
either discussion, however, it is useful to review the pricing mechanics of the fixed-income securities that
underlie these futures contracts.
8.2 UNDERLYING BONDS

Two types of interest rate or fixed-income securities underlie interest rate futures contracts. One type is a
zero-coupon or a discount bond. A discount bond provides no explicit interest payments. It is traded at prices
below the face value or par value of the security. Security income results from price appreciation. The other
type of fixed-income security is a coupon-bearing bond. Like a discount bond, a coupon-bearing bond pays
the face or par value at the end of the bond’s life. In addition, a coupon-bearing bond has prespecified
coupon interest payments at regular intervals throughout the bond’s life.
Zero-Coupon or Discount Bonds

The price of a discount bond, Bd , is computed by taking the present value of the promised payment of face
value, Fn , at the end of the bond’s life, n periods from now, that is,
Fn
Bd  (8.1)
(1  y ) n
where y is the yield to maturity or rate of return on the bond.

U.S. Treasury bills are discount bonds. T-bills are short-term debt instruments issued by the U.S.
Government. New 91-day and 182-day bills are issued every Thursday. New 364-day bills are issued every
fourth Thursday. The minimum face value of a T-bill is $10,000.
T-bill quotes are unusual in that their prices are not reported directly. Table 8.3 contains market
information about the T-bills that were active on November 13, 1991. Note that, in this table, bid and ask
“discounts” appear. These are bank discount price quotations. The definition of a bank discount is
Bank discount  (360/n) (1 00 - Bd ) , (8.2)
where n is the number of days to maturity of the bill, and Bd is the bill’s price expressed as a percentage of
par. (Traditionally, bankers have assumed a 360-day year.) To compute the price of the T-bill, simply
rearrange equation (8.2) to isolate Bd , that is,
Bd  100 - Bank discount (n / 360) . (8.3)
TABLE 8.3 Treasury bond, note, and bill prices.

GOVT. BONDS & NOTES
Maturity Ask
Rate Bid Asked Chg.
Mo/Yr Yld.
61/2 Nov 91n 100:00 100:02 0.00
81/2 Nov 91n 100:00 100:02 0.00
141/4 Nov 91n 100:01 100:03 0.00
73/4 Nov 91n 100:04 100:06 3.06
75/8 Dec 91n 100:11 100:13 4.24
81/4 Dec 91n 100:14 100:16 4.10
115/8 Jan 92n 101:03 101:05 -1 4.43
81/8 Jan 92n 100:22 100:24 -1 4.40
65/8 Feb 92n 100:13 100:15 4.65
91/8 Feb 92n 101:01 101:03 4.59
145/8 Feb 92n 102:13 102:15 -1 4.48
81/2 Feb 92n 101:01 101:03 4.61
77/8 Mar 92n 101:03 101:05 -1 4.69
81/2 Mar 92n 101:11 101:13 4.63
113/4 Apr 92n 102:25 102:27 -2 4.72
87/8 Apr 92n 101:26 101:28 4.68
65/8 May 92n 100:27 100:29 4.77
91/8 May 92n 102:00 102:02 –1 4.78
133/4 May 92n 104:11 104:13 -1 4.73
81/2 May 92n 101:28 101:30 -1 4.93
81/4 Jun 92n 102:01 102:03 -1 4.82
83/8 Jun 92n 102:03 102:05 4.84
103/8 Jul 92n 103:16 103:18 4.98
8 Jul 92n 102:02 102:04 -1 4.92
41/4 Aug 87-92 98:13 99:13 -1 5.07
71/4 Aug 92 101:18 101:22 4.94
77/8 Aug 92n 102:00 102:02 -1 5.04
81/4 Aug 92n 102:09 102:11 -1 5.03
TABLE 8.3 Cont. Treasury bond, note, and bill prices.

GOVT. BONDS & NOTES
Maturity Ask
Rate Bid Asked Chg.
Mo/Yr Yld.
81/4 Aug 92n 102:11 102:13 -1 4.99
81/8 Sep 92n 102:18 102:20 -1 5.02
83/4 Sep 92n 103:03 103:05 -1 5.02
93/4 Oct 92n 104:04 104:06 -1 5.02
73/4 Oct 92n 102:12 102:14 -1 5.11
73/4 Nov 92n 102:15 102:17 -1 5.12
83/8 Nov 92n 103:02 103:04 -1 5.13
101/2 Nov 92n 105:04 105:06 -1 5.11
73/8 Nov 92n 102:06 102:08 -1 5.13
71/4 Dec 92n 102:06 102:08 5.17
91/I Dec 92n 104:06 104:08 -1 5.19
83/4 Jan 93n 103:28 103:30 -2 5.23
7 Jan 93n 101:30 102:00 -1 5.27
4 Feb 88-93 96:22 97:22 –1 5.94
63/4 Feb 93 101:22 101:26 -1 5.24
77/8 Feb 93 103:00 103:04 -1 5.26
81/4 Feb 93n 103:15 103:17 -1 5.30
83/8 Feb 93n 103:20 103:22 -1 5.29
107/8 Feb 93h 106:19 106:21 -2 5.31
63/4 Feb 93n 101:22 101:24 -2 5.33
71/8 Mar 93n 102:08 102:10 -2 5.36
95/8 Mar 93n 105:17 105:19 -1 5.35
73/4 Apr 93n 102:20 102:22 -1 5.38
7 Apr 93n 102:05 102:07 -1 5.40
75/8 May 93n 101:02 103:04 5.43
85/8 May 93n 104:15 104:17 5.44
101/8 May 93n 106:18 106:20 -2 5.46
63/4 May 93n 101:26 101:28 -1 5.47
7 Jun 93n 102:08 102:10 -1 5.49
81/8 Jun 93n 103:31 104:01 -1 5.50
71/4 Jul 93n 102:21 102:23 -1 5.52
67/8 Jul 93n 102:02 102:04 -1 5.56
71/2 Aug 88-93 100:20 100:24 +2 7.04
8 Aug 93n 103:36 104:00 5.57
85/8 Aug 93 104:30 105:02 5.55
83/4 Aug 93n 105:05 105:07 5.58
117/8 Aug 93n 110:11 110:13 5.56
63/8 Aug 93n 101:09 101:11 5.58
61/8 Sep 93n 100:27 100:29 5.61
81/4 Sep 93n 104:19 104:21 -1 5.60
71/8 Oct 93n 102:21 102:23 -1 5.61
6 Oct 93n 100:21 100:23 5.61
73/4 Nov 93n 103:27 103:29 -1 5.66

GOVT. BONDS & NOTES
Maturity Ask
Rate Bid Asked Chg.
Mo/Yr Yld.
85/8 Nov 93 105:15 105:19 +1 5.63
9 Nov 93n 106:05 106:07 -1 5.67
113/4 Nov 93n 111:10 111:12 -1 5.65
75/8 Dec 93n 103:24 103:26 -1 5.70
7 Jan 94n 102:15 102:17 -2 5.74
67/8 Feb 94n 102:09 102:11 5.75
87/8 Feb 94n 106:10 106:12 -3 5.81
9 Feb 94 106:18 106:22 -2 5.79
81/2 Mar 94n 105:24 105:26 -2 5.84
7 Apr 94n 102:19 102:21 -1 5.81
41/8 May 89-94 96:24 97:24 -3 5.09
7 May 94n 102:17 102:19 -2 5.87
91/2 May 94n 108:06 106:06 -3 5.90
131/8 May 94n 116:17 116:19 -3 5.89
81/2 Jun 94n 106:03 106:05 -1 5.93
8 Jul 94n 104:30 105:00 -2 5.95
67/8 Aug 94n 102:08 102:10 -3 5.95
85/8 Aug 94n 106:16 106:18 -2 6.00
83/4 Aug 94 106:25 106:29 -2 5.99
125/8 Aug 94n 116:17 116:19 -2 5.99
81/2 Sep 94n 106:12 106:14 -1 6.03
91/2 Oct 94n 109:09 109:12 -2 6.06
6 Nov 94n* 100:00 100:01 -2 5.99
81/4 Nov 94n 105:27 105:29 -2 6.07
101/8 Nov 94 110:28 111:00 -3 6.06
115/8 Nov 94n 114:31 115:01 -2 6.07
75/8 Dec 94n 104:10 104:12 -2 6.07
85/8 Jan 95n 106:30 107:00 -2 6.16
3 Feb 95 97:00 98:00 -3 3.66
73/4 Feb 95n 104:17 104:19 -3 6.17
101/2 Feb 95 112:10 112:14 -3 6.22
111/4 Feb 95n 114:17 114:19 -5 6.22
83/8 Apr 95n 106:06 106:10 -4 6.29
81/2 May 95n 106:24 106:26 -2 6.30
103/8 May 95 112:14 112:18 -3 6.32
111/4 May 95n 115:03 115:05 -4 6.35
125/8 May 95 119:14 119:18 -2 6.31
87/8 Jul 95n 107:28 107:30 -4 6.41
81/2 Aug 95n 106:24 106:26 -4 6.43
101/2 Aug 95n 113:07 113:09 -4 6.46
85/8 Oct 95n 107:06 107:10 -4 6.4
81/2 Nov 95n 106:28 106:30 -4 6.50
91/2 Nov 95n 110:10 110:12 -6 6.51

GOVT. BONDS & NOTES
Maturity Ask
Rate Bid Asked Chg.
Mo/Yr Yld.
111/2 Nov 95 117:12 117:16 -4 6.47
91/4 Jan 96n 109:18 109:20 -5 6.57
71/2 Jan 96n 103:09 103:11 -5 6.58
77/8 Feb 96n 104:18 104:20 -5 6.61
87/8 Feb 96n 108:07 108:09 -6 6.61
71/2 Feb 96n 103:07 103:09 -5 6.61
73/4 Mar 96n 104:03 104:05 -5 6.64
93/8 Apr 96n 110:08 110:10 -5 6.64
75/8 Apr 96n 103:22 103:24 -5 6.64,
73/8 May 96n 102:22 102:24 -5 6.66
75/8 May 96n 103:21 103:23 -5 6.66
77/8 Jun 96n 104:18 104:20 -5 6.70
77/8 Jul 96n 104:19 104:21 -6 6.70
77/8 Jul 96n 104:23 104:25 -5 6.68
71/4 Aug 96n 102:06 102:08 -5 6.69
7 Sept 96n 101:06 101:08 -5 6.70
8 Oct 96n 105:06 105:08 -5 6.73
67/8 Oct 96n 100:26 100:28 -5 6.66
71/4 Nov 96n 102:04 102:06 -5 6.73
8 Jan 97n 105:04 105:06 -5 6.79
81/2 Apr 97n 107:07 107:09 -5 6.87
81/2 May 97n 107:07 107:09 -5 6.89
81/2 Jul 97n 107:07 107:09 -5 6.92
85/8 Aug 97n 107:26 107:28 -6 6.94
83/4 Oct 97n 108:16 108:18 -4 6.96
87/8 Nov 97n 109:04 109:06 -7 6.97
77/8 Jan 98n 104:06 104:08 -7 7.01
81/8 Feb 98n 105:13 105:15 -6 7.03
77/8 Apr 98n 104:06 104:08 -5 7.04
7 May 93-98 100:16 100:24 6.47
9 May 98n 109:27 109:29 -8 7.07
81/4 Jul 98n 105:30 106:00 -7 7.10
91/4 Aug 98n 111:04 111:06 -8 7.13
71/8 Oct 98n 100:09 100:11 -5 7.06
31/2 Nov 98 97:24 98:24 +24 3.70
87/8 Nov 98n 109:09 109:11 -6 7.15
87/8 Feb 9,9n 109:09 109:11 -6 7.20
81/2 May 94-99 105:00 105:08 -4 6.20
91/8 May 99n 110:23 110:25 -7 7.24
8 Aug 99n 104:08 104:10 -6 7.26
77/8 Nov 99n 103:15 103:17 -6 7.28
77/8 Feb 95-00 102:15 102:19 -1 6.97
81/2 Feb 00n 107:01 107:03 -9 7.34

GOVT. BONDS & NOTES
Maturity Ask
Rate Bid Asked Chg.
Mo/Yr Yld.
87/8 May 00n 109:06 109:08 -12 7.39
83/8 Aug 95-00 104:18 104:22 -3 6.93
83/4 Aug 00n 108:12 108:14 -13 7.42
81/2 Nov 00n 106:27 106:29 -13 7.43
73/4 Feb 0ln 101:31 102:01 -11 7.44
113/4 Feb 01 128:14 128:18 -10 7.43
8 May 01n 103:22 103:24 -10 7.44
131/8 May 01 138:04 138:08 -13 7.44
77/8 Aug 01n 102:29 102:31 -14 7.44
8 Aug 96-01 103:19 103:23 -15 7.07
133/8 Aug 01 140:16 140:20 -8 7.44
71/2 Nov 01n* 100:19 100:20 -11 7.41
153/4 Nov 01 157:21 157:25 -10 7.45
141/4 Feb 02 147:25 147:29 -14 7.48
115/8 Nov 02 130:10 130:14 -13 7.51
103/4 Feb 03 123:31 124:03 -12 7.54
103/4 May 03 124:06 124:10 -12 7.55
111/8 Aug 03 127:0.4 127:08 -15 7.58
117/8 Nov 03 133:07 133:11 -23 7.59
123/8 May 04 137:30 138:02 -29 7.60
133/4 Aug 04 149:07 149:11 -29 7.63
115/9 Nov 04 132:02 132:06 -23 7.67
81/4 May 00-05 104:10 104:14 -8 7.53
12 May 05 135:23 135:27 -26 7.69
103/4 Aug 05 125:12 125:16 -19 7.71
93/8 Feb 06 114:14 114:18 -18 7.68
75/8 Feb 02-07 99:20 99:24 -16 7.65
77/8 Nov 02-07 101:14 101:18 -16 7.66
83/8 Aug 03-08 104:28 105:00 -12 7.72
83/4 Nov 03-08 107:19 107:23 -19 7.75
91/8 May 04-09 110:21 110:25 -24 7.76
103/8 Nov 04-09 120:28 121:00 -24 7.78
113/4 Feb 05-10 132:10 132:14 -25 7.78
10 May 05-10 118:15 118:19 -23 7.75
123/4 Nov 05-10 141:24 141:28 -29 7.79
137/8 May 06-11 152:06 152:10 -30 7.79
14 Nov 06-11 154:08 154:12 -31 7.79
103/8 Nov 07-12 122:13 122:17 -25 7.87
12 Aug 08-13 137:25 137:29 -30 7.88
131/4 May 09-14 150:07 150:11 -36 7.89
121/2 Aug 09-14 143:13 143:17 -34 7.90
113/4 Nov, 09-14 136:27 136:31 -35 7.87
111/4 Feb 15 135:01 135:05 -36 7.92

GOVT. BONDS & NOTES
Maturity Ask
Rate Bid Asked Chg.
Mo/Yr Yld.
105/8 Aug 15 128:19 128:2 -3.1 7.92
97/8 Nov 15 120:23 120:25 -30 7.93
91/4 Feb 16 113:30 114:00 -33 7.94
71/4 May 16 92:21 92:23 -27 7.93
71/2 Nov16 95:07 95:09 -28 7.94
83/4 May17 108:25 108:27 -31 7.94
87/8 Aug 17 110:05 110:07 -31 7.94
91/8 May 18 113:00 113:02 -32 7.94
9 Nov 18 111:21 111:23 -32 7.94
87/8 Feb 19 110:09 110:11 -32 7.94
81/8 Aug 19 101:31 102:01 -30 7.94
81/2 Feb 20 106:06 106:08 -32 7.94
83/4 May 20 109:02 109:04 -32 7.94
83/4 Aug 20 109:02 109:04 -30 7.94
77/8 Feb 21 99:07 44:09 -30 7.94
81/8 May 21 102:04 102:06 -31 7.93
81/8 Aug 21 102:09 102:11 -30 7.92
8 Nov 21* 101:12 101:13 -28 7.88

U.S. TREASURY STRIPS
Mat Type Bid Ask Chg. Bid Yield
Feb 92 ci 98:26 98:26 1 4.87
May 92 ci 97:18 97:19 1 4.99
Aug 92 ci 96: 11 96:11 1 5.04
Nov 92 ci 95:04 95:05 2 5.05
Feb 93 ci 93:19 93:20 1 5.38
May 93 ci 92:08 92:09 2 5.46
Aug 93 ci 90:26 90:27 3 5.59
Nov 93 ci 89:13 89:15 2 5.68.
Feb 94 ci 87:24 87:26 1 5.89
May 94 ci 86:13 86:15 1 5.93
Aug 94 ci 84:30 85:00 3 6.03
Nov 94 ci 83:12 83:15 3 6.15
Nov 94 np 83:08 93:11 3 6.20
Feb 95 ci 81:18 81:20 4 6.37
Feb 95 np 81:24 81:26 4 6.30
May 95 ci 80:02 80:05 5 6.45
May 95 np 80:01 80:03 4 6.47
Aug 95 ci 78:20 78:22 4 6.52
Aug 95 np 78:14 78:17 5 6.58
Nov 95 ci 77:08 77:11 5 6.56
Nov 95 np 77:02 77:05 5 6.62

Feb 96 ci 75:14 75:17 5 6.74
Feb 96 np 75:09 75:12 5 6.79
May 96 ci 74:04 74:07 5 6.77
May 96 np 74:01 74:05 6 6.79
Aug 96 ci 72:19 72:22 5 6.86
Nov 96 ci 71:14 71:17 6 6.94
Nov 96 np 70:10 70:13 3 7.17
Feb 97 ci 69:2b 69:23 4 7.02
May 97 ci 68:07 68:10 3 7.08
May 97 np 68:05 68:09 4 7.09
Aug 97 ci 66:27 66:31 4 7.13
Aug 97 np 66:26 66:29 3 7.14
Nov 97 ci 65:24 65:28 1 7.11
Nov 97 np 65:18 65:22 -4 7.16
Feb 98 ci 63:31 64:02 -8 7.28
Feb 98 np 63:23 63:27 -4 7.34
May 98 ci 62:24 62:28 -8 7.30
May 98 np 62:16 62:20 -4 7.36
Aug 98 ci 61:15 61:19 -8 7.34
Aug 98 np 61:06 61:10 -4 7.41
Nov 98 ci 60:12 60:16 -8 7.34
Nov 98 np 60:05 60:09 -7 7.39
Feb 99 ci 58:19 58:23 -7 7.51
Feb 99 np 58:14 58:18 -8 7.55
May 99 ci 57:14 57:18 -7 7.53
May 99 np 57:09 57:13 -8 7.57
Aug 99 ci 56:07 56:11 -7 7.57
Aug 99 np 56:04 56:08 7 7.59
Nov 99 ci 55:06 55:10 8 7.57
Nov 99 np 55:03 55:07 -7 7.59
Feb 00 ci 53:28 54:00 -8 7.64
Feb 00 np 53:22 53:27 -9 7.68
May 00 ci 52:28 53:00 -7 7.64
May 00 np 52:22 52:27 -7 7.68
Aug 00 ci. 51:26 51:30 -7 7.66
Aug 00 np 51:23 51:27 -7 7.68
Nov 00 ci 50:24 50:28 -8 7.68
Nov 00 np 50:24 50:28 -7 7.68
Feb 01 ci 49:14 49:19 -7 7.76
Feb 01 np 49:09 49:13 -8 7.80
May 01 ci 48:14 48:18 -8 7.78
May 01 np 48:08 48:12 -9 7.82
Aug 01 ci 47:16 47:21 -9 7.78
Aug 01 np 47:11 47:15 -8 7.82

Nov 01 ci 46:20 46:24 -7 7.78
Feb 02 ci 45:09 45:13 -12 7.88
May 02 ci 44:11 44:15 -11 7.90
Aug 02 ci 43:12 43:17 -12 7.92
Nov 02 ci 42:15 42:19 -11 7.94
Feb 03 ci 41:16 41:21 -12 7.97
May 03 ci 40:21 40:26 -12 7.98
Aug 03 ci 39:27 39:31 -11 7.99
Nov 03 ci 39:00 39:05 -12 8.00
Feb 04 ci 38:04 38:08 -11 8.03
May 04 ci 37:09 37:13 -12 8.45
Aug 04 ci 36:16 36:21 -12 8.06
Nov 04 ci 35:24 35:28 -12 8.07
Nov 04 bp 35:23 35:27 -11 8.08
Feb 05 ci 34:31 35:03 -11 8.09
May 05 ci 34:08 34:12 -11 8.10
May 05 bp 34:09 34:13 -11 8.09
Aug 05 ci 33:18 33:22 -11 8.10
Aug 05 bp 33:19 33:24 -12 8.09
Nov 05 ci 32:29 33:01 -11 8.10
Feb 06 ci 32:05 32:10 -12 8.12
Feb 06 bp 32:04 32:08 -11 8.13
May 06 ci 31:17 31:21 -11 8.12
Aug 06 ci 30:29 31:01 -11 8.12
Nov 06 ci 30:10 30:14 -11 8.12
Feb 07 ci 29:16 29:20 -12 8.17
May 07 ci 28:29 29:01 -12 8.17
Aug 07 ci 28:11 28:15 -12 8.17
Nov 07 ci 27:25 27:29 -12 8.17
Feb 08 ci 27:03 27:07 -11 8.20
May 08 ci 26:15 26:19 -11 8.22
Aug 08 ci 25:30 26:02 -11 8.22
Nov 08 ci 25:14 25:18 -10 8.22
Feb 09 ci 24:25 24:29 -11 6.25
May 09 ci 24:10 24:13 -10 8.25
Aug 09 ci 23:26 23:30 -10 8.25
Nov 09 ci 23:11 23:15 -10 8.25
Nov 09 bp 22:30 23:02 -9 8.35
Feb 10 ci 22:28 23:00 -19 8.25
May 10 ci 22:13 22:17 -10 8.25
Aug 10 ci 21:30 22:01 -10 8.26
Nov 10 ci 21:15 21:19 -11 8.26
Feb 11 ci 21:02 21:05 -19 8.26
May 11 ci 20:20 20:24 -10 8.26

Aug 11 ci 20:07 20:11 -10 8.26
Nov 11 ci 19:26 19:30 -10 8.26
Feb 12 ci 19:12 19:16 -10 8.17
May 12 ci 19:00 19:03 -9 8.27
Aug 12 ci 18:20 18:23 -9 8.27
Nov 12 ci 18:08 18:11 -9 8.27
Feb 13 ci 17:28 17:31 -9 8.17
May 13 ci 17:16 17:20 -10 8.27
Aug 13 ci 17:05 17:09 -9 8.27
Nov 13 ci 16:26 16:30 -9 8.27
Feb 14 ci 16:15 16:19 -9 8.27
May 14 ci 16:05 16:08 -9 8.27
Aug 14 ci 15:26 15:30 -9 8.17
Nov 14 ci 15:16 15:20 -9 8.27
Feb 15 ci 15:06 15:10 -9 8.27
Feb 15 bp 15:07 15:11 -8 8.26
May 15 ci 14:29 15:00 -8 8.27
Aug 15 ci 14:19 14:22 -9 8.27
Aug 15 bp 14:20 14:23 -8 8.26
Nov 15 ci 14:10 14:13 -8 8.27
Nov 15 bp 14:11 14:14 -7 8.26
Feb 16 ci 14:00 14:04 -9 8.27
Feb 16 bp 14:03 14:06 -8 8.26
May 16 ci 13:23 13:27 -9 8.27
May 16 bp 14:00 14:03 -8 8.19
Aug 16 ci 13:15 13:18 -8 8.27
Nov 16 ci 13:06 13:09 -6 8.27
Nov 16 bp 13:12 13:15 -6 8.21
Feb 17 ci 12:31 13:02 -8 8.26
May 17 ci 12:22 12:25 -8 8.26
May 17 bp 12:26 12:29 -8 8.26
Aug 17 ci 12:14 12:17 -8 8.26
Aug 17 bp 12:18 12:21 -8 8.22
Nov 17 ci 12:05 12:08 -8 8.27
Feb 18 ci 11:31 12:02 -8 8.25
May 18 ci 11:24 11:26 -7 8.25
May 18 bp 11:27 11:30 -8 8.21
Aug 18 ci 11:16 11:19 -8 8.25
Nov 18 ci 11:09 11:12 -7 8.25
Nov 18 bp 11:12 11:15 -9 8.21
Feb 19 ci 11:03 11:06 -8 8.23
Feb 19 bp 11:08 11:11 -8 8.18
May 19 ci 10:29 11:00 -8 8.22
Aug 19 ci 10:22 10:25 -8 8.22

Aug 19 bp 10:27 10:30 -7 8.17
Nov 19 ci 10:16 10:19 -8 8.21
Feb 20 ci 10:11 10:13 -7 8.20
Feb 20 bp 10:14 10:17 -7 8.16
May 20 ci 10:05 10:08 -7 8.19
May 20 bp 10:08 10:10 -7 8.18
Aug 20 ci 9:30 10:01 -7 8.19
Aug 20 bp 10:02 10:05 -9 8.15
Nov 20 ci 9:26 9:28 -7 8.17
Feb 21 ci 9:20 9:22 -7 8.17
Feb 21 bp 9:23 9:26 -8 8.13
May 21 ci 9:16 9:19 -7 8.14
May 21 bp 9:19 9:21 -7 8.11
Aug 21 ci 9:19 9:21 -7 8.04
Aug 21 bp 9:18 9:20 -9 8.05
Nov 21 bp 9:19 9:22 -6 7.97

TREASURY BILLS
Days
Maturity to Bid Asked Chg. Ask Yld.
Mat.
Nov 21 '91 6 4.68 4.58 -0.02 4.65
Nov 29 '91 14 4.54 4.44 -0.01 4.52
Dec 05 '91 20 4.41 4.31 -0.03 4.39
Dec 02 '91 27 4.36 4.26 -0.08 4.34
Dec 09 '91 34 4.39 4.35 -0.05 4.43
Dec 26 '91 41 4.40 4.36 -0.04 4.45
Jan 02 '92 48 4.43 4.39 -0.02 4.49
Jan 09 '92 55 4.50 4.46 -0.02 4.57
Jan 16 '92 62 4.55 4.53 -0.01 4.63
Jan 23 '92 69 4.57 4.55 -0.01 4.67
Jan 30 '92 76 4.57 4.55 -0.01 4.67
Feb 06 '92 93 4.62 4.60 4.73
Feb 13 '92 44 4.63 4.61 4.73
Feb 20 '92 97 4.62 4.60 4.74.
Feb 27 '92 104 4.62 4.60 4.74
Mar 05 '92 111 4.6.4 4.62 4.76
Mar 12 '92 118 4.66 4.64 +0.01 4.79
Mar 19 '92 125 4.66. 4.64 +0.01 4.79
Mar 26 '92 132 4.64 4.62 4.78
Apr 02 '92 139 4.66 4.64 4.80
Apr 09 '92 146 4.69 4.67 4.84
Apr 16 '92 153 4.69 4-67 +0.01 4.84

TREASURY BILLS
Days
Mat.
Apr 23 '92 160 4.72 4.70 4.88
Apr 30 '92 167 4.68 4-66 4.84
May 07 '92 174 4.72 4.70 +0.02 4.89
May 14 '92 181 4.70 4.68 -0.01 4.87
Jun 04 '92 202 4.65 4.63 +0.02 4.82
Jul 02 '92 230 4.71 4.69 +0.01 4.89
Jul 30 '92 258 4.74 4.72 +0.03 4.93
Aug 27 '92 286 4.74 4.72 +0.02 4.94
Sep 24 '92 314 4.73 4.71 +0.04 4.94
Oct 22 '92 342 4.75 4.73 +0.03 4.98
To illustrate the bank discount convention, consider the T-bill in Table 8.3 that matures on February 13,
1992. If this bill is purchased, it would be at the ask price,63 as reflected by the ask discount of 4.61. The
number of days to maturity of this bill is 92. Therefore, the ask price of the bill is
Bd  100  4.61(92 / 360)  98.822
percent of par. If the par value is $10,000, the price of the T-bill is $9,882.20. To compare the rate of return
on the T-bill with the rate of return on other instruments, the effective annual rate of return is often
computed. If the price of a 92-day T-bill is 98.822, the effective annual rate of return compounded on a daily
basis64 is
r  (100 / 98.822)365 / 92  1  4.81% .
Eurodollar certificates of deposit (CD’s) are also discount bonds. Eurodollar deposits are U.S. dollar deposits
in any foreign bank, although Eurodollar deposits are most typically thought of as being U.S. dollar deposits
in London. The rate on these CD’s is referred to as the London Interbank Offer Rate (LIBOR). The usual
denomination of Eurodollar CD’s is $1,000,000. The usual maturities are in the three- to six-month range,
however, maturities as long as five years are not uncommon. Simple interest on Eurodollar deposits is
calculated for the actual number of days on a 360-day year basis and is paid at maturity on deposits with
terms of less than one year. For longer-term deposits, interest is paid annually.65
Sample Eurodollar rates are contained in Table 8.4. The rates reported are averages of the rates
quoted by five major banks in London. The three-month rate, for example, is reported to be 51/8 percent.
That implies that a $1,000,000 Eurodollar deposit on November 13, 1991, will be worth $1,000,000
63
The bid/ask spread is the cost of immediate trade execution. A market-maker stands ready to buy immediately at her bid price
and sell immediately at her ask price. The bid/ask spread is her compensation for providing market liquidity.
64
The effective annual return compounded continuously is r  ln (100/98.822)(365/92)  4.70% .
65
For more information on Eurodollar certificates of deposit, see Stigum (1990).
+1,000,000(0.05125)(90/360), or $1,012,812.50, on February 11, 1992.66 The effective annual rate of return
on this investment may be computed as follows:
r  (1, 012,812.50 /1, 000, 000)365/ 90  1  5.30% .
Note that although this Eurodollar deposit has approximately the same maturity as the T-bill in the previous
example, the effective annual rate of return is 49 basis points higher.67 This reflects the higher default risk of
these deposits vis-à-vis U.S. government securities.
T-bills and CD’s are short-term discount bonds. Table 8.3 also contains prices of Treasury strips, which are
long-term discount bonds. Strips get their name from the fact that, for each year, coupons of several
government coupon bonds are stripped” from the principal and combined to create a discount bond that
makes only one payment. Table 8.3 provides the prices and yields to maturity for these bonds.
TABLE 8.4 Money market rates.

MONEY RATES
Wednesday, November 13, 1991. The key U.S. and foreign annual Interest rates below are a guide to general
levels but don’t always represent actual transactions.
PRIME RATE: 71/2%. The base rate on corporate loans at U.S. money market banks.
FEDERAL FUNDS: 51/e% high, 2% low, 2% near closing bid, 21/2% offered. Reserves traded among
commercial banks for overnight use In amounts of $1 million or more. Source: Babcock Fulton Prebon
(U.S.A.) Inc.
DISCOUNT RATE: 4.50%. The charge on loans to depository Institutions by the Federal Reserve Banks.
CALL MONEY-. 63/d% to 7%. The charge on loans to brokers on stock exchange collateral.
COMMERCIAL PAPER placed directly by General Electric Capital Corp.: 4.85% 15 to 36 days; 4.70% 37
to 59 days; 4.93% 60 to 89 days; 4.90% 90 to 149 days; 4.87% 150 to 179 days; 4.70% 180 to 189 days;
4.87% 190 to 270 days. Commercial Paper Placed directly by General Motors Acceptance Corp.: 4.85% 30
to 44 days;- 4.80% 45 to 59 days; 5% 60 to 270 days.
COMMERCIAL PAPER: High-grade unsecured notes sold through dealers by mayor corporations In
multiples of $1,000: 5% 30 days; 5.10% 60 days; 5.02% 90 days.
CERTIFICATES OF DEPOSIT: 4.48% one month; 4.61% two months; 4.60% three months; 4.62% six
months; 4.98% one year. Average of too rates paid by mayor New York banks on Primary new Issues of
negotiable C.D.s, usually on amounts of $1 million and more. The minimum unit is $100,000. Typical rates
in the secondary market: 4.90% one month; 5% three months; 5% six months.
BANKERS ACCEPTANCES: 4.80% 30 days; 5.03% 60 days; 4.91% 90 days; 4.87% 120 days; 4.85% 150
days; 4.81% 180 days. Negotiable, bank-backed business credit instruments typically financing an Import
order.
LONDON LATE EURODOLLARS: 5% - 47/8% one month; 51/2% - 51/4% two months; 5 3/16% - 5
1/16% three months; 5 3/16% - 5 1/16% four months; 5 3/16% - 5 1/16% five months; 51/8% - 5% six
months.
LONDON INTERBANK OFFERED RATES (LIBOR): 4 15/16% one month; 51/8% three months; 51/8%
six months; 51/4% one year. The average of Interbank offered rates for dollar deposits In the London market
based on quotations at five mayor banks. Effective rate for contracts entered into two days from date
appearing at top of this column.
FOREIGN PRIME RATES: Canada 8.50%; Germany 11.50%; Japan 7%; Switzerland 10%; Britain 10.50%.
These rate Indications aren’t directly comparable; lending practices vary widely by location.
66
The number of days from November 13, 1991, to February 11, 1992, is 90.
67
A basis point is 1/100 of one percent.
TREASURY BILLS: Results of the Tuesday, November 12, 1991, auction of short-term U.S. government
bills, sold at a discount from face value in units of $10,000 to $1 million: 4.64% 13 weeks; 4.71% 26 weeks.
FEDERAL HOME LOAN MORTGAGE CORP. (Freddie Mac): Posted Yields on 30-year mortgage
commitments. Delivery within 30 days 8.54%, 60 days 8.60%, standard conventional fixed-rate mortgages;
6%, 2% rate capped one-year adjustable rate mortgages. Source: Telerate Systems Inc.
FEDERAL NATIONAL MORTGAGE ASSOCIATION (Fannie Mae): Posted yields on 30-year mortgage
commitments for delivery within 30 days (Priced at par). 8.47%, standard conventional fixed rate-mortgages;
6.10%, 6/2 rate capped one-year adjustable rate mortgages. Source: Telerate Systems Inc.
MERRILL LYNCH READY ASSETS TRUST: 4.90%. Annualized average rate of return after expenses for
the past 30 days; not a forecast of future returns.
Coupon Bonds
The price of a coupon-bearing bond, B , is computed by summing the present values of the fixed periodic
coupon payments, Ct , and the present value of the terminal face value, Fn , that is,
Ct Fn
B  t 1
n
 (8.4)
(1  y ) t
(1  y ) n
In general, the coupon payments are the same fixed amount each period, so the bond price may also be
written as the present value of an annuity plus the present value of the final amount:
B  (C / y )[1  (1  y )  n ]  F (1  y )  n . (8.5)
The Treasury bonds and notes in Table 8.3 are coupon-bearing bonds. Treasury notes are issued with
maturities of two to ten years and Treasury bonds with maturities longer than ten years. The denominations
of bonds and notes range from $1,000 to $1 million. Their prices are quoted as a percentage of par, so a
reported price of 96:00 for a $100,000 face-value bond is actually 96 percent of $100,000 or $96,000. In
addition, the decimal part of a T-bond or T-note price is the number of 32nds, so a reported price of 94:8 is
actually 948/32, or 94.25 in decimal form.
Finally, all T-bonds and T-notes have semiannual coupon payments. The “9’s of November 2018” in
Table 8.3, for example, pay coupon interest at a rate of 4.5 percent of par in May and November of each
year. The last coupon is paid with the repayment of the face value in November 201868. The fact that
Treasury bonds pay semiannual coupons implies that the yield to maturity, y , in (8.4) and (8.5) is an
effective rate over a six-month period. To annualize this rate, the effective annual yield to maturity, y A , may
be computed as
y A  (1  y ) 2  1 (8.6)
While (8.6) is technically correct, some people prefer to use the bond equivalent yield, which is simply
2  y , to measure of the expected annual rate of return on the bond.69
One last convention about Treasury bonds and notes must be discussed. The bond price reported in
the financial press (i.e., the quoted bond price) excludes the accrued interest of the current coupon period.
68
Fabozzi and Fabozzi (1989, pp. 83-84) provide a summary of the various types of bill, note, and bond issues of the U.S.
government.
69
Yet others rely on coupon yield, which is simply the annual coupon amount divided by the bond price.
The price that you would pay for the bond if you decided to buy it would be the quoted price plus the coupon
interest that has accrued on a straight-line basis during the current coupon period. That is, accrued interest
equals the proportion of the current coupon period that has elapsed times the amount of the coupon,
 number of days since last coupon was paid 

AI  C   . (8.7)
 total number of days in current coupon period 
To illustrate, consider the 9s of November 2018, whose price is reported in Table 8.3. The bond is reported
to have a bid price of 111:21 and an ask price of 111:23. Treasury bonds pay coupon interest every six
months and do so on the 15th of the month. For the 9s of November 2018, coupon interest of 4.5 percent of
par is paid on May 15th and November 15th. On November 13, 1991, the number of days since the last
coupon payment is 182. The total number of days from May 15, 1991, to November 15, 1991, is 184. The
accrued interest on this bond as of November 13 is, therefore, 4.5(182/184) or 4.45. If this bond were bought
at the ask price, we would pay the reported ask price of 11123/32 plus accrued interest of 4.45, or a total
amount 116.17.
To verify that 116.17 is actually the ask price of the 9s of November 2018, compute the present value
of the promised coupons and face value as of November 13, 1991, using the reported bond equivalent yield
of 7.94 percent. We cannot apply the bond valuation formulas (8.4) and (8.5) directly because those formulas
assume that the next coupon payment is in exactly six months. For the 9s of November 2018, we are part
way through the coupon period, so the bond valuation formula needs to be modified:
 n Ct Fn 
B  (1  y ) p t 1  , (8.8)
 (1  y ) (1  y ) n 
t
where p is the ratio of the number of days elapsed in the current coupon period to the total number of days
during the current coupon period. The term in bracket represents the present value, just after the last coupon,
of all future payments to the bondholder. This amount is then capitalized forward p coupon periods to the
current date. Substituting the example parameters,
 n Ct Fn   55 4.5 100 
B  (1  y ) p t 1  n
 (1.0397) p t 1   116.17
 (1  y ) (1  y ) 
t
 (1.0397) (1.0397) n 
t
Subtracting the accrued interest yields the quoted ask price of the bond reported in Table 8.3.
Finally, some Treasury bonds do not have a fixed maturity date. The issues denoted by the hyphenated
maturity date in Table 8.3 are callable bonds, which the Treasury has the right to “call,” or redeem, at any
3/4
time during a prespecified period in the future. The 11 of August 2009-14, for example, are callable bonds
that may be redeemed during the period November 2009 through 2014. Callable bonds may be delivered on
certain futures contracts, as long as they satisfy some minimum amount of time before the first call date.
8.3 INTEREST RATE RISK

Bondholders (lenders) face the risk that interest rates will rise, causing a decline in the value of their bonds.
Bond issuers (borrowers) face the risk that interest rates will fall, causing an increase in the value of their
debt obligation. Interest rate risk is a key concern of all financial institutions that operate in the debt markets.
These include banks, pension funds, and insurance companies. In this section, we analyze
the risk of the default-free fixed income securities (such as U.S. Treasury securities) that these institutions
hold.70
Duration
One measure of a fixed income security’s interest rate risk is its duration. Duration specifies the sensitivity of
the bond price to movements in yield. A specific formula for computing duration may be obtained by taking
the derivative of B with respect to y in equation (8.4). First, for the sake of mathematical convenience,
rewrite (8.4)
B  t 1 Ct (1  y ) t . (8.9)
n
Under this specification, we let Ct  C for t  1,  , n  1 and Cn  C  Fn for t  n . Now, take the derivative
of B with respect to y ,
dB t 1
 t 1 tCt (1  y ) .
n
(8.10)
dy
Multiply (8.10) by (1  y ) ,
dB t
 t 1 tCt (1  y ) .
n
(1  y ) (8.11)
dy
Finally, divide (8.11) by B ,
dB / B  C (1  y )  t 
 t 1 t  t  t 1 twt  D , (8.12)
n n
dy /(1  y ) B 
 
where wt  Ct (1  y )  t / B and where duration is defined as
D  t 1 twt . (8.13)
n
Equations (8.12) and (8.13) tell us two important things about duration. First, as the left-hand side of (8.12)
indicates, minus duration,  D , can be interpreted as the percentage change in the bond price, dB / B ,
induced by a change in the bond’s yield to maturity, dy , scaled by 1 /(1  y ) . Second, duration is the
weighted average time to maturity of a bond, where the weights, wt , are the present values of the payments
in each period.71 For coupon bonds, duration is less than time to maturity because some of the bond
payments –the coupons– are received in years prior to maturity of the bond. For non-coupon bonds, duration
equals time to maturity.
The scale factor 1 /(1  y ) on the left-hand side of (8.12) is cumbersome to account for. Most fixed
income portfolio managers prefer to use a measure of duration that is simply the percentage change in bond
price, dB / B , induced by a change in yield, dy . To find this expression, divide (8.12) by (1  y ) ,
70
Default risk is also an important consideration in fixed-income security risk management. Recall that on Chapter 4, we showed
how to hedge both the interest rate and default risk exposure of Mobil Oil bonds using T-bond and stock index futures contracts.
t
n C (1  yt ) B
Note that by virtue of equation (8.9), the weights, wt , sum to one, that is,  t 1 wt   t 1 t
n
71
 1
B B
dB / B  C (1  y ) t 
 t 1 t  t  t 1 twt  D . (8.14)
n n
dy /(1  y ) B 
 
is called modified duration and is more commonly used in interest risk management strategies. It is simply
the duration, as defined in (8.13), divided by (1  y ) . We use modified duration in a hedging application later
in this chapter. We now turn to a numerical example that uses the duration formulas (8.13) and (8.14).
EXAMPLE 8.1
What are the duration and the modified duration of an 8 percent, ten-year, $1000 bond, assuming annual
coupon payments and a required yield to maturity of 7 percent? From equation (8.4), we know the bond price
80 1,000
is B  t 1
10
  $1,070.24
t
(1.07) (1.07)10
This present value consists of the present value of 10 payments. In tabular form,
t Ct (1  y ) t Ct (1  y ) t wt twt
1 80.00 0.9346 74.77 0.0699 0.0699
2 80.00 0.8734 69.88 0.0653 0.1306
3 80.00 0.8163 65.30 0.0610 0.1831
4 80.00 0.7629 61.03 0.0570 0.2281
5 80.00 0.7130 57.04 0.0533 0.2665
6 80.00 0.6663 53.31 0.0498 0.2989
7 80.00 0.6227 49.82 0.0466 0.3259
8 80.00 0.5820 46.56 0.0435 0.3480
9 80.00 0.5439 43.51 0.0407 0.3659
10 1080.00 0.5083 549.02 0.5130 5.1299
Total 1,070.24 1.0000 7.3466
The duration of the bond is 7.3466, and the modified duration is 7.3466/1.07 = 6.8660. The modified
duration figure computed in the example predicts that if interest rates increase by 100 basis points, the bond
price will change by
dB / B   Dm  dy  6.8660  0 .01  6.8660% .
Conversely, a decrease in yield of 100 basis points implies a 6.8660% increase in bond price.
The formula for the duration of a bond shows that duration –price sensitivity– depends on the
maturity of the bond, on the coupon level, and on the yield to maturity. First, the greater the maturity, the
greater the duration, holding constant other characteristics of the bond. Second, the larger a bond’s coupon,
the smaller the duration. Coupon payments cause weight to be put on the early years in the duration formula.
In the case of a zero-coupon bond, duration equals maturity. Third, duration decreases with increases in the
yield to maturity. This is because an increase in the yield has a greater effect on the present value of a distant
coupon than on the present value of a nearby coupon.
Convexity
Modified duration is only an approximation of the percentage change in bond price for a given change in
yield. In fact, it is accurate only for infinitesimal shifts in yield. To assess the degree of error that may be
introduced, reconsider the results of Example 8. 1. The level of modified duration predicts that the bond
price will increase to $1,070.24(l.068660) = $1,143.72 if the yield drops to 6 percent. By using the bond
valuation formula (8.4), however, we know that the bond value is exactly $1,147.20 at a yield of 6 percent.
The difference between these prices is attributable to the fact that bond price is a nonlinear function of yield
to maturity.
Figure 8.1 illustrates the approximation error in this example. At a yield of 7 percent, the bond’s price
is $1,070.24. Modified duration depends on the derivative, dB / dy , evaluated at 7 percent. The derivative is
depicted in the figure by the straight-line tangent to the bond price curve at 7 percent. To estimate the change
in bond price due to a 100 basis point drop in yield, we draw a vertical line from 6 percent on the horizontal
axis to the straight line depicting the derivative, and then draw across horizontally to the vertical axis. The
estimated value of the bond based on modified duration is $1,143.72.. If the vertical line is continued upward
to the bond price curve, we find the exact value of the bond at a 6-percent yield, $1,147.20. The pricing
error, $3.48, is attributable to the failure of modified duration to account for the convex nature of the bond
pricing function.
We can achieve greater precision in measuring the bond’s responsiveness to yield shifts by also
accounting for the bond’s convexity. To understand convexity, expand the bond price as represented by (8.9)
into a Taylor series:
dB 1 d 2B
dB  dy  2
(dy ) 2   . (8.15)
dy 2 dy
FIGURE 8.1 Bond Price as a Function of Yield

Bond price, B
$1,147.20
$1,143.72
$1,070.24
6% 7% Yield to maturity, y
The error term,  , recognizes the fact that we have used only the first two terms of the Taylor series
expansion. Hence, our approximation for bond price changes will remain only approximate. Now, we drop
the error term and divide both sides by B ,
dB dB / B 1 d 2B 1
 dy  2
(dy ) 2   . (8.16)
B dy 2 dy B
We now define convexity as
1 d 2B 1
convexity  , (8.17)
2 dy 2 B
and rewrite (8.16), the percentage change in bond price, using (8.14) and (8.17), as
dB
  Dm dy  Convexity(dy ) 2 . (8.18)
B
The value of the second derivative of the bond price with respect to a change in yield is
d 2B  Ct 
 t 1 t (t  1) 
n
t 2 
. (8.19)
 (1  y ) 
2
dy
EXAMPLE 8.2
What is the convexity of an 8 percent, ten-year, $1000 bond, assuming annual coupon payments and a
required yield to maturity of 7 percent? From Example (8. 1), we know the bond price is $1,070.24. The
components of convexity (8.19) are
Ct t (t  1)Ct
t t (t  1)
(1  y ) t 2 (1  y ) t 2
1 2 65.3038 130.61
2 6 61.0316 366.19
3 12 57.0389 684.47
4 20 53.3074 1,066.15
5 30 49.8200 1,494.60
6 42 46.5607 1,955.55
7 56 43.5147 2,436.82
8 72 40.6679 2,928.09
9 90 38.0074 3,420.67
10 110 479.5329 52,748.62
Total 67,1231.77
The value of d 2 B / dy 2 for this bond is 67,231.77, so the convexity is 1/2 x 67,231.77  1/1,070.24  31.4098 .
To complete our illustration, we now use convexity in conjunction with modified duration to arrive at
a more accurate estimate of the percentage change in bond price attributable to a 100 basis point fall in yield.
dB
  Dm dy  Convexity(dy ) 2  6.8662  (0.01)  31.4098  (0.01) 2  7.180% .
B
In other words, based upon modified duration and convexity, the bond price is expected to rise to
$1,147.08 if yield falls by 100 basis points. Note that the error in prediction has been reduced from $3.48 to
$0.12 by accounting for convexity.
All bonds with fixed payment schedules plot as convex curves in y , as shown in Figure 8.1, although
different bonds have different degrees of convexity. Mortgage backed securities (MBS’s), however,
sometimes exhibit “negative convexity.” That is, they plot as curves that are concave to the origin. This is
because their payment schedules are uncertain. The underlying mortgages have the option to prepay before
maturity, and they will choose to prepay just when it is most inconvenient for the holder of the MBS. When
interest rates fall, mortgages are prepaid, so the duration and price sensitivity of the MBS are reduced.
Consequently, while fixed-payment bonds increase in price, the price of MBS’s increases little and
may even decline. When interest rates rise, mortgage prepayments slow, resulting in duration and price,
sensitivity increases for the MBS. Consequently, the price decline resulting from an increase in interest rates
is relatively greater for a MBS than for a fixed-payment bond.
8.4 TERM STRUCTURE OF INTEREST RATES

Up to this point in the chapter, we have assumed that all cash flows of a fixed income security are discounted
at a single rate-the yield to maturity. This assumption may be inappropriate since lenders may demand
different rates on short-term loans than on long-term loans. The relation between the level of interest rates
and the time to maturity of the loan is the focus of this section.
Yield Curve
The simplest way to consider the relation between expected bond return and time to maturity is to plot the
yield to maturity of U.S. Treasury obligations versus the term to maturity (making sure that the bonds have
the same default risk, no imbedded options such as callability, and no differential tax privileges). This
relation is referred to as the yield curve. The yield curve is usually plotted using Treasury bonds and notes.
The yield to maturity differs by maturity because expected future interest rates are different for different
maturities and because risk premia differ by maturity.
The yield to maturity of a coupon-bearing bond can be a misleading reflection of its expected rate of
return, however. To see this, consider the following situation. Suppose there exist two discount Treasury
bonds, one with a one-year maturity and a price of 90.91, and the other with a two-year maturity and a price
of 81.16. Assuming each of these bonds is redeemed at a par value of 100 at their respective maturities, the
expected yield on the one-year bond is 100/90.91 - 1  10.00% , and the expected yield on the two-year bond
is (100/81 - 16)1/2 - 1  11.00% .
Note that if the Treasury decided to issue other one- and two-year discount bonds, they must have the
same prices as the existing issues, otherwise, costless arbitrage profits could be earned.
Now, suppose that the Treasury also has a two-year, 12-percent coupon bond. The 12-percent issue
pays 12 at the end of one year, and 112 at the end of two years. The price of this coupon-bearing Treasury
bond has to be
12 112
B   101.81 ,
1.10 (1.11) 2
otherwise, costless arbitrage profits are possible.72 But, if the two-year coupon-bearing bond has a price of
101.81, its yield to maturity is 10.94 percent, that is, the solution to
12 112
101.81  
(1  y ) (1  y ) 2
On the surface, it might appear that the two-year coupon bond expected yield to maturity of 10.94 percent
conflicts with the two-year discount bond expected yield to maturity of 11 percent. However, arbitrage
profits are not possible. The discrepancy between the rates arises because the two-year coupon bond has an
72
Note that buying 0.12 units of the one-year discount bond and 1.12 units of the two-year discount bond produces a cash flow
stream exactly the same as the two-year coupon-bearing bond at a cost of 0.12(90.91) + 1.12(81.16) = 101.81.
“actual” term to maturity of less than two years. If you buy the coupon bond, you will receive 12 at the end
of one year and 112 at the end of two years; so, in essence, you have a portfolio of two discount bonds. The
first discount bond is worth
12
Bd1   10.91
1.10
and the second discount bond is worth
112
Bd 2   90.90
(1.10) 2
The average term to maturity of the bond portfolio (or the coupon-bearing bond) is, therefore,
 10.91   90.90 
Average term to maturity   1   2  1.893 years .
 101.81   101.81 
The 10.94-percent yield is not an expected return on a bond with two years to maturity, but rather on a bond
with a maturity of approximately 1.893 years.
Two principles underlie this discussion. First, to value coupon-bearing bonds precisely, all cash flows
of the bond should not be discounted at the same rate. Instead, each cash flow should be discounted at the
zero-coupon or discount bond rate, rt , which coincides with the timing of the cash flow,
 n Ct Fn 
B  t 1  
 (1  rt ) (1  rn ) n 
t
The rate rt is called the spot rate of interest on a t - period loan. Second, to estimate the relation between
expected bond returns and time to maturity, only zero-coupon or spot rates of interest should be used. The
yields to maturity on coupon bonds should not be used because the term to maturity of the bond overstates its
true economic life. The economic life of a coupon bond is a weighted average life of its constituent discount
bonds, that is,
 1 n  Ct   Fn  
Average term to maturity    t 1  t   n .
n  
(8.21)
 (1  rt )   (1  rn )  
t
B
Note that this weighted average term to maturity is the same as the bond’s duration defined in (8.12).73 In
practice, the relation between zero-coupon or spot rates of interest and time to maturity is referred to as the
zero-coupon yield curve. In this section, we refer to the relation as the term structure of interest rates. The
term structure can be read directly from the yields to maturity of Treasury strips. For example, Table 8.3
shows that the term structure was upward sloping on November 13, 1991. The one-year, two-year, five-year,
and ten-year spot rates were 5.05, 5.68, 6.84, and 7.78, respectively.
The shape of the term structure of spot rates affects the relative values of bonds with different
coupons. We have already illustrated this point when we showed that a two-year coupon bond would yield
10.94 percent to maturity while a two-year discount bond yielded 11 percent. In general, for bonds of the
73
Duration and maturity are the same only for a discount bond because the coupon terms, Ct , in the summation of (8.21) are zero.
same maturity, when the term structure is upward sloping, the yield to maturity decreases as the coupon
payment increases; when the term structure is downward sloping, the yield to maturity increases with a larger
coupon. This result can be explained in two ways. First, a coupon bond has a duration that is less than its
maturity. Its yield to maturity corresponds to its, duration, not its maturity. Second, when the term structure
is upward sloping, investors prefer coupon bonds because it is expected that the coupon payments can be
invested at higher future rates. They, therefore, bid up prices of coupon bonds and lower their yield to
maturity. Correspondingly, when the term structure is downward sloping, investors prefer low coupon bonds
because it is expected that the coupons can be invested only at lower future rates. Only when the term
structure of interest rates is flat do all bonds have the same yield to maturity.
Spot Rates and Forward Rates

Spot rates of interest are rates observable today. The spot rate on a three-month, default-free security is the
rate of return promised on a three-month T-bill. The notation that we use to describe the spot rate on a
t - period bond is rt . Forward rates of interest are interest rates on loans in the future and are implied from
the current term structure of spot rates. The forward rate on a t - period loan in period n , denoted n f t , is
computed using
(1  rnt ) nt
(1  n f t ) t  . (8.22)
(1  rn ) n
The forward rate on a one-year loan today, for example, is computed as

(1  0 f1 )1  (1  r1 )1 ,
and the forward rate on a one-year loan in three years is
(1  r31 ) 31
(1  3 f1 )1  .
(1  r3 ) 3
EXAMPLE 8.3
Suppose the one-year spot rate of interest is 10 percent and the two-year spot rate is II percent. What is the
implied one-year forward rate in one year?
Substituting the parameters of the example into equation (8.22), we find that
(1  r2 ) 2
(1 1 f1 ) 
1
1 f1  12.01% .
(1  r1 )1
EXAMPLE 8.4
Using the values reported in Table 8.3, find the 91-day forward rate implied by selling the December 19,
1991 T-bill and buying the March 19, 1992 T-bill. The effective annual rate of return of the December 19th
365 / 360
 100 
T-bill implied is r     1  4.562 .
100  4.39(36 / 360) 
The effective annual rate of return of the March 19th T-bill is
365 / 127
 100 
r   1  4.858
100  4.64(127 / 360) 
Given these two spot rates of interest, the implied forward rate of interest on a 91-day T-bill in 36 days is the
solution to
(1.04858)127 / 365
(1  36 f 91 ) 
91/ 365
36 f 91  4.975%
(1.04562)36 / 365
The implied forward rate of interest is 4.975 percent.
8.5 SHORT-TERM INTEREST RATE FUTURES CONTRACTS

The most actively traded short-term interest rate futures contracts are the Chicago Mercantile Exchange’s T-
bill and Eurodollar futures contracts. The Treasury bill futures contract requires the delivery of a $1,000,000
face value T-bill with 91days to maturity. The contract expires the business day before the date on which the
new 91-day T-bill is issued. The newly issued 91-day T-bill, the seasoned 182day, and the seasoned 364-day
T-bills are eligible for delivery on this futures contract.
The price of a T-bill futures contract is an index value based on the bank discount. The price reported
for the December 1991 T-bill futures in Table 8.2, for example, is 95.34. This does not mean that the futures
buyer will pay 95.34 percent of par when she buys the T-bill at the futures maturity. The index price, 95.34,
implies that the bank discount on the bill is 100 - 95.34 = 4.66 on an annualized basis. Since the futures
contract requires the delivery of a 91-day T-bill, the annualized discount is adjusted to a 91-day period using
the banker’s convention of a 360-day year. The 91-day discount is 4.66(91/360) = 1.178. Thus, if we bought
the December 1991 T-bill futures contract on November 13, 1991, at the reported price of 95.34, we would
be entering into a commitment to buy a 91-day, $1,000,000 T-bill on March 19, 1992, at a price of 98.822
percent of the face value of the T-bill or $988,220. The implied forward rate of interest on this T-bill is
365 / 91
 100 
   1  4.866% .
 98.822 
It is interesting to note that the implied forward rate of interest from the futures contract, 4.868 percent, is
less than the 4.975 percent rate implied by the T-bills in Example 8.4. Both rates apply to the 91-day period
beginning on December 19, 1991. If one could borrow at 4.868 percent and lend at 4.975 percent, an
arbitrage profit could be earned. In effect, this can be done by the following arbitrage transactions: (a) sell
the December 19, 1991 T-bill for $9,956.10; (b) buy the March 19, 1992 T-bill for $9,836.31; and (c) sell the
December 1991 T-bill futures contract at $9,882.20. Transactions (a) and (b) have the net effect of lending
money over the period December 19, 1991 to March 19, 1992; transaction (c) commits the investor to
borrowing money for the same period. The borrowing and lending, however, may never occur because the
transactions can be closed out on December 19, 1991. On that date, the arbitrageur covers the short position
in the maturing December 19 T-bill by paying $10,000, thereby incurring a cost of $43.90. The March 19 T-
bill purchased on November 13 now has 91 days to maturity and can be delivered against the futures contract
for $9,882.20, a net gain of $45-89. Finally, the net proceeds from the T-bill transactions on November 13,
1991, $119.79, have earned interest at, say, 5 percent and, after 36 days, are now worth $120.37. The
arbitrage profit realized on December 19, therefore, totals
$120.37  $45.89 - $43.90  $122.36 .
The Eurodollar futures contract is a commitment to transact a $ 1,000,000, three-month Eurodollar deposit.
Delivery never takes place since a Eurodollar futures contract is cash settled. Cash settlement occurs on the
second London business day before the third Wednesday of the contract month. The settlement price of the
Eurodollar futures contract on the expiration day is computed in such a way as to minimize the variation in
the quoted Eurodollar deposit rates. During the last day of trading, a random sample of approximately twelve
rates are taken from the twenty-plus approved banks in the London Euro-dollar market. The rates are then
ranked from highest to lowest, and the highest and lowest rates are discarded. The remaining ten rates are
averaged, and the average rate is subtracted from 100 to determine the settlement price.
As noted in the previous paragraph, quoted Eurodollar futures prices are actually index values, that is,
the value reported in the financial press is 100 less the Eurodollar interest rate. Thus, if we buy the March
1992 Eurodollar futures at the price reported in Table 8.2, 94.94, the implicit agreement that we are entering
into is to buy a $1,000,000 three-month Eurodollar certificate of deposit on March 16 1992 (the second
London business day before the third Wednesday of the future’s contract month), where the stated interest
rate on the deposit is 100 - 94.94, or 5.06 percent. The effective three-month forward interest rate on such a
deposit is, therefore,
365 / 92
100  5.06(92 / 360) 
r   1  5.23% .
 100
Note that the three-month interval from March 16, 1992, through June 16, 1992, has 92 days.
8.6 LONG-TERM INTEREST RATE FUTURES CONTRACTS

The most active long-ten-n interest rate futures contract is the T-bond contract on the Chicago Board of
Trade.74 The CBT’s Treasury bond futures contract is a commitment to deliver a nominal 8 percent,
$100,000 face-value U.S. Treasury bond with a least fifteen years to maturity or to first call date, whichever
comes first. The seller of the futures contract has the option to deliver any of the eligible issues on any date
during the delivery month. Whether an 8-percent coupon issue is available for delivery is unimportant since
the futures contract allows for the delivery of any T-bond with a long enough maturity. To remove the effects
of different bonds having different coupon rates, the CBT designed a system of conversion factors.
Conversion Factor and Invoice Price

To understand the CBT’s system of conversion, recall first that the lower the coupon rate, the lower the
bond’s price, other factors being held constant [see equation (8.4)]. Since the seller of the T-bond futures
contract can deliver any U.S. Treasury bond with at least fifteen years to maturity or to the first call date, a
method of conversion is needed to offset the economic incentive to deliver the lowest coupon bond. The
CBT’s system adjusts the futures price, which is based on an 8-percent coupon, to a price that corresponds to
the coupon of the issue being delivered. To illustrate the principle underlying the conversion, consider the
price of an 8-percent, fifteen-year bond with annual coupons and a yield to maturity of 8 percent. Using
equation (8.4), the bond price is
8 100
B  t 1
15
  100.00 .
(1.08) (1.08)15
t
Now, consider the price of a 12-percent, fifteen-year bond with fifteen years to maturity and the same 8-
percent yield. The bond price is
12 100
B  t 1
15
  134.24 .
(1.08) (1.08)15
t
Note that the only difference between the two bonds is that the second bond has higher coupon payments.
Owning the 12-percent coupon bond is like owning 1.3424 8-percent bonds. Since the futures price is based
74
For an interesting analysis of why this contract supplanted the earlier GNMA contract, see Johnston and McConnell (1989).
on an 8-percent coupon bond, the futures price is multiplied by a conversion factor of 1.3424 to compute the
amount paid (delivery price) by the long to the short if the short delivers the 12-percent coupon issue.
The actual formula for computing the conversion factor is slightly more complex than what is demonstrated
in the above example because coupons are paid on a semi-annual basis, and, in general, the next coupon
payment is made in less than six months (i.e., we are part of the way through the current coupon period). The
actual formula for the conversion factor, CF , is
 C C   C (6  X )
CF  (1  y / 2)  X / 6    [1  (1  y / 2) 2 n ]  1(1  y / 2) 2 n    , (8.23)
 2  y   2 6
where C is the annual coupon rate of the bond in decimal form, y  0.08 , n is the number of whole years to
first call, if the bond is callable, or the number of years to maturity, and X is the number of months that the
maturity exceeds n , rounded down to the nearest quarter (e.g., X  0, 3, 6, 9 ). Note that if X  0, 3, 6 , the
formula (8.23) is used directly. If X  9 set 2n  2n  1, X  3 , and calculate as above.75 Computer programs
are available to perform the computation (8.23). The values in Table 8.5, for example, were generated using
a program called OPTVAL. Alternatively, the CBT and others publish and distribute conversion factor
tables.
TABLE 8.5 Conversion factors for the U.S. Treasury Bonds eligible for delivery on the CBT’s T-bond
futures contract. These factors convert different coupon issues to yield 8 percent.
1/8 1/4 3/8 1/2 5/8 3/4 7/8
Years-Months 9% 9 % 9 % 9 % 9 % 9 % 9 % 9 %
25-0 1.1074 1.1208 1.1343 1.1477 1.1611 1.1745 1.1880 1.2014

25-3 1.1075 1.1210 1.1345 1.1479 1.1614 1.1749 1.1883 1.2018
25-6 1.1081 1.1216 1.1351 1.1486 1.1621 1.1756 1.1892 1.2027
25-9 1.1082 1.1217 1.1353 1.1488 1.1624 1.1759 1.1895 1.2030
26-0 1.1087 1.1223 1.1359 1.1495 1.1631 1.1767 1.1903 1.2039

26-3 1.1088 1.1225 1.1361 1.1497 1.1633 1.1770 1.1906 1.2042
26-6 1.1094 1.1230 1.1367 1.1504 1.1640 1.1777 1.1914 1.2051
26-9 1.1094 1.1232 1.1369 1.1506 1.1643 1.1780 1.1917 1.2054
27-0 1.1100 1.1237 1.1375 1.1512 1.1649 1.1787 1.1924 1.2062

27-3 1.1100 1.1238 1.1376 1.1514 1.1652 1.1789 1.1927 1.2065
27-6 1.1105 1.1244 1.1382 1.1520 1.1658 1.1796 1.1935 1.2073
27-9 1.1106 1.1245 1.1383 1.1522 1.1660 1.1799 1.1937 1.2076
28-0 1.1111 1.1250 1.1389 1.1528 1.1666 1.1805 1.1944 1.2083

28-3 1.1111 1.1251 1.1390 1.1529 1.1668 1.1807 1.1947 1.2086
28-6 1.1116 1.1256 1.1395 1.1535 1.1675 1.1814 1.1954 1.2093
28-9 1.1117 1.1257 1.1396 1.1536 1.1676 1.1816 1.1956 1.2096
C 
75
Note that if X  0 the formula (8.23) reduces to CF   1  (1  y / 2)2n   (1  y / 2)2n 
y 
TABLE 8.5 Cont. Conversion factors for the U.S. Treasury Bonds eligible for delivery on the CBT’s T-
bond futures contract. These factors convert different coupon issues to yield 8 percent.
1/8 1/4 3/8 1/2 5/8 3/4 7/8
Years-Months 9% 9 % 9 % 9 % 9 % 9 % 9 % 9 %
29-0 1.1121 1.1262 1.1402 1.1542 1.1682 1.1822 1.1963 1.2103

29-3 1.1122 1.1262 1.1403 1.1543 1.1684 1.1824 1.1965 1.2105
29-6 1.1126 1.1267 1.1408 1.1549 1.1690 1.1830 1.1971 1.2112
29-9 1.1127 1.1268 1.1409 1.1550 1.1691 1.1832 1.1973 1.2114
30-0 1.1131 1.1273 1.1414 1.1555 1.1697 1.1838 1.1980 1.2121

30-3 1.1131 1.1273 1.1415 1.1556 1.1698 1.1840 1.1981 1.2123
30-6 1.1136 1.1278 1.1420 1.1562 1.1704 1.1846 1.1988 1.2130
30-9 1.1136 1.1278 1.1420 1.1562 1.1705 1.1847 1.1989 1.2131
To illustrate the use of the conversion factor system, suppose that we are considering delivery of the 9s of
November 2018 on the March 1992 T-bond futures contract. This bond is eligible for delivery because on
March 1, 1992, it has more than fifteen years to maturity. Specifically, on March 1, 1992, the 9s of
November 2018 have 26.50 years to maturity (rounded down to the nearest quarter). Using Table 8.5, the
conversion factor of this bond is 1.1094. In other words, in place of delivering the hypothetical 8-percent,
fifteen-year bond on the March 1992 futures contract, we can deliver the 9s of November 2018, but the buyer
is going to have to pay 1.1094 times the prevailing futures price.
On the delivery date, the seller of the T-bond futures delivers an eligible T-bond to the buyer of the
T-bond futures contract. In return, the buyer must pay the invoice price to the bond seller. The amount of the
invoice price will be the sum of the futures price times the conversion factor of the delivered bond and the
accrued interest on the delivered bond. For example, suppose that on March 15, 1992, the March 1992
futures contract is priced at 96-18. Like the underlying bonds, the decimal part of the price is the number of
32nds, so the futures price is 96.5625. If we sell the futures and promptly deliver the 9s of November 2018 to
the futures contract buyer, the invoice price paid by the buyer equals .965625 (the futures price in decimal
form) times 100,000 (the denomination of the futures contract) times 1. 1094 (the conversion factor of the 9s
of November 2018 as of March 1, 1992), or $107,126.44, plus the accrued interest on the 9s of November
2018 as of March 15, 1992, $2,991.76 [i.e., 0.045(121/182)(100,000)]. The total invoice price is
$107,126.44  $2, 991.76  $110, 118.20 .
Cheapest to Deliver
In principle, the system of conversion factors is intended to make the short indifferent about which bond he
delivers. This means that if we are at time T –the end of the futures contract life– the profits from selling the
futures and buying and delivering bond i , computed as
 i ,T  FT (CFi )  AI i ,T  Bi ,T  AI i ,T  FT (CFi )  Bi ,T ,
where FT (CFi )  AI i ,T , is the invoice price received from delivering bond i and Bi ,T  AI i ,T , is the price paid
for the purchase of bond i , should equal zero.
In practice, however, one of the eligible delivery bonds is “cheapest to deliver” because the system of
conversion factors is not exact. Each bond will have a different value of  i,T , the bond with the highest  i,T
is the cheapest to deliver.
Its value of  i ,T , however, will be equal to zero. If it were positive, costless arbitrage profits could be
earned by the short at the expense of the long. If it were negative, costless arbitrage profits could be earned
by the long at the expense of the short.
The computed profits for all other deliverable bonds will be negative. However, negative profits for
these bonds do not imply arbitrage opportunities. To capture these “gains,” the long would need to take
delivery of a bond issue that is not the cheapest to deliver. Since it is not rational for the short to deliver any
issue other than the cheapest to deliver, no arbitrage gains are possible.
A cheapest-to-deliver issue arises because the conversion factors are derived by discounting the cash
flows of all bonds at 8 percent. Using 8 percent assumes that coupon payments can be reinvested at 8
percent. If the average future interest rate at which the coupons can be reinvested exceeds 8 percent,
however, investors prefer high-coupon bonds over low-coupon bonds (when each bond is valued by
discounting at 8 percent). The cheapest-to-deliver bond is, therefore, the low-coupon bond. If the average
future interest rate at which coupons can be reinvested is less than 8 percent, investors prefer low-coupon
bonds over high-coupon bonds, so the cheapest to deliver is the high-coupon bond. Only when the term
structure of interest rates is flat at 8 percent, will all bonds be equally desirable for delivery. If the yield
curve is above 8 percent, low-coupon bonds are the cheapest to deliver, and if the yield curve is below 8
percent, high-coupon bonds are the cheapest to deliver.
8.7 COST-OF-CARRY RELATION

Under the assumptions that the cheapest-to-deliver bond issue i is known and that it does not change
through time, the cost-of-carry relation between the futures price and the cheapest-to-deliver bond may be
written
( Bi ,0  AI i ,0 )e rT  AI i ,T  t 0 Ci ,t e r (T t )
T
F0  . (8.24)
CFi
The left-hand side of the equation is the futures price at time 0. The maturity of the futures contract is T
periods hence. The first term in the numerator of the right-hand side of the equation, ( Bi , 0  AI i ,0 )e rT . is the
time 0 cost of the bond taken forward to time T at the riskless rate of interest (i.e., the bond purchase is
financed at the short-term riskless interest rate). The second and third terms in the numerator represent
interest earned on the bond-accrued interest, AI i ,T , received when the bond is delivered against the futures

T
contract and the future value of the coupons received (if any), t 0
Ci ,t e r (T t ) . The conversion factor CFi in
the denominator “converts” bond i into the hypothetical 8-percent coupon issue upon which the futures
contract is designed.
Prior to discussing the cost-of-carry relation in more detail, it is worth noting that the “short-term
riskless rate of interest” used to finance the purchase of bonds required in the arbitrage transactions that drive
(8.24) is usually the rate on a repurchase agreement or “repo.”. Repurchase agreements are collateralized
loans. They involve a commitment to sell and then later to buy back a specific bond issue (presumably at the
maturity of the futures contract).76 The agreement specifies the date on which the bond will be repurchased,77
as well as the interest rate that will be paid on the loan. The dollar interest paid on the loan is computed as
Days repo is outstanding

Interest  Principal amount  Repo rate 
360
76
Repurchase and reverse repurchase agreements are discussed at length in Stigum (1990).
77
When the term of the loan is one day, it is called an overnight repo. Terms of greater than one day are term repos.
Note that, under this arrangement, the lender has committed to consummating actions opposite the borrower,
that is, the lender has entered into an agreement to buy and then later to sell the underlying bond. For this
reason, the lender is said to have a reverse repurchase agreement, a “reverse repo” or, simply, a “reverse.”
Both the borrower and lender gain from these agreements. The borrower gets a lower rate than he might
otherwise get at the bank, and the lender gets a higher rate than he might otherwise get on short-term, highly
liquid investments.
Before maturity, as at maturity, the futures price is based on the price of the cheapest to deliver, and
the cheapest to deliver is determined by finding the bond with the highest “cash and carry” portfolio profit,
 i ,0  F0 (CFi )  AI i ,T  t 0 Ci ,t e r (T t )  ( Bi , 0  AI i ,0 )e rT .
T
(8.25)
Again, the highest value of profit equals zero; otherwise arbitrage profits are possible. This identifies
the cheapest-to-deliver issue. All other profits will be less than zero. In other words, the futures price will be
less than in (8.24) for all bonds other than the cheapest to deliver.
Although (8.25) allows us to identify which bond is cheapest to deliver at time 0, there is no assurance that
this bond will also be cheapest to deliver at time T. Since the identity of the cheapest to deliver at time T is
uncertain, (8.24) does not hold as an equality, even for the bond that is currently cheapest to deliver. Indeed,
the short has a valuable quality option that gives him the right to choose which bond to deliver at time T.
Although the short may have entered a cash-and-carry position when bond i was cheapest to deliver, if, at
maturity, bond j is cheapest, the short can profit by selling bond i, buying bond j, and then delivering bond j
on his short futures commitment. Because this option to switch bonds is valuable, the investor doing cash-
and-carry arbitrage is willing to sell futures at a price below the price specified by the cost of carry on the
right-hand side of (8.24), that is,
( Bi ,0  AI i , 0 )e rT  AI i ,T  t 0 Ci ,t e r (T t )
T
F0  , (8.26)
CFi
or, alternatively,
( Bi ,0  AI i ,0 )e rT  AI i ,T  t 0 Ci ,t e r (T t )
T
F0   Quality option , (8.27)

CFi
where bond i is the current cheapest to deliver. The value of the quality option embedded in the T-bond
futures contract is estimated in Chapter 11.
The short futures also has a timing option that allows a choice about when during the contract month to
deliver. The most valuable element in the timing option is called the wildcard option. In the delivery month,
the futures price at which delivery is made is the settlement price established when the market closes at 2:00
PM. The short has until 8:00 PM to declare delivery. Obviously, if news arrives that justifies a decline in
bond prices, the short will choose to make delivery at the already established settlement price.
8.8 HEDGING WITH INTEREST RATE FUTURES CONTRACTS
Short-Term, Long Hedge

Interest rate futures can be used to lock in forward interest rates. Suppose, for example, that on November
13, 199 1, a company anticipates a cash inflow of $ 1,000,000 on March 16, 1992. The cash, when it is
received, will be placed in a three-month certificate of deposit until summer when it will be used to partially
finance a major capital expenditure that the firm plans. Suppose also that the company’s financial analyst
expects three-month CD rates to fall to a level of 4 percent by March, while the current implied three-month
forward rate of the March 1992 Eurodollar futures, based on its reported price of 94.94, is 5.06 percent. What
can the company do to lock in the higher rate of interest?
A simple solution to this problem is to buy the March 1992 Eurodollar futures contract at the reported price
of 94.94. When the $1,000,000 is received on March 16, 1992, the price of the Eurodollar futures will be
approximately the same as the spot Eurodollar rate since the futures is near expiration. Assume that our
analyst is correct in her prediction, and the spot rate is 4 percent. When we sell the futures position, our profit
is (9600 - 9494) basis points times $25, or $2,650. The Eurodollar deposit on March 16, 1992, is, therefore,
$1,002,650, which, at a 4-percent rate implies a deposit balance of $1,012,787.91 on June 15, 1992. Thus, in
spite of the fact that the nominal rate is lower, the earned interest income amounts to a nominal rate of
1.2788(360/91), or 5.06 percent, exactly the desired result. Note that it does not matter what the spot rate is
on March 16, 1992-the 5.06-percent rate is locked in regardless.
Long-Term, Short Hedge

Earlier we developed the concept of modified duration to assess the interest rate risk of a bond. Recall that
modified duration is an approximation for the percentage change in bond price with respect to a change in
yield. From a fixed-income security portfolio risk management standpoint, it is useful to recognize that the
duration of a portfolio of bonds or fixed-income securities is simply the market-value-weighted average of
the durations of the constituent bonds. It is also useful to know that, in the absence of the options imbedded
in the T-bond futures contract, the duration of the T-bond futures is approximately equal to the duration of
the cheapest-to-deliver T-bond.
To develop a framework for using T-bond futures to manage the risk of a fixed-income security
portfolio, define the following notation:
DP  modified duration of fixed - income security portfolio P.

DF  modified duration of T - bond futures contract.
D*  modified desired duration exposure for fixed - income portfolio.
P  current market value of fixed - income portfolio.
F  current futures price.
h  optimal number of futures contracts to buy (sell).
Under this notation, the dollar change in the value of the unhedged fixed-income portfolio is DP P times the
interest rate change. If we buy h futures contracts against his fixed-income investment, the dollar change in
the overall portfolio is ( DP P  hDF F ) times the interest rate change, which we equate to the dollar change in
the hedged portfolio at the desired duration level, D* P , that is,
D * P  DP P  hDF F , (8.28)
in order to determine the optimal hedge ratio. Rearranging to isolate h , we get
P( D *  DP )
h . (8.29)
DF F
EXAMPLE 8.5
Suppose that the cheapest-to-deliver bond (and hence the T-bond futures) has a duration of 12.50. Suppose
also that the duration of the bond portfolio that we are managing is 10.00 and that the market value of the
portfolio is $50,000,000. If today’s date is November 13, 1991, and we wish to hedge completely against
movements in the level of long-ten-n rates until the end of February 1992, how many March 1992 T-bond
futures contracts should we sell?
50,000,000(0 - 10.00)
Substituting the example values into equation (8.29), we get h   - 403.15
12.50  0.9921875  100, 000
Asset Allocation
At its most basic level, portfolio management involves a decision concerning what types of assets should be
purchased. For example, a fund manager might choose to invest 40 percent of the fund in stocks, 40 percent
in bonds, and 20 percent in real estate. Deciding what proportion of fund wealth to place in each asset
category is called the asset allocation decision.
Once the asset allocation decision is made and fund wealth is invested, dramatic changes to the
allocation are usually avoided because the transaction costs of liquidating assets in one category and buying
assets in another are excessive. Instead, fund managers use futures contracts to change the asset allocation
indirectly. To demonstrate, assume that a fund consists of S in stocks and B in bonds, for a total value of
V  S  B . Now, suppose the fund manager wants to change the amount invested in long-term bonds from
B to B * . The bond portfolio has a modified duration of DB . Rather than selling (buying) stocks to buy (sell)
bonds, the portfolio manager can effectuate the change by buying (selling) T-bond futures contracts. She
wants her bond portfolio to have income DB B * if interest rates fall 1 percent. She plans on generating that
amount with income from the current bond portfolio, DB B , and income from a T-bond futures position,
hDF F , that is,
DB B*  DB B  hDF F . (8.30)
Rearranging (8.30) to isolate h , we get

D ( B*  B)
h B . (8.31)
DF F
Note that if the dollar investment in bonds is to be reduced (i.e., B  B * ), T-bonds futures contracts are sold,
and, if the dollar investment in bonds is to be increased (i.e., B  B * ), T-bond futures contracts are
purchased. Presumably, the reduction (increase) in bond investment is then transferred to stocks through
buying (selling) stock index futures contracts.
EXAMPLE 8.6
A fund manager currently has $50,000,000 in a stock portfolio whose composition matches the S&P 500 and
$50,000,000 in a bond portfolio whose modified duration is 12.00. Believing that stocks are going to do
extraordinarily well over the next three months, the fund manager wants to take advantage of the impending
stock market rise and to eliminate his interest rate risk exposure. Unfortunately, liquidating bonds and buying
stocks is expensive, particularly if, at the end of the three months, the manager wants to return to his fifty-
fifty portfolio mix. How can the fund manager use T-bond and S&P 500 futures to carry out his plans?
Assume that the cheapest-to-deliver bond (and hence the T-bond futures) has a duration of 9.00 and that the
price of a three-month T-bond futures contract is 96.00. Also, assume that the three-month S&P 500 futures
is priced at 325.
First, with respect to eliminating the interest rate exposure, the number of T-bond futures to sell is 694.44:
12.00(0 - 50,000, 000)

h  694.44 .
9 x 0.96 x 100,000
This action is tantamount to liquidating the bond investment. Second, to take a long position of $50,000,000
in stocks using the S&P 500 futures, the number of contracts to buy is
50,000, 000
h  694.44 .
325.00  500
8.9 SUMMARY
Following an introduction to the particular interest rate futures markets in the U.S., specific pricing and yield
conventions governing the trading of fixed-income securities are discussed. The major focus of this chapter
is interest rate risk management. To this end, the notions of modified duration and convexity are introduced
and applied. Following that, the relation between short-term and long-ten-n rates, that is, the term structure
of spot interest rates, is presented. From the spot rates, forward rates of interest are derived, and the forward
rates implied by cash and futures prices are compared. A detailed discussion of the specifics of T-bond
futures contract delivery and pricing is provided. The chapter concludes with two interest rate risk
management illustrations.
9 CURRENCY FUTURES CONTRACTS
9.1 MARKETS FOR CURRENCIES

 Interbank Spot and Forward Market
 Futures Market
9.2 CURRENCY QUOTATIONS AND TRIANGULAR ARBITRAGE
9.3 STRUCTURE OF CURRENCY FUTURES PRICES
9.4 FACTORS AFFECTING EXCHANGE RATES

 Purchasing Power Parity
 The Monetarist Approach to Exchange Rate Determination
 Balance of Payments Approach to Exchange Rate Determination
 The Asset Market Approach to Exchange Rate Determination
9.5 RETURNS TO SPECULATORS IN CURRENCIES

 Spot Speculation and Futures Speculation
 Forecasting Exchange Rates
9.6 HEDGING CURRENCY RISK

 Transactions Risk
 Balance Sheet Risk
9.7 SUMMARY
9 CURRENCY FUTURES CONTRACTS
Currency futures contracts were developed in response to the shift from fixed to flexible exchange rates in
1971. At that time, the United States ceased redeeming dollars for gold. By 1973, most countries stopped
maintaining the price of their currencies with respect to the dollar and allowed market forces to determine
their exchange rates. The increased volatility of exchange rates created a demand for currency futures
markets both as speculative and hedging vehicles.78
9.1 MARKETS FOR CURRENCIES79
Futures Market
The U.S. futures market in currencies operates like futures in any other item. Table 9.1 gives contract terms
for the major foreign currency contracts, all of which are traded on the Chicago Mercantile Exchange's
International Monetary Market. Prices are quoted in U.S. dollars per unit of foreign currency. An example of
currency futures price quotes is contained in Table 9.2. While Table 9.1 indicates that eight maturity months
are available, only the nearby months are actively traded. Unlike other futures contracts, delivery of
currencies has relatively few complications. Only one "grade" of currency is available, delivery takes place
on a specific date during the delivery month, and no transportation costs are incurred.
TABLE 9.1 Currency futures contracts specifications (most active contracts in U.S. markets).
Trading
Contract Months80 Last Trading Day Delivery
Hours
7:20-2:00 1,3,4,6,7,9,10,12 Two business days prior to third
Third Wednesday
(CST) current Wednesday
Security (Exchange) Units/ Minimum Price Fluctuation
Canadian Dollar (CME) 100,000Can$/0.0001 ($10)
Deutsche Mark (CME) 125,000DM/0.0001 ($12.50)
French Franc (CME) 250,000FF/0.00005 ($12.50)
Japanese Yen (CME) 12,500,000yen/0.000001 ($12.50)
British Pounds (CME) 62,500pounds/0.0002 ($12.50)
Swiss Franc (CME) 125,000/0.0001 ($12.5.0)
Interbank Spot and Forward Market

By far the largest market in currencies is the interbank market. Major banks around the world trade
currencies on a 24-hour basis. Banks supply currencies to their business customers and even out their
positions by trading with other banks. Table 9.3 contains an example of prices from the interbank market.
The interbank market trades spot and forward currencies. Spot transactions call for delivery and
payment within two days. Forward transactions call for delivery and payment at the time specified in the
forward contract. Actively traded maturities are 30, 90, and 180 days, but other maturities are available.
Forward contracts are contracts between two banks without the intervention of a clearinghouse. To reverse a
forward contract, another contract must be entered into that reverses the first contract. For example, suppose
a bank purchased British pounds 90 days forward. Ten days later, the bank wishes to offset that position. It
could do so by selling British pounds 80 days forward.
78
While flexible exchange rate systems have short run variability, they do not have the periodic large price changes and trade
dislocations that were typical of fixed exchange rate systems.
79
See Grabbe (1986) and Solnik (1988) for greater detail on currency markets and international bond and stock markets.
80
TABLE 9.2 Currency futures contract prices.

FUTURES
Lifetime
Open
Open High Low Settle Change High Low
Interest
JAPAN YEN (IMM)-12.5 million yen; $ per Yen (0.00)
Dec .7691 .7699 .7671 .7679 -0.0012 .7770 .6997 69,869
Mr92 .7666 .7694 .7659 .7665 -0.0011 .7737 .7000 3,572
June … … …. .7659 -0.0010 .7730 .7015 917
Sept … … … .7659 -0.0010 .7710 .7265 599
Dec … … …. .7662 -0.0009 .7700 .7512 1,290
DEUTSCHEMARK (IMM)-125,000 marks; $ per mark
Dec .6088 .6108 .6060 .6088 +.0007 .6770 .5365 72,328
Mr92 .6012 .6045 .5998 .6024 +.0007 .6065 .5353 6,380
June .5965 .5970 .5960 .5963 +.0007 .5985 .5322 715
Est vol 56,177; vol Tues 36,905; open lnt 79,626, -1,188.
CANADIAN DOLLAR (IMM)-100,000 dlrs.; $ per Can $
Dec .8771 .8817 .8763 .8812 +.0019 .8906 .8175 20,341
Mr92 .8720 .8769 .8713 .8767 +.0020 .8857 .8253 4,840
June .8675 .8725 .8675 .8725 +.0018 .8820 .8330 734
Sept .8630 .8685 .8630 .8685 +.0016 .8774 .8348 105
Est vol 13,890; vol Tues 7,534; open Int 26,078, -782.
BRITISH POUND (IMM)-62,SM pds.; $ w Pound
Dec 1.7640 1.7696 1.7560 1.7650 +.0024 1.7900 1.5670 27,784
Mr92 1.7430 1.7490 1.7370 1.7436 +.0024 1.7570 1.5560 2,964
Est vol 13,M; vol Tues 7,681; open Int 30,780, - 899.
SWISS FRANC (IMM)-125,000 francs; $ per franc
Dec .6877 .6905 .6849 .6869 +.0002 .8090 .6235 29,074
Mr92 .6829 .6952 .6797 .6819 +.0004 .6995 .6225 2,196
June .6766 .6795 .6750 .6771 +.0004 .6840 .6546 296
Est vol 73,401; vol Tues 16,SU; open Int 31,566, -1,353.
AUSTRALIAN DOLLAR (IMM) - 100,000 dlrs.; $ per A.$
Dec .7825 .7839 .7822 .7832 -.0006 .7960 .7380 11,204
Est vol 113; vol Tues 164; open lnt 1,221, -246.
U.S. DOLLAR INDEX (FINEX)-SM times USDX
Dec 89.04 89.40 88.78 88.99 -.11 98.96 88.47 5,019
Mr 92 90.07 90.30 89.90 90.08 -.10 98.90 89.60 1,045
The Index: High 88.86; Low 88.34; Close 88.54 -.08
TABLE 9.2 Cont. Currency futures contract prices.
OTHER FUTURES
Settlement prices of selected contracts. Volume and open Interest of all contract months.
British Pound (MCE) IZSM pounds; $ per pound

Dec 1.7650 +.0024; Est. vol. 120; Open Int. 422
Japanese Yen (MCE) 6.25 million yen; $ per yen (.00)
Dec .7679 -.0012; Est. vol. 240; Open Int. 353
Swiss Franc (MCE) 6ZSM francs; $ per franc
Dec .6869 +.0002; Est. vol. 1,OX; Open Int. 253
Deutschemark (MCE) 6ZSM marks; $ per mark
Dec .6088 +.0007; Est. vol. 360; Open Int. 837
BP/DM Cross Rate (IMM) US $50,000 times BP/DM
DM/JY Cross Rate (IMM) US $125,000 times DM/JY
Dec .7928 +.0022; Est. vol. 6; Open lnt. 583
FINEX-Financial Instrument Exchange, a division of the New York Cotton Exchange. IMM-International
Monetary Market at the Chicago Mercantile Exchange. MCE MidAmerica Commodity Exchange.
Forward contracts are subject to two kinds of default risk –credit risk and country risk. Credit risk is the risk
that the contra bank has poor credit. This risk is low because only the most creditworthy international banks
participate in the interbank market. In the futures market, clearinghouse margin requirements serve to limit
credit risk. Country risk is the risk that a country will impose restrictions on the transfer of currencies and
thereby make it impossible for two banks to carry out the terms of their forward contract. Such restrictions
were quite frequent under fixed exchange rate systems, but are less frequent today.
9.2 CURRENCY QUOTATIONS AND TRIANGULAR ARBITRAGE

We will follow the convention of quoting the price of a foreign currency just as we quote the price of any
other commodity, in dollars per unit. Thus, the price of the German mark might be 0.57 dollars per mark.
Sometimes foreign currency prices are quoted as the number of units of foreign currency per dollar. In the
case of the German mark, that would be 1/0.57 Deutsche marks per dollar.81 Table 9.4 contains spot currency
prices from both perspectives for some key currencies.
The dollar cost of a German mark should be the same if one first purchased francs and then used the francs to
buy marks or if one purchased marks directly.
In Table 9.4, the dollar price of a German mark is 0.612000 dollars. The French Franc price of a
German mark is 3.4171 francs. The number of dollars needed to buy 3.4171 francs (which buy one mark) is
(0.1791)(3.4171) = 0.612003. The two methods of buying one mark are virtually identical and differ only in
the sixth decimal. If the two methods gave different answers, an opportunity for triangular arbitrage would
exist. For example, suppose the dollar price of a French franc is only 0.1720 dollars and the franc price of the
mark is as shown in the table. Then a mark could be purchased for (0.1720)(3.4171) = 0.5877 dollars. An
arbitrageur could borrow marks and sell them for 0.6120 dollars per mark and simultaneously take 0.5877
dollars to buy back the marks needed to repay the borrowings. The profit from this triangular arbitrage would
81
In principle, the price of wheat could be quoted both as the dollar price of wheat, say four dollars per bushel, or the wheat price
of dollars, which would be 1/4 bushels per dollar.
be 0.6120 - 0.5877 = 0.0243 dollars per mark. To avoid triangular arbitrage opportunities, the following
condition must be met for all trios of currencies
S1, 2  S1,3 S3, 2 , (9.1)
where Si , j is the number of units of the i - th currency required to purchase one unit of the j - th currency.
TABLE 9.3 Spot and forward currency prices from the Interbank market.
EXCHANGE RATES: Wednesday, November 13, 1991. The New York foreign exchange selling
rates below apply to trading among banks In amounts of $1 million and more, as quoted at 3 p.m.
Eastern time by Bankers Trust Co. and other sources. Retail transactions provide fewer units of foreign
currency Per-dollar.
U.S. $ equiv. Currency per U.S. $
Country Wed. Tues. Wed. Tues.
Argentina (Austral) 0.0001008 0.0001008 9918.67 9918.67
Australia (Dollar) 0.7860 0.7870 1.2723 1.2706
Austria (Schilling) 0.08681 0.08681 11.52 11.52
BMrsin (Dlnar) 2.6539 2.6539 0.3768 0.3768
Belgium (Franc) 0.02966 0.02966 33.72 33.72
Brazil (Cruzeiro) 0.00144 0.00146 694.71 695.60
Britain (Pound) 1.7730 1.7725 0.5640 0.5642
30-Day Forward 1.7648 1.7640 0.5666 0.5669
90-Day Forward 1.7504 1.74% 0.5713 0.5716
180-Day Forward 1.7299 1.7291 0.5781 0.5783
Canada (Dollar) 0.8842 0.8838 1.1310 1.1315
30-Day Forward 0.8815 0.8814 1.1344 1.1346
90-Day Forward 0.8784 0.8779 1.1384 1.1391
180-Day Forward 0.8737 0.8733 1.1445 1.1451
Chile (Peso) 0.002844 0.002780 351.56 359.65
China (Renmlnbi) 0.185642 0.185642 5.3867 5.3867
Colombia (Peso) 0.001753 0.001753 570.38 570.38
Denmark (Krone) 0.1573 0.1573 6.3570 6.3555
Ecuador (Sucre) Floating rate 0.000966 0.000966 1035.00 10335.00
Finland (Markka) 0.24984 0.24941 4.0025 4.0095
France- (Franc) 0.17881 0.17879 5.5925 5.5930
30-Day Forward 0.17813 0.17808 5.6140 5.6156
90-Day Forward 0.17690 0.17695 5.6529 5.6545
180-Day Forward 0.17510 0.17504 5.7110 5.7130
Germany (Mark) 0.6112 0.6111 1.6362 1.6365
30-Day Forward 0.6090 0.6088 1.6421 l.6426
90-Day Forward 0.6045 0.6044 1.6543 1.6544
180-Day Forward 0.5982 0.5902 1.6717 1.6718
Greece (Drachma) 0.005405 0.005405 185.00 185.00
Hong Kong (Dollar) 0.12884 0.12884 7.7615 7.7615
India (Rupee) 0.03890 0.03890 25.77 25.77
Indonesia (Rupiah) 0.0005056 0.0005056 1978.00 1978.00
Ireland (Punt) 1.6330 1.6318 0.6124 0.6128
TABLE 9.3 Cont. Spot and forward currency prices from the Interbank market.
Israel (Shekel) 0.4308 0.4321 2.3215 2.3142
Italy (Lira) 0.0006121 0.0006117 1231.41 1232.01
Japan (Yen) 0.007686 0.007707 130.10 129.75
30-Day Forward 0.007678 0.007696 130.24 129.90
90-Day Forward 0.007666 0.007686 130.45 130.10
180-Day Forward 0.007656 0.007677 130.62 130.26
Jordan (Dinar) 1.4500 1.4500 0.6897 0.6897
Kuwait (Dinar) 3.4965 3.4965 0.2860 0.2860
Lebanon (Pound) 0.001134 0.001134 881.50 881.50
Malaysia (Ringgit) 0.3650 0.3647 2.7400 2.7400
Matta (Lira) 3.1250 3.1250 0.3200 0.3200
Mexico (Peso) Floating rate 0.0003254 0.0003254 3073.01 3073.01
Netherland (Guilder) 0.5423 0.5423 1.9445 1.9445
New Zeland (Dollar) 0.5610 0.5620 1.7825 1.7794
Norway (Krone) 0.1558 0.1558 6.4175 6.4195
Pakistan (Rupee) 0.0405 0.0405 24.72 24.72
Peru (New Sol) 1.0152 1.0051 0.99 0.99
Philippines (Peso) 0.03839 0.03839 26.05 26.05
Portugal (Escudo) 0.007067 0.007063 141.50 141.59
Saudi Arabia (Rlyal) 0.26663 0.26663 3.7505 3.7505
Singapore (Dollar) 0.5958 0.5959 1.6795 1.6780
South Africa (Rand) Commercial rate 0.3568 0.3574 2.8023 2.7961
Financial rate 0.3240 0.3240 3.0790 3.0790
South Korea (Won) 0.0013310 0.0013310 751.30 751.30
Spain (Peseta) 0.009723 0.009699 102.85 103.10
Sweden (Krona) 0.1673 0.1672 5.9775 5.9615
Switzerland (Franc) 0.6888 0.6892 1.4517 1.4510
30-Day Forward 0.6872 0.6675 1.4552 1.454
90-Day Forward 0.6835 0.6839 1.4631 1.4621
120-day Forward 0.6788 0.6792 1.4732 1.4724
Taiwan (Dollar) 0.038650 0.037906 25.74 26.38
Thailand (Baht) 0.03926 0.03926 25.47 25.47
Turkey (Lira) 0.0002044 0.0002020 4892.01 4950.00
United Arab (Dirham) 0.2723 0.2723 3.6725 3.6725
Uruguay (New Peso) Financial 0.000425 0.000425 2352.94 2352.94
Venezuela (Bolivar) Floating rate 0.01695 0.01661 59.00 60.20
SDR 1.38023 1.38189 0.72452 0.72365
ECU 1.24952 1.25088
Special Drawing Rights (SDR) are based on exchange rates for the U.S., German, British, French and
Japanese currencies. Source: International Monetary Fund.
European Currency Unit (ECU) Is based on a basket of community currencies. Source: European
Community Commission.
TABLE 9.4 Key currency cross rates. Late New York Trading Nov. 13, 1991
Dollar Pound Sfranc Guilder Yen Lira D-Mark Ffranc CdnDol
Canada 1.1313 2.0098 0.78021 0.61434 0.00870 0.00092 0.69235 0.20261 …
France 5.5835 9.919 3.8507 3.0320 0.04292 0.00454 3.4171 … 4.9355
Germany 1.6340 2.9028 1.1269 0.88732 0.01256 0.00133 … 0.29265 1.4444
Italy 1229.3 2183.8 847.76 667.53 9.449 … 752.29 220.16 1086.6
Japan 130.10 231.12 89.724 70.649 … 0.10584 79.621 23.301 115.00
Netherlands 1.8415 3.2714 1.2700 … .01415 0.00150 1.1270 0.32981 1.6278
Switzerland 1.4500 2.5759 … 0.78740 0.01115 0.00118 0.88739 0.25969 1.2817
U.K 0.56290 … 0.39821 0.30568 0.00433 0.00046 0.34449 0.10082 0.49757
U.S … 1.7765 0.68966 0.54304 0.00769 0.00081 0.61200 0.17910 0.88394
Source: Telerate
9.3 STRUCTURE OF CURRENCY FUTURES PRICES

Unlike agricultural commodities, which are sometimes in short supply, currencies are always in large supply.
As a result, currency futures obey the cost-of-carry equilibrium developed in Chapter 3: Ft  St eb (T t ) . In
currencies, the cost of carry is the difference between the interest cost of the dollars invested in a currency
and the interest earned on the foreign currency. The cost-of-carry relation can therefore be written as
( rd r f )(T t )
Ft  St e , (9.2)
where rd (rf ) is the continuously compounded, riskless rate of interest in the domestic (foreign) currency and
T is the maturity of the futures contract.
In international finance, the relation (9.2) is called the interest rate parity (IRP) relation. It is
instructive to derive IRP in a somewhat different way. Consider an investor who has one U.S. dollar to
invest. If the dollar is invested domestically at the riskless rate, the value at maturity is e rd (T t ) .
On the other hand, the dollar may also be used to purchase foreign currency that is then invested at
the foreign riskless rate of interest. If the proceeds at maturity of the foreign investment are sold in the
futures market, a dollar return can be guaranteed at time t . At maturity, the dollar cash proceeds of a hedged
r (T t )
foreign investment are (1 / St )e f Ft
For example, consider an investment in Germany. Suppose St  0.57 , rf  0.05 per year,
T  t  0.25 years, and Ft  0.58 . The hedged dollar proceeds on the investment in Germany are 1.03034
dollars.
In equilibrium, the two ways of investing the dollar-directly in the U.S. or indirectly in a foreign
country-must have the same value at maturity, assuming that both investments offer a riskless return and that
there is no default risk on the futures or forward contract. In other words, the absence of costless arbitrage
opportunities in the marketplace ensures that
r (T t )
e rd (T t )  (1 / St )e f Ft .
This expression is easily manipulated to give (9.2). Dividing by S, and subtracting one from each side,
equation (9.2) can also be written as
Ft  St ( r r )(T t )
e d f 1 . (9.3)
St
The left-hand side of this expression, which in other futures markets we call the percentage basis, is called
the forward premium or the swap rate in the currency markets. The term swap rate comes from the fact that
investors frequently buy a foreign currency and agree to swap it back for dollars. The swap rate specifies the
gain or loss on such a transaction. The right-hand side is the interest differential between the two countries
over the time period T  t .
The IRP relation can also be derived for the case in which interest is earned discretely over the life of
the futures contract. Suppose rd* and rf* are the U.S. and foreign riskless rates of interest for the time over
which funds are invested, T  t .
Then an investment in the U.S. is worth $( 1  rd* ) at maturity, and a hedged investment in the foreign country
is worth $(1 / St )(1  rf* ) Ft at maturity. To eliminate riskless arbitrage opportunities, these outcomes must be
equal, which implies
F  S rd  rf
* *
 . (9.4)
S 1  rf*
This equilibrium relation is plotted in Figure 9.1 as the 45-degree line. Arbitrage opportunities arise if an
observation is not on the line.
For example, consider point A. At A, exchange rates are S  0.50 and F  0.51 , yielding a forward
premium ( F  S ) / S  0.02 ; and three-month interest rates are rd*  0.03 and rf*  0.02 , yielding an interest
differential of (rd*  rf* ) /(1  rf* )  0.01 . Although the interest differential is in favor of the U.S., it is profitable
to borrow in the U.S. and invest in the foreign country because the profit on the foreign exchange transaction
exceeds the loss on the interest differential. An arbitrageur could borrow one million dollars in the U.S.,
thereby incurring an obligation to repay (1  rd* )  1.03 million dollars at maturity; and she would invest in
the hedged foreign investment, which pays $(1 / S )(1  rf* ) F  1.0404 million dollars at maturity. The
arbitrage profit is 1.0404 million minus 1.03 million or 10,400 dollars. The activity of arbitrageurs results in
a capital outflow from the U.S. that raises U.S. interest rates, raises the spot exchange rate, lowers foreign
interest rates, and lowers the futures price, until equilibrium is restored. In fact, all points above the line in
the figure imply a capital outflow from the U.S. Points below the line, such as point B, imply a capital
inflow into the United States. At point B, the interest differential is in favor of the U.S., but the forward
premium is zero. Arbitrage profits could be earned by borrowing in the foreign country at rf* , selling the
foreign currency at S , investing in the U.S. at rd* , and entering into a futures contract to purchase the foreign
currency at F to repay the loan. The dollar value of the face amount of the foreign loan is $(1 / S )(1  rf* ) F ,
and the value of the dollar investment is (1  rd* ) . Assuming that S  0.50; rd*  0.03; rf*  0.02 and
F  0.50 , the profit is 0.01 for every dollar borrowed and invested in the United States.
In today's highly efficient international financial markets, arbitrage is instantaneous, and deviations
from IRP are rarely observed except for very small deviations caused by transaction costs. Apparent
deviations may be observed for certain interest rates. Each country has only one spot rate and one futures rate
for any maturity, but each country has many short-term interest rates that reflect different degrees of risk.
Thus, it is possible to find two interest rates that cause IRP to be violated. Differences in risk or differences
in transaction costs, however, usually explain deviations from IRP.
F S
S
FIGURE 9.1 Interest Rate Parity
A
0.02
0.01 rd*  rf*

1  rf*
9.4 FACTORS AFFECTING EXCHANGE RATES

Many factors affect exchange rates. While it is beyond the scope of this book to consider all of these factors
in detail, understanding some of the key factors that analysts consider is useful from a practical standpoint.82
Purchasing Power Parity

An important factor determining a country's currency value is the rate of inflation in the country. The higher
the rate of inflation, the greater the decline in the purchasing power of one unit of the currency, and the lower
the price of the currency. The theory of Purchasing Power Parity (PPP) is a macroeconomic version of the
Law of One Price (LOP) discussed in Chapter 6. The LOP takes the exchange rate as given and says that the
prices of a commodity in two different countries must be the same after adjustment for the exchange rate:
Gd  SG f
where Gd and G f are the U.S. and foreign prices of the commodity and S is the spot exchange rate (the
dollar price of the foreign currency). If commodity arbitrage takes place for every commodity, the LOP must
hold for bundles of the same commodities in the two countries. Changes in the price of the bundle in one
country due to inflation without a corresponding change in the price of the bundle in the other country must
imply a change in the exchange rate. In other words, under PPP, the exchange rate is the dependent variable,
and the prices of commodities in the two countries are independent variables.
The absolute version of PPP is written as
Pd ,t
St  (9.5)
Pf ,t
The variable Pd ,t , is the dollar price of a bundle of commodities in the U.S. and Pf ,t is the foreign currency
price of the same bundle. For example, at t  1991 , Pd ,t  500 dollars and Pf ,t  2,000 units of foreign
currency. That implies an exchange rate of 0.25 according to PPP. The relative version of PPP looks at the
change in the exchange rate as a function of the relative changes in the prices of the bundle of commodities
82
For detailed discussions of factors determining exchange rates, see Lessard (1985).
in each country. We can derive the relative version of PPP by writing equation (9.5) for a base period, time
0. Dividing the base period equation into (9.5) yields
St P /P P /P St I d ,t
 d ,t f ,t  d ,t d ,0 or  (9.6)
S0 Pd ,0 / Pf ,0 Pf ,t / Pf ,0 S0 I f ,t
where I d ,t and I f ,t , are the U.S. and foreign country price indexes with a time 0 base year. Equation (9.6) is
the relative version of PPP. Suppose the base year is 1967 and that in that year the bundle costs Pd ,0  200 in
the U.S. and Pf ,0  400 in the foreign country. That implies an exchange rate of .50 dollars in 1967. Using
the 1991 bundle prices assumed above, the price indexes are I d ,0  2.50 , in the U.S. and I f ,0  5.00 in the
foreign country. The ratio of these indexes implies a ratio of exchange rates in 1991 versus 1967 of 0.5,
which, according to PPP, means the exchange rate is 0.25 in 1991.
In practice, PPP does not seem to hold as well as in the example we have just presented. A difficulty arises
from measurement problems. Price indexes in different countries do not include the same commodities or
have the same weights when the same commodities are included. Difficulties also arise in measuring prices
of commodities accurately and at the same time in the different countries. Another difficulty arises from the
fact that transportation costs are high for many commodities, so that the law of one price cannot be
established. Indeed, transport costs are prohibitive for certain commodities. Moreover, many items such as
services are simply not traded. Services (say, haircuts) can only be traded by moving labor, but restrictions
on international migration prevent arbitrage of services. Neither of these difficulties is severe if the source of
a change in currency values is inflation that affects the prices of all commodities-traded and non-traded-in
the same way. In that case, any index will be representative. PPP does not hold in practice, however, because
the prices of different commodities move in different ways in different countries. It is possible, for example,
for a country's traded goods not to increase in price (because of cost cutting measures in the traded goods
sector of the economy), while the prices of non-traded goods increase considerably more. The exchange rate
of that country will not depreciate as much as would be predicted by the aggregate inflation in the country
because it has remained competitive in those goods that are traded internationally.
Despite these difficulties, PPP remains an important determinant of exchange rate changes over longer
periods of time, particularly when comparing countries with significantly different inflation rates. Table 9.5
presents some data for the exchange rates and inflation of seven industrial nations for the period 19.67 to
1983 in which substantial differences in inflation arose. The second column gives the ratio of the dollar price
of the foreign currency in 1983 to the dollar price in 1967.
During this period, the dollar appreciated relative to every country except West Germany and Japan.
For example, the price of the British pound fell by 44.9 percent, whereas the price of the German mark rose
by 56.4 percent. The third column shows that much of the change in exchange rates can be explained by
differences in inflation rates. Typically, when U.S. inflation is less than foreign inflation, the foreign
currency falls, and when U.S. inflation is greater than foreign inflation, the price of the foreign currency
rises. Japan is the only exception. According to the consumer price indexes for the U.S. and Japan, inflation
was about equal in the two countries, yet the Japanese yen appreciated by 52.6 percent. Part of the
explanation for this discrepancy lies in the fact that the consumer price index is not representative of the
price of Japanese traded goods. Consumer goods within Japan have risen in price, but goods traded
internationally have not risen in price to the extent implied by the Japanese consumer price index.
The Monetarist Approach to Exchange Rate Determination

The monetarist approach is an extension of purchasing power parity. Under the monetarist approach, the
price level in each country is determined by monetary factors. The exchange rate then depends on the factors
that determine the price level in each country.
TABLE 9.5 Exchange rates and inflation, 1967-1983.

Country S83 / S67 Id / I f Real Exchange Rate
Canada 0.875 0.929 0.94
France 0.649 0.732 0.89
West Germany 1.564 1.468 1.07
Italy 0.412 0.472 0.87
Japan 1.526 0.963 1.58
United Kingdom 0.551 0.554 0.99
Source: International Monetary Fund, International Financial Statistics, monthly.
The second column is the ratio of the dollar price of the foreign currency in 1983 to the dollar price of the
foreign currency in 1967. The third column is the ratio of the consumer price index in the United States to
the consumer price index in the foreign country. The base year for both indexes is 1967. The fourth column
is the ratio of the second column to the third column. A number greater than (less than) one indicates that the
currency had a real appreciation (depreciation) relative to the dollar.
The quantity theory of money states that
mt v  I t yt , (9.7)
where mt , is the money supply at time t divided by the money supply in the base period, v is the velocity of
money, I t is the price index, and yt is real income at time t divided by real income in the base period. The
price index can be written as
I t  (mt v) / yt , (9.8)
which shows that, under the assumption that the velocity of money is stable, the quantity theory states that
inflation occurs if the money supply expands faster than real income.
Under the quantity theory, the price level may be written for both the domestic and foreign countries:
I d ,t  (md ,t v) / yd ,t (9.8a)
I f ,t  (m f ,t v) / y f ,t (9.8b)
The base period for the variables is assumed to be the same in the two countries. Substituting (9.8a) and
(9.8b) in (9.6) yields
St I d ,t md ,t vd y f ,t
  . (9.9)
S0 I f ,t m f ,t v f yd ,t
Equation (9.9) says that the change in the exchange rate depends on relative money supply growths and
relative real income growths in the two countries, assuming constant velocities. Suppose that velocities in
two countries are one, that real income grew 100 percent in the foreign country ( y f ,t  2 ) and 50 percent in
the domestic economy ( yd ,t  1.5 ), that the money supply grew 300 percent in the foreign country ( m f ,t  4 )
and 100 percent in the domestic economy ( md ,t  2 ). That would imply St / S0  0.67 , a 33 percent decline in
the price of the foreign currency. The monetary approach can become a good deal more complicated as other
factors that affect the impact of money supply changes on the economy are considered.
Like PPP, the monetarist approach is less successful in explaining short-run changes in exchange
rates than in explaining long-run changes. Using the monetarist approach to predict exchange rates requires
predicting money supply growths in the two countries and other variables that affect inflation, something that
is not an easy task over short intervals.
Balance of Payments Approach to Exchange Rate Determination

Another set of factors considered by analysts is the balance of payments, particularly the current account.
The current account balance of a country is the exports of goods and services minus imports. A country has a
current account surplus when its exports exceed its imports, and it has a current account deficit when imports
exceed exports. The U.S., for example, had a large current account deficit in the 1980s. Analysts who take
the balance of payments approach argue that a deficit increases the demand for foreign exchange and raises
the price of foreign currencies, whereas a surplus lowers the price of foreign currencies. It has been difficult
to show, however, that exchange rates are, in fact, related to the current account deficit in this way. For
example, the U.S. dollar appreciated during some periods in the 1980s when the current account deficit was
large.
Part of the problem is that the current account is endogenous, that is, it depends on fundamental forces that
also affect the exchange rate. For example, the current account depends on monetary and fiscal policy that
also have a direct effect on the exchange rate. A country might import more because its real income has
grown. A growth in real income could be consistent with a decline in the price of foreign currencies (a
strengthening of the domestic exchange rate) as shown above for the monetarist approach.
Analysts often emphasize the current account, but the balance of payments also includes a capital
account. Overall, the balance of payments must balance. A current account deficit must be matched by a
capital account surplus (assuming government reserves and borrowings do not change), and a current
account surplus must be matched by a capital account deficit. If a country, imports more than it exports, the
cost of the net imports must be financed by borrowing from abroad. If a country exports more than it
imports, the foreign currency earnings must be invested abroad (or used to reduce foreign debt). Some
analysts argue that capital flows are exogenous and that the current account is endogenous. Under this
argument, the U.S. trade deficit results from a large capital inflow to the U.S. in response to higher U.S.
interest rates and other factors. The capital account surplus, in turn, made resources available to the U.S.,
some of which were spent on imports, thereby generating the current account deficit. Under this scenario, the
price of foreign currencies declined even though the U.S. ran a current account deficit.
Because of capital flows and other macroeconomic factors, the balance of payments approach, with
its focus on the current account, has not proved adequate to explain the behavior of exchange rates.
The Asset Market Approach to Exchange Rate Determination

The asset approach to exchange rate determination argues that investors throughout the world allocate
investment according to anticipated real returns and anticipated exchange rate changes. If Japan has a low
expected real return relative to the U.S., funds will flow to the U.S., and the U.S. dollar will appreciate. The
asset approach thus focuses on capital flows and the factors that determine capital flows. In addition to
expected real returns, capital flows respond to risk.
In summary, a variety of complicated forces affect exchange rates. PPP is important, but short-run deviations
from PPP are prevalent. These short-run deviations depend on monetary and fiscal policy, exogenous
changes in the demand for imports and exports, and exogenous changes in capital flows.
9.5 RETURNS TO SPECULATORS IN CURRENCIES

As in other financial markets, it is difficult for currency speculators to make abnormal returns because
competition among traders causes the currency markets to be efficient. One potential source of inefficiency
is the intervention by central banks in the currency markets to "stabilize" their exchange rates. Such
stabilization sometimes slows adjustment of exchange rates and makes it possible to predict future exchange
rates. Most of the evidence, however, indicates that, like stock prices, exchange rate changes are hard to
predict.
Spot Speculation and Futures Speculation

The fact that exchange rate changes are hard to predict does not keep people from trying. Speculation on
exchange rate changes can be done in either the spot market or the futures market. In the futures market, a
speculator would buy currency futures if she expects the spot rate at maturity to exceed the current futures
price, and she would sell currency futures if she expected the spot rate at maturity to be less than the futures
~
price. Her expected profit is E ( ST )  Ft (T ) . In equilibrium, competition among speculators would eliminate
profits and cause the futures price to equal the expected spot price:
~
Ft (T )  E ( ST ) . (9.10)
We noted in Chapter 4 that the trading pressures of hedgers might cause speculators to demand a risk
premium that brings about a divergence between the futures price and the expected spot price. In currency
markets, the risk premium could easily be positive or negative since hedgers could be buying foreign
currency or selling foreign currency. In the absence of a reason to assume the risk premium is a particular
sign, we shall assume that condition (9.10) is met and that no risk premium exists.
A speculator who expects a foreign currency to appreciate could also borrow dollars and buy the foreign
currency. The speculation is profitable if the currency appreciates more than the cost of holding the currency.
The mechanics of spot speculation are as follows: borrow a dollar, buy 1 / St , units of foreign currency, invest
~
the foreign currency at the foreign interest rate, rf* , sell the foreign currency at the future spot rate, ST , and
pay back the dollar plus interest of rd* . The expected profit is
~
(1 / St )(1  rf* ) E ( ST )  (1  rd* )
In the absence of a risk premium, equilibrium requires a zero-expected profit, which implies
(1  rd* ) ~
St  E ( ST ) . (9.11)
(1  rf )
*
It is worth noting that the left-hand sides of (9. 1 0) and (9.1 1), taken together, yield IRP. Under IRP, spot
speculation and futures speculation are equivalent.
The International Fisher Effect
Relation (9.1 1) is often called the international Fisher effect (IFE). The IFE is usually written as
~
E ( ST ) (1  rd* )

St (1  rf* )
The IFE says that the expected change in the exchange rate equals the interest rate differential between the
countries. For example, if the one-year interest rate in the U.S. is 8 percent and the one-year interest rate in
Brazil is 80 percent, the IFE says the price of the Brazilian cruzeiro is expected to be 60 percent of its
current value.
The IFE takes its name from Irving Fisher, who argued that the domestic interest rate is
approximately equal to the real rate of interest plus the expected rate of inflation. The IFE assumes that
international differences in interest rates reflect differences in expected inflation rates. According to Fisher,
(1  rd* )  (1  ad* )[1  E (~d* )] ; (1  rf* )  (1  a *f )[1  E (~ *f )] ,
where ad* and a*f are the domestic and foreign real rates of interest, and  d* and  *f ; are the U.S. and foreign
inflation rates. If ad*  a*f , the IFE becomes
~
E ( ST ) 1  E (~d* )
 . (9.12)
St 1  E (~ *f )
In this form, the IFE is very much like the relative PPP equation (9.6), except that PPP is an ex post relation
between the realized exchange rate change and realized inflation while the IFE is an ex ante relation
between the expected exchange rate change and expected inflation.
Forecasting Exchange Rates

In an efficient market without inside information, it is impossible to make better forecasts of exchange rates
than the forecasts available from the financial markets. As we have just seen, financial markets provide two
forecasts: the futures price and the difference in interest rates. In the absence of forecasting skill or inside
information and an understanding of the actions of world monetary authorities, these readily available
forecasts are as useful as more sophisticated forecasting. Careful analysis of fundamental factors such as
inflation, monetary policy, balance of payments, and asset flows may result in superior forecasts, but
competition among forecasters tends to eliminate abnormal returns.83 Market-based forecasts are not
necessarily accurate since unexpected events have a way of altering outcomes from the outcome that was
expected, but they are unbiased and readily available.
9.6 HEDGING CURRENCY RISK

Given the uncertainties about the future value of a currency, companies engaged in international operations
frequently want to hedge against the risks of exchange rate changes. Currency risk can arise because of a
particular import or export transaction or because a company's balance sheet contains assets and liabilities
denominated in a foreign currency.84
83
Evidence on forecasting ability is provided by Levich, “Evaluating the Performance of the Forecasters,” in Lessard (1985).
84
For greater detail on hedging and on other financial issues in managing international operations see Shapiro (1989) and Eiteman
and Stonehill (1986).
Transactions Risk
Transactions risk refers to the foreign currency risk of a particular future transaction denominated in a
foreign currency. For example, a U.S. company sells a product to a German importer who agrees to pay for
the product in three months, when shipment is received. Payment is specified in German marks. The dollar
value of that account receivable for the U.S. Company is uncertain. This transaction can be hedged by selling
forward the German marks to be received in three months. Such a forward sale is usually done through a
bank because the quantity and other terms of the forward contract can be tailored to the specific transaction
incurring currency risk. Futures contracts could also be used, but the size and maturity of a futures contract
may not match the hedging need exactly.
An example of transactions risk might be a U.S. airplane manufacturer who has committed to deliver
a jet to Lufthansa in one year for 40,000,000 marks, with payment to be made in one year. The spot price of
the mark is assumed to be St  $0.6000 and the one-year forward price, Ft  $0.61132075 . If left unhedged,
the dollar value of the contract is subject to fluctuations in the value of the German mark. At the current
forward price, the contract is worth $24,452,830, but if the spot price should fall two percent below its
current value to $0.5880, the contract would be worth $23,520,000, a loss of nearly one million dollars. The
manufacturer can hedge foreign exchange risk by entering a forward contract today to sell 40,000,000 marks
a year from now at the forward price. At maturity, when payment is made, the German mark proceeds from
the export of the jet are delivered to the bank in return for dollars. The manufacturer has a problem if there
are delays in delivery and payment. In that case, the forward contract has to be rolled over at some cost.
An alternative to the forward market hedge is a money market hedge. Under interest rate parity, a
money market hedge is equivalent to a forward market hedge. U.S. and German interest rates consistent with
the spot, and forward rates in the example are rd*  0.08 and rf*  0.06 . Under a money market hedge, the
U.S. jet manufacturer borrows German marks against the proceeds of the sale of the jet, which amounts to
40,000,000/(1.06) = 37,735,849 marks, and converts the marks to dollars at the current spot rate to get
(0.60)(37,735,849) = 22,641,509. If this amount is invested at 8 percent, the proceeds at maturity are
(22,641,509)(1.08) = 24,452,830, the same as under the forward contract. In practice, the two approaches
may not be the same since the short-term interest rates that cause interest rate parity to hold may not be the
rates at which the manufacturer can borrow and lend. If the manufacturer is in need of funds, the money
market hedge is preferable to the forward market hedge.
Balance Sheet Risk

In many cases, a company faces foreign exchange risk that is less specifically tied to a particular transaction.
For example, a company exporting to Germany may build up accounts receivable with many different
maturity dates, or it may hold inventory in Germany which is exposed to currency risk. Indeed, a company's
balance sheet may have both assets and liabilities denominated in various foreign currencies. Appropriate
hedging procedures for this more general case are quite complex. They depend on whether the foreign
currency obligation is contractual or not. A contractual obligation denominated in a foreign currency is
subject to foreign exchange risk because the contract price is set and cannot be changed, as in the case of the
U.S. exporter of jets. If the value of the currency falls, the foreign currency price cannot be adjusted. On the
other hand, a non-contractual business operation is less subject to exchange risk because changes in
exchange rates can, perhaps, be offset by price changes in the foreign currency. A U.S. toothpaste
manufacturer with an Italian subsidiary, for example, may find it possible to offset declines in the value of
the Italian lira with increases in the price of toothpaste sold in Italy.
A simple example of a balance sheet hedge of a contractual asset is that of a British subsidiary with a
cash balance of 5,000 pounds. The U.S. parent, worried about a possible decline in the British pound, decides
to hedge that position by selling futures in the amount of 5,000 pounds. If the value of the pound falls, the
decline in the dollar value of the cash is offset by the profit on the short futures position. This hedge is
equivalent to a hedge of a particular transaction, in the sense that both are contractual. It is different in that
the cash balance has no maturity and is changed daily as a result of business operations.
An example of a balance sheet hedge of a non-contractual item is hedging a finished goods inventory
position valued at 5,000 pounds at the current price of the product in Britain. This hedge is more complicated
because a decline in the value of the British pound that would reduce the dollar value of the inventory might
be offset by an increase in the price of the finished good in the United Kingdom. Indeed, under PPP, that is
exactly what would be expected. Suppose declines in the dollar price of the British pound can be partially
offset by increasing product prices by 50 percent of the amount they would increase under PPP. If repricing
offsets half the decline in the price of the pound, the optimal hedge is to sell short about 2,500 pounds. In
effect, the ability to reprice inventory in the U.K. provides a partial natural hedge (for 50 percent of the
decline). The futures market hedge hedges the other half of the exchange rate decline. Table 9.6 details this
example.
9.7 SUMMARY
This chapter begins by describing currency markets and currency quotations. The cost-of-carry model for
currency futures is then derived and shown to be the same as the well-known interest rate parity condition.
Theories of the determinants of exchange rates-the purchasing power parity approach, the monetarist
approach, the balance of payments approach, and the asset approach-are described.
We distinguish futures speculation and spot speculation. Competition among speculators in the futures
market implies that the futures price is a good estimate of the expected spot price. Similarly, competition in
spot speculation leads to the international Fisher effect-that the expected change in the exchange rate equals
the interest differential between two countries.
Currency futures may be used to hedge against fluctuations in exchange rates. The chapter concludes
with descriptions of hedging transaction risk and balance sheet risk.
TABLE 9.6 Hedging foreign exchange risk of 1000 units of finished goods inventory when 50 percent
of the exchange rate change can be offset by raising the sale price.
Inventory Futures Market
Futures
Value Value Futures Value
Date 85 Exch. Rate Position
(pounds) (dollars) Price (dollars)
(pounds)
Sept. 1 5,000 1.60 8,000 -2,500 1.60 -4,000
Dec. 1 5,833 1.20 7,000 -2,500 1.20 -3,000
Gain -1,000 1,000
85
The initial price of inventory is assumed to be 5 pounds per unit. The old exchange rate is 1/3 greater than the new exchange
rate. The table assumes the selling price can be raised by half the amount that would offset the change in the exchange rate, or by
0.5(1/3)(5).
10. ESTRUCTURA DE PRECIOS DE OPCIONES: RELACIONES DE ARBITRAJE
10.1 NOMENCLATURA, NOTACIÓN Y DEFINICIONES DE LAS OPCIONES
10.2 DIAGRAMAS DE GANANCIA Y NOTACIÓN VECTORIAL
10.3 DISCUSIÓN GENERAL DE LÍMITES DE PRECIO INFERIORES PARA LAS

OPCIONES
 Opciones Europeas
 Opciones Americanas
10.4 EVIDENCIA DE ARBITRAJE DE LOS LÍMITES DE PRECIO INFERIORES

 Put Europea
 Call Europea
10.5 EJERCICIO TEMPRANO O ANTES DEL VENCIMIENTO

 Call Americana
 Put Americana
 Opciones Americanas de Futuros
10.6 DISCUSIÓN GENERAL SOBRE LA PARIDAD DE PUT Y CALL
10.7 PRUEBAS DE LAS RELACIONES DE PARIDAD PUT Y CALL

10.8 OPCIONES DE FUTUROS Y OPCIONES DE COMMODITIES

10.9 RESUMEN
10. ESTRUCTURA DE PRECIOS DE OPCIONES: RELACIONES DE ARBITRAJE

Los contratos de opciones confieren el derecho, pero no la obligación, de consumar una transacción en una
mercancía en particular, a un precio particular, en una fecha futura. A igual que los futuros, los contratos de
opciones vinculan el presente y el futuro. Sin embargo, a diferencia de los futuros, el tenedor de una opción
puede elegir si quiere o no ejercer su derecho de comprar o vender el subyacente.
Este capítulo es la contrapartida del Capítulo 3 sobre precios de las opciones, en el cual se examinó la
estructura de precios de futuros. Nosotros derivamos y explicamos los vínculos de arbitraje entre opciones de
mercancías, opciones de futuros, futuros de mercancías y mercancías; tal como se describe en las líneas
horizontales y verticales de la Figura 10.1. En este capítulo, nosotros explicamos cómo se relacionan entre sí
los precios de estos instrumentos, y establecemos límites para los precios de las opciones con relación a los
precios de la mercancía subyacente. No especificamos los factores que determinan el nivel de precios de las
opciones. Los determinantes del nivel de precios de las opciones se discuten en el Capítulo 11.
Como un recordatorio, las premisas que fundamentan los argumentos de arbitraje son (a) las personas
son racionales; (b) los mercados están libres de fricciones (es decir, no existen costos de transacción, etc.);
(c) las personas pueden prestar y pedir prestado sin ningún riesgo a una tasa de interés de capitalización
continua, r ; y (d) el costo de acarreo de la mercancía subyacente de los futuros y los contratos de opciones
es una tasa conocida continua, b . Además, por comodidad, asumimos que (e) los contratos de futuros, de
opción de la mercancía, y de opciones de futuros tiene el mismo tiempo hasta su vencimiento, T .
10.1 NOMENCLATURA, NOTACIÓN Y DEFINICIONES DE LAS OPCIONES

Una opción es un contrato que proporciona a su tenedor el derecho a comprar o vender una mercancía a un
precio especificado dentro de un período de tiempo especificado. Usualmente las opciones llevan el adjetivo
de la naturaleza de la mercancía subyacente del contrato de opciones, esto es, una opción de acciones es el
derecho a comprar o vender acciones comunes, una opción de moneda extranjera es el derecho a comprar y
vender una moneda, una opción de bonos es el derecho a comprar y vender un bono, una opción de futuros
es el derecho a comprar y vender un contrato de futuros, y así sucesivamente.
FIGURA 10.1 Las Interrelaciones Entre el Mercado de

Mercancías y los Mercados para los Instrumentos
Derivados de las Mercancías
Mercando de Mercado de
Mercancías Opciones
Mercado de Mercado de
Futuros Opciones de
Futuros
Un contrato que provee a su tenedor el derecho a comprar la mercancía subyacente se llama opción de
compra (call); un contrato que provee el derecho de vender se llama opción de venta (put). En el contrato de
opciones, el precio especificado al cual se puede comprar o vender la mercancía se llama precio de ejercicio
o precio de cierre de una opción. Si el precio actual de la mercancía excede el precio de ejercicio de la
opción, la call es in-the-money y la put es out-of-the-money. Si el precio actual de la mercancía está por
debajo del precio de ejercicio de la opción, la call esta out-of-the-money y la put está in-the-money. Cuando
el precio actual de la mercancía es aproximadamente igual al precio de ejercicio de las opciones, tanto la call
como la de venta están at-the-money.
Dos tipos diferentes o estilos de contratos de opciones se negocian, las opciones Europeas y las
Americanas. Estos contratos de opciones son parecidos en muchos aspectos, excepto que las opciones
Europeas pueden ejercerse sólo al vencimiento, mientras que las opciones Americanas pueden ejercerse en
cualquier momento incluyendo el día de vencimiento.
La notación más comúnmente utilizada para representar los parámetros relacionados con el precio de
las opciones es la siguiente:
S ( ST )  precio actual (terminal aleatorio) de la mercancía
F ( FT )  precio actual (terminal aleatorio) de los futuros
X precio de ejercicio o strike
T tiempo hasta el vencimiento de la opción
c(S, T, X)  Call europea con precio de ejercicio X y tiempo hasta el vencimiento T .
p(S, T, X)  Put europea con precio de ejercicio X y tiempo hasta el vencimiento T .
C(S, T, X)  Call americana con precio de ejercicio X y tiempo hasta el vencimiento T .
P(S, T, X)  Put americana con precio de ejercicio X y tiempo hasta el vencimiento T .
Cuando el primer termino en el paréntesis de la notación de la opción es S [por ejemplo, c(S, T, X) ], la

opción es una opción sobre mercancía y el precio actual de la mercancía es S . Cuando el primer término en
el paréntesis de la notación de la opción es F [por ejemplo, c(F, T, X) ], la opción es una opción de futuros y
el precio actual de los futuros es F .
10.2 DIAGRAMAS DE GANANCIA Y NOTACIÓN VECTORIAL

En el Capítulo 1, ilustramos las ganancias al vencimiento para diferentes posiciones en opciones. En esta
sección, introducimos una notación vectorial simple que promueve el entendimiento de las contingencias de
ganancia de las posiciones complejas opción/mercancía.
La Figura 10.2 diagrama la ganancia al vencimiento, T , de diferentes posiciones como una función
del precio de la mercancía subyacente al vencimiento, ST (las primas de las opciones son ignoradas en esta
discusión). Nosotros asumimos que todas las opciones tienen en mismo precio de ejercicio, X . Una call es
rentable si ST  X y no es rentable si el precio de la mercancía no llega a exceder el precio de ejercicio al
vencimiento. La representación vectorial de este resultado es 0,1 . La primera posición indica la ganancia en
dólares por cada dólar en el precio de la mercancía que cae por debajo del precio de ejercicio al vencimiento,
y la segunda posición en el vector indica la ganancia en dólares por cada dólar en el precio de la mercancía
que excede el precio de ejercicio al vencimiento  Ganancia S  X , Ganancia S  X  . La notación vectorial
para el vendedor de una call es 0,1 . El vendedor de una call pierde si ST  X . Los resultados para el
comprador y el vendedor de una put también se muestran en la Figura 10.2. La notación vectorial para el
comprador de una put es 1,0  dado que el comprador hace dinero si el precio de la mercancía al vencimiento
está por debajo del precio de ejercicio. La notación vectorial para el vendedor de una put es  1,0 dado que
el vendedor de la put pierde si el precio de la mercancía al vencimiento cae por debajo del precio de
ejercicio.
Finalmente, la figura muestra las ganancias de una posición larga y una posición corta en la
mercancía, cada una establecida inicialmente a un precio, S , donde S  X . Nosotros asumimos que no se
incurre en ningún costo para almacenar la mercancía y que no se dan dividendos ni ningún otro pago a los
propietarios de la mercancía. La notación vectorial para la posición larga de futuros es  1,1 y para la
posición corta de futuros es 1,1 . La notación vectorial es útil para determinar la ganancia de
combinaciones de posiciones establecidas en un solo precio de ejercicio dado que se pueden agregar las
posiciones correspondientes en los vectores. Por ejemplo, el equivalente de una posición larga en futuros de
una mercancía puede establecerse comprando una call y vendiendo una put : 0,1   1,0    1,1 . Esta es
una posición larga sintética en futuros. Otra manera de ver esto es sumar verticalmente las ganancias de las
posiciones de compra de la call y la venta de la put en la Figura 10.2. Se puede establecer el equivalente de
una posición corta en futuros, una posición corta sintética en futuros, vendiendo una call y comprando una
put: 0,1  1,0   1,1 .
FIGURA 10.2a Flujo de caja al Vencimiento de las Posiciones
Larga y Corta de las Calls (Ignorando la Prima Inicial).
Utilidad en el momento T
Comprar Call  0,1
SX
Precio de la Mercancía, S
Vender una Call  0, 1
FIGURA 10.2b Flujo de caja al Vencimiento de las Posiciones

Larga y Corta de las Puts (Ignorando la Prima Inicial).
Utilidad en el momento T
Comprar Put 1, 0 
SX Precio de la Mercancía, S
Vender una Put  1, 0 
FIGURA 10.2c Flujo de caja al vencimiento para

posiciones larga y corta de Futuros
Ganancia en T
Compra Futuros  1,1
SX F
Precio de Commodity, S
Vende Futures 1, 1
Así como se pueden utilizar las opciones para sintetizar las posiciones en futuros para la mercancía
subyacente, se puede utilizar una posición en la mercancía subyacente combinada con una opción para
obtener otra opción. Una posición larga en una call 0,1 , por ejemplo, se puede repetir comprando futuros de
la mercancía y comprando una put:  1,1  1,0   0,1 . Similarmente, una posición larga en una put, 1,0  se
puede repetir vendiendo corto los futuros de la mercancía y comprando una call: 1,1  0,1  1,0  . Una
estrategia de inversión frecuente es vender una call contra una posición larga en futuros de la mercancía
subyacente, 0,1   1,1 lo cual produce la misma ganancia que vender una put, 0,1 .
Una posición que produce dinero sin importar la dirección en que cambie el precio es un “straddle”
(call y put al mismo precio de ejercicio). Un “straddle” consiste en comprar una call y una put:
0,1  1,0  1,1 . Por supuesto, esta posición tiene un precio alto porque el vendedor del straddle debe ser
compensado por las pérdidas esperadas por vender el straddle. Alguien que piensa que es probable que se
eleve el precio de la mercancía subyacente podría comprar dos calls. La notación vectorial para eso es  0, 2 
porque el valor de las opciones sube dos dólares por cada dólar de incremento en el precio de la mercancía
por encima del precio de ejercicio.
10.3 DISCUSIÓN GENERAL DE LÍMITES DE PRECIO INFERIORES PARA LAS

OPCIONES
El límite de precio inferior para una opción es el precio más bajo al que se puede vender
una opción correspondiente a cada precio de la mercancía subyacente. Antes del
vencimiento, usualmente una opción se venderá por más que el límite inferior porque la
opción tiene un potencial de ganancia que excede su potencial de pérdida. En esta
sección, desarrollamos la intuición que fundamenta los límites de precio inferiores para
las calls y puts Europeas y Americanas. En la siguiente sección, mostramos las
transacciones de la cartera de arbitraje que aseguran que se mantenga cada límite.
Opciones Europeas
Al vencimiento, los límites inferiores de las opciones Europeas son dados por el diagrama de ganancia de la
Figura 10.2. La call se vende por su valor de ejercicio, esto es, c  max[0, ST  X ] y la put se vende por su
valor de ejercicio, p  max[0, X  ST ] . Antes del vencimiento, una opción europea se vende en por lo menos
el valor actual de su valor de ejercicio más una asignación por el valor actual del costo de almacenar la
mercancía. El precio de la call debe satisfacer la siguiente condición para una opción in-the-money:
S (ebT  1) S  X
c( S , T ; X )   rT (10.1)
e rT e
donde S es el precio del activo subyacente en algún momento antes del vencimiento y T es el momento
hasta el vencimiento.86 Si el precio de la mercancía al vencimiento fuese S , el valor de ejercicio de la call
sería S  X . El valor antes del vencimiento es el valor descontado, el segundo término en (10.1). La call
también debe reflejar los costos de almacenamiento asociados con una posición larga en la mercancía ya que
al comprar una opción se evitan estos costos de almacenamiento. El valor actual de los costos de
almacenamiento, incluyendo el costo del interés de los fondos atados en S , es el primer término de (10.1).
Dado que la opción podría estar out-of-the-money, condición completa del límite inferior, la cual se muestra
en la Figura 10.3, es
 S (ebT  1) S  X 
c( S , T ; X )  max 0,  rT  . (10.2a)
 e rT e 
La condición establece que una call europea nunca se puede vender por menos de la cantidad dada en el lado
derecho de (10.2a). Correspondientemente, el límite inferior para una put europea es
  S (ebT  1)  S  X 
p ( S , T ; X )  max 0,  (10.2b)
 e rT e rT 
FIGURA 10.3 Limite Inferior del Precio de una Call Europea
Precio de la "Call", c(S,T; X)
Precio de "Commodity"
X/e bT S
S (ebT  1) S  X
Para S  X / ebT , c(S , T ; X )   rT
erT e
Para los valores de put que no son cero, el límite inferior es el valor actual del valor de ejercicio de la put
más una ajuste por costos de almacenamiento.
Opciones Americanas
Una característica singular de las opciones Americanas, a diferencia de las opciones Europeas, es que se
pueden ejercer antes del vencimiento. Dado que el derecho de ejercer antes no puede tener un valor negativo
(es decir, a usted no hay que pagarle para inducirlo a asumir un privilegio), se aplican las siguientes dos
condiciones:
86
El lado derecho de (10.1) se puede simplificar a Se ( br )T  Xe  rT , lo cual se hará más adelante en el capítulo.
C ( S , T ; X )  c( S , T ; X ) (10.3a)
y P( S , T ; X )  p( S , T ; X ) (10.3b)
Estas condiciones no quieren decir que las opciones Americanas tengan mayor valor que las Europeas –sólo
que no pueden tener valores menores. El derecho de ejercer antes puede no tener valor positivo, pero nunca
tendrá valor negativo.
Dado que las opciones Americanas no se pueden vender a menos que las opciones Europeas, el límite
inferior de las opciones Americanas es el límite inferior de la opción europea correspondiente al mismo
precio de ejercicio y vencimiento. Una opción americana tiene el beneficio adicional que puede ser ejercido
inmediatamente para recibir el valor de ejercicio S  X para la call americana y X  S para la put americana.
Esto significa que el límite inferior para la opción americana es el límite inferior para la opción europea o el
valor de ejercicio actual, el que fuese mayor. Los límites inferiores son
 S (ebT  1) S  X 
C ( S , T ; X )  max 0, rT
 rT , S  X  (10.4a)
 e e 
  S (e bT  1)  S  X 
P ( S , T ; X )  max 0, rT
 rT
, X  S (10.4b)
 e e 
10.4 PRUEBAS POR ARBITRAJE DE LOS LÍMITES INFERIORES DE PRECIO

Antes de continuar con los límites de precio de las opciones de mercancías, vale la pena re-introducir el
concepto de una posición “rollover”. Recuerden que en el Capítulo 3 introdujimos un fondo de
almacenamiento que era invertido en la mercancía. La suma del fondo de almacenamiento y la posición en
una unidad de la mercancía se llama una posición de rollover en la mercancía. En el Capítulo 3, utilizamos
una cartera que consistía de una posición corta en futuros y una posición de rollover larga en la mercancía
subyacente para demostrar el arbitraje de base. En este capítulo, utilizamos una posición de rollover de la
mercancía para demostrar los vínculos entre el precio de una opción y la mercancía subyacente. Una posición
de rollover larga en la mercancía subyacente se inicia con una inversión de Se(b  r )T en la mercancía
subyacente al final del día 0. Al final de cada día subsiguiente, la posición es reducida (incrementada) por el
factor Se  ( br ) si b  r (b  r ) . Al final del día T , exactamente una unidad de la mercancía está disponible.
También vale la pena señalar que un contrato de futuros es una mercancía que no cuesta nada mantener, esto
es, la tasa de costo-de-acarreo, b , es igual a cero. Por lo tanto, una posición de futuros rollover se inicia con
e  rT contratos de futuros al final del día 0 y se incrementa cada día por el factor e r .
Call Europea
El límite de precio inferior de una call europea se puede determinar considerando el valor inicial y final de
una cartera que consiste de una posición larga en una call europea c( S , T ; X ) , una posición larga de
Xe rT bonos libres de riesgo, y una posición rollover corta en la mercancía Se(b  r )T como se muestra en la
Tabla 10.1.87 La cantidad de préstamo libre de riesgo se determina por el hecho que necesitamos disponible
X en el momento T para ejercer la call. Si, al vencimiento de una opción, el precio de la mercancía excede
el precio de ejercicio, la call es ejercida y la unidad de la mercancía subyacente recibida es utilizada para
cubrir la posición corta de la mercancía. Para pagar el precio de ejercicio de la call, se utilizan exactamente
los bonos libres de riesgo. Por lo tanto, el valor final neto de la cartera es cero. Por otro lado, si al
vencimiento el precio de la mercancía cae por debajo del precio de ejercicio, la call vence sin valer nada y el
valor de los bonos libres de riesgo excede el valor necesario para cubrir la posición corta de la mercancía. En
87
Noten que por comodidad en la tabla se han suprimido los términos de la notación de precio de opción.
este caso, el valor final de la cartera es positivo. Dado que el tenedor de la cartera se asegura de no tener un
valor final negativo por su cartera, el valor inicial debe ser no-positivo, esto es,
 c( S , T ; X )  Se( b  r )T  Xe  rT  0
o
c( S , T ; X )  Se( b  r )T  Xe  rT (10.5a)
de lo contrario, son posibles ganancias de arbitraje. Esta condición es la misma que (10.1), la cual se puede
observar sumando y restando Se  rT en el lado derecho de (10.5a).
La condición (10.5a) muestra sólo uno de los límites de precio inferiores de la call europea. En un
mercado racional, el precio de la opción nunca será negativo ya que es un derecho y no una obligación. Por
lo tanto, la condición completa del límite de precio inferior para una call europea es
c( S , T ; X )  max[0, Se( b  r )T  Xe rT ] (10.6a)
la cual se puede demostrar es igual a (10.2a).
TABLA 10.1 Operaciones de Arbitraje para establecer el límite inferior de precio de una call europea
Valor Final
Posición Valor Inicial ~ ~
ST  X ST  X
c ~
Comprar Call europea 0 S X T
 rT
Comprar bonos libres de riesgo  Xe X X
~ ~
Vender posición rollover en mercancía Se(b  r )T  ST  ST
~
Valor neto de la cartera  c  Se (br )T  Xe  rT X  ST 0
Call Americana
Dado que la opción americana siempre se vende por más que la correspondiente opción europea, la call
americana está limitada desde abajo por (10.6a). También es limitada desde abajo por los ingresos de ejercer
inmediatamente la call, S  X . De lo contrario, se podrían obtener ganancias libres de riesgo al comprar la
call y ejercerla inmediatamente. Por lo tanto, la condición completa del límite de precio inferior para la
opción americana es
C ( S , T ; X )  max[0, Se( b  r )T  Xe  rT , S  X ] (10.7a)
que se puede demostrar es igual que (10.4a).
Si b  r el segundo término en los corchetes de (10.7a) excede al tercero, lo cual significa que el límite
inferior de la call americana es Se(b  r )T  Xe  rT , el límite inferior europeo. En otras palabras, cuando b  r ,
una call americana se comporta como una call europea; no se ejercerá hasta el vencimiento. Uno puede
escribir la condición de manera que el segundo término en corchetes de (10.7a) exceda el tercer término en
corchetes:
1  e rT
SX (10.8)
1  e ( b  r )T
Si se satisface la condición b  r , la call americana tendrá el mismo límite inferior que la call europea. La
condición, (10.8) se cumplirá si b  r (teniendo en mente que estamos buscando los valores de S para los
cuales el límite inferior de la call exceda a cero). Por otro lado, para los valores b  r , el segundo término en
los corchetes de (10.7a) puede ser menor que el tercer término con lo cual no se cumple (10.8). Esto implica
que el límite de la call americana es S  X . Las opciones que no satisfacen (10.8) pueden ejercerse antes del
vencimiento y tendrán un valor mayor que la opción europea correspondiente.
Put Europea
El límite de precio inferior para una put europea se puede derivar considerando el valor inicial y final de una
cartera que consiste de una posición larga en una put europea, p( S , T ; X ) , una posición rollover larga en la
mercancía subyacente, Se(b  r )T y una posición corta de Xe rT bonos libres de riesgo, como se muestra en la
Tabla 10.2. La cantidad de préstamo libre de riesgo se determina por el hecho que el ejercicio de la put al
vencimiento proporciona X . Por lo tanto, si al vencimiento de la put el precio de la mercancía cae por debajo
del precio de ejercicio, la mercancía disponible es vendida por el precio de ejercicio de la put al ejercer la
put. Los ingresos del ejercicio son entonces utilizados para pagar el préstamo libre de riesgo. El efecto neto
es que el valor final de la cartera será igual a 0. Por otro lado, si al vencimiento el precio de la mercancía
excede el precio de ejercicio, el valor neto de la cartera será positivo porque la put vence sin valer nada y el
precio de la mercancía excede la cantidad necesaria para pagar el préstamo libre de riesgo. Dado que la
cartera proporciona un valor final no-negativo, entonces es valor inicial debe ser no-positivo. Si el valor
inicial neto de la cartera es no-positivo, esto es si  p( S , T ; X )  Se( b  r )T  Xe  rT  0 , Entonces
p( S , T ; X )  Xe  rT  Se( b  r )T (10.5b)
Agregando la restricción de no-negatividad del valor de la put europea,

p( S , T ; X )  max[0, Xe  rT  Se( b  r )T ] (10.6b)
que se puede demostrar es igual a (10.2b).
TABLA 10.2 Operaciones de Arbitraje para establecer el límite inferior de precio de una put europea
Valor Final
ST  X ST  X
p ~
Comprar Put europea X  ST 0
~ ~
Comprar posición rollover en mercancía  Se( b  r )T ST ST
Pedir prestado Xe  rT Xe  rT X X
~
Valor neto de la cartera  p  Se( b  r )T  Xe  rT 0 ST  X
Put Americana
Naturalmente, la condición del límite de precio inferior para la put europea también se aplica a la put
americana. Pero, los ingresos del ejercicio de la put americana, X  S , pueden ser mayores que
Xe  rT  Se( b  r )T y sabemos que en un funcionamiento racional la put americana está limitada desde abajo por
X  S . De otra manera, se podrían obtener ganancias de arbitraje libres de costo al comprar una put y
ejerciéndola inmediatamente. De manera que la condición completa del límite inferior para la put americana
es
P( S , T ; X )  max[0, Xe rT  Se( b  r )T , X  S ] (10.7b)
que es igual que (10.4b).

Vale la pena notar que por lo menos parte del límite de precio inferior de la put europea está en el
lado izquierdo de los ingresos del ejercicio de la put americana, independientemente de si el costo de acarreo
es mayor o menor que la tasa de interés libre de riesgo. (Esto se muestra en la Figura 10.5.) La implicancia es
que, para las opciones de mercancías, no se puede decir que la put americana nunca será ejercida antes.
Opciones de Futuros
La Tabla 10.3 contiene un resumen de las condiciones de límite de precio inferior para opciones de
mercancías y futuros. Recuerden que anteriormente en este capítulo vimos que un contrato de futuros es una
mercancía con una tasa de costo de acarreo cero. Los límites de precio en la última columna realizan esta
sustitución.
TABLA 10.3 Resumen de los límites de precio inferior para las opciones Europeas y Americanas
de mercancías y futuros
Tipo de Opción Opción de Mercancía Opción de Futuros
( b  r )T  rT
Call europea max[0, Se  Xe ] max[0, ( X  F )e  rT ]
Call americana max[0, Se( b  r )T  Xe rT , S  X ] max[0, F  X ]
Put europea max[0, Xe  rT  Se (b  r )T ] max[0, ( F  X )e  rT ]
Put americana max[0, Xe  rT  Se( b  r )T , X  S ] max[0, X  F ]
10.5 EJERCICIO TEMPRANO O ANTES DEL VENCIMIENTO

Anteriormente, vimos que el derecho de ejercer temprano de las opciones Americanas es un privilegio y debe
tener un valor no-negativo. Si consideramos el valor de las opciones Americanas como la suma de sus
contrapartes Europeas más sus respectivas primas de ejercicio temprano,  C y  P , esto es,
C ( S , T ; X )  c( S , T ; X )   C ( S , T ; X ) (10.9a)
y
P( S , T ; X )  p( S , T ; X )   P ( S , T ; X ) (10.9b)
es obvio que las opciones Americanas se venden en por lo menos tanto como sus contrapartes Europeas.
Sin embargo, esto no quiere decir que las opciones Americanas siempre se ejercerán temprano, o
inclusive que puedan serlo. Por ejemplo, en el caso de las opciones sobre acciones que no pagan dividendos,
la call americana nunca se ejercerá temprano.
Call Americana
El que la call americana escrita sobre una mercancía puede ser ejercida temprano depende de la tasa del
costo de acarreo, b . Si b  r , la call americana no se ejercerá temprano. Para ver esto, consideren la
condición de límite inferior (10.7a) o, alternativamente, la Figura 10.4a. Si b  r , los ingresos del ejercicio
de la call, S  X , son siempre menos que el valor mínimo por el cual se negocia la call en el mercado,
Se(b  r )T  Xe  rT .
Dado que la call americana vale más cuando no se ejerce o “viva” que ejercida o “muerta”, nunca se ejercerá
antes del vencimiento. Por lo tanto, si b  r , la prima de ejercicio temprano de la call americana no vale
nada, esto es
 C (S ,T ; X )  0 (10.10)
y, de la ecuación (10.9a), la call americana tiene un valor igual a la call europea.
C ( S , T ; X )  c( S , T ; X ) . (10.11)
La intuición del hecho que la opción americana no será ejercida temprano cuando b  r se pude desarrollar
más fácilmente considerando la cantidad mínima perdida por el ejercicio temprano, esto es
S  X  [ Se( b  r )T  Xe  rT ] (10.12)
Despejando (10.12) para aislar los términos S y X , tenemos

S [1  e(b  r )T ]  X [1  e  rT ] (10.13)
La expresión (10.13) dice lo siguiente: Si la call americana se ejerce ahora en lugar de al vencimiento, el
tenedor de la call americana pierde de dos maneras. Primero, incurre en el valor actual de los costos de
almacenamiento que tendrá que pagar como resultado de aceptar la entrega de la mercancía subyacente,
S [1  e( b  r )T ] 88. Al mantener la opción, el tenedor de la opción no tiene inversión directa en la mercancía
subyacente –sólo el derecho de compra la mercancía en el futuro. Si acepta la entrega, ahora él enfrenta el
prospecto de almacenar la mercancía, asegurarla, etc. Segundo, él incurre en el valor actual del interés sobre
el precio de ejercicio de la opción, X [1  e  rT ] . Si la opción se ejerce ahora, el tenedor de la opción está
obligado a realizar el pago de la cantidad X ahora en lugar de más adelante, renunciando así al ingreso por
interés que hubiese podido ganar sobre el precio de ejercicio de la opción. La Figura 10.4a resume
convenientemente estos efectos mostrando que el límite de precio inferior de la call europea excede los
ingresos del ejercicio de la call americana para todos los valores posibles del precio de la opción.
FIGURA 10.4a Precios de call europea y americana como función del precio
del subyacente cuando el costo de acarreo (b) excede la tasa de interés (r)
( b  r ) de manera que la call americana no se ejerza óptimamente antes del
vencimiento
Precio de la call,
c(S,T; X) and C(S,T; X)
Límite Inferior del Precio de la

Call Europea, Se( br )T  Xe rT
C(S,T;X)  c(S,T;X) Valor de Ejercicio de la

Call Americana, S  X
Precio de la
X/e bT
X Mercancía, S
En el caso donde la tasa del costo de acarreo, b , es menor que la tasa de interés libre de riesgo, r ,
puede ser óptimo ejercer temprano la call americana, tal como se ve al examinar la expresión (10.13).
Cuando b  r , existen influencias compensadoras que afectan la decisión sobre el ejercicio temprano. Por un
lado, el diferir el ejercicio temprano permite al tenedor de la call obtener implícitamente interés sobre el
precio de ejercicio de la opción, como vimos anteriormente. Por otro lado, b  r significa que el tenedor
obtiene alguna forma de rendimiento sobre la mercancía subyacente. Por ejemplo, supongamos que la call es
escrita sobre un índice de acciones y que la cartera del índice de acciones paga dividendos a una tasa
conocida.89 Diferir el ejercicio significa que se está obteniendo un interés, pero que se renuncia al dividendo.
Noten que mientras mayor es el valor del precio actual de la mercancía, S , mayor será el valor de la
88
Dado que b  r , este término es no-positivo.
89
Recuerden que la tasa de costo de acarreo, b , consiste de la tasa de interés, r , más el costo de almacenamiento, seguro, etc.
Dado que el único “costo” aparte del interés involucrado en el mantenimiento de una cartera de acciones el es pago de dividendos
(es decir, un costo negativo), b  r .
expresión (10.13) y mayor será la rentabilidad del ejercicio temprano. El hecho se refleja en la Figura 10.4b,
mediante la creciente distancia entre el valor de ejercicio de la call americana y el límite de precio inferior de
la call europea conforme aumenta el precio de la mercancía subyacente. Sin embargo, en cualquier nivel de
precio de la mercancía existe un beneficio no-cero para el ejercicio temprano si b  r , de manera que la
prima de ejercicio temprano tiene un valor positivo,
 C (S ,T ; X )  0 (10.14)
y la call americana tiene un mayor valor que la call europea,
C ( S , T ; X )  c( S , T ; X ) (10.15).
FIGURA 10.4b Precio de la call europea y americana como función del precio
del subyacente si el costo de acarreo (b) es menor que la tasa de interés (r)
b  r de manera que la opción americana pueda ser ejercida temprano.
Precio de la Call, Límite Inferior de Precio de la

c(S,T; X) y C(S,T; X) Call Europea, Se( br )T  Xe rT
C(S * ,T;X)
C(S,T;X)
Valor de Ejercicio de la
Call Americana, S  X
c(S,T;X)
 C (S,T;X)
Precio de la Mercancía, S
X(e -rT -1 )
X/e bT X S*
e(b-r)T -1
La Tabla 10.4 contiene un ejemplo de una call para la cual el ejercicio temprano puede ser óptimo. La
mercancía subyacente en este ejemplo es una moneda extranjera. La tasa de interés de Estados Unidos se
asume como 8 porciento anual y la tasa de interés extranjera se asume como 12 porciento anual. Como
resultado, el costo de acarreo es negativo, esto es, -4 porciento anual. El precio de ejercicio de la call es 150
centavos y el vencimiento es en 30 días. En este ejemplo, el ejercicio temprano ocurre si el precio de la
moneda alcanza los 165 centavos. En este punto, el precio de la opción americana es igual al límite
americano porque la opción es valorada sobre la premisa que será ejercida. El ejercicio temprano es deseable
porque el tomar posesión de la moneda e invirtiéndola en el país extranjero proporciona una tasa
relativamente alta de ingreso por interés (12 porciento versus 8 porciento localmente). Si la opción se
mantuviese hasta el vencimiento, su valor es sólo 14.5692 (es decir, el valor de la opción europea). No
importa que, para precios entre 161 y 164, el límite inferior americano exceda el precio de la opción europea
porque el ejercicio temprano no es óptimo. En este rango, el precio de mercado de la opción americana
excede el límite inferior de la opción americana, por lo tanto el ejercicio no es óptimo.
Put Americana
La condición (10.7b), así como la Figura 10.5, muestran que siempre existe una oportunidad para que
cualquier put americana sea ejercida óptimamente antes del vencimiento. Siempre hay alguna región de los
precios de las mercancías sobre el cual los ingresos del ejercicio de la put serán mayores que la condición del
límite de precio inferior de la put europea. Para lograr una intuición sobre la naturaleza del intercambio
involucrado aquí, veamos la diferencia de los ingresos del ejercicio de la put americana y el límite de precio
inferior de la put europea, como lo hicimos con la call americana en la expresión (10.13). La diferencia es
X  S  [ Xe  rT  Se( b  r )T ] (10.16)
o, simplemente
X [1  e  rT ]  S [1  e (b  r )T ] (10.17)
TABLA 10.4. Límites Inferiores y Precios para las Calls Europeas y Americanas de moneda
extranjera. El precio de ejercicio de la opción ( X ) es 150, y el tiempo al vencimiento ( T ) es 30 días
(0.08219 años). La tasa de interés libre de riesgo doméstica ( rd ) es 8 porciento y la tasa de interés libre
de riesgo extranjera ( r f ) es 12 porciento. La tasa de costo de acarreo es por lo tanto –4 porciento. (La
moneda subyacente tiene una tasa de volatilidad anual de 20 porciento).
Límite Inferior Límite Inferior Precio de Opción Precio de Opción
Precio Actual
Europeo Americano Europea Americana
150 0.0000 0.0000 3.1637 3.1991
151 0.5011 1.0000 3.6704 3.7125
152 1.4913 2.0000 4.2224 4.2726
153 2.4814 3.0000 4.8784 4.8189
154 3.4716 4.0000 5.4581 5.5288
155 4.4618 5.0000 6.1381 6.2219
156 5.4520 6.0000 6.8564 6.9556
157 6.4422 7.0000 7.6103 7.7276
158 7.4324 8.0000 8.3969 8.5355
159 8.4226 9.0000 9.2132 9.3767
160 9.4127 10.0000 10.0560 10.2489
161 10.4029 11.0000 10.9225 11.1496
162 11.3931 12.0000 11.8097 12.0769
163 12.3833 13.0000 12.7149 13.0289
164 13.3735 14.0000 13.6355 14.0041
165 14.3637 15.0000 14.5692 15.0000
166 15.3538 16.0000 15.5138 16.0000
167 16.3440 17.0000 16.4677 17.0000
168 17.3342 18.0000 17.4291 18.0000
169 18.3244 19.0000 18.3967 19.0000
170 19.3146 20.0000 19.3693 20.0000
El primer término de (10.17) es el valor actual del interés que se puede obtener si se ejerce la opción
inmediatamente. Si el tenedor de la opción ejerce su put, él recibe X y entrega la mercancía subyacente que
vale S .
Los ingresos del ejercicio se puede invertir inmediatamente para obtener un interés. El efecto neto del
segundo término puede ser positivo o negativo, dependiendo de si b  r o b  r . En el primer caso, ejercer la
opción significa que el tenedor de la opción puede entregar la mercancía subyacente y renunciar a los costos
de almacenamiento involucrados con la mercancía que actualmente le está proporcionando algún tipo de
rendimiento. Puede estar renuente a hacerlo, pero la expresión (10.17) será positiva para los casos donde el
rendimiento sobre la mercancía es menor que la tasa de interés libre de riesgo.
La posibilidad de ejercicio temprano de una put americana también se puede ver geométricamente en
la Figura 10.5. Independientemente del valor de b , siempre existe un rango de precios de la mercancía por
encima de los ingresos del ejercicio de la put americana, X  S , en el cual son mayores que aquellos de la
put europea; por lo tanto, siempre existe la posibilidad que la put americana sea ejercida antes del
vencimiento. Para ver si este es el caso, consideren qué sucede si el precio de la mercancía cae por debajo de
cero. El valor de la put americana es igual al precio de ejercicio de la put dado que el tenedor de la put
americana tiene el derecho a vender una mercancía con precio cero por X en cualquier momento.
FIGURA 10.5 Precios de Puts Europea y Americana como Función del precio del
Subyacente
Precios de Puts,
p(S,T; X) y P(S,T; X)
X
Limite inferior del Precio de
P(S**,T;X)
Put Europea, Xe-rT - Se(b - r)T
Xe -rT
P(S,T;X) Valor de Ejercicio de

Put Americana, X - S
p(S,T;X)  P(S,T;X)
0 S ** Xe -bT X Precio de Commodity, S
De hecho, en el caso que el precio de la mercancía caiga por debajo de cero, el tenedor de la put
americana ejerce su opción inmediatamente porque (a) él puede comenzar a ganar un interés sobre los
ingresos del ejercicio, y (b) el precio de la mercancía puede elevarse en cuyo caso el precio de la put caerá.
Sin embargo, en S  0 , la put europea tiene un valor de Xe  rT . Para reconocer esto, consideren las
condiciones de límite impuestas sobre el precio de la put. El límite de precio inferior está dado por la
condición (10.2b). En S  0 , el valor mínimo para la put es Xe rT . Por otro lado, dado que el precio de la
mercancía no puede ser menor que cero al vencimiento de la opción, el valor actual de los ingresos máximos
del ejercicio es Xe  rT . Si el precio de la put europea está limitado por arriba y por abajo por Xe  rT se deduce
que el precio es Xe  rT . Dado que la put americana puede ejercerse inmediatamente para ingresos iguales a
X mientras que la opción europea tiene un valor menor, la prima de ejercicio temprano debe ser positiva
para S  0 . En general, mientras exista alguna posibilidad de ejercicio temprano, la prima de ejercicio
temprano tiene valor positivo.
 P (S ,T ; X )  0 (10.18)
y la put americana vale más que la put europea,

P( S , T ; X )  p( S , T ; X ) (10.19)
Opciones Americanas sobre Futuros

A diferencia de la opción americana de una mercancía subyacente, no existen condiciones bajo las cuales se
puede decir que la call o la put americana de un contrato de futuros no sea ejercida antes del vencimiento.
Uno podría pensar en un contrato de futuros como una mercancía con costo de acarreo cero (b  0) y
analizar la pérdida mínima del ejercicio temprano para calls (10.13) y puts (10.17). Cuando b  0 , ambas
expresiones son positivas, lo cual significa que el ejercicio temprano de una opción americana de futuros in-
the-money podría ser rentable.
La intuición para el posible ejercicio temprano de las opciones de futuros Americanas es directa.
Consideren una call sobre un contrato de futuros. Si se ejerce la call, se establece una posición larga en
futuros para el comprador de la call al precio del ejercicio de la call. Pero el pago del precio de ejercicio no
se requiere así como el precio de los futuros no se requiere cuando se inicia un contrato de futuros; sólo se
pagan las ganancias o las pérdidas. Si la opción está in-the-money, las ganancias son pagadas al comprador
al ejercicio de la call. Por ejemplo, consideren futuros de trigo a X  $3.00 por tonelada, asuman que el
precio actual de los futuros es $3.50 por tonelada. Si se ejerce una call sobre un contrato de futuros, el
comprado de la call asume una posición en futuros a un precio de futuros de $3.00 Dado que el precio actual
es $3.50, él ha pagado la ganancia de 50 centavos por tonelada, o $2,500 sobre un contrato de 5,000
toneladas. Estas ganancias pueden ser invertidas inmediatamente para obtener un interés. Al mismo tiempo,
la propiedad del contrato de futuros no impone el costo de acarreo que se incurriría si se poseyera una
mercancía subyacente. El ejercicio temprano podría ser deseable porque permite que las ganancias sean
invertidas antes. Por supuesto, normalmente el deseo de ejercicio temprano se compensa por la pérdida de la
protección contra la baja que proporciona la opción. La intuición es la misma para una put de futuros. Al
ejercer una put sobre un contrato de mercancías cuyo precio ha caído, las ganancias se pueden recibir antes y
se puede obtener un interés.
10.6 DISCUSIÓN GENERAL SOBRE LA PARIDAD DE PUT Y CALL

La paridad de put y call se refiere a la relación entre el precio de una put y el precio de una call. Mostramos
antes que el resultado al vencimiento de la compra de una call podía ser igualado por una posición larga en la
mercancía más la compra de una put. Dado que uno puede comprar una call directa o indirectamente al
comprar una put y asumir una posición larga, los precios de las puts y las calls deben estar claramente
relacionados. Aquí limitamos nuestra discusión a las opciones Europeas que son mantenidas hasta el
vencimiento.
Los convertidores de opciones toman ventaja de las discrepancias en los precios de las puts y las
calls. Si los precios de las calls son demasiado altos, ellos pueden vender las calls, comprar puts y asumir una
posición larga en futuros sobre la mercancía subyacente. La notación vectorial para la posición resultante es
0,1  1,0   1,1  0,0 .
Vender una call y comprar una put es una posición corta sintética. Al asumir una posición larga en
una mercancía subyacente, se establece una cobertura perfecta, con el resultado que sin importar los cambios
en el precio de la mercancía, no se dará ninguna pérdida o ganancia. Dado que la posición final está libre de
riesgo, las ganancias para los convertidores deben ser cero en equilibrio. Nosotros asumimos primero que el
precio de ejercicio de las opciones, X , es igual al precio actual de la mercancía, S . En ese caso, el valor
actual de los flujos de efectivo asociados con la posición protegida es el ingreso de vender la call, c( S , T ; X ) ,
el costo de la put,  p( S , T ; X ) , y el valor actual del costo de asumir una posición larga en la mercancía,
 S (ebT  1) / e rT . El costo de estar largo incluye el costo del interés de los fondos atados más cualquier costo
de almacenamiento menos cualquier pago de ingreso. El equilibrio requiere que la suma de estos flujos de
efectivo sea cero:
S (ebT  1)
c( S , T ; X )  p( S , T ; X )  0 o
e rT
S (ebT  1)
c( S , T ; X )  p ( S , T ; X )  (10.20)
e rT
Si los precios de las puts fuesen altos con relación a los precios de las calls, los convertidores venderían puts,
 1,0 , comprarían calls, 0,1 y asumirías posiciones cortas en futuros sobre la mercancía subyacente,
1,1 , para establecer una cobertura perfecta. Bajo la premisa que la posición corta puede obtener ingresos
de almacenaje que paga la posición larga, se da la misma relación de equilibrio.
Se requiere un ligero ajuste a la relación de paridad de put y call si el precio en el cual se establece la
mercancía subyacente difiere del precio de ejercicio de las opciones. Si S  X , la relación de paridad put y
call las opciones Europeas es
S (ebT  1) S  X
c( S , T ; X )  p( S , T ; X )   rT (10.21)
e rT e
donde el último término da cuenta que la call o la put está in-the-money. Aunque la ecuación de la paridad
put y call (10.21) se ve complicada, la idea básica es muy simple. La relación de paridad put y call
simplemente dice que el precio de la call menos el precio de la put es el valor actual del costo de mantener la
mercancía subyacente hasta el vencimiento de las opciones más el valor actual de la cantidad por la cual el
precio de la mercancía excede el precio de ejercicio. Nosotros asumimos que los costos de almacenamiento
son incurridos a una tasa continua, b , pero se puede realizar otras suposiciones o premisas. Por ejemplo, los
costos de mantenimiento podrían ser pagados al inicio como una suma abultada, B . En ese caso, el primer
término en el lado derecho de (10.21) sería B .
Para ilustrar la paridad put y call para las opciones Europeas, consideren el ejemplo de la Tabla 10.4.
Específicamente, cuando el precio de la moneda subyacente es 155, el precio de la call europea es 6.1381.
Sobre la base de este precio, el precio de la put europea se puede calcular como
155(e 0.04 ( 0.08219 )  1)  155  150
6.1381  p ( S , T ; X )   4.4618
e0.08( 0.08219 )
Por lo tanto, el valor de la put europea dado por la paridad put y call es 1.6763.
La relación put y call arriba establecida puede no mantenerse exactamente para las opciones
Americanas porque el ejercicio temprano de una opción americana puede romper la cobertura libre de riesgo.
Por ejemplo, el convertidor que vende puts, compra calls y se mantiene corto en la mercancía subyacente,
puede ejercer la put. La mercancía entregada al convertidor al precio de ejercicio puede utilizarse para pagar
la mercancía prestada para la venta corta, pero el convertidor también debe liquidar la inversión de los
ingresos de la venta corta (los cuales son necesarios para pagar la mercancía entregada) y eso se podría hacer
a pérdida. Como resultado, para las opciones Americanas, se podrían establecer límites sobre la diferencia
entre los precios de las calls y las puts, algo que se hace en la siguiente sección.
10.7 PRUEBAS DE LAS RELACIONES DE PARIDAD PUT Y CALL
Opciones Europeas
La relación de paridad put y call, (10.21) establecida anteriormente para las opciones de mercancías
Europeas, también se puede escribir como:
c( S , T ; X )  p( S , T ; X )  Se( b  r )T  Xe  rT (10.22)
Para entender como se deriva esta relación. Consideren una cartera de inversión que consiste de vender una
call europea, comprar una put europea al mismo precio de ejercicio, comprar una posición rollover en la
mercancía subyacente comenzando con e(b  r )T unidades, y pidiendo prestado Xe rT a la tasa de interés libre
de riesgo. El valor inicial y final de esta cartera se presentan en la Tabla 10.5. Noten que los valores finales
son iguales a cero, independientemente de si el precio final de la mercancía está por encima o por debajo del
precio de ejercicio de las opciones. Si al vencimiento la call está in-the-money, se requiere que entreguemos
una unidad de la mercancía y recibamos X . Por virtud de la posición rollover, tenemos disponible una
unidad de la mercancía para realizar la entrega. Los ingresos del ejercicio son utilizados para cubrir los
préstamos libres de riesgo, y la put vence sin valer nada. Si al vencimiento la put está in-the-money,
ejercemos la put, entregando la mercancía y recibiendo X . Las transacciones restantes son como las arriba
descritas. Dado que los valores finales son con seguridad iguales a cero, debe ser el caso que nadie pagará un
precio distinto a cero por asumir la cartera. El establecer el valor neto inicial de la cartera igual a cero
produce la ecuación (10.22).
TABLA 10.5 Operaciones de Arbitraje para establecer la paridad put y call para las opciones
Europeas
Valor Terminal
ST  X ST  X
~
Vender call europea c  ( ST  X )
p ~
Comprar put europea X  ST
~ ~
Comprar posición rollover en la mercancía  Se( b  r )T ST ST
Pedir prestado Xe  rT Xe  rT X X
Valor neto de la cartera c  p  Se(b  r )T  Xe  rT 0 0
La relación de paridad put y call para opciones de futuros Europeas es una caso especial de (10.22) donde la
tasa de costo de acarreo, b , es igual a cero. (Recuerden que las posiciones en futuros no requieren de ningún
desembolso de inversión inicial.) Le relación es
c( F , T ; X )  p( F , T ; X )  e  rT ( F  X ) (10.23)
Esta relación apareció primero en Black (1976).
Opciones Americanas
La característica del ejercicio temprano de las opciones Americanas causa que la especificación de la
relación de paridad put y call sea diferente a la de las opciones Europeas. Las relaciones que vinculan el
precio de la mercancía y los precios de las opciones Europeas de mercancías,
S  X  C ( S , T ; X )  P( S , T ; X )  Se( b  r )T  Xe  rT , if b  r (10.24a)
y
( b  r )T
Se  X  C ( S , T ; X )  P( S , T ; X )  S  Xe  rT , if b  r , (10-24b)
deben desarrollarse mediante dos conjuntos separados de transacciones de arbitraje. Nosotros consideramos
una desigualdad a la vez.
El lado izquierdo de la condición (10.24a) puede ser derivado considerando los valores de una cartera
que consiste de comprar una call, vender una put, prestar X libre de riesgo, y vender una posición rollover en
una mercancía comenzando con una unidad y disminuyendo la posición por el factor e  ( br ) cada día y
prestando X libre de riesgo. La Tabla 10.6 contiene estos valores de cartera. Noten que ahora en la tabla
existe una columna adicional con el título “Valor Intermedio”. Dado que el tenedor de la cartera está corto en
una opción, corre el riesgo que se le asigne la entrega sobre la opción antes del vencimiento. Debemos tener
en cuenta esta posibilidad al derivar los límites de precio racionales de las opciones Americanas.
En la Tabla 10.6, se puede ver que, si todas las posiciones del título valor se mantienen abiertas hasta
el vencimiento, el valor final de la cartera será positivo, independientemente de si el precio final de la
mercancía está por encima o por debajo del precio de ejercicio de las opciones. Si el precio final de la
mercancía está por encima del precio de ejercicio, se ejerce la call, y la mercancía adquirida al precio de
ejercicio X se utiliza para la entrega, en parte, contra la posición corta en la mercancía. Si el precio final de la
mercancía está por debajo del precio de ejercicio, el tenedor de la put ejerce su opción de vendernos la
mercancía subyacente al precio de ejercicio X . A su vez, utilizamos la mercancía para entregarla contra
nuestra posición corta en la mercancía establecida desde el inicio. Por lo tanto, si se mantienen las posiciones
de las opciones hasta el vencimiento, el valor final de la cartera es con seguridad positivo.
En el caso que el tenedor de la put decida ejercer su opción antes del vencimiento en un momento t ,
la inversión en bonos libres de riesgo es más que suficiente para cubrir el pago del precio del ejercicio al
tenedor de la put, y la mercancía recibida del ejercicio de la put es utilizada para cubrir la mercancía vendida
cuando se formó la cartera. Además, aún mantenemos la call, la cual puede tener un valor significativo. En
otras palabras, formando una cartera de títulos valores en las proporciones antes mencionadas, hemos
formado una cartera que nunca tendrá un valor futuro negativo. Si el valor futuro se asegura como no-
negativo, el valor inicial se asegura como no-positivo, o  C ( S , T ; X )  P( S , T ; X )  S  X  0
TABLA 10.6. Operaciones de Arbitraje para establecer la paridad put y call para las opciones
Americanas, donde b  r .
Put Ejercida
Put Ejercida Temprano
Temprano
Valor Inicial Valor Intermedian Valor Final
~ ~
Posición ST  X ST  X
~ ~
Comprar call americana C Ct ST  X
~ ~
Vender put americana P  ( X  St )  ( X  ST )
Vender posición rollover en ~ ~ ~
S  S t e  ( br ) t  ST e  (b  r )T  ST e  (b  r )T
mercancía
Prestar X X Xe  rt Xe  rT Xe  rT
~
Ct  X [e rt  1] X [e rT  1] X [e rT  1] 
Valor neto de la cartera C  P S  X
 St [1  e  ( b  r ) t ]  ST [1  e  ( b  r )T ] ST [1  e  (b  r )T ]
Despejando se obtiene el lado izquierdo de la ecuación (10.24a). El lado izquierdo de la ecuación (10.24b)
puede establecerse utilizando operaciones de arbitraje y argumentos similares a aquellos en la Tabla 10.6,
excepto que la posición rollover en la mercancía comienza con una inversión de e( b  r )t unidades.
El lado derecho de (10.24a) se puede derivar considerando la cartera utilizada para probar la paridad put y
call europea. Cambiando la notación para reflejar el hecho que estamos discutiendo opciones Americanas e
introduciendo la columna “Valor Intermedio” para reflejar el prospecto de un ejercicio temprano, la tabla del
valor de cartera se convierte en la Tabla 10.7. Aquí, el valor final de la cartera es con seguridad igual a cero,
si las posiciones de la opción se mantienen abiertas hasta ese momento. Las posiciones de la opción son
compensadas por la posición de la mercancía y los préstamos libres de riesgo son compensados por los
precios de ejercicio de las opciones. En el caso que el tenedor de la opción americana decida ejercer su
opción antes del vencimiento, el tenedor de la cartera utiliza su posición larga en la mercancía para cubrir su
obligación de mercancía en la call ejercida y utiliza los ingresos del ejercicio para retirar su deuda pendiente.
Después que se realizan estas acciones, aún tiene una posición abierta larga en la put, el efectivo en la
cantidad de X [1  e  r (T  t ) ] , y una posición en la mercancía que vale S [e( b  r )(T  t )  1] . Dado que es seguro que
la cartera tendrá resultados no-negativos, el valor inicial debe ser no-positivo o
C ( S , T ; X )  P( S , T ; X )  Se( b  r )T  Xe rT  0 . (10.25)
Despejando se obtiene el lado derecho de la condición (10.24a) de la relación de paridad put y call
americana. El lado derecho de (10.24b) se establece considerando la cartera en la Tabla 10.7, con la
excepción que la posición rollover de la mercancía se inicia con una unidad en lugar de e( b  r )T .
La relación de paridad put y call para las opciones Americanas es un caso especial de (10.24b). Dado
que la tasa de acarreo, b , es igual a cero, la relación se convierte en
Fe  rT  X  C ( S , T ; X )  P( S , T ; X )  F  Xe rT (10.26)
La Tabla 10.8 contiene un resumen de las relaciones de paridad put y call desarrolladas en esta sección.
TABLA 10.7 Transacciones de arbitraje para establecer la paridad put y call Americanas,
cuando b  r. C ( S , T ; X )  P( S , T ; X )  Se( b  r )T  Xe  rT .
Call Ejercida Call Ejercida al
Temprano Vencimiento
Valor Inicial Valor Intermedian Valor Final
~ ~
Posición ST  X ST  X
~ ~
Vender call americana C  ( St  X )  ( ST  X )
~ ~
Comprar put americana P Pt X  ST 0
Comprar posición rollover ~ ~ ~
 Se  ( b  r )T
St e(b  r )(T  t ) ST ST
en mercancía
Pedir prestado Xe  rT Xe  rT  Xe  r (T  t ) X X
~  r (T t )
C  P  Se  ( br )T
Pt  X [1  e ]
Valor neto de la cartera 0 0
 Xe rT  S [e ( br )(T t )  1]
TABLA 10.8 Resumen de las relaciones de paridad de puts y calls para Commodities y Futuros.
Tipo de Opción Opciones sobre Subyacente
Europea c( S , T ; X )  p( S , T ; X )  Se ( br )T  Xe  rT
S  X  C ( S , T ; X )  P( S , T ; X )  Se( b  r )T  Xe  rT , if b  r
Americana
Se(b  r )T  X  C ( S , T ; X )  P( S , T ; X )  S  Xe  rT , if b  r
Opciones sobre Futuros del Subyacente
Europea c( F , T ; X )  p( F , T ; X )  e  rT ( F  X )
Americana Fe  rT  X  C ( F , T ; X )  P( F , T ; X )  F  Xe  rT
10.8 OPCIONES DE COMMODITY VERSUS OPCIONES DE FUTUROS

Hasta ahora en este capitulo hemos discutido los vínculos de arbitraje entre opciones y el subyacente como
aparece en las líneas horizontales en la Figura 10.1. En esta sección completamos la discusión al
concentrarnos en las relaciones de arbitraje que existen entre las opciones sobre el subyacente y las opciones
sobre futuros del subyacente.90
Opciones Europeas
La relación entre las opciones Europeas sobre el subyacente y las opciones Europeas sobre el futuro es muy
simple porque a término el subyacente y el futuro tienen el mismo precio. La call (put) sobre el futuro tendrá
exactamente el mismo valor que la call (put) europea sobre el subyacente. Así que
c( S , T ; X )  c( F , T ; X ) (1 0.27a)
y
p( S , T ; X )  p( F , T ; X ) . (10.27b)
En el caso de opciones Europeas las opciones sobre el subyacente y los futuros son sustitutos perfectos.
Opciones Americanas
La igualdad de las opciones Europeas se debe a que no pueden ser ejercidas antes del vencimiento y a que
entonces los valores del subyacente y los futuros son iguales. Para opciones Americanas, que si se pueden
ejercer antes, el precio de los futuros y el subyacente puede ser diferente. Cuando el precio de los futuros es
por lo menos el precio del subyacente (es decir F  SebT , F  S if b  0 ), la call americana escrita sobre los
futuros vale por lo menos lo que vale la call sobre el subyacente.
C ( F ,T ; X )  C (S , T ; X ) (10.28a)
Para ver esto considere los valores iniciales, intermedio y final de una cartera que consiste en una posición
larga en la call sobre futuros y una posición corta en la call sobre el subyacente como se ve en la tabla 10.9.
Si ambas opciones se mantienen hasta termino el valor neto de la cartera es cero. Si las opciones están out-
of-the-money terminan sin valor y si están in-the-money los flujos de caja se cancelan mutuamente. Si la
~ ~
opción americana se ejerce antes contra el tenedor de la cartera, el valor de la cartera es Ct  St  X . Pero el
~
límite inferior del precio de la call es Ft  X . Como hemos asumido que Ft  St , el valor intermedio de la
cartera es no-negativo. Para eliminar oportunidades de arbitraje el valor inicial de la cartera debe ser no-
positivo, así que la condición (10.28a) debe cumplirse.
Un argumento similar se pude desarrollar para las puts Americanas. Como las puts son derechos de venta del
subyacente el instrumento con el precio más bajo representa el mayor precio de las opciones. Por lo tanto
F S,
P( S , T ; X )  P( F , T ; X ) . (10.28b)
Las condiciones (10.28a) y (10.28b) son las relaciones de precios entre opciones Americanas sobre el
subyacente y los futuros del subyacente cuando el costo de acarreo, b , es positivo o igual a cero. In algunos
mercados b puede ser negativo. Por ejemplo en monedas b  0 cuando la tasa de interés foránea es mayor
que la doméstica. Cuando esto sucede se puede demostrar fácilmente que las condiciones (10.28a) y (10.28b)
serán invertidas. La tabla 10.10 nos da un resumen de las relaciones de precios desarrolladas en esta sección.
90
En el Capitulo 1 indicamos que estos contratos existen para muchas monedas como el Marco Alemán y para índices bursátiles
como el S&P 500.
TABLA 10.9 Operaciones de Arbitraje para establecer la relación entre los precios de las opciones
Americanas escritas sobre el subyacente y sobre los futuros cuando b  0 . C ( F , T ; X )  C ( S , T ; X )
Call Ejercida Temprano Call Ejercida a la Expiración
Valor Terminal
Posición Valor Inicial Valor Intermedio ~ ~
ST  X ST  X
Compre opción ~ ~
 C (F ,T ; X ) C ( Ft , T  t ; X ) 0 ST  X
de futuros
Venda opción
~ ~
sobre el C (S , T ; X )  ( St  X ) 0  ( ST  X )
subyacente
Valor Neto de ~ ~
C (S ,T ; X )  C (F ,T ; X ) C ( Ft , T  t ; X )  ( St  X ) 0 0
la Cartera
TABLA 10.10. Resumen de las relaciones de precios entre opciones sobre subyacente y futuros.
Tipo de Opción Call Put
Europea c ( S , T ; X )  c ( F , T ; X ) p ( S , T ; X )  p ( F , T ;X)
C ( S , T ; X )  C ( F , T ; X ), si b  0 P ( S , T ; X )  P( F , T ; X ), si b  0
Americana
C ( S , T ; X )  C ( F , T ; X ), si b  0 P ( S , T ; X )  P( F , T ; X ), si b  0
10.9 RESUMEN
Primero mostramos como posiciones de opciones se pueden representar usando notación de vectores. Luego
presentamos y explicamos los limites inferiores de los precios de las opciones.
La diferencia entre opciones Americanas y Europeas es el derecho de ejercer temprano que tienen las
opciones Americanas. Discutimos luego las condiciones bajo las cuales este derecho tiene valor. Primero
ejercer temprano tiene sentido si la opción esta substancialmente in-the-money. Segundo, para calls ejercer
temprano requiere que el costo de acarreo sea “pequeño” comparado con la tasa de interés; para puts ejercer
temprano requiere que el costo de acarreo sea “alto” comparado con la tasa de interés.
La relación de paridad de puts y calls se derivo para opciones Americanas y Europeas y para opciones
sobre le subyacente y sobre futuros del subyacente. Finalmente, se derivo la relación entre los precios de las
opciones sobre le subyacente y sobre futuros del subyacente.
11 VALUACIÓN DE OPCIONES
11.1 VALORIZACION NEUTRAL AL RIESGO

 Cartera libre de riesgo Utilizando un Modelo Binomial Simple
 Valuación Neutral al Riesgo Utilizando el Modelo Binomial
 Valuación Adversa al Riesgo Utilizando el Modelo Binomial
11.2 PRECIO DE LA MERCANCÍA Y LA DISTRIBUCION DE SU RETORNO
11.3 VALUACIÓN NEUTRAL AL RIESGO DE LA CALL EUROPEA

 Evaluación de X Pr( ST  X )
 Evaluación de E ( S | S  X )
T T
 Evaluación de E (cT )
 Valor Actual de la Call
11.4 VALUACIÓN NEUTRAL AL RIESGO DE UNA PUT EUROPEA
11.5 PROPIEDADES DEL PRECIO DE LA CALL Y PUT EUROPEA (GRIEGAS)

 Cambio en el Precio de la Mercancía
 Cambio Porcentual en el Precio de la Mercancía
 Cambio en el Delta
 Cambios en el Precio del Ejercicio
 Cambio en la Tasa del Costo de Acarreo
 Cambio en la Tasa de Interés
 Cambio en la Volatilidad
 Cambio en el Tiempo Hasta el Vencimiento
11.6 OPCIÓN EUROPEA DE INTERCAMBIO
11.7 VALUACIÓN DE OPCIONES AMERICANAS
11.8 ESTIMACIÓN DE LOS PARÁMETROS DE PRECIOS DE LAS OPCIONES

 Cálculo de la Volatilidad Histórica
 Cálculo de la Volatilidad Implícita
11.9 RESUMEN
11 VALUACIÓN DE OPCIONES
En el último capítulo examinamos los precios de las opciones implícitos en la ausencia de oportunidades de
arbitraje. El enfoque de ese capítulo proporcionó relaciones de precios muy interesantes, pero los resultados
tomaron la forma de límites en los precios de las opciones en lugar de las ecuaciones de valuación.
El enfoque utilizado aquí asume que las opciones son valorizadas como si todas las personas en la
economía fuesen neutrales al riesgo. Esta premisa es razonable porque el valor de la opción no depende de la
tasa de retorno esperada de la mercancía subyacente.
En la sección 1 se explica el concepto de valuación neutral del riesgo y su equivalencia a las
valuaciones adversas al riesgo. La sección 2 examina las implicancias de la premisa que los precios de las
mercancías están distribuidos lognormalmente. En la tercera sección las premisas de la lognormalidad y la
neutralidad ante el riesgo son utilizadas para poner precio a una call europea y en la cuarta sección a la put
europea. La sección 5 describe la sensibilidad de los precios de las opciones ante los cambios en los
determinantes subyacentes de las opciones. La sección 6 presenta la ecuación de valuación para una opción
que permite al tenedor intercambiar una mercancía riesgosa por otra. Esta opción, llamada opción de
intercambio, se introduce en muchos tipos de contratos de futuros. Los métodos de aproximaciones de las
valuaciones para las opciones americanas se discuten brevemente en la sección 7. La sección 8 describe
cómo se pueden calcular los parámetros de la ecuación de valuación, y la sección 9 concluye con un breve
resumen.
11.1 PRECIOS NEUTRALES AL RIESGO

El valor de una opción al vencimiento depende del valor de la mercancía subyacente. Antes del vencimiento,
uno puede calcular el valor esperado de una opción basándose en la distribución de la probabilidad del valor
final de la mercancía subyacente de la opción. En la Figura 11.1 se muestra una distribución de probabilidad,
y se muestra también el valor de ejercicio, X . El valor esperado de una call al vencimiento es la utilidad de
la call por su probabilidad, sumando todos los valores posibles de la mercancía subyacente. Dado que a la
izquierda de X el valor de la call es cero, el valor esperado de la call es la expectativa parcial a la derecha de
X . Para ilustrar este principio del precio, supongamos más bien una distribución de probabilidad pareja
como la que se muestra en la Figura 11.1, la mercancía subyacente sólo puede tomar los valores 80, 90, 100,
110 y 120, con las probabilidades correspondientes de 0.15, 0.20, 0.30, 0.20 y 0.15. También, supongamos
que el precio de ejercicio de la call es 100. El valor esperado de la call es, por lo tanto, (110-100)0.20 + (120-
100)0.15 = 5.0.
El valor actual de la call es el valor descontado de su valor esperado al vencimiento. Determinar este
valor actual es un problema que nos deja perplejos. Según el enfoque tradicional de las personas adversas al
riesgo en la economía, el valor actual de la call se calcula descontando el valor esperado de la call al
vencimiento a la tasa de retorno ajustada al riesgo de la Call. Este es el enfoque derivado por Samuelson en
1965. Desafortunadamente, su enfoque requiere el cálculo de al tasa de retorno esperada sobre la mercancía
y la tasa de retorno esperada sobre la call. En la práctica, el cálculo confiable de estos parámetros es
extremadamente difícil.
Durante muchos años, la valuación de las opciones no pudo superar esta dificultad. En 1973 se dio un
avance con un artículo escrito por Black y Scholes. Ellos mostraban que uno podía establecer una cartera
libre de riesgo entre una opciones de acciones y las acciones subyacentes, de manera que la opción es libre
de riesgo con relación a las acciones.91 Cox y Ross (1976) demostraron que podían encontrar el valor
correcto de la opción si asumían que el retorno esperado sobre las acciones y la tasa de retorno esperada
sobre la call eran la tasa libre de riesgo, siempre que se mantuviera la distribución de la probabilidad del
valor final de las acciones. Para ilustrar, consideren las dos distribuciones de probabilidades trazadas en la
Figura 11.2. La distribución de la probabilidad sobre la derecha es la distribución implícita si las personas
son adversas al riesgo, y la distribución de la probabilidad a la izquierda es la distribución implícita si las
91
Para un recuento histórico del desarrollo del modelo de precios de opciones Black-Scholes, ver Black (1989).
personas son neutrales al riesgo. Las variaciones de las dos distribuciones son las mismas, pero el valor
esperado de la distribución neutral al riesgo es menor que el valor esperado de la distribución adversa al
riesgo. Según en enfoque Cox-Ross, el valor esperado de la distribución neutral al riesgo se descuenta a la
tasa de interés libre de riesgo, y según el enfoque de Samuelson, el valor esperado de la distribución adversa
al riesgo se descuenta a la tasa de retorno ajustada al riesgo. Al final, ambos enfoques proporcionan el mismo
valor actual para la call.
Figura 11.1 Distribución del Precio de la Mercancía en el Momento T
Probabilidad
X ST
Figura 11.2 Distribuciones del Precio de la Mercancía para las Personas Neutrales al Riesgo y las Adversas
al Riesgo
Probabilidad
Neutral al Riesgo
Adversa al Riesgo
ST
En este capítulo, utilizamos el enfoque de valuación neutral al riesgo por su ductilidad matemática. Sin
embargo, antes de hacerlo, mediante una ilustración utilizando un modelo binomial simple demostraremos
que los dos enfoques producen el mismo resultado. Primero, demostraremos el concepto de cobertura libre
de riesgo. Segundo, mostramos la valuación neutral al riesgo. Finalmente, mostramos la equivalencia de la
valuación adversa al riesgo y la valuación neutral al riesgo.
Cartera libre de riesgo Utilizando un Modelo Binomial Simple

La perspectiva clave en la derivación de las formula de precios de opciones presentada en este capítulo es
que se puede formar una cartera libre de riesgo entre la opción y la mercancía subyacente. Para entender el
concepto de cobertura libre de riesgo, consideren el siguiente problema numérico simple. Supongamos que el
precio actual de la mercancía es $40 y que al final de tres meses el precio de la mercancía será $45 o $35. La
figura a continuación ilustra los posibles movimientos en el precio de la mercancía.
$45
$40
$35
Ahora, consideren una call europea escrita sobre esta mercancía. Esta call tiene un precio de ejercicio de $40
y vence en exactamente tres meses. Al vencimiento, esta call tendrá un valor de $5 o $0, dependiendo de si el
precio de la mercancía es $45 o $35, como vemos en la figura a continuación.
$5
$c
$0
Ahora, supongamos que vamos a comprar una unidad de la mercancía y vender nc calls . El valor final de
esta cartera es $45 - 5nc si el precio de la mercancía se eleva y $35 si el precio de la mercancía baja. La
incertidumbre del valor final de la cartera puede eliminarse completamente estableciendo nc de manera que
$45 - 5nc  35 or nc  2 .
En otras palabras, si compramos una unidad de la mercancía y vendemos dos calls, el valor final de la cartera
es con seguridad $35. Este es el concepto de una cartera libre de riesgo. Debido a la existencia de la cartera
libre de riesgo, podemos poner precio a la call europea en el ejemplo anterior. El costo de formar esta cartera
de cartera libre de riesgo en el momento 0 es $40-$2 c . Dado que la inversión de $40 - 2nc proporciona un
valor final seguro de $35, entonces si invertimos alternativamente los $40 - 2nc en bonos libres de riesgo
también obtendríamos un valor final de $35. Si la tasa de interés libre de riesgo durante el intervalo de tres
meses es 2 porciento, entonces la ausencia de oportunidades de arbitraje libre de costo en el mercado
requiere que $(40-2 c )(1.02) = $35.
En otras palabras, el precio de la call europea es $2.84.
El hecho que se pueda formar una cartera libre de riesgo entre la opción y la mercancía subyacente tiene una
implicancia importante –el precio de la call riesgosa se puede derivar sin conocer la tasa de retorno esperada
sobre la mercancía. Aún cuando en el ejemplo anterior no se conocen las probabilidades que el precio de la
mercancía suba a $45 o baje a $35, podemos poner precio a la opción. En otras palabras, el valor de la call
con relación a la mercancía no se ve influenciado por las preferencias del inversionista. No importa si una
persona es adversa al riesgo o neutral al riesgo, ambas están dispuestas a pagar $2.84 por la call en el
ejemplo anterior.
Valuación Neutral al Riesgo Utilizando el Modelo Binomial

Cox y Ross (1976) llevan este argumento un paso más adelante. Ellos reconocen que, dado que el precio de
la call no varía con las preferencias del inversionista, no se pierde nada si asumimos que todos son neutrales
al riesgo. Según una premisa de neutralidad ante el riesgo, podemos encontrar que las probabilidades
“neutrales al riesgo” de una subida o bajada en el ejemplo anterior. En un mundo neutral al riesgo, el valor
final esperado de la mercancía es simplemente su precio actual por uno más la tasa de interés libre de riesgo.
Por lo tanto,
$40(l.02)  $45 p  $35(l - p) ,
o p igual a 58 porciento. Luego utilizamos esta probabilidad para calcular el valor final esperado de la call,
esto es,
E (cT )  $5(0.58)  $0(0.42)  $2.90 .
Finalmente, el valor actual de la call es simplemente el valor actual del valor final esperado. Según la
premisa de neutralidad ante el riesgo, la tasa de descuento es la tasa de interés libre de riesgo, de manera que
el precio actual de la call es,
$2.90
c  $2.84 ,
1.02
Exactamente el resultado que obtuvimos utilizando la cartera de cobertura libre de riesgo. Es importante
recordar que este enfoque pone precio a la opción con relación al precio actual de la mercancía, el cual se
asume es “correcto”.
Valuación Adversa al Riesgo Utilizando el Modelo Binomial

El precio de la opción calculado utilizando el enfoque de valuación neutral al riesgo es el mismo que el
precio calculado utilizando una economía donde se asume que las personas son adversas al riesgo. Para ver
esto, consideren un marco referencial binomial donde el precio de la mercancía es actualmente $40 y tiene
probabilidades “adversas al riesgo”, p ' , de elevarse a $45 y 1  p ' de caer a $35. Supongamos que la tasa de
retorno esperada sobre la mercancía es 4 porciento durante los siguientes tres meses, y la tasa de interés libre
de riesgo es 2 porciento.
La diferencia entre las dos tasas refleja que la prima de riesgo es un 2 porciento exigido por las
personas por mantener la mercancía riesgosa. Si la tasa de retorno esperada sobre la mercancía es 4
porciento, entonces las probabilidades adversas al riesgo se determinan por
$40(l.04)  $45 p'  $35(l - p' ) ,
esto es, p ' es 66 porciento. La mayor probabilidad de un alza refleja el hecho que la persona adversa al
riesgo exige una mayor recompensa por asumir el riesgo que una persona neutral al riesgo. El precio
esperado de la opción al vencimiento es, por lo tanto,
E (cT )  $5(0.66)  $0(0.34)  $3.30 .
El siguiente paso en la valuación es determinar la tasa de descuento ajustada al riesgo apropiada para la call.
En una economía neutral al riesgo, la tasa es simplemente la tasa de interés libre de riesgo dado que las
personas son indiferentes ante el riesgo. Sin embargo, las personas adversas al riesgo se preocupan por el
riesgo y exigen tasas de retorno mayores por activos o mercancías más riesgosas. Por ejemplo, según el
modelo de precios de activos de capital (CAPM), la tasa de retorno esperada sobre la mercancía es
ES  r  ( EM  r )  S ,
Donde ES y EM son los retornos esperados de la mercancía y la cartera de mercado, respectivamente, r es la

tasa de retorno libre de riesgo y  S es el coeficiente de riesgo sistemático relativo de la mercancía.
Sustituyendo los valores del ejemplo, encontramos
0.04  0.02  ( EM - 0.02)  S o ( EM - 0.02)  S  0.02 .
Dado que  S representa el cambio porcentual en el precio de la mercancía con relación al cambio porcentual
en la cartera de mercado, podemos multiplicar  S por el cambio porcentual en el precio de la call con
respecto a un cambio porcentual en el precio de la mercancía para obtener el beta de la call y, por lo tanto, la
tasa de retorno esperada. Esto es,
 dc / c 
Ec  r  ( E M  r )  c , Ec  r  ( E M  r )  S  
 dS / S 
Pero, en el caso de nuestra ilustración, el cambio porcentual en el precio de la opción es

dc / c dc S 5  0 40 20
    
dS / S dS c 45  35 c c
Por lo tanto, sustituyendo ( EM  0.02)  S , encontramos Ec  0.02  0.02(20 / c)
E (ct ) 3.30
Por lo tanto, el valor actual de la call es c  
1  Ec 1  0.02  0.40 / c
3.30  0.40
de manera que el precio de la call es c   2.84 .
1.02
exactamente el mismo resultado obtenido para la economía neutral al riesgo.
11.2 PRECIO DE LA MERCANCÍA Y LA DISTRIBUCION DE SU RETORNO

La valuación de las calls europeas es casi tan simple como se muestra en la ilustración. La única
característica adicional que se debe incorporar en la valuación de una call europea es una premisa más
realista sobre la distribución del precio de la mercancía en el momento que la opción vence. Esta sección
trata el punto de las características de la distribución de los precios y los retornos de la mercancía.
Antes de discutir las propiedades específicas de la distribución, se requieren unas cuantas definiciones
básicas. Primero, consideren una secuencia de precio periódicos de una mercancía que comienza hoy y
continúa a lo largo del tiempo T ,92
S0 , S1 , S2 ,  , ST 1 , ST
La tasa de retorno aleatoria sobre la mercancía durante T períodos se define como el precio relativo menos
uno, esto es
~
ST / S 0  1
La tasa de retorno aleatoria y compuesta continuamente durante los T períodos es
~ ~
x  ln( ST / S0 )  1 (11.2)
o, alternativamente, el precio final aleatorio de la mercancía es

~ ~
ST  S 0 e x (11.3)
Noten que el retorno compuesto continuamente de T períodos es la suma de los retornos periódicos
compuestos continuamente, esto es
~ ~ ~
ln( ST / S0 )  t 1 ln(St / St 1 ) (11.4)
T
Una premisa que se utiliza comúnmente en el desarrollo de modelos financieros es que los retornos de los
activos son distribuidos independiente e idénticamente en cada período. Por lo tanto, el retorno periódico
esperado compuesto continuamente es
E[ln( St / St 1 )]   (11.5)
y, por la ecuación (11.4), el retorno esperado compuesto continuamente de 0 a T es
~ ~ ~
E[ln(ST / S 0 )]  t 1 E[ln(S t / St 1 )]  T
T
(11.6)
92
Estos precios de la mercancía son observados a intervalos iguales a lo largo del tiempo.
Similarmente, la varianza del retorno periódico compuesto continuamente es

~ ~
Var[ln(St / St 1 )]   2 (11.7)
de manera que la varianza del retorno compuesto continuamente de 0 a T es
~ ~ ~
Var[ln(ST / S0 )]  t 1 Var[ln(St / St 1 )]  2T
T
(11.8)
El primer y segundo término en (11.8) son iguales por la independencia entre los retornos en períodos
diferentes. La desviación estándar del retorno compuesto continuamente de 0 a T es  T . La segunda
premisa que invocamos es que las tasas de retorno periódicas compuestas continuamente están distribuidas
normalmente con una media  y una varianza  2 . En este caso, el retorno compuesto continuamente de 0 a
T también está distribuido normalmente con una media T y una varianza  2T . También implica que la
distribución de los precios de los activos es lognormal con una media
~
E ( ST )  S 0 eT , (11.9)
donde      2 / 2 (11.9a)
(Podemos ver una prueba de esto en el Apéndice 11.1)

2
e x / 2
Figura 11.3(a) Distribución Normal f ( x) 
2
Distribución Normal para Retornos
0.4
Pr
ob 0.3
abi
0.2
lid
ad 0.1
0
-5 -3 -1 1 3 5
Retornos del Subayacente
1  log( x / m) 2 
Figura 11.3(b) Distribución Lognormal: f ( x; m,  )  2
exp , x  0
2 L2
L
x L 2  
2 2 2
m  e  0.5 ,  L2  e 2   (e  1);  ,  los parametros de la distribución normal.
Distribución Lognormal: Precios
Pr 0.5
ob 0.4
abi 0.3
lid
0.2
ad
0.1
0
0 5 10 15 20 25 30 35
Precios del Subyacente
Las figuras 11.3a y 11.3b contienen ilustraciones de las dos distribuciones que implícitamente estamos
utilizando. La primera es la distribución normal para ~
x , que tiene una media T y una varianza  2T . La
~
segunda es una distribución lognormal de ST , que tiene una media de S 0 eT . Noten que la distribución del
precio tiene la característica intuitivamente atractiva que el precio final de la mercancía no puede caer por
debajo de cero. Si los precios finales se asumen como distribuidos normalmente, existiría una posibilidad que
el precio de la mercancía caiga por debajo de cero. Nuestro uso de la distribución normal se facilita más
transformando el retorno compuesto continuamente x o ln( ST / S0 ) , en una variable estándar distribuida
normalmente, ~ z , que tiene una media de cero y una varianza de uno, esto es
~ ~
~ x  T ln( ST / S 0 )  T
z  (11.10)
 T  T
que también se puede escribir en términos del precio final de la mercancía
~ ~
ST  S 0 e T  T z (11.11)
~
La variable z tiene la función de densidad
2
e z / 2
n( z )  (11.12)
2
La probabilidad que un resultado de esta distribución normal unitaria produzca un valor menor que la
constante, d , es
 z2 / 2
d e
~
Pr( z  d )  N (d )   dz (11.13)

2
Para evaluar la probabilidad N (d ) en (11.13), se puede utilizar una variedad de modelos. Las
aproximaciones polinómicas son populares porque son simples de programar. El Apéndice 11.2 contiene dos
de estas aproximaciones y sus niveles de exactitud. Otra alternativa es utilizar los valores de las
probabilidades normales tabuladas en los libros de texto de estadística y otras publicaciones. El Apéndice
11.3 contiene las probabilidades normales tabuladas en rango de d , desde –4.99 hasta +4.99.
Dos propiedades de la función de densidad normal unitaria serán útiles más adelante en este capítulo.
Primero, la probabilidad de obtener un valor mayor que d de una distribución normal unitaria es igual a uno
menos la probabilidad de obtener un valor menor que d ,93 esto es,
Pr( ~
z  d )  1  N (d ) . (11.14)
Segundo, dado que la distribución normal unitaria es simétrica alrededor de 0, la probabilidad de

obtener un valor menor que d es igual a uno menos la probabilidad de obtener un valor menor que  d ,94
esto es,
N (d )  1  N (d ) (11.15)
EJEMPLO 11.1
Calcule las probabilidades que el resultado de una distribución normal estándar proporcione un valor que
esté (a) dentro de una desviación estándar de la media, (b) dentro de dos desviaciones estándar de la media y
(c) dentro de tres desviaciones estándar de la media. Primero, se debe notar que cualquier variable distribuida
normalmente, ~ x , puede transformarse en una variable unitaria distribuida normalmente (es decir, una
variable con una media de cero y una varianza de uno) al aplicar la transformación (11.10). Segundo,
nosotros evaluamos las probabilidades utilizando los valores tabulados para la distribución normal unitaria
acumulada. (Ver Apéndice 11.3)
93
Este resultado se deduce simplemente de Prob (z  d)  Prob (z  d)  1 y (11.13).
2 2 2 2
d e z /2  e  z / 2  e  z / 2 d e z /2
   
94
Este resultado se deriva de la siguiente manera: N (d )  dz  dz  dz  dz  1  N (d )
 2 d 2  2  2
Pr(1  ~
z  1)  Pr( ~
z  1)  Pr( ~
z  1)  1  2 Pr( z  1) = 0.84134 - 0.15866 = 0.68268
Pr(2  ~
z  2)  Pr( ~
z  2)  Pr( ~
z  2)  1  2 Pr( z  2) = 0.97725 - 0.02275 = 0.95450
Pr(3  ~
z  3)  Pr( ~
z  3)  Pr( ~
z  3)  1  2 Pr( z  3) = 0.99865 - 0.00135 = 0.99730
EJEMPLO 11.2
Asuman que el precio actual de la mercancía es $50 y que la tasa de retorno compuesta continuamente tiene
una media anualizada de 16 porciento y una desviación estándar de 32 porciento. Calcular la probabilidad de
que el precio de la mercancía exceda 75 al final de tres meses.
Primero, utilizamos la ecuación (11.10) para transformar el precio final lognormal en un valor
variable normal unitario. Específicamente,
ln(75 / 50)  0.16(0.25)
d  2.28416
0.32 0.25
Segundo, redondeamos d al centésimo más cercano y utilizamos las tablas de probabilidades:
~
Pr( ST  75)  Pr( ~
z  2.28)  0.98870 .
Noten que estamos evaluando la probabilidad que el precio final de la mercancía sea menos que 75 porque
las tablas encuentran el área bajo la función de densidad normal unitaria menos infinito hasta el límite d .
Para calcular la probabilidad que el precio final de la mercancía sea mayor que 75, simplemente tomamos el
~ ~
complemento o Pr( ST  75)  1  Pr( ST  75)  1  Pr( ~
z  2.28)  1  0.98870  0.01130 .
Noten que estamos introduciendo cierto error como resultado de redondear el límite integral superior
d a la centésima más cercana cuando utilizamos las tablas. Podríamos interpolar entre los valores de las
tablas para lograr una mayor exactitud, o podríamos utilizar una de las aproximaciones polinómicas del
Apéndice 11.2. Utilizando la segunda aproximación polinómica del apéndice obtenemos
N (2.28416)  0.98882 .
EJEMPLO 11.3
Utilizando los parámetros del Ejemplo 11.2, calcular la probabilidad que el precio de la mercancía caiga
entre 40 y 60 al final de seis meses.
Una vez más, utilizamos la ecuación (11.10) para transformar el precio final de la mercancía a la
variable unitaria distribuida normalmente. Los límites de la integración son
ln(60 / 50)  0.16(0.5) ln(40 / 50)  0.16(0.5)

d1   0.45220 d2   1.33972
0.32 0.5 0.32 0.5
Las probabilidades son Pr(~z  0.45220)  N (0.45220)  0.67444 y

~
Pr( z  1.33972)  N (1.33972)  0.09017 .
una vez más las probabilidades fueron calculadas utilizando la segunda aproximación polinómica del
Apéndice 11.2. El paso final involucra diferenciar las probabilidades, esto es,
~
Pr(40  ST  60)  Pr(1.33972  ~
z  0.45220)  Pr(~
z  0.45220)  Pr(~z  1.33972)
 0.67444  0.09017  0.58427 =58.427%
EJEMPLO 11.4
Utilizando los parámetros del Ejemplo 11.2, calcular el rango del precio de la mercancía en tres meses
asumiendo que estará entre dos desviaciones estándar de su nivel actual.
Utilice la ecuación (11.11) y establezca z igual a  2 . Los dos precios finales de las mercancías son
ST 1  50e 0.16 ( 0.25)0.32 0.25 ( 2 )
 37.78919
ST 2  50e 0.16 ( 0.25)0.32 0.25 ( 2 )
 71.66647 .
EJEMPLO 11.5
Supongamos que existe una call europea a tres meses sobre la mercancía en el Ejemplo 11.2 y su precio de
ejercicio es 50. Calcular la probabilidad que la call esté in-the-money al vencimiento. El límite integral
superior d es
ln(50 / 50)  0.16(0.5)
d1   0.25000
0.32 0.25
La probabilidad que el precio final de la mercancía sea menor que el precio de ejercicio es
Pr( ST  50)  Pr( zT  0.25000)  N (0.25000)  0.40129
de manera que la probabilidad que el precio de la mercancía exceda el precio de ejercicio es

~
Pr( ST  50)  1  N (0.25000)  0.59871 =59.871%.
EJEMPLO 11.6
Calcular la tasa de retorno esperada sobre la mercancía durante un intervalo de tres meses y el precio
esperado de la mercancía en ese momento en el tiempo. Por la ecuación (11.9a), sabemos que la tasa de
retorno esperada sobre la mercancía es igual a la media más la mitad de la varianza de la distribución del
~
logaritmo del ratio del precio de la mercancía, ln(ST / S ) , esto es,      2 / 2  0.16  0.32 2 / 2  0.2112
~
Por lo tanto, el precio final esperado de la mercancía es, E ( ST )  S 0 eT  50e 02112(0.25)  52.71094
11.3 VALUACIÓN NEUTRAL AL RIESGO DE LA CALL EUROPEA

Ahora se deriva la ecuación de valuación de la call europea según las premisas de distribución discutidas en
la sección anterior. El enfoque de valuación es consistente con la ilustración numérica utilizada en la Sección
11.1 –primero, calculamos el valor final esperado de la Call, y luego descontamos al presente –a la tasa de
riesgo cero– el valor final esperado. El precio teórico de la call es simplemente
c  e  rT E (c~T ) (11.16)
Para evaluar el precio final esperado de la Call, asumimos que la tasa de retorno esperada sobre la mercancía
es igual a la tasa de interés libre de riesgo (neutralidad ante el riesgo) y los precios de las mercancías están
distribuidos de manera lognormal al vencimiento de la opción. Para descontar el precio final esperado de la
call al presente, asumimos que la tasa de retorno esperada sobre la call es igual a la tasa de interés libre de
riesgo (neutralidad ante el riesgo).
Para poder hacer que la ecuación (11.16) sea operacional, necesitamos evaluar el término E (c~T ) , el
~
valor final esperado de la call. Si se asume que ST está distribuido lognormalmente, la distribución del
precio final de la Call, c~T , es conocido dado que c~T es simplemente
~ ~
~ ST  X for ST  X
cT   ~ (11.17)
0 for ST  X .
Con el precio final de la mercancía con una distribución lognormal, la condición (11.17) implica que el
precio final de la call tiene una distribución lognormal truncada y que la media de la distribución del precio
de la call es:
~ ~
E (c~T )  E ( ST  X | ST  X )  E (0 | ST  X )  E ( ST  X | ST  X )
~ ~
 E ( ST | ST  X )  E ( X | ST  X )  E ( ST | ST  X )  X Pr( ST  X )
Donde Pr( ST  X ) es la probabilidad que al vencimiento, el precio de la mercancía exceda el precio de

ejercicio de la opción. Para evaluar E (c~T ) , debemos evaluar cada uno de los dos términos del lado derecho
de (11.18). Comenzaremos por el segundo término.
Evaluación de X Pr( ST  X )
Sea L( ST ) la función de densidad lognormal de ST , el término X Pr( ST  X ) , es

X Pr( ST  X )   L( ST )dST
X
La manera más fácil de evaluar el integral es realizar un cambio de variables en ST . La ecuación (11.10) nos
muestra la transformación que aplicamos a ST . El límite inferior y superior de la integración para la nueva
variable z se obtienen sustituyendo  y X por ST en (11.10). Por lo tanto, los límites son  y
[ln( X / S )  T ] /  T , respectivamente. Por lo tanto,
[ln( X / S ) T ]

X Pr( ST  X )  [ln( X / S )T ] n( z )dz    T n( z )dz  XN (d 2 ) (11.19)

 T
donde d 2  [ln( X / S )  T ] /  T . En otras palabras, el valor N (d 2 ) es la probabilidad que el precio de la

mercancía exceda el precio de ejercicio al vencimiento de la opción.
~
Evaluación de E ( ST | ST  X )
~
La evaluación del primer término de la ecuación (11.18), E ( ST | ST  X ) , es ligeramente más difícil. Los
pasos iniciales son los siguientes:
2
~   Se T  T z e  z / 2
E ( ST | ST  X )   ST L( ST )dST  [ln( X / S )T ] dz
X
 T 2
2
T / 2 2T / 2 T z  z 2 / 2 T  z )2 / 2
 Se T  2T / 2   Se  (
 
T
 Se [ln( X / S ) T ] dz  Se [ln( X / S ) T ] dz
 T 2  T 2
[ln( X / S ) T ] 2
Se  ( y ) / 2

T  2T / 2 2
 Se  T dy  Se T  T / 2 N (d1 )

2
donde d1   T  [ln( X / S )  T ] /  T  d 2   T . Los pasos en (11.20) son los siguientes: (a) el valor
esperado condicional se expresa en forma de integral donde L( ST ) es la función de densidad lognormal para
ST ; (b) se realiza un cambio de variables en ST , y se elimina la función de densidad de la variable normal
estandarizada, z ; (c) Se T se factoriza fuera de la integral y se completa la raíz en el exponente dentro del
2
integral; (d) e T / 2 se factoriza fuera del integral y se simplifica la expresión restante en el exponente dentro
del integral; (e) se realiza un cambio de variables y   T  z y se redefinen los límites de integración;95 y
(f) la expresión se simplifica.
Evaluación de E (c~T )
Para resumir, estamos tratando de valorizar una call europea según la premisa que los precios de la
mercancía están distribuidos lognormalmente y que las personas son neutrales al riesgo. Estamos en el
proceso de valorizar el valor final esperado de la call, E (c~T ) . Sustituyendo las ecuaciones (11.20) y (11.19)
en la ecuación (11.18), ahora tenemos
E (c~T )  Se T  T N (d1 )  XN (d 2 )
2
(11.21)
donde
d 2  d1   T (11.21a)
ln( S / X )  T
d1  (11.21b)
 T
Sin embargo, no nos detendremos aquí. La tasa de retorno esperada de la mercancía en los límites del
~ ~
integral d1 y d 2 es la media de un logaritmo de los ratios del precio de la mercancía – E[ln(St / St 1 )] .
~
Quisiéramos expresar la tasa de retorno de la mercancía en términos de los precios relativos – E ( ST / S ) .
Sabemos que
~ 2
E ( ST / S )  eT  e (   / 2)T (11.22)
Ahora, recuerden que hemos invocado una premisa de neutralidad ante el riesgo. El valor de  en (11.22) es
la tasa de retorno esperada sobre la mercancía y, en un mundo neutral al riesgo, la tasa de retorno esperada
sobre la mercancía es igual a la tasa del costo de acarreo, b (es decir, el costo del interés más cualquier costo
adicional de almacenamiento). Sustituyendo b por  en (11.9a) y aislando  , obtenemos
  b   2 / 2 . (11.23)
Substituyendo esto en (11.21), obtenemos
E (c~T )  SebT N (d1 )  XN (d 2 ) , (11.24)

donde
ln( S / X )  (b  0.5 2 )T
d1  (11.24a)
 T
ln(S / X )  (b  0.5 2 )T
d2  d 2  d1   T (11.24b)
 T
Valor Actual de la Call

Teniendo E (c~T ) a la mano, podemos sustituir en la ecuación (11.16) para encontrar la fórmula de valuación
para la call europea sobre una mercancía con una tasa de costo de acarreo b , esto es,
c( S , T ; X )  Se( br )T N (d1 )  Xe  rT N (d 2 ) , (11.25)
95
Donde y = -z, se mantiene la siguiente propiedad:
La propiedad se utiliza al simplificar (11.20).
ln( S / X )  (b  0.5 2 )T
donde d1  (11.25a)
 T
d 2  d1   T . (11.25b)
La interpretación de los términos en (11.25) es bastante directa dado nuestro enfoque de valuación neutral al
riesgo. El término Se ( br )T N (d1 ) es el valor actual del beneficio esperado al ejercer la call al vencimiento,
condicional a que el precio final de la mercancía sea mayor que el precio de ejercicio al vencimiento de la
opción. El término N (d 2 ) es la probabilidad que el precio de la mercancía sea mayor que el precio de
ejercicio al vencimiento. El valor actual del costo esperado de ejercer la condicional de la call siempre y
cuando la call esté in-the-money al vencimiento es Xe  rT N (d 2 ) .
EJEMPLO 11.7
Calcular el precio de una call europea de moneda extranjera a tres meses con un precio de ejercicio de 40. El
tipo de cambio spot actualmente es 40, la tasa de interés doméstica es 8 porciento anual, la tasa de interés
extranjera es 12 porciento anual, y la desviación estándar del retorno compuesto continuamente es 30
porciento anual. Noten que la tasa de costo de acarreo, b , por lo tanto es –4 porciento.
c  40e ( 0.040.08) 0.25 N (d1 )  40e 0.08( 0.25) N (d 2 ) , donde

ln(40 / 40)  [0.04  0.5(0.30) 2 ](0.25)
d1   0.0083 ; d 2  d1  0.30 0.25  0.1417 .
0.30 0.25
Los valores de N ( d1 ) y N ( d 2 ) son 0.5033 y 0.4437, respectivamente, de manera que el precio de la call
europea es c  38.818(0.5033) - 39.208(0.4437)  2.14 .
11.4 VALUACIÓN NEUTRAL AL RIESGO DE UNA PUT EUROPEA

El enfoque de valuación neutral al riesgo se puede aplicar al problema de poner precio a una put europea y
encontrar la ecuación de valuación de la put. Sin embargo, una manera más sencilla es combinar la relación
de paridad put-call europea del capítulo anterior con la ecuación de valuación de la call europea (11.25). En
ausencia de oportunidades de arbitraje, la relación de paridad put-call europea,
c( S , T ; X )  p( S , T ; X )  Se (br )T  Xe  rT (11.26)
se mantiene todo el tiempo. Aislando p( S , T ; X ) en (11.26), obtenemos

p( S , T ; X )  c( S , T ; X )  Se (br )T  Xe  rT (11.27)
Sustituyendo la ecuación de la call europea (11.25) por el término c( S , T ; X ) , encontramos que la ecuación
de la valuación de la put europea sobre una mercancía es
p( S , T ; X )  Se ( br )T N (d1 )  Xe  rT N (d 2 )  Se( br )T  Xe  rT
p( S , T ; X )  Xe  rT [1  N (d 2 )]  Se ( br )T [1  N (d1 )]
p( S , T ; X )  Xe  rT N (d 2 )  Se ( br )T N (d1 ) , (11.28)
donde
ln( S / X )  (b  0.5 2 )T
d1  (11.28a)
 T
d 2  d1   T (11.28b)
Por lo tanto, la valuación de la put europea se deduce directamente de la paridad put-call europea y la
valuación de la call europea.
La interpretación de los términos en (11.28) es igual a la interpretación neutral al riesgo para la call.
El término Xe  rT N (d 2 ) es el valor actual del beneficio esperado del precio de ejercicio de la put al
vencimiento, condicionado a que el precio final de la mercancía sea menor que el precio de ejercicio al
vencimiento de la opción. Recuerden que la put proporciona el derecho a vender la mercancía de manera que
el beneficio de mantener la opción es el efectivo que nosotros recibimos cuando ejercemos la opción, esto es,
X . N ( d 2 ) es la probabilidad que el precio de la mercancía sea menor que el precio de ejercicio al
vencimiento. Noten que es el complemento de N (d 2 ) , la probabilidad que el precio final de la mercancía
exceda el precio de ejercicio. El valor actual del costo esperado de ejercer la put condicionado a que el
vencimiento la put esté in-the-money es Se ( br )T N (d1 ) . Si ejercemos la put, debemos entregar el subyacente
como cumplimiento de nuestra obligación de manera que el valor actual del precio final esperado de la
mercancía condicional al ejercicio sea nuestro costo hoy.
EJEMPLO 11.8
Calcular el precio de una put europea de moneda extranjera a tres meses con un precio de ejercicio de 40. El
tipo de cambio spot actual es 40, la tasa de interés doméstica es 8 porciento anual, la tasa de interés
extranjera es 12 porciento anual, y la desviación estándar del retorno compuesto continuamente es 30
porciento sobre una base anualizada. Noten que la tasa de costo de acarreo, b , es por lo tanto, -4 porciento.
p  40e 0.08( 0.25) N (d 2 )  40e ( 0.040.08)( 0.25) N (d1 ) , donde
ln(40 / 40)  [0.04  0.5(0.30) 2 ](0.25)

d1   0.0083 ; d 2  d1  0.30 0.25  0.1417
0.30 0.25
Los valores de N (d1 ) y N (d 2 ) son 0.5563 y 0.4967, respectivamente, de manera que el precio de la put
europea es
p  39.208(0.5563) - 38.818(0.4967) = 2.53 .
Noten que este resultado, junto con el resultado del ejercicio 11.7 verifica la relación de paridad put-call
(11.26), esto es,
2.14 – 2.53 = 38.818 – 39.208 = -0.39
11.5 PROPIEDADES DEL PRECIO DE LA CALL Y LA PUT EUROPEA (GRIEGAS)

Las ecuaciones de valuación para la call y put europea son
c( S , T ; X )  Se ( br )T N (d1 )  Xe  rT N (d 2 ) (11.25)
p( S , T ; X )  Xe  rT N (d 2 )  Se ( br )T N (d1 ) (11.28)
ln( S / X )  (b  0.5 2 )T
respectivamente, donde d1  y d 2  d1   T
 T
El precio de la opción depende de seis variables – S , X , b, r ,  , T . En esta sección, analizamos cómo los
precios de la call y put europea responden a los cambios en las variables que definen la opción. Se discutirá
cada una de las variables, comenzando por el precio de la mercancía. Las derivaciones de cada una de las
expresiones a continuación están contenidas en el Apéndice 11.4.
Cambio en el Precio de la Mercancía: DELTA

El cambio en el precio de la opción con respecto a un cambio en el precio de la mercancía se conoce como el
delta de la opción. El delta de la call europea es
c
c   e ( br )T N (d1 )  0 (11.29a)
S
El delta de la call es positivo, lo que implica que un incremento en el precio de la mercancía causa que se
incremente el precio de la Call. El resultado es intuitivo dado que la call conlleva el derecho a comprar la
mercancía subyacente a un precio fijo y la mercancía subyacente recién se ha vuelto más valiosa. La Figura
11.4 muestra cómo cambia el valor de la call europea conforme cambia el precio de la mercancía subyacente.
La opción tiene tres meses hasta el vencimiento. Noten que cuando la call está out-of-the-money, su
pendiente es casi horizontal. Las calls out-of-the-money tienen de valores delta muy bajos; esto es, no
responden rápidamente a los cambios en el precio de la mercancía. Conforme el precio de la mercancía se
incrementa y la call se vuelve at-the-money y luego in-the-money, la pendiente se hace cada vez más
empinada. Cuando la opción está muy in-the-money, el valor delta es casi uno, y el precio de la call cambia
en una proporción uno-a-uno con el precio de la mercancía. La Figura 11.5 muestra el valor delta de la
opción como una función del precio de la mercancía. El delta de la put es
p
p   e (br )T N (d1 )  0 (11.29b)
S
Esta derivada es negativa porque un incremento en el precio de la mercancía hace que la put sea menos
valiosa. Una vez más, se puede demostrar que la sensibilidad del precio de la put ante los cambios en el
precio de la mercancía subyacente es en sí mismo sensible al “moneyness” de la opción. La Figura 11.6
muestra esta sensibilidad para una put europea con tres meses hasta el vencimiento. La opción in-the-money
tiene una pendiente más empinada que la opción at-the-money, la cual a su vez, tiene una pendiente más
empinada que la opción out-of-the-money. La Figura 11.7 muestra el valor delta de la put como una función
del precio de la mercancía subyacente.
Figura 11.4 Precio de una call europea ( c )como una función del precio de la mercancía subyacente ( S ). El
rango del precio de la mercancía es de 50 a 150. La opción tiene un precio de ejercicio ( X ) de 100 y un
tiempo hasta el vencimiento ( T ) de tres meses. La tasa de costo de acarreo (b) es 8 porciento y la tasa de
interés libre de riesgo ( r ) es 8 porciento. La desviación estándar del logaritmo de los ratios del precio de la
mercancía (  ) es 30 porciento.
Call and Lower Bound
60
50
40
30
20
10
Commodity P rice s (S )
FIGURA 11.5 El delta (  c ) de la call europea como una función del precio de la mercancía subyacente ( S ).
El rango del precio de la mercancía está entre 50 y 150. La opción tiene un precio de ejercicio ( X ) de 100 y
un tiempo al vencimiento ( T ) de tres meses. La tasa del costo de acarreo ( b ) es 8 porciento y la tasa de
Call Delta
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
50 70 90 110 130 150
Commodity P rice (S )
Cambio Porcentual en el Precio de la Mercancía: ETA

Con frecuencia se da el caso que en lugar de interesarse por el cambio en dólares en el precio de la opción
con respecto a un cambio en dólares en el precio de la mercancía, uno se interesa en la elasticidad del precio
de la opción con respecto al precio de la mercancía. Esta elasticidad, llamada el eta de la opción, es el
cambio porcentual en el precio de la opción con respecto al cambio porcentual en el precio de la mercancía.
La elasticidad del precio de la call con respecto al precio de la mercancía es mayor que uno,96 esto es
96
Se puede demostrar que la elasticidad del precio de la call con respecto al precio de la mercancía es mayor que uno al re-escribir
(11.30a) como c  1
 rT
Xe N ( d 2 )
1
Se( b  r )T N ( d1 )
El último término en el denominador es menor que uno porque el precio de la call europeo no puede ser menos que cero, por lo
tanto, el valor de  c , debe ser mayor que uno.
S S
c  e ( br )T N (d1 )  1 (11.30a)
c c
y la elasticidad del precio de la put con respecto al precio de la mercancía es menos que uno,97 esto es
S S
p  e ( br )T N (d1 )  1 (11.30b)
p p
FIGURA 11.6 El precio de la put europea ( p ) como una función del precio de la mercancía subyacente ( S ).
Put and Lower Bound
50
45
40
35
30
25
20
15
10
5
0
Commodity Price (S)
Put y Límite Inferior; 2) Precio de la Mercancía ( S )

FIGURA 11.7 El delta (  p ) de la put europea como una función del precio de la mercancía subyacente ( S ).
PUT DELTA
0
-0.1
-0.2
-0.3
-0.4
-0.5
-0.6
-0.7
-0.8
-0.9
-1
97
La evidencia que  p  1 se deduce según las mismas líneas que la evidencia que c  1 en la nota a pie de página anterior.
La Tabla 11.1 contiene los precios de las opciones, los valores delta y las elasticidades para precios
alternativos de la mercancía subyacente. Es interesante notar que (i) las elasticidades tiene magnitudes
grandes y (ii) las magnitudes son mayores para las opciones más out-of-the-money. Si alguien está casi
seguro que el precio de una mercancía individual se va a elevar, una inversión en una call proporcionará una
tasa de retorno mayor que una inversión directamente en la mercancía y, más aún, una inversión en una call
out-of-the-money proporcionará una mayor tasa de retorno que una call in-the-money.
Tabla 11.1 Simulación de Precios de Calls y Puts de Índice de Acciones, deltas, etas y gammas de
los precios de las opciones. Parámetros: X = 100, b = 0.08, r = 0.08, T = 0.25, y  = 0.30.
Precio de
Precio de Delta Eta Gamma Precio de Delta Eta Gamma
la
la Call c c c la Put p p p
Mercancía p
c
S
80 0.537 0.100 14.952 0.014 18.557 -0.899 -3.878 0.014
90 2.494 0.310 11.207 0.026 10.514 -0.689 -5.900 0.026
100 6.961 0.582 8.367 0.026 4.981 -0.417 -8.380 0.026
110 13.954 0.800 6.310 0.016 1.974 -0.199 -11.109 0.016
120 22.645 0.922 4.889 0.008 0.665 -0.077 -13.924 0.008
Sin embargo, estas tasas de retornos mayores no vienen sin un correspondiente incremente en el riesgo. De
hecho, así como las tasas de retorno sobre las opciones están proporcionalmente relacionadas con la tasa de
retorno sobre la mercancía, el riesgo o “beta” de una opción está proporcionalmente relacionado con el beta
de la mercancía, esto es,  c  c  S y  p   p  S . El incremento en la tasa de retorno esperada como
resultado de mantener una call es exactamente lo que se justifica sobre la base del modelo de precio del
activo de capital.
Cambio en el Delta: GAMMA

Anteriormente describimos el delta de la opción, la manera en que el precio de la opción cambia conforme
cambia el precio de la mercancía. El gamma de la opción está relacionado con este concepto –el cambio en
delta conforme cambia el precio de la mercancía. La expresión para el gamma de una call es
 c e (br )T n(d1 )
c    0 (11.31a)
S S T
el gamma para una put es
 p e (br )T n(d1 )
p     c  0 (11.31b)
S S T
donde n(d1 ) es la densidad en d1 . En resumen, este valor nos dice cuán rápidamente cambia delta conforme
cambia el precio de la mercancía. Dado que el gamma de una opción es mayor cuando las opciones están
aproximadamente at-the-money, estas opciones son las más difíciles de proteger. Además, si usted piensa
que el precio de la mercancía está por moverse en una u otra dirección (recuerden que la motivación para
colocar un margen de volatilidad), el margen at-the-money maximizará la respuesta en dólares de la cartera a
los movimientos del precio de la mercancía subyacente. La Figura 11.8 muestra el gamma de la opción como
una función del precio de la mercancía subyacente.
Cambios en el Precio del Ejercicio

Las derivadas parciales de los precios de la call y put con respecto al precio de ejercicio de la opción son
c
 e rT N (d 2 )  0 (11.32a)
X
y
p
 e rT N (d 2 )  0 , (11.32b)
X
respectivamente. Noten que si el precio de ejercicio de las opciones se incrementa el valor de la call
disminuye y el valor de la put se incrementa. Esto se deduce del hecho que la call se pondría más out-of-the-
money y la put más in-the-money.
Las derivadas parciales de los precios de la opción, con respecto al precio de ejercicio, son de muy
poco valor práctico en el sentido que una vez creada la opción, el precio de ejercicio no cambia. Aquí sólo
son expresadas para completar la idea.
FIGURA 11.8 El gamma (  c   p ) de la opción europea como una función del precio de la mercancía
subyacente ( S ). El rango del precio de la mercancía está entre 50 y 150. La opción tiene un precio de
ejercicio ( X ) de 100 y un tiempo al vencimiento ( T ) de tres meses. La tasa del costo de acarreo ( b ) es 8
porciento y la tasa de interés libre de riesgo ( r ) es 8 porciento. La desviación estándar del logaritmo de los
ratios del precio de la mercancía (  ) es 30 porciento.
Call and Put O ptions Gamma
0.03
0.025
0.02
0.015
0.01
0.005
0
50 70 90 110 130 150
Cambio en la Tasa del Costo de Acarreo

El cambio en el precio de la opción con respecto al cambio en la tasa de acarreo es
c
 TSe (br )T N (d1 )  0 . (11.33a)
b
Conforme se incrementa la tasa del costo de acarreo, el valor de la call se incrementa, manteniendo constante
el precio spot y otras variables. Mientras mayor sea el costo de acarrear la mercancía subyacente, mayor será
la tasa de apreciación en el precio de la mercancía y por lo tanto mayor será el valor de la call. Sin embargo,
la magnitud de la derivada es pequeña. Para la call de moneda extranjera valuada en el Ejemplo 11.7, la
derivada parcial con respecto a la tasa del costo de acarreo es 4.884. En otras palabras, si la tasa del costo de
acarreo sobre la mercancía subyacente se incrementa en 100 puntos base, el precio de la call se incrementará
en aproximadamente cinco centavos. La derivada parcial del precio de la put con respecto a la tasa del costo
de acarreo es
p
 TSe ( br )T N (d1 )  0 . (11.33b)
b
Conforme se incrementa la tasa del costo de acarreo, la tasa de apreciación esperada en el precio de la
mercancía se incrementa y, por lo tanto, disminuye el valor de la put. El valor numérico de esta derivada
parcial para la put en el Ejemplo 11.8 es –4.8200.
Cambio en la Tasa de Interés

El cambio en el precio de la call con respecto al cambio en la tasa de interés libre de riesgo es
c
 TXe rT N (d 2 )  0 . (11.34a)
r
El precio de la call se incrementa con un incremento en la tasa de interés porque el valor actual del precio de
ejercicio disminuye. El valor de esta derivada es 4.3489 para la call en el Ejemplo 11.7. La derivada parcial
del precio de la put con respecto a la tasa de interés libre de riesgo es
p
 TXe rT N (d 2 )  0 . (11.34b)
r
Aquí el signo es negativo porque, conforme se incrementa la tasa de interés libre de riesgo, disminuye el
valor actual del precio de ejercicio recibido al ejercer la opción. El valor de la derivada para la put en el
Ejemplo 11.8 es –5.4531, lo que implica que un incremento de 100 puntos base en la tasa de interés reduce el
valor de la opción en casi cinco centavos.
Cambio en la Volatilidad: VEGA

Al cambio en el precio de la opción con respecto al cambio en la volatilidad 98 del retorno de la mercancía
subyacente se le conoce como el vega de la opción. El vega de la call europea es
c
Vegac   Se ( br )T n(d1 ) T  0 . (11.35a)

El signo de la derivada es positivo, indicando que conforme se incrementa la volatilidad del retorno de la
mercancía subyacente, se incrementa el valor de la call. La intuición para este resultado es que un incremento
en la tasa de volatilidad incrementa la probabilidad de grandes movimientos ascendentes en el precio de la
mercancía subyacente. La probabilidad de grandes movimientos descendentes en el precio de la mercancía se
incrementa; sin embargo, no es de consecuencia ya que el tenedor de la opción tiene responsabilidad
limitada.
El valor numérico del vega de la call implica que el precio de las opciones es más sensible a la
volatilidad que a la tasa del costo de acarreo o la tasa de interés. La opción en el Ejemplo 11.7 tiene un vega
de 7.7428. Un incremento en la volatilidad de 100 puntos incrementa el precio de la call en casi ocho
centavos.
El vega de la put es el mismo que el de la Call, esto es
p
Vega p   Se (br )T n(d1 ) T  Vegac  0 . (11.35b)

98
Hasta este punto, hemos utilizado el término “desviación estándar” para describir la dispersión de los retornos de la mercancía,
 . En la industria, a  normalmente se le conoce como la volatilidad o la tasa de volatilidad de los retornos de la mercancía
subyacente, y nosotros adoptamos esa idea en el resto del capítulo.
El valor de la put también se incrementa con un incremento en la volatilidad dado que se incrementa la
probabilidad de una gran disminución en el precio de la mercancía. El valor numérico del vega para la put en
el Ejemplo 11.8, por lo tanto, también es 7.7428.
La Figura 11.9 muestra el vega de la opción como una función del precio de la mercancía. Noten que
el vega tiene su valor más alto cuando la opción está aproximadamente at-the-money.
Cambio en el Tiempo Hasta el Vencimiento: THETA

La derivada parcial del precio de la opción con respecto al parámetro del tiempo hasta el vencimiento se
llama el theta de la opción. El theta de la call es
c 
c   Se ( br )T n(d1 )  (b  r ) Se ( br )T N (d1 )  rXe rT N (d 2 )  0 (11.36a)
T 2 T
La expresión muestra que la sensibilidad del precio de la call ante cambios en el tiempo hasta el vencimiento
de la opción es la suma de tres componentes. El primer término al lado derecho es positivo y refleja el
incremento en el precio de la opción debido al hecho que un incremento en el tiempo hasta el vencimiento
incrementa la probabilidad de movimiento ascendentes en el precio de la mercancía y, por lo tanto,
incrementa el valor de la opción. El segundo término puede ser positivo o negativo dependiendo de si la tasa
de costo de acarreo, b , es mayor o menor que la tasa de interés, r . Si b  r , el término es positivo dado que
conforme se incrementa el tiempo hasta el vencimiento, el valor actual del precio final esperado de la
mercancía aumenta (recuerden que el precio de la mercancía subyacente aumenta a la tasa b mientras que la
tasa de descuento del valor final de la opción es r ). Finalmente, el tercer término es positivo. Conforme se
incrementa el tiempo hasta el vencimiento, el valor actual del precio de ejercicio se hace más pequeño. Noten
que el único caso donde el valor global de theta es ambiguamente positivo es cuando b  r .
Para la call en el Ejemplo 11.7, b  r , de manera que sabemos que el theta no necesita ser positivo.
El valor de theta es, no obstante, positivo en 3.6927. En otras palabras, el precio de la opción se incrementa
conforme se incrementa el tiempo hasta el vencimiento. Para ver el origen de este resultado, examinamos los
valores de cada uno de los tres términos en la derivada: 4.6457, -2.3446 y 1.3916. El mayor componente del
theta de la call en esta ilustración, 4.6457, viene de una mayor probabilidad de grandes movimientos en el
precio. Dado que b  r , el segundo término es negativo. Conforme aumenta el tiempo hasta el vencimiento,
el valor de la call cae debido a que se espera que el precio de la mercancía se incremente a una tasa menor
que la tasa de interés libre de riesgo. El valor de este componente es –2.3466. Finalmente, el valor del tercer
término es 1.3916, indicando que el valor de la call se incrementa porque el valor actual del precio del
ejercicio se reduce conforme aumenta el tiempo hasta el vencimiento.
Theta proporciona información sobre el deterioro del valor de la opción conforme pasa el tiempo
hasta el vencimiento. El theta de la call en el Ejemplo 11.7 es 3.6927. Esto implica que la disminución de
tiempo en esta opción es 3.6927 x 1/365 o ligeramente algo más que un centavo por encima del día siguiente
y 3.6927 x 7/365 o casi siete centavos por encima de la próxima semana, manteniendo los otros factores
constantes.
El theta de la put europea es
p 
c   Se( br )T n(d1 )  (b  r ) Se ( br )T N (d1 )  rXe rT N (d 2 )  0 . (11.36b)
T 2 T
Existe un paralelo entre la interpretación de los términos en la expresión del theta de la put y aquellos de la
call. El primer término es el incremento en el valor de la put resultante de la perspectiva de grandes
movimientos en el precio de la mercancía cuando el tiempo hasta el vencimiento es más largo. El segundo
término es negativo si b  r . En el caso de la put, el valor de la opción se incrementa cuando la tasa del costo
de acarreo está por debajo de la tasa de interés. El tercer término es positivo. Refleja el hecho que un
incremento en el tiempo hasta el vencimiento demora la recepción del precio de ejercicio y por lo tanto
reduce el valor de la put. El valor del theta de la put en el Ejercicio 11.8 es 5.2143, siendo los componentes
individuales de la suma 4.6457, 2.3136 y –1.7450.
Figura 11.9 El vega de la opción europea como una función del precio de la mercancía subyacente ( S ). El
rango del precio de la mercancía está entre 50 y 150. La opción tiene un precio de ejercicio ( X ) de 100 y un
tiempo al vencimiento ( T ) de tres meses. La tasa del costo de acarreo ( b ) es 8 porciento y la tasa de interés
libre de riesgo ( r ) es 8 porciento. La desviación estándar del logaritmo de los ratios del precio de la
Call and Put O ptions V E GA
35
30
25
20
15
10
5
0
50 70 90 110 130 150
Commodity Price (S)
11.6 OPCIÓN EUROPEA DE INTERCAMBIO

Estrechamente relacionadas con las opciones europeas de mercancías con un precio de ejercicio fijo están las
opciones europeas que le otorgan al tenedor el derecho de intercambiar una mercancía por otra. Estas
opciones son algo común, aunque usualmente son introducidas en algún tipo de contrato. Por ejemplo, en el
Capítulo 8 discutimos el hecho que el contrato de futuros de bonos del Tesoro permite la entrega de un
número cualquiera de emisiones elegibles del bono-T y que la posición corta realizará la entrega más barata.
En esta instancia, la posición corta tiene una opción de intercambio que le permite intercambiar el bono–T
que actualmente posee por una emisión más barata, si es que se pone a disposición una emisión más barata.
Muchos contratos de futuros agrícolas también cuentan con esta opción de intercambio u opción de calidad
dentro del diseño del contrato.
La derivación de la fórmula de opción de intercambio puede seguir el mismo enfoque de valuación
neutral al riesgo que utilizamos anteriormente en el capítulo, de manera que aquí no repetiremos en enfoque.
La ecuación de valuación de una opción de intercambio europea que permite al tenedor intercambiar la
mercancía 2 por la mercancía 1, esto es, “comprar” la mercancía 1 con la mercancía 2, es
c( S1 , T ; S 2 )  S1e ( b1 r )T N (d1 )  S 2 e ( b2 r )T N (d 2 ) (11.37)

donde
ln(S1 / S 2 )  (b1  b2  0.5 2 )T
d1  (11.37a)
 T
d 2  d1   T (11.37b)
   1   2  2 1, 2 1 2
2 2 2
(11.37c)
Las variables con subíndice 1 se aplican a la mercancía 1 y las variables con subíndice 2 se aplican a la
mercancía 2. El significado de cada variable se describió anteriormente en este capítulo.
Una observación importante con relación a (11.37) es que fórmula de call descrita anteriormente en
este capítulo es simplemente un caso de esta ecuación de valuación. Supongamos que permitimos que la
mercancía 2 sea un activo libre de riesgo. Por lo tanto, el precio actual de la mercancía 2, S 2 , es Xe rT , la
tasa del costo de acarreo, b2 , la tasa de interés libre de riesgo, r y la desviación estándar del retorno,  2 , es
igual a cero. Con estas sustituciones, la ecuación (11.37) se convierte en la fórmula de la call europea
(11.25).
Otra observación importante con relación a (11.39) es que el valor de una call para “comprar” la
mercancía 1 con la mercancía 2, c( S1 , T ; S 2 ) , es igual al valor de una put para “vender” la mercancía 2 por la
mercancía 1, p ( S 2 , T ; S1 ) . En el caso de una call, la opción se ejerce al vencimiento si los ingresos
S1,T  S 2,T  0 , esto es, si el precio final para la mercancía 1 excede el precio final de la mercancía 2; de otra
manera, no se ejerce. En el caso de un put, la opción se ejerce al vencimiento si los ingresos S1,T  S 2,T  0 ;
esto es, si la mercancía 2 es más barata que la mercancía 1; de otra manera, no se ejerce. Pero las estructuras
de estos dos problemas de valuación son idénticas, entonces
c( S1 , T ; S 2 )  p ( S 2 , T ; S1 ) . (11.38)
Regresando a la especificación del contrato de futuros de bonos-T, recuerden que al final del Capítulo 8
dijimos que el precio de los futuros es igual al precio del bono a entregarse más barato menos el valor de la
opción de calidad. Si el contrato de futuros de bonos-T tiene sólo dos emisiones de bonos-T elegibles para
entregarse, la fórmula de valuación (11.38) puede utilizarse para valorizar la opción de calidad. Con más
emisiones elegibles, el modelo se debe generalizar.99
EJEMPLO 11.9
Supongamos que existen dos bonos elegibles para entregarse sobre un contrato de futuros de bonos-T. El
tiempo al vencimiento de los futuros es de tres meses. Actualmente el Bono 1 es el más barato de entregar.
Su precio es de 99 y su cupón es 6 porciento. El Bono 2 tiene un precio de 102 y su cupón es 9 porciento. La
desviación estándar del retorno compuesto continuamente es 15 porciento para el Bono 1 y 12 porciento para
el Bono 2. La correlación entre sus tasas de retorno es 0.9. La tasa de interés libre de riesgo es 7 porciento.
Calcular el valor de la opción de intercambio. Por simplicidad, asuman que ambos bonos tiene factores de
conversión iguales a uno y que el interés del cupón es pagado continuamente a lo largo de la vida restante del
contrato de futuros
p( B2 , T ; B1 )  99e ( 0.070.060.07 ) 0.25 N (d1 )  102e ( 0.070.090.07 ) 0.25 N (d 2 )
ln(99 / 102)  (0.07  0.06  0.07  0.09  0.5 2 )0.25

donde: d1  , d 2  d1   0.25
 0.25
y  2   12   22  2 1, 2 1 2  0.152  0.12 2  2(0.9)(0.15)(0.12)  0.0045
Sustituyendo   0.0671 en las expresiones d1 y d 2 , y luego en los valores d1  0.6495 y d 2  0.6830 en

la fórmula de la opción se obtiene que el valor de la opción de intercambio es 0.50.
99
La fórmula de la opción de intercambio para el caso de dos activos donde ambos activos tiene una tasa de costo de acarreo igual
a la tasa de interés libre de riesgo fue derivada por Margrabe (1978). La fórmula aquí presentada generaliza el resultado Margrabe
para permitir que los activos tengan diferentes tasas de acarreo. La opción de intercambio de n  activos fue desarrollada más
tarde por Margrabe (1982).
11.7 VALUACIÓN DE OPCIONES AMERICANAS

En el Capítulo 10 vimos que la opción americana vale por lo menos tanto como la opción europea por el
hecho que la opción americana puede ejercerse antes del vencimiento. Por lo tanto, el valor de la call y put
americana se puede escribir como
C ( S , T ; X )  c( S , T ; X )   C ( S , T ; X ) , (10.9a)
P( S , T ; X )  p( S , T ; X )   P ( S , T ; X ) . (1 0.9b)
El valor del privilegio de ejercicio antes del vencimiento,  C ( S , T ; X ) , depende de la relación entre la tasa de
costo de acarreo, b , y la tasa libre de riesgo, r . En el caso de una call, el privilegio de ejercicio temprano o
antes del vencimiento tiene valor sólo si b  r . En este caso, el costo de acarreo de la mercancía subyacente
es menor que el costo de los fondos atados en la mercancía. Como resultado, puede ser beneficioso ejercer la
call y tomar posesión de la mercancía porque las ganancias sobre la mercancía exceden el costo de los
fondos atados en la mercancía. Por ejemplo, puede ser deseable ejercer la call sobre moneda extranjera para
obtener un interés sobre la moneda extranjera si la tasa de la moneda extranjera excede la tasa
estadounidense. Si b  r , el ejercicio temprano de una call no es óptimo porque existe el costo de mantener
la mercancía.
Al continuar manteniendo la call, también se pueden lograr todas las ganancias potenciales que se
podrían obtener por mantener la mercancía , y se evita el costo de mantener la mercancía. Por ejemplo, nunca
es óptimo ejercer antes del vencimiento una opción sobre una acción que no paga dividendos (es decir,
b  r ).
En el caso de la put, el ejercicio temprano siempre es una posibilidad. Intuitivamente, el ejercicio
temprano es deseable si la utilidad de la put es lo suficientemente grande como para que el interés que se
pudiera obtener al invertir la utilidad exceda la posibilidad de una utilidad aún mayor si se continúa
manteniendo el put.
Se desconocen las soluciones analíticas explícitas para el precio de las opciones americanas. Si la
opción americana no se ejerce temprano, se mantienen las fórmulas para las opciones europeas. Pero si es
deseable un ejercicio temprano, el valor de la opción americana excede el valor europea por una cantidad
(con frecuencia bastante pequeña) que sólo puede ser aproximada por técnicas numéricas. En la práctica
comúnmente se aplican dos técnicas de aproximación que son el método del binomio del Cox, Ross y
Rubinstein (1979), y el método de aproximación cuadrática de Barone-Adesi y Whaley (1987). El método
del binomio se utiliza en el Capítulo 13 para valorizar una put sobre una acción que paga dividendos y en el
Capítulo 14 se utiliza la aproximación cuadrática para valorizar opciones sobre el índice de acciones y
futuros del índice de acciones.
11.8 ESTIMACIÓN DE LOS PARÁMETROS DE PRECIOS DE LAS OPCIONES

Los modelos de precio de las opciones europeas (11.25) y (11.28) y los métodos de aproximación de
opciones americanas que se discutirán en los Capítulo 13 y 14 son, en general, muy fáciles de usar. El precio
de ejercicio, X , y el tiempo hasta el vencimiento, T , son términos del contrato de la opción. El precio de la
mercancía, S , es la tasa de interés libre de riesgo, r , y la tasa de costo de acarreo, b , son valores
determinados por el mercado y de fácil acceso.100 El parámetro más difícil de calcular (y, por lo tanto, el
100
Normalmente se utilizan las tasas de los pagarés del Tesoro como un sustituto de la tasa de interés libre de riesgo. Recuerden
que del pie de página 2 del Capítulo 8 que la tasa de interés efectiva anual compuesta continuamente, r , se obtiene
r  ln(100 / Bd ) / T . Donde Bd es el precio del pagaré-T (por ejemplo, 100 menos el promedio del descuento de oferta y demanda
del pagaré-T) y T es el tiempo hasta el vencimiento del pagaré-T expresado en años.
parámetro sobre el cual más comúnmente están en desacuerdo los inversionistas) es la desviación estándar de
la tasa de retorno de la mercancía subyacente. En general, se utilizan dos métodos –el cálculo de la
volatilidad histórica y el cálculo de la volatilidad implícita.
Cálculo de la Volatilidad Histórica

Anteriormente en este capítulo, nosotros asumimos que la media y la desviación estándar de las tasas de
retornos compuestas continuamente, Rt  ln( St / St 1 ) , son constantes a lo largo del tiempo. El parámetro de
volatilidad en la fórmula de precio de las opciones es la tasa de volatilidad futura de la mercancía. Sin
embargo, si el parámetro de volatilidad es estacionario a lo largo del tiempo podemos utilizar retornos
anteriores para calcular la volatilidad histórica y luego utilizar la volatilidad histórica como un cálculo de la
volatilidad futura. El factor de cálculo más comúnmente utilizado es la varianza de la tasa de retorno sobre la
mercancía,  h2
1

T
ˆ h2  ( Rt  ˆ ) 2 (11.39)
T 1 t 1
donde T es el número de observaciones del retorno de las series de tiempo en el cálculo,101 Rt es la tasa de
retorno compuesta continuamente sobre la mercancía en el mes t [es decir, ln(St / St 1 ) ], y ̂ es el cálculo de
la tasa de retorno media,
1 T
̂  t 1 Rt (11.40)
T
Se puede obtener un estimado de la desviación estándar del retorno sobre la mercancía sacando la raíz
cuadrada de la varianza estimada, esto es, ˆ h  ˆ h2 .
Las tasas de retorno utilizadas en las ecuaciones (11.39) y (11.40) pueden durar cualquier período –
un día, una semana o un mes. En general, es mejor mientras más corta sea la distancia entre las
observaciones de precios ya que se utiliza más información para el cálculo. De manera que los retornos
semanales son efectivamente mejores que los retornos mensuales para el cálculo de la volatilidad,
manteniendo constante la duración global del período de cálculo.
Siguiendo la misma lógica, parecería que los retornos diarios son una mejor fuente de información
que los retornos semanales. Sin embargo, generalmente este no es el caso. Por ejemplo, los retornos de las
acciones demuestran estacionalidad por día de la semana.102 En general, los retornos de las acciones de los
viernes son significativamente mayores que el promedio, y los de los lunes son significativamente menores
que el promedio. Otras mercancías también tienen estacionalidad por día de la semana, pero la estacionalidad
tiene una estructura diferente. Es más, independiente de la mercancía subyacente, utilizando datos diarios, el
investigador decide cómo se deben tomar en cuenta los retornos del fin de semana. ¿La tasa de retorno del
cierre del viernes al cierre del lunes debe tomarse en cuenta como una tasa de retorno de tres días (días
calendario) o una tasa de retorno de 1 día (días comerciales)? Dadas las anomalías empíricas asociadas con
los retornos diarios y dado que el problema del manejo de los retornos del fin de semana no se han resuelto
satisfactoriamente, probablemente los retornos semanales son los mejores cuando se trata de calcular la
volatilidad histórica de los retornos de las mercancías. Otro punto que surge cuando se utiliza un factor
histórico se relaciona con cuántas observaciones de retornos utilizar en el cálculo de la volatilidad. Por un
lado, mientras más información se utilice en el proceso de cálculo, más preciso es el cálculo. Por otro lado,
mientras más largo sea el período sobre el cual se calcula la volatilidad, mayor será la probabilidad que se
viole la premisa estacionaria, en cuyo caso, la fórmula (11.39) ya no es un cálculo no parcializado de la
101
Noten que se necesitan T + 1 observaciones de precio pata generar las tasas de retorno T .
102
Ver, por ejemplo, French (1980) y Gibbons y Hess (1981).
varianza de la tasa de retorno de la mercancía. En la ausencia de la información que indica que se ha violado
la premisa estacionaria, veintiséis semanas de observaciones de los retornos probablemente son suficientes
para asegurar un cálculo razonablemente exacto de la volatilidad.
El cálculo de la volatilidad realizado utilizando la ecuación (11.39) calcula la varianza de la tasa de
retorno para el intervalo de tiempo entre las observaciones de precios utilizadas para calcular las tasas de
retorno. Por lo tanto, si se utilizan retornos semanales, el cálculo de la varianza de (11.39) es la varianza de
la tasa de retorno sobre una semana. Para anualizar este valor, tenemos que multiplicar la varianza por el
número de semanas en el año, esto es,  ha 2
 52 hw
2
, donde los subíndice w y a denotan la semana y el año,
respectivamente. Por lo tanto, la transformación para la anualización de la desviación estándar semanal es
 ha  52 hw .103
Vale la pena notar un último punto con relación al cálculo de la volatilidad histórica. Los factores de
cálculo arriba mostrados generalmente utilizar información de precio cierre-a-cierre al generar las tasas de
retorno. Esto ha sido aceptado como una práctica común ya que tradicionalmente la historia registrada de los
precios de la mercancía son precios reportados para la última transacción del día. Con el surgimiento de la
tecnología de cómputo y base de datos sofisticada, ahora se ha vuelto más fácil registrar y mantener
conjuntos de información más grandes, con la mayoría de las bolsas de valores y mercancías manteniendo
archivos de los precios de las transacciones. Esta información más refinada permite un cálculo más preciso
de la volatilidad. Por ejemplo, Parkinson (1980) y Garman y Klass (1980) desarrollaron factores de cálculo
alternativos para la varianza que utilizan los precios alto, bajo y de cierre de la mercancía y muestran que
estos factores de cálculo son ocho veces “mejores” que el factor de cálculo tradicional (11.39).
Cálculo de la Volatilidad Implícita

Un procedimiento de cálculo de la volatilidad implícita surge del modelo de precio de las opciones. Dado
que todos los parámetros del modelo de precio de las opciones, excepto  , son conocidos o se pueden
calcular con poca incertidumbre, uno sólo necesita igualar el precio de mercado observado de la opción con
su valor de fórmula, esto es
V j  Vˆ j ( j ) (11.41)
donde V es el precio observado de la opción, V es el precio modelo de la opción y hay que despejar  . Se
puede derivar una expresión analítica para el parámetro de variación; sin embargo, la aproximación exacta es
posible mediante “prueba-y-error”, casi de la misma manera en que se despeja el rendimiento al vencimiento
de un bono. Las volatilidades calculadas de esta manera son llamadas “volatilidades implícitas” o
“desviaciones estándar implícitas”. Pueden ser interpretadas como la volatilidad consenso del mercado en el
sentido que se utiliza el precio de mercado de la opción para imputar el cálculo de volatilidad. Si alguien
considera todas las opciones escritas sobre una mercancía dada, parecería razonable pensar que todas
rendirán el mismo estimado de volatilidad sobre la mercancía subyacente. Sin embargo, este no es el caso.
Existe una variedad de razones que causan que los estimados sean diferentes.
103
Para transformar el cálculo de volatilidad estimado utilizando retornos diarios a una volatilidad anual, el cálculo diario
usualmente se multiplica por la raíz cuadrada del número de días comerciales en el año (típicamente, 253 ), en lugar que el
número de días calendario en el año ( 365 ). El motivo para este ajuste es que los estudios de los retornos diarios de las acciones
indican que la volatilidad del retorno del cierre del viernes al cierre del lunes (tres días) es casi igual que la volatilidad de cierre-a-
cierre durante cualquier otro par de días comerciales adyacentes (un día). Ver, por ejemplo, Stoll y Whaley (1990a). Por lo tanto, el
tratar los fines de semana como un solo día de negociación proporciona el ajuste más apropiado para las volatilidades de los
retornos diarios de las acciones. Sin embargo, la evidencia empírica relacionada con la volatilidad de fin de semana de los
mercados son acciones es escasa, de manera que la generalidad de este resultado para otras mercancías todavía es desconocida.
Para los mercados sin acciones, un procedimiento más seguro podría ser utilizar los retornos semanales, como vimos anteriormente
en esta sección.
No-Simultaneidad de los Precios. El cálculo de volatilidad implícita utilizando (11.41) asume que el
precio de la opción y el precio de la mercancía son observados en el mismo instante en el tiempo. Con
frecuencia, se da el caso que la única información disponible es el precio de la opción y la mercancía en
momentos en que fueron negociados por última vez. Es poco probable que estas negociaciones, una en el
mercado de opciones y una en el mercado de mercancías, ocurrieran en el mismo instante, hasta el punto en
que no son contemporáneas, habría un error en el cálculo de la volatilidad.
Precios de Oferta y Demanda. Inclusive si las observaciones en el precio de la opción y la mercancía
utilizadas en (11.41) son simultáneas, existe un problema sobre qué representan los precios. Si los mercados
son perfectamente líquidos y libres de fricción, las negociaciones son liquidadas al precio de equilibrio del
título valor. Sin embargo, ninguno de los descriptores es verdadero. Los elaboradores de mercado
proporcionan liquidez manteniéndose listos para inmediatamente compra o vender título valores. Dado que
los elaboradores de mercados tienen capital (tanto de inversión como humano) ocupado en sus operaciones,
ellos exigen una tasa de retorno sobre su capital, la cual extraen estableciendo un precio de oferta del título
valor por debajo del precio solicitado. Por lo tanto, cuando se ejecutan las órdenes del mercado están en el
precio ofertado o solicitado, dependiendo de si la persona que ingresa al mercado quiere comprar o vender.
Dado que no hay manera de discernir la motivación del negociador que estuvo involucrado en la última
transacción observada, los cálculos de la volatilidad implícita tienen error cuando el precio de oferta de la
opción es igual al precio solicitado por la mercancía y viceversa.
Especificación Errada del Modelo.

El utilizar (11.41) para calcular la volatilidad también está sujeto a una mala especificación del modelo. La
técnica asume que el modelo de precio de la opción utilizado para Vˆ j ( j ) está correctamente especificado.
Si no, entonces obviamente va a existir error en el cálculo de la desviación estándar de la tasa de retorno
sobre la mercancía. Por ejemplo, las premisas del modelo que podrían ser violadas son la premisa de
lognormalidad de los precios de las acciones o la independencia de los retornos.
Para mitigar los problemas asociados con utilizar un solo cálculo de la volatilidad implícita para
representar la volatilidad de la mercancía subyacente, las volatilidades implícitas para diversas opciones
sobre la misma mercancía son promediadas para formar un cálculo global. La naturaleza de los esquemas de
promedio varía, de manera que es mejor comenzar con un enunciado general de la volatilidad implícita
promedio, esto es,
ˆ   j 1 jˆ j /  j 1 j

n n
(11.42)
donde  j es el ponderado aplicado al estimado j de la volatilidad y n es el número de opciones para las

cuales se obtuvo los cálculos de volatilidad.
Los esquemas de ponderación utilizados en al literatura han sido muchos y variados. Schmalensee y Trippi
(1978) y Patell y Wolfson (1979), por ejemplo, utilizan igual promedio ponderado,  j  1 / n, j  1, , n .
Su motivación para hacerlo es que cada cálculo de volatilidad es igualmente valiosos en la
determinación de la volatilidad global para la mercancía. Latane y Rendleman (1976), por otro lado,
ponderan de acuerdo a la derivada parcial del precio call con respecto a la desviación estándar del retorno de
la mercancía, esto es, V j / ˆ j , j  1,, n . Al hacerlo, los cálculos de la desviación estándar de las opciones
que son teóricamente más sensibles al valor de  son sopesados con mayor peso que aquellos que no lo son.
Chiras y Manaster (1978) siguen una lógica similar al utilizar la elasticidad del precio de la call con respecto
a la desviación estándar, V jˆ j / ˆ jV j , j  1,, n . Desafortunadamente este esquema tiene serias
deficiencias. Utilizar la elasticidad como la base del esquema de ponderación implica que los cálculos de la
volatilidad para las opciones out-of-the-money reciben el peso más alto. Las opciones out-of-the-money
generalmente no producen estimados de volatilidad muy exactos porque los mercados para estas opciones
son relativamente ilíquidos (induciendo a problemas serios de no-simultaneidad) y las opciones mismas
tienen márgenes altos de oferta-demanda (induciendo a errores oferta-demanda). Finalmente, Whaley (1982)
utiliza una regresión no lineal para calcular un valor de  utilizando simultáneamente toda la información de
precios de opciones disponible, esto es
V j  Vˆ j ( )   j (11.43)
Las propiedades del estimador de probabilidad máxima de (11.43) son, tal vez, las alternativas que mejor se
entienden.
Sin embargo, sin importar el esquema de ponderación, parece existir un fuerte respaldo empírico a
favor de la medida de la volatilidad implícita. Latane y Rendleman y Chiras y Manaster correlacionan las
medidas históricas e implícitas sobre la desviación estándar real del retorno y concluyen que el cálculo
implícito es un pronosticador notablemente superior. Aparentemente el mercado utiliza más información que
un simple estimado histórico al evaluar la volatilidad esperada de la mercancía.
11.9 RESUMEN
En este capítulo se han derivado en detalle las ecuaciones de precio de las opciones europeas sobre diferentes
tipos de mercancías subyacentes. El capítulo comienza con una discusión intuitiva del enfoque de valuación
neutral al riesgo utilizado para derivar las fórmulas de precio. Luego se describieron las distribuciones de
retorno y precio asumidas. En la sección 3, la valuación neutral al riesgo de la opción europea es llevado a
cabo en detalle y se muestran varias ecuaciones de valuación básica para diversos tipos de mercancías
subyacentes. Luego utilizando la paridad put-call, se derivaron las ecuaciones de valuación.
El precio de una opción sobre una mercancía depende del precio spot de la mercancía, el precio de
ejercicio de la opción, el costo de acarreo de la mercancía, la tasa libre de riesgo, la desviación estándar del
retorno de la mercancía y el tiempo hasta el vencimiento. En la sección 5, se analizó el efecto de los cambios
en cada una de estas variables sobre el precio de la opción.
La sección 6 presenta la ecuación de valuación que permite al tenedor intercambiar una mercancía
riesgosa por otra. Esta opción, llamada opción de intercambio, es introducida en muchos tipos de contratos
de futuros y se introdujo en capítulos anteriores. Se utiliza como ejemplo la valuación de una opción de
entrega introducida en un contrato de futuros de bonos del Tesoro (bono-T).
Los factores subyacentes al valor de una opción americana son los mismos subyacentes a una opción
europea excepto que la opción americana tiene el beneficio adicional del ejercicio antes del vencimiento. La
sección 7 nombra dos enfoques populares para la valuación de opciones americanas. Los métodos se
describen en detalle en los Capítulos 13 y 14.
En la práctica, la variable más importante que afecta el precio de una opción es la volatilidad de la
mercancía subyacente. La sección 8 explica los dos enfoques para calcular la volatilidad histórica y la
volatilidad implícita.
APPENDIX 11.1
~
Proof that E ( ST )  S 0 eT ,      2 / 2 where  and  are the mean and the variance of the normally
distributed continuously compounded rate of return.
Begin by rewriting the expected terminal price as the expected price relative,
~ ~
E ( ST / S 0 )  eT  E (e x ) , (Al.1)
where ~x is the normally distributed, continuously compounded rate of return from 0 through T . ~
x can be re-
expressed in terms of  ,  , and the unit normally distributed variable z . Using (11.10),
~ ~ T ~z
ST / S 0  e x  e T 
Substituting this result into (Al.1),
T~ T ~z
eT  E (e T  z
)  e T E (e ) (Al.2)
T ~z
The term E (e ) in (Al.2) may be simplified as follows:
T z  z2 / 2 T zz2 / 2 2
T / 2 2T / 2 T z  z 2 / 2
 T ~z  e e  e  e
E (e ) dz   dz   dz

2 
2 
2
2
T / 2 T z  z 2 / 2 T  z )2 / 2
 2T / 2   e   2T / 2   e  (
 
2
e dz  e dz  e T /2
(Al.3)

2 
2
Substituting (Al.3) into (Al.2), taking the logarithm of both sides, and then factoring T gives
    2 / 2 (Al.4)
APPENDIX 11.2 APPROXIMATIONS FOR THE CUMULATIVE NORMAL DENSITY

FUNCTION
The probability that a drawing from a unit normal distribution will produce a value less than the constant d
is
 z2 / 2
d e
~
Pr( z  d )  N (d )   dz

2
Below are two polynomials that provide reasonably accurate approximations for the above integral.
Approximation 1
2
N ( d )  a0 e  d /2
(a1t  a2t 2  a3t 3 )
where: t  1/(l  0.33267d ); a0  0.3989423; a1  0.4361836; a2  - 0.1201676; a3  0.9372980
With this approximation method, the value of d must be greater than or equal to 0. The maximum absolute
error of this approximation method is 0.00001.
Approximation 2
2
N ( d )  a0 e  d /2
(a1t  a2t 2  a3t 3  a4t 4  a5t 5 )
where: t  1/(l  0.2316419 d ); a 0  0.3989423; a1  0.31938153 0; a 2  - 0.35656378 2

a 3  1.78147793 7; a 4  - 1.82125597 8; a 5  1.33027442 9
With this approximation method, the value of d must be greater than or equal to 0. The maximum absolute
error of this approximation method is 0.000000075.
APPENDIX 11.3 CUMULATIVE NORMAL PROBABILITY TABLES
The probability that a drawing from a unit normal distribution will produce a value less than the constant d is
 z2 / 2
d e
~
Pr( z  d )  N (d )   dz

2
Range of d : -4.99  d  -2.50
d -0.00 -0.01 -0.02 -0.03 -0.04 -0.05 -0.06 -0.07 -0-08 -0.09
-4.90 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000
-4.80 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000
-4.70 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000
-4.60 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000
-4.50 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000
-4.40 0.00001 0.00001 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000 0.00000
-4.30 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001
-4.20 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001 0.00001
-4.10 0.00002 0.00002 0.00002 0.00002 0.00002 0.00002 0.00002 0.00002 0.00001 0.00001
-4.00 0.00003 0.00003 0.00003 0.00003 0.00003 0.00003 0.00002 0.00002 0.00002 0.00002
-3.90 0.00005 0.00005 0.00004 0.00004 0.00004 0.00004 0.00004 0.00004 0.00003 0.00003
-3.80 0.00007 0.00007 0.00007 0.00006 0.00006 0.00006 0.00006 0.00005 0.00005 0.00005
-3.70 0.00011 0.00010 0.00010 0.00010 0.00009 0.00009 0.00008 0.00008 0.00008 0.00008
-3.60 0.00016 0,00015 0.00015 0.00014 0.00014 0.00013 0.00013 0.00012 0.00012 0.00011
-3.50 0.00023 0.00022 0.00022 0.00021 0.00020 0.00019 0.00019 0.00018 0.00017 0.00017
-3.40 0.00034 0.00032 0.00031 0.00030 0.00029 0.00028 0.00027 0.00026 0,00025 0.00024
-3.30 0.00048 0.00047 0.00045 0.00043 0.00042 0.00040 0.00039 0.00038 0.00036 0.00035
-3.20 0.00069 0.00066 0.00064 0.00062 0.00060 0.00058 0.00056 0.00054 0.00052 0.00050
-3.10 0.00097 0.00094 0.00090 0.00087 0.00084 0.00082 0.00079 0.00076 0.00074 0.00071
-3.00 0.00135 0.00131 0.00126 0.00122 0.00118 0.00114 0.00111 0.00107 0.00104 0.00100
-2.90 0.00187 0.00181 0.00175 0.00169 0.00164 0.00159 0.00154 0.00149 0.00144 0.00139
-2.80 0.00256 0.00248 0.00240 0.00233 0.00226 0.00219 0.00212 0.00205 0.00199 0.00193
-2.70 0.00347 0.00336 0.00326 0.00317 0.00307 0.00298 0.00289 0.00280 0.00272 0.00264
-2.60 0.00466 0.00453 0.00440 0.00427 0.00415 0.00402 0.00391 0.00379 0.00368 0.00357
-2.50 0.00621 0.00604 0.00587 0.00570 0.00554 0.00539 0.00523 0.00508 0.00494 0.00480
CUMULATIVE NORMAL PROBABILITY TABLES
 z2 / 2
d e
~
Pr( z  d )  N (d )   dz

2
Range of d : -2.49  d  0.00
d -0.00 -0.01 -0.02 -0-03 -0.04 -0.05 -0.06 -0.07 -0-08 -0.09
-2.40 0.00820 0.00798 0.00776 0.00755 0.00734 0.00714 0.00695 0.00676 0.00657 0.00639
-2.30 0.01072 0.01044 0.01017 0.00990 0.00964 0.00939 0.00914 0.00889 0.00866 0.00842
-2.20 0.01390 0.01355 0.01321 0.01287 0.01255 0.01222 0.01191 0.01160 0.01130 0.01101
-2.10 0.01786 0.01743 0.01700 0.01659 0.01618 0.01578 0.01539 0.01500 0.01463 0.01426
-2.00 0.02275 0.02222 0.02169 0.02118 0.02068 0.02018 0.01970 0.01923 0.01876 0.01831
-1.90 0.02872 0.02807 0.02743 0.02680 0.02619 0.02559 0.02500 0.02442 0.02385 0.02330
-1-80 0.03593 0.03515 0.03438 0.03362 0.03288 0.03216 0.03144 0.03074 0.03005 0.02938
-1.70 0.04457 0.04363 0.04272 0.04182 0.04093 0.04006 0.03920 0.03836 0.03754 0.03673
-1-60 0.05480 0.05370 0.05262 0.05155 0.05050 0.04947 0.04846 0.04746 0.04648 0.04551
-1-50 0.06681 0.06552 0.06426 0.06301 0.06178 0.06057 0.05938 0.05821 0.05705 0.05592
-1.40 0.08076 0.07927 0.07780 0.07636 0.07493 0.07353 0.07215 0.07078 0.06944 0.06811
-1-30 0.09680 0.09510 0.09342 0.09176 0.09012 0.08851 0.08691 0.08534 0.08379 0.08226
-1.20 0.11507 0.11314 0.11123 0.10935 0.10749 0.10565 0.10383 0.10204 0.10027 0.09853
-1.10 0.13567 0.13350 0.13136 0.12924 0.12714 0.12507 0.12302 0.12100 0.11900 0.11702
-1.00 0.15866 0.15625 0.15386 0.15150 0.14917 0.14686 0.14457 0.14231 0.14007 0.13786
-0.90 0.18406 0.18141 0.17879 0.17619 0.17361 0.17106 0.16853 0.16602 0.16354 0.16109
-0.80 0.21186 0.20897 0.20611 0.20327 0.20045 0.19766 0.19489 0.19215 0.18943 0.18673
-0.70 0.24196 0.23885 0.23576 0.23270 0.22965 0.22663 0.22363 0.22065 0.21770 0.21476
-0.60 0.27425 0.27093 0.26763 0.26435 0.26109 0.25785 0.25463 0.25143 0.24825 0.24510
-0.50 0.30854 0.30503 0.30153 0.29806 0.29460 0.29116 0.28774 0.28434 0.28096 0.27760
-0.40 0.34458 0.34090 0.33724 0.33360 0.32997 0.32636 0.32276 0.31918 0.31561 0.31207
-0-30 0.38209 0.37828 0.37448 0.37070 0.36693 0.36317 0.35942 0.35569 0.35197 0.34827
-0.20 0.42074 0.41683 0.41294 0.40905 0.40517 0.40129 0.39743 0.39358 0.38974 0.38591
-0.10 0.46017 0.45620 0.45224 0.44828 0.44433 0.44038 0.43644 0.43251 0.42858 0.42465
0.00 0.50000 0.49601 0.49202 0.48803 0.48405 0.48006 0.47608 0.47210 0.46812 0.46414
 z2 / 2
d e
~
Pr( z  d )  N (d )   dz 0

2
Range of d : 0.00  d  2.49
d 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
0.00 0.50000 0.50399 0.50798 0.51197 0.51595 0.51994 0.52392 0.52790 0.53188 0.53586
0.10 0.53983 0.54380 0.54776 0.55172 0.55567 0.55962 0.56356 0.56749 0.57142 0.57535
0.20 0.57926 0.58317 0.58706 0.59095 0.59483 0.59871 0.60257 0.60642 0.61026 0.61409
0.30 0.61791 0.62172 0.62552 0.62930 0.63307 0.63683 0.64058 0.64431 0.64803 0.65173
0.40 0.65542 0.65910 0.66276 0.66640 0.67003 0.67364 0.67724 0.68082 0.68439 0.68793
0.50 0.69146 0.69497 0.69847 0.70194 0.70540 0.70884 0.71226 0.71566 0.71904 0.72240
0.60 0.72575 0.72907 0.73237 0.73565 0.73891 0.74215 0.74537 0.74857 0.75175 0.75490
0.70 0.75804 0.76115 0.76424 0.76730 0.77035 0.77337 0.77637 0.77935 0.78230 0.78524
0.80 0.78814 0.79103 0.79389 0.79673 0.79955 0.80234 0.80511 0.80785 0.81057 0.81327
0.90 0.81594 0.81859 0.82121 0.82381 0.82639 0.82894 0.83147 0.83398 0.83046 0.83891
1.00 0.84134 0.84375 0.84614 0.84850 0.85083 0.85314 0.85543 0.85769 0.85993 0.86214
1.10 0.86433 0.86650 0.86864 0.87076 0.87286 0.87493 0.87698 0.87900 0.88100 0.88298
1.20 0.88493 0.88686 0.88877 0.89065 0.89251 0.89435 0.89617 0.89796 0.89973 0.90147
1.30 0.90320 0.90490 0.90658 0.90824 0.90988 0.91149 0.91309 0.91466 0.91621 0.91774
1.40 0.91924 0.92073 0.92220 0.92364 0.92507 0.92647 0.92785 0.92922 0.93056 0.93189
1.50 0.93319 0.93448 0.93574 0.93699 0.93822 0.93943 0.94062 0.94179 0.94295 0.94408
1.60 0.94520 0.94630 0.94738 0.94845 0.94950 0.95053 0.95154 0.95254 0.95352 0.95449
1.70 0.95543 0.95637 0.95728 0.95818 0.95907 0.95994 0.96080 0.96164 0.96246 0.96327
1.80 0.96407 0.96485 0.96562 0.96637 0.96712 0.96784 0.96856 0.96926 0.96995 0.97062
1.90 0.97128 0.97193 0.97257 0.97320 0.97381 0.97441 0.97500 0.97558 0.97615 0.97670
2.00 0.97725 0.97778 0.97831 0.97882 0.97932 0.97982 0.98030 0.98077 0.98124 0.98169
2.10 0.98214 0.98257 0.98300 0.98341 0.98382 0.98422 0.98461 0.98500 0.98537 0.98574
2.20 0.98610 0.98645 0.98679 0.98713 0.98745 0.98778 0.98809 0.98840 0.98870 0.98899
2.30 0.98928 0.98956 0.98983 0.99010 0.99036 0.99061 0.99086 0.99111 0.99134 0.99158
2.40 0.99180 0.99202 0.99224 0.99245 0.99266 0.99286 0.99305 0.99324 0.99343 0.99361
 z2 / 2
d e
~
Pr( z  d )  N (d )   dz 0

2
Range of d : 2.50  d  4.99
d 0.00 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09
2.50 0.99379 0.99396 0.99413 0.99430 0.99446 0.99461 0.99477 0.99492 0.99506 0.99520
2.60 0.99534 0.99547 0.99560 0.99573 0.99585 0.99598 0.99609 0.99621 0.99632 0.99643
2.70 0.99653 0.99664 0.99674 0.99683 0.99693 0.99702 0.99711 0.99720 0.99728 0.99736
2.80 0.99744 0.99752 0.99760 0.99767 0.99774 0.99781 0.99788 0.99795 0.99801 0.99807
2.90 0.99813 0.99819 0.99825 0.99831 0.99836 0.99841 0.99846 0.99851 0.99856 0.99861
3.00 0.99865 0.99869 0.99874 0.99878 0.99882 0.99886 0.99889 0.99893 0.99897 0.99900
3.10 0.99903 0.99906 0.99910 0.99913 0.99916 0.99918 0.99921 0.99924 0.99926 0.99929
3.20 0.99931 0.99934 0.99936 0.99938 0.99940 0.99942 0.99944 0.99946 0.99948 0.99950
3.30 0.99952 0.99953 0.99955 0.99957 0.99958 0.99960 0.99961 0.99962 0.99964 0.99965
3.40 0.99966 0.99968 0.99969 0.99970 0.99971 0.99972 0.99973 0.99974 0.99975 0.99976
3.50 0.99977 0.99978 0.99978 0.99979 0.99980 0.99981 0.99981 0.99982 0.99983 0.99983
3.60 0.99984 0.99985 0.99985 0.99986 0.99986 0.99987 0.99987 0.99988 0.99988 0.99989
3.70 0.99989 0.99990 0.99990 0.99990 0.99991 0.99991 0.99992 0.99992 0.99992 0.99992
3.80 0.99993 0.99993 0.99993 0.99994 0.99994 0.99994 0.99994 0.99995 0.99995 0.99995
3.90 0.99995 0.99995 0.99996 0.99996 0.99996 0.99996 0.99996 0.99996 0.99997 0.99997
4.00 0.99997 0.99997 0.99997 0.99997 0.99997 0.99997 0.99998 0.99998 0.99998 0.99998
4.10 0.99998 0.99998 0.99998 0.99998 0.99998 0.99998 0.99998 0.99998 0.99999 0.99999
4.20 0.99999 0.99999 0.99999 0.99999 0.99999 0.99999 0.99999 0.99999 0.99999 0.99999
4.30 0.99999 0.99999 0.99999 0.99999 0.99999 0.99999 0.99999 0.99999 0.99999 0.99999
4.40 0.99999 0.99999 1.00000 1.00000 1,00000 1.00000 1.00000 1.00000 1.00000 1.00000
4.50 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000
4.60 1,00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000
4.70 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000
4.80 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000
4.90 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000 1.00000
APPENDIX 11.4 PARTIAL DERIVATIVES OF EUROPEAN COMMODITY

OPTION VALUATION EQUATIONS
The valuation equations for the European call and put options are
c( S , T ; X )  Se ( br )T N (d1 )  Xe  rT N (d 2 ) (A4.1)
and
p ( S , T ; X )  Xe  rT N (d 2 )  Se ( br )T N (d1 ) , (A4.2)
respectively, where
ln( S / X )  (b  0.5 2 )T
d1  and d 2  d1   T
 T
d 2  d1   T (A4.3)
d 2  d 1  2d1 T   2T
2 2
(A4.4)
d 2  d 1  2[ln(S / X )  bT  0.5 2T ]   2T  d 12  2[ln(Se bT / X )]
2 2
d 22  d 12  2[ln(Se bT / X ) (A4.5)
 d 22 /2
e
n( d 2 )  (A4.6)
2
 d 12 / 2[ln( SebT / X )  d 12 [ln( SebT / X )
e e e
n( d 2 )    n(d1 ) Se bT / X (A4.7)
2 2
n(d1 )  n(d 2 ) X / Se bT (A4.8)
c N (d1 ) N (d 2 )
c   e (br )T N (d1 )  Se ( br )T  Xe rT
S S S
c N ( d ) d N (d 2 ) d 2
c   e ( br )T N (d1 )  Se (br )T 1 1
 Xe rT
S d1 S d 2 S
c d d
c   e ( br )T N (d1 )  Se ( br )T n(d1 ) 1  Xe rT n(d 2 ) 2
S S S
c d d
c   e ( br )T N (d1 )  Se ( br )T n(d1 ) 1  Xe rT 1 n(d1 ) Se bT / X
S S S
c
c   e ( br )T N (d1 )  0 (A4.9a)
S
p N (d 2 ) ( br )T N (d1 )

p   Xe rT e N (d1 )  Se ( br )T
S S S
p d d
p   e ( br )T N (d1 )  [ Xe rT n(d1 ) Se bT / X ] 1  Se ( br )T n(d1 ) 1
S S S
p
p   e ( br )T N (d1 )  0 (A4.9b)
S
c / c S S
c    c  e ( br )T N (d1 )  1 (A4.10a)
S / S c c
p / p S S
p    p  e ( br )T N (d1 )  1 (A4.10b)
S / S p p
 c e ( br )T N (d1 ) d e ( b  r ) T n ( d1 )
c    e ( b  r ) T n ( d1 ) 1  0 (A4.11a)
S S S S T
  e ( br )T N ( d1 ) d e ( b  r ) T n ( d1 )
p  p   e ( b  r ) T n ( d1 ) 1  c  0 (A4.11b)
S S S S T
c
 e rT N (d 2 )  0 (A4.12a)
X
p
 e rT N (d 2 )  0 (A4.12b)
X
c N (d1 ) N (d 2 )
 TSe (br )T N (d1 )  Se ( br )T  Xe rT  TSe (br )T N (d1 )  0 (A4.13a)
b b b
p N ( d 2 ) N (d1 )
 Xe rT  TSe ( br )T N (d1 )  Se ( br )T  TSe ( br )T N (d1 )  0 (A4.13b)
b b b
c  N ( d1) N (d 2 )
 TSe ( br )T N (d1 )  TSe ( br )T N (d1 )  Se ( br )T  TXe rT N (d 2 )  Xe rT N (d 2 )
r r r
 rT
 TXe N (d 2 )  0 (A4.14a)
p N ( d 2 ) N (d1 )
 TXe rT N (d 2 )  Xe rT N (d 2 )  TSe ( br )T N ( d1 )  TSe ( br )T N (d1 )  Se (br )T
r r r
 rT
 TXe N (d 2 )  0 (A4.14b)
c N (d1 ) N (d 2 ) d d
Vegac   Se ( br )T  Xe rT  Se (br )T n(d1 ) 1  Xe rT n(d 2 ) 2
    
c  d  d  d d 
Vegac   Se ( br )T n(d1 ) 1  [ Xe rT n(d1 ) Se bT / X ] 2  Se (br )T n(d1 )  1  2 
      
( b  r )T
 Se n(d1 ) T  0 (A4.15a)
p N (d 2 ) N (d1 )  d   d 
Vega p   Xe rT  Se ( br )T  Xe rT n(d 2 )  2   Se ( br )T n(d1 )  1 
        
p  d d 
Vega p   Se ( br )T n(d1 )  1  2   Se ( br )T n(d1 ) T  Vegac  0 (A4.15b)
    
c N(d1) N(d2 )
c   (b  r)Se(br )T N(d1)  Se(br )T  rXerT N(d2 )  XerT
T  
c  d d 
c   Se (br )T n(d1 )  1  2   (b  r ) Se ( br )T N (d1 )  rXe rT N (d 2 )
T  T T 
c 
c   Se (br )T n(d1 )  (b  r ) Se (br )T N (d1 )  rXe rT N (d 2 )   0 (A4.16a)
T 2 T
p N (d 2 ) N (d1 )
p    rXe rT N (d 2 )  Xe rT  (b  r ) Se ( br )T N (d1 )  Se (br )T
T T T
p  d d 
p   rXe rT N (d 2 )  (b  r ) Se (br )T N (d1 )  Se ( br )T n(d1 )  1  2 
T  T T 
p 
p   Se ( br )T n(d1 )  (b  r ) Se ( br )T N (d1 )  rXe rT N (d 2 )   0 (A4.16b)
T 2 T
 d1 d 2   ln(S / X ) b    ln(S / X ) b   
 T  T    2T 3 / 2  2 T  4 T    2T 3 / 2  2 T  4 T   2 T
   
12 ESTRATEGIAS DE NEGOCIACIÓN DE OPCIONES
12.1 ESTRATEGIAS DE NEGOCIACIÓN Y DIAGRAMAS DE GANANCIA

 Premisas
 Seis Posiciones Básicas
 Estrategias de Arbitraje Libres de Riesgo
 Posiciones Sintéticas
 Posiciones con Múltiples Opciones/Mercancías
 Estrategias de Margen
12.2 CALCULO DE PROBABILIDADES DE EQUILIBRIO Y GANANCIAS

ESPERADAS
 Probabilidad de Equilibrio
 Ganancia esperada a Termino
12.3 REPLICACIÓN DE OPCIONES A LARGO PLAZO
12.4 ADMINISTRACIÓN DINAMICA DE CARTERAS

 Riesgo y Retorno Esperado
 administración de cambios inesperados en el commodity
 administración de cambios inesperados en la volatilidad
 administración de la caída por tiempo
12.5 RESUMEN
12 ESTRATEGIAS DE NEGOCIACIÓN DE OPCIONES

En el Capítulo 10 desarrollamos las relaciones de precio sin arbitraje entre las opciones y el subyacente. En
el Capítulo 11 desarrollamos las ecuaciones de valuación de opciones. En este capítulo, se utilizan los
elementos esenciales de estos capítulo para describir y analizar las posiciones de una cartera de
opciones/mercancías. Examinamos en detalle los seis diagramas básicos de ganancia final de la
opción/mercancía que son: las posiciones largas y cortas en la call, la put y el subyacente. Combinaciones de
estas posiciones crean una variedad amplia de carteras con diferentes precios de equilibrio en las mercancías,
pérdidas máximas, ganancias máximas. Se consideran las posiciones sintéticas, estrategias de márgenes
(spreads), estrategias para escribir opciones y estrategias especulativas. En la segunda sección, con
distribución de precios lognormal realizamos el análisis de diagrama de ganancia: probabilidades de pérdidas
máximas y ganancias máximas, así como las ganancias esperadas para las diferentes estrategias. La tercera
sección discute una metodología para simular posiciones de opciones a largo plazo utilizando opciones a
corto plazo. Esto es particularmente útil en mercados donde opciones a largo plazo no se negocian
activamente. En la cuarta sección, discutimos el manejo retorno/riesgo. Primero, discutimos las relaciones
retorno-esperado/riesgo creadas al utilizar sólo opciones y opciones en combinación con inversión en el
subyacente. Luego nos enfocamos en la administración del riesgo de cartera en intervalos de tiempo cortos y
mostramos cómo se pueden reunir las exposiciones al riesgo de las posiciones de opciones individuales para
encontrar las exposiciones al riesgo de la cartera. Una selección juiciosa de las inversiones en opciones,
futuros y el subyacente puede manejar con efectividad estos riesgos.
12.1 ESTRATEGIAS DE NEGOCIACIÓN Y DIAGRAMAS DE GANANCIA

Esta sección se enfoca en las estrategias si las posiciones de la opción/mercancía se mantienen hasta el
vencimiento. Después de enunciar las premisas del análisis, presentamos seis posiciones básicas sobre las
que se basan las estrategias de negociación. Los diagramas se utilizan primero para confirmar la relación de
paridad put-call utilizando el arbitraje de conversión y conversión inversa. Después presentamos las
estrategias para escribir opciones, junto con algunas estrategias especulativas. Las estrategias aquí
presentadas están diseñadas sólo como ejemplos, ya que el número de estrategias de negociación posibles es
ilimitado. Sin embargo, el marco referencial del análisis es lo suficientemente general como para que el
lector pueda analizar carteras más complejas con las herramientas presentadas.
Premisas
La única premisa nueva en esta sección es que todas las posiciones, incluyendo cualquier posición en el
subyacente, se mantiene hasta el vencimiento de las opciones, T . Como en los capítulos anteriores, el costo
de acarreo del subyacente ocurre a la tasa b . Una posición larga (corta) implica que el tenedor paga (recibe)
el costo de acarreo. También, el costo de acarreo incluye, además del interés, los cargos (recibos) adicionales
por mantener la mercancía. El costo de acarreo de un contrato de opciones es sólo la tasa de interés libre de
riesgo, r . El precio inicial de la mercancía se denota como S y el precio final se denota como ST . Los
precios iniciales de la call y la put se denotan como c y p , respectivamente. Si la opción es comprada, el
precio de compra se financia a la tasa r , y si la opción es vendida, los ingresos de la venta son invertidos a la
tasa r hasta el vencimiento de la opción. A no ser que se especifique de otra manera, se asume que la call y
la put tienen el mismo precio de ejercicio. Una ganancia ocurre cuando una posición gana más que el costo
del interés de los fondos atados en la posición.
El análisis de cada estrategia es en dos pasos. Primero, presentamos la función de ganancia al
vencimiento,  T , y luego mostramos la función de ganancia en un diagrama. Luego describimos los precios
finales de punto de equilibrio de la mercancía (cuando la estrategia tiene una ganancia cero al vencimiento),
la pérdida máxima y la ganancia máxima. Los diagramas trazan la posición de ganancia cero como una línea
sólida horizontal y trazan la función de ganancia como otra línea sólida. La intersección de las líneas sólidas
representa la posición de ganancia cero.
Seis Posiciones Básicas: Las funciones de ganancia final de las seis posiciones básicas opción/mercancía
son:
1. Posición Larga en Futuros de Mercancías (o Spot): La mercancía es comprada y mantenida a la tasa b

hasta el vencimiento de la opción en T .
 T  ST  SebT (12.1)
Figura 12.1a: Posición Larga en Futuros de Mercancías
Long Futures Position
60
40
20
Payoff
0
-20 0 50 100 150 200
-40
-60
Terminal Commodity Pirce
Punto de equilibrio: ST  SebT .

Pérdida Máxima: SebT si ST cae a cero.
Ganancia Máxima: Ilimitada, si ST se eleva sin límite.
2. Posición Corta en Futuros de Mercancías (o Spot): La mercancía es vendida y los ingresos de la venta
obtienen una tasa b hasta el vencimiento en T .
 T  SebT  ST (12.1b)
Figura 12.1b: Posición Corta en Futuros de Mercancías
Short Commodity Futures
60
40
20
Payoff
0
-20 0 50 100 150 200
-40
-60
Terminal Commodity Futures
Punto de equilibrio: ST  SebT .

Ganancia Máxima: SebT si ST cae a cero.
Pérdida Máxima: Ilimitada, si ST se eleva sin límite.
3. Call Larga: La call es comprada y mantenida a la tasa r hasta el vencimiento de la opción en T .

 S  X  cerT si ST  X
 T   T rT . (12.2a)
ce si ST  X
Figura 12.2a: Call Larga
Long Call Profit
50
40
30
Profit
20
10
0
-10 50 70 90 110 130 150
Terminal Commodity Price
Punto de equilibrio: ST  X  ce rT .
Pérdida Máxima: cerT si ST  X .
4. Call Corta: La call es vendida, y los ingresos de la venta ganan r hasta el vencimiento en T .
 ST  X  cerT si ST  X
 T   rT . (12.2b)
ce si ST  X
Figura 12.2b: Call Corto
Short Call Profit
10
0
-10 50 70 90 110 130 150
Profit
-20
-30
-40
-50
Punto de equilibrio: ST  X  ce rT .
Ganancia Máxima: cerT si ST  X .
5. Put Larga: La put es comprada y mantenida a la tasa r hasta el vencimiento de la opción en T .

 perT si ST  X
T   . (12.3a)
 X  ST  pe si ST  X
rT
Figura 12.3a: Put Larga
Long Put Profit
50
40
Profit 30
20
10
0
-10 50 70 90 110 130 150
Punto de equilibrio: ST  X  perT .

Pérdida Máxima: pe rT si ST  X .
Ganancia Máxima: X  pe rT , si ST cae a cero.
6. Put Corta: La put es vendida, y los ingresos de la venta ganan r hasta el vencimiento de la opción en T .
 perT si ST  X
T   (12.3b)
 X  ST  pe si ST  X
rT
Figura 12.3b: Put Corta

Short Put Profit
10
0
-10 50 70 90 110 130 150
Profit
-20
-30
-40
-50
Punto de equilibrio: ST  X  perT

Ganancia Máxima: pe rT si ST  X
Pérdida Máxima: X  pe rT , si ST cae a cero
En las posiciones de las opciones descritas arriba y más adelante en el capítulo, la prima pagada al
comprador o recibida por el vendedor depende del precio de la mercancía en el momento que el contrato fue
escrito. Al trazar los diagramas de ganancia, hacemos supuestos razonables sobre las primas iniciales de la
put o la call, pero la forma de los diagramas de ganancia no se ve afectada por ninguna prima inicial pagada.
Para posiciones que involucran un solo precio de ejercicio, nosotros generalmente asumimos que estamos at-
the-money cuando comenzamos a negociar. Para posiciones que involucran precios de ejercicio diferentes,
generalmente asumimos que una opción está at-the-money y la otra no.
Estrategias de Arbitraje Libres de Riesgo

Las primeras dos estrategias, llamadas arbitraje de conversión y arbitraje de conversión inversa, se
discutieron en el Capítulo 10 en la sección de la paridad put-call. Estas estrategias de negociación libres de
riesgo son diseñadas para tomar ventaja de los errores temporales entre los precios entre de la call, put y el
subyacente. Para determinar las funciones de ganancia de estos, así como las subsiguientes estrategias de
negociación discutidas en esta sección simplemente sumamos las ganancias correspondientes de los activos
presentados en las ecuaciones (12.1) a (12.3).
7. Conversión: Comprar la mercancía, comprar la put, vender la call.

 T   SebT  X  (c  p)e rT (12.4)
Figura 12.4: Conversión
Conversion Profit
15
10
5
Porfit
0
-5 70 90 110 130
-10
-15
Punto de equilibrio: ninguno o infinito.

Pérdida Máxima:  T   SebT  X  (c  p)e rT .
Ganancia Máxima:  T   SebT  X  (c  p)e rT .
La función de ganancia (12.4) muestra que la ganancia de la cartera al vencimiento de la opción está
asegurada porque el precio final de la mercancía, ST , no aparece en la expresión. Esto también se puede ver
en la Figura 12.4, si la ganancia de la cartera es descrita por la línea horizontal superior. Dado que esta línea
nunca cruza la línea de ganancia cero debajo de ella, la ganancia está asegurada.
El significado económico de este resultado de ganancia segura es importante. Recuerden que todas
las posiciones en la mercancía/opción son completamente autofinanciadas. Por lo tanto, se puede asumir que
si las opciones y la mercancía tienen un precio apropiado en el mercado, la ganancia final de la estrategia es
cero. Y, si la ganancia final es cero, entonces la relación de paridad put-call
(c  p)  Se (br )T  Xe  rT (10.22)
que fue derivada en el Capítulo 10, se mantiene. Cuando la paridad put-call se mantiene, las dos líneas
sólidas en la Figura 12.4 coinciden.
El arbitraje de conversión entra en juego cuando hay errores temporales en los precios. Por ejemplo,
un negociador institucional podría comprar un gran número de calls del índice en reacción a información
nueva sobre el mercado de valores. Como resultado del exceso de la presión de compra sobre los call, el
precio podría incrementar en más de lo que está garantizado con relación a los precios de la put y el índice de
valores subyacente. Conforme el elaborador del mercado vende los calls al negociador institucional,
simultáneamente compra los puts y la cartera del índice para deshacerse del riesgo del precio de la mercancía
de la posición call corta, capturando así la violación temporal de la paridad put-call. Esta actividad de
negociación asegura una cierta ganancia positiva.
8. Conversión inversa: Vender mercancía, vender put y comprar call

Esta estrategia es la inversa de la Estrategia 7 utilizada si se viola la paridad en la otra dirección.
Posiciones Sintéticas
Las conversiones explotan las oportunidades de arbitraje al crear una posición sintética en cualquier activo –
a partir de posiciones juiciosamente seleccionadas en otros dos activos. En una conversión, por ejemplo, una
posición larga en una mercancía, combinada con una posición put larga, es equivalente a una posición call
larga. Cuando esta posición call larga se combina con una posición call corta, se tiene como resultado una
cartera libre de riesgo. Para demostrar esta idea con más detalle, ahora mostramos cómo se pueden crear
posiciones sintéticas largas y cortas en el subyacente utilizando calls y put.
9. Posición larga sintética: Comprar call y vender put.

 T  ST  X  (c  p)e rT (12.5a)
Figura 12.5a: Ganancia en Posición Larga Sintética
Profit Synthetic Long Futures
15
10
5
Profit
0
-5 80 90 100 110 120
-10
-15
Punto de equilibrio: ST  X  (c  p)e rT .

Pérdida Máxima: X  (c  p)e rT si ST cae hasta cero.
La función de ganancia de la posición call larga/put corta es idéntica a la posición larga de la mercancía en la
Figura 12.1a. La única diferencia entre las funciones de ganancia (12.1a) y (12.5a) es que (12.1a) tiene el
término SebT , mientras que(12.5a) tiene el término X  (c  p)e rT . Pero esto es esperado. Si la paridad put-
call (10.22) se mantiene, estos dos valores son iguales. Si por cualquier razón, SebT  X  (c  p)e rT , es más
barato crear sintéticamente una posición larga en la mercancía que comprarla.
10. Posición Corta Sintética en la Mercancía: Vender call y comprar put.

 T   ST  X  (c  p)e rT (12.5b)
Figura 12.5b: Ganancia en Posición Corta Sintética
Profit Synthetic Short Futures
15
10
5
Profit 0
-5 80 90 100 110 120
-10
-15
Punto de equilibrio: ST  X  (c  p)e rT .

Ganancia Máxima: X  (c  p )erT si ST cae hasta cero.
La posición corta sintética en la mercancía es completamente análoga a la posición corta en la mercancía

discutida anteriormente. Esta estrategia es particularmente adecuada si la venta corta del subyacente es difícil
o muy costosa (por ejemplo, venta corta del índice S&P 500). En estos casos, la cartera call corta/put larga se
convierte en una alternativa atractiva.
Las posiciones sintéticas largas y cortas de las opciones también son posibles. Dado que el análisis de
estas posiciones es bastante directo, sólo describimos las siguientes composiciones de cartera:
11. Call larga sintética: Mercancía larga y put larga

12. Call corta sintética: Mercancía corta y put corta.
13. Put larga sintética: Mercancía corta y call larga.
14. Put corta sintética: Mercancía larga y call corta.
Posiciones con Mercancías/Opciones-Múltiples

Se necesita una última idea básica antes de continuar con el análisis de estrategias de opciones más
complejas. ¿Qué sucede cuando se compran o venden varias opciones? Para entender esto, recuerden la
función de ganancia para la posición larga. Para generalizarla para la compra de n calls, simplemente
multiplicamos (12.2a) por n , esto es,
n( ST  X  cerT ) si ST  X
n T   (12.6)
nce si ST  X
rT
La Figura 12.6 muestra los diagramas de ganancia para la compra de una, dos y tres calls. Noten que el punto
de equilibrio es independiente del número de calls comprados. El cambio importante es que se pierde más de
la prima de la opción si la opción vence out-of-the-money, y la tasa de ganancia por dólar del precio de la
mercancía se incrementa si la opción vence in-the-money. Este concepto es utilizado más adelante en el
capítulo cuando se discuten varias estrategias de margen de ratios y estrategias para escribir ratios.
Figura 12.6: Múltiples Calls Largos; Calls largos 1, 2 y 3
Profit Long Multiple Calls
130
80
Profit 30
-20 50 70 90 110 130 150

Estrategias de Margen (Spreads)

Las estrategias de margen son estrategias en las que el riesgo de una posición se compensa, en cierto grado,
por otra posición; esto es, una posición se beneficia por el incremento en el precio de una mercancía y la otra
posición pierde. Una estrategia usualmente será neutral, bull (apuesta al alza) o bear (apuesta a la baja). Una
estrategia neutral es una en la que la estrategia es rentable cuando el precio de la mercancía no se mueve
mucho después de haber asumido la posición. Una estrategia bull es una que obtiene ganancia en el caso que
el precio del subyacente suba, y una estrategia bear obtiene ganancia cuando el precio de la mercancía cae.
Las estrategias de márgenes normalmente involucran las posiciones compensadoras con diferentes
precios de ejercicio o vencimientos diferentes. Uno podría compra una call con un precio de ejercicio o un
vencimiento y vender otra call con otro precio de ejercicio o vencimiento. Las posiciones de margen posibles
se describen bajo cuatro categorías: márgenes de volatilidad, márgenes basados en diferencias en el precio de
ejercicio, márgenes calendario y márgenes diagonales.
Márgenes de Volatilidad. Los márgenes de volatilidad involucran la compra de una put y una call o la venta
de una put y una call.
15. Un “straddle” largo o un margen de volatilidad largo: Comprar call y comprar put.
 S  X  (c  p)erT si ST  X
T   T (12.7)
 X  ST  (c  p)e si ST  X
rT
Puntos de equilibrio: ST  X  (c  p)e rT y (b) ST  X  (c  p)e rT .

Pérdida máxima: (c  p)e rT si ST  X .
Ganancia Máxima: (a) X  (c  p)e rT , si ST cae hasta cero, (b) ilimitada, si ST se eleva sin límite.
Como lo demuestran la función de ganancia (12.7) y la Figura 12.7, la estrategia produce ganancias positivas
cuando el precio del subyacente sube o baja en una cantidad suficiente. Por esta razón, con frecuencia el
comprar un “straddle” se le llama comprar volatilidad. La estrategia pierde dinero cuando el precio final de
la mercancía, ST , está dentro de la banda X  (c  p)e rT al vencimiento de las opciones.
16. Straddle corto o margen de volatilidad corto: Vender call y vender put.
Esta estrategia es inversa al straddle largo. Los inversionistas realizan straddles cortos o venden volatilidad
cuando piensan que el precio de la mercancía no se moverá mucho antes del vencimiento de las opciones.
Figura 12.7: Straddle Largo
Profit Long Stradle
25
15
Profit
5
-5 70 90 110 130
-15
17. Strangle largo: Comprar call y comprar put, con el precio de ejercicio de la put, X p , menor que el precio
del ejercicio de la call X c , (es decir, X p  X c )
 ST  X c  (c  p )e rT si ST  X c

 T  (c  p)erT si X p  ST  X c . (12.8)

 X p  ST  (c  p )e si ST  X p
rT
Figura 12.8: Strangle Largo
Profit Long Strangle
10
5
Profit
0
70 90 110 130
-5
-10
Puntos de equilibrio: (a) ST  X p  (c  p)e rT y (b) ST  X c  (c  p)e rT

Pérdida máxima: (c  p)e rT si X p  ST  X c
Ganancia Máxima: (a) X p  (c  p)e rT , si ST cae hasta cero, (b) ilimitada, si ST se eleva sin límite
Un strangle y un straddle largo tienen el mismo objetivo. La diferencia es que el strangle requiere
menor inversión porque la call o la put están más out-of-the-money que con el straddle. Sin embargo, con un
costo de inversión menor viene una región más amplia en la cual la estrategia no es rentable. El precio de la
mercancía debe moverse más que con el straddle para que el strangle sea rentable al vencimiento. Una
variación en el strangle largo es que el precio de ejercicio de la call está por debajo del precio de ejercicio de
la put. Como tanto la put como la call están in-the-money, el costo y la rentabilidad son mayores.
18. Strangle Corto: Vender call y vender put, con X p  X c .
Combinación de Opciones con Precios de Ejercicio Diferentes. Las calls o puts con precios de ejercicio
diferentes pueden combinarse para crea márgenes bull o bear, márgenes de ratio o ratio inverso y márgenes
“butterfly” largos o cortos.
19. Margen Bull con Calls. Comprar una call con un precio de ejercicio menor, X l , y vender una call
idéntica excepto con un precio de ejercicio mayor, X h (es decir, X l  X h ).
 X h  X l  (cl  ch )erT si ST  X h

 T   ST  X l  (cl  ch )erT si X l  ST  X h . (12.9a)

(cl  ch )e si ST  X l
rT
Figura 12.9a: Margen Bull con Calls
Bull Spread with Calls
20
15
10
Profit
5
0
-5 70 90 110
-10
Punto de equilibrio: ST  X l  (cl  ch )e rT .

Pérdida máxima: (cl  ch )e rT , si ST  X l .
Ganancia máxima: X h  X l  (cl  ch )e rT , si ST  X h .
El término “bull” se debe a que esta estrategia obtiene ganancia cuando se incrementa el precio del
subyacente. La estrategia es bastante conservadora en el sentido que si los inversionistas están confiados en
que se va a incrementar el precio de la mercancía, una posición call larga sería más rentable. El beneficio de
comprar el margen bull es que el vender la call out-of-the-money proporciona un ingreso que compensa el
costo de comprar otra call. El costo es la pérdida de la ganancia potencial si es que el precio de la mercancía
se eleva drásticamente.
Antes del vencimiento, un margen bull toma ventaja del hecho que el delta el mayor para la call in-
the-money que para la call out-of-the-money. Conforme se incrementa el precio de la mercancía, la ganancia
sobre la Call Larga in-the-money sobrepasa la pérdida sobre la call corto out-of-the-money. Conforme se
eleva más el precio de la mercancía y ambas calls se vuelven más in-the-money o conforme se acerca el
vencimiento, la ganancia y la pérdida sobre las dos posiciones se compensan completamente.
20. Margen Bear con Calls: Vender una call con un precio de ejercicio menor, X l , y comprar una call
idéntica excepto con un precio de ejercicio mayor, X h (es decir, X l  X h ).
 X l  X h  (cl  ch )e rT si ST  X h

 T   X l  ST  (cl  ch )erT si X l  ST  X h . (12.9b)

(cl  ch )e si ST  X l
rT
Figura 12.9b: Margen Bear con Calls

Bear Spread with Calls
20
15
10
5
Profit
0
-5 70 90 110 130
-10
-15
-20
Punto de equilibrio: ST  X l  (cl  ch )e rT .

Ganancia máxima: (cl  ch )e rT , si ST  X l
Pérdida máxima: X h  X l  (cl  ch )e rT , si ST  X h
21. Margen Bull con Puts: Comprar una put con un precio de ejercicio menor, X l , y vender una put idéntica
excepto con un precio de ejercicio mayor, X h (es decir, X l  X h ).
( ph  pl )e rT si ST  X h

 T   ST  X h  ( ph  pl )erT si X l  ST  X h (12.10a)

 X l  X h  ( ph  pl )e si ST  X l
rT
Punto de equilibrio: ST  X h  ( ph  pl )e rT
Pérdida máxima: X l  X h  ( ph  pl )e rT , si ST  X l
Ganancia máxima: ( ph  pl )e rT , si ST  X h
El margen “bull” con puts es equivalente al margen bull utilizando calls, aunque las fuentes de los resultados
son diferentes. Para el margen bull utilizando put, la ganancia, si se incrementa el precio de la mercancía,
viene de la prima sobre la put vendida a un precio de ejercicio mayor. Para el margen bull utilizando call, la
ganancia, si se incrementa el precio de la mercancía, viene del hecho que la ganancia sobre la Call Larga a
un precio de ejercicio menor excede la pérdida sobre la call corto a un precio de ejercicio mayor.
Figura 12.10a: Margen Bull con Puts
Bull Spread with Puts
20
15
10
5
Profit
0
-5 50 70 90 110 130 150
-10
-15
-20
22. Margen “Bear” con Puts: Vender la put a un precio de ejercicio menor, X l , y comprar una put idéntica
excepto con un precio de ejercicio mayor, X h , (es decir, X l  X h ).
( ph  pl )e rT si ST  X h

 T   X h  ST  ( ph  pl )erT si X l  ST  X h (12.10b)

 X h  X l  ( ph  pl )e si ST  X l
rT
Figura 12.10b: Margen Bear con Puts

Bear Spread with Puts
20
15
10
5
Profit
0
-5 70 90 110 130
-10
-15
-20
Punto de equilibrio: ST  X h  ( ph  pl )e rT
Pérdida máxima: X l  X h  ( ph  pl )e rT , si ST  X l
Ganancia máxima:  ( ph  pl )e rT , si ST  X h
23. Margen de Ratio con Calls: Vender la call a un precio de ejercicio menor, X l , y vender n calls idénticas
excepto con un precio de ejercicio mayor, X h , (es decir, X l  X h ).
 nX h  X l  (n  1) ST  (nch  cl )e rT si ST  X h

 T   ST  X l  (nch  cl )erT si X l  ST  X h (12.11a)

(nch  cl )e si ST  X l
rT
Figura 12.11a: Margen de Ratio con Calls
2:1 Ratio Spread with Calls
30
20
10
Profit
0
70 90 110 130
-10
-20
nX h  X l  ( nc h  cl )e rT
Puntos de equilibrio: (a) S T  . (b) ST  X l  (nch  cl )e rT si nch  cl .
( n  1)
Pérdida máxima: (a) ilimitada, si ST se eleva sin límite. (b) (nch  cl )e rT , si nch  cl y ST  X l .
Ganancia máxima: X h  X l  (nch  cl )e rT , si ST  X h .
Un margen de ratio call es como un margen bull utilizando calls excepto que varias opciones idénticas son
vendidas al precio de ejercicio mayor.
El “ratio” del margen de ratio se define en términos de la cantidad de call vendidos, n , por call
comprado, esto es, un margen de ratio n : 1 . El margen de ratio descrito en la Figura 12.1 es un margen de
ratio 2:1, el cual involucra la venta de dos calls con un precio de ejercicio mayor contra la compra de una call
con un precio de ejercicio menor.
El margen de ratio es más rentable cuando el precio de la mercancía no se mueve mucho. Si el precio
de la mercancía cae, el riesgo de bajada es fijo. Si el precio de la mercancía cae por debajo del precio de
ejercicio, ambas opciones vencen sin valer nada. La inversión inicial puede ser un débito neto o un crédito
neto, dependiendo de las magnitudes relativas sobre las primas y el ratio del margen. La Figura 12.11a indica
que el margen de ratio 2:1 mostrado tiene un crédito neto inicial. Si el precio de la mercancía sube sin límite,
la estrategia pierde dinero sin límite.
24. Margen de ratio inverso con Calls: Vender la call con un precio de ejercicio menor, X l , comprar n calls
idénticas excepto que tienen un precio de ejercicio mayor, X h (es decir, X l  X h ).
(n  1) ST - nX h  X l  (nch  cl )erT si ST  X h

 T   X l  ST  (nch  cl )erT si X l  ST  X h . (12.11b)

(nch  cl )e si ST  X l
rT
nX h  X l  (nch  cl )e rT
Puntos de equilibrio: (a) ST  (b) ST  X l  (nch  cl )e rT , si nch  cl
(n  1)
Ganancia máxima: (a) ilimitada, si ST se eleva sin límite (b) (nch  cl )e rT , si nch  cl y ST  X l
Pérdida máxima: X h  X l  (nch  cl )e rT , si ST  X h
Figura 12.11b: Margen de Ratio Inverso con Calls
2:1 Reverse Ratio Spread with Calls
40
30
20
10
Profit
0
-10 50 70 90 110 130 150
-20
-30
-40
Terminal Commodity Value
25. Margen de Ratio con Puts: Comprar la put con un precio de ejercicio mayor, X h , vender n puts idénticas
excepto con un precio de ejercicio menor, X l (es decir, X l  X h ).
(npl  ph )erT si ST  X h

 T   X h  ST  (npl  ph )erT si X l  ST  X h . (12.12a)

 X h  nX l  (n  1) ST  (npl  ph )e si ST  X l
rT
Figura 12.12a: Margen de Ratio con Puts

2:1 Ratio Spread with Puts
20
15
10
5
Profit
0
-5 60 80 100 120
-10
-15
-20
nX l  X h  (npl  ph )e rT
Puntos de equilibrio: (a) ST  (b) ST  X h  (npl  ph )e rT , si npl  ph
(n  1)
Pérdida máxima: (a) X h  nX l  (npl  ph )e rT , si ST cae hasta cero (b) (npl  ph )e rT , si npl  ph y ST  X h .
Ganancia máxima: X h  X l  (npl  ph )e rT , si ST  X l
26. Margen de Ratio Inverso con Puts: Vender la put con un precio de ejercicio mayor, X h , vender n puts
idénticos excepto con un precio de ejercicio menor, X l (es decir, X l  X h ).
(npl  ph )erT si ST  X h

n T   ST  X h  (npl  ph )e rT si X l  ST  X h . (12.12b)

nX l  X h  (n  1) ST  (npl  ph )e si ST  X l
rT
Figura 12.12b: Margen de Ratio Inverso con Puts

Reverse Ratio Spread with Puts
20
15
10
5
Profits
0
-5 60 80 100 120
-10
-15
-20
nX l  X h  (npl  ph )e rT
Puntos de equilibrio: (a) ST  (b) ST  X h  (npl  ph )e rT , si npl  ph
(n  1)
Pérdida máxima: (a) X h  nX l  (npl  ph )e rT , si ST cae hasta cero (b) (npl  ph )e rT si npl  ph , y ST  X h .
Ganancia máxima: X h  X l  (npl  ph )e rT si ST  X l
27. Margen “Butterfly” Largo con Calls: Vender la call con un precio de ejercicio menor, X l , comprar dos
calls con el precio de ejercicio medio, X m , y vender la call con un precio de ejercicio mayor, X h .
 X l  2 X m  X h  (cl  2cm  ch )e rT si ST  X h

 S  (2 X m  X l )  (cl  2cm  ch )e si X m  ST  X h
rT
T   T (12.13)
 X l  ST  (cl  2cm  ch )e si X l  ST  X m
rT
(c  2c  c )e rT si ST  X l
 l m h
Puntos de equilibrio: (a) ST  2 X m  X l  (cl  2cm  ch )e rT . (b) ST  X l  (cl  2cm  ch )e rT .

Pérdida Máxima: X l  X m  (cl  2cm  ch )e rT si ST  X m .
Ganancia Máxima: X l  2 X m  X h  (cl  2cm  ch )e rT si ST  X h . (b) (cl  2cm  ch )e rT si ST  X l .
Un margen butterfly largo combina un margen bear y un margen bull. Las ganancias son similares a las de
un margen bull si el precio de la mercancía se incrementa y a aquellas de un margen bear si el precio de la
mercancía cae. El inversionista pierde dinero cuando el precio de la mercancía se mantiene neutral. El
diagrama de ganancia resultante (Figura 12.13) se asemeja a una mariposa (butterfly) –de allí el nombre.
Figura 12.13: Margen Butterfly Largo con Calls
Long Butterfly Spread with Calls
15
10
5
Profit
0
-5 70 90 110 130
-10
-15
28. Margen “Butterfly” Corto con Calls: Comprar la call con un precio de ejercicio menor, X l , vender dos
calls con el precio de ejercicio medio, X m , y vender la call con un precio de ejercicio mayor, X h .
29. Margen “Butterfly” Largo con Puts: Vender la put con un precio de ejercicio menor, X l , comprar dos
puts con el precio de ejercicio medio, X m , y vender la put con un precio de ejercicio mayor, X h .
30. Margen “Butterfly” Corto con Puts: Comprar la put con un precio de ejercicio menor, X l , vender dos
puts con el precio de ejercicio medio, X m , y comprar la put con un precio de ejercicio mayor, X h .
Márgenes calendario: Un margen calendario requiere la compra de una call o put con un vencimiento y la
venta de una opción idéntica con un vencimiento diferente. Presentar una función de ganancia para un
margen calendario es problemático porque se requiere una ecuación de precio para mostrar el valor de una
opción distante en la fecha de vencimiento más cercana.104 Por estar razón, vamos inmediatamente al
diagrama de ganancia.
31. Margen Calendario Largo: Comprar una call con un vencimiento distante, y vender una call idéntica con
un vencimiento cercano.
Figura 12.14: Margen Calendario Largo con Calls
Long Calendar Spread with Calls
15
10
5
Profit
0
-5 70 90 110 130
-10
-15
104
Las ecuaciones de la call y put europea (11.25) y (11.28) se utilizan para poner precio a la opción distante en la fecha de
vencimiento más cercana.
Las opciones at-the-money son utilizadas para generar el margen calendario de la Figura 12.14, y los
resultados son trazados al vencimiento de una call cercana. Este margen es neutral dado que se obtienen
ganancias positivas siempre y cuando el precio de la mercancía no se mueva mucho debido al vencimiento
de la opción cercana.105 Dado que la opción de plazo más largo tiene un precio mayor, esta estrategia tiene
una posición de débito neto (es decir, pagamos la diferencia entre los precios de la opción cuando se forma la
posición). Sin embargo, la pérdida máxima se limita a la cantidad de débito neto. La ganancia máxima ocurre
cuando el precio de la mercancía es igual al precio de ejercicio de la opción cercana, pero la cantidad no está
clara dado que depende de la vida restante de la opción distante y la volatilidad del retorno de la mercancía.
La amplitud del rango de ganancia y los puntos de equilibrio también son funciones de la volatilidad y el
tiempo hasta el vencimiento.
Manteniendo los otros factores constantes, la rentabilidad del margen calendario se debe a la
disminución de tiempo de las primas de las opciones. Como demostraremos más adelante en el capítulo, la
tasa de disminución de tiempo (es decir, el theta de la opción) es mayor mientras menor sea el tiempo hasta
el vencimiento. En un margen calendario largo, se establece una posición corta en una opción cercana para
poder capturar su disminución de tiempo a costa de la disminución de tiempo en la opción distante.
32. Margen Calendario Corto con Calls: Vender la call con un vencimiento distante, y comprar una call
idéntica con un vencimiento cercano.
33. Margen Calendario Largo con Puts: Comprar la put con un vencimiento distante, y comprar una put
34. Margen Calendario Corto con Puts: Vender la put con un vencimiento distante, y comprar una put
35. Margen Calendario de Ratio Largo con Calls: Comprar una call con un vencimiento distante, y vender
más de una call idéntica con un vencimiento cercano.
Figura 12.15: Margen Calendario de Ratio Largo con Calls
Long Ratio Calendar Spread with

Calls
30
20
10
Profit
0
-10 50 70 90 110 130 150
-20
-30
Al escribir más de una call con vencimiento cercano, el que realiza un margen calendario usualmente recibe
un crédito inicial (él recibe dinero cuando se forma una posición). El crédito neto incrementa la ganancia si
el precio de la mercancía cae por debajo del precio de ejercicio antes del vencimiento de la opción cercana.
Los incrementos en el precio de la mercancía más allá del precio de ejercicio de la opción cercana reducen la
cantidad de la ganancia dado que, cuando ambas opciones están muy in-the-money, el que realiza el margen
está sintéticamente corto en la mercancía. En general, la posición tiende a la baja. La Figura 12.15 muestra el
diagrama de ganancia de un margen calendario con ratio 2:1.
105
Si se utilizan calls out-of-the-money para formar el margen calendario, la posición tiende ligeramente al alza.
36. Margen Calendario de Ratio Corto con Calls: Vender una call con un vencimiento distante, y comprar
más de una call idéntica con un vencimiento cercano.
37. Margen Calendario de Ratio Largo con Puts: Comprar una put con un vencimiento distante, y vender más
de una put idéntica con un vencimiento cercano.
38. Margen Calendario de Ratio Corto con Puts: Vender una put con un vencimiento distante, y comprar más
de una put idéntica con un vencimiento cercano.
Márgenes Diagonales. En general, los márgenes diagonales son cualquier posición de margen conformada
con precios de ejercicio diferentes y vencimientos diferentes. Una margen diagonal largo requiere que la
opción distante sea comprada y la opción cercana sea vendida (comprada). Si el ratio de margen es 1:1, se
tiene como resultado un margen bull (bear) diagonal, dependiendo de si la opción distante tiene un precio de
ejercicio menor (mayor). Los márgenes diagonales largos y cortos que utilizan otros ratios producen una
amplia variedad de posiciones bull y bear. A continuación se describe un margen diagonal posible.
39. Margen bull diagonal: Comprar una call con un precio de ejercicio menor y vencimiento distante, y
vender una call idéntica con un precio de ejercicio mayor y un vencimiento cercano.
Como lo muestra la Figura 12.6, un margen bull diagonal es muy similar al margen bull descrito
anteriormente en esta sección. La pérdida máxima se limita a la diferencia entre los precios de opciones
distantes y cercanas (es decir, la cantidad de débito neto). La ganancia máxima se da cuando el precio de la
mercancía es igual al precio de ejercicio de la opción cercana al vencimiento de la opción cercana. Más allá
de ese nivel, los incrementos en el precio de la mercancía reducen el nivel de ganancia a la diferencia entre
los precios de ejercicio y la cantidad de débito neto.
Figura 12.16: Margen Bull Diagonal Largo

Long Diagonal Bull Spread
15
10
5
Profit
0
-5 75 95 115 135
-10
-15
Terminal Commoidty Price
Venta de Opciones y Estrategias Especulativas

En esta sección, examinamos los efectos de comprar y vender opciones contra una posición en el subyacente.
En general, nosotros discutimos las estrategias que reducen el riesgo de una posición larga en la mercancía
escribiendo calls o comprando puts. Pero también consideramos los efectos de comprar calls y escribir puts
para poder incrementar el apalancamiento.
40. Escribir call cubierta: Vender una call contra una posición larga en el subyacente.
 X  Se  ce si ST  X
bT rT
T   (12.14)
 ST  Se  ce si ST  X
bT rT
Figura 12.17: Escribir Call Cubierta

Covered Call Writing
15
10
5
Profit
0
-5 80 90 100 110 120
-10
-15
Punto de equilibrio: ST  SebT  ce rT

Pérdida Máxima: SebT  ce rT si ST cae hasta cero
Ganancia Máxima: X  SebT  ce rT , si ST  X
La Figura 12.17 muestra que el que escribe una call cubierta recibe la prima de la opción a cambio de un
mayor potencial de una posición larga en la mercancía. La posición es equivalente a vender una put sin
subyacente. Esta estrategia sólo tiene sentido si un inversionista piensa que el precio de la mercancía no se
moverá mucho durante la vida de la opción. No se beneficia si el precio de la mercancía se eleva, y la prima
de la opción es un consuelo pequeño si el precio de la mercancía cae drásticamente.
Los grandes fondos de acciones con frecuencia se involucran en una forma especial en escribir calls
cubiertas llamada “overwriting” de opciones. En el caso usual, el fondo tiene administradores de cartera de
acciones y opciones separados. El gerente de la cartera de acciones maneja la inversión en acciones y
aconseja al que realiza el overwrite de la opción sobre la composición actual de la cartera de acciones. El que
realiza el overwrite de la opción luego escribe calls contra las acciones. En el caso que una call se ejerza
contra el que realiza el overwrite de la opción, el que realiza el overwrite de la opción debe comprar las
acciones para entregarlas sobre la opción, ya que él no tiene autoridad para entregar una posición de acciones
existentes. Sin embargo, el propietario del fondo, debe esperar que se tengan que liquidar algunas de sus
acciones, ya que el escribir opciones contra las acciones es una estrategia para reducir el riesgo.
41. Combinación escribir call cubierta: Vender calls in-the-money a la mitad de la posición de la mercancía,
y vender calls out-of-the-money contra la otra mitad.
Figura 12.18: Combinación Call cubierta
Combination Covered Call
30
20
10
Profit
0
-10 70 90 110 130
-20
-30
Esta estrategia es, generalmente, la misma que la estrategia de escribir call cubierta. La Figura 12.18 muestra
que la ganancia es sólo ligeramente diferente. A lo largo del rango de precio de la mercancía entre los
precios de ejercicio, el que escribe la opción comparte la mitad de cualquier ganancia realizada en el precio
de la acción. Sin embargo, como en la estrategia de la call cubierta anterior, el mayor potencial de una
posición larga en la mercancía se niega completamente una vez que el precio de la mercancía excede cierto
nivel, en este caso el precio de ejercicio de la opción out-of-the-money.
42. Escribir ratio de call: Vender más de una call contra una posición larga en el subyacente.
nX  (n  1) ST  nce rT si ST  X
T   . (12.15)
 ST  Se  nce si ST  X
bT rT
Punto de equilibrio: ST  SebT  nce rT .

Ganancia Máxima: nce rT , si ST  X al vencimiento.
Figura 12.19: Escribir Ratio de Calls

Ratio Call Writing
15
10
5
Profit
0
-5 70 90 110 130
-10
-15
Como los márgenes de ratio, los ratios escritos se expresan en términos del número opciones c vendidas, n ,
por unidad del subyacente. Por lo tanto, un ratio de 2:1 se refiere a escribir dos calls contra una unidad de la
mercancía. En un ratio 2:1, la mitad de los calls están cubiertos mientras que la otra mitad no. Un ratio de 2:1
como el que se muestra en la Figura 12.19 crea un diagrama de retribución que se ve exactamente como si
hubiésemos escrito un straddle. La ganancia máxima ocurre cuando el precio de la mercancía es igual al
precio de ejercicio al vencimiento de la opción. Sin embargo, grandes cambios en cualquier dirección en el
precio de la mercancía, producen pérdidas.
Usualmente los que piensan que el precio de la mercancía no va a moverse durante la vida de la
opción escriben un ratio cuando se trata de lograr un ingreso adicional. Los calls son escritos al precio de
ejercicio más cercano al precio actual de la mercancía. Las ganancia se obtienen si el precio de la mercancía
se mantiene relativamente sin cambio. Sin embargo, las pérdidas pueden ser significativas si el precio del
subyacente cambia significativamente.
43. Escribir ratio variable: Vender calls in-the-money y calls out-of-the-money contra una posición larga en
una mercancía, de manera que la posición larga en la mercancía sea menos que suficiente para cubrir la
entrega si se ejercen las opciones.
Figura 12.20: Escribir Ratio Variable
Variable Ratio Writing
15
10
5
Profit
0
-5 70 90 110 130
-10
-15
La estrategia de escribir un rateo variable 2:1 se muestra en la Figura 12.20. Como lo muestra la ilustración,
el escribir un ratio variable puede producir un diagrama de ganancia que se ve exactamente como una
posición strangle corta. La ganancia máxima se obtiene cuando el precio de la mercancía cae entre los dos
precios de ejercicio al vencimiento de las opciones. Los grandes movimientos de precios en cualquier
dirección producen pérdidas.
44. Venta corta protegida: Comprar una call contra una posición corta en el subyacente.
Figura 12.21: Venta Corta Protegida
Protected Short Sale
15
10
5
Profit
0
-5 70 90 110 130
-10
-15
Ocasionalmente, un inversionista está corto en la mercancía y quiere asegurarse a sí mismo contra grandes
incrementos en el precio del subyacente. Comprar una call proporciona esta seguridad. Como lo muestra la
Figura 12.21, comprar una call contra una posición corta en la mercancía produce una estructura de ganancia
de cartera que se ve exactamente como una posición put larga. La posición también es lo opuesto a una call
cubierta. La ganancia máxima es igual a SebT  ce rT si el precio de la mercancía cae hasta cero. La pérdida
máxima es SebT  X  cerT que debe ser aproximadamente igual al valor de una put con un precio de
ejercicio X y un tiempo hasta el vencimiento T .
45. Cobertura inversa o straddle simulado: Comprar más de una call contra una posición corta en el
subyacente.
Figura 12.22: Cobertura Inversa
Reverse Hedge
20
15
10
5
Profit
0
-5 70 90 110 130
-10
-15
-20
Comprar dos calls contra una posición corta en el subyacente crea un diagrama de ganancia que se ve
exactamente como un straddle largo. Por esta razón, esta estrategia algunas veces se conoce como un
straddle simulado. La posición es también opuesta a la posición de escribir una ratio de call descrita
anteriormente. La pérdida máxima se da cuando el precio de la mercancía es igual al precio de ejercicio de
las opciones al vencimiento. La ganancia en el lado de movimientos ascendentes es ilimitada, si el precio de
la mercancía se eleva sin límite. Los movimientos descendentes en el precio de la mercancía también son
beneficiosos, dado que las opciones vencen sin valor y el inversionista tiene una posición corta en la
mercancía.
46. Posición protegida en la mercancía: Comprar una put contra una posición larga en la mercancía.
 ST  SebT  pe rT si ST  X
T   . (12.16)
 X  Se  pe si ST  X
bT rT
Figura 12.23: Posición de Mercancía Protegida

Protected Commodity Position
20
15
10
5
Profit
0
-5 70 90 110 130
-10
-15
-20
Punto de equilibrio: ST  SebT  pe rT .

Pérdida Máxima: X  SebT  pe rT si ST  X .
Comprar put protectoras es una forma muy común de asegurar la mercancía. Como lo indica la Figura 12.23,
un inversionista con una posición larga en la mercancía está bien protegido en el caso que el precio de la
mercancía caiga drásticamente. El costo de este seguro es la prima de la opción put. La posición resultante es
la misma que comprar una call.
47. Escribir una call cubierta protegida: Comprar una put contra una call cubierta.
Figura 12.24: Escribir Call Protegida

Protected Covered Call Write
15
10
5
Profit
0
-5 70 90 110 130
-10
-15
Escribir una call cubierta significa que el inversionista mantiene una posición larga en la mercancía/corta en
la call. En el caso que haya pasado algún tiempo desde que se formó la call cubierta y el precio de la
mercancía no se haya movido, el inversionista puede querer asegurar su ganancia proveniente de la
disminución de tiempo de la call comprando una put. Cuando lo hace, en efecto, crea un arbitraje de
conversión. Luego, independientemente de la dirección en que se mueve el precio de la mercancía, la
ganancia de la cartera no cambia. La Figura 12.24 lo demuestra claramente.
 X  SebT  (c  p)erT si ST  X
T   . (12.17)
2 ST  Se  X  (c  p)e si ST  X
bT rT
Figura 12.25: Escribir Straddle Cubierto

Covered Straddle Write
20
15
10
5
Profit
0
-5 70 90 110 130
-10
-15
-20
SebT  X  (c  p)e rT
Punto de equilibrio: ST  .
2
Pérdida Máxima:  SebT  X  (c  p)e rT , si ST cae hasta cero.
Ganancia Máxima: X  SebT  (c  p)e rT si ST  X .
En este caso, el inversionista ha escrito una call y una put contra una posición en el subyacente. Él ha
cobrado dos primas, lo que es igual a la cantidad de la ganancia de cartera si el precio de la mercancía está
por encima del precio de ejercicio al vencimiento de la opción. Sin embargo, si el precio de la mercancía cae,
la ganancia de la cartera cae en el doble de la cantidad, dado que el inversionista no pierde solamente sobre
la posición larga en la mercancía sino también en la posición corta de la put.
49. Comprar call contra la mercancía: Comprar una call contra una posición larga en la mercancía.
Figura 12.26a: Posición Larga en la Mercancía y Call Larga

Long Commodity and Long Call
Position
20
10
Profit
0
70 90 110 130
-10
-20
50. Vender put contra la mercancía: Vender una put contra una posición larga en la mercancía.
Figura 12.26b: Posición Larga en la Mercancía/ Put Corta
Long Commodity and Short Put

Position
20
10
Profit
0
70 90 110 130
-10
-20
El comprar una call contra una posición larga y vender una put contra una posición larga en la mercancía
sirve como apalancamiento para la tasa de las ganancias de la cartera. La posición larga en la mercancía/call
larga de la Figura 12.26a, por ejemplo, muestra que por debajo de cierto nivel en el precio de la mercancía, la
ganancia de la cartera es menor que la posición larga en la mercancía, dado que se tiene que comprar la call.
Sin embargo, por encima de cierto nivel en el precio de la mercancía, la ganancia de la cartera se incrementa
al doble de la tasa que el precio de la mercancía por sí misma. Por lo tanto, hemos incrementado el
apalancamiento de la estrategia.
El escribir una put contra una posición larga en la mercancía tiene un efecto similar. Los ingresos de
la venta de la opción aumentan la ganancia de la cartera en el lado superior. En el lado inferior, si el precio
de la mercancía cae, el inversionista pierde tanto sobre la posición larga en la mercancía como en la posición
corta del put.
12.2 CÁLCULO DE PROBABILIDADES DEL PUNTO DE EQUILIBRIO Y GANANCIAS ESPERADAS

Dos conceptos útiles para analizar las estrategias mercancía/opción que acabamos de discutir son la
probabilidad que la cartera sea rentable al vencimiento de las opciones y que la ganancia esperada de la
estrategia de negociación. Estos dos conceptos se basan en los procedimientos de precio de las opciones
presentados en el Capítulo 11. Dado que el concepto de ganancia esperada se basa en el cálculo de
probabilidades, primero veremos el cálculo de probabilidades.106
Probabilidades de Punto de Equilibrio

Para calcular la probabilidad que una estrategia particular sea rentable al vencimiento, primero necesitamos
establecer todos los puntos de equilibrio de la estrategia. Por ejemplo, para una posición straddle larga,
Estrategia de Negociación 15, existen dos puntos de equilibrio. Un punto de equilibrio es donde el precio
final de la mercancía, ST , es BEa  X  (c  p)e rT , y el otro donde ST es BEb  X  (c  p)e rT . La Figura
12.7 muestra que un straddle largo tiene una ganancia positiva, si ST  BEa , o si ST  BEb . Si nosotros
asumimos que el precio de la mercancía está distribuido de manera lognormal, como lo hicimos en el
Capítulo 11, la probabilidad neutral al riesgo que el straddle sea rentable al vencimiento,
~ ~
Pr( ST  BEa or ST  BEb ) , se puede encontrar utilizando la función de distribución acumulativa estándar,
esto es,
~ ~
Pr( ST  BEa or ST  BEb )  N ( d a )  N (d b ) ,
ln(S / BEa )  (b  0.5 2 )T ln(S / BEb )  (b  0.5 2 )T

donde d a  db  . Recuerden que un signo menos en el
 T  T
término d implica que el cálculo de la probabilidad es para la región por debajo del precio final crítico de la
mercancía. Recuerden también que la expresión para d transforma el precio de la mercancía distribuido
lognormalmente en una distribución normal estándar.
EJEMPLO 12.1
Asuman que el precio actual de la mercancía es $50 y que los precios de las opciones a tres meses at-the-
money son $3.35 para la call y $2.90 para la put. Calculen la probabilidad que una posición straddle larga
utilizando estas opciones sea rentable al final de los tres meses. Asuman que la tasa de costo de acarreo para
el subyacente es 4 porciento, la tasa de volatilidad d el subyacente es 32 porciento y la tasa de interés libre de
riesgo es 6 porciento.
El primer paso es calcular los puntos de equilibrio

BEa  50 - (3.35  2.90)e 0.06(0.25)  43.656 y BEb  50  (3.35  2.90)e0.06(0.25)  56.344 .
El segundo paso es transformar los puntos de equilibrio del precio de la mercancía a puntos de equilibrio en
términos de la distribución normal unitaria, esto es,
ln(50 / 43.656)  (0.04  0.50.32 2 )0.25
da   0.8305 y
0.32 0.25
ln(50 / 56.344)  (0.04  0.50.32 2 )0.25
db   0.7641 .
0.32 0.25
Finalmente, la probabilidad que un straddle sea rentable al final de tres meses es

Pr( ST  BEa or ST  BEb )  N (0.8305)  N (0.7641)  0.4255 o 42.55 porciento.
106
En esta sección asumimos que estamos en un mundo neutral al riesgo en el que los precios futuros de las mercancías son
descontados o capitalizados a una tasa libre de riesgo.
Ganancia Final Esperada

La ganancia final esperada de una posición de cartera mercancía/opción es nuestro mejor supuesto sobre cuál
será la ganancia de la cartera al vencimiento. Se puede calcular multiplicando la ganancia de la cartera a cada
precio final posible de la mercancía por la probabilidad que ocurra el precio de la mercancía y luego sumar
todos estos productos. Conceptualmente, aunque el procedimiento mencionado es directo, no podemos
utilizar un número infinito de posiciones de ganancia de la opción. Una alternativa práctica es definir el
rango de posibles precios futuros de la mercancía como (4 desviaciones estándar del precio esperado de la
mercancía, SebT , lo cual, de acuerdo con el Apéndice 11.3, debe dar cuenta del 99.994 porciento de la
distribución del precio de la mercancía. El rango de precios futuros de la mercancía dado por las 4
desviaciones estándar alrededor del precio esperado es definido por
ln( S / S min )  (b  0.5 2 )T ln( S / S max )  (b  0.5 2 )T
4  y 4 
 T  T
Reordenando, las expresiones para el máximo y el mínimo del rango del precio de la mercancía son
2
S min  Se (b0.5 )T 4 T (12.1 8a)
y
2
S max  Se ( b0.5 )T 4 T (12.18b)
Una segunda consideración tiene que ver con el cálculo de la ganancia para una probabilidad dada. Incluso
con un rango pre-especificado del precio final de la mercancía, ST , existe un número infinito de precios de la
mercancía y, por lo tanto, un número infinito de ganancias y probabilidades de cartera. El cálculo es práctico
si aproximamos la distribución continua de los precios finales de la mercancía con una distribución discreta.
Para hacerlo, dividimos la distribución del precio final de la mercancía en n incrementos iguales de Sinc ,
donde
S  S min
Sinc  max (12.19)
n 1
Luego comenzamos por el precio más bajo de la mercancía y asumimos que, durante el primer
intervalo de S min  0.5Sinc a Smin  0.5Sinc , el precio de la mercancía es S min . Más generalmente, el precio de
la mercancía se asume como Si durante el intervalo i  ésimo , el cual tiene un rango de Si ,T  0.5Sinc ,
donde
Si ,T  S min  (i  1) Sinc (12.20)
La probabilidad que el precio final de la mercancía caiga en este rango
~
Pr[ Si ,T  0.5Sinc  ST  Si ,T  0.5Sinc ]  N (d l ,i )  N (d u ,i ) , (12.21)
ln[S /( Si ,T  0.5Sinc )]  (b  0.5 2 )T ln[S /( Si ,T  0.5Sinc )]  (b  0.5 2 )T

donde d l ,i  y d u ,i 
 T  T
Por lo tanto, el precio final de la mercancía se calcula como

~
E ( ST )  i1[ N (d l ,i )  N (d u ,i )]Si ,T (12.22)
n
EJEMPLO 12.2
Calcular el precio esperado de la mercancía en tres meses, asumiendo que el precio actual de la mercancía es
$50, la tasa de costo de acarreo es 4 porciento y la tasa de volatilidad es 32 porciento.
El primer paso es calcular el rango de precio mínimo y máximo utilizando (12.18):

2
S min  50e[ 0.040.5( 0.32 ) ]0.254( 0.32 ) 0.25  26.2909
2
S max  50e[ 0.040.5( 0.32 ) ]0.25 4 ( 0.32 ) 0.25
 94.5589 .
El siguiente paso es dividir el rango de los precios de la mercancía en intervalos de igual espacio. Eligiendo
n  11 , el tamaño de cada intervalo es
94.5589  26.2909
Sinc   6.8268
11  1
El punto medio en cada intervalo se asume como Si ,T  26.2909  6.8268(i  1) , cuyos valores son reportados
en la segunda columna de la Tabla 12.1.
Los puntos finales de cada intervalo entonces son definidos como Sl ,i  Si  0.5Sinc y Su ,i  Si  0.5Sinc para
i  1,, n . Basados en los valores de punto final del intervalo, los límites de la integral normal unitaria son
calculados y reportados como la tercera y cuarta columna de la Tabla 12.1. Basados en estos límites, la
probabilidad que el precio final de la mercancía caiga en el intervalo i  ésimo se calcula utilizando (12.21) y
se reportan en la quinta columna.
La última columna contiene el producto del precio final de la mercancía y su probabilidad respectiva.
Sumando los valores reportados en la última columna, encontramos que el precio final de la mercancía es
~
E ( ST )  i 1[ N (d l ,i )  N (d u ,i )]Si ,T  $50.5013
11
Noten que este valor corresponde muy estrechamente con el verdadero precio final de la mercancía, por el
Capítulo 11 sabemos que S  $50e 0.04 ( 0.25)  50.5025
La pequeña discrepancia se debe a que el método numérico para calcular el precio final esperado de
la mercancía es sólo una aproximación, aunque una muy exacta en esta ilustración. Se puede obtener una
mayor exactitud estableciendo n en un valor mayor o expandiendo el rango posible de los precios finales de
la mercancía considerados.
Extendiendo este enfoque para calcular la ganancia final de una cartera de opciones es un método
directo: simplemente reemplazando el precio final de la mercancía Si ,T en (12.22) con una ganancia de
cartera de opciones, dado un precio de la mercancía Si ,T , esto es,
E (~T )  i1[ N (d l ,i )  N (d u ,i )] ( Si ,T )  $50.5013 (12.23)
11
En la última sección se presentaron las funciones de ganancia  () para una amplia variedad de estrategias.
EJEMPLO 12.3
Calcular la ganancia final esperada de una call at-the-money, asumiendo que el precio actual de la mercancía
es $50, la tasa de costo de acarreo es 4 porciento y la tasa de volatilidad es 32 porciento. La tasa libre de
riesgo se asume como 6 porciento y el precio actual de la call es $3.410.
Todos los pasos en este ejemplo son los mismos que los del Ejemplo 12.2, excepto que en lugar de
multiplicar la probabilidad por el precio final de la mercancía en el intervalo, multiplicamos la probabilidad
por la ganancia de la call condicionada al precio final de la mercancía, como se muestra en la Tabla 12.2. La
ganancia es la diferencia entre el valor de ejercicio de la call y el precio inicial de la call ajustado por el
interés. Noten que la ganancia final esperada de la cartera de la call es aproximadamente $0.1937, lo cual
parece indicar un error en el precio.
El valor teórico de esta call utilizando la valuación (11.25) es $3.410, lo mismo que el valor de la put,
lo que significa que no hay error. La ganancia positiva surge de la aproximación implícita en la Tabla 12.2 y
el hecho que la ganancia de la call es una función no lineal del precio final de la mercancía. Para rectificar
este problema, debemos tener cuidado en establecer n mayor. Con un número mayor de pasos, la
discrepancia se reducirá. Por ejemplo si n  500 la ganancia final esperada es 0.0010 –un error de
aproximación de un décimo de centavo.
Tabla 12.1 Cálculo del precio final esperado de la mercancía, utilizando el enfoque de la
distribución discreta del precio de la mercancía con espacios iguales: S  50 b  0.04 T  0.25 , y
  0.32 .
(1) Límite Integral Límite Integral
(2) (1)
Intervalo Precio de la Inferior Superior
N (d l ,i )  N (du ,i )
108 Por
No. Mercancía d l ,i d u ,i
107 (2)
1 26.2909 4.8692 3.2371 0.0006 0.0159
2 33.1177 3.2371 1.9441 0.0253 0.8391
3 39.9445 1.9441 0.8733 0.1653 6.6028
4 46.7713 0.8733 .-0.0405 0.3249 15.1975
5 53.5981 -0.0406 -0.8377 0.2827 15.1536
6 60.4249 -0.8377 -1.5446 0.1399 8.4521
7 67.2517 -1.5446 -2.1796 0.0466 3.1326
8 74.0785 -2.1796 -2.7559 0.0117 0.8681
9 80.9053 -2.7559 -3.2836 0.0024 0.1953
10 87.7321 -3.2836 -3.7702 0.0004 0.0378
11 94.5589 -3.7702 -4.2216 0.0001 0.0066
~
E ( S T )  50.5013
12.3 REPLICACIÓN DE OPCIONES A LARGO PLAZO

Ocasionalmente, los administradores de cartera quieren comprar o vender opciones a largo plazo, pero estas
opciones no son cotizadas o los mercados de las opciones no son muy activos. En estos casos, es posible
imitar una opción a largo plazo con una cartera que consiste de opciones a corto plazo y un activo libre de
riesgo a corto plazo, como los Pagarés del Tesoro109. Las herramientas necesarias para llevar a cabo este
duplicación son los procedimientos de ganancia esperada de la sección anterior, junto con una regresión
lineal múltiple.
El enfoque es simple. Primero, como en la sección anterior, se encuentra un rango posible de precios
de la mercancía al final de la vida de las opciones a corto plazo, t . Utilizando las expresiones (12.18a) y
(12.18b), se identifica un rango que abarque el 99.994 porciento de la distribución de la probabilidad en t .
Segundo, se dividel rango en n incrementos iguales utilizando (12.19), y se identifican los precios de la
mercancía, Si ,t , en el punto medio de cada intervalo, utilizando (12.20). Tercero, se encuentra la
probabilidad que el precio final caiga dentro del intervalo i  ésimo , utilizando (12.21). Hasta ahora, todo es
igual que en la sección anterior.
Sobre la base de los precios de la mercancía creados en el segundo paso del último párrafo,
Si ,t , i  1,, n , calcular los valores del valor de la opción a largo plazo, VLT ( Si ,t ) , así como los valores
107
Si ,T es el precio final de la mercancía en el punto medio del intervalo i  ésimo .
108
N (dl ,i )  N (d u ,i ) es la probabilidad que el precio final de la mercancía caiga en el intervalo i  ésimo .
109
El rebalanceo dinámico de una cartera que consiste de la mercancía y pagarés del tesoro es otra manera de imitar o duplicar una
opción a largo plazo. Discutimos esta posibilidad en el Capítulo 14 bajo el título “Seguro de una Cartera Dinámica”.
finales de todas las opciones m a corto plazo, VSTj ( Si ,t ) , que se asumen como disponibles, j  1,, m . Se
utiliza los valores de la opción a largo plazo como la variable dependiente y los valores de la opción a corto
plazo como las variables independientes, y se realiza una regresión que minimiza la suma de los errores
2
cuadrados, Mini1 pi [VLT ( Si ,t )  b0   j 1 b j [VST , j ( Si ,t )]
n m
Los coeficientes de regresión estimados, bˆ j , j  1,, m , son las cantidades de las inversiones en las opciones
a corto plazo. El término de intersección estimado, b0 es la cantidad invertida en el activo libre de riesgo. Se
puede chequear lo bien que está funcionando la técnica comparando el valor actual de la cartera de opciones
a corto plazo con el valor teórico de la opción a largo plazo.110
Tabla 12.2 Cálculo de la ganancia final esperada de una posición call larga, cuyo precio es
c  3.410 . Los parámetros de precio son: S  50 , X  50 , b  0.04 , T  0.25 , r  0.06 , y
  0.32 .
(2)
Intervalo Precio de la (1)
112 Ganancia (1) Por (2)
No. Mercancía111 Si ,T N (d l ,i )  N (du ,i )
 113 i,T
1 26.2909 0.0006 -3.4615 -0.0021

2 33.1177 0.0253 -3.4615 -0.0877
3 39.9445 0.1653 -3.4615 -0-5722
4 46.7713 0.3249 -3.4615 -1.1248
5 53.5981 0.2827 0.1366 0.0386
6 60.4249 0.1399 6.9634 0.9740
7 67.2517 0.0466 13.7902 0.6424
8 74.0785 0.0117 20.6170 0.2416
9 80-9053 0.0024 27.4438 0.0662
10 87.7321 0.0004 34.2706 0.0148
11 94.5589 0.0001 41.0974 0.0028
~
E ( i ,T )  0.1937
EJEMPLO 12.4
Asuman un inversionista que posee una cartera de mercancía y quiere comprar una put europea con un precio
de ejercicio de 100 y un vencimiento de un año. El precio actual de la mercancía es 100, la tasa de costo de
acarreo es 4 porciento y la volatilidad es 32 porciento. La tasa de interés libre de riesgo es 6 porciento. El
precio teórico de esta opción es $10.3887 sobre la base de la ecuación de valuación de opciones europeas
(11.28). Sin embargo, no existe este tipo de opción a largo plazo.
El inversionista está considerando comprar una cartera de opciones put a tres meses que se pueda
utilizar para duplicar la opción a un año durante los próximos tres meses. En tres meses, se puede establecer
110
Choie y Novomestsky (1989) señalan que si el valor final de la cartera de la opción a corto plazo corresponde al valor de la
opción a largo plazo para todos los niveles del precio de la mercancía en el momento t, entonces en la ausencia de oportunidades
de arbitraje libre de costo en el mercado, el valor actual de la cartera de la opción a corto plazo es igual al valor actual de la opción
a largo plazo.
111
Si ,T es el precio final de la mercancía en el punto medio del intervalo i  esim o .
112
N ( d l ,i )  N ( d u ,i ) es la probabilidad que el precio final de la mercancía caiga en el intervalo i  esim o .
113
 i ,T  max(0, Si ,T  X )  ce rT
una nueva posición a corto plazo para imitar la entonces opción de nueve meses.114 Están disponibles siete
opciones a tres meses, la composición de la cartera es
Precio de Ejercicio Precio de la Opción

85 1.0438
90 2.0681
95 3.6432
100 5.8302
105 8.6265
110 11.9752
115 15.7855
Estableciendo el número de incrementos, n , en 300, se aplica el procedimiento de duplicación arriba

descrito. Primero se calculan los valores de la opción a largo plazo en los tres meses para los posibles valores
del subyacente en los meses. Estos valores son regresionados contra los posibles valores al vencimiento de
las opciones a corto plazo. Utilizando los coeficientes de regresión estimados, la composición de la cartera es
Precio de (1) (2)

(1) Por (2)
Ejercicio Precio de la Opción Coeficiente Estimado
85 1.0438 0.2286 0.2386
90 2.0681 0.0018 0.0036
95 3.6432 0.1004 0.3657
100 5.8302 0.0362 0.2109
105 8.6265 0.1730 1.4924
110 11.9752 -0.4009 -4.8009
115 15.7855 -0.6136 9.6860
T-bill 0.9851 3.2406 3.1923
Total 10.3887
Con la excepción de recortar la put de 110, todas las otras puts son compradas. La suma de los pesos de la
cartera por el precio de los activos, 10.3887 es igual al precio de la put a largo plazo. (El precio del pagaré
del Tesoro se asume como e 0.06( 0.25)  0.9851 . En la Figura 12.27 podemos ver una comparación del valor de
la put a largo plazo real con el valor simulado de la put. Noten lo cercanos que son los valores hasta que el
precio de la mercancía se vuelve muy alto.
Este procedimiento puede ser refinado para tomar en cuenta las restricciones de no-negatividad,
liquidez del mercado y los errores en precios observados de las opciones a corto plazo. También son posibles
cambios en la volatilidad a lo largo de la vida de la opción a largo plazo.115 Nuestro enfoque asume que las
compras y las ventas de opciones a corto plazo se permiten libremente en cualquier cantidad demandada.
También asumimos que los precios de la opción están de acuerdo con las ecuaciones de valuación de
opciones europeas (11.25) y (11.28) y que la tasa de volatilidad es constante durante la vida de la opción a
largo plazo.
Los administradores de cartera interesados en las opciones con fechas a largo plazo pueden crearlas
como se indicó anteriormente o, como se da con más frecuencia, comprarlas a bancos de inversión en una
114
En la práctica, el uso de opciones a corto plazo con más de una fecha de vencimiento (por ejemplo, opciones a tres meses y seis
meses) y/o el refinanciamiento de posiciones de opciones a corto plazo antes de su vencimiento puede proporcionar una
duplicación más efectiva de la posición de la opción a largo plazo.
115
Ver, por ejemplo, Jamshidian y Zhu (1990).
transacción fuera de bolsa. El banco de inversión vende la opción por su cuenta y cubre su posición tomando
una posición de compensación en la cartera de imitación de opciones a corto plazo.
Figura 12.27 Opciones Simuladas versus Precio de la Opción a Largo Plazo
Precio de la Opción
Opciones a largo plazo

simuladas utilizando opciones a
corto plazo
Precio de la Opción a Largo Plazo
Mercancía S
12.4 ADMINISTRACIÓN DINÁMICA DE RIESGO DE CARTERA

Hasta este punto en el capítulo, las posiciones de la opción se han mantenido hasta el vencimiento. En esta
sección, tratamos la administración dinámica de riesgo, esto es, una administración de riesgo de cartera que
toma en cuenta los movimientos en el precio del subyacente en el corto plazo, los cambios de la volatilidad
en el corto plazo y la erosión natural de la prima de la opción conforme disminuye el tiempo hasta el
vencimiento. En este contexto, nos basamos particularmente en las derivadas parciales de las fórmulas de
precio de opciones europeas que derivamos en el Capítulo 11. Mostraremos que los deltas, gammas, etas,
thetas y vegas de la opción son valiosas en el manejo del retorno esperado y el riesgo de una cartera de
opciones y el subyacente.
Riesgo y Retorno Esperado

Para comenzar, es útil entender claramente las características del riesgo/retorno-esperado de las posiciones
de la opción. En el Capítulo 11, mostramos que el beta de una opción es igual al valor absoluto de la
elasticidad del precio de la opción con respecto al precio de la mercancía por el beta del subyacente, esto es
 c  c  S , y  p   p  S . También hicimos notar que la volatilidad de la opción es igual al valor absoluto de
la elasticidad del precio de la opción con respecto al precio de la mercancía por la volatilidad del subyacente,
esto es,  c | c |  S y  p |  p |  S . Recuerden que la elasticidad depende de los precios de la mercancía y
la opción y de otras variables en la fórmula del precio de la opción, como la volatilidad, tiempo hasta el
vencimiento y otros. Como resultado, las características de riesgo y retorno de las posiciones
opción/mercancía cambian a lo largo del tiempo conforme cambian estas variables. Si el riesgo de una
posición permanece sin cambio a lo largo del tiempo, la posición debe ser re-equilibrada inapropiadamente.
Ahora examinaremos las características riesgo/retorno de una cartera de opciones y el subyacente con
la ayuda de una ilustración numérica simple. Asumimos que el precio de la mercancía es 50, la tasa de
retorno esperado de la mercancía es 16 porciento, el beta de la mercancía es 1.20, y la volatilidad del retorno
de la mercancía es 40 porciento. Nosotros asumimos una tasa de costo de acarreo de 4 porciento y una tasa
libre de riesgo de 6 porciento. Existen calls y puts europeas a tres meses con precios de ejercicio de 45, 50 y
55, y todas tienen precios iguales a los valores teóricos, dados por las ecuaciones del capítulo anterior.
Figura 12.28a Relación Entre Retorno y Beta Esperado
Expected Return and Beta
100
In-the-money call
80
60
Out
Expected Return
In-the-money put 40 At
20
0 Commodity
-15 -10 -5 -20 0 5 10 15
At
-40
-60 Riskless Asset
Out
-80
Beta
Enfocándose primero en la medida de riesgo beta, podemos encontrar los atributos riesgo/retorno-esperado
de las opciones al encontrar primero sus betas respectivos, y luego encontrar los retornos esperados en
equilibrio basados en sus betas. Por ejemplo, el beta para la call in-the-money, que tiene una elasticidad de
c  5.289 , se puede calcular como  c  c  S = 5.289(l.20) = 6.35.
Asumiendo que el mercado de capital está en equilibrio, el retorno esperado sobre la mercancía es
E ( RS )  r  [ E ( RM )  r ] S .
Sustituyendo E ( RS ) , el retorno esperado de la mercancía, y por r , la tasa de interés libre de riesgo,
encontramos que [ E ( RM )  r ] S es igual a 0.10. Para encontrar el retorno esperado para la call in-the-
money, utilizamos una vez más la línea del mercado de valores del modelo de precio del activo de capital:
E ( Rc )  r  [ E ( RM )  r ] c  r  [ E ( RM )  r ]c  S  0.06  0.10c  0.06  0.10(5.289)  58.89%
Utilizando un procedimiento similar para las opciones restantes, encontramos los siguientes retornos
esperados y betas:
Opción Precio Delta Elasticidad Retorno Esperado Beta

Call 45 7.061 0.747 5.289 58.89 6.35
Call 50 4.196 0.557 6.636 72.36 7.96
Call 55 2.294 0.370 8.069 86.69 9.68
Put 45 1.640 -0.248 -7.559 -69.59 -9.07
Put 50 3.701 -0.438 -5.920 -53.20 -7.10
Put 55 6.724 -0.625 -4.645 -40.45 -5.57
Mercancía 1 1 16.00 1.20
Activo Libre de Riesgo 0 0 6.00 0.00
La Figura 12.28a ilustra estos resultados. La relación beta/retorno-esperado descrita en la Figura 12.28a es
asombrosa. Los betas y retornos esperados de las opciones son drásticamente diferentes del retorno-esperado
y beta del subyacente. Las posiciones de la call larga, por ejemplo, tienen retornos esperados y betas muy
altos –de hecho, varias veces mayores que los del subyacente. La ilustración también muestra que el retorno-
esperado y el beta de la call largo se incrementan conforme la call se hace cada vez más out-of-the-money.
Por otro lado, vemos que las puts generalmente tienen retornos esperados negativos y betas negativos, y
mientras más out-of-the-money esté la put, menor (más negativo) es su retorno y su beta esperado.
Los administradores de cartera también están interesados en conocer el nivel de la volatilidad del
retorno. Como vimos anteriormente, la volatilidad del retorno de la opción es simplemente la elasticidad del
precio de la opción con respecto al precio de la mercancía por la volatilidad del retorno del subyacente. En la
ilustración, la volatilidad del retorno de la mercancía es 40 porciento. Por lo tanto las volatilidades de las
opciones son
Opción Elasticidad Volatilidad del Retorno
Call 45 5.289 211.56
Call 50 6.636 265.44
Call 55 8.069 322.76
Put 45 -7.559 302.36
Put 50 -5.920 236.80
Put 55 -4.645 185.80
Mercancía 1 40.00
Activo Libre de Riesgo 0 0
La Figura 11.28b ilustra estos resultados. El grado de riesgo extremo de las opciones se confirma aún más
con estos valores. Donde la volatilidad del retorno del subyacente es 40 porciento, las volatilidades del
retorno de la opción exceden, en algunos casos, varios cientos porciento.
Sin embargo, combinando opciones con una posición en el subyacente puede ayudar a reducir el
riesgo. Por ejemplo, una estrategia de venta de call cubierta (es decir, escribir una call contra una mercancía
subyacente) o una estrategia put protectora (es decir, comprar una put contra el subyacente) reduce el riesgo
de la posición global. El retorno esperado, el beta y la volatilidad del retorno de una cartera que consiste de
una opción y una mercancía subyacente pueden calcularse utilizando las siguientes ecuaciones.
E ( R p )  X S E ( RS )  (1  X S ) E ( Ro ) , (12.25)
 p  X S  S  (1  X S )  o , (12.26)
y
 p  X 2 2  (1  X S ) 2  2  2d o X (1  X S )   o
S S o S S
(12.27)
dondel subíndice o indica opción y la variable indicadora d o  1 es para calls y d o  1 es para puts (es
decir, los retornos de la call [put] están perfecta y positivamente [negativamente] correlacionados con los
retornos de la mercancía). El peso de X S es la proporción de dólares S invertidos directamente en la
mercancía, esto es,
S  noOo
XS  (12.28)
S
donde no es el número de opciones compradas (es decir, un valor positivo de no indica que las opciones son
compradas y un valor negativo indica que las opciones son vendidas) y Oo es el valor de mercado de cada
opción. Noten que cuando no es igual a cero, todo el valor de la cartera es invertido en la mercancía. El valor
1  X S es la proporción de la inversión original en las opciones.
Para reforzar estos procedimientos, reconsideren la ilustración anterior y asuman que se crea una call
cubierta al vender la call in-the-money contra una posición larga en la mercancía. Los ingresos de la venta de
la call son invertidos en la mercancía de manera que la inversión total en la mercancía es
50-(-1)(7.061)=57.061.
Por lo tanto, la proporción de la inversión original en el subyacente es X S  57.061 / 50  1.141 . La

proporción del valor de la cartera invertido en la call es (1  X S )  0.141 . Conociendo los pesos, podemos
calcular el retorno esperado, el beta y la volatilidad de la posición call cubierta utilizando las expresiones
(12.25), (12.26) y (12.27):
E ( RP )  1.141(16.00) - 0.141(58.89)  9.94% ;

 P  1.141(l.20) - 0.141(6.35)  0.47 ,
y
 P  1.1412 (0.40)2  (-0.141)2 (2.1156)2  2(1.141)(-0.141)(0.40)(2.1156)  15.81% .
Figura 12.28b Relación Entre el Retorno Esperado y la Volatilidad
Expected Return and Volatility
100
80 Out
Commodity
60
Expected Return
At
40
Riskless Asset In-the-money call
20
0
-20 0 100 200 300 400
-40 Out
-60 In-the-money put
At
-80
Volatility
En otras palabras, el escribir la call in-the-money contra el subyacente reduce el retorno esperado y el riesgo
de la cartera subyacente. De hecho, todos y cada una de las ventas de calls cubiertos y todas y cada una de
las compras de put protectores compartirán estos atributos. Para nuestra ilustración, las características de la
mercancía y las seis cartera mercancía /opción son
Inversión en XS Retorno Volatilidad

Opción Beta
Mercancía Esperado del Retorno
Ninguna opción 50.000 1.000 16.00 1.20 40.00
Venta call 45 57.061 1.141 9.94 0.47 15.81
Venta call 50 54.196 1.084 11.27 0.63 21.06
Venta call 55 52.294 1.046 12.75 0.81 26.99
Compra 45 48.360 0.967 13.18 0.86 28.69
Compra put 50 46.299 0.926 10.88 0.59 19.52
Compra put 55 43.276 0.866 8.44 0.29 9.74
Los resultados de la tabla son trazados en las Figuras 12.29a y 12.29b.
Figura 12.29a Relación Entre Retorno Esperado y beta

Commodity
18 6
16 4
2
Expected Return
14
12
10
8 5
6 3
4 1
2 Riskless Asset
0
0 0.5 1
Beta
1) In-the-money covered call. 2) In-the-money protective Put

3) At-the-money covered call. 4) At-the-money protective Put
5) Out-of-the-money covered call. 6) Out-of-the-money protective Put
Figura 12.29b Relación Entre Retorno Esperado y Volatilidad

Commodity
18
16 4
2 6
Expected Return
14
12
10 5
8
6 3
4 1
2
Riskless Asset
0
0 10 20 30 40
Volatility
1) In-the-money covered call. 2) In-the-money protective Put

3) At-the-money covered call. 4) At-the-money protective Put
5) Out-of-the-money covered call. 6) Out-of-the-money protective Put
Estas cantidades muestran claramente que las ventas de la call cubierto y las compras de put protector sirven
para reducir el retorno esperado y el riesgo de la cartera.116 Para los calls cubiertos, la reducción
retorno/riesgo se hace mayor mientras más in-the-money esté la opción. Esto es simplemente porque el que
vende la opción está dispuesto a aceptar más efectivo (es decir, prima de la opción) a cambio del potencial de
movimientos hacia arriba en el precio de la mercancía. Esta actividad es completamente análoga a retirar la
inversión de la mercancía y e invertir en el activo libre de riesgo. De hecho, dado que la call 45 tiene un delta
de .747, la cartera de la call cubierto 45 tiene un valor delta neto de 1.141-0.747=0.394. Si creamos una
cartera que consiste de 0.394 en la mercancía y 0.606 en el activo libre de riesgo, el retorno esperado, el beta
y la volatilidad del retorno son:
E ( RP )  1.141(16.00) - 0.141(58.89)  9.94% ;
116
Recuerden que el retorno esperado y el beta de las posiciones de la opción se mantienen para la siguiente instancia en la cual las
variables subyacentes no cambian mucho.
 P  1.141(l.20) - 0.141(6.35)  0.47 , y
 P  1.1412 (0.40) 2  (-0.141)2 (2.1156) 2  2(1.141)(-0.141)(0.40)(2.1156)  15.81% .
los mismos atributos retorno/riesgo de la venta de la call 45 cubierto.

Esto no quiere decir que la venta de calls contra el subyacente siempre reduzca el riesgo. Si se venden
demasiadas calls, el riesgo de la cartera puede incrementarse. Consideren una venta de call con ratio 4:1. La
inversión total en la mercancía es
50+4(7.061)= 78.244.
Por lo tanto, la proporción de la inversión original en el subyacente es, X S  78.244/50  1.565 . La

proporción del valor de la cartera invertido en la call es (1  X S )  - 0.565 . El retorno esperado sobre esta
cartera es E ( RP )  1.565(16.00) - 0.565(58.90)  - 8.24%
y la volatilidad es  P  (1.565) 2 (0.40) 2  (2.1156) 2 (0.565) 2 - 2(1.565)(0.565)(0.40)(2.1156)  56.93%
Donde la volatilidad de una cartera que consiste exclusivamente de la mercancía es 40 porciento, la

volatilidad de una venta de ratio 4:1 es 56.93 porciento. En esta situación, estamos sobre protegidos.
La Figura 12.30 ayuda a aclarar este punto.
Figura 12.30 Relación Entre el Retorno Esperado y la Volatilidad
40
30 1
Expected Return
20 3 2
5 6 6
10
4
0
0 20 40 60 80 100
-10
-20
Volatility
1) Commodity 2) 1:1 Call Write

3) Riskless Hedge 4) Zero Return Hedge
5) 4:1 Call Write 6) Long Commodity/ Long call option Portfolios
En la Figura, vemos la mercancía con un retorno esperado de 16 porciento y una volatilidad de 40 porciento.
El punto llamado “venta call 1:1” muestra que el retorno esperado y la volatilidad de la cartera se reducen
cuando se escribe una sola call 45 contra la mercancía. Conforme se escriben o venden más calls, el retorno
esperado y la volatilidad continúan disminuyendo hasta que finalmente, cuando se escriben 1.65 calls contra
la mercancía, la cartera está libre de riesgo.117 Más allá de este número, si se escriben más calls, el retorno
esperado continúa disminuyendo y la volatilidad se incrementa.
Eventualmente, el retorno esperado de la cartera se vuelve negativo, y, si el número de calls escritos
continúa aumentando, la volatilidad comienza a exceder el 40 porciento. La venta de la call 4:1 se mantiene
aislada. La cartera, por ejemplo, tiene una mayor volatilidad que el número crítico de la mercancía de los
calls escritos contra la mercancía para generar una volatilidad de 40 porciento y es 3.30, exactamente el
doble del número de calls para generar la cobertura libre de riesgo.118
Manejando Cambios Inesperados en el Precio de la Mercancía

El riesgo de una cartera de opciones está sujeto a cambios conforme cambia el precio del subyacente. En
consecuencia, un administrador de cartera no sólo necesita identificar el riesgo actual de la cartera, como se
muestran en la sección precedente, sino también debe manejar la cartera para minimizar los efectos de
cambios inesperados en el precio de la mercancía. El conocimiento de los deltas y gammas de la opción
proporciona las herramientas necesarias para inmunizar las carteras contra movimientos adversos en el
precio del subyacente.
En el Capítulo 11, desarrollamos expresiones para las derivadas parciales de las opciones europeas
individuales, basadas en las ecuaciones de valuación (11.25) y (11.28). Por ejemplo, el delta es la derivada
parcial del precio de la opción con respecto al cambio en el precio del subyacente. La pregunta que surge en
el manejo de la cartera de la opción es cómo cambia el valor de la cartera como resultado de un cambio en el
precio de la mercancía. Para entender la respuesta a esta pregunta, primero desarrollamos una respuesta
simple e intuitiva para el valor de la cartera. Asumimos que la cartera consiste de N posiciones de opción
diferentes, una posición del subyacente y una inversión en el activo libre de riesgo. Cada opción tiene ni
contratos al precio actual de Oi . Sumando las posiciones y agregando nS unidades de la mercancía al precio
S y el activo libre de riesgo, B , el valor de la cartera es
V  i1 ni Oi  nS S  B
N
(12.29)
La derivada parcial del valor de cartera con respecto a un cambio en los determinantes de la opción, k , es
V O S B
 i1 ni i  nS
N
 (12.30)
k k k k
En otras palabras, para encontrar el cambio en el valor de la cartera resultante de un cambio en k , un

término genérico que representa variables, simplemente calculamos cómo cambia el valor de cada opción al
cambiar k , multiplicamos por el número de contratos de esa opción y luego sumamos todas las posiciones
de la opción. Las posiciones de la mercancía y el activo libre de riesgo también pueden afectar el valor de la
cartera. El mismo resultado se mantiene cuando examinamos la segunda derivada parcial:
117
Antes ilustramos que una posición call cubierta era análoga a una cartera que consiste de alguna fortuna en la mercancía y
alguna fortuna en el activo libre de riesgo. La cantidad invertida en la mercancía es ( S  nc c  nc S  c ) . Si establecemos este valor
igual a cero (es decir, toda el valor se invierte en el activo libre de riesgo) y despejamos para nc , obtenemos nc  1/( c  c / S ) .
Sustituyendo los valores del ejemplo, nc  1/(0.747 7.061/50)  1.65 .
118
En el último pie de página, utilizamos el hecho que un cal cubierto es como una cartera que consiste de alguna fortuna en la
mercancía y alguna fortuna en el activo libre de riesgo para poder deducir la composición de la cobertura libre de riesgo. La
cantidad invertida en la mercancía es ( S  nc c  nc S  c ) . Si establecemos este valor igual a S (es decir, una venta corta de la
mercancía subyacente) y despejamos pata nc , obtenemos nc  2 /( c  c / S ) . Sustituyendo en los valores del ejemplo,
nc  2/(0.747 - 7.061/50)  3.30 .
 2V  2Oi 2S 2B

i1 i k 2 S k 2  k 2
N
 n  n (12.31)
k 2
Ahora, regresemos al problema de manejar los cambios en el precio de la mercancía. El cambio en el valor
de la cartera con respecto a un cambio en el precio de la mercancía (es decir, el delta de la cartera) es
V  i 1 ni  o  nS (12.31)
N
Noten que por premisa, el valor del activo de riesgo no cambio con los cambios en el precio de la mercancía.
Para inmunizar esta cartera contra los cambios en el precio de la mercancía, simplemente calculamos V y
luego tomamos una posición en opciones o el subyacente que forma el valor cero del delta de la cartera.
EJEMPLO 12.5
Asuman un creador de mercado de opciones de futuros tiene posiciones largas de 150 calls con un precio de
ejercicio de 45 y un tiempo hasta el vencimiento de dos meses, 200 puts con un precio de ejercicio de 50 y
tiempo hasta el vencimiento de tres meses, y 225 calls con un precio de ejercicio de 55 y un tiempo hasta
vencimiento de tres meses. En lugar de enfrentar el riesgo que el precio de los futuros subyacentes pueda
moverse significativamente de la noche a la mañana, él decide cubrir la posición utilizando futuros o calls
con un precio de ejercicio de 50 y un tiempo hasta el vencimiento de tres meses. Comparen la efectividad de
utilizar futuros y la call 50 al crear una cobertura neutral al riesgo. Asuman que el precio actual de los futuros
es $50, la call 50 tiene un precio de $2.455 y tiene un delta de 0.5171, la tasa de interés libre de riesgo es 6
porciento y la volatilidad es 25 porciento.
Tipo de Precio de Tiempo Hasta el Precio de la

Cantidad Delta
Opción Ejercicio Vencimiento Opción
150 Call 45 0.16667 5.325 0.852
200 Put 50 0.25 2.455 -0.468
225 Call 55 0.25 0.828 0.238
Por lo tanto, el valor de la cartera es V  150(5.325)  200(2.455)  225(0.828)  1, 476.05 ,
y el delta de la cartera es V  150(0.852)  200(-0.468)  225(0.238)  87.75 .
El delta de la cartera de 87.75 nos dice que mantener esta cartera es como mantener una posición larga en
87.75 contratos de futuros. Para crear una cartera neutral al delta, podemos (a) vender 87.75 contratos de
futuros o (b) vender 87.75/0.5171=169.70 calls. La Figura 12.31 muestra la efectividad de cada cobertura.
La Figura 12.31 demuestra que tanto la cobertura de futuros neutral al delta como la cobertura de
opción neutral al delta reducen el rango de los valores de cartera posibles. La cartera sin cobertura tiene un
rango de valor entre 1400 y 3600 sobre el rango de los precios de la mercancía –la cobertura de futuros tiene
entre 1475 y 3000 y la cobertura de opciones entre 1475 y 2500. Claramente, la cobertura de opciones es la
más efectiva.
La razón por la que la cobertura de opciones resulta ser la más efectiva es que tiene que ver con
gamma –el cambio en delta conforme cambia el precio de la mercancía. Conforme cambia el precio de la
mercancía, el valor del delta de la cartera de opciones cambia. De hecho, el gamma de la cartera de la opción
es 30.08, como se muestra a continuación. El contrato de futuros tiene un gamma igual a cero, de manera que
la cobertura de futuros neutral al delta aún tiene un gamma de 30.08. Por otro lado, las 169.70 calls que
vendimos tiene un gamma de –169.70(0.062758)=-10.65, de manera que el gamma de la cobertura de la
opción neutral al delta es 19.43.
No obstante, la reducción del gamma con la cobertura de opciones es incidental en este caso, de
manera que ahora ilustraremos cómo tomar en cuenta tanto el delta como el gamma en la cobertura de una
cartera contra los movimientos en el precio.
EJEMPLO 12.26
Una vez más, estamos considerando al creador de mercado descrito en el Ejemplo 12.5. Su posición de
cartera es
Cantidad Delta Gamma
150 Call 45 0.16667 5.325 0.852 0.04304
200 Put 50 0.25 2.455 -0-468 0.06276
225 Call 55 0.25 0.828 0.238 0.04922
Como en el ejemplo anterior, el valor de la cartera es 1476.05 y el delta de la cartera es 87.75. El gamma de
la cartera es 87.75. El gamma de la cartera es
 V  150(0.04304)  200(0.06276)  225(0.04922)  30.08 .
Para cubrir tanto el riesgo del delta como del gamma se requieren por lo menos dos opciones (es decir, los
futuros tienen un gamma igual a cero, de manera que no son efectivos para ajustar el riesgo de gamma de la
cartera). Además de la call 50, la cual está disponible en el Ejemplo 12.5, también asumimos que está
disponible una put 55 a tres meses. Su precio es $7.754, su delta es -0.7468 y su gamma es –0.04922. La call
50 tiene un gamma de 0.06276.
Para calcular la cobertura óptima neutral al delta/gamma de estas dos opciones, resolvemos el
siguiente sistema de ecuaciones. Queremos que la cartera sea neutral al delta, de manera que
nc (0.5171)  n p (-0.7468)  - 87.75 .
Figura 12.31Valor de Cartera como una Función del Precio de la Mercancía
Portfolio Value
Unhedged Portfolio
Delta-hedged Portfolio –futures
Delta-hedged Portfolio –options
Commodity Price
También queremos que la cartera sea neutral al gamma, de manera que

nc (0.06276)  n p (0.04922)  - 30.08 .
Para despejar, podemos aislar nc de la primera ecuación, sustituir en la segunda y despejar n p . El valor de n p
es –138.93. Luego sustituimos para n p en la primera ecuación y encontramos que nc es –370.34.
El valor de la cartera neutral al delta y al gamma se traza en la Figura 12.32, junto con el valor de la
cartera no cubierta. Claramente, la cobertura es efectiva. Donde la cartera no cubierta varía en más de 2000
sobre el rango del precio de la mercancía mostrado, la cartera protegida parece variar en menos de 200.
Noten que en el Ejemplo 12.6, se asumen que sólo están disponibles dos opciones para establecer la
cobertura neutral al delta/neutral al gama. En la práctica, están disponibles muchas opciones con las cuales
establecer la cobertura. Se necesitan por lo menos dos opciones para ejecutar una cobertura, pero se pueden
utilizar más opciones. Algunas veces se utiliza la programación lineal para encontrar el conjunto de opciones
de costo mínimo que eliminará el riesgo del delta y gamma.
Figura 12.32 Valor de Cartera como una Función del Precio de la Mercancía
Portfolio Value
Unhedged Portfolio
Delta-hedged and Gamma-Hedged

Portfolio
Commodity Price
Manejando los Cambios Inesperados en la Volatilidad

Junto con el riesgo del precio de la mercancía, los comerciantes también se encuentran en una posición
donde sus carteras de opciones pueden sufrir grandes pérdidas si la volatilidad subyacente de las opciones
cambia. Por ejemplo, un creador de mercado puede estar corto en calls y puts y, aunque la posición es neutral
al delta, un incremento repentino en la volatilidad causaría que el creador de mercado incurra en pérdidas
significativas. Al igual que la cobertura delta mostrada arriba, el creador de mercado puede cubrir la
volatilidad mediante una cobertura vega.
EJEMPLO 12.7
Una vez más, estamos considerando al creador de mercado en los contratos de opciones de futuros descrito
en el Ejemplo 12.5. Sin embargo, la cartera de este creador de mercado es diferente. Él está corto 180 calls
50 a tres meses y 200 puts 50 a tres meses.

Cantidad Delta Vega
-180 Call 50 0.25 2.455 0.517 9.806
-200 Put 50 0.25 2.455 -0.468 9.806
Noten que las opciones de futuros at-the-money tienen el mismo precio. La cartera es claramente neutral al
delta, como se muestra a continuación:
V  -180(0.517)  - 200(-0.468)  0.54 .
Desafortunadamente, la exposición al vega es sustancial:

VegaV  - 180(9.806)  - 200(9.806)  - 3726.28 .
Figura 12.33 Valor de Cartera como una Función de la Volatilidad
Portfolio Value
Unhedged Portfolio
Vega-hedged Portfolio
Commodity Volatility( %)
Esto significa que si la volatilidad se incrementa de su nivel actual de 25 porciento a, digamos, 26 porciento,
el valor de la cartera disminuirá en 37.26.
Para cubrir este riesgo, asuman que está disponible la put 55 a tres meses del Ejemplo 12.6. La put 55
tiene un vega de 7.69. Para poder eliminar el riesgo de vega de la cartera debemos comprar
3,726.28
np   484.56 puts. La Figura 12.33 muestra la efectividad de este procedimiento.
7.69
Claramente, la cobertura del vega se muestra como efectiva para eliminar los efectos de los cambios en la
volatilidad. Noten, sin embargo, que la compra de puts 55 afecta drásticamente el delta de la cartera. Ahora
está en un nivel de 484.56(-0.7468)+0.54-361.33. Este ejemplo está diseñado sólo para mostrar cómo se
maneja el riesgo del vega. Obviamente, considerar simultáneamente el delta, el gamma y el vega puede ser
razonable; esto se pude realizar con tres o más opciones disponibles.
Manejando la Deterioro del Tiempo

Nuestra discusión final tiene que ver con el deterioro del tiempo. Las posiciones largas en las opciones
deterioran su valor a lo largo del tiempo conforme disminuye el prospecto de grandes movimientos en el
tiempo. La Figura 12.34 ilustra cómo una call at-the-money disminuye en valor conforme se acerca su fecha
de vencimiento.
La pendiente de la curva mostrada es el theta de la opción, esto es, el cambio en el valor de la opción
conforme cambia el tiempo al vencimiento. Primero la tasa de disminución es lenta. En los días finales antes
del vencimiento, la tasa es considerablemente más rápida. Una implicancia de esta figura es que el manejar el
deterioro del tiempo es más difícil conforme se acerca la fecha de vencimiento de la opción.
Para manejar el deterioro del tiempo, utilizamos los mismos procedimientos que utilizamos para las
otras derivadas parciales. En lugar de utilizar el delta, gamma o vega; encontramos una opción u opciones
para eliminar el theta de la cartera.
Figura 12.34 Deterioro del Tiempo de una Call At-the-money
Option Price
Days to Expiration
EJEMPLO 12.8
Consideren la cartera del creador del mercado de una opción de futuros del Ejemplo 12.7. ¿Cómo puede esta
persona eliminar el deterioro del tiempo en su cartera utilizando la call 50?
Las posiciones de la cartera

Cantidad Theta
150 Call 45 0.16667 5.325 3.043
200 Put 50 0.25 2.455 4.756
225 Call 55 0.25 0.828 7.690
Por lo tanto el theta de la cartera es   150(3.043)  200(4.756)  225(7.690)  3,137.90 ,
lo cual significa que, durante el siguiente día, el valor de la cartera disminuirá en 3,137.90(1/365)=8.597.
Se puede crear una cobertura theta utilizando la call 50. Su theta es 4.756. Para eliminar el deterioro del
tiempo, debemos vender 3,137.90/4.756=659.78 contratos.
En esta sección, hemos mostrado cómo se pueden utilizar los valores delta, gamma, vega y theta de
una cartera para manejar de manera efectiva el valor de una cartera de opciones. Aunque las ilustraciones se
enfocan típicamente en una sola dimensión a la vez, está claro que es necesario un mantenimiento diario de
las posiciones de la cartera.
12.5 RESUMEN
En este capítulo, hemos logrado cuatro cosas. Primero, hemos desarrollado y analizado más de cincuenta
estrategias de negociación de opciones. Cada estrategia se formó desde sus componentes básicos y se ilustró
con un diagrama de ganancia. Se proporcionaron los precios de las mercancías, las pérdidas máximas y
ganancias máximas. Segundo, mostramos que utilizando la premisa de distribución lognormal del precio de
la mercancía del Capítulo 11, podemos calcular las probabilidades de pérdidas y ganancias, así como la
ganancia esperada para cada estrategia de negociación. En la tercera sección, mostramos que se puede
utilizar un enfoque de regresión para crear opciones a largo plazo a partir de opciones a corto plazo.
Finalmente, hemos discutido el manejo del riesgo diario de las carteras de opciones.
13 COMMON STOCK OPTION CONTRACTS
13.1 COMMON STOCK OPTION MARKETS
13.2 PRICE BOUNDS AND ARBITRAGE RELATIONS

 Non-Dividend-Paying Stocks
 Dividend-Paying Stocks
13.3 VALUATION OF CALL OPTIONS ON STOCKS

 European Call Option on Non-Dividend-Paying Stocks
 European Call Option on Dividend-Paying Stocks
 American Call Option on Non-Dividend-Paying Stocks
 American Call Option on Dividend-Paying Stocks
 Dividend Spreads
13.4 VALUATION OF PUT OPTIONS ON STOCKS

 European Put Option on Non-Dividend-Paying Stocks
 European Put Option on Dividend-Paying Stocks
 American Put Option on Non-Dividend-Paying Stocks
 American Put Option on Dividend-Paying Stocks
13.5 RIGHTS AND WARRANTS
13.6 SUMMARY
APPENDIX: APPROXIMATION FOR THE CUMULATIVE BIVARIATE NORMAL DENSITY

FUNCTION
13 COMMON STOCK OPTION CONTRACTS
Common stock option contracts have been traded in the U.S. for many decades. Trading began in the over-
the-counter market more than fifty years ago. In April 1973, the Chicago Board Options Exchange (CBOE)
became the first organized secondary market in call options on sixteen NYSE common stocks. In 1977, the
CBOE introduced put options on stocks. Today both call and put options are traded on over five hundred
different stocks and on five exchanges. In addition to the CBOE, active secondary stock option markets exist
on the American Stock Exchange, the Pacific Coast Exchange, the Philadelphia Stock, Exchange and the
New York Stock Exchange.
This chapter focuses on stock options. In the first section, the stock option market is described. In the
second section, we adapt the arbitrage pricing, relations of Chapter 10 to stock option contracts. The
principles are modified somewhat to account for the fact that common stocks typically pay discrete dividends
during the option’s life. In section 3, we value European and American call options on dividend-paying,
stocks. Even though an American call option can, be exercised early, a valuation equation for this option
exists. For American put options on stocks, no valuation equation exists. In section 4, the binomial method is
used to approximate the value of American put options on dividend-paying, stocks. Although the application
is specific, the binomial method can be applied to valuation of virtually any type of option. This method is
applied again in chapter 15, for example, to value interest rate options. Finally in section 5, warrants used to
raise new capital are studied. Warrants are exercised, the company’s equity may be diluted, something that
must be incorporated into warrant valuation.
13.1 COMMON STOCK OPTION MARKETS

In the U.S., stock option contracts are written in denominations of one hundred shares, expire on the
Saturday after the third Friday of the contract month, and are American-style. A call option on a stock
represents the right to buy one hundred shares of the underlying stock, and a put option represents the right
to sell one hundred shares. Although the contract denomination is one hundred shares, the option prices are
quoted on a per-share basis. The exercise prices of stock options are in $5 increments above $25 and in $2.50
increments below.
To illustrate these conventions, consider the option prices reported in Table 13.1. The IBM call
option with an exercise price of 100 and a January maturity has a price of $3.625. To buy this contract, it
would cost $3.625 x 100 = $362.50. The implied terms of this contract are that the buyer has the right to buy
one hundred shares of IBM for a total amount of $10,000 at any time between November 13, 1991, and
January 17, 1992 (i.e., the third Friday of January).119
In reporting stock option prices, a number of conventions are used. In Table 13.1, a number of cells
of the table have the entry “ r ”. This means that the particular option did not trade on that day. Other cells
have the entry “ s ” implying that the particular option contract does not exist. Also, beneath the firm’s ticker
symbol, the closing stock price is reported. This is done for convenience, so the reader does not have to turn
to the stock market page to find the closing price of the underlying stock.
One final note about Table 13.1 is that the most actively traded stock options on the various
exchanges are listed. These are simply the option contracts with the greatest number of transactions on that
day. Interspersed among the stock options are usually some stock index options, the subject of the next
chapter. In the CBOE active option list, for example, nine of the ten most active options are written on the
S&P 100 index.
119
For practical purposes, assuming the option expires Friday seems prudent since both the stock market and the option market are
closed on Saturday.
13.2 PRICE BOUNDS AND ARBITRAGE RELATIONS

The lower price bounds and the put-call parity relations for stock options are presented for non-dividend- and
dividend-paying stocks. Without dividends, the arbitrage relations are straightforward since a common stock
is a commodity for which the cost-of-carry-rate, b , equals the riskless rate of interest, r . With dividends, the
relations are modified slightly to account for the fact that the underlying stock drops in price at the ex-
dividend instant, while the exercise price does not.120 As a result, an anticipated cash dividend generally
reduces the value of calls and increases the value of puts.
TABLE 13.1 Stock option contract prices.

CHICAGO BOARD
Option and Strike
NY Close price Calls-Last Puts-Last
Alcoa 55 r s 8 1/4 r s r
617/8 60 r 3 4 1/16 1 1/16 1 13/16
61 7/8 65 r 7/8 1114 1 3/4 3 1/4 4 1/2
61 7/8 70 r r 1/2 r r 8 1/4
61 7/8 75 s s 1/8 s s r
AmGenl 40 r r 3 7/8 r r 1/2
43 ½ 45 1/16 r 7/8 r r r
Chryslr 10 r r 3 1/8 r 1/16 1/8
13 12½ 9/16 ¾ 1 1/6 1/16 1/2 3/4
13 15 r 1/8 5/16 r 2 3/8 2 3/8
Cisco 45 7 7/8 8 3/4 r r r 7/8
54 ¾ 50 4 1/2 6 6 5/8 r r r
54 ¾ 55 r 3 1/4 4 r 3 r
Citicp 10 s 1 3/4 1 7/8 s r 5/16
11 ½ 12½ 1/16 7/16 1/2 1 1/16 r 1 5/16
11 ½ 15 r s 1/8 3 3/4 s 35/8
Homstk 12½ r r 2 7/8 r r r
15 3/8 15 7/16 15/16 1 1/8 1/16 5/16 11/16
15 3/8 17½ r 1/16 3/8 r r r
IBM 90 r s 10 1/4 r s 11/16
98 7/8 95 4 5 3/8 6 5/8 1/16 15/16 1 11/16
98 7/8 100 3/8 2½ 3 5/8 1 3/8 2 15/16 3¾
98 7/8 105 1/16 7/8 1 13/16 6 3/8 6¼ r
98 7/8 110 r s 13/16 r s 11 3/8
98 7/8 115 r s 3/8 r s r
InPap 65 8 s r r s r
74 3/8 70 3¼ 3¾ r r ¾ 1 7/8
Boeing 40 9 s r r s r
49¼ 45 4¼ 4½ 5½ r ¼ 7/8
49¼ 50 1/8 1 1/8 2 3/16 1 1 5/8 2½
49¼ 55 1/16 3/16 5/8 6 5 7/8 6½
120
The exercise price is, however, adjusted for stock splits and stock dividends. Where the stock split/dividend produces a
fractional result, the exercise price is rounded to the nearest eighth.
TABLE 13.1 Cont. Stock option contract prices.

CHICAGO BOARD
Option and Strike
Bois C 22½ r r r r r ½
25½ 25 9/16 r 1 5/8 1/16 r r
CBS 150 5 r r r 1 7/8 r
155½ 155 1 9/16 4 3/8 10 1 1/8 r 7 3/8
155½ 160 ¼ 2 7/8 7 4½ 7 r
Honwll 55 5¼ r 6½ r r 1 5/8
60 60 ¾ 2¼ 3¾ 5/8 r r
60 65 r r 1½ r r r
Humana 25 2 5/8 r 3¼ r ½ r
27 3/8 30 r 9/16 r 2 1 5/16 3 r
27 3/8 35 r r r 7 5/8 r r
ImunRs 35 r r r 3/16 2 3 5/8
46 40 9 9 1/8 14 3/8 ½ 3¼ 5¼
46 45 3 1/8 7 9½ 1 3/8 5 3/8 7
TelMex 35 r r 11 r r 5/16
45¼ 40 5¼ 5¾ 7 r 3/8 r
45¼ 45 11/16 2¼ 3 7/8 r 1 13/16 2 15/16
45¼ 50 r 5/8 1 13/16
Tribune 40 4 3/8 1 3/8 r r r
TritEn 40 8½ 8¾ 11 1/8 r 7/8 r
48 45 3½ 5½ 7 3/8 ¼ 2½ r
48 50 9/16 3½ 5¾ 2 3/6 r r
48 55 r 1 5/8 4¼ r r r
UAL 20 r s r 1/16 s r
131 7/8 25 6¾ s r r s 5¼
Oracle 7½ s r 7½ s r r
15½ 12½ 2 5/8 3 r r r r
15½ 15 ½ 1 5/16 2 3/8 r 5/8 r
15½ 17½ r r 1 1/8 r r 3
Pall 30 r r r r r 3/8
38 1/8 35 r r r r r r
38 1/8 40 r r r r r r
38 1/8 45 s r 9/16 s r r
ParaCm 35 6 3/8 6½ r r r ½
41¼ 40 1 5/16 2 3/8 4 1/16 7/8 2¼

AMERICAN
Option and Strike
Aetna 30 r r r r r 1/8
41 3/8 35 r r 6 /78 r r 3/8
41 3/8 40 1½ 2¼ 3 1/8 7/8 1¼
Cabltr 40 4 3/8 5¼ r 1/16 r r
43½ 45 3/8 3¼ 3¾ r r r
43½ 50 r 1 1/16 r r r r
Cetus 12½ s s 6½ s s r
19 15 3 7/8 r 4½ r r r
19 17½ 1 ½ 2 1/16 2 5/8 r 3/8 ¾
19 20 1/16 ½ 1 3/16 r r r
Chips 7½ r r 1 1/8 r r r
8 3/8 10 r r 5/16 r r r
Chiron 60 r s 13 ½ r s r

CHICAGO BOARD
Sales Last Chg. N.Y. Close
CALLS
SP 100 NOV 370 50767 2 5/16 +3/16 371.20
SP 100 NOV 365 30573 6 3/8 +1/2 371.20
SP 100 NOV 375 9832 3/8 -1/16 371.20
SP 100 NOV 360 9570 11 ¼ +3/4 371.20
SP 100 DEC 370 8047 7 3/8 +3/4 371.20
SP 100 DEC 375 7012 4 3/8 +1/2 371.20
SP 100 NOV 380 5476 1/16 -1/16 371.20
IBM NOV 100 4699 3/8 -3/16 98 7/8
SP 100 DEC 365 4243 10 5/8 +1/8 371.20
TelMex FEB 50 3172 1 13/16 -1/16 45 ¼
PUTS
SP 100 NOV 370 33791 1 1/16 -1 1/16 371.20
SP 100 NOV 365 28524 5/16 -3/8 371.20
SP 100 NOV 360 13976 1/8 -1/8 371.20
SP 100 DEC 365 10083 3 5/8 -1/2 371.20
SP 100 DEC 36 7759 2½ -5/16 371.20
SP 100 NOV 355 5404 1/16 -1/16 371.20
SP 100 NOV 375 5266 4 1/8 -1 ¼ 371.20
SP 100 DEC 370 4556 5¼ -1/2 371.20
SP 100 DEC 340 3482 5/8 -1/16 371.20
SP 500 NOV 390 3445 ¼ -3/16 397.41

AMERICAN
CALLS
Glaxo Dec 30 3534 1 11/16 +1 1/16 30 3/8
RJR Nb Nov 71/2 2964 2 5/8 -1/16 10 1/8
MMIdx Nov 32S 2496 31/ 8 +1 327.25
Dig Eq Nov 60 2214 2 7/8 +5/8 62 3/4
Ph Mor Nov 70 1946 ¾ -1/8 70 5/8
RJR Nb Dec 71/2 1900 2 5/8 -1/16 10 1/8
TelMex Feb 50 1821 1 7/8 +3/16 45 1/4
MMIdx Nov 320 1632 7½ +1 ¼ 327.25
Gillet Dec 45 1599 1 3/8 +1/8 44 3/4
TritEn Nov50 1549 ¾ +9/16 48
PUTS
Disney Nov 110 2089 1 +5/8 1103
RJR Nb Dec 10 2010 1/4 101/8
RJR Nb Dec 7½ 1900 1/16. 10 1/8
RJR Nb Nov 7½ 1900 1/16 10 1/8
MMIdx Nov 325 1627 11/16 - 13/16 327.25
Disney Dec 110 1054 3¼ +1 110 3/8
MMldx Nov 320 l035 1/8 -1/4 327.25
Chase Nov 15 1000 1/8 18 1/4
Ph Mor Nov 70 925 3/16 70 5/8
Pfizer Nov 70 779 11/16 +5/16 70 1/8
PHILADELPHIA
CALLS
NCNB Dec 35 2225 41/8 +1/4 39
NCNB Nov 35 2130 4 +1/8 39
RJR Nb Nov 10 1800 3/16 -1/16 10 1/8
RJR Nb Dec 10 1750 7/16 10 1/8
RJR Nb Nov 7½ 1049 2 5/8 +1/16 10 1/e
Synrgn Nov 65 965 3/8 -2½ 59 7/8
HomeD Nov 60 805 1 -3/8 60 3/4
HomeD Dec 60 775 3 1/8 -1/8 60 3/4
Synrsn Nov60 619 1 13/16 -4 5/16 59 7/8
Abbt L Dec 5S 571 6 3/8 -3/8 61 1/8

PHILADELPHIA
PUTS
HomeD Dec 55 2187 ¾ +1/4 4503/4
Comeric Jan 50 2000 1 -1 3/8 52 1/4
Synrgn Nov 65 1074 5½ +3 5/8 597/s
BkBost Nov 12½ 1020 1/4 12 ½
SecPac Dec 35 578 4 3/8 +1/8 31 3/8
RJR Nb Nov 12½ 515 2 3/8 103/e
HomeD Dec 50 505 l/4 +1/16 60 3/4
RJR Nb Dec 10 500 1/4 10 l/8
Abbt L Feb 60 434 2 -3/16 61 1/8
HomeD Nov 60 426 1/4 ... 60 3/4
Non-Dividend-Paying Stocks
To derive the lower price bounds and the put-call parity relations for options on non-dividend-paying stocks,
simply set the cost-of-carry rate, b , equal to the riskless rate of interest, r , in the relations presented in
Chapter 10. The only cost of carrying the stock is interest. The lower price bounds for the European call and
put options are
c( S , T ; X )  max[0, S  Xe  rT ] (13.1a)
and
p( S , T ; X )  max[0, Xe  rT  S ] , (13.1b)
respectively, and the lower price bounds for the American call and put options are
C ( S , T ; X )  max[0, S  X ] (13.2a)
and
P( S , T ; X )  max[0, Xe  rT  S ] (13.2b)
respectively. The put-call parity relation for non-dividend-paying European stock options121 is
c( S , T ; X )  p( S , T ; X )  S  Xe rT , (13.3a)
and the put-call parity relation for American options on non-dividend-paying stocks is
S  X  C ( S , T ; X )  P( S , T ; X )  S  Xe rT (13.3 b)
For non-dividend-paying stock options, the American call option will not rationally be exercised early, while
the American put option may be.122
121
The original formulation of put-call parity, for European stock options is contained in Stoll (1969).
122
For proofs of any of the relations (13.1a) through (13.3b), see Chapter 10.
Dividend-Paying Stocks
If dividends are paid during the option’s life, the above relations must reflect the stock’s drop in value when
the dividends are paid. To manage this modification, we assume that the underlying stock pays a single
dividend during the option’s life at a time that is known with certainty. The dividend amount is D and the
time to ex-dividend is t .123
If the amount and the timing of the dividend payment is known, the lower price bound for the
European call option on a stock is
c( S , T ; X )  max[0, S  De  rt  Xe  rT ] . (13.4a)
In this relation, the current stock price is reduced by the present value of the promised dividend. Because a
European-style option cannot be exercised before maturity, the call option holder has no opportunity to
exercise the option while the stock is selling cum dividend. In other words, to the call option holder, the
current value of the underlying stock is its observed market price less the amount that the promised dividend
contributes to the current stock value, that is, S  De rt . To prove this pricing relation, we use the same
arbitrage transactions as in Chapter 10, except we use the reduced stock price S  De  rt in place of S .
The lower price bound for the European put option on a stock is
p( S , T ; X )  max[0, Xe  rT  S  De  rt ] . (1 3.4b)
Again, the stock price is reduced by the present value of the promised dividend. Unlike the call option case,
however, this serves to increase the lower price bound of the European put option. Because the put option is
the right to sell the underlying stock at a fixed price, a discrete drop in the stock price such as that induced by
the payment of a dividend serves to increase the value of the option. An arbitrage proof of this relation is
straightforward when the stock price, net of the present value of the dividend is used in place of the
commodity price.
The lower price bounds for American stock options are slightly more complex. In the case of the
American call option, for example, it may be optimal to exercise just prior to the dividend payment because
the stock price falls by an amount D when the dividend is paid. The lower price bound of an American call
option expiring at the ex-dividend instant would be 0 or S  Xe  rt , whichever is greater. On the other hand, it
may be optimal to wait until the call option’s expiration to exercise. The lower price bound for a call option
expiring normally is (13.4a).124 Combining the two results, we get
C ( S , T ; X )  max[0, S  Xe rt , S  De  rt  Xe  rT ] (1 3.5a)
The last two terms on the right-hand side of (13.5a) provide important guidance in deciding whether to
exercise the American call option early, just prior to the ex-dividend instant. The second term in the squared
brackets is the present value of the early exercise proceeds of the call. If the amount is less than the lower
price bound of the call that expires normally, that is, if
S  Xe rt  S  De rt  Xe rT (13.6)
the American call option will not be exercised just prior to the ex-dividend instant.
123
Generalizations of the results to cases where there are more than one known dividend are derived in the same manner as the
single dividend results shown here.
124
Recall that in Chapter 10 we showed that an American call is never optimally exercised early if b  r . Between dividends, the
cost-of-carry rate is r , so exercise is not optimal. At the ex-dividend instant, however, the call option holder may wish to exercise
to capture the dividend.
To see why, simply rewrite (13.6) so it reads
D  X [1  e  r (T  t ) ] (13.7)
In other words, the American call will not be exercised early if the dividend captured by exercising prior to
the ex-dividend date is less than the interest implicitly earned by deferring exercise until expiration.
Figure 13.1 depicts a case in which early exercise could occur at the ex-dividend instant, t . Just prior
to ex-dividend, the call option may be exercised yielding proceeds St  D  X , where St is the ex-dividend
stock price. An instant later, the option is left unexercised with value c( St , T  t; X ) , where c() is the
European call option formula. Thus, if the ex-dividend stock price, St , is above the critical ex-dividend stock
price where the two functions intersect, S *t , the option holder will choose to exercise her option early just
prior to the ex-dividend instant. On the other hand, if S t  S *t , the option holder will choose to leave her
position open until the option’s expiration.
FIGURE 13.1 American call option price as a function of the ex-

dividend stock price immediately prior to the ex-dividend instant.
Early exercise may be optimal.
Call option price,

C(S,T;X)
c(S,T  t;X)
Exercise value of American

Call Option, St  D  X
c(St* ,T  t;X)
Lower Price Bound for European

Call Option, S t  Xe  r (T t )
Ex - D stock Price S t
X/e r (T t ) X  D S t*
Figure 13.2 depicts a case in which early exercise will not occur at the ex-dividend instant, t . Early exercise
will not occur if the functions, St  D  X and c(St , T - t; X) do not intersect, as is depicted in Figure 13.2.
 r(T - t)
In this case, the lower boundary condition of the European call, S t  Xe , lies above the early exercise
proceeds, St  D  X , and hence the call option will not be exercised early. Stated explicitly, early exercise
is not rational if
St  D  X  St  Xe  r(T - t)
FIGURE 13.2 American call option price as a function of the ex-

dividend stock price immediately prior to the ex-dividend instant.
Early exercise will not be optimal.
Call option price,

C(S,T;X)
Lower Price Bound for European

Call Option, S t  Xe  r (T t )
Exercise value of
c(S,T  t;X)
America Call Option, S t  D  X
Ex - D stock Price S t
X/e r (T t ) X D
This condition for no early exercise is the same as (13.6), where St is the ex-dividend stock price and where
the investor is standing at the ex-dividend instant, t . The condition can also be written as
D  X [1  e  r(T - t) ] (13.7)
In words, if the ex-dividend stock price decline-the dividend-is less than the present value of the interest
income that would be earned by deferring exercise until expiration, early exercise will not occur. When
condition (I 3.7) is met, the value of the American call is simply the value of the corresponding European
call.
The lower price bound of an American put option is somewhat different. In the absence of a dividend,
an American put may be exercised early. In the presence of a dividend payment, however, there is a period
just prior to the ex-dividend date when early exercise is suboptimal. In that period, the interest earned on the
exercise proceeds of the option is less than the drop in the stock price from the payment of the dividend. If tn
represents a time prior to the dividend payment at time t , early exercise is suboptimal, where ( St  X )e r(t - t n )
is less than ( X  S  D ). Rearranging, early exercise will not occur between tn and t if125
ln(1  XD S )
tn  t  (13.8)
r
125
It is possible that the dividend payment is so large that early exercise prior to the dividend payment is completely precluded.
For example, consider the case where X  50, S  40, D  1, t  0.25 and r  0.10 . Early exercise is precluded if
tn  0.25  ln[1  1/(50  40)]/ 0.10  0.7031 . Because the value is negative, the implication is that there is no time during the current
dividend period (i.e., from 0 to t ) where it will not pay the American put option holder to wait until the dividend is paid to exercise
his option.
Early exercise will become a possibility again immediately after the dividend is paid. Overall, the lower
price bound of the American put option is
P( S , T ; X )  max[0, X  ( S  De  rt )] . (13.5b)
Put-call parity for European options on dividend-paying stocks also reflects the fact that the current stock
price is deflated by the present value of the promised dividend, that is,
c( S , T ; X )  p( S , T ; X )  S  De  rt  Xe  rT . (13.9)
That the presence of the dividend reduces the value of the call and increases the value of the put is again
reflected here by the fact that the term on the right-hand side of (13.9) is smaller than it would be if the stock
paid no dividend.
Put-call parity for American options on dividend-paying stocks is represented by a pair of
inequalities, that is,
S  De  rt  X  C ( S , T ; X )  P( S , T ; X )  S  De  rt  Xe  rT . (13.10)
To prove the put-call parity relation (13.10), we consider each inequality in turn.
The left-hand side condition of (13.10) can be derived by considering the values of a portfolio that consists
of buying a call, selling a put, selling the stock, and lending X  De  rt risklessly. Table 13.2 contains these
portfolio values.
In Table 13.2, it can be seen that, if all of the security positions stay open until expiration, the
terminal value of the portfolio will be positive, independent of whether the terminal stock price is above or
below the exercise price of the options. If the terminal stock price is above the exercise price, the call option
is exercised, and the stock acquired at exercise price X is used to deliver, in part, against the short stock
position. If the terminal stock price is below the exercise price, the put is exercised. The stock received in the
exercise of the put is used to cover the short stock position established at the outset. In the event the put is
exercised early at time  , the investment in the riskless bonds is more than sufficient to cover the payment
of the exercise price to the put option holder, and the stock received from the exercise of the put is used to
cover the stock sold when the portfolio was formed. In addition, an open call option position that may still
have value remains. In other words, by forming the portfolio of securities in the proportions noted above, we
have formed a portfolio that will never have a negative future value. If the future value is certain to be
nonnegative, the initial value must be nonpositive, or the left-hand inequality of (13.10) holds.
The right-hand side of (13.10) may be derived by considering the portfolio used to prove European
put-call parity. Table 13.3 contains the arbitrage portfolio transactions. In this case, the terminal value of the
portfolio is certain to equal zero, should the option positions stay open until that time. In the event the
American call option holder decides to exercise the call option early, the portfolio holder uses his long stock
position to cover his stock obligation on the exercised call and uses the exercise proceeds to retire his
outstanding debt. After these actions are taken, the portfolio holder still has. an open long put position and
cash in the amount of X [1  e  r (T  t ) ] . Since the portfolio is certain to have nonnegative outcomes, the initial
value must be nonpositive or the right-hand inequality of (13.10) must hold.
TABLE 13.2 Arbitrage transactions for establishing put-call parity for American stock options.
S  De  rt  X  C ( S , T ; X )  P( S , T ; X )
Ex-Dividend Put Exercised Put Exercised
Day ( t ) Early (  ) Normally ( T )
Intermediate Terminal Value
Position Initial Value ~ ~
Value ST  X ST  X
~ ~
Buy American Call C C 0 ST  X
~ ~
Sell American Put P  ( X  S )  ( X  ST ) 0
~ ~ ~
Sell Stock S D  S  ST  ST
Lend De  rt  De  rt D
Lend X X Xe r XerT Xe rT
~
Net Portfolio Value  C  P  S  De  rt  X 0 C  X [e r  1] X [e rT  1] X [e rT  1]
13.3 VALUATION OF CALL OPTIONS ON STOCKS

European Call Option on Non-Dividend-Paying Stocks
A non-dividend-paying stock has a cost-of-carry rate, b , equal to the riskless rate of interest, r , so, using
equation (11.25) from Chapter 11, the valuation equation
of a European call option on a non-dividend-paying stock126 is
c( S , T ; X )  SN (d1 )  Xe rT N (d 2 ) , (13.11)
ln(S / X  (r  0.5 2 )T
where d1  , and d 2  d1   T
 T
TABLE 13.3 Arbitrage transactions for establishing put-call parity for American stock options.
C ( S , T ; X )  P( S , T ; X )  S  De  rt  Xe rT
Ex-
Put Exercised Put Exercised
Dividend
Early (  ) Normally ( T )
Day ( t )
Intermediate Terminal Value
Position Initial Value ~ ~
Value ST  X ST  X
~ ~
Sell American Call C  ( S  X ) 0  ( ST  X )
~ ~
Buy American Put P P X  ST 0
~ ~ ~
Buy Stock S D S ST ST
Borrow De  rt De  rt D
Borrow Xe  rT Xe  rT  Xe r (T  ) X X
~
Net Portfolio Value C  P  S  De rt  Xe rT 0 P  X [1  e  r (T  ) ] 0 0
126
This equation is often referred to as simply the Black-Scholes formula, given the important impact that the Black-Scholes
(1973) paper has had on the theory of option pricing and, more generally, the practice of finance.
European Call Option on Dividend-Paying Stocks

In the presence of a known discrete dividend, option valuation remains relatively straightforward. Instead of
assuming that future stock prices are lognormally distributed, we now assume that stock prices, net of the
present value of the escrowed dividend, are lognormally distributed. The dividend to be paid during the
option’s life is a known amount on a known date. At the ex-dividend instant, the stock price drops by the
amount of the dividend. The stock price net of the present value of the dividend remains unchanged.
Under this assumption, the valuation equations for European options maintain the same structure as
in Chapter 11. The only change is that the current stock price net of the present value of the promised
dividend,
S x  S  De  rt (13.12)
is substituted for the stock price parameter in the European call option pricing formula (13.11). The valuation
equation of a European call option on a dividend-paying stock is
c( S , T ; X )  S x N (d1 )  Xe  rT N (d 2 ) , (13.13)
ln(S x / X  (r  0.5 2 )T
where: d1  , and d 2  d1   T
 T
Naturally, the value of the call decreases as a result of the cash disbursement on the stock.
American Call Option on Non-Dividend-Paying Stocks

In Chapter 11, we showed that if the cost-of-carry rate of the underlying commodity is greater than or equal
to the riskless rate of interest (i.e., b  r ), the American call option on the commodity will never be
optimally exercised early and its value equals the value of a European call option. In the case of a non-
dividend-paying stock, the cost-of-carry rate equals the riskless rate of interest. Hence, the valuation equation
of an American call option on a non-dividend-paying stock is (13.11).
American Call Option on Dividend-Paying Stocks

When a stock pays a dividend, the problem of valuing an American call is more complex. Whereas an
American call option on a non-dividend-paying stock will never optimally be exercised prior to expiration,
an American call option on a dividend-paying stock may be. This situation arises because the stock price
falls at the ex-dividend instant, which causes the call to drop in value. It may be optimal for the American
call option holder to exercise his option just prior to this ex-dividend stock price drop.
The easiest way to derive the valuation equation for the American call option on a dividend-paying
stock is to use the risk-neutral option valuation. Consider the American call option holder’s dilemma as
depicted in Figure 13.1. At time t , the American option holder will exercise his option if St  St* and will
leave his position open if St  St* . And, if he leaves his position option at t , his option will have terminal
~
values ST  X if ST  X and 0 if ST  X . Recognizing that the option’s payoffs will occur at one of two
points in time (i.e., just prior to the early exercise instant or at expiration) allows us to write the current value
of the call as the present value of the expected future payoffs, that is,
~ ~
C ( S , T ; X )  e  rt E (Ct | St  St* )  e  rT E (Ct | St  St* )
~ ~
 e  rt E ( St  D  X | St  St* )  e  rT E ( ST  X | St  St* and ST  X )
~ ~
 e  rt E ( St | St  St* )  ( X  D) Pr( St  St* )  e  rT E ( ST  X | St  St* and ST  X )
~
 e  rT {E ( ST | St  St* and ST  X )  X Pr( St  St* and ST  X )} (13.13)
Assuming that the future stock price net of the present value of the promised dividend is lognonnally
distributed, the expected values on the right-hand side of (13.14) become the valuation equation of an
American call option on a dividend-paying stock:
C ( S , T ; X )  e  rt [ S x e rT N1 (b1 )  ( X  D) N1 (b2 )]
 e  rT [ S x e rT N 2 (a1 ,b1; t / T )  XN 2 (a2 ,b2 ; t / T )] (13.15a)
C ( S , T ; X )  S x [ N1 (b1 )  N 2 (a1 ,b1; t / T )]

 Xe  rT [ N1 (b2 )e r (T t )  N 2 (a2 ,b2 ; t / T )]  De  rT N1 (b2 ) (13.15b)
where
ln(S x / X )  (r  0.5 2 )T ln(S x / St* )  (r  0.5 2 )t
a1  , and a2  a1   T ; b1  , and b2  b1   t
 T  t
N1 (b) is the cumulative univariate normal density function with upper integral limit b 127 and N 2 (a, b;  ) is
the cumulative bivariate normal density function with upper integral limits, a and b , and correlation
coefficient,  .128 St* is the ex-dividend stock price for which
c( St* , T  t ; X )  St*  D  X (13.15c)
as noted earlier in the discussion of Figure 13.1.129

In equation (13.15a), the American call formula is the sum of the present values of two conditional
expected values-the present value of the expected early exercise value of the option conditional on early
exercise, St* N1 (b1 )  ( X  D)e  rt N1 (b2 ) , and the present value of the expected terminal exercise value of the
call conditional on no early exercise, St* N1 (a1 ,b1 ; t / T )  Xe  rt N 2 (a2 ,b2 ; t / T ) . The term N1 (b1 ) is the
probability that the call option will be exercised early, and the term N 2 (a2 ,b2 ; t / T ) is the joint probability
that the call option will not be exercised early and will be in-the-money at expiration.
Note that as the amount of the dividend approaches the present value of the interest income that
would be earned by deferring exercise until expiration, that is, D  X [1  e  r (T  t ) ] , the value of St*
approaches   , the values of N1 (b1 ) and N1 (b2 ) approach 0, the values of N 2 (a1 ,b1; t / T ) and
N 2 (a2 ,b2 ; t / T ) approach N1 (a1 ) and N1 (a2 ) , respectively, and the American call (13.15b) becomes the
dividend-adjusted Black-Scholes European option formula call option formula (13.13).
EXAMPLE 13.1
Compute the value of an American-style call option whose exercise price is $50 and whose time to
expiration is 90 days. Assume the riskless rate of interest is 10 percent annually, the underlying stock price is
$50, the standard deviation of the rate of return of the stock is 30 percent, and the stock pays a dividend of $2
in exactly 60 days.
127
Recall that the function N (b) was used for the first time in Chapter 11. The subscript 1 is used here only to contrast the
univariate integral from the bivariate integral.
128
More details about the bivariate normal probability, as well as a method of computation and a numerical example, are contained
in Appendix 13.1.
129
Roll (1977) provides a framework for analytically valuing the American call option. The valuation formula (13.15b) is from
Whaley (1981).
We compute the value of this call first using the European formula (13.13) and then using the
American formula (13.15b). In this way, we can identify the value of the early exercise premium on the call.
The current stock price net of the present value of the promised dividend is
S x  50  2e 0.10 ( 60 / 365)  48.033 , so the European call value can be computed as

c  48.033N1 (d1 )  50e 0.10 ( 90 / 365) N1 (d 2 ) where
[ln(48.033 / 50)  (0.10  0.5(0.30) 2 (90 / 365)]

d1   0.029 and d 2  0.029  0.149  0.178 . The
(0.30) (90 / 365)
probabilities N1 (0.029) and N1 (0.178) are 0.488 and 0.429, so the European call value is
c  48.033(0.488)  48.782(0.429)  2.51 .
Prior to applying the valuation equation for the American call option on a dividend-paying stock
(13.15b), we must determine if there is a chance of early exercise. Recall that if condition (13.7) holds, the
call will not be exercised early. Substituting the exercise values into (13.7), we find
2  50[1  e 0.10 ( 90 / 365) ]  0.409
showing that early exercise is not precluded and that formula (13.15b) should be used.
The value of the American call is now computed as
C  48.033 [ N1 (b1 )  N 2 ( a1 , b1 ;  t / T )]

 50e 0.10 ( 90 / 365) [ N1 (b2 )e0.10 ( 30 / 365)  N 2 (a2 ,b2 ; t / T )]  2e 0.10 ( 60 / 365) N1 (b2 ) ,
where t / T  (60 / 365)(90 / 365)  (60 / 90)  0.6667 ,

[ln(48.033 / 50)  (0.10  0.5(0.30) 2 (90 / 365)]
a1   0.029
(0.30) (90 / 365)
[ln(48.033 / 49.824)  (0.10  0.5(0.30) 2 (60 / 365)]
b1   0.105 , and
(0.30) (60 / 365)
b2  0.105  0.30 60 / 365  0.227 .
The values b1 and b2 depend on the critical ex-dividend stock price St* , which is determined by
c( St* ,30 / 365;50)  St*  2  50
and, in this example, equals 49.824. The bivariate normal probabilities are N 2 (a1 ,b1;  )  0.1135 and
N 2 (a2 ,b2 ;  )  0.1056 , and the univariate normal probabilities are N1 (b1 )  0.4581 and N1 (b2 )  0.4103 .
The value of the American call is 2.931; hence, the early exercise premium on the American option is 2.931 -
2.513 = 0.418.
TABLE 13.4 Simulated American and European call option values oil a stock with a known discrete
dividend. The call option has an exercise price of $50 and a time to expiration of 90 days. The
riskless rate of interest is 10 percent, and the standard deviation of the stock return is 30 percent.
The stock pays a dividend of $2 in 60 days.
Stock Price European Call Price American Call Price Early Exercise Premium
S c( S , T ; X ) C (S ,T ; X ) C
40 0.126 0.136 0.011
45 0.760 0.867 0.107
50 2.515 2.931 0.418
55 5.610 6.481 0.871
60 9.726 10.974 1.248
The size of the early exercise premium of an American call option on a dividend-paying stock becomes
larger as the option goes deeper in the money. In Table 13.4, we extend the results of Example 13.1 by
allowing the stock price to vary from $40 to $60. It is interesting to note that the dividend payment induces a
fairly large early exercise premium on the call option, particularly when the call is deep in-the-money. At a
$60 stock price, for example, the value of the early exercise premium is about $1.25, more than 11 percent of
the call’s overall value.
Dividend Spreads
In practice, not all call options are exercised when they should be. And, when they are not, there are ways to
profit risklessly. Consider, for example, two in-the-money call options written on a stock that is about to pay
a dividend. Assume the deeper in-the-money call is sold and the other is purchased. Now, on the day before
ex-dividend, exercise the purchased option and wait. If the holder of the deeper in-the-money call exercises
her option before ex-dividend, deliver the stock received from the exercise of the purchased option and pay
the net difference between the exercise prices of the options. On the other hand, suppose the holder of the
deeper in-the-money call option forgets to exercise her option. In this case, sell the stock the next morning
and buy back the remaining option. In the first case, profit equals zero, and, in the second, a profit in the
amount of the dividend would be received. A numerical example may serve to clarify this strategy.
Assume that a stock is currently priced at $60 and will pay a $2.00 dividend tomorrow. Call options
with exercise prices of 50 and 55 and time to expiration of 30 days are priced at $10.01 and $5.01,
respectively. (The riskless rate of interest is assumed to be 10 percent, and the standard deviation of stock
returns is 30 percent.) Now, assume the 50 call is sold, and the 55 call is purchased, yielding net proceeds at
the outset of $5.00. At the end of the day, the investor exercises the 55 call, receiving proceeds St 1 + 2.00 -
55. (Day t  1 is the day before ex-dividend, and the notation St 1 is the stock price net of the value of the
escrowed dividend.) If the 50 call option is exercised against the investor before ex-dividend (which will not
be known until the next day before market opening), the investor’s obligation is - ( St + 2.00 - 50), and the
net terminal value is St + 2.00 - 55 - St - 2.00 + 50 or - $5, exactly a wash, considering $5 was collected up
front. However, if the 50 call option is not exercised, the investor goes into the next morning with a long
position in the stock (acquired from exercising the 55 call) and a short position in the 50 call, which has a
value St  2.00  55  Ct . For all intents and purposes, this position is riskless because the uncertainty in the
value of the long stock position is offset by that of the short in-the-money call. The outstanding call, being
deep in-the-money, is selling for about its floor value of S, - 50, so the net value of the position is 2.00 - 55 +
50 or - $3. Net of the initial cash receipt of $5, the profit is $2.00, exactly the amount of the dividend. This
strategy is usually called a dividend capture or a dividend spread.
13.4 VALUATION OF PUT OPTIONS ON STOCKS
European Put Option on Non-Dividend-Paying Stocks

The valuation equation of a European put option on a non-dividend-paying stock is
p ( S , T ; X )  Xe  rT N ( d 2 )  SN (d1 ) (13-16)
where
ln(S / X )  (r  0.5 2 )T
d1  , and d 2  d1   T
 T
This result follows straightforwardly from substituting the riskless rate of interest, r , for the cost-of-carry
rate, b , in equation (11.28) from Chapter 11.
European Put Option on Dividend-Paying Stocks

Like the European call option on a dividend-paying stock, the European put option on a dividend-paying
stock is obtained by substituting the stock price net of the present value of the dividend, S x  S  De rt , for
the stock price parameter in the European option pricing formula. The valuation equation of a European put
option on a dividend-paying stock is
p ( S , T ; X )  Xe  rT N ( d 2 )  S x N (d1 ) , (13.17)
ln(S x / X )  (r  0.5 2 )T
where d1  , and d 2  d1   T
 T
Note that the put value increases as a result of the cash disbursement on the stock.
American Put Option on Non-Dividend-Paying Stocks

The American put option on a non-dividend-paying stock is not a tractable problem from a mathematical
standpoint, as was noted in Chapter 11. For this reason, numerical methods must be used to approximate the
value of the put. The approach used here is the binomial method.130 The binomial method assumes the stock
price moves in discrete jumps over discrete intervals of time. The up-and-down steps in stock price are
defined as a proportion of the stock price at the previous interval. If the current stock price is S0 , the stock
price at the end of the first interval is either uS0 or dS0 . If the total number of time steps is defined as n ,
where t  T / n and T is the time to expiration of the option, there are n  1 possible stock prices at the
option’s expiration. This binomial lattice is illustrated in Figure 13.3. The length of each interval or time step
in the figure is t . The factors u and d are defined as131
u  e t (1 3.18a)
and d  1/ u (13.18b)
The risk-neutral probabilities of up and down movements are

r*  d
p . (13.18c)
ud
130
See Cox, Ross, and Rubinstein (1979).
131
The factors are consistent with the Black-Scholes model. See Cox, Ross, and Rubinstein (1979) for details.
and 1  p , respectively, where r *  e rt .
FIGURE 13.3 Possible paths that the stock price may follow under the binomial model.
0 1 2 3 4  n (even) n (odd)
 u n S0 u n S0
u 4 S0   
3
u S0   
2 2
u S0 u S0   
uS0 uS0   uS0
S0 S0 S0  S0
dS0 dS0   dS0
2 2
d S0 d S0   
3
d S0   
4
d S0   
 d n S0 d n S0
FIGURE 13.4 Possible paths that the stock price may follow under the binomial model, where the current
stock price is 50, the riskless rate of interest is 10 percent annually, and the standard deviation of stock
returns is 30 percent annually. The time to expiration of the option is 90 days, and the number of time steps
is also 90. The time increment t is, therefore, 1 day or 0.00274 years.
Figure 13.4 provides a numerical illustration of the binomial lattice for the stock price. The current stock
price, S0 , is 50, the riskless rate of interest, r , is 10 percent, and the standard deviation of stock returns,  ,
is 30 percent. The time to expiration of the option, T , is 90 days, and the number of time steps, n , is 90. The
size of the time increment t is, therefore, one day or 0.00274 years.
0 1 2 3 4  90 days
 205.46
53.24  
52.41  
51.60 51.60  
50.79 50.79  
50 50 50  50
49.22 49.22  
48.45 48.45  
47.70  
46.96  
 12.17
Based on this information, the factors u and d are
u  e 0.30 0.00274
 1.01583 and d  1 / 1.01583  0.98442
e0.10 ( 0.00274 )  0.98442

Also, the probabilities of up-and-down movements are p   0.5048 and
1.01583  0.98442
1  p  0.4952 . Note that, in Figure 13.4, possible stock prices range from S0 d n  50(0.98442)90  12.17 to
S 0u n  50(1.0158267)90  205.46 at the option’s expiration.
With the stock price lattice computed, the approximation method starts at the end of the option’s life
and works back to the present, one increment, t , at a time. At the end of the option’s life (column n in the
figure), the option value at each stock price node is given by the intrinsic value of the put option, that is,
0 where S n , j  X
Pn , j ( S n , j )   (13.19)
 X - S n , j where S n , j  X
The option values one step back in time (in column n  1 ) are computed by taking the present value of the
expected future value of the option. At any point j in column n  1 , the stock price can move up with
probability p or down with probability 1  p . The value of the option at time n if the stock price moves up
is Pnj 1 , and if the stock price moves down is Pnj . The present value of the expected future value of the
option is, therefore,
( p ) Pn , j 1  (1  p ) Pn , j
Pn 1, j  (13.20)
r*
where r *  e rt . Using this present value formulation, all of the option values in column n  1 may be
identified.
Before proceeding back another time increment, t , in the valuation, it is necessary to see if any of
the computed option values are below their early exercise proceeds at the respective nodes, X  S n 1, j . If the
exercise proceeds are greater than the computed option value, the computed value is replaced with the early
exercise proceeds. If they are not, the value is left undisturbed. If this step is not performed, the procedure
will produce the value of a European put option.
Once the checks are performed, we go to column n  2 , repeat the steps and so on back through time.
Eventually, we will work our way back to time 0, and the current value of the American put option (in
column 0) will be identified.
To complete the binomial method illustration, suppose that the stock price lattice shown in Figure
13.4 underlies a 90-day American put option with an exercise price of 50. Applying the binomial method, the
value of the American put is $2.475. The value of this put using the European formula (13.16) is $2.364,
which means that the early exercise premium of the American put is worth about 11.10. Note that the value
produced by the binomial method for the European put (by not checking for the early exercise constraints) is
$2.355, which is different from the $2.364 obtained using the European formula. This is error due to the fact
that the binomial method is only an approximation. In general, the accuracy of the binomial method
increases with the number of time steps, holding other factors constant.
American Put Option on Dividend-Paying Stocks

The binomial method is also well suited to handle the case of valuing an American put option on a dividend-
paying stock. If the dividends paid during the put’s life are known with certainty, we first subtract the present
value of the dividends from the current stock price, that is,
S 0x  S 0  i 1 Di e  rt i (13.21)
n
where Di (ti ) is the amount of (time to) the i  th dividend paid during the option’s life and S0 is the current
stock price. Next, we set up the binomial lattice, beginning with S0x rather than S0 . That is, if the current
stock price net of dividends is S0x , the stock price at the end of the first time interval is either uS0x , or dS0x .
The values of u, d and p are computed as (13.18a), (13.18b), and (13.18c).
With the stock price lattice (net of dividends) computed, the approximation method starts at the end
of the option’s life and works back to the present. At the end of the option’s life, the option value at each
stock price node is given by the intrinsic value of the put (13.19), where S x replaces S . The option values
one step back in time (at time n  1 ) are computed by taking the present value of the expected future value of
the option (13.20). Before stepping back another time increment, it is again necessary to see if any of the
option values are below their early exercise value. Here is where dividends may enter the picture again. If a
dividend is paid at time n  1 , then the early exercise value is simply the ex-dividends are paid at time n  1 ,
however, the early exercise proceeds equal the exercise price less the lattice stock price less the dividend. If
any of the computed option values are below the exercise proceeds, they are replaced with the value of the
exercise proceeds.
As we repeat the process and step back further in time, we must keep track of the sum of the present
values of the dividends paid during the option’s remaining life. At time n  1 , there was only one dividend
and it was paid at time n  1 , so the sum equals the value of the single dividend paid at time n  1 . If we are
at time n  2 and there is a dividend paid at time n  2 as well as a dividend paid at time n  1 , the sum of
the present values of the promised dividends that should be included in the early exercise boundary check at
time n  2 is
Dn 1
PVDn  2  Dn  2  . (13.22)
r*
In other words, the early exercise boundary at time n  2 is X  S nx 2, j  PVDn  2 . By the time the iterative
procedure is complete, the early exercise boundary used to check the option price corresponding to the time
0 stock index level node will include the present value of all promised dividends as in equation (13.21).
EXAMPLE 13.2
Compute the value of an American-style put option that has an exercise price of $50 and a time to expiration
of 90 days. Assume the riskless rate of interest is 10 percent annually, the stock price is $50, the standard
deviation of the rate of return of the stock is 30 percent per year, and the stock pays a dividend of $2 in
exactly 60 days.
We proceed in two distinct steps. First we compute the European put option value using the formula
(13.17), and then we compute the American put option value using the binomial method. In this way, we can
identify the value of the early
exercise premium on the put.

The current stock price net of the present value of the promised dividend is
S x  50  2e 0.10 ( 60 / 365)  48.033 ,
so the European put value can be computed as p  50e 0.10 ( 90 / 365) N1 ( d 2 )  48.033N1 (d1 )
where
(ln(48.033 / 50)  (0.10  0.5(0.30) 2 )(90 / 365)

d1   0.029
0.30 90 / 365
and d 2  0.029  0.149  0.178 . The probabilities N (0.029) and N (0.178) are 0.512 and 0.571, so the
European put value is
p  48.782(0.571)  48.033(0.512)  3.26 .
The value of the American put is computed using the binomial method. The number of time steps is set equal
to 90, so the time increment t is one day or 0.00274 years. The values of the factors u and d are
u  e 0.30 0.00274  1.01583 , and d  1 / 1.01583  0.98442 , with probabilities
p  (e 0.10 ( 0.00274 )  0.98442) /(1.01583  0.98442)  0.5048 and 1  p  0.4952 , respectively. The possible
values of the stock price (net of dividends) at the option’s expiration range from 11.69 to 197.38. The value
of the American put is 3.393, hence the early exercise premium on the American option is 3.393 - 3.262 =
0.131.
Table 13.5 demonstrates how the value of the early exercise premium increases as the put option goes deeper
in-the-money. At a stock price of $40, for example, the early exercise premium is about 350, about 3 percent
of the overall option value.
Finally, it is worthwhile to note that the dividend-adjusted binomial procedure outlined above not
only handles an American put option on a dividend-paying stock but also handles American call options.
Where the stock pays only a single dividend during the option’s life, the American call option valuation
equation (13.15b) is the most computationally efficient. Where the stock pays multiple dividends, however,
the dividend-adjusted binomial method is much faster. We address this issue again when we value the
American-style S&P 100 index options in the next chapter.
13.5 RIGHTS AND WARRANTS

Rights and warrants are securities issued by the firm. Usually, they are attached to debt offerings by the firm
in order to reduce the coupon interest cost. Like call options, rights and warrants provide holders with the
right to buy the underlying stock at a predetermined price within a specified period of time. Unlike call
options, however, warrants are issued by the firm. Since the firm has a fixed amount of assets, the exercise of
rights or warrants means that there will be more stockholders sharing the same “pie,” hence the value of
existing shares will be diluted. Warrant valuation must account for this dilutionary effect.
In this section, we value rights and warrants explicitly recognizing the effects of dilution.132
Effectively, there is little distinction between rights and warrants from a valuation standpoint. Rights tend to
be short-term and at-the-money when they are issued; warrants tend to be long-term and out-of-the-money.
For convenience, we use only the term “warrants” in the remaining part of this section.133 The valuation
approach is like that used in Chapter 11. Because a riskless hedge can be formed between the warrant and the
value of the firm, we suffer no loss of generality and gain considerable mathematical tractability if we invoke
an assumption of risk-neutrality.
132
The approach used here is based on Smith (1976).
133
Executive stock options, a popular form of management incentive compensation, can be considered warrants from a valuation
standpoint.
TABLE 13.5 Simulated American and European put option values on a stock with a known discrete
dividend. The put option has an exercise price of $50 and a time to expiration of 90 days. The riskless
rate of interest is 10 percent, and the standard deviation of the stock return is 30 percent. The stock
pays a dividend of $2 in 60 days.
Stock Price European Put Price American Put Price Early Exercise Premium
S p ( S , T ; X ) P ( S , T ; X ) P
40 10.875 11.230 0.355
45 6.510 6.757 0.247
50 3.264 3.393 0.129
55 1.360 1.406 0.046
60 0.476 0.492 0.016
Let S be the aggregate market value of the shares of the common stock currently outstanding; W ,
the aggregate market value of all warrants; r , the riskless rate of interest; and, V , the total market value of
the firm. The firm is assumed to have only two sources of financing-stock and warrants-so the total market
value of the firm may be defined as V  S  W . The number of shares of stock outstanding is nS , and nW is
the number of shares of stock sold if warrants are exercised. One warrant is assumed to provide the right to
buy one share. The dilution factor possible due to the presence of the warrants is   nW /(nS  nW ) . The
stock is assumed to pay no dividends during the warrant’s life. The standard deviation of the overall rate of
return on the firm is  . Finally, the warrant contract parameters are T , the time to expiration of warrants,
and X , the aggregate amount paid by warrant holders to acquire the stock (i.e., the aggregate exercise price).
The assumptions used in the development of the warrant valuation equation are the same as those underlying
the European call option except it is assumed the total market value of the firm is lognorrnally distributed at
~
the warrants’ expiration, not the firm’s share price (i.e., ln(VT / V ) ) is normally distributed with variance
 2 ).
Using risk-neutral valuation, the value of the firm’s warrants today is the present value of the
expected future terminal value, that is,
~
W  e  rT E (WT ) . (13.23)
The terminal value of the warrants, in turn, is the proportion of the firm that the warrant holders will own if
the warrants are exercised,  , times the terminal value of the firm after the warrant holders pay the cash
~
exercise amount, VT  X , less the cash exercise amount, X , that is,
~
~  (V  X )  X if  (VT  X )  X
WT   T which can be rewritten
0 if  (VT  X )  X
~
~  (VT )  (1   ) X if  (VT )  ( 1  γ)X
WT   (13.25)
0 if  (VT )  ( 1  γ)X
Note the similarity between the structure of the warrants payoffs in (13.25) and the European call option
payoffs discussed in Chapter 11, that is,
~
~ ST  X if ST  X
cT   (13.26)
0 if ST  X
In Chapter 11, we showed that the expected terminal value of the European call option for a non-dividend-
paying common stock is
E (c~T )  e rT SN (d1 )  XN (d 2 ) (13.27)
ln(S / X )  (r  0.5 2 )T
where d1  and d 2  d1   T
 T
where ST is lognormally distributed. In the warrant valuation problem,  (VT ) , corresponds to the terminal
stock price ST and is assumed to be lognormally distributed. The term (1   ) X is certain and corresponds to
the exercise price of the stock option. It therefore follows that
~
E (WT )  e rT ( )VN (d1 )  (1   ) XN (d 2 ) , (13.28)
ln(( )V /(1   ) X )  (r  0.5 )T 2
 T
Substituting (13.28) into (13.23), we find-that the aggregate market value of the warrants of the firm is
W  ( )VN (d1 )  e  rT (1   ) XN (d 2 ) , (13.29)
ln(( )V /(1   ) X )  (r  0.5 2 )T

 T
The market value per warrant is simply W from (13.29) divided by nW .

One problem with applying (13.29) to value warrants is that it is difficult to estimate the volatility
rate of the firm,  . To estimate this parameter using historical data requires a time series of prices for both
the stock and the warrant. Since the warrants may not actively trade, acquiring the historical price series for
the warrant may be difficult. On the other hand, estimating the volatility rate of the stock,  S , is much easier
since stocks are more actively traded and historical daily prices are available from a variety of sources. In
addition, if the stock has listed options, the stock option pricing model can be used to compute the implied
volatility rate of the stock.
Fortunately, the warrant valuation equation can be reformulated in terms of the stock’s volatility rate
rather than the volatility rate of the firm.134 Since the rate of return on the stock is perfectly correlated with
the rate of return on the overall firm,
 S  SV  , (13.30)
where  SV is the elasticity of the stock price with respect to the value of the firm, that is, the percentage
change in stock price for a given percentage change in the value of the firm. To estimate  SV first recall that
by assumption the firm’s value is the sum of the market value of the stock and the market value of warrants.
Thus,
134
This idea was first suggested by Schulz and Trautmann (1991).
S  V W
The change in the market value of the stock for a given change in the value of the firm is, therefore,
S W
1
V V
Second, from the valuation equation (13.29), we know that
W
 ( ) N (d1 ) so
V
S
 1  ( ) N (d1 ) . (13.31)
V
Finally, multiplying this term by V / S , we find that the elasticity of the stock price with respect to the value
of the firm is
S / S V
SV   [1  ( ) N (d1 )] (13.32)
V / V S
Hence, to value the warrant as a function of the stock’s volatility rate, we simply substitute the following
term for the  in (13.29):
1 S
 S  S (13.33)
SV [1  ( ) N (d1 )]V
Another somewhat perplexing consideration in applying (13.29) to value warrants is that the warrant
value appears on both sides of the equation, directly on the left-hand side and indirectly through V (i.e., W
is embedded in V ) on the right-hand side. This poses no great difficulty. We simply find the value of W that
satisfies the equation through some sort of numerical search procedure, just as we do when finding the yield
to maturity of a coupon-bearing bond.
EXAMPLE 13.3
Compute the value of a one-year warrant whose exercise price is $50. The current stock price is $50, and the
stock pays no dividends. The firm has only two sources of financing: 2,000 shares of stock and 500 warrants.
One warrant entitles its holder to one share of common stock. Assume that the riskless rate of interest is 6
percent, and that the standard deviation of the rate of return on the firm is 30 percent.
The dilution factor posed by the warrants is
500
   0.2
2,000  5000
the aggregate exercise proceeds to the firm are

X = 500 x 50 = 25, 000,
and the market value of the firm is
V  2,000  50  W  100,000  W .
The aggregate market value of the Warrants is computed by solving
W  0.20(100,000  W ) N (d1 )  e 0.06 (1) 0.80(25,000) N (d 2 )

ln[0.2(100,000  W ) /(0.8(25,000))]  (0.06  0.5(0.30) 2 )1
where d1   0.4611 and d 2  d1  0.30 1
0.30 1
The solution to this problem is obtained iteratively. The values of d1 and d 2 on the final iteration are
0.4611 and 0.1611, respectively. The probabilities are N (0.4611)  0.6776 and N (0.1611)  0.5640 . The
aggregate market value of the warrants is $3,389.46, so the price per warrant is $6.78.
In the interest of completeness, it is worthwhile to note that the volatility rate of the stock in this exercise
equals
[1  ( ) N (d1 )]V [1  0.2(0.6776)]103,389.46

S    0.30  0.2681 .
S 100,000
The volatility rate of the stock is lower than the volatility rate of the overall firm so the volatility rate of the
warrants must be higher. Since the returns of the stock and the warrant are perfectly correlated,
S W 
   S    W  
V  V 
  100,000  103,389.46 
Therefore, the warrant volatility rate is  0.30  0.2661    1.3002
  103,389.46  3,389.46 
13.6 SUMMARY
The focus of this chapter is stock options. Following a description of exchange-traded stock options in the
first section, we adapt the general pricing principles of Chapters IO and II to value call and put options on
non-dividend-paying stocks. The principles are modified somewhat to account for the fact that common
stocks typically pay discrete dividends during the option’s life. We assume that the amount and the timing of
the dividends paid during the option’s life are known with certainty.
In this chapter, we also introduce the use of the binomial method to price American options.
Although the specific application in this chapter is American-style options on stocks, the binomial method
can be applied to the valuation of virtually any type of option. We use it again in Chapter 15, for examples to
value interest rate options.
Finally, a special type of call option on common stock is considered. Specifically, firms often issue
fights or warrants to raise new capital. Like call options, these contracts provide the holder with the fight to
buy the common shares of the firm at a fixed price within a specified period of time. Unlike call options,
however, the firm sells (or gives away) the options, so, if the fights/warrants are exercised, the firm faces the
prospect of having the equity of the firm diluted. The prospect of dilution has an important effect on warrant
price.
APPROXIMATION FOR THE CUMULATIVE BIVARIATE NORMAL DENSITY FUNCTION
In Appendix 11.2, a cumulative univariate normal density function approximation was provided to help
compute the value of options on commodities with a constant, proportional cost-of-carry rate. In this chapter,
we found that if a common stock pays a discrete dividend during the option’s life, the American call option
valuation equation requires the evaluation of a cumulative bivariate normal density function.
While there are many available approximations for the cumulative bivariate normal distribution, the
approximation provided here relies on Gaussian quadratures.135 The approach is straightforward and
efficient, and its maximum absolute error is 0.00000055.
The probability that x is less than a and that y is less than b for a standardized bivariate normal
distribution is
1 a b x 2  2 xy  y 2
Pr( x  a and y  b)   
2 1   2    
exp[
2(1   2 )
]dxdy  N 2 (a, b;  )
where  is the correlation between the random variables x and y .

The first step in the approximation of the bivariate normal probability N 2 (a, b;  ) involves
developing a routine that evaluates the function  (a, b;  ) below:
 (a, b;  )  0.31830989 (1   2 ) i1  j 1 wi w j f ( xi , x j )

5 5
where
f ( xi , x j )  exp[a1 (2 xi  a1 )  b1 (2 x j  b1 )  2(  )( xi  a1 )( x j  b1 )] (2)
the pairs of weights ( w ) and corresponding abscissa values ( x ) are:
i, j w x
1 0.24840615 0.10024215
2 0.39233107 0.48281397
3 0.21141819 1.0609498
4 0.0033246660 1.7797294
5 0.00082485334 2.6697604
and the coefficients a1 and b1 are computed using

a b
a1  and b1  .
2(1   )
2
2(1   2 )
The second step in the approximation involves computing the product, ab .
If ab  0 , compute the bivariate normal probability, N 2 (a, b;  ) , using the following rules:
1.If a  0, b  0, and   0, then N 2 (a, b;  )   (a, b;  )

2. If a  0, b  0, and   0, then N 2 (a, b;  )  N1 (a )   (a, b;   )
3. If a  0, b  0, and   0, then N 2 (a, b;  )  N1 (b)   (a, b;   )
4. If a  0, b  0, and   0, then N 2 (a, b;  )  N1 (a)  N1 (b)  1   (a, b;  )
135
The Gaussian quadrature method for approximating the bivariate normal is from Drezner (1978), and the Gaussian quadratures
for the integral are from Steen, Byme, and Gelbard (1969). For a contingency table approach to this problem, see Wang (1987).
If ab   0 , compute the bivariate normal probability, N 2 (a, b;  ) , as:
N 2 (a, b;  )  N 2 (a, 0;  ab )  N 2 (b, 0;  ab )  
where the values of N 2 () on the right-hand side are computed from the rules for ab   0 ,
(  a  b) Sign(a ) (  b  a ) Sign(b) 1  Sign(a )  Sign(b)

 ab  ,  ab   and
a 2  2  ab  b 2 a 2  2  ab  b 2 4
1, x  0
Sign( x)  
0, x  0
N1 (d ) is the cumulative univariate normal probability. An approximation for N1 (d ) is contained in

Appendix 11.2.
Finally, to assist those who may attempt to code this algorithm, sample computations for the bivariate
normal probabilities are shown in the next table:
a b  N 2 (a, b;  )
-1.00 -1.00 -0.50 0.003782
-1.00 1.00 -0.50 0.096141
1.00 -1.00 -0.50 0.096141
1.00 1.00 -0.50 0.686472
-1.00 -1.00 0.50 0.062514
-1.00 1.00 0.50 0.154873
1.00 -1.00 0.50 0.154873
1.00 1.00 0.50 0.745203
0.00 0.00 0.00 0.250000
0.00 0.00 -0.50 0.166667
0.00 0.00 0.50 0.333333
EXAMPLE 13.A
Compute the risk-neutral probability that IBM and GM will have stock prices above $120 and $60,
respectively, at the end of two months. The current price of IBM is $107 and the current price of GM is $48.
Assume the riskless rate of interest is 10 percent annually, IBM and GM returns are bivariate normally
distributed, IBM has a return volatility of 33 percent annually, GM has a return volatility of 36 percent
annually, and the correlation between the returns of the two stocks is 0.6. Neither stock pays a dividend
during the next two months.
The first step in finding this probability is to compute the upper integral limits for the standardized
normal bivariate density function. For IBM, the computation is
[ln(107 /120)  (0.10  0.5(0.33) 2 )(2 /12)]
a  0.7948 and, for GM, the computation is
(0.33) (2 /12)
[ln(48 / 60)  (0.10  0.5(0.36) 2 )(2 /12)]
b  1.4784
(0.36) (2 /12)
The next step is to apply an approximation method to compute the bivariate normal probability. Applying the
procedure described above, the probability is
N 2 (a, b;  )  N 2 (0.7948, 1.4784;0.6)  0.0463

The probability that in two months IBM will have a stock price above 120 and that GM will have a stock
price above 60 is slightly more than 4.6 percent.
It is instructive to note that the individual probabilities of each stock realizing its critical future value
are
N1 (a )  N1 (0.7948)  0.2134 and N1 (b)  N1 (1.4784)  0.0697 ,
for IBM and GM, respectively. Thus, if the returns of IBM and GM were independent (i.e., their return
correlation is 0), the probability that in two months IBM will have a stock price above 120 and that GM will
have a stock price above 60 is 0.2134 x 0.0697 or about 1.49 percent. The reason that this probability is less
than the 4.6 percent where the correlation is 0.6 is that, with high positive return correlation, an upward
movement in IBM’s stock price implies that GM’s stock price a will tend to move upward also. In the
extreme case where these two stocks have a perfect positive correlation (i.e.,   1 ), the probability that in
two months IBM will have a stock price above 120 and that GM will have a stock price above 60 is the
lower of the two univariate probabilities, 6.97 percent.
14 STOCK INDEX AND STOCK INDEX FUTURES OPTION CONTRACTS
14.1 STOCK INDEX AND STOCK INDEX FUTURES OPTION MARKETS

 Stock Index Options
 Stock Index Futures Options
14.2 LOWER PRICE BOUNDS AND PUT-CALL PARITY

 Stock Index Options
 Stock Index Futures Options
 Index Options Versus Index Futures Options
14.3 VALUATION EQUATIONS

 European Stock Index Options
 American Stock Index Options
 European Stock Index Futures Options
 American Stock Index Futures Options
14.4 DISCRETE DIVIDENDS AND THE S&P 100 INDEX OPTION

 Discrete Dividends
 S&P 100 Index Option Valuation
 The Wildcard Option
14.5 PORTFOLIO INSURANCE

 Static Portfolio Insurance
 Dynamic Portfolio Insurance
14.6 SUMMARY
14 STOCK INDEX AND STOCK INDEX FUTURES OPTION

CONTRACTS
In Chapter 7, we discussed the use of stock index futures contracts in the management of stock portfolios.
We saw how fund managers could quickly and inexpensively hedge the market risk of their portfolios by
selling index futures contracts. In this chapter, we look at another means of tailoring the risk/return
characteristics of stock portfolios by using options on stock indexes and stock index futures. We begin with a
description of current index option and index futures option contracts. In sections 2 and 3, arbitrage price
relations and valuation equations for index option are described. For the most part, we maintain the
assumption that the dividend yield rate on the stock index portfolio is a constant, continuous rate. In section
4, we discuss the effect that discrete cash dividends have on the valuation of the American-style S&P 100
index options, and we address the wildcard option embedded in the S&P 100 index option contract. Section 5
contains a discussion of a popular application of index option contracts-portfolio insurance. The chapter is
summarized in section 6.
14.1 STOCK INDEX AND STOCK INDEX FUTURES OPTION MARKETS

The first stock index options to trade in the U.S. were the Chicago Mercantile Exchange’s S&P 500 and the
New York Futures Exchange’s NYSE Composite futures option contracts. They began trading on January
28, 1983. The Chicago Board Options Exchange followed quickly with the introduction of the S&P 100
option contract on March 11, 1983. The American Exchange introduced an option on the Major Market
Index on July 1, 1983, and the New York Stock Exchange introduced options on the NYSE Composite Index
on September 23, 1983. Since that time, a number of option contracts on more narrowly based stock indexes
have been introduced, but most of these have failed.
Tables 14.1 and 14.2 contain Wall Street Journal listings of the currently active index option and index
futures option contracts, respectively. - The index options in Table 14.1 are all cash settlement contracts.
When the option expires, usually at the close of trading on the third Friday of the contract month, the option
seller pays the option buyer an amount of cash equal to the difference between the closing index price and
the exercise price of the option, assuming the option is in-the-money. No delivery takes place. When the
index futures options in Table 14.2 expire, the option holder receives a position in the underlying futures
contract.
Stock Index Options

All index options, except the CBOE’s S&P 100 contract, are European-style. For the S&P 100 index options,
exercise may occur at any time prior to expiration. If an option buyer exercises early, she receives the
difference between the closing index level on that day and the exercise price of the option. The offsetting
option seller, who is obliged to make the cash payment to the buyer, is randomly chosen from all of the open
short positions in that option.
The prices reported in Table 14.1 are quoted in the same units as the underlying stock index, although
the value of the contract is 100 times the reported price. For example, the January 360 call option on the S&P
100 index had a reported closing price of 17 on November 13, 1991, and the S&P 100 index closed at
371.21. To buy this call, one would pay 17 X $100 or $1,700. The call would provide us with the right to buy
317.21 x $100 or $37,121 worth of the S&P 100 stock portfolio for an amount of cash equal to 360.00 x
$100 or $36,000 at any time between November 13, 1991, and the third Friday of January 1992. Again, the
contract is cash-settled, so delivery does not actually take place. If the S&P 100 index level is, say, $380.50
at exercise, the call option buyer would receive in cash an amount equal to (380.50 - 360.00) x $100 or
$2,050.
TABLE 14.1 Stock index option contract prices.

OPINONS: CHICAGO BOARD
S&P 100 INDEX-$100 times Index
Calls-Last Puts-Last
Strike
Price
335 1/16
340 30 311/2 311/4 1/16 5/8 1/2
345 23 27 28 1/16 7/8 21/8
350 211/2 211/8 1/16 13/16 3
355 163/8 19 195/8 1/16 1 3/4 37/8
360 111/4 143/4 17 1/8 21/2 41/2
365 63/8 105/8 131/2 5/16 35/8 6
370 25/16 73/8 101/2 11/16 51/4 73/4
375 3/8 43/8 73/8 41/8 75/8 107/8
380 1/16 27/16 51/8 91/4 107/8 133/4
385 1/16 11/4 31/8 15/4
390 1/16 5/8 17/8 185/8 20 203/4
The Index: High 371.22; Low 367.54; Close 371.21, +0.92
S&P 50 INDEX-SIDO times Index
Strike
Nov Dec Mar Nov Dec Mar
Price
325 71 13/8
345 3/16
350 1/4 27/8
355 1/2
360 353/4 1/2 4
365 1/16 3/4 41/2
370 251/2 255/8 1/16 1
375 225/8 1/16 11/4 73/8
380 141/4 193/4 1/16 1 3/4 6
385 123/4 157/8 1/8 21/2 10
390 73/8 113/4 19 1/4 35/8 103/4
395 27/8 83/8 17 5/8 5 113/4
400 1/2 51/4 4 7
405 1/8 3 101/8
410 11/2 123/4 131/8
415 7/8
420 51/2 27
425 3/16 33/8 275/8 281/2
TABLE 14.1 Cont. Stock index option contract prices.

LEAPS-S&P 100 INDEX
Strike Dec Dec Dec Dec Dec Dec
Price 92 93 94 92 93 94
30 11/16
321/2 11/4
35 2 211/16
371/2 35/8
Total call volume 0 Total call open Int. 25,487
Total put volume 81 Total put open Int. 86,28.4
LEAPS-S&P 500 INDEX
Strike Dec Dec Dec Dec Dec Dec
Price 92 93 94 92 93 94
30 3/4
35 19/16
371/2 7/8
40 3
Total put volume 366 Total put open Int. 72,715
The Index: High 39.74; Low 39.40; Close 39.74, + 0.07
INSTITUTIONAL INDEX
Strike
Price
335 1/16
340 1/8
360 11/16
370 15/16
375 1/2 15/16
380 327/8 3/4
385 13/16 21/8
390 1/16 15/16 211/16
395 1/16 15/8 43/8
400 15 1/8 29/16 43/8
405 71/2 3/16
410 41/8 91/8 11/16 43/4
415 15/16 47/8 21/2 61/2
420 1/8 3 57/8 61/2
425 17/8
430 11/16

JAPAN INDEX
Strike
Price
225 3/8
230 13/16
235 21/4
240 1/4
245 61/8 93/8 11/4 41/4
250 3/16 43/4 65/8
260 1
Total call volume 220; Total call open Int. 7,624
Total put volume 531; Total put open Int. 21,223
The Index: Close 245.50, -2.52
PHILADELPHIA
GOLD/SILVER INDEX
Strike
Price
70 117/8
80 ½ 27/8
85 1/8 13/8
90 3/4
The Index: High 80.61; Low 79.06; Close 79.53, -0.03
VALUE LINE INDEX OPTIONS
Strike
Price
315 93/4 1/8
320 61/2 121/4 5/16 31/2
325 21/2 11/2 5 75/8
330 41/2
Total put volume 397 Total put open Int. 5,140
UTILITIES INDEX
Strike
Price
250 41/4
Total put volume 0; Total put open Int. 78
The Index: High 255.00; Low 2.54.24; Close 254.95, +0.63

OPINONS
AMERICAN
MAJOR MARKET INDEX OPTIONS
Strike
Price
270 1/16
280 45 46 7/16
295 1/2
300 251/8 251/2 9/16 11/2
305 21 1/16 7/8
310 161/2 161/2 1/16 13/8
315 10 141/8 1/8 13/4
320 71/2 81/2 123/4 1/8 23/4 55/8
325 31/9 65/8 8 11/16 4
330 7/16 4 51/2 33/8
335 1/16 2 41/8
340 15/16 151/4
345 3/8 15/16
Total call volume 7,617 Total call open Int. 39, lS8
LT-20 INDEX OPTIONS
Strike
Dec 92 Dec 93 Dec 94 Dec 92 Dec 93 Dec 94
Price
25 3/8
PACIFIC
FINANCIAL NEWS COMPOSITE INDEX OPTIONS
Strike
Price
260 10
270 7/8 115/16
275 61/4
Total put volume 44; Total put open Int. 374

OPINONS
NEW YORK
NYSE INDEX OPTIONS
Strike
Price
210 1/8 101/4
215 43/8 61/2 ¼ 113/16
2171/2 21/8 3/16
220 5/16 13/8 43/8
225 77/8
The Index: High 219.37; Low 217.64; Close 219.371 + 0.37
TABLE 14.2 Stock index futures option contract prices.

INDEX TRADING
FUTURES OPTIONS
S&P 500 STOCK INDEX (CME) $500 times premium
Calls-Settle Puts- Settle
Strike
Nov-c Dec-c Jan-c Nov-p Dec-p Jan-p
Price
390 8.50 11.60 15.50 0.20 3.35 4.25
395 3.90 8.15 12.05 0.60 4.85 6.75
400 0.85 5.35 9.00 2.55 7.05
405 1.01 3.20 6.45 6.80 9.85
410 0.0000 1.65 4.35 11.70 13.30
415 0.0000 0.80 2.75 17.40
Est. vol. 6,929; Tues vol 4,989 calls; 8,151 puts
Open Interest Tues; 36,719 calls; 69,190 puts
OTHER FUTURES OPTIONS
NYSE COMPOSITE INDEX (NYFE) $500 times premium
Strike
Price
220 0.80 2.95 4.80 0.75 3.05 4.00
Est. vol. 136; Tues vol 116 calls; 121 puts
Open Interest Tues; 985 calls; 1,075 puts
TABLE 14.2 Cont. Stock index futures option contract prices.

NIKKEI 225 STOCK AVERAGE (CME) $5 times NSA
Strike
Price
24500 260 550 60 350
Est. vol. 185; Tues vol 0 calls; 20 puts
Open Interest Tues; 265 calls; 666 puts
CBT-Chicago Board of Trade. CME-Chicago Mercantile
Exchange. NYFE-New York Futures Exchange, a unit of the New
York Stock Exchange
The figures in Table 14.1 show that there is a wide disparity in the trading activity of different index option
contracts. The CBOE’s S&P 100 contract, for example, is by far the most active. The CBOE’s S&P 500
contract and the AMEX’s MMI contract are moderately active. The rest of the index option contracts are
fairly inactive. Given the substitutability of the broad-based index option contracts, it is not surprising that
the broad-based contracts introduced earliest have been the most successful. The lack of success of the
options on narrow indexes reflects the smaller market for these options.
The compositions of the stock portfolios that underlie each of the indexes with the most active option
contracts were discussed in Chapter 7, except for the S&P 100. The S&P 100 index has a fairly short history.
When the CBOE was considering an index option contract in the early 1980s, it decided on a value weighted
index of the one hundred largest stocks for which CBOE stock options existed. Originally, the index was
called the “CBOE 100.” Later, the CBOE reached an agreement to have Standard & Poors track the portfolio
composition, at which time the index was renamed the S&P 100.

Table 14.2 contains the two active index futures option contracts. These options are American-style,
denominated as 500 times the current index value, and written on the index futures contract. Unlike index
options, index futures options are delivery contracts. For example, ownership of the December 390 call on
the S&P 500 futures would give one the right to buy the underlying December 1991 futures contract at a
price of 390.00 at any time between November 13, 1991, and the close of trading on the third Thursday of
December 1991. This option is currently in-the-money since the current price of the December S&P 500
futures is 398.30 as seen in Table 14.3. If this call were exercised on November 13, one would receive a long
December S&P 500 futures position at a price of 390.00. A seller of the option, selected randomly, would
receive the offsetting short position in the futures contract. When the futures contract is marked-to-market on
that day, the long is allowed to withdraw 8.30 x $500 or $4,150, and the short pays in $4,150, but both would
still have open futures positions. To withdraw fully from the market, the futures contracts established as a
result of exercise would have to be sold. An exception to this procedure occurs on the expiration date of the
futures contract. Because the December futures option contract expires at the same time as the December
futures contract and because the December futures contract is cash-settled to the index value at expiration,
exercising the option at expiration amounts to cash settlement in that no futures position is left open.
Table 14.2 also shows futures option contracts with a November expiration, while Table 14.3 reveals
that there is no November futures contract. The November index futures options, as well as the December
futures options, are written on the December futures contracts. When an in-the-money November futures
option expires on the third Friday of November, for example, a position in the underlying December futures
is assumed. Again, to exit the market fully, the option holder must reverse the futures position that he
receives upon exercising his option.
14.2 LOWER PRICE BOUNDS AND PUT-CALL PARITY

The arbitrage price relations for stock index and stock index futures options follow straightforwardly from
the results of Chapter 10 if the underlying index is assumed to pay a constant proportional dividend yield
rate. Recall that this was the assumption made in the discussion of stock index futures contracts in Chapter 7.
In this case, the cost-of-carry rate, b , of the underlying stock index is the difference between the riskless rate
of interest, r , and the dividend yield rate, d , on the index. Substituting r  d for the cost-of-carry rate
provides us with the arbitrage relations for stock index options. The cost-of-carry for a futures contract is
zero, and substituting zero for the cost-of-carry rate, b , provides us with the arbitrage relations for stock
index futures options.
TABLE 14.3 Stock index futures contract prices.

FUTURES
S&P 500 INDEX (CME) $500 times index
Dec 395.00 398.50 394.30 398.30 +1.00 401.50 316.50 139,341
Mr92 396.80 400.50 396.50 400.35 +1.00 404.00 374.70 7,544
June 398.30 402.35 398.30 402.20 +1.10 407.00 379.00 1,102
Est vol 42,125; vol Tues 41,413; open Int 148,048, + 916.
Index prelim High 397.42; Low 394.01; Close 397.42 +0.68
NIKKEI 225 Stock Average (CME)-$5 times NSA
Dec 24690. 24700. 24600. 24700. -340. 28900. 22380. 10,969
Mr92 25250. 25250. 25170. 25230. -340. 26725. 23000. 2,423
The index- High 24814.35; Low 24416.23; Close 24416.23 -251.50
NYSE COMPOSITE INDEX (NYFE) $500 times Index
Dec 218.00 220.10 217.75 220.05 +.70 220.10 175.50 5,026
Mr92 218.80 221.00 218.90 221.00 +.80 221.00 207.60 746
June ... ... ... 222.00 +.80 220.10 208.90 172
Sept ... ... ... 223.00 +.80 221.00 217.50 123
Est vol 5,0571 –vol Tues 5,996; open Int 6,067, +344.
MAJOR MKT INDEX (CBT) $500 times Index
Nov 323.70 327-40 323.25 327.25 +1.55 327.40 315.20 2,819
Dec 323-50 327.70 323-50 327.70 +1.50 327-70 315.75 746
Est vol 2,500 –vol Tues 1,163 open Int 3,598, +172.
MGMI BASE METAL INDEX (FOX) 100 times Index
Nov ... ... ... 134.50 ... 140.50 132.50 2,246
Dec ... ... ... 134.90 ... 182.70 133.00 8,662
Ja92 ... ... ... 134.90 ... 137.10 132.30 120
Mar ... ... ... 135.30 ... 160.20 132.60 2,643
June ... ... ... 136.00 ... 155.90 134.50 962
Est vol 0; vol Tues 0; open Int 14,712, .
TABLE 14.3 Cont. Stock index futures contract prices.

OTHERFUTURES
Settlement price of selected contract. Volume and open Interest of all contract months.
KC Mini Value Line (KC) –100 times Index
Dec 328.60 +0.85; Est. vol. 100; Open Int. 254
KC Value Line Index (KC) –500 times Index
CRB Index (NYFE) –500 times Index
The Index: High 214.43; Low 213.94; Close 214.20 +.26
CBT –Chicago Board of Trade. CME –Chicago Mercantile Exchange. KC –Kansas City Board
of Trade. NYFE –New York Futures Exchange, a unit of the New York Stock Exchange.
Stock Index Options

The lower price bounds for European stock index options are
c( S , T ; X )  max[0, Se  dT  Xe  rT ] (14.1a)
and
p( S , T ; X )  max[0, Xe rT  Se  dT ] (14.1b)
for the call and the put, respectively. The lower price bounds for the American call and put options on stock
indexes are
C ( S , T ; X )  max[0, Se dT  Xe  rT , S  X ] (14.2a) and
P ( S , T ; X )  max[0, Xe  rT  Se dT , X  S ] , (14.2b)
respectively. The put-call parity relation for European stock index options is
c( S , T ; X )  p( S , T ; X )  Se  dT  Xe  rT (14.3)
and the put-call parity relation for American stock index options is
Se  dT  X  C ( S , T ; X )  P ( S , T ; X )  S  Xe  rT . (14.4)
Note that in all of these relations, there is an implicit assumption that the dividend income received (or paid)
while the arbitrage portfolio is held is automatically reinvested in the stock index portfolio.

The lower price bounds for European stock index futures options are
c( F , T ; X )  max[0, e rT ( F  X )] (14.5a)
and
p( F , T ; X )  max[0, e  rT ( X  F )] , (14.5b)
for the call and the put, respectively. The lower price bounds for the American call and put options on stock
index futures are
C ( F , T ; X )  max[0, ( F  X )] (14.6a)
and
P ( F , T ; X )  max[0, ( X  F )] , (14.6b)
respectively. The put-call parity relation for European stock index futures options is
c( F , T ; X )  p( F , T ; X )  e  rT ( F  X ) , (14.7)
and the put-call parity relation for American stock index futures options is
Fe  rT  X  C ( F , T ; X )  P ( F , T ; X )  F  e rT X (14.8)
In all of these relations, there is an implicit assumption that the futures position is the rollover position
originally described in Chapter 3.
Index Options Versus Index Futures Options

For some of the stock indexes, like the S&P 500, index futures contracts, index option contracts, and index
futures option contracts trade simultaneously. For example, if the index options and the index futures options
have the same expiration date and if they are both European, it follows that
c( F , T ; X )  c( S , T ; X ) (14.9a)
and
p( S , T ; X )  p( F , T ; X ) (14.9b)
In this case, index options and index futures options are perfect substitutes for one another. If the options are
American, and if the dividend yield rate is below the riskless rate of interest, the price relations are
C ( F , T ; X )  C (S , T ; X ) (14.10a)
and
P( S , T ; X )  P( F , T ; X ) . (14.10b)
Violation of the conditions (14.9a) through (14.10b) implies that costless arbitrage profits may be earned.
It is also important to recognize that stock index options may be priced in relation to the futures contracts.
The put-call parity equations, (14.3) and (14.4), can be expressed in relation to the price of the index futures
contract (rather than bT the underlying index) by using the cost-of-carry relation, F  Sebt , to substitute for
S . For European index options, (14.3) becomes
c( F , T ; X )  p( F , T ; X )  e  rT ( F  X ) , (14.11)
and, for American index options, (14.4) becomes

Fe  rT  X  C ( F , T ; X )  P( F , T ; X )  Fe ( r d )T  Xe rT . (14.12)
These relations are particularly important for index option market makers who use index futures as a way of
laying off the risk of index option portfolios they hold. It is considerably less expensive to buy and sell
futures contracts on the index than it is the index portfolio itself.
Likewise, index futures options may be priced in relation to the underlying index. For European index
futures options, the relation is
c( S , T ; X )  p( S , T ; X )  Se  dT  Xe  rT (14.13)
and, for American index futures options, the relation is
Se  dT  X  C ( S , T ; X )  P( S , T ; X )  Se( r d )T  Xe  rT . (14.14)
14.3 VALUATION EQUATIONS

European Stock Index Options
Under the assumption of a constant, proportional dividend yield rate of the stock index portfolio, valuation
equations for European stock index options follow directly from equations (11.25) and (11.28) in Chapter 11.
The pricing equation of a European call option on a stock index136 is
c( S , T ; X )  Se dT N (d1 )  Xe  rT N (d 2 ) , (14.15)
and the pricing equation for a European put option on a stock index is
p( S , T ; X )  Xe  rT N (d 2 )  Se  dT N (d1 ) , (14.16)
ln( S / X )  [(r  d )  0.5 2 ]T

where d1  , and d 2  d1   T
 T
American Stock Index Options

With a constant proportional dividend yield rate, both the American call and the American put options on
stock indexes may be rationally exercised early. As was noted in Chapter 11, pricing equations for these
options have not been derived, and approximation methods must be used to price such options. Many
different approximation methods are possible. We discussed the binomial method in Chapter 13. It is well
suited for American-style options where the underlying commodity makes cash disbursements during the
option’s life. The simplest and quickest approximation method for pricing American-style options with a
constant, continuous cost-of-carry rate, however, is the quadratic approximation method of Barone-Adesi
and Whaley (1987). Below their method is adapted to the stock index and stock index futures option
valuation problems.
For an American call option on a stock index, the quadratic approximation
c( S , T ; X )  A2 ( S / S ) 2 if S  S
* q *
C (S , T ; X )   (14.17)
 S  X if S  S *
where
136
This model is called the constant, proportional dividend yield model and first appeared in Merton (1973).
S *{1  e  dT N [d1 ( S * )]} ln( S / X )  [(r  d )  0.5 2 ]T

A2  , d1 ( S )  ,
q2  T
1  n  (n  1) 2  4k 2(r  d ) 2r
q2  , n , k
2 2  (1  e rT )
2
S * is the critical index level above which the American call should be exercised immediately and is the
solution to
S *  X  c( S * , T ; X )  {1  e  dT N [d1 ( S * )]}S * / q2 , (14.17a)
where c( S , T ; X ) is the European call option formula (14.15). For an American put option on a stock index,
the quadratic approximation is
 p( S , T ; X )  A1 ( S / S ** ) q1 if S  S **
P(S , T ; X )   (14.18)
 X  S if S  S **
where
S **{1  e  dT N [d1 ( S ** )]} 1  n  (n  1) 2  4k

A1   , q1  ,
q1 2
S ** is the critical index level below which the American put should be exercised immediately and is the
solution to
X  S **  p ( S ** , T ; X )  {1  e  dT N [d1 ( S ** )]}S ** / q1 , (1 4.1 8a)
where p( S , T ; X ) is the European put option formula (14.16). All other notation is as defined for the
American call option.
EXAMPLE 14.1
Compute the price of a 90-day S&P 100 index put option with an exercise price of 350. The current S&P 100
index level is 355.00, the dividend yield on the index is 3.50 percent annually, and the volatility rate is 25
percent annually. The riskless rate of interest is 6 percent.
The S&P 100 index option is an American-style option, but, since the quadratic approximation
requires the value of the corresponding European option, we will compute it first. Applying
p  350e 0.035(90 / 365) N (d1 )  355e0.06(90 / 365) N (d 2 )
ln(355 / 350)  [0.06  0.035  0.5(0.25) 2 ](90 / 365)

where: d1   0.2260 and
(0.25) (90 / 365)
d 2  d1  (0.25) (90 / 365)  0.1018
The values of N (d1 ) and N (d 2 ) are 0.4106 and 0.4594, respectively, so the European put option price is
p = 344.860(0.4594) - 351-949(0.4106) = 13.918 .
Applying the quadratic approximation (14.18), we find that the critical index level, S ** , below which the
index put should be exercised immediately is 285.347. Since the index level is currently 355.00, we are not
in the early exercise region. The value of A, is 1.778; the value of q , is - 11.335. The approximate value of
the S&P 100 option is, therefore,
P ( S , T ; X )  p ( S , T ; X )  A1 ( S / S ** ) q1 = 13.918 + 1.778(355.00/285.347)-11.335 = 14.068 .
The value of the S&P 100 index put is 14.068, and the value of the right to early exercise is 0.15.
European Stock Index Futures Options

The cost of carrying a futures contract is zero, independent of whether it is a stock index futures, a foreign
currency futures, and so on. Using equations (11.25) and (11.28) from Chapter 11, the value of a European
call on all index futures contract is
c( F , T ; X )  e  rT [ FN (d1 )  XN (d 2 )] , (14.19)
and the value of a European put option on an index futures contract is
p( F , T ; X )  e  rT [ XN (d 2 )  FN (d1 )] , (14.20)
where
ln( F / X )  0.5 2T
d1  , and d 2  d1   T
 T
American Stock Index Futures Options

American options on stock index futures contracts can also be valued with the help of the quadratic
approximation. In the case of the American call option on stock index futures, the approximation is
c( F , T ; X )  A2 ( F / F ) 2 if F  F
* q *
C(F ,T ; X )   (14.21)
 F  X if F  F *
F *{1  e  dT N [d1 ( F * )]} ln( F / X )  0.5 2T
where A2  , d1 ( F )  ,
q2  T
1  1  4k 2r
q2  , k 2
2  (1  e rT )
F * is the critical futures price above which the American call should be exercised immediately and is the
solution to
F *  X  c( F * , T ; X )  {1  e  dT N [d1 ( F * )]}F * / q2 , (14.21a)
where c( F , T ; X ) is the European call option formula (14.19). For an American put option on a stock index
futures contract, the approximation is
 p ( F , T ; X )  A1 ( F / F ** ) q1 if F  F **
P( F , T ; X )   (14.22)
 X  F if F  F **
where
F **{1  e  dT N [d1 ( F ** )]} 1  12  4k

A1   , q1  ,
q1 2
F ** is the critical futures price below above which the American put should be exercised immediately and is
the solution to
X  F **  p( F ** , T ; X )  {1  e dT N [d1 ( F ** )]}F ** / q1 , (1 4.22a)
where p ( F , T ; X ) is the European put option pricing formula (14.20). All other notation is as defined for the
American call option.
EXAMPLE 14.2
Compute the price of a 90-day call option on the S&P 500 futures contract. Assume the call has an exercise
price of 380 and the current futures price is 400.00. The riskless rate of interest is 6 percent annually and the
volatility rate of the futures contract is 25 percent annually.
The S&P 500 futures option contract is an American-style option, so we use the quadratic
approximation. As an intermediate step, however, we first compute the value of the corresponding European-
style futures option contract using equation (14.19).
c  e 0.06(90 / 365) [400 N (d1 )  380e0.06(90 / 365) N (d 2 )]
ln(400 / 380)  0.5(0.25) 2 (90 / 365)

where: d1   0.475 and d 2  d1  (0.25) (90 / 365)  0.351
(0.25) (90 / 365)
The values of N (d1 ) and N (d 2 ) are 0.683 and 0.637, respectively, so the European call option price is
c = 394.126(0.683) - 374.419(0.637) = 30.68 .
Applying the quadratic approximation (14.21), we find that the critical index level, F ** , above which the call
should be exercised immediately is 491.458. Since the futures price is currently 400.00, we are not in the
immediate early exercise region. The value of A2 is 1.270, the value of q2 is 11.945. The approximate value
of the S&P 500 futures option is, therefore,
C ( F , T ; X )  c( F , T ; X )  A2 ( F / F * ) q2 = 30.68 + 1.370(400.00/491.458)1.94 = 30.80 .
The value of the call option on the S&P 500 futures is 30.80. Hence the value of the right to early exercise is
about 12¢.
14.4 DISCRETE DIVIDENDS AND THE S&P 100 INDEX OPTION
Discrete Dividends
For many stock indexes, the assumption that dividends are paid at a constant, continuous rate is unrealistic.
To value S&P 100 index options precisely, for example, it is necessary to account for the fact that the
dividends on the index are discrete and seasonal in their nature. Finns tend to pay dividends in quarterly
cycles and generally use a December 31 fiscal year-end. For these reasons, monthly seasonality in index
dividend payments is expected.
Figure 14.1 summarizes the monthly dividend patterns of the S&P 100 index during the period March
1983 through December 1989.137 It shows that the most popular ex-dividend months are February, May,
August, and November. The average daily dividend payments are higher in these months, and the frequency
of zero-dividend days is lower. Of these four months, the average daily dividend during the month of
February is highest, 6.090, probably as a result of the fact that extra dividend payments are typically declared
in the last quarter of the fiscal year. Of the total number of February trading days in this sample, less than 32
percent have zero dividends. The least popular months are January, April, July, and October. During these
months, the average daily dividend is less than half of the average dividend across all days. In addition, the
frequency of zero-dividend days during these months exceeds 55 percent of the total trading days.
Figure 14.1 also shows that average dividends have generally increased over the seven-year sample
period. In August 1983, for example, the average daily payout was 4.450, and, in August 1989, the average
payout was 6.700. The growth in the payouts also occurred in non-peak months. In April 1983, the average
daily dividend was 0.430, and, in April 1989, the average dividend was 2.150.
Aside from the monthly seasonal pattern in S&P 100 dividends, there is also a daily seasonal pattern.
Figure 14.2 presents average dividends by day of the week. Monday has the largest average payment, 4.940.
In addition, of the total number of Mondays in the sample period, only 18 percent are zero-dividend days. At
the other extreme, Wednesday appears to be the least popular day to pay dividends. In more than 55 percent
of the total number of Wednesdays during the sample period, no dividends are paid. The average dividend
payment across all Wednesdays is 1.760. Moreover, Wednesday’s popularity as an ex-dividend day appears
to be declining over the sample period. Regarding the remaining days of the week, Friday is more popular
than Tuesday, which, in turn, is more popular than Thursday. For Tuesdays and Thursdays, there are more
zero-dividend days than non-zero-dividend days.
The above dividend descriptions are written in terms of average dividend amounts. The importance of
dividends in S&P 100 index option valuation has to do not only with the average level of dividends but also
the magnitudes of individual daily dividend payments. Many cash dividends on the S&P 100 index are quite
large. The largest cash dividend on the index during the March 11, 1983December 29, 1989, sample period
was 55.130 on Thursday, November 5, 1987.
On Friday, December 23, 1983, the cash dividend on the S&P 100 index was 46.830, and, on Thursday, May
7, 1987, it was 46.100. The holder of an S&P 100 index call, for example, may find it optimal to exercise her
option the day immediately before such a large dividend is to be paid, and the holder of an S&P 100 index
put may find it optimal to exercise his option just after.
137
This figure and the dividend information were obtained from Harvey and Whaley (1992b).
S&P 100 Index Option Valuation

Assuming that the cash dividend series for the S&P 100 index is known, call and put options on the S&P 100
index can be quickly and efficiently valued using the dividend-adjusted binomial method described in the
last chapter. The binomial method accounts for the possibility that large cash dividends may induce exercise
of the option prior to expiration. Since the approximation method was described in detail in the last chapter,
no further description is provided here.
While the dividend-adjusted binomial method accounts for the possibility of early exercise from a
theoretical perspective, it is useful to know that early exercise is a practical matter. Figures 14.3 and 14.4
show the number of call and put options on the S&P 100 index that were exercised early during the period
March 12, 1983, through November 30, 1989. Figure 14.3 shows that considerable early exercise activity for
calls occurs in the last week before expiration, although exercise activity extends out to as much as 134 days
prior to expiration. It is doubtful that the early exercise activity thirty days or more before expiration is
rational. However, the sheer number of exercises in the weeks prior to expiration indicate that early exercise
of index call options is not uncommon. Indeed, it may be likely if a large dividend is paid during the option’s
remaining life.
Figure 14.4 shows the number of S&P 100 index put contracts exercised prior to expiration. Early
exercise of index puts is more frequent than calls. Early exercise activity for puts in the weeks prior to
expiration appears commonplace and extends as far out as 109 days.
The Wildcard Option

While some of the early exercise activity shown in Figures 14.3 and 14.4 may be dividend-induced, some
may also be attributable to the wildcard option embedded in the S&P 100 option contract. The wildcard
option arises because the proceeds from exercise are based on the difference between the index level
established at 3:00 PM CST when the NYSE closes and the exercise price of the options. However, the index
option market stays open until 3:15 PM, giving the option holder until 3:15 PM to decide whether to exercise
the option. The fact that stock prices can move during the fifteen-minute interval after the NYSE close gives
the option holder a “wildcard option,” that is, in the case of a put, an option providing the right to put the
index option to the writer after 3:00 PM for cash proceeds established at 3:00 PM.
14.5 PORTFOLIO INSURANCE

Portfolio insurance protects the principal value of a portfolio against price declines. C3 4 In effect, a
portfolio insurance policy is a put option for which the exercise price of the put is the principal value of the
portfolio. The portfolio being insured may consist of any number of assets or securities, although it is most
usually discussed in terms of stock portfolios. The discussion in this section focuses on stock portfolio
insurance.
Portfolio insurance may be purchased directly in the form of exchange-traded options or indirectly
through dynamic trading strategies. The first is static portfolio insurance, where the fund manager simply
purchases exchange-traded put options. The second is dynamic portfolio insurance, where the fund manager
mimics the payoffs of the put by dynamically rebalancing the composition of the portfolio between stocks
and short-term debt as the market goes up or down.
Static Portfolio Insurance

Static portfolio insurance of a stock portfolio involves the purchase of a protective put option. If the value of
the stock portfolio is S and the price of a European put option with exercise price X and time to expiration
T is p( S , T ; X ) , the terminal value of the insured portfolio is as depicted in Table 14.4. Note that with
portfolio insurance, the value of the overall portfolio does not fall below X , the exercise value of the put or,
in the context of this application, the face amount of the insurance policy. The dividend income on the stock
portfolio is assumed to be zero for the sake of simplicity.
To illustrate static portfolio insurance, consider an index portfolio with a current value of 100.
Suppose the riskless rate of interest is 8 percent, the standard deviation of the instantaneous rate of return of
the index portfolio is 20 percent, and the stock portfolio pays no dividends. To insure the value of this
portfolio, the fund manager would buy a European put option on the index (we are assuming such an option
exists), where the exercise price of the put is 100 and the time to expiration of the put is six months. Note
that the insurance policy that the manager is choosing insures that the value of the overall fund is at least 100
at the end of six months.
Now, let us consider both the existing value of the insured portfolio as well as the value of the
portfolio if the index level rises or falls immediately.
TABLE 14.4 Terminal value of stock portfolio with static portfolio insurance.
Terminal Value
Initial
Position ST  X ST  X
Value
Own Stock Portfolio S ST ST
Buy Index Put p X  S T
Net Portfolio Value S p X ST
These values are represented in Table 14.5. The current position is where the index level is 100. The
increments up (down) from 100 in the first column of the table are 5 percent of the preceding (succeeding)
index level. The second column is the value of the European put at the different index levels. Note that the
original price of the portfolio insurance is 3.79. Finally, the last column is the overall value of the insured
portfolio. Note that the overall portfolio value never goes below 96.08. This value is the present value of 100
over the six-month interval, that is, 96.08e0.08(0.5)  100 . Recall that the policy the manager purchased insured
a portfolio value of at least 100 at the end of six months.
Dynamic Portfolio Insurance

Dynamic portfolio insurance does the same thing as the static insurance, except a put option is not purchased
directly. Instead, the fund manager dynamically rebalances a portfolio consisting of stocks and short-term
riskless debt such that the payoff contingencies of the portfolio are exactly the same as the payoff
contingencies of the portfolio with static insurance.
To be more precise, under the static portfolio insurance scheme above, the value of the insured portfolio is
S  p . If we substitute the value of a European put option on the index (14.16), we get
S  p  Xe  rT N (d 2 )  SN (d1 )  S[1  N (d1 )]  Xe rT N (d 2 )  SN (d1 )  Xe  rT N ( d 2 )

 Sw1  Xe  rT w2 , (14.23)
where w1  N (d1 ) is the number of units of the index and w2  N (d 2 ) is the number of units of T-bills.
Note that both w, and W2 Will change as the index level, S, and the time to expiration, T , change. In fact, in
theory, the weights must change as soon as we have an infinitesimal change in either variable.
TABLE 14.5 Insured stock portfolio value at alternative stock index levels, using static portfolio
insurance.138
Index Level Put Option Portfolio Value
S p Sp
59.87 36.21 96.0’8
63.02 33.06 96.08
66.34 29.75 96.09
69.83 26.29 96.13
73.51 22.70 96.21
77.38 19.03 96.41
81.45 15.38 96.83
85.74 11.87 97.61
90.25 8.67 98.92
95.00 5.94 100.94
100.00 3.79 103.79
105.00 2.29 107.29

110.25 1.27 111.52
115.76 0.65 116.42
121.55 0.31 121.86
127.63 0.13 127.76
134.01 0.05 134.06
140.71 0.02 140.73
147.75 0.01 147.75
155.13 0.00 155.13
162.89 0.00 162.89
To demonstrate the dynamic portfolio insurance scheme more clearly, consider Table 14.6. The first column
is the index level and the second column is the price of the short-term debt instrument. The third and fourth
columns are the portfolio weights w1 and w2 , and the final column is the portfolio value. Note that as -AV
the index level falls, funds are removed from stocks and placed in T-bills, and, as the index level rises, funds
are moved from T-bills and placed in stock. Note also that the portfolio values are identical to the values in
Table 14.5, showing that the two portfolio insurance schemes provide identical results in theory. The
problem with the dynamic approach is that, from a practical perspective, continuous rebalancing is not
possible.
Table 14.7 gives some perspective on what happens when the portfolio is not rebalanced. The
weights are established when the index level is 100 and are held at that level independent of the direction the
index moves. Note that when the index falls, we do not remove funds from stocks to place in T-bills, so the
portfolio value falls below the present value of the 100 we want to have on hand in six months. On the other
138
In the valuation of the European put, the index is assumed to pay no dividends, the riskless rate of interest is 8 percent, and the
standard deviation of the index return is 20 percent. The put option has an exercise price of 100 and a time to expiration of six
months.
hand, if the index level rises, we do not profit by as much because we continue to hold a large proportion of
T-bills.
TABLE 14.6 Insured stock portfolio value at alternative stock index levels, using dynamic
portfolio insurance with no rebalancing with continuous rebalancing.139
Index Level T-Bill Price Stock Portfolio Weight T-Bill Weight
 rT Portfolio Value
S Xe w1 w2
59.87 96.08 0.001 1.000 96.08
63.02 96.08 0.002 0.999 96.08
66.34 96.08 0.005 0.996 96.09
69.83 96.08 0.014 0.990 96.13
73.51 96.08 0.034 0.975 96.21
77.38 96.08 0.072 0.945 96.41
81.45 96-08 0.136 0.892 96.83
85.74 96.08 0.231 0.809 97.61
90.25 96.08 0.355 0.696 98.92
95.00 96-08 0.496 0.560 100.94
100.00 96.08 0.638 0.416 103.79
105.00 96.08 0.758 0.289 107.29

110.25 96.08 0.852 0.183 111.52
115.76 96.08 0.918 0.106 116.42
121.55 96.08 0.959 0.056 121.86
127.63 96.08 0.981 0.026 127.76
134.01 96.08 0.992 0.011 134.06
140.71 96.08 0.997 0.004 140.73
147.75 96.08 0.999 0.001 147.75
155.13 96.08 1.000 0.000 155.13
162.89 96.08 1.000 0.000 162.89
As a compromise between continuous rebalancing and no rebalancing, dynamic portfolio insurers usually
rebalance when the index level moves by a certain percent from the time the portfolio was last rebalanced. In
Table 14.8, we simulate this activity using a 5-percent trigger point. When the stock index is at a level of
100, we have the original portfolio weights seen in the previous tables. If the index moves down by 5
percent, for example, the new portfolio value will be 100.59 (i.e., 0.638 X 95.00 + 0.416 X 96.08), and the
portfolio will be rebalanced with the new weights, 0.495 and 0.558, for stocks and T-bills, respectively. Note
that this scheme does not completely insure downside protection. If the stock index level falls 15 percent, the
portfolio value falls below the 96.08 necessary to insure that 100 is on hand in six months. On the other side,
because we are slow to rebalance when the index level is going up, the upside potential is not as great as it
was in the continuous rebalancing case.
139
months.
TABLE 14.7 Insured stock portfolio value at alternative stock index levels, using
dynamic portfolio insurance with no rebalancing.
Stock Portfolio T-Bill
Index Level T-Bill Price Weight Weight Portfolio Value
S Xe  rT w1 w2
59.87 96-08 0.638 0.416 78.18
63.02 96.08 0.638 0.416 80.19
66.34 96.08 0.638 0.416 82.31
69.83 96.08 0.638 0.416 84.53
73.51 96.08 0.638 0.416 86.88
77.38 96.08 0.638 0.416 89.35
81.45 96-08 0.638 0.416 91.95
85.74 96.08 0.638 0.416 94.68
90.25 96-08 0.638 0.416 97.56
95.00 96.08 0.638 0.416 100.59
100.00 96.08 0.638 0.416 103.79
105.00 96-08 0.638 0.416 106.98

110.25 96.08 0.638 0.416 110.33
115.76 96.08 0.638 0.416 113.84
121.55 96.08 0.638 0.416 117.54
127.63 96-08 0.638 0.416 121.42
134.01 96.08 0.638 0.416 125.49
140.71 96-08 0.638 0.416 129.76
147.75 96.08 0.638 0.416 134.25
155-13 96.08 0.638 0.416 138.97
162.89 96.08 0.638 0.416 143.92
a. In the valuation of the European put, the index is assumed to pay no dividends, the riskless rate of interest
is 8 percent, and the standard deviation of the index return is 20 percent. The put option has an exercise price
of 100 and a time to expiration of six months.
14.6 SUMMARY
This chapter focuses on stock index option contracts. Options on stock indexes and on stock index futures
are traded in the U.S. In sections 2 and 3, we discuss the pricing principles for these options, where the cash
index is assumed to pay dividends at a constant, continuous rate through time.
TABLE 14.8 Insured stock portfolio value at alternative stock index levels, using
dynamic portfolio insurance with discrete rebalancing.140
Stock Portfolio
Index Level T-Bill Price T-Bill Weight
 rT
Weight Portfolio Value
S Xe w2
w1
59.87 96.08 0.001 0.985 94.65
63.02 96.08 0.002 0.984 94.66
66.34 96.08 0.005 0.982 94.67
69.83 96.08 0.014 0.976 94.72
73.51 96.08 0.034 0.961 94.85
77.38 96.08 0.071 0.933 95.12
81.45 96.08 0.135 0.882 95.67
85.74 96.08 0.229 0.802 96.65
90.25 96.08 0.353 0.691 98.24
95.00 96.08 0.495 0.558 100.59
100.00 96.08 0.638 0.416 103.79
105.00 96.08 0.755 0.288 106.98

110.25 96.08 0.847 0.183 110.94
115.76 96.08 0.911 0.105 115.61
121.55 96.08 0.951 0.055 120.89
127.63 96.08 0.973 0.026 126.67
134.01 96.08 0.984 0.011 132.87
140.71 96.08 0.988 0.004 139.46
147.75 96.08 0.990 0.001 146.42
155.13 96.08 0.991 0.000 153.73
162.89 96.08 0.991 0.000 161.41
Section 4 examines the effect of discrete cash dividend payments. The cash dividend payments on the S&P
100 index portfolio are shown to be discrete and seasonal. Some of the cash dividends are large enough to
induce early exercise. The wildcard option embedded in the S&P 100 option contract is also discussed.
The final section deals with portfolio insurance. Both static portfolio insurance using index put
options and dynamic portfolio insurance using portfolio rebalancing are discussed.
140
months. Rebalancing occurs if the index level moves by 5 percent.
15 INTEREST RATE AND INTEREST RATE FUTURES OPTIONS

CONTRACTS
15.1 INTEREST RATE OPTION MARKETS
15.2 SHORT-TERM INTEREST RATE OPTION PRICING

 Relation Between Price Volatility and Yield Volatility
15.3 TREASURY BOND OPTION PRICING-PRICE-BASED VALUATION
15.4 TREASURY BOND OPTION PRICING: YIELD-BASED VALUATION
15.5 MANAGING DURATION AND CONVEXITY
15.6 SUMMARY
15 INTEREST RATE AND INTEREST RATE FUTURES OPTIONS

CONTRACTS
Options on Treasury instruments began trading in October 1982. These instruments offer important new
ways for managing the interest rate risk of fixed-income portfolios. In particular, interest rate options provide
an effective means of managing the convexity risk-a subject first discussed in Chapter 8. In the first section
of this chapter, we describe the interest rate option and futures option markets that are currently active in the
United States. Of the exchange-traded option contracts, the T-bond futures option is clearly the most popular.
Options on T-bonds and T-notes are not as actively traded, although OTC markets for these contracts
continue to proliferate. We then proceed, in section 2, with a discussion of short-term interest rate option
valuation. Short-term interest rate options provide an interesting new valuation challenge. Using the standard
lognormal price distribution assumption from the previous chapters is clearly inappropriate because the
underlying asset value cannot exceed a predetermined level.141 In its place, we substitute the assumption that
yield is lognormally distributed and then rederive the European option valuation equations. The third and the
fourth sections discuss T-bond and T-bond futures option valuation under the lognormal price and the
lognormal yield assumptions, respectively. Section 5 is a detailed discussion of duration/convexity risk
management.
15.1 INTEREST RATE OPTION MARKETS

The first interest rate option contracts to trade in the United States were the Chicago Board of Trade’s
(CBT’s) Treasury bond futures option contract and the Chicago Mercantile Exchange’s (CME’s) Eurodollar
futures option contract on October 1, 1982. The Chicago Board Options Exchange (CBOE) introduced
options on Treasury bonds, and the American Exchange introduced options on Treasury notes and bills on
October 22, 1982. Other interest rate options have been introduced subsequently.
Table 15.1 contains a listing of interest rate instruments from the Wall Street Journal. Options on Treasury
bonds and notes are traded at the CBOE. These contracts expire on the Saturday after the third Friday of the
contract month, are American-style, and require the delivery of a specific Treasury bond or Treasury note.
Exchange-traded options on T-bonds and notes are not very active, however. In fact, no active contracts are
listed in Table 15. 1. The table reports only prices of relatively inactive CBOE options on short-term and
long-term bond indexes.
The most active interest rate options are those written on interest rate futures contracts. Of these, the
CBT’s Treasury bond futures option and the CME’s Eurodollar futures option contracts have the greatest
trading volume and open interest, as shown in Table 15.1. Upon exercising a T-bond futures option, a long or
short position in the nearby T-bond futures contract is assumed. These options are American-style and thus
may be exercised at any time up to and including the expiration day. The last day of trading is the Friday
preceding, by at least five business days, the first notice day for the corresponding T-bond futures contract.
In general, the first notice day of the futures is the first business day of the contract month.
The Eurodollar futures option is also American-style. The expiration day of the Eurodollar futures
option contract is the second London business day before the third Wednesday of the contract month, the
same as that of the underlying Euro-dollar future’s. Exercise of the Eurodollar option results in delivery of a
position in the Eurodollar futures contract of the same maturity. The Eurodollar futures contract, in turn,
fixes the price (or, equivalently, the yield) on a three-month Eurodollar deposit.
Table 15.2 on pages 370-371 shows the large number of Treasury issues outstanding on a given date.
Not all of these, nor even the majority of these, have exchange-traded options. If individuals want to buy or
sell options on particular bond issues, they usually go to OTC markets where bond option contracts can be
141
Recall that the assumption of a lognormal price distribution permits the price to rise without limit.
tailored in any manner. It is commonplace to see both European- and American-style OTC bond options,
including ones with times to expiration of several years.
In Chapter 8, we discussed a number of conventions regarding T-bonds and T-bond price reporting.
For example, the decimal part of the reported price represents 32nds. Thus, the reported bid price (in Table
15.2) of 107:07 for the 81/2s of May 1997 is actually 1077/32. Two other conventions are that the face value
of a Treasury bond is $100,000 and the bond price is reported as a percentage of par. Thus, the bid price of
the 8/2s of May 1997 is actually 1077/32 X $1000 or $107,218.75. Finally, the reported bond price does not
include the accrued interest for the current coupon period. For the 81/2s of May 1997, this means that from
the
TABLE 15.1 Interest rate options and futures

INTEREST RATE INSTRUMENTS. Wednesday, November 13,
1991. For Notes and Bonds, decimals in closing prices represent
32nds; 1.01 means 1 1/32. For Bills, decimals In closing prices
represent basis points; $25 per 0.01.
OPTIONS
CHICAGO BOARD
OPTIONS ON SHORT-TERM INTEREST RATES
Strike Price Nov Dec Jan Nov Dec Jan
45 1 5/16
Total call volume 45. Total call open Int. 1,787
Total put volume 0. Total put open Int. 130
IRX levels: High 47.50; Low 46.20; ciose 46.20, unch
OPTIONS ON LONG-TERM INTEREST RATES
75 1 5/16 7/16
771/2 1/2
Total call volume 17. Total call open Int. 2,012
Total put volume 6. Total put open Int. 1,470
LTX levels: High 75.40; Low 74.08; ciose 74.73, +0.58
TABLE 15.1 Cont. Interest rate options and futures

FUTURES
Yield Open
Settle Interest
TREASURY BONDS (CBT)-$100,000; Pts. 32nds of 10096
Dec 101-04 101-05 99-15 100-03 -29 7.991 +0.091 280,772
Mr92 101-04 100-10 98-20 99-07 -30 8.079 +0.095 31,420
June 99-12 99-12 97-30 98-10 -30 8.173 +0.097 10,197
Sept 98-13 98-13 97-06 97-14 -31 8.264 +0.101 2,717
Dec 97-19 97-19 96-10 96-20 -31 8.350 +0.102 4,487
Mr93 96-00 96-05 95-28 95-28 -32 8.430 +0.107 511
Est vol 370,000; Vol Tues 245,19 1 ; op Int 330,091, +8,975.

FUTURES
Yield Open
Settle Interest
TREASURY BONDS (MCE)-$100,000; Pts. 32nds of 10096
Dec 101-02 101-02 99-15 100-04 -32 7.987 +0.100 13,542
Est Vol 6,600; Vol Tues 6,396; open lnt 13,641, -132.
T-BONDS (LIFFE) U.S. $100,000; pts of 10096
Dec 101-02 101-03 99-21 100-05 0-23 101-03 96-24 5,443
Est Vol 2,273; Vol Tues 4,066; open Int 5,480, +721.
GERMAN GOV’T. BOND (LIFFE) 250,000 marks; $ per mark (0.01)
Dec 86.73 86.25 86.02 86.19 +0.02 86.44 83.73 75,176
Mar n.a. n.a. n.a. n.a. n.a. 86.70 85.39 6,879
Est Vol 47,392; Vol Tues 51,218; open lnt 82,055, -2,487.
TREASURY NOTES (CBT)-$100,000; pts. 32nds of 10096
Dec 103-23 103-73 102-18 103-06 -14 7.540 +0.061 86,289
Mr92 102-29 102-29 101-26 102-13 -14 7.651 +0.062 12,194
June … … … 101-18 -13 7.772 +0.058 418
5 YR TREAS NOTES (CBT)-$100,000; pts. 32nds of 10096
Dec 104-275 104-28 104-12 04-215 -5.5 6.880 +0.040 91,919
Mr92 104-015 104-015 103-19 03-275 -6.0 7.071 +0.045 10,105
2 YR TREAS NOTES (CBT)-$20,000, pts. 32nds of 10096
Dec 103-26 103-26 103-17 03-255 -1/4 … … 13,800
Mr92 103-11 103-11 03-057 03-105 -1/2 … … 3,553
Est Vol 1,500; Vol Tues 895; open lnt 17,353, +153.
30-DAY INTEREST RATE (CBT)-$5 million; pts. of 10096
Nov 95.14 95.14 95.13 95.14 -0.01 4.86 +0.01 1,254
Dec 95.12 95.12 95.08 95.09 -0.05 4.91 +0.05 1,232
Ja92 95.16 95.17 95.15 95.17 -0.05 4.83 +0.05 1,100
Feb 95.23 95.27 95.23 95.26 -0.04 4.74 +0.04 962
Mar 95.18 95.21 95.18 95.21 -0.05 4.79 +0.05 570
Apr 95.20 95.20 95.20 95.20 -0.05 4.80 +0.05 107
June 95.10 95.11 95.10 95.11 -0.04 4.89 +0.04 189
Est Vol 725; Vol Tues 529; open lnt 5,464, +185.
TREASURY BILLS (IMM)-$l mil.; pts. of 10096
Discount Open
Settle Interest
Dec 95.38 95.38 95.31 95.34 -0.03 4.66 +0.03 21,996
Mr92 95.52 95.52 95.42 95.50 -0.02 4.50 +0.02 79,371
June 95.23 95.34 95.23 95.32 -0.03 4.68 +0.03 3,8912
Sept 95.09 95.09 95.05 95.06 -0.02 4.92 +0.02 338
Dec 94.64 94.64 94.64 94.64 … 5.36 … 156
Est vol 6,378; Vol Tues 5,830; open lnt 55,771, + 261.

FUTURES
Open
Interest
LIBOR-1 MO. (IMM)-3,000,000; points of 10096
Nov 95.05 95.05 94.98 95.02 -0.04 4.918 +0.04 6,950
Dec 94.60 94.60 94.50 94.55 -0.10 5.45 +0.10 7,336
Ja92 95.11 95.11 95.02 95.06 -0.05 4.912 +0.05 9,651
Feb 95.01 95.08 94.99 95.06 -0.05 4.94 +0.05 2,515
Mar 94.96 95.01 94.94 95.00 -0.05 5.00 +0.05 1,390
Apr … … … 95.01 -0.04 4.99 +0.04 163
Est vol 1,577; vol Open Int 2,06,3; open lnt 78,005, + 300.
MUNI BOND INDEX (CBT)-$1,000; times Bond Buyer MBI
Dec 95-23 95-23 95-01 95-06 15 95-25 88-16 12,542
Mr92 95-04 95-04 94-07 94-13 19 95-04 88-00 827
The Index: ciose 95-09; Yield 6.82.
EURODOLLAR (IMM)-$1 million; Pts of 10096
Dec 94.85 94.8.6 94.75 94.80 -0.07 5.20 +0.07 242,049
Apr92 94.99 94.99 94.84 94.94 -0.05 5.06 +0.05 252,314
June 94.77 94.78 94.62 94.74 -0.04 5.26 +0.04 145,943
Sept 94.48 94.50 94.34 94.46 -0.03 5.54 +0.03 100,739
Dec 93.94 93.96 93.93 93.93 -0.03 6.07 +0.03 71,656
Mr93 93.76 93.17 93.65 93.74 -0.03 6.26 +0.03 55,473
June 93.46 93.46 93.36 93.43 -0.02 6.57 +0.02 44,243
Sept 93.17 93.19 93.09 93.17 -0.01 6.83 +0.01 31,658
Dec 92.79 92.82 92.72 92.82 +0.02 7.18 -0.02 21,132
Mr94 92.76 92.81 92.71 92.80 +.03 7.20 -0.03 27,888
June 92.50 92.58 92.46 92.57 +0.05 7.43 -0.05 17,956
Sept 92.28 92.37 92.23 92.35 +0.06 7.65 -0.06 11,828
Dec 91.97 92.06 91.92 92.05 +0.07 7.95 -0.07 9,816
Mr95 91.96 92.05 91.93 92.04 +0.07 7.96 -0.07 7,178
June 91.94 91.93 91.82 91.92 +0.07 8.08 -0.07 6,868
Sept 91.69 91.78 91.67 91.77 +0.07 8.23 -0.07 6,419
Est vol 283,796; vol Tues 130,7D9; open Int I,O:S3,2T7, + 4,629.

FUTURES
Lifetime
Open
Interest
EURODOLLAR (LIFFE)-$l million; Pt96 of 10096
Dec 94.86 94.87 94.75 94.93 -0.02 94.94 90.58 17,546
Mr92 94.97 94.99 94.95 94.96 … 95.06 90.60 10,098
June 94.77 94.78 94.66 94.75 … 94.83 90.97 5,226
Sept 94.49 94.49 94.47 94.46 … 94.53 90.97 2,724
Dec 93.97 93.97 93.95 93.93 … 94.00 91.54 614
Mr93 93.80 913.80 93.78 93.74 +0.02 93.80 91.55 545
June … … … 93.43 -0.23 93.44 9,2.60 405
Sept … … … 93.17 +0.08 93.09 92.82 137
STERLING (LIFFE) - GBP500,000; pts of 10096
Dec 89.79 89.82 89.78 89.81 +0.02 90.35 86.52 52,369
Mr92 90.25 90.29 90.24 90.28 +0.04 90.49 86.68 45,90
June 90.34 90.37 90.33 90.36 +0.03 90.46 87.45 34,626
Sept 90-31 90-33 90-29 90.32 +0.02 90.41 87.46 10,M
Dec 90.22 90.23 90.20 90.23 +0.02 90.32 87.55 6,664
Mr93 90.07 90.10 90.07 90.09 +0.02 90.16 87.50 4,548
June 89.97 89.97 89.97 89.98 +0.02 90.09 87.58 2,095
Sept 89.913 89.95 89.93 89.95 +0.02 90.08 88.20 1,746
Dec 89.88 89-90 89-88 89-92 +0.04 90.02 98.95 1,641
Est vol 20,529; vol 20,507; open lnt 160,391, - 1,238.
LONG GILT (LIFFE) –GBP 50,000; 32nds of 10096
Dec 95-17 95-23 95-09 95-17 +0-03 97-17 89-10 43,299
Mr92 95-24 95-24 95-18 94-22 +0-05 96-06 94-18 2,524
Est vol 24,368; vol Tues 23,953; open Int 45,823, +1,160.
OTHER INTEREST RATE FUTURES
Settlement Prices of selected contracts. Volume and open Interest of all contract months.
Mortgage-Backed (CBT)-$100,000, pts. & 64ths of 10096
Nov Cpn 8.5 102-04 -6; Est. vol. 0; Open Int. 90
5-Yr. Int. Rate Swap (CBT)-$25 per 1/2 b.p.; Pts of 10096
Dec 92.770 -0.010; Est. vol. 50; Open I nt. 707
3-Yr. Int. Rate Swap (CBT)-$25 per 1/2 b.p.; pts of 10096
Dec 93.490 -0.010; Est. vol. 0; Open Int. 456
Treas. Auction 5 Yr (FINEX)-$250,000, 100 minus yield
Dec 93.22 - 4.0; Est. vol. 100; Open Int. 4
CBT –Chicago Board of Trade. FINEX –Financial Instrument Exchange, a division of the New York Cotton
Exchange. IMM –International Monetary Market at Chicago Mercantile Exchange. LIFFE –London
International Financial Futures Exchange. MCE –MidAmerica Commodity Exchange

FUTURES OPTIONS
T-BONDS (CBT) $100,000; points and 64ths of 10096
916 4-07 3-57 3-48 c7 0-44 1-28
918 2-06 2-31 2-40 0-02 1-19 2-19
100 0-25 1-31 1-47 0-21 2-17 3-20
102 0-02 0-51 1-06 1-60 3-32 4-42
104 0-01 0-26 0-43 3-58 5-05 6-12
106 c2 0-12 0-25 5-58 6-54 7-54
Est. Vol. 100,000, Tues Vol. 57,179 calls, 29,063 puts
Open Interest Tue 351,393 calls, 279,518 puts
T-NOTES (CBT) $100,000; points and 64ths of 10096
101 2-13 2-14 0-01 0-53
102 1-14 1-38 0-02 1-13
103 0-23 1-08 0-11 1-45
104 0-04 0-48 0-55
105 0-02
106
Est. Vol. 7,5.00, Tues Vol. 2,665 calls, 1,850 puts
Open Interest Tues 38,546 calls, 34,9101 puts
MUNICIPAL BOND INDEX (CBT) $100,000; Pts. & 64ths of
100%
93 2-19 2-00 1-56 0-05 0-40 1-13
94 1-26 1-26 0-11 1-00
95 0-44 0-28
97 0-18 0-44 1-03 2-57
98
Est. vol. 10, Tues vol. 64 calls, 2 puts
5-YR TREAS NOTES (CBT) $100,000; Points and 64ths of 100%
10350 1-11 0-62 0-01 0-39
10400 0-44 0-45 0-01 0-54
10450 0-16 0-32 0-05
10500 0-02 0-24
10550 0-01 0- 14
10600
Est. vol. 2,500, Tues vol. 315 calls, 960 puts

FUTURES OPTIONS
5 YR INT. RATE SWAP (CBT) $12.50 Per 1/4 b.p.; pts of 100%
Calls-Last Puts – Last
9260
9270
9280
9290
9300
9310
Open Interest Tues 115 calls, 140 puts
EURODOLLAR (IMM) S million; pts. of 100%
9425 0.56 0.72 0.56 0.01 0.03 0.09
9450 0.32 0.50 0.39 0.02 0.06 0.16
9475 0.12 0.30 0.25 0.07 0.11 0.26
9500 0.02 0.15 0.14 0.22 0.21 0.39
9525 0.0004 0.07 0.06 0.45 0.38
9550 0.03 0.04 0.70
Est. vol. 62,887, Tues vol. 15,575 calls, 7,506 puts
It is important to review these pricing conventions since similar conventions are used in bond option pricing.
For example, prices of T-bond options traded at the CBOE are also reported in 32nds and as a percentage of
par. Thus, a reported T-bond option price of 2-24 means that the cost of the option is 2 24/32 x $1000, or
$2,750. To provide a finer demarcation in option price, some OTC T-bond option bond futures dealers quote
prices in 64ths. The CBOT uses a similar practice for T bond futures options.142
The accrued interest convention also has an impact on T-bond option pricing mechanics. Assume, for
example, that a call option with an exercise price of 107 is written on the 81/2s of May 1997. If this option
were to be exercised on November 13, 1991, the bondholder would pay the exercise price and receive a T-
bond that she could immediately sell for $111,422.55. The exercise price, however, is not simply the stated
exercise price times $1,000. If the call is exercised, the option holder must not only pay the stated exercise
price but also the accrued interest as of the exercise date. Thus, the total exercise price on November 13,
1991, is (107 + 4.20380) x $1000, or $111,203.80.
The price of the Eurodollar futures option is expressed in decimal form, representing basis points.
Exercising the Eurodollar futures option implies that a futures position is assumed. The December 94.75 call
option contract implies that the option holder may buy a Eurodollar futures contract at an index level of
94.75.143 Each basis point of the price of the option is worth $25, so the price of the December is 12 x $25, or
$300. (See Table 15.1.) The $25 value comes from 94.75 call the value of 0.01 percent of $1 million for
three months (i.e., .0001 x $1,000,000 x 90/360 = $25).
142
Recall that the CBOT's T-bond futures contract is quoted in 32nds.
143
The translation of the index level to the yield on the $1,000,000 Eurodollar deposit is provided in Chapter 8.
15.2 SHORT-TERM INTEREST RATE OPTION PRICING

The assumption that the underlying commodity has a lognormal price distribution at the option’s expiration
works well for most commodities, but it is inappropriate for short-term debt instruments such as a T-bill or
Eurodollar deposits. The lognormal distribution allows for the possibility of infinitely large prices. T-bills
and Eurodollar deposits, which generally mature three months after the option expiration, have a
predetermined future value. This fixed future value makes the possibility of an infinitely large price only
three months before maturity unreasonable.
To circumvent this problem, we assume that the yield, rather than the price, of the short-term debt
instrument is lognormally distributed at the option’s expiration. Under this assumption, the yield can fall to
zero, in which case the market price of the short-term debt instrument becomes its predetermined maturity
value. On the other hand, if the yield rises without limit, the market price of the debt instrument converges to
zero.
To make the valuation approach as specific as possible, we focus on the most popular short-term debt
option –the CME’s Eurodollar futures option. The pricing principles developed here, however, can easily be
extended to the other options on short-term debt instruments. Aside from the assumption of lognormal yield,
we invoke all of the same assumptions used in Chapter 11. In particular, the assumption of risk-neutral
pricing greatly simplifies the development.
Using the risk-neutral valuation approach, the value of the Eurodollar futures option today is the
present value of the expected terminal value, that is,
c( F , T ; X )  e rT E (cT ) (15.1)
The terminal value of the option is, in turn, a function of the Eurodollar futures index level, FT , that is,
 F  X if FT  X
cT   T (15.2)
0 if FT  X
TABLE 15.2 Treasury instruments.

TREASURY BONDS, NOTES & BILLS
GOVT. BONDS & NOTES
Maturity Ask
Rate Bid Asked Chg.
Mo/Yr Yld.
61/2 Nov 91n 100:00 100:02 0.00
81/2 Nov 91n 100:00 100:02 0.00
141/4 Nov 91n 100:01 100:03 0.00
73/4 Nov 91n 100:04 100:06 3.06
75/8 Dec 91n 100:11 100:13 4.24
81/4 Dec 91n 100:14 100:16 4.10
115/8 Jan 92n 101:03 101:05 -1 4.43
81/8 Jan 92n 100:22 100:24 -1 4.40
65/8 Feb 92n 100:13 100:15 4.65
91/8 Feb 92n 101:01 101:03 4.59
145/8 Feb 92n 102:13 102:15 -1 4.48
81/2 Feb 92n 101:01 101:03 4.61
TABLE 15.2 Cont. Treasury instruments.

GOVT. BONDS & NOTES
Maturity Ask
Rate Bid Asked Chg.
Mo/Yr Yld.
77/8 Mar 92n 101:03 101:05 -1 4.69
81/2 Mar 92n 101:11 101:13 4.63
113/4 Apr 92n 102:25 102:27 -2 4.72
87/8 Apr 92n 101:26 101:28 4.68
65/8 May 92n 100:27 100:29 4.77
91/8 May 92n 102:00 102:02 –1 4.78
133/4 May 92n 104:11 104:13 -1 4.73
81/2 May 92n 101:28 101:30 -1 4.93
81/4 Jun 92n 102:01 102:03 -1 4.82
83/8 Jun 92n 102:03 102:05 4.84
103/8 Jul 92n 103:16 103:18 4.98
8 Jul 92n 102:02 102:04 -1 4.92
41/4 Aug 87-92 98:13 99:13 -1 5.07
71/4 Aug 92 101:18 101:22 4.94
77/8 Aug 92n 102:00 102:02 -1 5.04
81/4 Aug 92n 102:09 102:11 -1 5.03
81/4 Aug 92n 102:11 102:13 -1 4.99
81/8 Sep 92n 102:18 102:20 -1 5.02
83/4 Sep 92n 103:03 103:05 -1 5.02
93/4 Oct 92n 104:04 104:06 -1 5.02
73/4 Oct 92n 102:12 102:14 -1 5.11
73/4 Nov 92n 102:15 102:17 -1 5.12
83/8 Nov 92n 103:02 103:04 -1 5.13
101/2 Nov 92n 105:04 105:06 -1 5.11
73/8 Nov 92n 102:06 102:08 -1 5.13
71/4 Dec 92n 102:06 102:08 5.17
91/I Dec 92n 104:06 104:08 -1 5.19
83/4 Jan 93n 103:28 103:30 -2 5.23
7 Jan 93n 101:30 102:00 -1 5.27
4 Feb 88-93 96:22 97:22 –1 5.94
63/4 Feb 93 101:22 101:26 -1 5.24
77/8 Feb 93 103:00 103:04 -1 5.26
81/4 Feb 93n 103:15 103:17 -1 5.30
83/8 Feb 93n 103:20 103:22 -1 5.29
107/8 Feb 93h 106:19 106:21 -2 5.31
63/4 Feb 93n 101:22 101:24 -2 5.33
71/8 Mar 93n 102:08 102:10 -2 5.36
95/8 Mar 93n 105:17 105:19 -1 5.35
73/4 Apr 93n 102:20 102:22 -1 5.38
7 Apr 93n 102:05 102:07 -1 5.40
75/8 May 93n 101:02 103:04 5.43
85/8 May 93n 104:15 104:17 5.44

GOVT. BONDS & NOTES
Maturity Ask
Rate Bid Asked Chg.
Mo/Yr Yld.
101/8 May 93n 106:18 106:20 -2 5.46
63/4 May 93n 101:26 101:28 -1 5.47
7 Jun 93n 102:08 102:10 -1 5.49
81/8 Jun 93n 103:31 104:01 -1 5.50
71/4 Jul 93n 102:21 102:23 -1 5.52
67/8 Jul 93n 102:02 102:04 -1 5.56
71/2 Aug 88-93 100:20 100:24 +2 7.04
8 Aug 93n 103:36 104:00 5.57
85/8 Aug 93 104:30 105:02 5.55
83/4 Aug 93n 105:05 105:07 5.58
117/8 Aug 93n 110:11 110:13 5.56
63/8 Aug 93n 101:09 101:11 5.58
61/8 Sep 93n 100:27 100:29 5.61
81/4 Sep 93n 104:19 104:21 -1 5.60
71/8 Oct 93n 102:21 102:23 -1 5.61
6 Oct 93n 100:21 100:23 5.61
73/4 Nov 93n 103:27 103:29 -1 5.66
85/8 Nov 93 105:15 105:19 +1 5.63
9 Nov 93n 106:05 106:07 -1 5.67
113/4 Nov 93n 111:10 111:12 -1 5.65
75/8 Dec 93n 103:24 103:26 -1 5.70
7 Jan 94n 102:15 102:17 -2 5.74
67/8 Feb 94n 102:09 102:11 5.75
87/8 Feb 94n 106:10 106:12 -3 5.81
9 Feb 94 106:18 106:22 -2 5.79
81/2 Mar 94n 105:24 105:26 -2 5.84
7 Apr 94n 102:19 102:21 -1 5.81
41/8 May 89-94 96:24 97:24 -3 5.09
7 May 94n 102:17 102:19 -2 5.87
91/2 May 94n 108:06 106:06 -3 5.90
131/8 May 94n 116:17 116:19 -3 5.89
81/2 Jun 94n 106:03 106:05 -1 5.93
8 Jul 94n 104:30 105:00 -2 5.95
67/8 Aug 94n 102:08 102:10 -3 5.95
85/8 Aug 94n 106:16 106:18 -2 6.00
83/4 Aug 94 106:25 106:29 -2 5.99
125/8 Aug 94n 116:17 116:19 -2 5.99
81/2 Sep 94n 106: 106:14 -1 6.03
91/2 Oct 94n 109:9 109:02 -2 6.06
6 Nov 94n* 100:00 100:01 -2 5.99
81/4 Nov 94n 105:27 105:29 -2 6.07
101/8 Nov 94 110:28 111:00 -3 6.06

GOVT. BONDS & NOTES
Maturity Ask
Rate Bid Asked Chg.
Mo/Yr Yld.
115/8 Nov 94n 114:31 115:01 -2 6.07
75/8 Dec 94n 104:10 104:12 -2 6.07
85/8 Jan 95n 106:30 107:00 -2 6.16
3 Feb 95 97:00 98:00 -3 3.66
73/4 Feb 95n 104:17 104:19 -3 6.17
101/2 Feb 95 112:10 112:14 -3 6.22
111/4 Feb 95n 114:17 114:19 -5 6.22
83/8 Apr 95n 106:06 106:10 -4 6.29
81/2 May 95n 106:24 106:26 -2 6.30
103/8 May 95 112:14 112:18 -3 6.32
111/4 May 95n 115:03 115:05 -4 6.35
125/8 May 95 119:14 119:18 -2 6.31
87/8 Jul 95n 107:28 107:30 -4 6.41
81/2 Aug 95n 106:24 106:26 -4 6.43
101/2 Aug 95n 113:07 113:09 -4 6.46
85/8 Oct 95n 107:06 107:10 -4 6.4
81/2 Nov 95n 106:28 106:30 -4 6.50
91/2 Nov 95n 110:10 110:12 -6 6.51
111/2 Nov 95 117:12 117:16 -4 6.47
91/4 Jan 96n 109:18 109:20 -5 6.57
71/2 Jan 96n 103:09 103:11 -5 6.58
77/8 Feb 96n 104:18 104:20 -5 6.61
87/8 Feb 96n 108:07 108:09 -6 6.61
71/2 Feb 96n 103:07 103:09 -5 6.61
73/4 Mar 96n 104:03 104:05 -5 6.64
93/8 Apr 96n 110:08 110:10 -5 6.64
75/8 Apr 96n 103:22 103:24 -5 6.64,
73/8 May 96n 102:22 102:24 -5 6.66
75/8 May 96n 103:21 103:23 -5 6.66
77/8 Jun 96n 104:18 104:20 -5 6.70
77/8 Jul 96n 104:19 104:21 -6 6.70
77/8 Jul 96n 104:23 104:25 -5 6.68
71/4 Aug 96n 102:06 102:08 -5 6.69
7 Sept 96n 101:06 101:08 -5 6.70
8 Oct 96n 105:06 105:08 -5 6.73
67/8 Oct 96n 100:26 100:28 -5 6.66
71/4 Nov 96n 102:04 102:06 -5 6.73
8 Jan 97n 105:04 105:06 -5 6.79
81/2 Apr 97n 107:07 107:09 -5 6.87
81/2 May 97n 107:07 107:09 -5 6.89
81/2 Jul 97n 107:07 107:09 -5 6.92
85/8 Aug 97n 107:26 107:28 -6 6.94

GOVT. BONDS & NOTES
Maturity Ask
Rate Bid Asked Chg.
Mo/Yr Yld.
83/4 Oct 97n 108:16 108:18 -4 6.96
87/8 Nov 97n 109:04 109:06 -7 6.97
77/8 Jan 98n 104:06 104:08 -7 7.01
81/8 Feb 98n 105:13 105:15 -6 7.03
77/8 Apr 98n 104:06 104:08 -5 7.04
7 May 93-98 100:16 100:24 6.47
9 May 98n 109:27 109:29 -8 7.07
81/4 Jul 98n 105:30 106:00 -7 7.10
91/4 Aug 98n 111:04 111:06 -8 7.13
71/8 Oct 98n 100:09 100:11 -5 7.06
31/2 Nov 98 97:24 98:24 +24 3.70
87/8 Nov 98n 109:09 109:11 -6 7.15
87/8 Feb 9,9n 109:09 109:11 -6 7.20
81/2 May 94-99 105:00 105:08 -4 6.20
91/8 May 99n 110:23 110:25 -7 7.24
8 Aug 99n 104:08 104:10 -6 7.26
77/8 Nov 99n 103:15 103:17 -6 7.28
77/8 Feb 95-00 102:15 102:19 -1 6.97
81/2 Feb 00n 107:01 107:03 -9 7.34
87/8 May 00n 109:06 109:08 -12 7.39
83/8 Aug 95-00 104:18 104:22 -3 6.93
83/4 Aug 00n 108:12 108:14 -13 7.42
81/2 Nov 00n 106:27 106:29 -13 7.43
73/4 Feb 0ln 101:31 102:01 -11 7.44
113/4 Feb 01 128:14 128:18 -10 7.43
8 May 01n 103:22 103:24 -10 7.44
131/8 May 01 138:04 138:08 -13 7.44
77/8 Aug 01n 102:29 102:31 -14 7.44
8 Aug 96-01 103:19 103:23 -15 7.07
133/8 Aug 01 140:16 140:20 -8 7.44
71/2 Nov 01n* 100:19 100:20 -11 7.41
153/4 Nov 01 157:21 157:25 -10 7.45
141/4 Feb 02 147:25 147:29 -14 7.48
115/8 Nov 02 130:10 130:14 -13 7.51
103/4 Feb 03 123:31 124:03 -12 7.54
103/4 May 03 124:06 124:10 -12 7.55
111/8 Aug 03 127:0.4 127:08 -15 7.58
117/8 Nov 03 133:07 133:11 -23 7.59
123/8 May 04 137:30 138:02 -29 7.60
133/4 Aug 04 149:07 149:11 -29 7.63
115/9 Nov 04 132:02 132:06 -23 7.67
81/4 May 00-05 104:10 104:14 -8 7.53

GOVT. BONDS & NOTES
Maturity Ask
Rate Bid Asked Chg.
Mo/Yr Yld.
12 May 05 135:23 135:27 -26 7.69
103/4 Aug 05 125:12 125:16 -19 7.71
93/8 Feb 06 114:14 114:18 -18 7.68
75/8 Feb 02-07 99:20 99:24 -16 7.65
77/8 Nov 02-07 101:14 101:18 -16 7.66
83/8 Aug 03-08 104:28 105:00 -12 7.72
83/4 Nov 03-08 107:19 107:23 -19 7.75
91/8 May 04-09 110:21 110:25 -24 7.76
103/8 Nov 04-09 120:28 121:00 -24 7.78
113/4 Feb 05-10 132:10 132:14 -25 7.78
10 May 05-10 118:15 118:19 -23 7.75
123/4 Nov 05-10 141:24 141:28 -29 7.79
137/8 May 06-11 152:06 152:10 -30 7.79
14 Nov 06-11 154:08 154:12 -31 7.79
103/8 Nov 07-12 122:13 122:17 -25 7.87
12 Aug 08-13 137:25 137:29 -30 7.88
131/4 May 09-14 150:07 150:11 -36 7.89
121/2 Aug 09-14 143:13 143:17 -34 7.90
113/4 Nov, 09-14 136:27 136:31 -35 7.87
111/4 Feb 15 135:01 135:05 -36 7.92
105/8 Aug 15 128:19 128:2 -3.1 7.92
97/8 Nov 15 120:23 120:25 -30 7.93
91/4 Feb 16 113:30 114:00 -33 7.94
71/4 May 16 92:21 92:23 -27 7.93
71/2 Nov16 95:07 95:09 -28 7.94
83/4 May17 108:25 108:27 -31 7.94
87/8 Aug 17 110:05 110:07 -31 7.94
91/8 May 18 113:00 113:02 -32 7.94
9 Nov 18 111:21 111:23 -32 7.94
87/8 Feb 19 110:09 110:11 -32 7.94
81/8 Aug 19 101:31 102:01 -30 7.94
81/2 Feb 20 106:06 106:08 -32 7.94
83/4 May 20 109:02 109:04 -32 7.94
83/4 Aug 20 109:02 109:04 -30 7.94
77/8 Feb 21 99:07 44:09 -30 7.94
81/8 May 21 102:04 102:06 -31 7.93
81/8 Aug 21 102:09 102:11 -30 7.92
8 Nov 21* 101:12 101:13 -28 7.88

Feb 92 ci 98:26 98:26 1 4.87
May 92 ci 97:18 97:19 1 4.99
Aug 92 ci 96: 11 96:11 1 5.04
Nov 92 ci 95:04 95:05 2 5.05
Feb 93 ci 93:19 93:20 1 5.38
May 93 ci 92:08 92:09 2 5.46
Aug 93 ci 90:26 90:27 3 5.59
Nov 93 ci 89:13 89:15 2 5.68.
Feb 94 ci 87:24 87:26 1 5.89
May 94 ci 86:13 86:15 1 5.93
Aug 94 ci 84:30 85:00 3 6.03
Nov 94 ci 83:12 83:15 3 6.15
Nov 94 np 83:08 93:11 3 6.20
Feb 95 ci 81:18 81:20 4 6.37
Feb 95 np 81:24 81:26 4 6.30
May 95 ci 80:02 80:05 5 6.45
May 95 np 80:01 80:03 4 6.47
Aug 95 ci 78:20 78:22 4 6.52
Aug 95 np 78:14 78:17 5 6.58
Nov 95 ci 77:08 77:11 5 6.56
Nov 95 np 77:02 77:05 5 6.62
Feb 96 ci 75:14 75:17 5 6.74
Feb 96 np 75:09 75:12 5 6.79
May 96 ci 74:04 74:07 5 6.77
May 96 np 74:01 74:05 6 6.79
Aug 96 ci 72:19 72:22 5 6.86
Nov 96 ci 71:14 71:17 6 6.94
Nov 96 np 70:10 70:13 3 7.17
Feb 97 ci 69:2b 69:23 4 7.02
May 97 ci 68:07 68:10 3 7.08
May 97 np 68:05 68:09 4 7.09
Aug 97 ci 66:27 66:31 4 7.13
Aug 97 np 66:26 66:29 3 7.14
Nov 97 ci 65:24 65:28 1 7.11
Nov 97 np 65:18 65:22 -4 7.16
Feb 98 ci 63:31 64:02 -8 7.28
Feb 98 np 63:23 63:27 -4 7.34
May 98 ci 62:24 62:28 -8 7.30
May 98 np 62:16 62:20 -4 7.36
Aug 98 ci 61:15 61:19 -8 7.34
Aug 98 np 61:06 61:10 -4 7.41
Nov 98 ci 60:12 60:16 -8 7.34
Nov 98 np 60:05 60:09 -7 7.39
Feb 99 ci 58:19 58:23 -7 7.51

Feb 99 np 58:14 58:18 -8 7.55
May 99 ci 57:14 57:18 -7 7.53
May 99 np 57:09 57:13 -8 7.57
Aug 99 ci 56:07 56:11 -7 7.57
Aug 99 np 56:0.4 56:08 7 7.59
Nov 99 ci 55:06 55:10 8 7.57
Nov 99 np 55:03 55:07 -7 7.59
Feb 00 ci 53:28 54:00 -8 7.64
Feb 00 np 53:22 53:27 -9 7.68
May 00 ci 52:28 53:00 -7 7.64
May 00 np 52:22 52:27 -7 7.68
Aug 00 ci. 51:26 51:30 -7 7.66
Aug 00 np 51:23 51:27 -7 7.68
Nov 00 ci 50:24 50:28 -8 7.68
Nov 00 np 50:24 50:28 -7 7.68
Feb 01 ci 49:14 49:19 -7 7.76
Feb 01 np 49:09 49:13 -8 7.80
May 01 ci 48:14 48:18 -8 7.78
May 01 np 48:08 48:12 -9 7.82
Aug 01 ci 47:16 47:21 -9 7.78
Aug 01 np 47:11 47:15 -8 7.82
Nov 01 ci 46:20 46:24 -7 7.78
Feb 02 ci 45:09 45:13 -12 7.88
May 02 ci 44:11 44:15 -11 7.90
Aug 02 ci 43:12 43:17 -12 7.92
Nov 02 ci 42:15 42:19 -11 7.94
Feb 03 ci 41:16 41:21 -12 7.97
May 03 ci 40:21 40:26 -12 7.98
Aug 03 ci 39:27 39:31 -11 7.99
Nov 03 ci 39:00 39:05 -12 8.00
Feb 04 ci 38:04 38:08 -11 8.03
May 04 ci 37:09 37:13 -12 8.45
Aug 04 ci 36:16 36:21 -12 8.06
Nov 04 ci 35:24 35:28 -12 8.07
Nov 04 bp 35:23 35:27 -11 8.08
Feb 05 ci 34:31 35:03 -11 8.09
May 05 ci 34:08 34:12 -11 8.10
May 05 bp 34:09 34:13 -11 8.09
Aug 05 ci 33:18 33:22 -11 8.10
Aug 05 bp 33:19 33:24 -12 8.09
Nov 05 ci 32:29 33:01 -11 8.10
Feb 06 ci 32:05 32:10 -12 8.12
Feb 06 bp 32:04 32:08 -11 8.13

May 06 ci 31:17 31:21 -11 8.12
Aug 06 ci 30:29 31:01 -11 8.12
Nov 06 ci 30:10 30:14 -11 8.12
Feb 07 ci 29:16 29:20 -12 8.17
May 07 ci 28:29 29:01 -12 8.17
Aug 07 ci 28:11 28:15 -12 8.17
Nov 07 ci 27:25 27:29 -12 8.17
Feb 08 ci 27:03 27:07 -11 8.20
May 08 ci 26:15 26:19 -11 8.22
Aug 08 ci 25:30 26:02 -11 8.22
Nov 08 ci 25:14 25:18 -10 8.22
Feb 09 ci 24:25 24:29 -11 6.25
May 09 ci 24:10 24:13 -10 8.25
Aug 09 ci 23:26 23:30 -10 8.25
Nov 09 ci 23:11 23:15 -10 8.25
Nov 09 bp 22:30 23:02 -9 8.35
Feb 10 ci 22:28 23:00 -19 8.25
May 10 ci 22:13 22:17 -10 8.25
Aug 10 ci 21:30 22:01 -10 8.26
Nov 10 ci 21:15 21:19 -11 8.26
Feb 11 ci 21:02 21:05 -19 8.26
May 11 ci 20:20 20:24 -10 8.26
Aug 11 ci 20:07 20:11 -10 8.26
Nov 11 ci 19:26 19:30 -10 8.26
Feb 12 ci 19:12 19:16 -10 8.17
May 12 ci 19:00 19:03 -9 8.27
Aug 12 ci 18:20 18:23 -9 8.27
Nov 12 ci 18:08 18:11 -9 8.27
Feb 13 ci 17:28 17:31 -9 8.17
May 13 ci 17:16 17:20 -10 8.27
Aug 13 ci 17:05 17:09 -9 8.27
Nov 13 ci 16:26 16:30 -9 8.27
Feb 14 ci 16:15 16:19 -9 8.27
May 14 ci 16:05 16:08 -9 8.27
Aug 14 ci 15:26 15:30 -9 8.17
Nov 14 ci 15:16 15:20 -9 8.27
Feb 15 ci 15:06 15:10 -9 8.27
Feb 15 bp 15:07 15:11 -8 8.26
May 15 ci 14:29 15:00 -8 8.27
Aug 15 ci 14:19 14:22 -9 8.27
Aug 15 bp 14:20 14:23 -8 8.26
Nov 15 ci 14:10 14:13 -8 8.27
Nov 15 bp 14:11 14:14 -7 8.26

Feb 16 ci 14:00 14:04 -9 8.27
Feb 16 bp 14:03 14:06 -8 8.26
May 16 ci 13:23 13:27 -9 8.27
May 16 bp 14:00 14:03 -8 8.19
Aug 16 ci 13:15 13:18 -8 8.27
Nov 16 ci 13:06 13:09 -6 8.27
Nov 16 bp 13:12 13:15 -6 8.21
Feb 17 ci 12:31 13:02 -8 8.26
May 17 ci 12:22 12:25 -8 8.26
May 17 bp 12:26 12:29 -8 8.26
Aug 17 ci 12:14 12:17 -8 8.26
Aug 17 bp 12:18 12:21 -8 8.22
Nov 17 ci 12:05 12: 08 -8 8.27
Feb 18 ci 11:31 12:02 -8 8.25
May 18 ci 11:24 11:26 -7 8.25
May 18 bp 11:27 11:30 -8 8.21
Aug 18 ci 11:16 11:19 8 8.25
Nov 18 ci 11:09 11:12 7 8.25
Nov 18 bp 11:12 11:15 9 8.21
Feb 19 ci 11:03 11:06 8 8.23
Feb 19 bp 11:08 11:11 8 8.18
May 19 ci 10:29 11:00 8 8.22
Aug 19 ci 10:22 10:25 8 8.22
Aug 19 bp 10:27 10:30 -7 8.17
Nov 19 ci 10:16 10:19 -8 8.21
Feb 20 ci 10:11 10:13 -7 8.20
Feb 20 bp 10:14 10:17 -7 8.16
May 20 ci 10:05 10:08 -7 8.19
May 20 bp 10:08 10:10 -7 8.18
Aug 20 ci 9:30 10:01 -7 8.19
Aug 20 bp 10:02 10:05 -9 8.15
Nov 20 ci 9:26 9:28 -7 8.17
Feb 21 ci 9:20 9:22 -7 8.17
Feb 21 bp 9:23 9:26 -8 8.13
May 21 ci 9:16 9:19 -7 8.14
May 21 bp 9:19 9:21 -7 8.11
Aug 21 ci 9:19 9:21 -7 8.04
Aug 21 bp 9:18 9:20 -9 8.05
Nov 21 bp 9:19 9:22 -6 7.97

TREASURY BILLS
Days
Mat.
Nov 21 ‘91 6 4.68 4.58 -0.02 4.65
Nov 29 ‘91 14 4.54 4.44 -0.01 4.52
Dec 05 ‘91 20 4.41 4.31 -0.03 4.39
Dec 02 ‘91 27 4.36 4.26 -0.08 4.34
Dec 09 ‘91 34 4.39 4.35 -0.05 4.43
Dec 26 ‘91 41 4.40 4.36 -0.04 4.45
Jan 02 ‘92 48 4.43 4.39 -0.02 4.49
Jan 09 ‘92 55 4.50 4.46 -0.02 4.57
Jan 16 ‘92 62 4.55 4.53 -0.01 4.63
Jan 23 ‘92 69 4.57 4.55 -0.01 4.67
Jan 30 ‘92 76 4.57 4.55 -0.01 4.67
Feb 06 ‘92 93 4.62 4.60 4.73
Feb 13 ‘92 44 4.63 4.61 4.73
Feb 20 ‘92 97 4.62 4.60 4.74.
Feb 27 ‘92 104 4.62 4.60 4.74
Mar 05 ‘92 111 4.6.4 4.62 4.76
Mar 12 ‘92 118 4.66 4.64 +0.01 4.79
Mar 19 ‘92 125 4.66. 4.64 +0.01 4.79
Mar 26 ‘92 132 4.64 4.62 4.78
Apr 02 ‘92 139 4.66 4.64 4.80
Apr 09 ‘92 146 4.69 4.67 4.84
Apr 16 ‘92 153 4.69 4-67 +0.01 4.84
Apr 23 ‘92 160 4.72 4.70 4.88
Apr 30 ‘92 167 4.68 4-66 4.84
May 07 ‘92 174 4.72 4.70 +0.02 4.89
May 14 ‘92 181 4.70 4.68 -0.01 4.87
Jun 04 ‘92 202 4.65 4.63 +0.02 4.82
Jul 02 ‘92 230 4.71 4.69 +0.01 4.89
Jul 30 ‘92 258 4.74 4.72 +0.03 4.93
Aug 27 ‘92 286 4.74 4.72 +0.02 4.94
Sep 24 ‘92 314 4.73 4.71 +0.04 4.94
Oct 22 ‘92 342 4.75 4.73 +0.03 4.98
If we assume the terminal futures price is lognormally distributed, we would evaluate E (cT ) in the same
manner we did in Chapter 11, and substitute this into (15.1). The valuation equation would be (11.25) with
the cost-of-carry rate, b , set to zero because the underlying instrument is a futures contract.
The assumption of lognormally distributed yield requires a modification of the terminal value
function, (15.2), for the call. In Chapter 8, we discussed the fact that the Eurodollar futures price is an index
level computed by subtracting the yield on the Eurodollar deposit from 100. In other words, the futures price
is F  100  y . If we substitute this definition into (15.2) and rearrange, we find that the terminal call price
can be expressed as
(100  X )  yT if yT  (100  X )

cT   (15.3)
0 if yT  (100  X )
But equation (15.3) looks surprisingly familiar. It is the terminal value function of a European put option,
where yT , has replaced ST and where ( 100  X ) has replaced X . Since yT is lognormally distributed, the
European put formula (11.28) of Chapter 11 can be applied directly. Using the fact that y  (100  F ) , the
expected terminal call price is
E (cT )  (100  X ) N (d 2 )  (100  F ) N (d1 ) , (15.4)

ln[(100  F ) /(100  X )]  0.5 T 2
where d1  y
, d 2  d1   y T
y T
and  y is the standard deviation of the logarithm of the yield ratios, ln( yt / yt 1 ) . Substituting (15.4) into
(15.1), the price of a European call option on a Eurodollar futures contract144 is
c( F , T ; X )  e  rT [(100  X ) N (d 2 )  (100  F ) N (d1 )] . (15.5)
By put-call parity for European futures options, the price of a European put option on a Eurodollar futures
contract is
p( F , T ; X )  e rT [(100  F ) N (d1 )  (100  X ) N (d 2 )] (15.6)
EXAMPLE 15.1
Using the values reported in Tables 15.1 and 15.2, compute the implied yield volatility of the March 95.00
call option on the Eurodollar futures contract. According to the tables, the call price is .15, the underlying
futures price is 94.94, and the riskless rate of interest is about 4.6 percent. As of November 13, 1991t the
option has 124 days remaining to expiration.
The implied yield volatility for this call is computed by solving
c( F , T ; X )  e  rT [(100  X ) N (d 2 )  (100  F ) N (d1 )]  0.15

0.15  e 0.046(124 / 365) [(100  95) N (d 2 )  (100  94.94) N (d1 )]
ln[(100  94.94) /(100  95)]  0.5 y2 (124 / 365)

where d1  , d 2  d1   y (124 / 365)
 y (124 / 365)
Without showing the steps of the iterative search that is used to find the implied volatility, the solution is
 y  10.26% .
Note that this volatility is upward biased since Eurodollar futures options are American style.
144
This approach to Eurodollar futures option valuation is described in detail in Emanuel (1985).
Relation Between Price Volatility and Yield Volatility

Prior to this chapter, volatility has been defined as the standard deviation of the logarithm of commodity
price ratios, ln( St / St 1 ) , or the standard deviation of the logarithm of futures price ratios, ln( Ft / Ft 1 ) . The
volatility parameter used in (15.5) and (15.6), however, is the standard deviation of the logarithm of yield
ratios, ln( yt / yt 1 ) . Since the scale of these two volatilities appears so different, it is important to understand
how these two measures are linked. The yield of the Eurodollar deposit is
y  (100  F ) , (15.7)
so the relation between a yield change and a price change is
dy  dF (15.8)
Multiplying the left-hand side by y / y and the right-hand side by F / F , we have

dy dF
y F
y F
which can be rearranged as
dy dF  F 
   (15.9)
y F  100  F 
In other words, the rate of change in yield equals the rate of change in the index level scaled by the factor
F /(100  F ) . The yield volatility,  y , therefore, equals the return volatility,  F , times the factor
F /(100  F ) , that is,
 F 
y F   . (15.10)
 100  F 
EXAMPLE 15.2
In Example 15.1, the implied yield volatility rate from the March 95 Eurodollar call is shown to be 10.26
percent. Compute the implied return volatility based on this estimate.
 94.94 
The implied futures price volatility is the solution to 0.1026   F   which implies that
 100  94.94 
 F  0.55% .
15.3 TREASURY BOND OPTION PRICING-PRICE-BASED VALUATION

In Table 15.1, only two options are bond options-the CBOE’s short-term and long-term interest rate index
contracts. These options are American-style and are written on specific Treasury issues (see Table 15.2 for
the price of the underlying T-bond or T-note). If we are willing to accept the assumption that long-term bond
prices are lognormally distributed at the option’s expiration,145 these options can be priced using the
continuous cost-of-carry commodity option framework developed in Chapters 10 and 11.
In doing so, we must account for the treatment of accrued interest. We noted earlier that the reported
bond price excludes accrued interest for the current coupon period and, to find the cost of the bond, the
145
For short-term options of a year or less, this assumption is plausible, particularly if the underlying bond has a long time to
maturity. For long-term options and/or for short-term T-bonds and T-notes, this assumption is less tenable.
accrued interest must be added to the reported bond price. We also noted that the exercise price of a bond
option is increased by the accrued interest. Since accrued interest is added to the reported bond price and to
the exercise price, one can simply ignore it and use the reported bond price and the stated exercise price in
the bond option pricing formula.
The option pricing formulas require the cost of carry for a bond, which is the short-term riskless rate
of interest less the coupon yield, that is, b  r  y . To compute the annualized coupon yield of a bond with
price B for use in the option pricing equations, recognize that a coupon payment, C , is received each half
year,
BC  C
e y (0.5)  or y  2 ln 1   (15.11)
B  B
It is worth noting that bonds with high coupon yields tend to depreciate in price (since they initially sell
above par) and that bonds with low coupon yields tend to appreciate in price (since they initially sell below
par). The value of an option, in turn, depends on the price appreciation or depreciation. The higher (lower)
the rate of price appreciation on the bond, the higher (lower) the call price and the lower (higher) the put
price. We now apply the commodity option pricing results of Chapters 10 and 11 using this cost-of-carry
parameter.
EXAMPLE 15.3
Assume there exists a European-style call option on the 8½ s of May 1997 that we discussed earlier in the
chapter. The call has an exercise price of 107, a time to expiration of 100 days, and a current market price of
$12 1/32. Assume that the risk-less rate of interest is 4.6 percent. Compute the implied volatility of this
option based upon the average of the bid and ask bond prices.
The average of the bid and ask prices for the 8½s of May 1997 is 107 8/32 or 107.25. The coupon
yield on this bond is
 4.25 
y  2 ln 1    7.77%
 107.25 
Substituting into valuation equation (11.25), we get
1.656  107.25e(0.0460.07770.046)(100 / 365) N (d1 )  107e 0.046(100 / 365) N (d 2 )]
ln(107.25 /107)  [0.046  0.0777  0.5 y2 ](100 / 365)
where d1  , d 2  d1   y (100 / 365)
 y (100 / 365)
Without showing the steps of the iterative search that is used to find the implied volatility, the solution is
  8.97%
By far the most active long-term interest rate option market is the CBT’s Treasury bond futures options.
These options are American-style, and expire on the first Friday preceding, by at least five business days, the
first notice day for the corresponding T-bond futures contract. Also, the T-bond futures option has the
decimal part of its price reported in 64ths. Under the assumption that the futures price at the option’s
expiration is lognormally distributed, the valuation of these options is possible using the quadratic
approximation described in Chapter 14.
EXAMPLE 15.4
Compare the theoretical price of a March 1992 T-bond futures put option, with a strike price of 100 to the
quoted price in Table 15.1, 2 17/64. The price of the March 1992 futures is 99 7/32, and, given that the
option expires on February 21, 1992, the time to expiration is 100 days. Assume the riskless rate of interest is
4.6 percent and the volatility rate is 9.00 percent.
The T-bond futures option contract is an American-style option, so we use the quadratic
approximation. We begin by computing the value of the corresponding European-style futures put contract
using equation (14.20).
p  e 0.046(100 / 365) [100 N (d1 )  99.21875 N (d 2 )]
ln(99.21875 /100)  0.5(0.09) 2 (100 / 365)

where: d1   0.143, d 2  d1  (0.09) (100 / 365)  0.190
(0.09) (100 / 365)
The European put option value is 2.32.

Applying the quadratic approximation (14.22), we find that the critical index level F ** below which
the put should be exercised immediately is 90.490, considerably below the current futures price of 99.21875.
Hence, the value of the early exercise premium should be small. The value of A1 is 0.092, the value of q1 is
29.619 . The approximate value of the March 100 put option on the T-bond futures is, therefore,
P ( F , T ; X )  p ( F , T ; X )  A1 ( F / F ** ) q1 = 2.32 + 0.092(99.21875/90.490) -29.619 = 2.33 .
To compare this theoretical value to the observed price, we need to transform the decimal price 2.33 to 64ths,
that is, 2 + (0.33 x 64)/64 = 2 21/64. In other words, the put option appears 4/64 underpriced.
15.4 TREASURY BOND OPTION PRICING: YIELD-BASED VALUATION

Yield-based option pricing of T-bond and T-bond futures is facilitated by using the binomial approximation
method described in Chapter 13. Instead of modeling movement of the underlying commodity price over the
next interval of time, we model the up and down movement of the yield. The next period up and down state
yields are a proportion of the current yield. If the current yield is y0 , the yield at the end of the first interval
is either uy0 , or dy0 . If the total number of time steps is defined as n , where t  T / n and T is the time to
expiration of the option, there are n  1 yield nodes at the option’s expiration, with an odd number if n is
even and an even number of nodes if n is odd.
This binomial lattice is illustrated in Figure 15.1. The length of each interval or time step in the figure is t .
The factors u and d are defined as
u  e t
(15.12a)
1
and d (15.12b)
u
The risk-neutral probabilities of up and down movements are
1 d
p (15.12c)
ud
and 1  p , respectively.
Once the yield lattice is computed, it is necessary to compute the bond price at each node. Bond
valuation equations were presented in Chapter 8. Keep in mind that as you move forward through the lattice
computing bond prices at each yield node, the time to maturity of the bond decreases.
Figure 15.2 shows a binomial lattice with both yield and price computed at each node. The
underlying bond is assumed to be an 8-percent coupon bond with 20 years to maturity. It currently sells at
par, so the current yield to maturity, y0 , is 8 percent. The riskless rate of interest, r , is 6 percent, and the
yield volatility,  y , is 50 percent. The time to expiration of the option, T , is 90 days, and the number of time
steps, n , is 90. The size of the time increment, t , is, therefore, one day or 0.00274 years.
FIGURE 15.1 Possible Paths that the Yield may follow under the
Binomial Model
Yield at the end of Interval:
0 1 2 3 4  n (even) n (odd)
u n y0 u n y0
u 4 y0   
3
u y0   
2 2
u y0 u y0   
uy0 uy0  uy0
y0 y0 y0  y0
dy0 dy0  dy0
2 2
d y0 d y0   
3
d y0   
4
d y0   
n n
d y0 d y0
Based on this information, the factors u and d are:

u  e0.50 0.00274
 1.026517 d  (1/1.026517)  0.974168
and
1  0.974168
Also, the probabilities of up and down movements are p   0.493458
1.026517  0.974168
and 1  p = 0.506542. Note that, in Figure 15.2, possible yields range from u n y0  8.00%(0.974168)90 
0.759% to d n y0  8.00%(1.026517)90  84.339% at the option’s expiration and bond prices range from
237.385 to 9.627.
With the yield/bond price lattice computed, the approximation method starts at the end of the option’s
life and works back to the present. At the end of the option’s life (column n in the figure), the option value at
each yield/bond price node is given by the intrinsic value of the option. In the case of a put option,
0 if Bn , j  X
Pn , j ( Bn , j )   (15.13)
 X  Bn , j if Bn , j  X
The option values one step, t , back in time (in column n  1 ) are computed by taking the present value of
the expected future value of the option. At any point j in column n  1 , the yield can move up with
probability p or down with probability 1  p . The value of the option at time n if the yield moves up is Pn , j 1 ,
and if the yield moves down is Pn , j . The present value of the expected future value of the option is, therefore,
( p ) Pn , j 1  (1  p) Pn , j
Pn1, j  (15.14)
r*
where r *  e rt . Using this present value formulation, all of the option values in column n I may be
identified.
FIGURE 15.2 Possible paths that yield/bond price may follow under the binomial model where the
current price of the 8 percent, 20-year bond is 100, the riskless rate of interest is 6 percent annually,
and the yield volatility is 50 percent annually. The time to expiration of the option is 90 days, and the
number of time steps is also 90. The time increment, t , is, therefore, 1 day or 0.00274 years.
Yield at the end of Interval:
0 1 2 3 4  n (even) n (odd)
8.43/95.94 8.43/95.94
8.88/91.93   
8.65/93.93   
84.34/9.63 84.34/9.63   
8.21/97.97 8.21/97.97  8.21/97.97
8.00/100.00 8.00/100.00 8.00/100.00  8.00/100.00
7.79/102.08 7.79/102.08  7.79/102.08
7.59/104.16 7.59/104.16   
7.40/106.26   
7.20/108.36   
0.76/237.38 0.76/237.38
Before proceeding back another time increment, t , in the valuation, it is necessary to see if any of
the computed option values are below their early exercise proceeds at the respective nodes, X  Bn1, j . If the
exercise proceeds are greater than the computed option value, the computed value is replaced with the early
exercise proceeds. If they are not, the value is left unchanged. Note that if this step is not performed, the
procedure will produce the value of a European put option.
Once the checks are performed, we go to column n  2 and repeat the steps, and so on back through
time. Eventually, we will work our way back to time 0, and the current value of the American put option (in
column 0) will be identified.
To complete the binomial method illustration, suppose that the bond price lattice shown in Figure
15.2 underlies a 90 -day American put option with an exercise price of 100. Applying the yield-based
binomial method, the value of the American put is $6.157. The value of the corresponding European-style
put option using the yield-based method is $5.713. The early exercise premium of the American put is,
therefore, worth about 44.40.
15.5 MANAGING DURATION AND CONVEXITY

In Chapter 8, we discussed the duration and convexity of a fixed-income portfolio. Modified duration
measures the percentage change in the value of a bond for a given change in yield,
dB / B
Dm   (15.15)
dy
and convexity indicates how duration changes for a given change in yield,
1 d 2B 1
Convexity  (15.16)
2 dy 2 B
The keys to these expressions are the first and second derivatives of bond price with respect to a change in
yield, that is, dB / dy and d 2 B / dy 2 , respectively. In this sense, duration and convexity are like the delta and
gamma of an option. In fact, we will now show that bond option deltas and gammas enable a fixed-income
portfolio manager to control the duration and convexity of the portfolio.
To understand how to tailor the duration and convexity exposure of a fixed-income portfolio, we need to
develop expressions for bond (or bond futures) option price changes as a function of yield changes. The first
derivative of option price with respect to a change in yield is
O O B B
  (15.17)
y B y y
where  is the delta value of the option. Thus, to change the dollar value exposure of a bond portfolio, we
simply combine the exposure in bonds, dB / dy , with the exposure in n bond options, dB / dy , that is,
dB
Desired dollar risk exposure = (1  n) . (15.18)
dy
To reduce the portfolio’s risk exposure to zero (for small changes in yields), the optimal number of options is
1
n

In the general case, one may use N different options to hedge a bond portfolio, that is,
Desired dollar risk exposure =

dB
dy

1   i 1 ni  i
N

(15.19)
Note that the dollar risk exposure is exactly zero only at the point where the derivative is taken.
EXAMPLE 15.5
Assume that a fixed income portfolio manager holds a 9-percent, 20-year bond whose current yield to
maturity is 8 percent. Its current market value is 109.82, its modified duration is 9.606, and its convexity is
70.450. Given the uncertainty about economic events, the manager decides to hedge the interest rate risk of
his portfolio by writing call options on this bond. The call has an exercise price of 110 and a time to
expiration of 3 months. The current price of this option is 2.900, its delta is 0.472, and its gamma is 0.0474.
Compute the number of call options to sell against this bond to immunize it from movements in the bond’s
yield.
1 1
The optimal number of calls to mitigate the duration exposure is: n    2.119 .
 0.472
If we sell this number of options against the underlying bond, the bond portfolio value for given changes in
the yield is shown in Figure 15.3.
Figure 15.3 demonstrates that the delta-neutral hedge reduces the range of possible portfolio values.
The unhedged portfolio ranges in value from 90 to 135 over the range of yields shown, while the hedged
portfolio ranges in value from 90 to 110. But, even with the delta-neutral hedge, the range of possible
portfolio values is large over this somewhat limited yield range. In addition, the hedged portfolio has reduced
value if the yield rises or falls. To improve upon this hedge, it is possible to use more than one option to
hedge both duration and convexity risk.
The zero-risk portfolio given by (15.19) is analogous to the delta-neutral portfolio discussed in Chapter 12.
In that chapter, we also noted that a change in delta brought about by a commodity price change introduces
gamma risk. A similar situation arises in the case of the hedged bond portfolio. Not only does the bond
option delta change as the yield changes, but so does the duration of the bond portfolio. To compensate for
these effects, we must consider the second partial derivative of the bond option price with respect to yield,
that is
 2O  Oy  B y
 
O B



O d 2 B dB  2O
  
d 2 B dB   OB 
  
d 2 B dB  B

y 2 y y y dy 2 dy By dy 2 dy y dy 2 dy y y
2
 d 2B   dB 
  2     (15.20)
 dy   dy 
where    / B is the gamma value of the option. Combining the convexity exposure of the bond with a
portfolio that consists of N bond options, we get dollar convexity exposure
Dollar convexity =
d 2B
dy 2
 
N
i 1
ni  i
 d 2B 
 2
 dy 
  
N
i 1
ni  i
 dB  d 2 B
 
 dy 

dy 2
1 
 
N
i 1
ni  i  f 
N
n
i 1 i i  (15.21)
  /   .
2
where f  dB
dy
d 2B
dy 2
EXAMPLE 15.6
Unsatisfied with the effectiveness of the hedge portfolio indicated in Example 15.5, the portfolio manager
decides to evaluate the effectiveness of a hedge of both the duration and the convexity risk of his position.
Aside from the 110 call described in the last exercise, a 105 call with three months to expiration is available.
The current price of the 105 call Is 5.720, its delta is 0.703, and its gamma is 0.0403. Compute the number of
calls to buy/sell against this bond to immunize it from movements in the bond’s yield.
To neutralize the duration and the convexity risk of the bond portfolio, we need to solve simultaneously the
following equations:
n11  n2  2  1 and n1 (1  f  1 )  n2 ( 2  f  2 )  1
To compute the coefficient f , we need to know the values of dB / dy , and d 2 B / dy 2 .
These can be obtained from the modified duration and the convexity figures reported for the bond. That is,
from equations (15.15) and (15.16), we know
dB
 BDm = (109.82)(9.606) = 1, 054.93 ,
dy
d 2B
and  2 BConvexity = 2  109.82  70.450 = 15, 473.64 .
dy 2
The value of f is, therefore, 71.921. The deltas and gammas of the individual options are known, and the
remaining task is only computational. The optimal composition of the duration-convexity hedge is to sell
3.315 105 calls and to buy 2.819 110 calls.
The effectiveness of this hedge relative to the unhedged portfolio and the duration-hedged portfolio from
Example 15.5 can be seen in Figure 15.4. The range of outcomes has been further diminished. Using both
calls generates a curve that is much more horizontal at 110, ranging from 103 to 116. In addition, the hedged
portfolio rises if the yield falls and falls more slowly if the yield rises. Clearly, this second hedge is more
effective than the hedge discussed in Example 15.5.
15.6 SUMMARY
This chapter focuses on the valuation of interest rate options. After reviewing the designs of the U.S.
exchange-traded interest option contracts, we discuss short-term interest rate option valuation. Using the
standard lognormal price distribution assumption is inappropriate for these options. In its place, we use the
assumption that yield is lognormally distributed and rederive the European option valuation equations. In the
third section, T-bond and T-bond futures option valuation under the lognormal price distribution assumption
is presented. The fourth section describes the same valuation but under the lognormal yield assumption. The
binomial method is also used. Finally, we show how T-bond option contracts can be used to control the
duration and the convexity of a fixed-income portfolio.
16 OTHER OPTION CONTRACTS
16.1 CURRENCY AND CURRENCY FUTURES OPTIONS

 Uses of Currency Options
16.2 PHYSICAL COMMODITY FUTURES OPTIONS
16.3 EXOTIC OPTIONS

 Options on the Maximum and the Minimum
 Options on Options
 Lookback Options
 Barrier Options
 Other Exotic Options
16.4 SUMMARY
16 OTHER OPTION CONTRACTS
The last three chapters have focused on specific option contracts on stocks, stock indexes, and interest rate
instruments. Where the valuation procedures of Chapters 10 and 11 did not directly apply to these specific
option contracts, the procedures were modified. For example, we showed how the quadratic approximation
could be used to price American-style index options and how the binomial method could be used to value
options on assets with discrete cash flows during the option's life.
In this chapter, we discuss options on currencies and on physical commodities. Since no new valuation
procedures are needed for these contracts, sections 1 and 2 review only the nature of the exchange-traded
options on currencies and physical commodities, and refer the reader to the appropriate valuation equations
from the previous chapters. The third section discusses some exotic OTC options that are currently traded.
Examples of these are options on options, options on the maximum and the minimum of two risky
commodities, lookback options and barrier options. For these four types of options, we present valuation
equations. Other types of exotic options include options on the average price of a commodity, deferred-start
options, deferred-payment American options, and all-or-nothing options. For these instruments, we describe
only the essence of the option contracts. The chapter concludes with a summary.
16.1 CURRENCY AND CURRENCY FUTURES OPTIONS
Option contracts on spot currencies have been actively traded on the Philadelphia Exchange since late 1982.
Both European-style and American-style option contracts are traded, with the American-style contracts
having the greatest trading volume and open interest. The most active contracts are for British pounds,
German marks, Japanese yen, and Swiss francs, although options on other currencies also trade. Table 16.1
contains a listing of the currently active, exchange-traded currency option contracts. The spot currency must
be delivered upon exercise of these options. The current spot prices are also reported in Table 16.1. Currency
options expire on the Saturday before the third Wednesday of the contract month.
Valuation of Currency Options

The valuation of currency options is relatively straightforward. First, all of the arbitrage pricing principles
developed in Chapter 10 apply. The cost-of-carry rate equals the difference between the domestic interest
rate, rd , and the foreign interest rate rf . Second, all of the valuation equations and approximations discussed
in Chapter 11 and the trading strategies discussed in Chapter 12 also apply, assuming that the spot exchange
rate has a lognormal price distribution at the option's expiration. For European-style currency options,
valuation equations (11.25) and (11.28) can be used, where b  rd  rf .146
For American-style currency options, the quadratic approximation from Chapter 13 is recommended.
Again, the cost-of-carry rate is set equal to the difference between the domestic and foreign interest rates.
Option contracts on currency futures were developed in late 1984 and early 1985 by the Chicago
Mercantile Exchange. The only such options available are American-style. Upon exercise, a long (call) or
short (put) position in the futures is obtained. The expiration date of these options is the second Friday before
the third Wednesday of the contract month. (The futures expires two business days before the third
Wednesday.) Like the currency options, the most active currency futures options are on British pounds,
German marks, Japanese yen, and Swiss francs. Table 16.2 contains current foreign currency futures and
futures option contract prices. The valuation of currency futures options is even more straightforward than
the valuation of currency options. The cost-of-carry rate for any currency futures is zero, so we set b = 0 in
all of the pricing relations of Chapters 10 and 11 and the trading strategies of Chapter 12.
146
For an approach to foreign currency option valuation that permits interest rates to be stochastic, see Grabbe (1983).
TABLE 16.1 Foreign currency exchange rates and foreign currency options.
CURRENCY TRADING
EXCHANGE RATES
Wednesday, November 13, 1991. The New York foreign exchange selling rates below apply
to trading among banks In amounts of $1 million and more, as quoted at 3 p.m. Eastern time
by Bankers Trust Co. and other sources. Retail transactions provide fewer units of foreign
currency Per-dollar.
Argentina (Austral) 0.0001008 0.0001008 9918.67 9918.67
Australia (Dollar) 0.7860 0.7870 1.2723 1.2706
Austria (Schilling) 0.08681 0.08681 11.52 11.52
Bahrain (Dinar) 2.6539 2.6539 0.3768 0.3768
Belgium (Franc) 0.02966 0.02966 33.72 33.72
Brazil (Cruzeiro) 0.00144 0.00146 694.71 695.60
Britain (Pound) 1.7730 1.7725 0.5640 0.5642
30-Day Forward 1.7648 1.7640 0.5666 0.5669
90-Day Forward 1.7504 1.74% 0.5713 0.5716
180-Day Forward 1.7299 1.7291 0.5781 0.5783
Canada (Dollar) 0.8842 0.8838 1.1310 1.1315
30-Day Forward 0.8815 0.8814 1.1344 1.1346
90-Day Forward 0.8784 0.8779 1.1384 1.1391
180-Day Forward 0.8737 0.8733 1.1445 1.1451
Chile (Peso) 0.002844 0.002780 351.56 359.65
China (Renminbi) 0.185642 0.185642 5.3867 5.3867
Colombia (Peso) 0.001753 0.001753 570.38 570.38
Denmark (Krone) 0.1573 0.1573 6.3570 6.3555
Ecuador (Sucre) Floating rate 0.000966 0.000966 1035.00 10335.00
Finland (Markka) 0.24984 0.24941 4.0025 4.0095
France- (Franc) 0.17881 0.17879 5.5925 5.5930
30-Day Forward 0.17813 0.17808 5.6140 5.6156
90-Day Forward 0.17690 0.17695 5.6529 5.6545
180-Day Forward 0.17510 0.17504 5.7110 5.7130
Germany (Mark) 0.6112 0.6111 1.6362 1.6365
30-Day Forward 0.6090 0.6088 1.6421 l.6426
90-Day Forward 0.6045 0.6044 1.6543 1.6544
180-Day Forward 0.5982 0.5902 1.6717 1.6718
Greece (Drachma) 0.005405 0.005405 185.00 185.00
Hong Kong (Dollar) 0.12884 0.12884 7.7615 7.7615
India (Rupee) 0.03890 0.03890 25.77 25.77
Indonesia (Rupiah) 0.0005056 0.0005056 1978.00 1978.00
Ireland (Punt) 1.6330 1.6318 0.6124 0.6128
Israel (Shekel) 0.4308 0.4321 2.3215 2.3142
Italy (Lira) 0.0006121 0.0006117 1231.41 1232.01
TABLE 16.1 Cont. Foreign currency exchange rates

Japan (Yen) 0.007686 0.007707 130.10 129.75
30-Day Forward 0.007678 0.007696 130.24 129.90
90-Day Forward 0.007666 0.007686 130.45 130.10
180-Day Forward 0.007656 0.007677 130.62 130.26
Jordan (Dinar) 1.4500 1.4500 0.6897 0.6897
Kuwait (Dinar) 3.4965 3.4965 0.2860 0.2860
Lebanon (Pound) 0.001134 0.001134 881.50 881.50
Malaysia (Ringgit) 0.3650 0.3647 2.7400 2.7400
Malta (Lira) 3.1250 3.1250 0.3200 0.3200
Mexico (Peso) Floating rate 0.0003254 0.0003254 3073.01 3073.01
Netherlands (Guilder) 0.5423 0.5423 1.9445 1.9445
New Zealand (Dollar) 0.5610 0.5620 1.7825 1.7794
Norway (Krone) 0.1558 0.1558 6.4175 6.4195
Pakistan (Rupee) 0.0405 0.0405 24.72 24.72
Peru (New Sol) 1.0152 1.0051 0.99 0.99
Philippines (Peso) 0.03839 0.03839 26.05 26.05
Portugal (Escudo) 0.007067 0.007063 141.50 141.59
Saudi Arabia (Riyal) 0.26663 0.26663 3.7505 3.7505
Singapore (Dollar) 0.5958 0.5959 1.6795 1.6780
South Africa (Rand) Commercial rate 0.3568 0.3574 2.8023 2.7961
Financial rate 0.3240 0.3240 3.0790 3.0790
South Korea (Won) 0.0013310 0.0013310 751.30 751.30
Spain (Peseta) 0.009723 0.009699 102.85 103.10
Sweden (Krona) 0.1673 0.1672 5.9775 5.9615
Switzerland (Franc) 0.6888 0.6892 1.4517 1.4510
30-Day Forward 0.6872 0.6675 1.4552 1.454
90-Day Forward 0.6835 0.6839 1.4631 1.4621
1WDay Forward 0.6788 0.6792 1.4732 1.4724
Taiwan (Dollar) 0.038650 0.037906 25.74 26.38
Thailand (Baht) 0.03926 0.03926 25.47 25.47
Turkey (Lira) 0.0002044 0.0002020 4892.01 4950.00
United Arab (Dirham) 0.2723 0.2723 3.6725 3.6725
Uruguay (New Peso) Financial 0.000425 0.000425 2352.94 2352.94
Venezuela (Bolivar) Floating rate 0.01695 0.01661 59.00 60.20
SDR 1.38023 1.38189 0.72452 0.72365
ECU 1.24952 1.25088
Special Drawing Rights (SDR) are based on exchange rates for the U.S., German, British, French and
Japanese currencies. Source: International Monetary Fund.
European Currency Unit (ECU) is based on a basket of community currencies. Source: European
Community Commission.
TABLE 16.1 Cont. Foreign currency options.

OPTIONS PHILADELPHIA
Wednesday, November 13, 1991
Option & Strike
Underlying Price
50,000 Australian Dollars-cents per unit.
Adollr 78 0.63 0.94 r r r r
78.66 81 r r r r r 3.53
78.66 93 r r r r 4.56 r
31,250 British Pounds-European Style.
BPound 150 r r 23.80 r r r
177.12 1671/2 r r r r r 2.20
177.12 175 r 3.00 4.25 r r r
177.12 1771/2 r r 3.25 r r r
177.12 1871/2 r r r r r 14.20
31,250 British Pounds -cents per unit.
BPound 1671/2 r 10.00 r r 0.18 r
1 T7.12 1721/2 r 4.90 r r 0.92 r
1 T7.12 175 2.20 2.95 r 0.27 1.70 r
177.12 1771/2 0.63 2.10 r 1.05 r r
177.12 180 0.07 r r r r r
177.12 1821/2 r 0.52 r r r r
177.12 190 5 r r $13.77 r
50,000 Canadian Dollars-European Style.
CDollar 871/2 r r r r 0.30 r
88.23 88 r r r 0.08 r r
88.23 881/2 r r r 0.25 r r
50,000 Canadian Dollars-cents per unit.
CDollr 841/2 r r r r r 0.12
88.23 85 r r r r r 0.17
88.23 86 r r r r r 0.41
88.23 871/2 r r r r 0.24 r
88.23 88 r 0.35 r 0.07 0.55 r
88.23 881/2 r 0.20 r 0.62 r r
88.23 89 r 0.13 0.34 0.70 r r
88.23 891/2 r r r r 1.48 r
TABLE 16.1 Cont. Foreign currency options.

Option & Strike
Underlying Price
62,500 German Marks-cents per unit.
DMark 56 r r r r 0.02 r
61.03 57 r 4.05 r r r 0.51
61.03 58 r r r r 0.10 0.81
61.03 581/2 2.59 r s r r s
61.03 59 2.03 2.19 r r 0.20 r
61.03 591/2 r r s r 0.29 s
61.03 60 r r r 0.04 0.44 r
61.03 601/2 0.68 r s 0.10 0.67 s
61.03 61 0.33 0.73 r 0.73 0.90 r
61.03 611/2 0.15 0.53 s r r s
61.03 62 0.06 0.46 r r r r
61.03 621/2 r 0.30 s r r s
6,250,000 Japanese Yen-100ths of a cent per unit.
Jyen 73 r r r r 0.04 r
76.99 74 2.96 r r r r r
76.99 75 r 1.98 r r r r
76.99 76 r r r 0.04 0.43 r
76.99 761/2 r r s 0.13 0.62 s
76.99 77 0.22 r r 0.34 0.86 r
76.99 771/2 0.06 0.46 5 r r s
76.919 78 0.03 0.31 r r r r
76.99 79 r r 0.82 r r r
76.99 85 r r 0.09 r r r
76.99 86 r r 0.07 r r r
6,250,000 Japanese Yen-European Style.
JYen 74 r r r r r 0.63
76.99 761/2 r r s 0.09 r s
76.99 78 r r 1.10 r r r
62,500 Swiss Francs-European Style.
SFranc 66 r r r r r 0.94
68.85 69 r r 1.60 r r r
62,500 Swiss Francs-cents per unit.
SFranc 65 r r r r 0.07 r
68.85 68 r r r r 0.58 r
68.85 681/2 r r s 0.17 0.79 s
68.a5 69 0.24 r r 0.25 1.06 r
68.85 781/2 s r r s 0.33 r
Total Call Vol 14,647 Call; Open Int 476,994
Total Put Vol 17,077 Put; Open Int 516,307
TABLE 16.2 Foreign currency futures and futures options

CURRENCY TRADING
FUTURES
Lifetime
Open High Low Settle Change High Low Open Interest
JAPAN YEN (IMM)-12.5 million yen; $ per Yen (0.00)
Dec .7691 .7699 .7671 .7679 -0.0012 .7770 .6997 69,869
Mr92 .7666 .7694 .7659 .7665 -0.0011 .7737 .7000 3,572
June … … …. .7659 -0.0010 .7730 .7015 917
Sept … … … .7659 -0.0010 .7710 .7265 599
Dec … … …. .7662 -0.0009 .7700 .7512 1,290
DEUTSCHEMARK (IMM)-125,000 marks; $ per mark
Dec .6088 .6108 .6060 .6088 +.0007 .6770 .5365 72,328
Mr92 .6012 .6045 .5998 .6024 +.0007 .6065 .5353 6,380
June .5965 .5970 .5960 .5963 +.0007 .5985 .5322 715
Est vol 56,177; vol Tues 36,905; open lnt 79,626, -1,188.
CANADIAN DOLLAR (IMM)-100,000 dlrs.; $ per Can $
Dec .8771 .8817 .8763 .8812 +.0019 .8906 .8175 20,341
Mr92 .8720 .8769 .8713 .8767 +.0020 .8857 .8253 4,840
June .8675 .8725 .8675 .8725 +.0018 .8820 .8330 734
Sept .8630 .8685 .8630 .8685 +.0016 .8774 .8348 105
BRITISH POUND (IMM)-62,500 pds.; $ per Pound
Dec 1.7640 1.7696 1.7560 1.7650 +.0024 1.7900 1.5670 27,784
Mr92 1.7430 1.7490 1.7370 1.7436 +.0024 1.7570 1.5560 2,964
Est vol 13,M; vol Tues 7,681; open Int 30,780, - 899.
SWISS FRANC (IMM)-125,000 francs; $ per franc
Dec .6877 .6905 .6849 .6869 +.0002 .8090 .6235 29,074
Mr92 .6829 .6952 .6797 .6819 +.0004 .6995 .6225 2,196
June .6766 .6795 .6750 .6771 +.0004 .6840 .6546 296
Est vol 73,401; vol Tues 16,SU; open Int 31,566, -1,353.
AUSTRALIAN DOLLAR (IMM) - 100,000 dlrs.; $ per A.$
Dec .7825 .7839 .7822 .7832 -.0006 .7960 .7380 11,204
Est vol 113; vol Tues 164; open lnt 1,221, -246.
TABLE 16.2 Cont. Foreign currency futures and futures options

CURRENCY TRADING
FUTURES
U.S. DOLLAR INDEX (FINEX)-SM times USDX
Dec 89.04 89.40 88.78 88.99 -.11 98.96 88.47 5,019
Mr 92 90.07 90.30 89.90 90.08 -.10 98.90 89.60 1,045
The Index: High 88.86; Low 88.34; Close 88.54 -.08
OTHER FUTURES
Settlement prices of selected contracts. Volume and open Interest of all contract months.
British Pound (MCE) IZSM pounds; $ per pound
Japanese Yen (MCE) 6.25 million yen; $ per yen (.00)
Dec .7679 -.0012; Est. vol. 240; Open Int. 353
Swiss Franc (MCE) 6ZSM francs; $ per franc
Dec .6869 +.0002; Est. vol. 1,020; Open Int. 253
Deutschemark (MCE) 6ZSM marks; $ per mark
Dec .6088 +.0007; Est. vol. 360; Open Int. 837
BP/DM Cross Rate (IMM) US $50,000 times BP/DM
DM/JY Cross Rate (IMM) US $125,000 times DM/JY
Dec .7928 +.0022; Est. vol. 6; Open lnt. 583
FINEX-Financial Instrument Exchange, a division of the New York Cotton Exchange. IMM-International
Monetary Market at the Chicago Mercantile Exchange. MCE MidAmerica Commodity Exchange.

CURRENCY TRADING
FUTURES OPTIONS
JAPANESE YEN (IMM) 12,500,000 yon; cents per 100 yen
Strike Price Calls-Settle Puts-Settle
Dec-c Jan-c Mar-c Dec-p Jan-p Mar-p
7550 1.50 1.68 0.21 0.54
7600 1.13 1.36 1.92 0.34 0.72 1.28
7650 0.81 1.09 0.52 0.94
7700 0.56 0.95 1.42 0.77 1.20 1.76
7750 0.39 0.66
7800 0.26 0.50 1.03 1.46 2.36

CURRENCY TRADING
FUTURES OPTIONS
DEUTSCHEMARK (IMM) 125,000 marks; cents Per mark
6000 1.26 1.20 1.75 0.38 0.96 1.51
6050 0.94 0.97 0.56 1.22
6100 0.66 0.77 1.30 0.80 2.04
6150 0.48 0.60 1.10
6200 0.32 0.46 0.95 1.44 2.68
6250 0.21 1.83
FUTURES OPTIONS
CANADIAN DOLLAR (IMM) 100,000 Can.$, cents per Can.$
8700 1.20 1.12 0.09 0.26 0.51
8750 0.78 0.60 0.84 0.16 0.46 0.70
8800 0.43 0.37 0.61 0.31 0.72 0.96
8850 0.21 0.21 0.43 0.59 1.06 1.26
8900 0.06 0.11 0.29 0.97 1.62
8950 0.02 0.18
Est. vol. 4,010, Tues vol. 913 calls, 1,294 puts
Open Interest Tues 13,MS calls, 17,860 puts
BRITISH POUND (IMM) 62,500 pounds; cents per pound
1725 4.64 5.54 0.66 3.68
1750 2.88 2.78 4.28 1.40 3.40 4.Se
1775 1.60 1.90 3.24 2.60 4.92 6.30
1800 0.82 1.12 2.40 4.28 7.92
1825 0.36 0.68 1.74 6.34
1850 0.16 1.22 8.62 11.66
Est. vol. 1,706, Tues vol. &% calls, 681 puts
SWISS FRANC (IMM) 125,000 francs; cents per franc
6750 1.55 0.35 0.91
6800 1.20 1.32 1.92 0.50 1.13 1.73
6850 0.91 1.08 0.72 1.39
6900 0.68 0.87 1.46 0.98 2.27
6950 0.49 0.70 1.28
7000 0.34 0.56 1.10 1.63
Est. vol. 1,746, Tues vol. 965 calls, 706 puts

CURRENCY TRADING
FUTURES OPTIONS
U.S. DOLLAR INDEX (FINEX) 540 times Index
Dec-c Jan-c Feb-c Dec-p Jan-p Feb-p
87 3.40 0.27
88 1.51 2.6.4 0.52 0.59
89 0.92 1.97 0.93 0.92
90 0.52 1.42 1.53 1.35
91 0.27 0.96 2.28
92 0.13 0.66 3.13
Open Interest Tues 12,05.4 calls, 16,739 puts
Final or settlement prices of selected contracts. Volume and open
Interest are totals in all contract months.
Australian Dollar (IMM) $100,000; $ per $
7850 0.47 0.65
Est. vol. 3. Tues vol. 55. Op. Int. 1,668.
FINEX-Financial Instrument Exchange, a division of the New York
Cotton Exchange. IMM-international Monetary Market at Chicago
Mercantile Exchange . LIFFE-London International Financing Futures
Exchange.
Uses of Currency Options

Currency options are helpful in managing foreign exchange risk that arises in international trade or in the
management of international investment portfolios. Currency options allow more flexible hedging of
exchange risk than is possible with currency futures alone.
As an example, consider a U.S. importer of German machinery that costs DM500,000 and is to be
delivered in March 1992. Payment in German marks is to be made upon delivery. At the March futures price
of $0.6024 shown in Table 16.2, the dollar cost of the machinery will be $301,200. Of course, if the price of
the German mark increases, the dollar cost of the machinery will increase. One way to hedge this exchange
risk is to buy four futures contracts (each contract applies to DMI25,000) or to enter into an appropriate
forward contract with a bank. If the price of the D-mark does increase, the increased equivalent dollar cost of
the machinery is offset by the profit on the futures position. On the other hand, if the price of the D-mark
decreases, the lower dollar cost is offset by the loss on the futures position.
An alternative way to hedge against D-mark price increases is to buy March call options on the D-
mark spot currency or on the D-mark futures. If the D-mark appreciates, the increase in the dollar cost of the
machinery is offset by the profit on the call options. On the other hand, if the D-mark depreciates, the lower
cost of the machinery is a pure gain. This type of hedge provides insurance against increases in the exchange
rate without an offsetting penalty should the exchange rate drop. Naturally, the call premium reflects the
value of this insurance. To reduce the premium cost, one might buy out-of-the-money D-mark futures
options. For example, Table 16.2 indicates it would have been possible to buy call options on the D-mark
futures with an exercise price of 0.61 at a cost of 1.30 cents per mark.
Since each contract at the Chicago Mercantile Exchange is DM 1 25,000, four contracts are
necessary, and the total premium is $6,500. This option position would cap the total dollar cost at
(0.61)(500,000) = $305,000, while retaining the possibility of gain if the D-mark should depreciate.
Increases in the D-mark above 0.61 would be offset by profits on the futures option position.
Options can provide a useful hedge if there is uncertainty about the underlying import or export
contract. For example, consider a U.S. company that bids a price of 350,000 pounds to install a computer
system in Great Britain and suppose the British company has a month in which to accept or reject the bid.
The U.S. company is concerned about a depreciation of the British pound, but if it sells futures to hedge the
foreign exchange risk, and the bid is not accepted, the company is left with an open currency futures position
that may have to be liquidated at a loss. An alternative hedge is for the U.S. company to buy put contracts on
350,000 British pounds. By purchasing puts, the company guarantees the price at which pounds can be sold
if the bid is accepted. If the bid is rejected, the put option is not exercised and is sold. In effect, the U.S.
company is using an option to hedge a contract that has an option feature. The U.S. company has given the
British company the put option to sell 350,000 pounds to the U.S. company in return for the computer
system. The U.S. company hedges that risk by buying a put.147
In addition to hedging import and export contracts, currency options are useful in international
investment and portfolio management. Investment in a fore-_ii country exposes a portfolio to exchange rate
risk as well as the usual risk of capital losses. Currency options can be used to modify that risk. For example,
an investor in Australian bonds could hedge principal and/or interest payments by purchasing puts on the
Australian dollar. Over-the-counter options written by banks are frequently used to tailor such hedges to the
needs of the investor, particularly when longer maturities are necessary and/or when a sequence of options is
required (as when a stream of coupon payments is hedged). Some fixed-income securities are offered with
imbedded currency options. For example, a bond might offer to pay interest and/or principal in either of two
currencies at a fixed exchange rate, with the investor having the option to choose the currency. Complex or
exotic options, which are discussed below, are often created to deal with currency risk. For example, a bond
could offer to pay principal and interest in dollars or in two other currencies at fixed exchange rates
established in the bond indenture. The holder of the bond thus has a dollar bond plus the option of choosing
the most valuable of the three currencies in which payment may be received.
16.2 PHYSICAL COMMODITY FUTURES OPTIONS

Markets for option contracts on physical commodity futures became active in 1982 with the introduction of
sugar futures options by the Coffee, Sugar, and Cocoa Exchange and of gold futures options by the
Commodity Exchange. These option contracts are American-style and settle through delivery of a position in
the underlying futures. The grain contracts trade predominantly on the Chicago Board of Trade; the livestock
contracts on the Chicago ' Mercantile Exchange; oil and oil-related products on the New York Mercantile
Exchange; and metals at the Commodity Exchange. Table 16.3 contains a listing of the currently active,
exchange-traded commodity options. As with all futures option contracts, the valuation principles follow
from Chapters 10 through 12 once the cost-of-carry rate is set to zero ( b = 0).
16.3 EXOTIC OPTIONS

Exotic options are complex options that typically incorporate two or more option features. A compound
option, for example, is considered an exotic option. It provides its holder with the right to buy another
option. Options on the maximum of two (or more) risky commodities are also considered to be exotic
options. With this option, the investor has the right to buy “the better of two commodities.” Because exotic
147
The hedging uses of currency options in the kind of situation described here are also discussed in Giddy (1983) and in Feiger
and Jacquillat (1979).
options are complex and are often tailored to the needs of the customer, they are available primarily in the
OTC market.
TABLE 16.3 Commodity futures options

COMMODITY FUTURES OPTIONS (Wed. 11/13/1991)
Wednesday, November 13, 1991.
-AGRICULTURAL-
CORN (CBT) 5,000 bu.; cents per bu.
Strike
Calls- Settle Puts – Settle
Price
Dec-c Mar-c May-c Dec-p Mar-p May-p
230 16 251/2 1/8 1/2 ¾
240 61/4 17 24 5/8 21/4 21/2
250 7/8 111/4 171/8 5 51/2 5
260 C4 7 12 141/8 115/8 93/4
270 C2 41/8 8 24 181/2 151/4
280 C1 23/8 51/4 34 27 23
SOYBEANS (CBT) 5,000 bu.; cents per bu.
Jan-c Mar-c May-c Jan-p Mar-p May-p
500 571/2 651/2 1/4 1 2
525 331/4 431/2 11/2 33/4 51/2
550 157/8 26 35 8 11 3/4 131/4
575 6 151/4 231/2 23/2 25 251/2
600 23/8 9 161/2 441/2 43 411/2
625 1 53/4 103/4 681/4 65 601/2
SOYBEAN MEAL (CBT) 100 tons; $ per ton
170 10.90 9.30 0.05 0.50 2.35
115 5.90 5.55 6.35 0.30 1.75 4.60
150 2.00 3.50 4.55 1.10 4.20 7.60
185 0.40 1.70 3.25 4.40 7.75 11.20
190 C3 1.00 2.20 9.10 12.00
105 0.05 0.55 1.60 14.10
TABLE 16.3 Cont. Commodity futures options

SOYBEAN OIL (CBT) 60,000 lbs.; cents per lb.
Strike
Price
17 1.880 0.005
18 0.900 0.020 0.060 0.220
19 0.100 0.400 0.220 0.470 0.490
20 0.010 0.140 0.450 1.130 1.200 1.180
21 0.005 0.050 0.210 2.130 2.150 1.950
22 0.005 0.020 0.130 6.130 3.100 2.850
WHEAT (CBT) 5,000 bu.; cents per bu.
330 221/2 263/4 211/2 1/8 4 131/2
340 125/8 193/4 16 1/2 7 18
350 41/2 141/4 12 2 103/4 24
360 5/8 10 10 81/4 161/4
370 1/8 7 61/2 171/2 223/4
380 1/8 43/4 51/4 301/2
Open interest Tues 28,890 calls, 32,465 puts
WHEAT (KC) 5,000 bu.; cents per bu.
340 16 201/2 161/2 5/8 6 163/4
350 6 15 131/2 11/4 10
360 1 111/2 83/4 51/4 141/4
370 3/8 61/2 61/2 143/4 211/2
380 4 5 241/2
390 21/2 31/4
Est. vol. 372, Tues vol. 168 calls, 1,030 puts
COTTON (CTN) 50,000 lbs.; cents per lb.
Mar-c May-c Jul-c Mar-p May-p Jul-p
57 1.25 1.50
58 2.90 1.60 1.90
59 3.25 2.02 2.25
60 1.90 2.75 2.60 2.75
61 1.50 2.30 3.20 3.30 3.50
62 1.15 1.95 2.55 3.95 3.85
Est. vol. 1,400; Tues vol. 1,065 calls; 534 puts

ORANGE JUICE (CTN).15,000 lbs.; cents per lb.
Strike
Price
165 13.25 18.05 1.60 7.50 11.10
170 9.65 16.40 3.00 9.75 13.25
175 6.65 13.85 12.20 15.50
180 4.35 11.70
185 1.95 9.65
190 1.55
Est. vol. 375; Tues vol. 98 calls; 116 puts
COFFEE (CSCE) 37,500 lbs.; cents per lb.
75 9.03 11.75 14.10 0.78 0.95 0.95
50 5.38 8.20 10.40 2.25 2.40 2.25
85 3.32 5.15 6.80 5.00 4.60 4.20
90 2.10 3.45 4.95 8.78 7.65 6.80
95 1.33 2.35 3.50 13.06 11.55 10.35
100 0.90 1.60 2.53 17.65 15.80 14.38
Est. vol. 745; Tues vol. 930 calls; 68 puts
Open Interest Tues; 15,495 calls; puts
SUGAR-WORLD (CSCE) 112,000 lbs.; cents per lb.
7.50 0.83 0.98 0.06 0.21
8.00 0.42 0.68 0.75 0.16 0.40 0.51
8.50 0.19 0.45 0.43 0.67
9.00 0.07 0.30 0.41 0.79 1.00 1.11
9.50 0.03 0.20 1.26 1.43
10.00 0.02 0.12 0.20 1.75 1.95 1.94
Est. vol. 7,719; Tues vol. 1,431 calls; 498 puts
COCOA (CSCE) 10 metric tons; $ per ton
1100 174 214 257 10 14 23
1200 96 139 184 35 39 50
1300 50 94 121 90 84 91
1400 23 48 81 159 148 147
1500 12 29 54 248 229 220
1600 5 19 44 341 319 310
Est. vol. 408; Tues vol. 519. calls; 155 puts

-OIL-
CRUDE OIL (NYM) 1,000 bbls.; $ per bbl.
Strike
Price
Jan-c Feb-c Mar-c Jan-p Feb-p Mar-p
20 2.28 0.06 0.18 0.31
21 1.38 1.33 1.25 0.16 0.37 0.55
72 0.63 0.68 0.69 0.41 0.72 0.98
23 0.23 0.30 0.35 1.01 1.33 1.62
24 0.07 0.13 0.16 1.85 2.16
25 0.03 0.06 0.07
Est. vol. 18,873; Tues vol. 10,016 calls; 23,569 puts
HEATING OIL No.2 (NYM) 42,000 gal.; $ per gal.
62 0.0478 0.0454 0.0301 0.0030 0.0116 0.0255
64 0.0323 0.0329 0.0215 0.0075 0.0190 0.0369
66 0.0198 0.0219 0.0160 0.0150 0.0280 0.0514
68 0.0115 0.0160 0.0115 0.0267 0.0421
70 0.0070 0.0110 0.0085 0.0422 0.0571 0.0839
72 0.0042 0.0077 0.0065 0.0594
GASOLINE-Unleaded (NYM) 42,000 gal.; 4 per gal.
58 0.0344 0.0386 0.0030 0.0060 0.0065
60 0.0194 0.0246 0.0366 0.0080 0.0120 0.0120
62 0.0095 0.0145 0.0246 0.0181 0.0219 0.0200
64 0.0045 0.0080 0.0160 0.0331 0.0354 0.0314
66 0.0025 0.0043 0.0100 0.0511 0.0517 0.0454
68 0.0015 0.0022 0.0062
-LIVESTOCK-
CATTLE-FEEDER (CME) 44,000 lbs.; cents per lb.
Nov-c Jan-c Mar-c Nov-p Jan-p Mar-p
80 4.00 3.70 2.92 0.00 0.70 1.60
82 2.00 2.20 1.87 0.02 1.20 2.45
84 0.30 1.17 0.85 0.30 2.17 3.52
86 0.02 0.52 0.40 2.02 3.50 5.07
88 0.00 0.17 0.25 4.00 5.17
90 0.00 0.05 0.10 6.00 7.00

-LIVESTOCK-
CATTLE-LIVE (CME) 40,000 lbs.; cents per lb.
Strike
Price
Dec-c Feb-c Apr-c Dec-p Feb-p Apr-p
70 5.00 0.05 0.35 0.60
72 3.10 3.62 0.15 0.62 0.95
74 1.45 2.12 2.20 0.50 1.10 1.52
76 0.35 1.02 1.10 1.40 1.95 2.42
78 0.05 0.32 0.47 3.10
80 0.00 0.10 0.15 5.05
Open Interest Tues 17,%l calls, 31,OD4 puts
HOGS-LIVE (CME) 40,000 lbs.; cents per lb.
38 3.70 4.97 0.05 0.15 0.60
40 1.75 3.32 1.85 0.20 0.50 1.30
42 0.55 1.87 1.00 0.90 1.05 2.45
44 0.12 0.90 0.52 2.47 2.07 3.85
46 0.05 0.42 0.22 4.40 3.60 5.55
48 0.00 0.15 0.10 6.35 5.32
-METALS-
COPPER (CMX) 25,000 lbs.; cents per lb.
98 6.10 6.40 6.30 1.50 2.90 3.65
100 4.60 5.10 5.20 2.05 3.70 4.60
102 3.60 4.05 4.25 2.95 4.65 5.65
104 2.85 3.25 3.55 4.20 5.75 6.95
105 2.30 2.90 3.20 4.60 6.40 7.50
106 2.00 2.60 2.95 5.30 7.10 8.25
GOLD (CMX) 100 troy ounces; dollars per troy ounce
Jan-c Feb-c Apr-c Jan-p Feb-p Apr-p
340 20.20 20.50 23.70 0.20 0.60 1.50
350 10.60 11.70 15.50 0.60 1.70 3.30
360 3.30 5.00 9.10 3.20 4.90 6.70
370 0.70 1.90 5.00 10-60 11.80 12.50
380 0.30 0.80 2.70 20.10 20.50 19.90
390 0.20 0.50 1.60 30.00 30.00 28.60

-METALS-
SILVER (CMX) 3,000 troy ounces; cents per troy ounce
Strike
Price
350 60.7 61.0 61.7 0.1 0.4 1.0
375 36.2 37.0 39.1 0.5 1.8 3.4
400 14.4 17.5 21.0 3.7 7.0 10.3
425 3.5 7.0 21.0 3.7 7.0 10.3
450 1.1 3.0 4.8 40.3 42.0 44.1
475 0.7 1.50 2.7 64.8 65.0 67.0
Est. vol. 12,000, Tues vol. 2,739 calls, 603 puts
OTHER FUTURES OPTIONS. Final or settlement prices of selected
contracts. Volume and open Interest are totals in all contract months.
Lumber (CME) 160,000 bd.ft., $ per 1,000 bd.ft.
Strike Calls- Settle Puts – Settle
210 6.70 5.20
Est. vol. 10. Tues vol. 0. Op. Int. 149.
Oats (C BT) 5,000 bu.; cents per bu.
130 1/2 13/4 3
Est. vol. 5. Tues vol. 2. Op. lnt. 439.
Platinum (NYM) 50 troy oz.; $ per troy oz.
360 10.00 5.50
Est. vol. 76. Tues vol. 76. Op. Int. n.a..
Pork Bellies (CME) 40,000 lbs.; cents per lb.
40 2.60 3.05 0.75 2.92 3.42
Silver (CBT) 1,000 troy oz.; cents per troy oz.
400 4.0 23.0 1.5
Est. vol. 5. Tues vol. 3. Op. I nt. 208.
Soybeans (MCE) 1,000 bu.; cents per bu.
550 157/8 26 35 8 113/4 131/4
Wheat (MPLS) 5,000 bu.; cents per bu.
340 11/2 121/2 15 5 13
Est. vol. 5. Tues vol. 16. Op. Int. 712.
In this section, we apply the lognormal price distribution mechanics used in Chapters 11 and 12 to price
compound options (or, more commonly, “options on options”), options on the maximum or the minimum,
lookback options, and barrier options. Illustrations are provided. Following these discussions, we describe
some other types of exotic options that currently trade. Our list of exotic options is necessarily incomplete,
since new option contracts are designed and traded almost every day. The descriptions included will give a
flavor for the ingenuity of some current option contract designs.
Options on Options
Compound options or options on options fall in the category of exotic options. Call (put) options providing
the right to buy (sell) call options (i. e., calls on calls or puts on calls) and call (put) options providing the
right to buy (sell) put options (i.e., calls on puts or puts on puts) are the most common forms. To value these
options on options,148 we adopt the assumptions and notation used in Chapter 11. The critical assumptions
are that the terminal commodity price distribution is lognormal and that the principles of risk-neutral
valuation apply. The call and put options that we are valuing are assumed to be European-style with exercise
prices, ct* and pt* , respectively, and with time to expiration t . The notation representing the right to buy a
call at time t (i.e., a call on a call) is c(c, t ; ct* ) , the right to sell a call at time t (i.e., a put on a call) is
p(c, t ; ct* ) , the right to buy a put at time t (i.e., a call on a put) is c(c, t ; pt* ) , and the right to sell a put at time
t (i.e., a put on a put) is p(c, t ; pt* ) . The option received or delivered at expiration from the exercise of an
option on an option has exercise price X and time to expiration T . The notation used to describe the
underlying options is c( St , t ; X ) and p( St , t ; X ) , respectively. Conditional upon knowing St , these
European-style options can be valued using equations (11.25) and (11.28) from Chapter 11.
To demonstrate how to value a compound option, we use a call on a call. The first step in the risk-
neutral valuation approach is to formulate the option's payoff contingencies. For the call on a call, the payoff
contingencies at time t are
c( St , T ; X )  ct* if ct  ct*

ct   (16.1)
0 if ct  ct*
That is, if the value of the call to be received at time t , c( St , T ; X ) , is greater than the exercise price, ct* , the
call option holder will exercise his right to buy the call. If the value is less, he will let it expire worthless.
The second step involves restating the contingent payoffs in (16.1) in terms of the underlying commodity
price at time t , St , in order to make the problem more tractable mathematically. The commodity price above
which the call option holder will choose to exercise his call at time t is given by
c( S t* , T ; X )  ct* (16.2)
where c( St , T ; X ) represents the European-style option valuation equation (11.25) evaluated at St  S t* . Note
that the value of St* may be solved iteratively in the same manner that we have computed critical commodity
prices in earlier chapters.149 With St* known, the payoff contingencies expressed in (16.1) may be written as
148
The models presented in this section are based on the work of Geske (1979).
149
See, for example, the valuation of American-style call options on dividend-paying stocks in Chapter 13 or the valuation of
American-style options using the quadratic approximation method in Chapter 14.
c( St , T ; X )  ct* if St  St*

ct   (16.3)
0 if St  St*
Call on Call. Under risk-neutral valuation, the value of a call on a call may be written as the present value of
the expected terminal value of the option, where the discount rate is the riskless rate of interest, r :
c(ct , t ; ct* )  e  rT E[c( St , T ; X )  ct* | St  St* ]Pr( St  St* ) . (16.4)
Expressing c( St , T ; X ) in terms of its terminal commodity price payoffs and isolating the cost of exercising
the option at time t , equation (16.4) becomes
c(ct , t; ct* )  e r (t T ) E[ ST | ST  X and St  St* ]Pr( ST  X and St  St* )

(16.5)
e r (t T ) X Pr( ST  X and St  St* )  e r (t T ) ct* Pr( St  St* )
Under the assumption that future commodity prices are lognormally distributed, the value of a European-
style call on a call is
   
c(ct , t ; ct* )  Se( br )( t T ) N 2 a1 , b1 ; t /(t  T )  Xe r ( t T ) N 2 a2 , b2 ; t /(t  T )  e rt ct* N1  b2  (16.6)
where
ln( S / X )  (b  0.5 2 )(t  T )
a1  , a2  a1   (t  T )
 (t  T )
ln( S / St* )  (b  0.5 2 )t
b1  , b2  b1   t
 t
and N1 () and N 2 () , are the cumulative univariate and bivariate unit normal density functions described in
Chapters 11 and 13, respectively.
 
In equation (16.6), the term N 2 a1 , b1 ; t /(t  T ) is the delta value of the call option on a call option. It
describes the call option price movement for a small change in the commodity price. Recall that in Chapter
12 we showed how delta values are used for hedging purposes. The term N1 (b2 ) is the probability that the
commodity price will exceed the critical commodity price at time t . The term N 2 a2 , b2 ; t /(t  T ) is the  
probability that the commodity price will exceed St* at time t and the exercise price X at time t  T .
Put on Call. The simplest way to derive the valuation equation for a put on a call is to deduce the valuation
formula from known results. In Chapter 12, we showed that a long-call/short-commodity position is
tantamount to a long-put position. Here, the underlying commodity position is a call option, so a long-call-on
a-call/short-call position should be tantamount to a put on a call. Since we have the valuation equation for a
call on a call (16.6) and for a European-call (11.25), the valuation equation for a put on a call is
   
p (ct , t ; ct* )  Se( br )(t T ) N 2 a1 , b1 ; t /(t  T )  Xe  r ( t T ) N 2 a2 , b2 ; t /(t  T )  e  rt ct* N1  b2 
,
 Se( br )( t T ) N1  a1   Xe  r ( t T ) N1  a2   e rt ct*

p (ct , t ; ct* )  Xe r ( t T ) N 2 a2 , b2 ;  t /(t  T )  (16.7)
 
 Se( br )( t T ) N 2 a1 , b1 ;  t /(t  T )  e  rt ct* N1  b2 
where all notation is defined in (16.6).
Call on Put. The risk-neutral valuation framework shown above can also be applied to value a call on a put.
The value of a European-style call on a put is

c(ct , t ; pt* )  Xe  r (t T ) N 2 a2 , b2 ; t /(t  T )  (16.8)
 
 Se(br )( t T ) N 2 a1 , b1 ; t /(t  T )  e  rt pt* N1  b2 
The critical commodity price below which the call option holder will choose to exercise the call to buy the
put at time t is determined by solving
p ( St* , T ; X )  pt* (16.9)
p ( St , T ; X ) represents the European-style option valuation equation (11.28) evaluated at S  St* . All other
 
notation is as previously defined. The term N 2 a1 , b1 ; t /(t  T ) is the delta value of a call option on a
put option delta value. The term N 2 (b2 ) is the probability that the commodity price will be below the critical
 
commodity price at time t . The term N 2 a2 , b2 ; t /(t  T ) is the probability that the commodity price
will be below St* at time t and the exercise price X at time t  T .
Put on Put. A put on a put has the same payoff contingencies as a long-call on-a-put/short-put position.
Using equations (11.28) and (16.8), it can be shown that the value of a put on a put is

p (ct , t ; pt* )  Se( br )( t T ) N 2 a1 , b1 ;  t /(t  T )  , (16.10)
 
 Xe r ( t T ) N 2 a2 , b2 ;  t /(t  T )  e rt pt* N1  b2 
where all notation is defined above.
EXAMPLE 16.1
Consider a call option that provides its holder with the right to buy a put option on the S&P 500 index
portfolio. The put that would be delivered against the call if the call is exercised has an exercise price of
$400 and a time to expiration of six months. The call has an exercise price of $10 and a time to expiration of
three months. The S&P 500 index is currently at 390, pays dividends at a constant rate of 4 percent annually,
and has a volatility rate of 28 percent. The riskless rate of interest is 7 percent.
The first step in valuating the compound option is to compute the critical commodity price below
which the call will be exercised to take delivery of the put. This is done by solving
p ( St* , 0.5; 400)  10.00 .
The critical commodity price, St* , is 497.814. The next step is to apply the valuation formula (16.8). Here,
we get

c(ct , t ; pt* )  400e0.07(0.250.5) N 2 a2 , b2 ; 0.25 / 0.75 
 
390e0.03(0.250.5) N 2 a1 , b1 ; 0.25 / 0.75  e0.07(0.25)10 N1  b2   27.722
where
ln(390 / 400)  (0.03  0.5(0.28)2 )(0.75)
a1   0.1096, a2  0.1096  0.28 0.75  0.1329
0.28 0.75
ln(390 / 497.814)  (0.03  0.5(0.28) 2 )0.25
b1   1.620, b2  1.620  0.28 0.25  1.760
0.28 0.25
The probability that the commodity price will be below the critical commodity price at time t , N1 (b2 ) , is
0.961. The probability that the commodity price will be below St* at time t and below the exercise price, X ,
 
at time t  T , N 2 a2 , b2 ; t /(t  T ) , is 0.453. The value of a call on a call with the same terms as the put is
27.012. (The critical index price is 342.424.)
Options on the Maximum and the Minimum

Options on the maximum and the minimum of two or more risky commodities are popular exotic options.150
For example, someone may buy the right to buy the S&P 500 index or gold for $400, depending on which
commodity is worth more at the option's expiration. As in the case of compound options, options on the
maximum and the minimum can be valued straightforwardly, assuming that both commodity prices have
lognormal price distributions at the option's expiration. Under the risk-neutral valuation approach, the value
of a call option on the maximum, for example, may be written as
cmax ( S1 , S2 ; X )  e rT E ( S1,T | S1,T  X and S1,T  S2,T ) Pr( S1,T  X and S1,T  S2,T )
e  rT E ( S2,T | S2,T  X and S2,T  S1,T ) Pr( S2,T  X and S2,T  S1,T )
(16.11)
 Xe  rT Pr( S1,T  X or S2,T  X )
Under the assumption that future commodity prices are lognormally distributed, the value of a European-
style call on the maximum is
cmax ( S1 , S 2 ; X )  S1e( b1 r )T N 2 ( d11 , d1' ; 1' )  S 2 e( b2 r )T N 2 ( d12 , d 2' ;  2' )  Xe  rT [1  N 2 (  d 21 ,  d 22 ; 12 )] (16.12)
where
ln( S1 / X 1 )  (b1  0.5 12 )T
d11  , d 21  d11   1 T
1 T
ln( S 2 / X 2 )  (b2  0.5 22 )T
d12  , d 22  d12   2 T
2 T
ln( S1 / S2 )  (b1  b2  0.5 2 )T '
d1'  , d 2  (d1'   T )
 T
150
Other names for the option on the maximum are “the better of two assets” or “outperformance options.” The models presented
here are on the maximum or the minimum of two risky commodities and the valuation models are based on Stulz (1982). To
generalize these models to three or more risky assets, see Johnson (1987).
 1  12 2  2  12 1
 2   12   22  2 12 1 2 , 1'  and  2' 
 
In equation (16.12), the term [1  N 2 (d 21 ,  d 22 ; 12 )] is the probability that one of the two commodity prices
will exceed the exercise price at time T or, alternatively, one minus the probability that neither commodity
will have a price greater than the exercise price at the option's expiration.
EXAMPLE 16.2
Consider a call option that provides its holder the right to buy $100,000 worth of the S&P 500 index
portfolio at an exercise price of $400 or $100,000 worth of a particular T-bond at an exercise price of $ 1 00,
whichever is worth more at the end of three months. The S&P 500 index is currently priced at $360, pays
dividends at a rate of 4 percent annually, and has a return volatility of 28 percent. The T-bond is currently
priced at $98, pays a coupon yield of 10 percent, and has a return volatility of 15 percent. The correlation
between the rates of return of the S&P 500 and the T-bond is 0.5. The riskless rate of interest is 7 percent.
Before applying the option on the maximum formula, it is important to recognize that there are two
exercise prices in this problem: $400 for the S&P index portfolio and $ 1 00 for the T-bond. What this
implies is that we can buy $ 1 00,000/ $400 = 250 “units” of the index portfolio or $100,000/$100 = 1,000 T-
bond “units” at the end of three months, depending on which is worth more. At this juncture, we must decide
whether to work with the valuation equation (16.12) in units of the S&P 500 index portfolio, in which case
we multiply the current T-bond price and its exercise price by 4, and then multiply the computed option price
by 250, or to work with the valuation equation (16.12) in units of the T-bond, in which case we divide the
current S&P 500 price and the option's S&P 500 exercise price by 4, and then multiply the computed option
price by 1,000.151 In this exercise, we choose to work in units of the S&P 500 index portfolio, so we adjust
the T-bond prices: the current T-bond price is assumed to be 392, and the T-bond exercise price is 400. With
the units of the two underlying assets comparable, we now apply equation (16.12):
cmax  360e0.04(0.25) N 2 (d11 , d1' ; 1' )  392e0.10(0.25) N 2 (d12 , d 2' ;  2' )
400e0.07(0.25) [1  N 2 (d 21 , d 22
'
; 12 )]  11.962
where
ln(360 / 400)  (0.07  0.04  0.5(0.28) 2 )(0.25)
d11   0.6290,
(0.28) 0.25
d 21  0.6290  (0.28) 0.25  0.7690
ln(392 / 400)  (0.07  0.10  0.5(0.15)2 )(0.25)
d12   0.3319,
(0.15) 0.25
d 22  0.3319  (0.15) 0.25  0.4069
ln(360 / 392)  (0.06  0.5(0.2427) 2 )(0.25)
d1'   0.5175,
(0.2427) 0.25
d 2'  0.5175  (0.2427) 0.25  0.6388
  0.282  0.152  2(0.5)(0.28)(0.15)  0.2427,
0.28  0.5(0.15) 0.15  0.5(0.28)
1'   0.8447, and  2'   0.0412
0.2427 0.2427
151
These types of adjustments can be made freely because the option price is linearly homogeneous in both the commodity price
and the exercise price. See Merton (1973).
The computed option price is 11.962, which implies the value of the option contract is $11.962 x 250, or
$2,990.50. The probability that either or both components of the option are in-the-money at expiration is
1  N 2 (0.7690,0.4069;0.5) , or 42.72 percent.
Under the same assumptions, the value of a European-style call on the minimum is
cmin ( S1 , S 2 ; X )  S1e(b1 r )T N 2 (d11 , d1' ;  1' )  S2 e(b2 r )T N 2 (d12 , d 2' ;   2' )

, (16.13)
 Xe  rT [1  N 2 (d 21 , d 22
'
; 12 )]
where all notation is as previously defined.
Lookback Options
Aside from compound options and options on the maximum and the minimum, many other exotic options
trade in OTC markets. Some of the options are backward looking. A lookback call option provides its holder
with settlement proceeds equal to the difference between the highest commodity price during the life of the
option less the exercise price, and a lookback put option provides its holder with settlement proceeds equal to
the difference between the exercise price and the lowest commodity price during the life of the option. It
should come as no surprise, therefore, that these options are sometimes referred to as “no-regret options.”
In a sense, lookback options are like American-style options because the option holder is guaranteed
the most advantageous exercise price. Lookback call options can be valued analytically using the risk-neutral
valuation mechanics.152 The reason for this is that it never pays to exercise a lookback option prior to
expiration. Independent of how low the commodity price has been thus far during the option's life, there is
always some positive probability that it will fall further. For this reason, the option holder will always defer
early exercise in the hope of recognizing higher exercise proceeds in the future.
Under the assumptions of risk-neutral valuation and lognormally distributed future commodity prices, the
value of a lookback call may be written as style call option whose exercise price is the current minimum
value of the underlying commodity. This is the least the lookback call can be worth since the commodity
price may fall below X, thereby driving the “exercise price” down further.
  b[T  
2 ln( S / X )
]
cLB  Se(br )T N1 (d1 )  Xe  rT N1 (d 2 )  Se( br )T  e 2
N1 ( d3 )  N1 (  d1 )  (16.14)
 
where X is the current minimum price of the commodity during the life of the option,   0.5 2 / b ,
ln( S / X )  (b  0.5 2 )T (b  0.5 2 )T
d1  , d 2  d1   T , and d3  . Note that the first two terms of the
 T  T
option are the value of a European style call option whose exercise price is the current minimum value of the
underlying commodity. This is the least the lookback call can be worth since the commodity price may fall
below X , thereby driving the “exercise price” down further.
152
The pricing equations provided here are based on the work of Goldman, Sosin, and Gatto (1979).
EXAMPLE 16.3
Consider a lookback call option that provides its holder with the right to buy the S&P 500 index at any time
during the next three months. The S&P 500 index is currently at a level of 390, pays dividends at a constant
rate of 4 percent annually, and has a volatility rate of 28 percent. The riskless rate of interest is 7 percent.
The cost-of-carry rate is 0.07 - 0.04 = 0.03. The value of the lookback call is, therefore,
cLB  390e(0.030.07)(0.25) N1 (d1 )  390e0.07(0.25) N1 (d 2 )

 0.03[(0.25) ( 0.28)2 ] 
2 ln(390 / 390 )
390e(0.030.07)(0.25)  e N1 (d3 )  N1 (d1 )   42.583

 
where
  0.5 2 / b  0.5(0.28) 2 / 0.03  1.3067
ln(390 / 390)  (0.03  0.5(0.28) 2 )(0.25)
d1   0.1236; d 2  0.1236  0.28 0.25  0.0164;
0.28 0.25
(0.03  0.5(0.28) 2 )(0.25)
and d3   0.0164
0.28 0.25
Note that the price of the lookback call is considerably higher than an at-the-money index call option. The
value of a European-style call (i.e., the sum of the first two ten-ns in the valuation equation) is only 22.941.
The value of a European-style lookback put option is
 
2 ln( S / X )
 b[ T  ]
pLB  Xe  rT N1 (  d 2 )  Se( br )T N1 (  d1 )  Se( br )T   N1 ( d1 )  e 2
N1 (  d3 )  (16.15)
 
where all notation is as defined for the lookback call. Note that a standard European-style put option is the
lower bound for the price of the lookback put option. The third term is necessarily positive. Using the same
parameters as in Example 16.3, the value of a lookback put option is $43.468, with the underlying ordinary
European-style put being valued at $20.056.
Other backward-looking options are also traded. For example, average price or Asian options are
based on the average (either arithmetic or geometric) commodity price during the option's life. The average
commodity price may be used as the exercise price of the option, in which case the settlement value of the
call will be the terminal commodity price less the average price, or it may be used as the terminal commodity
price, in which case the settlement value will be the average price less the exercise price. Unfortunately,
most Asian options do not have closed-form valuation equations. Accurate pricing involves the use of
numerical methods.153
Barrier Options
Barrier options are options whose existence depends on the underlying commodity price. A down-and-out
call, for example, is a call that expires if the commodity price falls below a prespecified “out” barrier, H .154
At that time, the option buyer may receive a cash rebate, R . A down-and-in call is a call that comes into
existence if the commodity price falls below the “in” barrier at any time during the option's life. Note that if
153
There are a number of useful background readings for those interested in pricing Asian options. Among them are Boyle (1977)
and Boyle and Emanuel (1985).
154
The valuation equation for the down-and-out call option was first provided in Cox and Rubinstein (1985, Ch. 7). The valuation
equation presented here is a modified version of the formula presented in Rubinstein (1990).
we buy a down-and-out call and a down-and-in call with the same barrier price, H , exercise price, X , and
time to expiration, T , the portfolio has the same payoff contingencies as a standard call option. For this
reason, we automatically know how to value a down-and-in call if we can value a down-and-out call.
Under the assumptions of risk-neutral valuation and lognormally distributed commodity prices, the
valuation equation for a down-and-out call option is
cDO  Se(br )T N1 (a1 )  Xe rT N1 (b2 )  Se(br )T ( H / S )2( 1) N1 (b1 )

, (16.16)
 Xe rT ( H / S ) 2 N1 (b2 )  R( H / S )  N1 (c1 )  R( H / S )  N1 (c2 )
H is the barrier commodity price below which the call option life ends; R is the rebate, if any, received by
the option buyer should the option terminate,
b  0.5 2 ln( S / X )
 ;    2  (2r /  2 ); a1   (1   ) T , a2  a1   T
 2
 T
ln( H 2 / SX ) ln( H / S )
b1   (1   ) T ; b2  b1   T ; c1    T ; c2  c1   T
 T  T
The valuation equation for a down-and-in call is simply equation (11.25) less (16.16).
EXAMPLE 16.4
Consider a down-and-in call option that provides its holder with the right to buy the S&P 500 index at 380
any time during the next three months, should the index level fall below 375. The S&P 500 index is currently
at a level of 390, pays dividends at a constant rate of 4 percent annually, and has a volatility rate of 28
percent. The riskless rate of interest is 7 percent.
he cost-of-carry rate is 0.07 - 0.04 = 0.03. The value of the down-and-out call is
cDO  390e(0.030.07)(0.25) N1 (a1 )  380e0.07(0.25) N1 (b2 )  390e(0.030.07)(0.25) (375 / 390)2( 1) N1 (b1 )
380e0.07(0.25) (375 / 390)2 N1 (b2 )  14.817
where
0.03  0.5(0.28)2
 2
 0.1173;   (0.1173) 2  (2(0.07) /(0.28) 2 )  1.3414;
(0.28)
ln(390 / 380)
a1   (1  0.1173)(0.28) 0.25  0.3091, a2  0.3091  (0.28) 0.25  0.1691
(0.28) 0.25
ln((375) 2 / 390  380)
b1   (1  0.1173)(0.28) 0.25  0.2512; b2  0.2512  (0.28) 0.25  0.3912;
(0.28) 0.25
ln( H / S )
c1   (1.3414)(0.28) 0.25  0.0924; and c2  0.0924  (0.28) 0.25  0.4680
(0.28) 0.25
The value of a standard European-style call option is 28.151, using equation (11.25). The value of the down-
and-in call is, therefore, 28.151 - 14.817 = 13.334.
An up-and-out put and an up-and-in put can be valued in a similar manner. An up-and-out put is a put that
expires if the commodity price rises above the out barriers Its valuation equation is
pDO  Xe rT N1 (b2 )  Se(br )T N1 (a1 )  Se(br )T ( H / S ) 2( 1) N1 (b1 )

(16.17)
 Xe rT ( H / S )2 N1 (b2 )  R ( H / S )  N1 (c1 )  R( H / S )  N1 (c2 )
An up-and-in put comes into existence when the commodity price rises above H . Its valuation equation is
simply (11.28) less (16.17).
Other Exotic Options

Exotic options abound.155 Among those not yet mentioned are those involving deferred features. A deferred-
start option, for example, is an option which is purchased before its life actually begins. A deferred payment
American option is like a standard American-style option except, if the option is exercised early, the option
buyer does not receive the exercise proceeds until the end of the option's life. Yet others involve lump sum
payoffs. An all-or-nothing call (put) option, for example, pays a predetermined amount (i.e., the “all”) should
the underlying commodity price be above (below) the exercise price at the option's expiration. A one-touch
all-or-nothing call (put) pays a predetermined amount if the commodity price touches the exercise price at
any time during the option's life.
16.4 SUMMARY
This chapter concludes the presentation of option valuation principles and applications. First, we discussed
currency and currency futures options. Contract specifications were provided, and we noted that the
valuation of these options is a straightforward application of the constant cost-of-can-y framework developed
in Chapters 10 and 11. The cost-of-carry rate for currency options is the domestic rate of interest less the
foreign rate of interest, and the cost-of-carry rate for currency
futures options is zero. We discussed, as well, the use of currency options in hedging the currency risk that
arises in international trade or investment. Second, we discussed physical commodity futures options. In
general, no options on physical commodities trade, only options on physical commodity futures. Hence, the
valuation principles for these options also follow straightforwardly from the constant cost-of-carry
framework of the earlier chapters. The cost-of-carry rate for physical commodity futures is zero.
The remainder of the chapter focuses on exotic options. These are not exchange-traded options but
are unusual options that trade in OTC markets. We show how options on options, options on the maximum
and the minimum of two commodities, lookback options, and barrier options may be valued within a
lognormal price distribution framework. But, these are only four of a myriad of option contract designs that
exist in the OTC markets. We discuss others; however, the list is certainly incomplete given the pace with
which these new contracts are introduced.
155
For a brief review of a range of exotic options, see Hudson (1991).
Chapter 2: Mechanics of Future Markets
Section Description
2.0 Introduction
2.1 Background
2.2 Specifications of the Futures Contract
2.3 Convergence of Futures Price to Spot Price
2.4 Daily Settlement and Margins
2.5 Newspaper Quotes
2.6 Delivery
2.7 Types of Traders and Types of Orders
2.8 Regulation
2.9 Accounting and Tax
2.10 Forward Contracts vs. Futures Contracts
2.0 Introduction
This chapter covers:
 what futures are,
 how the futures markets operate,
 the convergence of futures prices to the spot price of the underlying asset, and
 the key differences between futures and forward contracts.
Futures and forward contracts are agreements to buy or sell an asset at a future time at a certain price.
2.1 Background
In March, an investor in NY calls a broker to buy 5,000 bushels of corn for delivery in July.
i. The broker passes these instructions to a trader on the floor of the CBOT.
ii. The broker requests a long position in one contract (each corn contract comprises exactly 5,000 bushels).
In March, another investor in Kansas might instruct a broker to sell 5,000 bushels of corn for July delivery. The broker
would pass instructions to short one contract to a trader on the floor of the CBOT.
The two floor traders meet, agree on a price to be paid for corn in July, and the deal is done; in this example:
 The investor in New York who agreed to buy has a long futures position in one contract.
 The investor in Kansas who agreed to sell has a short futures position in one contract.
 The price agreed to is the current futures price for July corn, which like any price, is determined by the laws of
supply and demand.
Closing Out Positions
The vast majority of futures contracts do not lead to delivery.
i. Most traders close out (take opposite trades from the original one) their positions prior to the delivery.
ii. For example, the New York investor can close out the position by selling (i.e., shorting) one July corn futures,
while the Kansas investor can close out the position on by buying one July contract.
2.2 Specifications of the Futures Contract

When developing a new contract, the exchange must specify:
 The asset
 The contract size (i.e., exactly how much of the asset will be delivered under one contract).
 Where delivery will be made, and
 When delivery will be made.
1. The Asset
A. When the asset is a commodity (e.g. cotton, orange juice, etc.), the exchange must stipulate the grade or grades
of the commodity that are acceptable.
B. For some commodities a range of grades can be delivered, the price received is adjusted depending on the
grade chosen.
C. Financial assets in futures contracts are well defined. However, some features of the Treasury bond and
Treasury note futures contracts are worth noting:
i. The underlying asset in the Treasury bond contract is any long–term U.S. Treasury bond, with a non–
callable 15 year provision and maturity of 15 years or longer.
ii. The underlying Treasury note asset is any long–term Treasury note with maturity between 6.5 and 10 years.
Options, Futures, and Other Derivatives 1/128

In these cases, the exchange uses a formula to adjust the price received according to the coupon and maturity date of
the bond delivered. This is discussed in Chapter 6 (Interest Rate Futures).
2. The Contract Size:
A. Specifies the amount of the asset that has to be delivered under one contract.
B. Is important since contracts that are too large or too small exclude many traders from hedging their exposures.
3. Delivery Arrangements
A. Although the vast majority of the futures contracts do not lead to delivery, delivery arrangements are important
in understanding the relationship between the futures price and the spot price of the asset (This is particularly
important for commodities where there may be significant transportation costs).
B. When alternative delivery locations are specified, the price received by the party with the short position is
sometimes adjusted according to the location chosen by that party.
4. Delivery Months
A. A futures contract is referenced by its delivery month.
B. Delivery months vary from contract to contract. For example:
1. currency futures have delivery months of March, June, September, and December;
2. corn futures traded on the CBOT have delivery months of March, May, July, September, and December.
Note: the exchange specifies the last day on which trading can take place, which is generally a few days before the last
day on which delivery can be made.
5. Price Quotes
A. Are given in ways that are convenient and easy to understand.
B. Examples:
1. Crude oil futures prices are quoted in dollars per barrel.
2. Bond and T–notes are quoted in dollars and 32nds of a dollar.
6. Daily Price Movement Limits (specified by the exchange)
Its purpose is to prevent large price movements from occurring because of speculative excesses.
A. If the price moves down by an amount equal to the daily price limit, the contract is said to be limit down (vice
versa for limit up).
B. Normally, trading ceases for the day once the contract is limit up or limit down.
7. Position Limits (specified by the exchange)
Its purpose is to prevent speculators from exercising undue influence on the market.
A. Position limits are the maximum number of contracts that a speculator may hold.
B. Bona fide hedgers are not affected by position limits.
2.3 Convergence of Futures Price to Spot Price

As the delivery month of a futures contract approaches, the futures price converges to the asset’s spot price.
The following examples explain why this is the case.
A. A futures price which remains above the spot price of the underlying asset during the delivery period will give rise
to an arbitrage opportunity.
i. An arbitrager would short a futures contract, buy the underlying asset, make delivery and secure a profit equal
to the amount by which the futures price exceeds the spot price.
ii. Arbitragers who exploit these situations cause the futures price to decrease.
B. A futures price which remains below the spot price of the underlying asset during the delivery period will give rise
to an arbitrage opportunity. A trader would enter into a long futures contract and wait for delivery to be made.
However, demand for the undervalued futures contract would raise the futures price.
Futures Price
Spot Price
Time
Conclusion: A futures price (above or below the spot price) will converge to the spot price of the underlying asset as
the delivery month of a futures contract draws near.

2.4 Daily Settlement and Margins

One of the key roles of the exchange: To organize trading so that contract defaults are minimized (which is where
margins come in).
1. Marking to Market – An Example
A. An investor contacts a broker on 6/3/XX to buy two December gold futures contracts (since the contract size is 100
ounces, the investor has contracted to buy a total of 200 ounces).
B. The current futures price is $400 per ounce.
C. The broker will require the investor to deposit funds in what is termed a margin account.
1. The initial margin (the amount that must be deposited at the time of the contract). Assume it is $2,000 per contract,
or $4,000 in total. At the end of each trading day, the margin account is adjusted to reflect the investor’s gain or loss
(known as marking to market the account).
Note: Marking to market can involve the broker (to cover an investor’s position at times), the exchange and the
investor.
2. Maintenance Margin
i. An investor may withdraw any balance in the margin account in excess of the initial margin.
ii. Maintenance margins (generally lower than initial margins) ensure that the balance in the margin account never
becomes negative.
a. If the balance in the margin account falls below the maintenance margin, the investor receives a margin call
(to bring margin to the initial level, the extra funds deposited being known as a variation margin).
b. Failure to provide the variation margin may result in the broker closing out the position by selling the contract.
The Table below gives an example of the operation of the margin account.
Assume: The maintenance margin is 1,500 per contract ($3,000 total).
Futures Price Daily Gain Cumulative Gain Margin Account Margin Call
Day (dollars) (Loss)(dollars) (Loss)(dollars) Balance (dollars) (dollars)
(1) (2) (3) (4) (5) (6)
400.00 4,000
June 3 397.00 (600) (600) 3,400
June 4 396.10 (180) (780) 3,220
June 5 398.20 420 (360) 3,640
June 6 397.10 (220) (580) 3,420
June 7 396.70 (80) (660) 3,340
June 10 395.40 (260) (920) 3,080
June 11 393.30 (420) (1,340) 2,660 1,340
June 12 393.60 60 (1,280) 4,060
June 13 391.80 (360) (1,640) 3,700
June 14 392.70 180 (1,460) 3,880
June 17 387.00 (1,140) (2,600) 2,740 1,260
June 18 387.00 0 (2,600) 4,000
June 19 388.10 220 (2,380) 4,220
June 20 388.70 120 (2,260) 4,340
June 21 391.00 460 (1,800) 4,800
June 24 392.30 260 (1,540) 5,060
(3) = [(2i + 1) – (2i)] × $200

(4) = Cumulative sum of (3)
(5) = [(5i – 1) + (3i)] + (6i – 1)
(6) = [If (5i) < 3,000, 4000 – (5i), 0]
Further Details
A. Some brokers allow investors to earn interest on the balance in their margin accounts.
B. To satisfy the initial margin requirements (but not subsequent margin calls), an investor can sometimes deposit
securities with the broker.
1. T–bills are used in lieu of cash (at about 90% of their face value).
2. Shares are also sometimes accepted in lieu of cash (at about 50% of their face value).

Additional Notes:
 Minimum levels for initial and maintenance margins are set by the exchange.
 A bona fide hedger (a company that produces the commodity on which the futures contract is written) is
often subject to lower margin requirements than a speculator.
 A day trade is one in which the position will be closed out in the same day.
 A spread transaction is one where the trader simultaneously takes a long position in a contract with one
delivery month and a short position in a contract on the same underlying asset with another delivery month.
 Margin requirements are the same on short futures positions as they are on long futures positions.
Clearinghouse and Clearing Margins
The exchange clearinghouse:
 acts as an intermediary or middleman in futures transactions.
 guarantees the performance of the parties to each transaction.
 has as its main task to track all the transactions that take place during a day so that it can calculate the net position
of each of its members.
A. A clearinghouse member is required to maintain a margin (known as a clearing margin) account with the
clearinghouse.
B. Every day, the account balance for each contract must equal the original margin times the number of contracts
outstanding.
C. The clearinghouse calculates the number of contracts outstanding on either a gross or a net basis.
1. The gross basis adds the total of all long positions to the total of all the short positions entered by clients.
2. The net basis allows these to be offset against each other (most exchanges currently use net margining).
Credit Risk
The purpose of the margining system is to ensure that traders do not walk away from their commitments.
 Overall the system has been successful as investors have always had their contracts honored.
 While some investors walked away from their long positions in S&P futures after the 20% decline in the S&P 500
index on October 19, 1987, which also caused some brokers to become bankrupt due to their inability to meet
margin calls on contracts they entered into on behalf of their clients, everyone who had a short futures position on
the S&P 500 got paid off.
Collateralization in OTC Markets
Credit risk has been a feature of the over–the–counter markets.
 There is always a chance that the party on the other side of an OTC trade will default.
 To reduce credit risk, the OTC market is now imitating the margining system adopted by exchanges with a
procedure known as collateralization.
Example: Consider company A and B who have an outstanding OTC counter contract.
 They could enter into a collateralization agreement where they value the contract each day using a pre–agreed
valuation method.
 Over time if the value of the contract to company A increases, company B is required to pay collateral equal to this
increase to company A.
 If the value of the contract to company A decreases, company A is required to pay collateral equal to the decrease
to company B.
Collateralization significantly reduces the credit risk in OTC contracts and is discussed further in Chapter 20.
Note: In the 1990s, collateralization agreements were used by a hedge fund, Long–Term Capital Management
(LTCM). This allowed LTCM to become highly leveraged, and although the contracts did provide credit risk
protection, the high leverage left the hedge fund vulnerable to other risks. LTCM would buy bond × (less liquid) and
short Bond Y (more liquid) and would wait for the prices of the two bonds to converge at some future time. However,
due to a major Russian default and its impact on illiquidity spread on bonds, the prices of the bonds LTCM had bought
went down and the prices of those it had shorted increased. It was required to post collateral on both. LTCM could not
make payments required under their collateralization agreements and this lead to their demise.


The WSJ’s futures quotations can currently be found in the Money and Investing section. The table below shows the
quotations for commodities as they appeared in the WSJ of Friday, 3/16/01.
LIFETIME OPEN
OPEN HIGH LOW SETTLE CHANGE HIGH LOW INTEREST
Corn (COT) 5,000 bu.; cents per bu.
May 2171/2 2173/4 2101/2 2103/4 7 2821/2 2061/2 186,129
:::
Est vol 103,000; vol Wed 60,060; Open int 431,377, +1,845
The quotes refer to the trading that took place on the previous day (3/15/01). Quotations for index futures, currency
futures, and interest rate futures are given in Chapters 3, 5 and 6 respectively.
The information in the above table is as follows:
i. The asset underlying the futures contract, the exchange that the contract is traded on, the contract size, and how the
price is quoted.
ii. The asset is corn, traded on the Chicago Board of Trade, with a contract size of 5,000 bushels, and with prices
quoted in cents per bushel.
iii. The months in which particular contracts are traded are shown in the first column.
Prices
The first three numbers in the table above are the opening price, and the highest and lowest price achieved during the
day. The opening price is representative of the prices at which contracts were trading at the opening bell.
Settlement Price
The 4th number is the settlement price, it is usually the price at which the contract traded immediately before the end of
trading for the day, and is used for calculating daily gains and losses and margin requirements.
The 5th number is the change in the settlement price from the previous day (e.g. for the May 2001 corn futures
contract, the settlement price was 210 ¾ cents on 3/15/01, down 7 cents from 3/14/01).
The settlement price is important because it is used for calculating daily gains and losses and margin requirements. In
the case of the May 2001 corn futures:
i. an investor with a long position in 1 contract would find his or her margin account balance reduced by $350 (=
5,000 × 7 cents) between 3/14/01 and 3/15/01.
ii. an investor with a short position in 1 contract would find that the margin balance increased by $350 between these
two dates.
Lifetime Highs and Lows
The 6th and 7th numbers show the highest and lowest futures price achieved in trading the contract over its lifetime.
Open Interest and Volume of Trading
The last column shows the open interest for each contract, which is the total number of contracts outstanding.
The open interest is the number of long or short positions, and is 1 trading day older than the price information.
Example. In the 3/16/01 WSJ, the open interest (186,129) is for the close of trading on 3/14/01.
At the end of each section, the above table shows:
a. the estimated volume of trading in contracts of all maturities on 3/15/01 , and
b. the actual volume of trading in these contracts on 3/14/01, and
c. the total open interest for all contracts on 3/14/01, and
d. the change in this open interest from the previous trading day.
Example: For all corn futures contracts, the estimated trading volume was 103,000 contracts on 3/15/01, the actual
trading volume was 60,060 contracts on 3/14/01, and the for all corn futures contracts was 431,377 on 3/14/01, up
1,845 from the previous trading day. Sometimes the volume of trading in a day is greater than the open interest at the
end of the day. If so, this is indicative of a large number of day trades.
2.6 Delivery
To close a position, the broker of the party with the short position sends a notice of intention to deliver to the
exchange. The exchange then selects a party with an outstanding long position to accept delivery.
Cash Settlement
Financial futures are settled in cash, since it is inconvenient or impossible to deliver the underlying asset (e.g. not
feasible to deliver a portfolio of 500 stocks in the case of a futures contract on the S&P 500).

The final settlement price is set equal to the spot price of the underlying asset at either the opening or close of trading
on that day (e.g. in the S&P 500 futures contract, all contracts are declared closed on the third Friday of the delivery
month and the final settlement price is the opening price of the index on that day).
2.7 Types of Traders and Types of Orders

A. Types of Traders. There are two main types of traders executing trades:
1. Commission brokers earn a fee for executing trades for other people.
Orders might include:
a. market order (to trade at the current market price).
b. limit order (specifies that the transaction be executed only if that price or a better one is obtained).
2. Locals trade for their own account.
Individuals taking positions (whether locals or the clients of commission brokers), can be categorized as:
a. Hedgers
b. Speculators. Speculators can be classified as:
i. Scalpers (who hold their position for only a few minutes after watching for short term trends and
attempting to profit from small changes in the contract price).
ii. Day traders (who hold their positions for less than one trading day).
iii. Position traders (who hold their positions for much longer periods of time).
c. Arbitrageurs
B. Types of Orders
1. Market order. The order is that the trade be executed immediately at the best price available in the market.
2. A limit order.
a. the order can be executed only at a particular price or at one more favorable to the investor.
b. there is no guarantee that the order will be executed at all, because the limit price may never be reached.
3. A stop order or stop–loss order.
a. the order is executed at the best available price once a bid or offer is made at that particular price or a less–
favorable price.
b. a stop order becomes a market order as soon as the specified price has been hit.
c. the purpose of a stop order is to close out a position and limit the loss that can be incurred.
4. A stop–limit order is a combination of a stop order and a limit order. The order becomes a limit order as soon as a
bid or offer is made at a price equal to or less favorable than the stop price.
5. A market–if–touched (MIT) order (a.k.a. board order).
a. is executed at the best available price after a trade occurs at a specified price or at a price more favorable than
the specified price.
b. an MIT order becomes a market order once the specified price has been hit.
c. an MIT order is designed to ensure that profits are taken if sufficiently favorable price movements occur.
6. A discretionary order or market–not–held order is traded as a market order except that execution may be delayed at
the broker's discretion in an attempt to get a better price.
Some orders specify time conditions.
a. Typically, an order is a day order and expires at the end of the trading day.
b. A time–of–day order specifies a period of time during the day when the order can be executed.
c. An open order or a good–till–canceled order is in effect until executed or until the end of trading in the
particular contract.
2.8 Regulation
Futures markets are regulated federally by the Commodity Futures Trading Commission (CFTC) which:
i. is responsible for licensing futures exchanges and approving contracts. For contracts to be approved, the contract
must have some useful economic purpose (e.g. it must serve the needs of hedgers as well as speculators).
ii. looks after the public interest, by ensuring that prices are communicated to the public and that futures traders
report their outstanding positions (if they are above certain levels)
iii. licenses all individuals who offer their services to the public in futures trading, conducts backgrounds checks and
determines minimum capital requirements.
iv. deals with complaints brought by the public and ensures that disciplinary action is taken when appropriate.

In 1982, the National Futures Association was formed and assumed some of the CFTC responsibilities. The NFA’s
objective is to prevent fraud and to ensure that the market operates in the best interests of the public. NFA requires its
members to pass an exam, and is authorized to monitor trading and take disciplinary action when appropriate.
The Securities and Exchange Commission (SEC), the Federal Reserve Board, and the U.S. Treasury Department also
have jurisdictional rights over some aspects of futures trading. These bodies are concern with the effects of futures
trading on the spot markets for securities such as stocks, Treasury bills, and Treasury bonds. Further, the SEC
currently has an effective veto over the approval of new stock or bond index futures contracts.
Trading Irregularities
One type of trading irregularity occurs when an investor group tries to "corner the market".
i. A huge long futures position is taken to exercise some control over the supply of the underlying commodity.
ii. As the maturity of the futures contracts approaches, an investor group may not close out its position, and thus, the
number of outstanding futures contracts would exceed the amount of the commodity available for delivery.
iii. Investors with short positions would realize that it is difficult to deliver, become desperate to close out their
positions, and this would result is a large rise in both futures and spot prices.
Regulators deal with this type of abuse by increasing margin requirements, imposing stricter position limits,
prohibiting trades that increase a speculator's open position, and forcing market participants to close out their positions.
Other types of trading irregularities involve traders on the floor of the exchange.
In 1989, the FBI carried out a two year investigation, due to offenses including overcharging customers, not paying
customers the full proceeds of sales, and traders using their knowledge of customer orders to trade first for themselves.
2.9 Accounting and Tax

Accounting: On 9/02 a trader takes a long position in a 3/03 corn futures contract and closes the position at the end of
2/03. Let the futures prices be 150 cents per bushel when the contract is entered into, 170 cents per bushel at the end of
2002, and 180 cents per bushel when the contract is closed out. One contract is for the delivery of 5,000 bushels.
A. If the trader is a speculator, the gains for accounting purposes are
5,000 × $0.20 = $1,000 in 2002 and
5,000 × $0.10 = $500 in 2003.
B. If the trader is hedging the purchase of 5,000 bushels of corn in 2/03, hedge accounting can be used; this:
i. means the entire $1,500 gain then appears in the 2003 income statement.
ii. allows the profit/loss from a hedging instrument to be recognized at the same time as the profit/loss from the
item being hedged.
FAS 133: Accounting for Derivative Instruments and Hedging Activities:
i. applies to all types of derivatives (including futures, forwards, swaps, and options).
ii. requires all derivatives appearing on the balance sheet to be at fair market value.
iii. increases disclosure requirements.
iv. gives companies less latitude in using hedge accounting, since it requires that the hedging instrument used to be
effective in offsetting exposures and an assessment of this effectiveness is required every 3 months.
Tax: Two key issues are the nature of a taxable gain or loss and the timing of the recognition of the gain or loss.
Note: Gains or losses are either classified as capital gains/losses or as ordinary income.
1. For a corporate taxpayer:
i. capital gains are taxed at the same rate as ordinary income.
ii. capital losses are deductible only to the extent of capital gains.
iii. A corporation may carry back a capital loss for 3 years and carry it forward for up to 5 years.
iv. Positions in futures contracts are treated as if they are closed out on the last day of the tax year, and gains and
losses are considered capital.
2. For a non-corporate taxpayer:
i. short–term capital gains are taxed at the same rate as ordinary income
ii. long–term capital gains are taxed at a lower rate than ordinary income. Make note of the definitions:
iii. Long–term capital gains are gains from the sale of a capital asset held for longer than 1 year, while Short
term capital gains are the gains from the sale of a capital asset held less than 1 year.
iv. capital losses are deductible to the extent of capital gains plus ordinary income up to $3,000 and can be carried
forward indefinitely.
v. Positions in futures contracts are considered 60% long term and 40% short term.
**Hedging transactions are exempt from this rule, and the definition of a hedge transaction for tax purposes is different
from that for accounting purposes.

Tax laws define hedging as a transaction entered-into in the normal course of business primarily for two reasons:
1. To reduce the risk of price changes/currency fluctuations with respect to property held or to be held by the
taxpayer for the purposes of producing ordinary income.
2. 2. To reduce the risk of price/interest rate/ currency fluctuations with respect to borrowings made by the taxpayer.
a. Gains or losses are treated as ordinary income.
b. The timing of the recognition of gains or losses matches that of income or deduction from the hedged items.
Special rules apply to foreign currency futures transactions. A taxpayer can elect to treat gains and losses as ordinary
income, regardless of whether the contract is entered into for hedging or speculative purposes.
If a taxpayer does not make this election, then foreign currencies futures transactions are treated in the same way as
other futures transactions.
2.10 Forward Contracts vs. Futures Contracts
Both forwards and futures are agreements to buy or sell an asset for a certain price at a certain future time.
The main differences between forward and futures contracts are summarized below.
Forward Futures
Private contract between two parties Traded on an exchange
Not standardized Standardized contract
Usually one specified delivery date Range of delivery dates
Settled at end of contract Settled daily
Delivery or cash settlement usually takes place Contract is usually closed out prior to maturity
Some credit risk Virtually no credit risk
Profits from Forward and Futures Contracts
Keep in mind the following:
 Under a forward contract, the whole gain or loss is realized at the end of the life of the contract.
 Under a futures contract, the gain or loss is realized day by day because of the daily settlement procedures.
Example: Assume the following:
 The sterling exchange rate for a 90–day forward contract is 1.6000 and that this is also the rate for the futures
price for a contract to be delivered in exactly 90 days.
 Investor A is long £1 million in a 90–day forward contract and
 Investor B is long £1 million in 90–day futures contracts. (Note: because each futures contract is for the
purchase or sale of £62,500, investor B must purchase a total of 16 contracts i.e. £62,500 × 16 contracts =
£1 million.)
 The spot exchange rate in 90 days proves to be 1.8000 dollars per pound.
What is the difference between the gains and losses under the two contracts?
 Investor A makes a gain of $200,000 on the 90th day [16 × £62,500 × ($1.80/£ – $1.60/£)].
 Investor B makes the same gain–but spread out over the 90–day period. On some days investor B may
realize a loss, whereas on other days he or she makes a gain. However, in total, when losses are netted
against gains, there is a gain of $200,000 over the 90–day period.
Foreign Exchange Quotes
a. Futures prices are always quoted as the number of U.S. dollars per unit of the foreign currency or as the number of
U.S. cents per unit of the foreign currency.
b. Forward prices are always quoted in the same way as spot prices. For other than British pound, the euro, the
Australian dollar, and the New Zealand dollar, forward quotes show the number of units of the foreign currency per
U.S. dollar (USD).
Example: Consider the Canadian dollar (CAD).
A futures price quote of 0.7050 USD per CAD corresponds to a forward price quote of 1.4184 CAD per USD (1.4184
= 1/0.7050).

Chapter 3: Hedging Strategies Using Futures
Section Description
3.0 Introduction
3.1 Basic Principles
3.2 Arguments For and Against Hedging
3.3 Basis Risk
3.4 Cross Hedging
3.5 Stock Index Futures
3.6 Rolling the Hedge Forward
Appendix Proof of Minimum Variance Hedge Ratio Formula
3.0 Introduction
Many participants in the futures markets are hedgers. The purpose of a futures hedge is to reduce risk by making the
outcome more certain. A perfect hedge is one that completely eliminates the risk. Perfect hedges are rare.
Some of the questions addressed in this chapter are:
a. When is a short or long futures position appropriate?
b. Which futures contract should be used?
c. What is the optimal size of the futures position for reducing risk?
Two additional notes:
a. The strategies discussed by the author can be described as hedge–and–forget strategies (i.e. no attempt is made to
adjust the hedge once it has been put in place).
b. The author treats futures contracts as forward contracts.
3.1 Basic Principles

The objective of a futures hedge is to take a position that neutralizes the risk as far as possible.
Example: If a company knows it will gain (lose) $10,000 for each 1 cent increase (decrease) in the price of a
commodity over a certain time period, the company should take a short futures position designed to offset this risk.
The futures position should lead to a loss (gain) of $10,000 for each 1 cent increase (decrease) in the price of the
commodity over the corresponding period of time.
Short Hedges:
A short hedge involves a short position in a futures contracts.
A short hedge is appropriate when:
a. the hedger owns an asset and expects to sell it.
b. the hedger does not currently own the asset but intends to own it some time in the future.
Assume we are given the following regarding a short hedge:
 On 5/15, an oil producer agrees to sell 1 million barrels of crude oil, at its market price, on 8/15.
 The spot price on 5/15 is $19 per barrel and the August crude oil futures price is $18.75 per barrel.
If the company is satisfied in locking in a price of $18.75 per barrel, the company can hedge its exposure to future
price changes by shorting 1,000 August futures contracts (since each futures contract is for the delivery of 1,000
barrels).
Scenario 1: Suppose the spot price on 8/15 is $17.50 per barrel. Note: the futures price on 8/15 should be very close to
the spot price of $17.50.
a. the company realizes $17.5 million for the oil under its sales contract.
b. the company gains $18.75 – $17.50 = $1.25 per barrel, or $1.25 million in total from the short futures position.
Thus, the strategy initiated in May “locked in” a price of $18.75 per barrel.
Scenario 2: Suppose the spot price on 8/15 is $19.50 per barrel.
a. the company realizes $19.5 million for the oil under its sales contract.
b. the company loses $19.50 – $18.75 = $0.75 per barrel, or $750,000 on its short futures position.
Thus, the strategy initiated in May “locked in” a price of $18.75 per barrel.
Long Hedges
A long hedge involves taking a long position in a futures contract. A long hedge is appropriate when:
a. a company knows it will have to purchase a certain asset in the future and wants to lock in a price now.
b. the goal is to partially offset an existing short position. For an investor who has shorted a certain stock, part of the
risk faced by the investor is related to a downturn in the market. This portion of the risk can be neutralized by
taking a long position in index futures contracts.

A long hedge example:

 On 1/15, a company knows it will require 100,000 pounds of copper on 5/15 to meet a certain contract.
 The spot price of copper is 140 cents per pound, and the May futures price is 120 cents per pound.
If the company is satisfied in locking in a price of 120 cents per pound, it can hedge its position by taking a long
position in four May futures contracts and closing its position on May 15. Note: four contracts are necessary since each
contract is for the delivery of 25,000 pounds of copper.
Scenario 1: Suppose that the price of copper on 5/15 proves to be 125 cents per pound. Note: the futures price on 5/15
should be very close to the spot price of 125 cents per pound
a. The company gains 100,000 × ($1.25 – $1.20) = $5,000 on its futures contract.
b. The company pays 100,000 × $1.25 = $125,000 for the copper
Thus, the net cost to the company is approximately $125,000 – $5, 000 = $120,000.
Scenario 2: Suppose that the price of copper on 5/15 proves to be 105 cents per pound. Note: the futures price on 5/15
should be very close to the spot price of 105 cents per pound.
a. The company loses 100,000 × ($1.20 – $1.05) = $15,000 on its futures contract.
b. The company pays 100,000 × $1.05 = $105,000 for the copper.
Thus, the net cost to the company is approximately $15,000 + $105,000 = $120,000.
Final Notes:
a. It makes sense for the company to use futures contracts vs. buying copper spot on 1/15, since the company would
have to pay 140 cents vs. 120 cents per pound and would incur interest and storage costs.
b. Making or taking delivery can be costly, and therefore hedgers with long positions usually close out their positions
before the delivery period.
c. The examples above assumed that a futures contract is the same as a forward contract. This is not completely
accurate since the “marking to market” for futures contracts has a small effect on the performance of a hedge.
3.2 Arguments For and Against Hedging

Most companies are in the business of manufacturing, retailing, wholesaling or providing a service. Since the have no
skills in predicting interest rates, exchange rates, and commodity prices, it makes sense for them to hedge these risks as
they arise.
However many risks are left unhedged. The reasons why are as follows:
1. Shareholders can hedge themselves. However:
a. this assumes shareholders have as much information about the risks faced by a company as does company
management, which in most instances, this is not the case.
b. commissions and other transactions costs are more expensive per dollar of hedging for small transactions than
for large transactions.
c. the size of futures contracts makes hedging by shareholders impossible in many situations.
2. Shareholders can diversify risk more easily than a corporation. Further, if companies act in the best interests of
well–diversified shareholders, hedging is unnecessary in many situations.
3. If hedging is not the norm in an industry, it may not make sense for a company hedge. Consider the danger in
hedging when competitors do not hedge. Example:
Change in Effect on price of Effect on profits of Effect on profits of
gold price gold jewelry Take a Chance Co. Safe and Sure Co.
Increase Increase None Increase
Decrease Decrease None Decrease
Thus, hedging using futures contracts can result in a decrease or an increase in a company’s profits relative to the
position it would be in with no hedging. Therefore, all implications of price changes on a company’s profitability
should be taken into account when designing a hedging strategy to protect against the price changes.
3.3 Basis Risk

In practice, hedging is not as straightforward as described in the previous examples.
Three reasons why hedging using futures contracts works less than perfectly in practice are
1. The asset whose price is being hedged and the asset underlying the futures contract may not be the same.
2. The exact date when the asset will be bought or sold may be uncertain.
3. The futures contract may need to be closed out well before its expiration date.
These problems give rise to what is termed basis risk.

The basis in a hedging situation is computed as follows:

Basis = Spot price of asset to be hedged – Futures price of contract used
Prior to expiration the basis may be positive or negative.
Strengthening of the basis occurs when the spot price increases by more than the futures price.
Weakening of the basis occurs when the futures price increases by more than the spot price.
Basis risk can be defined using the following variables:

S1: spot price at time t1 S2: spot price at time t2 F1: futures price at time t1
F2: futures price at time t2 b1: basis at time t1 b2: basis at time t2
Variation of basis over time

Spot Price
Future Price
Time
t1 t2
Example:
Assume a hedge is put in place at time t1 and closed out at time t2.
Given the following: S1 = 2.50, F1 = 2.20, S2 = 2.00, and F2 = 1.90.
Compute the basis as b1 = S1 – F1 = 2.50 – 2.20 = 0.30 and
b2 = S2 – F2 = 2.00 – 1.90 = 0.10
Scenario 1: Short Hedge: (the asset will be sold at time t2 and a short futures position is taken at t1).
a. The price realized for the asset is S2 and the profit on the futures position is F1 – F2.
The effective price obtained for the asset with hedging is S2 + F1 – F2 = F1 + b2 = 2.30
b. A perfect hedge (i.e., a hedge eliminating all uncertainty about the price obtained) results when b2 is known at the
time of the hedge. However, this is usually not the case.
Thus, basis risk is the uncertainty of the value of b2
Scenario 2. Long Hedge: (the asset will be bought at time t2 and a long futures position is taken at t1).
The price paid for the asset is S2 and the loss on the hedge is F1 – F2
The effective price obtained for the asset with hedging is S2 + F1 – F2 = F1 + b2 = 2.30
The basis risk for an investment asset stems from uncertainty as to the level of the risk–free interest rate and the asset’s
future yield.
Consider the case when the asset being hedged and the asset underlying the futures contract may not be the same. In
this case, the basis risk is then usually greater.
Define S 2* as the price of the asset underlying the futures contract at time t2.
Recall that S2 is the price of the asset being hedged at time t2.
By hedging, a company ensures that the price paid (or received) for the asset is S2 + F1 – F2 (which can be written as)
F1  ( S2*  F2 )  ( S2  S2* )
Thus, two components of the basis can be defined:
1. ( S 2*  F2 ) is the basis if the asset being hedged were the same as the asset underlying the futures contract.
2. S 2  S 2* is the basis arising from the difference between the two assets.
Note: basis risk can lead to an improvement or a worsening of a hedger’s position.
Choice of Contract
The choice of the futures contract is a key factor affecting basis risk. The choice has two components:
1. The choice of the asset underlying the futures contract.
If the asset being hedged does not match the asset underlying a futures contract, it is necessary to determine which of
the available futures contracts has futures prices most closely correlated with the price of the asset being hedged.
2. The choice of the delivery month (which is influenced by the following factors).
a. Price volatility during the month of delivery. Thus, a contract with a later delivery month is usually chosen.
b. Taking actual possession. A long hedger runs the risk of taking delivery of the physical asset if the contract is
held during the delivery month, and this can be expensive and inconvenient.

Thus, a good rule of thumb is to choose a delivery month that is as close as possible to, but later than, the expiration of
the hedge.
Example: Assuming the following information is given:
 On 3/1, a U.S. company is notified that it will receive 50 million Japanese yen at the end of July.
 Yen futures contracts have delivery months of March, June, September, and December, and one contract is for the
delivery of 12.5 million yen.
 The company shorts four September yen futures contracts on 3/1.
 The company closes out its position when the yen are received at the end of July.
 The futures price on 3/1 in cents per yen is 0.7800 and the spot and futures prices when the contract is closed out
are 0.7200 and 0.7250, respectively.
1. Compute the gain on the futures contract. 0.7800 – 0.7250 = 0.0550 cents per yen.
2. Compute the basis when the contract is closed out.
Basis equals 0.7200 – 0.7250 = –0.0050 cents per yen
3. Compute the effective price obtained in cents per yen:
Effective price equals the final spot price plus the gain on the futures: 0.7200 + 0.0550 = 0.7750
Effective price also equals the initial futures price plus the final basis: 0.7800 – 0.0050 = 0.7750
The total amount received by the company for the 50 million yen is 50 × 0.00775 million dollars, or $387,500.
Example: Assuming the following information:
 On 6/8, a company knows it will need to purchase 20,000 barrels of crude oil some time in October or November.
The futures price on 6/8 is $18.00 per barrel.
 Oil futures contracts are traded for delivery every month and the contract size is 1,000 barrels.
 The company decides to use the December contract and takes a long position in 20 December contracts.
 The company closes out its futures contract on 11/10, when the spot price and futures price are $20.00 per barrel
and $19.10 per barrel.
1. Compute the gain on the futures contract: 19.10 – 18.00 = $1.10 per barrel.
2. Compute the basis when the contract is closed out: 20.00 – 19.10 = $0.90 per barrel.
3. Compute the effective price paid (in dollars per barrel):
Effective price paid equals the final spot price less the gain on the futures: 20.00 – 1.10 = 18.90.
Effective price paid also equals the initial futures price plus the final basis: 18.00 + 0.90 = 18.90
The total price paid is 18.90 × 20,000 = $378,000.
3.4 Cross Hedging
Cross hedging occurs when the two assets are different. If an airline is concerned about the future price of jet fuel and
there is no futures contract on jet fuel, it might choose to use heating oil futures contracts to hedge its exposure.
A hedge ratio:
 is the ratio of the size of the position taken in futures contracts to the size of the exposure.
 of 1.0 is not necessarily optimal, if the objective of the hedger is to minimize risk.
The following notation is used to explain the hedge ratio:
S: the change in spot price, S, during the life of the hedge.
F: the change in futures price, F, during the life of the hedge.
S: the standard deviation of S
F: the standard deviation of F
: coefficient of correlation between S and F
h*: hedge ratio that minimizes the variance of the hedger’s position
Appendix A proves that h* is determined as follows: h* =  (S /F).
If  = 1 and F = S, the hedge ratio, h*, is 1.0 and the futures price mirrors the spot price.
If  = 1 and F = 2S, the hedge ratio h* is 0.5 so the futures price always changes by twice as much as the spot price.
The optimal hedge ratio, h*, is the slope of the best fit line when S is regressed against F. The hedge effectiveness is
 F2
defined as the proportion of the variance that is eliminated by hedging. This is 2, or h*2
 S2
Optimal Number of Contracts
To determine the optimal number of contracts, we first define the following variables:
NA: Size of position being hedged (units) QF: Size of one futures contract (units)
N *: Optimal number of futures contracts for hedging

The futures contracts used should have a face value of h*NA. The number of futures contracts is required is therefore
h* N A
given by N * 
QF
Example: Hedging the price of fuel by buying futures contracts on heating oil given:
 A company will buy 2 million gallons of jet fuel in 1 month.
 S (standard deviation of the change in the price per gallon of jet fuel) = 0.0263.
 F (standard deviation of the change in the futures price) = 0.0313
  (correlation coefficient between the one month change in the price of jet fuel and the futures price) = 0.928
Compute the optimal hedge ratio: h  0.928  0.0263  0.78
0.0313
Compute the optimal number of contracts to buy:
Given that a heating oil futures contract is on 42,000 gallons, the company should therefore buy
2 M gal
0.78   37.14  37 contracts
42 K gal/contract

Stock index futures and how they are used to hedge or manage exposures to equity prices
A stock index tracks changes in the value of a hypothetical portfolio of stocks.
 The weight of a stock in the portfolio equals the proportion of the portfolio invested in the stock.
 The percentage increase in the stock index over a small interval of time is set equal to the percentage increase in
the value of the hypothetical portfolio.
 Dividends are usually not included in the calculation so that the index tracks the capital gain/loss from investing in
the portfolio.
 If the portfolio of stocks remains fixed, the weights assigned to individual stocks do not remain fixed.
When the price of one stock in the portfolio rises more sharply than others, more weight is automatically given to that
stock.
Some indices are constructed by selecting one of each of a number of stocks.
The weights assigned to the stocks are then proportional to their market prices, with adjustments made when there are
stock splits.
Other indices are constructed so that weights are proportional to market capitalization (stock price × number of shares
outstanding).
The portfolio is automatically adjusted to reflect stock splits, stock dividends, and new equity issues.
Table 3.3 shows futures prices for contracts on a number of different stock indices as they were reported in the Wall
Street Journal of February 5, 2004. The prices refer to the close of trading on February 4, 2004.
Types of stock indices, reported in the Wall Street Journal, follows:
1. The Dow Jones Industrial Average consists of 30 blue–chip stocks in the U.S. The weights given to the stocks are
proportional to their prices.
2. The Chicago Board of Trade trades two contracts on the index. One is on $10 times the index. The other (the Mini
DJ Industrial Average) is on $5 times the index.
3. The Standard & Poor’s 500 (S&P 500) Index is based on a portfolio of 500 different stocks: 400 industrials, 40
utilities, 20 transportation companies, and 40 financial institutions. The weights of the stocks in the portfolio at any
given time are proportional to their market capitalizations. This index accounts for 80% of the market
capitalization of all the stocks listed on the NYSE
4. The Chicago Mercantile Exchange (CME) trades two contracts on the S&P 500. One is on $250 times the index;
the other (the Mini S&P 500 contract) is on $50 times the index.
5. The Standard & Poor’s MidCap 400 Index is similar to the S&P 500, but based on a portfolio of 400 stocks that
have lower market capitalizations.
6. The Nasdaq 100 is based on 100 stocks using the NASD Automatic Quotations Service. The CME trades two
contracts. One is on $100 times the index; the other the Mini Nasdaq 100 contract) is on $20 times the index.
7. The Russell 2000 Index is an index of the prices of 2,000 small capitalization stocks in the U.S.
8. The Russell 1000 Index is an index of the prices of the 1,000 largest capitalization stocks in the U.S.
9. The NYSE Composite Index is an index of all stocks trading on the New York Stock Exchange.

10. The US Dollar Index is a trade–weighted index of the values of six foreign currencies (the euro, yen, pound,
Canadian dollar, Swedish krona, and Swiss franc).
11. The Nikkei 225 Stock Average is based on a portfolio of 225 of the largest stocks trading on the Tokyo Stock
Exchange. Stocks are weighted according to their prices. One futures contract (traded on the CME) is on $5 times
the index.
12. The Share Price Index is the All Ordinaries Share Price Index, a broadly based index of Australian stocks. The
CAC–40 Index is based on 40 large stocks trading in France.
13. The Xetra DAX Index is based on 30 stocks trading in Germany.
14. The FTSE 100 Index is based on a portfolio of 100 major UK stocks listed on the London Stock Exchange.
15. The DJ Euro Stoxx 50 Index and the DJ Stoxx 50 Index are two different indices of blue–chip European stocks
compiled by Dow Jones and its European partners. The futures contracts on these indices trade on Eurex and are
on 10 times the values of the indices measured in euros.
16. The other indices shown in Table 3.3 are not stock indices.
 The DJ–AIG commodity index and the GSCI index futures contract track commodity prices.
 The TRAKRS long–short tech index is designed to reflect the performance of a portfolio that is long
individual technology stocks and short financial instruments representing technology sectors.
Recall:
 Futures contracts on stock indices are settled in cash, not by delivery of the underlying asset.
 All contracts are marked to market to either the opening price or the closing price of the index on the last trading
day, and the positions are then deemed to be closed. For example, contracts on the S&P 500 are closed out at the
opening price of the S&P 500 index on the 3rd Friday of the delivery month.
Stock index futures can be used to hedge an equity portfolio. We begin by defining the following notation:
P: the current value of the portfolio.
A: the current value of the stocks underlying one futures contract.
: the appropriate hedge ratio for a portfolio.
is the slope of the best fit line when excess return on the portfolio over the risk–free rate is regressed against the
excess return of the market over the risk–free rate.
N*: the optimal number of futures contracts for hedging.
1. If the portfolio mirrors the index, a hedge ratio of 1.0 is appropriate, and the number of futures contracts that
P
should be shorted is N * 
A
P
2. When the portfolio does not mirror the index, the N *  
A
When  = 1.0, the return on the portfolio tends to mirror the return on the market
When  = 2.0, the excess return on the portfolio tends to be twice as great as the excess return on the market.
When  = 0.5, it tends to be half as great.
Example: Compute the number of contracts that need to be shorted to hedge a portfolio of $1 million, given that the
current value of the index is 1.000, and each futures contract is on $250 times the index.
$1M
N *  1.0 4
$250  $1,000
The computation of Beta using the Capital Asset pricing model:
If the expected return–beta relationship holds, then the expected rate of return (as a %) on any portfolio i is
Cov( ri , rM )
E ( ri )  rf  i [ E ( Rm )  rf ] , where i 
M
2
Example: A futures contract on the S&P 500 with 4 months to maturity will be used to hedge the value of the portfolio
over the next 3 months. You are given the following information:
The value of S&P 500 index: 1,000; The value of portfolio: $5,000,000; The risk–free interest rate: 10% pa (2.5% per
3 months); The dividend yield on index: 4% pa (1.0% per 3 months); The Beta of the portfolio: 1.5.
One futures contract is for delivery of $250 times the index.
1. Compute the current futures price: 1,000e (0.10 – 0.04) × 4/12 = 1,020.20
5,000,000
2. Compute the number of futures contracts that should be shorted to hedge the portfolio: 1.5   30
250,000

3. If the index turns out to be 900 in 3 months, compute the futures price: 900e(0.10 – 0.04) × 1/12 = 904.51
4. Compute the gain from the short futures position: 30 × (1,020.20 – 904.51) × 250 = $867,676
5. Compute the expected return (as a %) on the portfolio.
a. The return on the index over a 3 month period is a loss of 10% [(900/1000) – 1.0] = –0.10
b. The return from the dividend yield on index is 1%.
c. The net return on the index is –0.10 + 0.01 = –0.09.
d. The expected rate of return on the portfolio can be computed using: E ( ri )  rf  i [ E ( Rm )  rf ]
The expected return (in percentage) on the portfolio equals 2.5 + [1.5 × (–9.0 – 2.5)] = –14.75
6. Compute the expected value of the portfolio (inclusive of dividends) at the end of the 3 months:
$5,000,000 × (1 – 0.1475) = $4,262,500.
7. Compute the expected value of the hedger’s position (including the gain on the hedge):
$4,262,500 + $867,676 = $5,130,176
The table below shows that the value of the position in 3 months depends on the value of the index in 3 months:
Value of index In three months 900.00 950.00 1,000.00 1,050.00 1,100.00
Three months total value of position ($000) 5,130,176 5,128,296 5,126,416 5,124,537 5,122,657
Conclusions:
a. The hedging scheme results in a value for the hedger’s position close to $5,125,000 at the end of 3 months.
b. The effect of the hedge results in the hedger’s position growing at the risk–free rate (2.5% per 3 months).
Reasons for Hedging an Equity Portfolio
Q. Why the hedger use futures contracts to hedge when the hedger can sell the portfolio and buy Treasury bills?
A1. The hedger feels that the stocks in the portfolio have been chosen well and will outperform the market.
Therefore, a hedge using index futures removes the risk arising from market moves and leaves the hedger exposed only
to the performance of the portfolio relative to the market.
A2. The hedger is planning to hold a portfolio for a long period of time and requires short–term protection.
Changing Beta of the hedgers’ portfolio
Based on the example above, the purchase of 30 futures contracts reduces the beta of the portfolio to zero.
To alter the beta of the portfolio to something other than zero, we adjust the formula to determine the number of
contracts purchased. Therefore,
a. To change the beta of the portfolio from  to  *, where  >  * a short position in ( –  *)(P/A) number of
contracts is required.
b. To change the beta of the portfolio from  to  *, where  <  * a long position in ( –  *)(P/A) number of
contracts is required.
Exposure to the Price of an Individual Stock
Some exchanges trade futures contracts on selected individual stocks, but in most cases a position in an individual
stock can only be hedged using a stock index futures contact.
Example: You are given the following information.
 An investor holding 20,000 IBM shares, worth $100 per share, is concerned about the market’s volatility over the
next month and chooses to hedge the position using an August futures contract on the S&P 500.
 Each contract is for delivery of $250 times the index.
 The  of IBM is 1.1.
 The current level of the index is 900, and the current futures price for the August contract is 908.
20,000  $100 2,000,000
Q1. Compute the number of contracts that should be shorted: 1.1   1.1   9.78  10
900  250 225,000
Assume IBM rises to $125 during the month, and the futures price of the S&P 500 rises to 1080.
Q1. Compute the effect of the hedge:
The investor’s gains $500,000 [20,000 × ($125 – $100)] on the shares of IBM; the investor loses $430,000 [10 × $250
× (1080 – 908)] on the futures index; the net effect is a gain to the investor of $70,000
Q2. What was the purpose of the hedge?
A. To reduce risk making unfavorable outcomes less unfavorable and favorable outcomes less favorable.
3.6 Rolling the Hedge Forward

When the hedge expiration date is later than the delivery date, short hedges must be rolled forward.
Example: A series of short hedges can be used to reduce the price risk for an asset to be received at time T.

The company can use the following strategy:

Time t1: Short futures on contract 1
Time t2: Close out futures on contract 1 and Short futures on contract 2
Time t3: Close out futures on contract 2 and Short futures on contract 3
:::::
Time tn–1: Close out futures on contract n–1 and Short futures contract n.
Time T: Close out futures on contract n
Notes:
1. There are n basis risks or sources of uncertainty.
2. Rollover basis: On each of the n – 1 occasions when the hedge is rolled forward, there is uncertainty about the
futures price for the contract being closed out and the futures price for the new contract being entered into.
3. Flexibility on the exact time when a switch is made from one contract to the next reduces rollover basis risk.
Example:
 In 4/02, you realize it will have to sell 100,000 barrels in 6/03 and decides to hedge with a hedge ratio of 1.0.
 Only six month futures contracts have sufficient liquidity to meet your needs.
 The current spot price, S0, is $19, and you short 100 – 10/02 contracts.
 In 9/02, you roll the hedge forward into a 3/03 contract, and in 2/03 roll the hedge forward into a 7/03 contract.
 Assume the following futures contracts can be shorted and closed out at the prices below:
i. the 10/02 futures contract can be shorted at $18.20 per barrel, closed out at $17.40 per barrel.
ii. the 3/03 contract can be shorted at $17.00 per barrel and closed out at $16.50 per barrel.
iii. the 7/03 contract can shorted at $16.30 per barrel and closed out at $15.90.
If the price of oil drops from $19 to $16 per barrel between 4/02 and 6/03, the profit on the series of short hedges.
A profit of $0.80 ($18.20 – $17.40) per barrel was made on the 10/02 futures contract.
Thus, the futures contracts provided a total of $1.70 per barrel compensation for the $3 per barrel oil price decline.
Metallgesellschaft
Rolling a hedge forward can lead to cash flow pressures, as Metallgesellschaft (MG) a German company experienced
in the early 1990s.
 MG sold a huge volume of 5– to 10–year heating oil and gasoline fixed–price supply contracts to its customers
 It hedged its exposure with long positions in short–dated futures contracts that were rolled over.
 When the price of oil fell, margin calls on the futures position were made. This lead to huge short–term cash flow
pressures. The company closed out all the hedge positions lost $1.33 billion.
Appendix Proof of Minimum Variance Hedge Ratio Formula

Suppose we expect to sell NA units of an asset at time t, and choose to hedge at time t1 by shorting futures contracts on
N
NF units of a similar asset. The hedge ratio, which we will denote by h, is h  F
NA
Denote the total amount realized for the asset when the profit or loss on the hedge is taken into account by Y, so that:
Y = S2 ×NA – (F2 – F1) ×NF; or Y = S1 ×NA + (S2 – S1) ×NF– (F2 – F1) ×NF
where S1 and S2 are the asset prices at times t1 and t2, and F1 and F2 are the futures prices at times t1 and t2. Using the
above equations, Y can be written as: Y = S1×NA + NA ×(S – h F), where S = S2 – S1 and F = F2 – F1
Because S1 and NA are known at time t1, the variance of Y is minimized when the variance of S – h F is minimized.
Proof of Minimum Variance Hedge Ratio Formula (continued):
The variance of S  hF is v   S2  h 2 F2  2h  s f
where s, f, and  are as defined in Section 3.4.
v v
Setting , we have  2h F2  2  S F  0 ; since  2 v / h 2 is positive, the value of h that minimizes the variance
h h
is h    S /  S  .

Chapter 5: Determination of Forward and Futures Prices
Section Description
5.0 Introduction
5.1 Investment Assets vs. Consumption Assets
5.2 Short Selling
5.3 Assumptions and Notation
5.4 Forward Price of an Investment Asset
5.5 Known Income
5.6 Known Yield
5.7 Valuing Forward Contracts
5.8 Are Forward and Future Prices Equal?
5.9 Futures Prices of Stock Indices
5.10 Forward and Future Contracts on Currencies
5.11 Futures on Commodities
5.12 Cost of Carry
5.13 Delivery Options
5.14 Futures Prices and the Expected Future Spot Price
5.0 Introduction
This chapter examines how forward and futures prices relate to the spot price of the underlying asset.
Forward contracts are settled at the end of the contract while futures contracts are settled daily. The settlement of a
forward contract by a single payment at maturity makes forward contracts easier to analyze than futures contracts.
However, forward and futures prices of an asset are equal when the maturities of the two contracts are the same.
5.1 Investment Assets vs. Consumption Assets

When pricing forward and future contracts, it is necessary to distinguish between two types of underlying assets:
Investment assets and Consumption assets
A. Investment assets are those held for investment purposes by a large number of investors. Stocks, bonds, gold and
silver are examples of investment assets.
B. Consumption assets are held for consumption, and usually not for investment purposes. Copper, oil and pork
bellies are examples of consumption assets.
5.2 Short Selling

Short selling (“shorting”) involves selling an asset that is not owned with the intention of buying it back later. Shorting
yields a profit when the price of the asset goes down and a loss when it goes up. Details of the short selling process:
A. An investor instructs a broker to short shares. The broker then borrows shares from another client, sells them in the
open market, and deposits the sale proceeds to the investor's account.
B. Any income (dividends or interest) received on the securities that have been shorted must be transferred to the
broker, who transfers this to the account of the client from whom the securities were borrowed.
C. Provided there are enough shares, the investor can maintain the short position.
D. An investor can be short–squeezed at any time during the contract, when the broker runs out of shares to borrow.
The investor must then close out the position immediately.
Example: Computing gains (or losses) on a short sale.
Gain (or loss) equals initial proceeds received from the short sale minus dividend or interest income received minus
amount paid by investor to cover the short sale.
Given: An investor shorts 500 shares of IBM in April when the price per share equals $120.
IBM pays a dividend of $1 per share in May.
The investor covers the short position in July when the price per share equals $100.
The net gain to the investor equals 500×($120 – $1 – $100) = $9,500.
The investor is required to maintain a margin account with the broker.
 The margin account consists of cash or marketable securities deposited by the investor with the broker to guarantee
that the investor will not walk away from the short position if the share price increases.
 An initial margin is required and if there are adverse movements (i.e., price increases in the asset that is being
shorted), additional margin may be required.

 Interest is paid on the balance in margin accounts and, if the interest rate offered is unacceptable, marketable
securities such as Treasury bills can be used to meet margin requirements.
 The proceeds of the sale of the asset belong to the investor and normally form part of the initial margin.
Examples of cash flows from short sale and purchase of shares.
Purchase of shares
April: Purchase 500 shares for $120 –$60,000
May: Receive dividend +$500
July: Sell 500 shares for $100 per share +$50,000
Net profit = –$9,500
Short sale of shares
April: Borrow 500 shares and sell them for $120 +$60,000
May: Pay dividend –$500
July: Buy 500 shares for $100 per share (replace shares and close position) –$50,000
Net profit = +$9,500
Note: Securities used to be sold short only on an uptick (when the most recent movement in the price is up).

Assume that there are some market participants that:
 are subject to no transactions costs when they trade.
 are subject to the same tax rate on all net trading profits.
 can borrow and lend money at the risk–free rate.
 take advantage of arbitrage opportunities when they occur.
The following symbols are defined and used throughout:

Symbol Definition
T Time until delivery date in a forward or futures contract (in years)
S0 Price of asset underlying the forward or futures contract today
F0 Forward or futures price today
r Zero risk–free rate of interest, with continuous compounding, for an investment maturing at time T
The risk–free rate, r, is in theory the rate at which money is borrowed or lent when there is no credit risk, so that the
money is certain to be repaid. Financial institutions in derivatives markets assume that LIBOR rates rather than
Treasury rates are risk–free rates (see Chapter 4).
5.4 Forward Price of an Investment Asset

Consider the relationship between the forward price and spot price of an investment asset that pays no income. Assume
you are asked to price a long forward contract to purchase a non–dividend–paying stock in three months.
Example1: Suppose that the forward price is relatively high at $43.
 An arbitrageur can borrow $40 at r = 5% per annum, buy one share, and short a forward contract to sell one share
in 3 months.
 At the end of the 3 months, the arbitrageur delivers the share and receives $43.
 The sum of money required to pay off the loan is 40e0.05×3/12 = $40.50
The arbitrageur locks in a profit of $43.00 – $40.50 = $2.50 at the end of the 3–month period.
Example 2: Suppose the forward price is relatively low at $39.
 An arbitrageur can short one share, invest the proceeds of the short sale at 5% per annum for 3 months, and take a
long position in a 3–month forward contract.
 The proceeds of the short sale grow to 40e0.05×3/12 = $40.50 in 3 months.
 At the end of the 3 months, the arbitrageur pays $39, takes delivery of the share under the terms of the forward
contract, and uses it to close out the short position.
A net gain of $40.50 – $39.00 = $1.50 is made at the end of the 3 months.
Arbitrage opportunities when forward price is out of line with spot price for asset providing no income. (Asset price =
$40; interest rate = 5%; maturity of forward contract = 3 months.)

Forward Price = $43 Forward Price = $39

Action now: Action now:
Borrow $40 at 5% for 3 months Short 1 unit of asset to realize $40
Buy one unit of asset Invest $40 at 5% for 3 months
Enter into forward contract to sell Enter into a forward contract to buy
asset in 3 months for $43 asset in 3 months for $39
Action in 3 months: Action in 3 months

Sell asset for $43 Buy asset for $39
Use $40.50 to repay loan with interest Close short position
Receive $40.50 from investment
Profit realized = $2.50 Profit realized = $1.50
Under what circumstances do arbitrage opportunities not exist?
 The first arbitrage works when the forward price is greater than $40.50.
 The second arbitrage works when the forward price is less than $40.50.
Thus, for there to be no arbitrage the forward price must be exactly $40.50.
Based on the above, we can determine the formula to compute the forward price as F0 = S0 erT
Example: Let S0 = 40, r = 0.05; T = 0.25. The forward price is F0 = 40e0.05×0.25 = 40.50.
If F0 > S0erT, arbitrageurs borrow at the risk free rate of interest, buy the asset, short forward contracts on the asset,
repay the loan at the end of the period equal to S0erT, deliver the shares at the end of the period, and lock in a profit
equal to F0 – S0erT.
If F0 < S0erT, arbitrageurs short the asset and buy forward contracts on the asset.
Example: An investor wants to price a 4 month forward contract to buy a zero–coupon bond maturing one year from
today. The current price of the bond is $930 and the 4 month risk–free rate of interest (continuously compounded) is
6% per annum. Thus, S0 = 930, r = 0.06; T = 0.333.
The forward price is F0 = 930e0.06×0.333 = 948.79.
Recapping Arbitrage Opportunities:
Assume the underlying asset in question is gold. Further, assume no storage costs or income.
Scenario A: If F0 > S0erT, an arbitrager is likely to:
1. Borrow S0 dollars at an interest rate r for T years.
2. Buy one ounce of gold.
3. Short a forward contract on one ounce of gold.
At time T:
1. One ounce of gold is sold for F0.
2. An amount equal to S0erT repays the loan.
3. The arbitrager makes a profit of F0 – S0erT.
Scenario B: If F0 < S0erT, an arbitrager with one ounce of gold can:
1. Sell the gold for S0.
2. Invest the proceeds at interest rate r for time T.
3. Take a long position in a forward contract on one ounce of gold.
At time T:
1. The cash invested has grown to S0erT.
2. The gold is repurchased for F0
3. The arbitrager makes a profit of S0erT – F0
In a competitive market, the relationship defined by F0 = S0erT holds, and therefore arbitrage opportunities fail to exist.
5.5 Known Income

Consider forward contracts on an investment asset that provides predictable cash income (e.g. a coupon bond).
Assume: A coupon bond has a current price of $900 and matures in 5 years. A forward contract under consideration by
arbitragers is a contract to purchase a four year bond in one year. Coupon payments of $40 are expected after 6 months
and 12 months. The 6 month and 1 year risk–free interest rates (continuously compounded) are 9% and 10% per
annum.
Scenario 1: Assume the forward price equals $930. Since this is above its theoretical value ($912.39):
1. An arbitrageur borrows $900 in total to buy the bond and short a forward contract.

2. The present value of the first coupon payment is ( 40  e 0.090.50  38.24 ). 38.24 is borrowed and invested at 9%
so that in six months it repays the first coupon payment.
3. 900 – 38.24 = 861.76 is borrowed at 10% for one year and 861.76 × 1.1 = $952.39 needs to be repaid at the end of
the year.
4. The second coupon provides $40 toward this amount, and $930 is received for the bond under the terms of the
forward contract. A net profit of $40 + $930 – $952.39 = $17.61 is made.
Scenario 2: Assume the forward price equals $905. Since this is below its theoretical value ($912.39):
1. An arbitrager who holds the bond can sell the bond and enter into a long forward contract.
2. Using the values in the previous scenario, the arbitrager’s gain is $952.39 – $40 – $905 = $7.39
Conclusions regarding the scenarios:
The first scenario a profit of $17.61 when the forward price ($930) is greater its theoretical price ($912.39).
The 2nd scenario produces a profit of $7.39 when the forward price ($905) is less than $912.39.
Therefore, when no arbitrage opportunities exist, the forward price must be $912.39.
A Generalization (for investments generating known income)
When an investment asset generates a known cash income during the life of a forward contract:
F0 = (S0 – I )erT , where I equals the present value of the income.
In the previous example, S0 = 900.00; I = 40e–0.09 × 0.5 + 40e–0.10 × 1 = 74.433; r = 0.10, and T = 1.0
Thus, F0 = (900.00 – 74.433)e0.10 × 1 = $912.39
5.6 Known Income
Example: Consider a 10–month forward contract on a stock with a price of $50.
Assume: The risk–free rate of interest (continuously compounded) is 8% per annum for all maturities.
Dividends of $0.75 per share are expected after 3, 6, and 9 months. Thus, S0 = 50, r = 0.08, T = 10/12
Compute the forward price.
Step 1: Compute the PV of the dividends, I = 0.75e–0.08 × 3/12 + 0.75e–0.08 × 6/12 + 0.75e–0.08 × 9/12 = 2.162
Step 2: Using the formula above and the value from Step 1, find the forward price: F0 = (50 – 2.162)e0.08 × 10/12 = $51.14
5.6 Known Yield

Consider the situation where the asset underlying a forward contract provides a dividend yield.
Define q as the average yield per annum on an asset during the life of a forward contract.
The forward price for an investment asset providing a continuous dividend yield at rate q is
F0 = S0 × e(r – q)T
Example: Consider a 6 month forward contract on an asset expected to provide income equal to 2% of the asset price
once during a six–month period. The yield is 4% per annum with semiannual compounding, and 3.96% per annum
with continuous compounding. Note: 2 ln(1.02) = 0.0396.
Let S0 = 25, r = 0.10, T = 0.5, q = 0.0396.
The forward price equals F0 = 25× e(0.10 – 0.0396)×0.50 = 25.77.
5.7 Valuing Outstanding Forward Contracts

This section describes how the value of forward contracts is determined. Additional notation is given first.
Symbol Definition
F0 the current forward price for contract that was negotiated some time ago
K the delivery price in the contract
f the value of a long forward contract today
The value of a forward contract is zero at inceptin and may be positive or negative, at a later point.
At time of Negotiation: As time passes:
F0 = K F0 changes
f=0 f changes
Valuing Forward Contracts
Formulas, The value of a long forward on an investment asset that provides no income can be expressed in two ways:
i. f  ( F0  K )e  rT
ii. f  S0  Ke  rT
Example: A long forward contract on a non–dividend–paying stock was entered into some time ago. The contract has 6
months to maturity.

The risk–free rate of interest (with continuous compounding) is 10% per annum.
The stock price is $25, and the delivery price is $24.
Thus, S0 = 25, r = 0.10, T = 0.5, and K = 24.
Compute:
The 6 month forward price. F0  S0  e rT  25  e0.100.50  26.28
The value of the forward contract (using the 1st formula): f  (26.28  24)  e 0.100.50  2.17
The value of the forward contract (using the 2nd formula): f  25  24  e 0.100.50  2.17
Other Formulas:
The value of a short forward contract with delivery price K: f  ( K  F0 )e  rT
The value of a long forward contract on an investment asset with the present value a known income (I):
f  S0  I  Ke  rT
The value of a long forward contract on an investment asset that provides a known dividend yield at rate q:
f  S0 e  qT  Ke  rT
5.8 Are Forward and Future Prices Equal?
Forward and futures prices in theory are no longer the same when interest rates vary. The following argument for the
relative prices of forwards and futures is based on the premise that S, the price of the underlying asset, is strongly
correlated with interest rates.
When S increases, an investor holding a long futures position makes an immediate gain because of the daily settlement
procedure. The gain is invested at a higher than average rate of interest.
When S decreases, an investor holding a long futures position incurs a loss. The loss is financed at a lower than
average rate of interest.
Based on the above argument,
When S is strongly positively correlated with interest rates, futures prices tend to be higher than forward prices.
When S is strongly negatively correlated with interest rates, forward prices tend to be higher than futures prices.
Additional views of the equality of forward and futures prices:
i. The differences between forward and futures prices for contracts that last only a few months are small.
ii. As the life of a futures contract increases, the differences between forward and futures contracts are liable to
become significant.
iii. Factors not reflected in theoretical models that may cause forward and futures prices to be different include taxes,
transactions costs, and the treatment of margins.
iv. Empirical research provides conflicting evidence as to whether statistical differences exist between forward and
futures prices.
5.9 Futures Prices of Stock Indices

A stock index tracks changes in the value of a portfolio of stocks.
 The weight of a stock equals the proportion of the portfolio invested in the stock.
 Dividends are usually not included in the calculation. Therefore, the index only tracks the capital gains and losses
from investing in the portfolio.
Notes:
 If the portfolio of stocks remains fixed, the weights assigned to individual stocks will vary over time. As prices
change, those that rise more sharply than others receive more weight.
 Some indices are constructed from a hypothetical portfolio consisting of one of each of a number of stocks. The
weights assigned to the stocks are proportional to their market prices, with adjustments being made when there are
stock splits.
 Other indices are constructed so that weights are proportional to market capitalization (stock price × number of
shares outstanding).
1. Futures Prices of Stock Indices
A stock index can be regards as the price of an investment asset that pays dividends. The asset is the portfolio of stocks
underlying the index and the dividends generated by the asset are those paid by stocks within the portfolio.
Since there are many stocks providing dividends at different times, an index can be considered as an asset providing a
continuous dividend yield. If q is the dividend yield rate, the futures price can be determined as F0  S0  e( r q )T .

Example: Consider a 3 month futures contract on the S&P 500.

Assume stocks underlying an index produce a dividend yield of 1% per annum, that the current value of the index is
800, and that the continuously compounded risk–free interest rate is 6% per annum.
Thus, r = 0.06, S0 = 800, T = 0.25, and q = 0.01.
Hence, the futures price is given by F0 = 800e(0.06 – 0.01) × 0.25 = $810.06
Note: In practice, the dividend yield underlying an index varies weekly throughout the year.
2. Index Arbitrage can be described by the following strategies when acted upon:
If F0  S0  e( r  q )T , profits are made by buying the stocks underlying the index and shorting futures contracts.
When the above holds true, index arbitrage is often done by a pension fund that owns an indexed portfolio of stocks.
If F0  S0  e( r  q )T , profits are made by shorting the stocks underlying the index and taking a long position in
futures contracts.
When the above holds true, it is often done by a corporation holding short–term money market investments.
Note: index arbitrage is often triggered through program trading.
5.10 Forward and Future Contracts on Currencies

To better understand forward and futures foreign currency contracts, the following notation is introduced:
Symbol Definition
S0 the current spot price in dollars of one unit of foreign currency
F0 the forward or futures price in dollars of one unit of the foreign currency
rf the foreign risk–free interest rate for money invested for time T
r the domestic risk–free rate for money invested for time T
Notes:
i. The underlying asset is a certain number of units of the foreign currency.
ii. A spot or forward exchange rate is normally quoted as the number of units of the currency that are equivalent to
one dollar. This is not the case for the British pound, euro, Australian dollar, and New Zealand dollar.
iii. A foreign currency has the property that the holder of the currency can earn interest at the risk–free interest rate
prevailing in the foreign country. For example, the holder can invest the currency in a foreign–denominated bond.
The interest rate parity relationship
( r  r )T
The relationship between F0 and S0, under interest rate parity is F0  S0 e f ; given the following:
The two–year interest rates in Australia and the United States are rf = 5% and r = 7%.
The spot exchange rate between the Australian dollar (AUD) and the U.S. dollar (USD) is 0.6200 USD per AUD.
Compute the two–year forward exchange rate.
The two–year forward exchange rate equals 0.6200e(0.07 – 0.05) × 2 = 0.6453
Arbitrage opportunities:
Scenario 1: Suppose the 2–year forward exchange rate is 0.6300 (less than 0.6453). An arbitrageur can:
1. Borrow 1,000 AUD at 5% per annum for two years, convert to 620 USD, and invest the USD at 7%
The 620 USD invested at 7% grows to 620e0.07 × 2 = 713.17 USD in two years.
The principal and interest on the 1,000 AUD that are borrowed is 1,000e0.05 × 2 = 1,105.17
2. Enter into a forward contract to buy 1,105.17 AUD for 1,105.17 × 0.63 = 696.26 USD.
696.26 USD are used to purchase 1,105.17 AUD under the terms of the forward contract.
Engaging in the above strategy will produce a riskless profit of 713.17 – 696.26 = 16.91 USD
Scenario 2: Suppose the 2–year forward exchange rate is 0.6600 (greater than 0.6453)
An arbitrageur can:
1. Borrow 1,000 USD at 7% pa for two years, convert to 1,000/0.6200 = 1,612.90 AUD, invest the AUD at 5%.
The 1,612.90 AUD that are invested at 5% grows to 1, 612.90e 005 × 2 = 1,782.53 AUD in two years.
The amount needed to pay off the USD borrowings is 1,000e 0.07 × 2 = 1,150.27 USD.
2. Enter into a forward contract to sell 1,782.53 AUD for 1,782.53 × 0.66 = 1,176.47 USD.
Engaging in the above strategy will produce a riskless profit of 1,176.47 – 1,150.27 = 26.20 USD.
A Foreign Currency as an Asset Providing a Known Yield
A foreign currency can be regarded as an investment asset paying a known yield. The yield is the risk–free rate of
interest in the foreign currency. Therefore, the following equations are similar:
( r  r )T
i. Under interest rate parity: F0  S0 e f
ii. The forward price for an investment asset providing a continuous dividend yield F0 = S0 e(r – q)T

Note: the value of interest paid in a foreign currency depends on the value of the foreign currency.
Example: Assume the interest rate on British pounds is 5% per annum.
To a US investor the British pound provides an income equal to 5% of the value of the British pound per annum. In
other words it is an asset that provides a yield of 5% per annum.
5.11 Futures on Commodities

Prices on commodity futures contracts are dependent upon whether the commodities are investment assets (e.g. gold
and silver) or consumption assets (e.g. oil).
Income and Storage Costs
If an investment asset has no storage costs and produces no income, the relationship between its futures price and its
spot price is F0 = S0 × erT
Storage costs can be regarded as negative income.
1. Let U represent the present value of all storage costs incurred during the life of a futures contract, then
F0 = (S0 + U) × erT
Example: Consider a 1–year futures contract on gold. Suppose gold storage costs per year are $2 per ounce, with
the payment being made at the end of the year. Let r = 0.07; S0 = 450; T = 1.
Compute U: U = 2 e–0.07×1 = 1.865
Compute the futures price: F0 = (450 + 1.865) × e–0.07×1 = 484.63
2. If storage costs incurred are proportional to the price of the commodity, they can be regarded as generating a
negative dividend yield. In this case, F0 = S0 e (r + u)T, where u is the storage costs per annum as a proportion of the
spot price.
Consumption Commodities
Commodities that are consumption assets rather than investment assets usually provide no income, but can be
subject to significant storage costs.
Arbitrage strategies used to determine futures prices from spot prices are described below.
Consider the equation F0 > (S0 + U)erT
To take advantage of this opportunity, an arbitrageur can:
1. Borrow S0 + U at the risk–free rate and use it to purchase one unit of the commodity and pay storage costs.
2. Short a forward contract on one unit of the commodity.
Treating a futures contract as a forward contract, this leads to a profit of F0 – (S0 + U)erT at time T. However, as
arbitrageurs continue to implement this strategy, there will be a tendency for S0 to increase and F0 to decrease
until the inequality vanishes. We conclude that the inequality cannot hold for a significant length of time.
Consider F0 < (S0 + U)erT
When the commodity is an investment asset, arbitragers will find it profitable to do the following:
1. Sell the commodity, save the storage costs, and invest the proceeds at the risk–free interest rate.
2. Take a long position in a forward contract.
This results in a riskless profit at maturity of (S0 + U)erT – F0
Again, this inequality cannot hold for any significant length of time.
This argument cannot be used for commodities that are not significantly held for investment.
 Individuals and companies who keep such a commodity in their inventory do so because of its consumption
value–not because of its value as an investment.
 They are reluctant to sell the commodity and buy forward contracts, because forward contracts cannot be
consumed.
Thus, all that can be asserted for a consumption commodity is F0  (S0 + U)erT
If storage costs are expressed as a proportion it of the spot price, the equivalent result is F0  S0 e (r + u)T.
Convenience Yields
When F0  (S0 + U)×erT, users of the commodity may feel that there are benefits from ownership that are not obtained
by the holder of a futures contract.
i. The benefits may include profiting from temporary local shortages or keeping a production process running.
ii. The benefits are sometimes referred to as the convenience yield provided by the product:
1. If storage costs are known and have a present value, U, the convenience yield, y, is F0 e yT = (S0 + U)×erT
2. If the storage costs per unit are a constant proportion, u, of the spot price, y is F0 e yT = S0×e(r + u)T
or F0 = S0×e(r + u - y)T .

Other comments about the convenience yield.

 The convenience yield must be zero for investment assets; otherwise, there are arbitrage opportunities.
 Figure 2.2 of Chapter 2 shows that the futures prices of crude oil tended to decrease as the time to maturity of the
contract increased on February 4, 2004. This pattern suggests that the convenience yield, y, is greater than r + u for
oil on this date.
The convenience yield reflects the market's expectations concerning the future availability of the commodity.
 The greater the possibility that shortages will occur, the higher the convenience yield.
 If users of the commodity have high inventories, there is little chance of shortages in the near future and the
convenience yield tends to be low.
 On the other hand, low inventories tend to lead to high convenience yields.
5.12 Cost of Carry

Futures prices and spot prices can be summarized in terms of the cost of carry.
The cost of carry measures the storage cost plus the interest that is paid to finance the asset less the income earned on
the asset. The cost of carry varies depending upon the asset under consideration:
 For a non–dividend paying stock, the cost of carry is r (no storage costs and no income is earned).
 For a stock index, it is (r – q) since income is earned at rate q on the asset.
 For a currency, it is r – rf
 For a commodity that provides income at rate q and requires storage costs at rate u, it is r – q + u
Define the cost of carry as c.
A. For an investment asset, the futures price is F0 = S0 × ecT
B. For a consumption asset, the futures price is F0 = S0 × e(c – y)T, where y is the convenience yield.
5.13 Delivery Options

The fact that a futures contract often allows the party with the short position to choose to deliver at any time gives rise
to problem in determining futures prices. Should the maturity of the futures contract be assumed to be the beginning,
middle, or end of the delivery period?
If the futures price is an increasing function of the time to maturity, c > y, so the benefits from holding the asset
(including convenience yield and net of storage costs) are less than the risk–free rate. In this case,
i. it usually optimal for the party with the short position to deliver as early as possible, because the interest earned on
the cash received outweighs the benefits of holding the asset.
ii. futures prices should be calculated on the basis that delivery will take place at the beginning of the delivery period.
If futures prices are decreasing as time to maturity increases (c < y), the reverse is true, and it is usually optimal for the
party with the short position to deliver as late as possible.
5.14 Futures Prices and the Expected Future Spot Price

Is the futures price of an asset equal to its expected future spot price?
1. Arguments according to Keynes and Hicks.
a. If hedgers tend to hold short positions and speculators tend to hold long positions, the futures price of an asset
will be below the expected spot price. This is because:
 Speculators require compensation for the risks they are bearing.
 Speculators will trade only if they can expect to make money on average.
 Hedgers will lose money on average, but they are likely to be prepared to accept this because the futures
contract reduces their risks.
b. If hedgers tend to hold long positions while speculators hold short positions, the futures price will be above the
expected spot price for a similar reason.
2. Arguments according to the Risk and Return concept
Another way to answer the above question is to consider the relationship between risk and expected return in the
economy. The higher the risk of an investment, the higher the expected return demanded by an investor.
The Risk in a Futures Position
Consider a speculator who takes a long futures position. Assume:
 that she puts the present value of the futures price into a risk–free investment, uses its accumulated value to buy the
asset on the delivery date, and then sells the asset immediately at its market price.
 that the futures contract can be treated as a forward contract.

The cash flows to the speculator are:

Time 0: – F0e–rT
Time T : +ST, where ST is the price of the asset at time T.
The present value of this investment is –F0e–rT + E(ST)e–kT, where k is the discount rate (i.e., the expected return
required by investors on the investment) and E denotes the expected value.
The Risk in a Futures Position (continued):
Assuming all investment opportunities in securities markets have zero net present value,
–F0e–rT + E(ST)e–kT = 0 and therefore F0 = E(ST)e(r – k)T
If the returns from this asset are uncorrelated with the stock market, the correct discount rate to use is the risk–free rate
r, so we should set k = r. Thus, F0 = E(ST)
This shows that the futures price is an unbiased estimate of the expected future spot price when the return from the
underlying asset is uncorrelated with the stock market.
The value of k depends on the systematic risk of the investment.
ST correlation to the market Relationship between k and r Relationship between F0 and E(ST)
Uncorrelated k=r F0 = E(ST)
Positively correlated k>r F0 < E(ST)
Negatively correlated k<r F0 > E(ST)
If the return from the asset is positively correlated with the stock market,
 k > r and F0 < E(ST)
 thus, when the asset underlying the futures contract has positive systematic risk, expect the futures price to
understate the expected future spot price.
 An example of an asset that has positive systematic risk is a stock index.
The expected return of investors on the stocks underlying an index is generally more than the risk–free rate, r. The
dividends provide a return of q.
The expected increase in the index must therefore be more than r – q.
This is consistent with the prediction that the futures price understates the expected future stock price for a stock index.
When the return from the asset is negatively correlated with the stock market:
 k < r and F0 > E(ST).
 thus, when the asset underlying the futures contract has positive systematic risk, expect the futures price to
overstate the expected future spot price.
Normal Backwardation and Contango
 When the futures price is below the expected future spot price, the situation is known as normal backwardation.
 When the futures price is above the expected future spot price, the situation is known as contango.


Chapter 6: Interest Rate Futures
Section Description
6.1 Day Count Conventions
6.2 Quotations for Treasury Bonds
6.1 Day Count Conventions

This section examines what day count conventions are used when interest rates are quoted. A generic formula to
determine the interest earned between two dates is:
Number of days between dates
 Interest earned in reference period
Number of days in reference period
Day count conventions in the US: (1) Actual/actual (in period); (2) 30/360; (3) Actual/360
The following information will be used to determine the interest earned between two periods.
Given: Bond principal of $100, having coupon payment on 3/1 and 9/1, with a coupon rate of 8%.
1. Actual/actual (in period).
# days from 3/1  7 / 3 124
a. Interest earned between 3/1 and 7/3:  Coupon   $4  $2.6957
# days from 3/1  9 /1 184
b. Actual/actual day count convention is used for U.S. Treasury bonds.
2. 30/360.
4  30  2 122
a. Compute the interest earned between 3/1 and 7/3:  Coupon   $4  $2.7111
6  30 180
b. 30/360 day count convention is used for U.S. corporate and municipal bonds.
3. Actual/360. Actual/360 is used for U.S. Treasury bills and other money market securities. This implies that:
a. The interest earned in 90 days is exactly one–fourth of the quoted rate.
b. The interest earned in a whole year is 365/360 times the quoted rate.
6.2 Quotations for Treasury Bonds

Price quotes and cash paid for an interest–bearing instrument are typically not the same. In general,
Cash price = Quoted price + Accrued interest since last coupon date
The quoted price (a.k.a. the clean price), is not the same as the cash price (a.k.a. the dirty price).
Examples of the cash price and quoted price for Treasury bonds and Treasury bills in the U.S are shown below.
1. Treasury Bonds (in the U.S.)
a. Treasury bond prices are quoted in dollars and thirty–seconds of a dollar.
b. The quoted price is for a bond with a face value of $100. Thus, a quote of 90–05 indicates a cash price of
$90,156.25 for a bond with a face value of $100,000.
Example:
Determine the cash price on 3/5/04 of a $100,000 bond with 11% coupon bond maturing on 7/10/09, with a quoted
price of 95–16 or $95.50. The most recent coupon payment was made on 1/10/04.
Step 1: Determine the accrued interest since the last coupon date. The last coupon payment was for $5.50 on January
10th. The number of days between 1/10/04 and 3/5/04 is 54. The next coupon payment is on 7/10/04.
54
Since actual/actual in period is used for Treasury bonds, accrued interest equals  $5.5  $1.64. .
181
Step 2: Using the formula Cash price = Quoted price + Accrued interest since last coupon date, compute the cash price.
Cash price per $100 face value for the bond $95.5 + $1.64 = $97.14.
Cash price of a $100,000 bond equals 1,000 × ($97.14) = $97,140.
2. Treasury Bills
a. As mentioned, the actual/360 day count convention is used for Treasury bills.
b. There is a difference between the cash price and quoted price for a Treasury bill.
i. If P is the cash price of a Treasury bill that has a face value of $100 and n days to maturity, the cash price is
360
P (100  y ) , where P is the cash price, Y is the quoted price, and n is the remaining life of the Treasury
n
bill measured in calendar days.
ii. The prices of money market instruments are sometimes quoted using a discount rate.

Chapter 6: Interest Rate Futures
This is the interest earned as a percentage of the final face value rather than as a percentage of the initial price paid for
the instrument. For example, if the price of a 91–day Treasury bill is quoted as 8, the annualized rate of interest earned
is 8% of the face value.
Example: Given a 91–day Treasury bill, quoted as 8, having a face value of $100.
1. Compute the amount of interest earned over the 91–day life.
Interest of $2.0222 (= $100 × 0.08 × 91/360) is earned over the 91–day life.
2. Compute the rate of interest earned over the 91–day life.
Rate of interest of 2.0222/(100 – 2.0222) = 2.064% for the 91–day period.

Chapter 7: Swaps
Section Description
7.0 Introduction
7.1 Mechanics of Interest Rate Swaps
7.2 Day Count Issues
7.3 Confirmations
7.4 The Comparative–Advantage Argument
7.5 The Nature of Swaps
7.6 Determining Libor/Swap Zero Rates
7.7 Valuation of Interest Rate Swaps
7.8 Currency Swaps
7.9 Valuation of Currency Swaps
7.10 Credit Risk
7.11 Other Types of Swaps
7.0 Introduction
A swap is an agreement by two companies to exchange cash flows in the future.
A forward contract may be viewed as a swap. For example:
On 3/1/02 a company enters into a 1 year forward contract to buy 100 ounces of gold for $300 per ounce.
This is equivalent to a swap where the company agrees that on 3/1/03, it will pay $30,000 and receive 100S, where S
is the market price of one ounce of gold on that date.
Unlike forwards, swaps typically lead to multiple cash flow exchanges taking place on several future dates.
7.1 Mechanics of Interest Rate Swaps

The “plain vanilla” interest rate swap is the most common type.
a. A company agrees to pay cash flows equal to a fixed rate of interest on a notional (notional meaning that the
principal itself will not be exchanged) principal for a number years.
b. The company will receive in return interest at a floating rate on the same notional principal for the same period of
time.
The London Interbank Offer Rate (LIBOR) is the floating rate used in many interest rate swap agreements.
The “plain vanilla” interest rate swap is best understood by example.
Consider a 3–year swap initiated on 3/5/03 between Microsoft and Intel.
Microsoft agrees to pay to Intel 5% (with semiannual compounding) on a notional principal of $100 million.
Intel agrees to pay Microsoft the 6–month LIBOR rate on the same notional principal.
The agreement specifies that payments are to be exchanged every six months.
Intel Pays interest at LIBOR Microsoft
Pays interest at 5%
The cash flows involved in the Swap are shown on the next page.
Cash flows (million of dollars) to Microsoft in a $100 million three–year interest rate swap when a fixed rate of 5% is
paid and LIBOR is received
6–month LIBOR Floating cash flow Fixed cash flow Net

Date rate (%) Received Paid cash flow
(1) (2) (3) = (1) – (2)
3/5/03 4.20
9/5/03 4.80 0.5 × 0.042 × $100 = +$2.1 –2.50 –0.40
3/5/04 5.30 +2.40 –2.50 –0.10
9/5/04 5.50 +2.65 –2.50 +0.15
3/5/05 5.60 +2.75 –2.50 +0.25
9/5/05 5.90 +2.80 –2.50 +0.30
3/5/06 6.40 +2.95 –2.50 +0.45

Chapter 7: Swaps
Notes: On the first payment, Intel pays Microsoft interest at the six–month LIBOR rate prevailing six months prior to
September 5, 2003. Intel pays Microsoft 0.5 × 0.042 × $100 = $2.1 million.
Interest rate swaps are structured so that one side remits the difference between the two payments to the other side
(e.g. on 9/5/03, Microsoft pays Intel $0.4 million (= $2.5M – $2.1M).
Using Swaps to Transform a Liability
For Microsoft, the above swap could be used to transform a floating–rate loan into a fixed–rate loan.
Suppose Microsoft borrows $100 million at LIBOR plus 10 basis points.
For Intel, the swap could have the effect of transforming a fixed–rate loan into a floating–rate loan.
Suppose Intel has a three–year $100 million loan outstanding on which it pays 5.2%.
The cash flows to the two companies are described verbally and shown graphically below:
Microsoft Cash Flows Intel Cash Flows
1. It pays LIBOR plus 0.1% to its outside lenders. 1. It pays 5.2% to its outside lenders.
2. It receives LIBOR under the terms of the swap. 2. It pays LIBOR under the terms of the swap.
3. It pays 5% under the terms of the swap. 3. It receives 5% under the terms of the swap.
Pays 5.2% Intel Pays interest at LIBOR Microsoft
Pays interest at 5% Pays LIBOR + 0.10%

For Microsoft, the swap could have the effect of transforming borrowings at a floating rate of LIBOR plus 10 basis
points into borrowings at a fixed rate of 5.1% (5% + LIBOR plus 0.1% – LIBOR).
For Intel, the swap could have the effect of transforming borrowings at a fixed rate of 5.2% into borrowings at a
floating rate of LIBOR plus 20 basis points. (5.2% + LIBOR – 5%).
Notes:
1. The same procedure can be used when using a swap to transform an asset.
2. If an intermediary is involved, the swap is structured so that the financial institution earns about 3 or 4 basis points
(0.03 to 0.04%) on a pair of offsetting transactions.
3. Since it is unlikely that two companies will want to take opposite positions in exactly the same swap, the financial
institutions act as market makers for swaps.
Market Makers
Large institutions are market makers in the swap market. They quote a bid and an offer for the fixed rate they will
exchange for LIBOR.
The bid is the fixed rate in a contract where the market maker will pay fixed and receive LIBOR.
The offer is the fixed rate in a swap where the market maker will receive fixed and pay LIBOR.
The average of the bid and offer fixed rates is known as the swap rate.
Bid and offer fixed rates in the swap market and swap rates (% per annum);
Maturity (years) Bid (%) Offer (%) Swap rate (%) Zero Rates
(1) (2) [(1) + (2)] ÷ 2
6 month 5.50%
1 year 5.75%
1.5 years 5.90%
2 6.03 6.06 6.045
3 6.21 6.24 6.225
4 6.35 6.39 6.370
5 6.47 6.51 6.490
7 6.65 6.68 6.665
10 6.83 6.87 6.850
A swap can be characterized as the difference between a fixed–rate bond and a floating–rate bond. Define:
Bfix : Value of fixed–rate bond underlying the swap
Bfl : Value of floating–rate bond underlying the swap
Since the value of a swap is generally zero, Bfl = Bfix.

Chapter 7: Swaps
7.2 Day Count Issues

Day count conventions affect payments on a swap.
Example: The 6–month LIBOR is quoted on an actual/360 basis because it is a money market rate.
 The first floating payment is based on a LIBOR rate of 4.2%, and is therefore $2.10 million.
 Because there are 184 days between March 5, 2004, and September 5, 2004, it should be 100×0.042×184/360 =
$2.1467 million
A LIBOR–based floating–rate cash flow on a swap payment date is calculated as L×R×n/360, where
L is the principal, R is the relevant LIBOR rate, and n is the number of days since the last payment date.
The fixed rate paid in a swap transaction is quoted with a particular day count basis being specified.
Thus, fixed payments may not be exactly equal on each payment date.
 The fixed rate is usually quoted as actual/365 or 30/360. It is not therefore directly comparable with LIBOR
because it applies to a full year.
 To make the rates comparable, either the 6–month LIBOR rate must be multiplied by 365/360 or the fixed rate
must be multiplied by 360/365.
7.3 Confirmations
A confirmation is the legal agreement underlying a swap and is signed by representatives of the two parties.
The confirmation specifies that the following business day convention is to be used and that the US calendar
determines which days are business days and which days are holidays.
 If a payment date falls on a weekend or a US holiday, the payment is made on the next business day.
 September 5, 2004, is a Sunday.
 The first exchange of payments in a swap is therefore on Monday September 6, 2004.

7.4 The Comparative–Advantage Argument
Consider the use of an interest rate swap to transform a liability.
Some companies have a comparative advantage when borrowing in fixed–rate markets, whereas others have a
comparative advantage in floating–rate markets. Thus, a company may borrow fixed when it wants floating, or borrow
floating when it wants fixed.
The swap is used to transform a fixed–rate loan into a floating–rate loan, and vice versa.
Example:
AAA-Corp and BBB-Corp (named after their credit rating) wish to borrow $10 million for 5 years
BBB-Corp wants to borrow at a fixed rate of interest; AAA-Corp wants to borrow at a floating rate.
Borrowing rates that provide a basis for the comparative–advantage argument
Fixed Floating
AAA-Corp 10.0% 6–month LIBOR + 0.3%
BBB-Corp 11.2% 6–month LIBOR + 1.0%
Notes: BBB-Corp pays 1.2% more than AAA-Corp in fixed–rate markets and only 0.7% more than AAA-Corp in
floating–rate markets.
BBB-Corp appears to have a comparative advantage in floating–rate markets, whereas AAA-Corp appears to have a
comparative advantage in fixed–rate markets.
They enter into a swap agreement to ensure that AAA-Corp ends up with floating–rate funds and BBB-Corp ends up
with fixed–rate funds.
Terms of the swap: AAA-Corp agrees to pay BBB-Corp interest at six–month LIBOR on $10 million.
In return, BBB-Corp agrees to pay AAA-Corp interest at a fixed rate of 9.95% per annum on $10 million.
Pays 10% AAA Corp Pays interest at LIBOR BBB Corp
Pays interest at 9.95% Pays LIBOR + 1%
AAA Corp Cash Flows BBB Cash Flows

1. It pays 10% per annum to outside lenders. 1. It pays LIBOR + 1.0% to outside lenders.
2. It receives 9.95% per annum from BBB-Corp. 2. It receives LIBOR from AAA-Corp.
3. It pays LIBOR to BBB-Corp. 3. It pays 9.95% per annum to AAA-Corp.
The net effect is that AAA-Corp pays LIBOR plus 0.05% (10% + LIBOR – 9.95%) per annum.

Chapter 7: Swaps
This is 0.25% (LIBOR + 0.3% – LIBOR – 0.05%) per annum less than it would pay in floating–rate markets.
The net effect is that BBB pays 10.95% (LIBOR + 1% + 9.95% – LIBOR) per annum.
This is 0.25% (11.2% – 10.95%) per annum less than it would pay in floating–rate markets.
Thus, the position of both AAA-Corp and BBB-Corp improves by 0.25% per annum.
The total gain is therefore 0.5% per annum, and the total apparent gain from this type of interest rate swap
arrangement is always a – b, where a is the difference between the interest rates facing the two companies in fixed–
rate markets, and b is the difference between the interest rates facing the two companies in floating–rate markets. In
this case, a = 1.2% and b = 0.70%.
Criticism of the Comparative–Advantage Argument
The spread differentials exist due to the nature of the contracts involved:
a. The 10.0% and 11.2% rates available to AAA-Corp and BBB-Corp in fixed–rate markets are 5 year rates (e.g., the
rates at which the companies can issue five–year fixed–rate bonds).
b. The LIBOR + 0.3% and LIBOR + 1.0% rates available to AAA-Corp and BBB-Corp in floating–rate markets are
6–month rates.
c. If the creditworthiness of AAA-Corp or BBB-Corp declines, the lender has the option of increasing to spread over
LIBOR that is charged. The providers of fixed–rate financing do not have the option to change the terms of to loan
in this way.
d. Further, the spreads between the rates offered to AAA-Corp and BBB-Corp are a reflection of the extent to which
BBB-Corp is more likely to default than AAA-Corp.
e. Finally, Because BBB-Corp’s spread over six–month LIBOR is more likely to rise than to fall, BBB-Corp’s
expected average borrowing rate when it enters into the swap is greater than 10.95%.
7.5 The Nature of Swaps

The nature of swap rates and the relationship between swap and LIBOR markets:
 LIBOR is the rate of interest at which AA–rated banks borrow for periods between 1 and 12 months from other
banks.
 A swap rate is the average of:
(a) the fixed rate that a swap market maker is prepared to pay in exchange for receiving LIBOR (its bid rate) and
(b) the fixed rate that it is prepared to receive in return for paying LIBOR (its offer rate).
Swap rates and LIBOR rates are not risk–free lending rates, however they are close to risk–free.
A financial institution can earn the 5–year swap rate on a certain principal by doing the following:
1. Lend the principal for the fast 6 months to a AA borrower and then re-lend it for successive 6 month periods to
other AA borrowers; and
2. Enter into a swap to exchange the LIBOR income for the 5–year swap rate.
Thus, the 5–year swap rate is an interest rate with a credit risk corresponding to the situation where 10 consecutive 6–
month LIBOR loans to AA companies are made.
Why Swap rates are less than AA borrowing rates.
It is more worthwhile to lend money for successive 6–month periods to borrowers who are always AA at the
beginning of the periods than to lend it to one borrower for the whole 5 years.
7.6 Determining Libor/Swap Zero Rates

Derivative traders use LIBOR rates as the risk–free rates when valuing derivatives (see Section 4.1).
 One problem with LIBOR rates is that direct observations are possible only for maturities out to 12 months.
 One way of extending the LIBOR zero curve beyond 12 months is to use Eurodollar futures. Eurodollar futures
are used to produce a LIBOR zero curve out to 2 years–and sometimes out to as far as 5 years (see Section 6.4).
 Traders then use swap rates to extend the LIBOR zero curve further.
 The resulting zero curve is sometimes referred to as the LIBOR zero curve and sometimes as the swap zero curve
(a.k.a. the LIBOR/swap zero curve)
How swap rates are used in the determination of the LIBOR/swap zero curve:
 The value of a newly issued floating–rate bond that pays 6–month LIBOR is always equal to its principal value (or
par value) when the LIBOR/swap zero curve is used for discounting.
The bond provides a rate of interest of LIBOR, and LIBOR is the discount rate. The interest on the bond exactly
matches the discount rate, and as a result the bond is fairly priced at par.
 For a newly issued swap where the fixed rate equals the swap rate, Bfix = Bfl.
 Since Bfix and Bfl equal the swap’s notional principal, swap rates define a set of par yield bonds.

Chapter 7: Swaps
For example, from the swap rates in section 7.1 can deduce that the 2–year LIBOR/swap par yield is 6.045%, the 3–
year LIBOR/swap par yield is 6.225%, etc.
The usual method for determining the LIBOR/swap zero curve is the bootstrap method which we used to determine
the Treasury zero curve (see section 4.5).
 LIBOR rates define the zero curve out to 1 year.
 Swap rates define par yield bonds that are used to determine longer–term rates.
Example:
Suppose that the 6–month, 12–month, and 18–month LIBOR/swap zero rates have been determined as 4%, 4.5%, and
4.8% with continuous compounding and that the 2–year swap rate (for a swap where payments are made
semiannually) is 5%.
This 5% swap rate means that a bond with a principal of $100 and a semiannual coupon of 5% per annum sells for par.
Compute R is the 2–year zero rate: 2.5e–0.04×0.5 + 2.5e –0.045×1.0 + 2.5e–0.048×1.5 + 102.5e2R = 100
Solving, R = 4.953%. (Note that this calculation is simplified in that it does not take the swap’s day count conventions
and holiday calendars into account. See Section 7.2).
7.7 Valuation of Interest Rate Swaps

An interest rate swap is worth zero when first initiated; after some time, its value may become positive or negative (the
example on the next page shows that it becomes negative).
Two valuation approaches for interest rate swaps:
1. a long position in one bond combined with a short position in another bond or
2. a portfolio of forward rate agreements.
1. Valuation in Terms of Bond Prices
Principal payments are not exchanged in an interest rate swap.
In section 7.1, we saw that principal payments are both received and paid at the end of the swap without changing its
value.
From the point of view of the floating–rate payer, a swap can be regarded as a long position in a fixed rate bond and a
short position in a floating–rate bond, so that: Vswap = Bfix – Bfl, where
Vswap is the value of the swap, Bfl is the value of the floating–rate bond (corresponding to payments that are made), and
Bfix is the value of the fixed–rate bond (corresponding to payments that are received).
From the point of view of the fixed–rate payer, a swap is a long position in a floating–rate bond and a short position in
a fixed–rate bond, so that the value of the swap is: Vswap = Bfl – Bfix
 The value of the fixed rate bond, Bfix can be determined as described in Section 4.4.
 To value the floating–rate bond, note that the bond is worth the notional principal immediately after an interest
payment. This is because at this time the bond is a “fair deal” where the borrower pays LIBOR for each
subsequent accrual period.
Let: L = the notional principal; t* = the next exchange of payments is at time; k* = the floating payment that will be
made at time t* (which was determined at the last payment date). Immediately after the payment Bfl = L
It follows that immediately before the payment Bfl = L + k*.
The floating–rate bond can therefore be regarded as an instrument providing a single cash flow of L + k* at time t*.
The value of the floating–rate bond today is (L + k*)e–r*t* where r* is the LIBOR/swap zero rate for a maturity of t*.
Example
A financial institution pays 6 month LIBOR and receives 8% per annum (with semiannual compounding) on a swap
with a notional principal of $100 million. The swap has a remaining life of 15 months, with remaining payment dates
in 3, 9, and 15 months.
The LIBOR rates with continuous compounding for 3–month, 9–month, and 15–month maturities are 10%, 10.5%,
and 11 %, respectively.
The 6–month LIBOR rate at the last payment date was 10.2% (with semiannual compounding).
a. Compute the value of the fixed–rate bond: First, compute k: k = (0.08)×(100M)×(0.50) = 4M
Next, compute Bfix: Bfix = 4e–0.1 × 3/12 + 4e–0.105 × 9/12 + 104e–0.11 × 15/12 = $98.24 million
b. Compute the value of the floating rate bond: First, compute k*: k* = (0.102)×(100M) × (0.50) = 5.1
Next, compute Bfl: Bfl = 5.1e–0.1×3/12 + 100e–0.1×3/12 = $102.51 million
c. Compute the value of the swap: 98.24 – 102.51 = –$4.27 million
Note: If the bank had been in the opposite position of paying fixed and receiving floating, the value of the swap would
be +$4.27 million.

Chapter 7: Swaps
2. Valuation in Terms of Forward Rate Agreements
The procedure used to determine the value of a swap in terms of FRAs is as follows:
1. Calculate forward rates for each of the LIBOR rates that will determine swap cash flows.
2. Calculate swap cash flows assuming that the LIBOR rates will equal the forward rates.
3. Set the swap value equal to the present value of these cash flows.
Step 1: Calculate forward rates for each of the LIBOR rates that will determine swap cash flows.
0.105  0.75  0.10  0.25
The forward rate between 3 and 9 months is  0.1075 , or 10.75% with continuous
0.5
compounding, and 11.044% 2×(e0.1075/2 –1) = 0.11044 with semiannual compounding.
0.11 1.25  0.105  0.75
The forward rate between 9 and 15 months is  0.1175 , or 11.75% with continuous
0.5
compounding and 12.102% 2×(e0.1175/2 –1) = 0.12102 with semiannual compounding.
Step 2: Calculate swap cash flows assuming that the LIBOR rates will equal the forward rates.
Based on the previous example, we know that the cash flows that will be exchanged in 3 months have already been
determined: A rate of 8% will be exchanged for 10.2%.
a. The value of the exchange to the financial institution is 0.5×100×(0.08 – 0.102)e–0.1 × 3/12 = –1.07
b. Using the forward rate between 3 and 9 months, compute the value of the FRA corresponding to the exchange in 9
months: 0.5×100×(0.08 – 0.11044)e–0.105 × 9/12 = –1.41
c. Using the forward rate between 9 and 15 months, compute the value of the FRA corresponding to the exchange in
15 months: 0.5×100×(0.08 – 0.12102)e–0.11 × 15/12 = –1.79
Step 3: Set the swap value equal to the present value of these cash flows.
The total value of the swap is –1.07 – 1.41 – 1.79 = –4.27 or –$4.27 million. This agrees to the amount computed
based on bond prices.
7.8 Currency Swaps

A currency swap involves exchanging principal and interest payments in one currency for principal and interest
payments in another currency.
The principal amounts are usually exchanged at the beginning and at the end of the life of the swap and are made
equivalent by using the exchange rate at the swap’s initiation.
A currency swap is best understood by example. The example below is a fixed–for–fixed currency swap.
Consider a 5–year swap initiated on 2/1/03 between IBM and British Petroleum.
Suppose IBM pays a fixed rate of interest of 11% in sterling and receives a fixed rate of interest of 8% in dollars from
British Petroleum.
Interest rate payments are made once a year and the principal amounts are $15 million and £10 million.
At the outset of the swap, IBM pays $15 million and receives £10 million. Each year, IBM receives $1.20 million and
pays £1.10 million. At the end of the swap, it pays a principal of £10 million and receives a principal of $15 million.
Cash flows to IBM in currency swap
Date Dollar cash flow (millions) Sterling cash flow (millions)
2/1/01 –15.00 +10.00
2/1/02 +1.20 = 8% × $15 –£1.10 = 11% × £10.
2/1/03 +1.20 –1.10
2/1/04 +1.20 –1.10
2/1/05 +1.20 –1.10
2/1/06 +16.20 –11.10
Use of a Currency Swap to Transform Loans and Assets
The swap just considered can be used to transform borrowings in one currency to borrowings in another currency. The
swap can also be used to transform the nature of assets. For example: Rather than investing £10 million pounds in the
U.K to yield 11% per annum, IBM may want to engage in a swap since it may feel that the U.S. dollar will strengthen
against sterling. The swap has the effect of transforming the U.K. investment into a $15 million investment in the U.S.
yielding 8%.
Comparative Advantage: Consider the use of a currency swap to transform borrowing rates.

Chapter 7: Swaps
Example: General Motors (GM) and Qantas Airways (QA) want to reduce their borrowing costs. Suppose that GM
wants to borrow 20 million AUD and QA wants to borrow 12 million USD and that the current exchange rate is 0.6
USD/AUD. Borrowing rates that provide a basis for the comparative–advantage argument under a currency swap:
USD AUD
General Motors 5.0% 12.6%
Qantas Airways 7.0% 13.0%
Notes: QA pays 2% more than GM in USD markets and only 0.4% more than GM in AUD markets.
QA appears to have a comparative advantage in the AUD market, whereas GM appears to have a comparative
advantage in the USD market.
GM and QA each borrow in the market where they have a comparative advantage.
The companies enter into a currency swap to transform GM’s loan into an AUD loan and QA’s loan into a USD loan.
What is the total effect of the swap?
Since the difference between the dollar interest rates is 2%, and the difference between the AUD interest rates is 0.4%,
the interest rate swap can be expected to produce a total gain to all parties of 2.0 – 0.4 = 1.6% per annum.
USD 5% USD 5% Financial

USD 6.3%
GM QA
Institution
AUD 11.9% AUD 13.0% AUD 13.0%
GM borrows USD:
The swap transforms borrowing at a USD interest rate of 5% per annum to an AUD interest rate of 11.9% per annum.
GM is 0.7% (12.6% – 11.9%) per annum better off than it would be if it went directly to AUD market.
QA borrows AUD.
The swap transforms borrowing at an AUD interest rate of 13% per annum to an USD interest rate of 6.3% per annum.
QA is 0.7% (7% – 6.3%) per annum better off than it would be if it went directly to AUD market
The financial institution that coordinates the swap:
Gains 1.3% (6.3% – 5%) per annum on its USD cash flows.
Loses 1.1% (11.9% – 13%) per annum on its AUD flows.
The net gain to the financial institution is 0.2% per annum.
The total gain to all parties is 1.6% (0.7% + 0.7% + 0.2%) per annum.
7.9 Valuation of Currency Swaps

1. Valuing Currency Swaps using Bonds
A currency swap can be valued as the difference in positions taken in two bonds (absent default risk).
If Vswap is defined as the value in U.S. dollars of a swap where dollars are received and a foreign currency is paid,
then Vswap = BD – S0BF, where
BF is the value, measured in the foreign currency, of the foreign–denominated bond underlying the swap
BD is the value of the U.S. dollar bond underlying the swap, and
S0 is the spot exchange rate (expressed as number of units of domestic currency per unit of foreign currency).
The value of a swap can therefore be determined from:
LIBOR rates in the two currencies
The term structure of interest rates in the domestic currency, and
The spot exchange rate.
The value of a swap where the foreign currency is received and dollars are paid is
Vswap = S0BF – BD
Example (value of a swap where the foreign currency is received and dollars are paid)
A financial institution enters into a 3 year currency swap in which it receives 5% per annum in yen and pays 8% per
annum in dollars once a year. Assume the following:
 The term structure of interest rates is flat in both Japan and the United States.
 The Japanese rate is 4% per annum and the U.S. rate is 9% per annum (both with continuous compounding).
 The principals in the two currencies are $10 million and 1,200 million yen.
 The spot exchange rate is 110 yen = $1.
Compute BD, the value of the U.S. dollar bond underlying the swap:
The cash flows paid by the institution on the US bond are 0.08×10M, 0.08×10M, 0.08×10M + 10M
Thus, BD = 0.8e–0.09 × 1 + 0.8e–0.09×2 + 10.8e–0.09×3 = 9.644 million dollars
Compute BF, the value of the foreign–denominated (FD) bond underlying the swap:

Chapter 7: Swaps
The cash flows received under the FD bond are 0.05×1200M, 0.05×1200M, 0.05×1200M + 1200M
Thus, BF = 60e–0.04 × 1 + 60e–0.04×2 + 1,260e–0.04×3 = 1,230.55 million yen
Therefore, the value of the swap in dollars is [1,230.55/110] – 9.644 = 1.543 million.
Note: If the financial institution had been paying yen and receiving dollars, the value of the swap would have been –
$1.543 million.
2. Valuing Currency Swaps using Forward Contracts
A currency swap can be valued as series of forward contracts.
Consider the IBM agreement previously described. On each payment date IBM has agreed to exchange an inflow of
$1.2 million and an outflow of £1.1 million. At the final payment date, it has agreed to exchange a $15 million inflow
for a £10 million outflow. Each of these exchanges represents a forward contract.
Example: Consider the situation in the previous example, where a financial institution enters into a 3 year currency
swap in which it receives 5% per annum in yen and pays 8% per annum in dollars once a year.
Step 1: Compute the one–year, two–year, and three–year forward exchange rates using the general formula between F0
and S0: F0 = S0e(r – rf)T
a. First, compute the current spot rate in dollars per yen: dollar/yen = [1/110] = 0.009091
b. Next, compute the difference between rates of interest paid and received: 0.09 – 0.04 = 0.05
Thus, the one–year, two–year, and three–year forward exchange rates are:
1 year forward exchange rate: 0.009091e0.05 × 1 = 0.009557
Step 2: Compute the values of the forward contracts (in millions of dollars) associated with the principals in the two
currencies and corresponding to the exchange of interest are computed above using the general formula f = (F0 – K)e–rT
a. Recall that the exchange of interest involves receiving 60 million yen and paying $0.8 million.
b. Recall that the risk–free interest rate in dollars is 9% per annum.
Thus, the values of the forward contracts are as follows:
1 year forward contract: (60 × 0.009557 – 0.8)e–0.09 × 1 = –0.2071
Step 3: Compute the value of the final exchange of principal (receiving 1,200 million yen and paying $10 million).
(1,200 × 0.010562 – 10)e–0.09 × 3 = 2.0416
Step 4: Compute the total value of the swap:
The total value of the swap is 2.0416 – 0.1269 – 0.1647 – 0.2071 = $1.543 million (which agrees to the result
produced by valuing currency swaps using positions taken in bonds).
7.10 Credit Risk

Swaps between two companies entail credit risks.
 When a financial institution enters into offsetting contracts with two companies, and neither party defaults, the
financial institution remains fully hedged, since a decline in the value of one contract will always be offset by an
increase in the value of the other contract.
 However, if one party to the contact defaults, the financial institution has to honor its contract with the other party.
A financial institution has credit–risk exposure from a swap only when the value of the swap to the financial institution
is positive. Although all three parties (the financial institution and the two companies) benefit from the net effects of
the swap, the financial institution is liable to lose the whole of the positive value in the swap when the company whose
contract gives rise to the positive portion of the swap defaults.
Potential losses from defaults on swaps:
 Potential losses from defaults on currency swaps are greater than on interest rate swaps. Since currency swaps
involve an exchange of principle amounts in two different currencies, the potential for default is greater than that
under an interest rate swap.
 Potential losses from defaults on a loan are also greater than potential losses from defaults on interest rate swaps
with the same principal since the value of the swap is usually only a small fraction of the value of the loan.
Distinguishing between credit risk and market risk to the financial institution involved in a swap. Credit risk arises
from the possibility of a default by the counterparty when the value of the contract to the financial institution is
positive. Market risk arises from the possibility that market variables (e.g. interest rates and exchange rates) will move
in such a way that the value of a contract to the financial institution becomes negative.
While market risks can be hedged by entering into offsetting contracts, credit risks are less easily hedged.

Chapter 7: Swaps
7.11 Other Types of Swaps

Variations on the Standard Interest Rate Swap
In fixed–for–floating interest rate swaps, LIBOR is the most common reference floating interest rate.
 The tenor (i.e., payment frequency) of LIBOR may be 1 month, 3 months, 6 months or 12 months.
 The tenor on the floating side does not have to match the tenor on the fixed side.
While LIBOR is the most common floating rate, others such as the commercial paper (CP) rate may be used.
Floating–for–floating interest rates swaps may be negotiated. For example,
 The 3–month CP rate plus 10 basis points might be exchanged for 3–month LIBOR with both being applied to the
same principal.
This would allow a company to hedge its exposure when assets and liabilities are subject to different floating rates.
Specific types of varied interest rate swaps:
a. A constant maturity swap (CMS swap) is an agreement to exchange a LIBOR rate for a swap rate.
An example: exchange 6–month LIBOR applied to a certain principal for the 10–year swap rate applied to the
same principal every 6 months for the next 5 years.
b. A constant maturity Treasury swap (CMT swap): exchange a LIBOR rate for a particular Treasury rate (e.g., the
10–year Treasury rate).
c. A compounding swap: interest on one or both sides is compounded forward to the end of the life of the swap
according to pre-agreed rules with only one payment date at the end of the life of the swap.
d. A LIBOR–in arrears swap: the LIBOR rate observed on a payment date is used to calculate the payment on that
date. Note: in a standard deal the LIBOR rate observed on one payment date is used to determine the payment on
the next payment date.
e. An accrual swap: the interest on one side of the swap accrues only if the floating rate is within a range.
The principal in a swap agreement can vary over the term of the swap to meet the needs of a counterparty.
 In an amortizing swap, the principal reduces in a predetermined way.
Designed to correspond to the amortization schedule on a loan.
 In a step–up swap, the principal increases in a predetermined way.
Designed to correspond to draw-downs on a loan agreement.
 Deferred swaps or forward swaps: parties do not begin to exchange interest payments until some future date, are
also sometimes arranged.
 Sometimes swaps are negotiated where the principal to which the fixed payments are applied is different from the
principal to which the floating payments are applied.
Other Currency Swaps
a. Fixed–for–floating currency swap: a floating rate (usually LIBOR) in one currency is exchanged for a fixed rate in
another currency.
This is a combination of a fixed–for–floating interest rate swap and a fixed–for–fixed currency swap and is known
as a cross–currency interest rate swap.
b. Floating–for–floating currency swap: floating rates in both currencies are exchanged.
c. A diff swap or a quanto.
 Sometimes a rate observed in one currency is applied to a principal amount in another currency.
 Example: A 3–month LIBOR observed in the U.S. is exchanged for 3–month LIBOR in Britain, with both
principals being applied to a principal of 10 million British pounds.
Equity Swaps
An equity swap: exchange the total return (dividends and capital gains) realized on an equity index for either a fixed
or a floating rate of interest.
 The total return on the S&P 500 in successive 6–month periods might be exchanged for LIBOR, with both being
applied to the same principal.
 Can be used by portfolio managers to convert returns from a fixed or floating investment to the returns from
investing in an equity index, and vice versa.
Options
Options may be embedded in a swap agreement.
 An extendable swap: one party has the option to extend the life of the swap beyond the specified period.
 A puttable swap: one party has the option to terminate the swap early.
Options on swaps, or swaptions: provide one party with the right at a future time to enter into a swap where a
predetermined fixed rate is exchanged for floating.

Chapter 7: Swaps
Commodity Swaps and Volatility Swaps
Commodity Swaps: a series of forward contracts on a commodity with different maturity dates and the same delivery
prices.
Volatility swaps incorporate a series of time periods.
 At the end of each period, one side pays a pre-agreed volatility, while the other side pays the historical volatility
realized during the period.
 Both volatilities are multiplied by the same notional principal in calculating payments.

Chapter 8: Mechanics of Options Markets
Section Description
8.0 Introduction
8.1 Types of Options
8.2 Option Positions
8.3 Underlying Assets
8.4 Specifications of Stock Options
8.6 Trading
8.7 Commissions
8.8 Margins
8.9 The Options Clearing Corporation
8.10 Regulation
8.11 Taxation
8.12 Warrants, Executive Stock Options, and Convertibles
8.13 Over the Counter Markets
8.0 Introduction
Options are different from forward, futures, and swap contracts:
 An option gives its holder the right to do something (which may include not exercising this right)
 In forward, futures, or swap contracts, the two parties have committed themselves to some action.
General definitions:
 A call option is the right to buy an asset for a certain price.
 A put option is the right to sell an asset for a certain price.
 A European option can be exercised only at the end of its life.
 An American option can be exercised at any time during its life.
Keep in mind that there are four types of option positions: a long position in a call, a long position in a put, a short
position in a call, and a short position in a put.
8.1 Types of Options

Call Options – An example – Consider the following:
 An investor buys a European call option with a strike price of $100 to purchase 100 eBay shares.
 The current stock price is $98, the expiration date of the option is in 4 months, and the price of an option to
purchase one share is $5.
 The initial investment is $500.
Since the option is European, the investor can exercise only on the expiration date.
 If the stock price on the expiration date is less than $100, the investor will choose not to exercise; why buy a share
for $100, when the market value of the share is less than $100?
The investor loses the entire initial investment of $500.
 If the stock price on the expiration date is above $100 on the expiration date, the option will be exercised.
Example: Suppose the stock price is $115. When the option is exercised, the investor buys 100 shares for $100 per
share. If the shares are sold immediately, the investor’s gain is $15 per share, or $1,500, ignoring transactions costs.
The investor’s net profit is $1,000.
Figure 1 shows how the investor’s net profit or loss on an option to purchase one share varies with the final stock price
in the example.
Note: Investors sometimes exercise options and make a loss overall. Consider eBay’s stock price is $102 at the
expiration of the option.
 The investor exercises the option for a gain of 100 × ($102 – $100) = $200
 The investor realizes a loss overall of $300 (Option price × 100 shares) – $200).
 If the investor does not exercise the option, an overall loss of $500 is realized (the amount of the initial
investment).

In general, call options should always be exercised at the expiration date if the stock price is above the strike price.
Figure 1: Profit from buying a European call option on one eBay share. Option price = $5; strike price = $100
Put Options
 The purchaser of a put option is hoping that it will decrease. Consider the following:
 An investor who buys a European put option to sell 100 shares in IBM with a strike price of $70.
 The current stock price is $65, the expiration date of the option is in 3 months, and the price of an option to sell
one share is $7.
 The initial investment is $700 (i.e. $7 × 100).
 The option is European, so it will be exercised only if the stock price is below $70 on the expiration date.
 Suppose that the stock price is $55 on the expiration date.
The investor buys 100 shares for $55 per share and, sells the same shares for $70 to realize a gain of $15 per share,
or $1,500 (transactions costs ignored.)
1,500 minus the $700 initial cost of the option yields the investor a net profit of $800.
 If the final stock price is above $70, the put option expires worthless, and the investor loses $700.
Figure 2: Profit from buying a European put option on one IBM share. Option price = $7; strike price = $70.
Early Exercise
Exchange–traded stock options are generally American rather than European. Thus, investors would not have to wait
until the expiration date before exercising the option.
Later in this chapter, the circumstances under which it is optimal to exercise American options prior to maturity will be
described.
8.2 Option Positions

There are two sides to every option contract.
 On one side is the investor who has taken the long position (i.e., has bought the option).
 On the other side is the investor who has taken a short position (i.e., has sold or written the option).
 The writer of an option receives cash up front, but has potential liabilities later.
 The writer’s profit or loss is the reverse of that for the purchaser of the option.
Figures 3 and 4 show the variation of the profit or loss with the final stock price for writers of the options considered in
Figures 1 and 2.

Figure 3: Profit from writing a European call option on eBay. Option price = $5; strike price = $100.
There are four types of option positions:

1. A long position in a call option
2. A long position in a put option
3. A short position in a call option
4. A short position in a put option
Figure 4: Profit from writing a European put option on IBM. Option price = $7; strike price = $70.
It is useful to characterize European option positions in terms of the terminal value or payoff to the investor at
maturity. The initial cost of the option is then not included in the calculation.
 If K is the strike price and ST is the final price of the underlying asset, the payoff from a long position in a
European call option is max(ST – K, 0)
This reflects the fact that the option will be exercised if ST > K and will not be exercised if ST  K.
 The payoff to the holder of a short position in the European call option is
–max(ST – K, 0) = min(K – ST, 0)
 The payoff to the holder of a long position in a European put option is max(K – ST, 0)
 The payoff from a short position in a European put option is –max(K – ST, 0) = min (ST – K, 0)
Figure 5: Payoff’s from positions in European options: (a) long call; (b) short call; (c) long put; (d) short put.
Strike price = K; price of asset at maturity = ST.

8.3 Underlying Assets

1. Stock Options:
The exchanges trading stock options include the Chicago Board Options Exchange (CBOE), the Philadelphia
Exchange (PHLX), the American Stock Exchange (AMEX), and the Pacific Stock Exchange (PSE).
One contract gives the holder the right to buy or sell 100 shares at the specified strike price.
2. Foreign Currency Options
The Philadelphia Exchange is the major exchange for trading foreign currency options.
 PHLX offers both European and American contracts on currencies such as the Australian dollar, British pound,
Canadian dollar, etc.
 The size of one contract depends on the currency.
3. Index Options
The most popular contracts in the U.S. are those on the S&P 500 Index (SPX), the S&P 100 Index (OEX), the Nasdaq
100 Index (NDX), and the Dow Jones Industrial Index (DJX).
 The S&P 500 option is European, while the S&P 100 option is American.
 Settlement is in cash, and based on the index value at the end of the day on which exercise instructions are issued.
4. Futures Options
The underlying asset is a futures contract (which usually matures shortly after the expiration of the option).
 When a call option is exercised, the holder of the option acquires from the writer of the option:
i. a long position in the underlying futures contract plus
ii. a cash amount equal to the excess of the futures price over the strike price.
 When a put option is exercised, the holder of the option acquires from the writer of the option:
i. a short position in the underlying futures contract plus
ii. a cash amount equal to the excess of the strike price over the futures price.
8.4 Specifications of Stock Options

1. Expiration Dates
An exchange–traded stock option contract is an American–style option contract to buy or sell 100 shares of the stock.
a. A stock option is referenced by the month its expiration date occurs on.
b. Stock options run on a January, February, or March cycle.
1. The January cycle consists of the months of January, April, July, and October.
2. The February cycle consists of the months of February, May, August, and November.
3. The March cycle consists of the months of March, June, September, and December.
c. Longer–term options, known as LEAPS (long–term equity anticipation securities), have expiration dates up to
three years into the future. The expiration dates for LEAPS on stocks are always in January.
2. Strike Prices
The CBOE chooses strike prices at which options can be written so that they are spaced $2.50, $5, or $10 apart. For
example:
At a low stock price (e.g. $12), options trade with strike prices of $10, $12.50, and $15.
At a higher stock price (e.g. $100), options trade with strike prices of $90, $95, $100, $105, and $110
3. Terminology
For any asset at any given time, many different option contracts may be trading.
The following terms are defined:
Option class. All options of the same type (calls or puts) are referred to as an option class.
Option series: All options in a given class with the same expiration date and strike price are known as option series.
For example, the IBM 110 January calls are an option series.
Relative Option Values: Let S be the stock price and K be the strike price:
a. An “in–the–money option” produces a positive cash flow to the holder if exercised when (S > K).
Note: Often it is optimal for the holder of an in–the–money American option to wait rather than exercise
immediately. The option is then said to have time value.
b. An “at–the–money option” leads to zero cash flow to the option holder if exercised when (S = K).
c. An “out–of–the–money option” leads to a negative cash flow if exercised when (S < K).
Thus, an option will be exercised only if it is “in the money”.

The intrinsic value of a call and put option are defined as follows:
 For a call option, it is defined as the max of (S – K, 0).
 For a put option, it is defined as the max of (K – S, 0).
The total value of an option can be thought of as the sum of its intrinsic value and its time value.
4. Flex Options
The CBOE offers flex options on stock and stock indices.
 These options give traders on the floor of the exchange the ability to agree to nonstandard terms (e.g. strike prices
or expiration dates different from those usually offered by the exchange).
 Flex options are an attempt by the exchanges to regain business from the over–the–counter markets.
5. Dividends and Stock Splits
Both exchange–traded and over–the–counter options are not generally adjusted for cash dividends.
Exchange–traded options are adjusted for stock splits: In general, an n–for–m stock split should cause:
a. the stock price to go down to m/n of its previous value.
b. the exercise price is reduced to m/n of its previous value, while the number of shares covered by one contract is
increased to n/m of its previous value.
Example: Consider a call option to buy 100 shares of a company for $30 per share.
After a 2–for– 1 stock split, the option holder has the right to purchase 200 shares for $15 per share.
Stock options are adjusted for stock dividends.
Example: Consider a put option to sell 100 shares of a company for $15 per share.
After a 25% stock dividend is declared (which is equivalent to a 5–for–4 stock split), the option holder has the right to
sell 125 shares for $12.
6. Position Limits (Exercise Limits)
Position limits and exercise limits are designed to prevent investors or groups of investors from having undue
influence on the market. Therefore,
 The CBOE specifies position limits for option contracts. This defines the maximum number of option contracts
that a trader can hold on one side of the market. Long calls and short puts are on the same side of the market.
 Options on the largest traded stocks have positions limits of 75,000 contracts, while options on smaller
capitalization stocks have position limits of 60,000, 31,500, 22,500, or 13,500 contracts.

An extract of stock option quotations from Wall Street Journal is shown and described as follows:
Column (1) list the company and its stock price
Columns (3) and (4) identify the strike price and the expiration month for the option.
Columns (4) and (5) show the volume of trading and price at last trade for the call option.
Columns (6) and (7) show the volume of trading and price at last trade for the put option.
(1) (2) (3) (4) (5) (6) (7)
Call Put
Option StrikeP Exp Vol last Vol Last
AT&T
23.35 22.50 Apr 2933 2 12689 1.10
8.6 Trading
Options exchanges use market makers to facilitate trading. A market maker will:
 quote both a bid and an offer price on the option.
i. The bid is the price at which the market maker is prepared to buy.
ii. The offer is the price at which the market maker is prepared to sell.
The offer is always higher than the bid, and the amount by which the offer exceeds the bid is referred to as the
bid–offer spread.
The exchange sets upper limits for the bid–offer spread (e.g. the spread can be no more than $0.25 for options
priced at less than $0.50, etc).
 ensure that buy and sell orders can always be executed at some price without delays.
Offsetting Orders (used to close out positions)
Buyers of options can close out their positions by issuing an offsetting order to sell the same option.
Option writers can close out their position by issuing an offsetting order to buy the same option.

8.7 Commissions
Commissions for option trades vary dependent upon broker type (e.g. discount vs. full service).
 The actual amount charged is usually a fixed cost plus a proportion of the dollar amount of the trade, subject to
maximum and minimum commissions.
 Based on the discount brokerage commission table below, the purchase of 8 contracts when the option price is $3
would cost $20 + [0.02 × ($3 × 8 × 100)] = $68 in commissions.
Dollar amount of Trade Commission*
< $2,500 $20 + 0.02 of the dollar amount
$2,500 to $10,000 $45 + 0.01 of the dollar amount
> $10,000 $120 + 0.0025 of the dollar amount
*Maximum commission is $30 per contract for the first five contracts plus $20 per contract for each additional
contract.
Minimum commission is $30 per contract for the first contract plus $2 per contract for each additional contract.
The market maker’s bid–offer spread is a hidden cost in option trading (and in stock trading). For example:
 Given a bid price of $4.00, an offer price of $4.50, we can assume that a “fair” price for the option is halfway
between the two amounts, or $4.25.
 The cost to the buyer and to the seller of the market maker system is the difference between the fair price and the
paid price (in this case, the cost is $4.50 – $4.25 = $0.25 per option, or $25 per contract).
8.8 Margins
Differences exist in how payment is rendered when purchasing stocks and options in the U.S.
When shares are purchased, an investor can either pay cash or borrow using a margin account.
The initial margin required is usually 50% of the value of the shares.
The maintenance margin is usually 25% of the value of the shares.
When call and put options are purchased, the option price must be paid in full since options are highly leveraged
investments. Buying on margin would raise this leverage to an unacceptable level.
Investors who write options are required to maintain funds in a margin account.
The size of the margin required depends on the circumstances.
Writing Naked Options
A naked option is one that is not combined with an offsetting position in the underlying stock.
The initial margin when writing a naked options is the greater of the results of the following two calculations:
1. 100% of the proceeds of the sale + 20% of the underlying share price – the amount by which the option is out of
the money (if any).
2. 100% of the proceeds of the sale + 10% of the underlying share price.
For options on a broadly based index, 15% (instead of 20% in the above calculation) is generally used, because an
index is usually less volatile than the price of an individual stock.
Example:
An investor writes 4 naked call option contracts on a stock.
The option price is $5, the strike price is $40, and the stock price is $38.
Note: the option is $2 out of the money.
The second calculation gives 400 × [$5 + (0.10 × $38)] = $3,520
The first calculation gives 400 × [$5 + (0.2 × $38) – $2] = 4,240 (and this is the initial margin requirement).
Notes:
If the investor wrote a put option, the margin requirement would be $5040 [400 ($5 + 0.2 x $38)] (since the put is
already $2 in the money).
In both cases, the proceeds of the sale, $2,000 (400 × $5), can be used to form part of the margin account.
Writing Covered Calls
A covered call occurs when call options are written on shares that might have to be delivered, but are already owned.
Covered calls are less risky than writing naked calls because the worst case the investor faces is selling shares already
owned at below their market value.
If covered call options are out of the money, no margin is required.
Example: An investor decides to buy 200 shares of a certain stock on margin and to write two call options.
The stock price is $63, the strike price is $65, and the price of the option is $7.

Compute the minimum cash required from an investor given the following trades.
To determine the minimum cash required, note the following:
i. the margin account allows the investor to borrow 50% of the price of the stock, or $6,300.
ii. the price received for the options is $1400 ($7 × 200) are used to purchase of the shares.
iii. the cost of the shares is $12,600 ($63 × 200).
Thus, the minimum cash required for his/her trades is: $12,600 – $6,300 – $1,400 = $4,900.
8.9 The Options Clearing Corporation

The Options Clearing Corporation (OCC) guarantees that option writers fulfill their obligations and keeps a record of
all long and short positions.
 All options trades must be cleared through a member of the OCC.
 Members are required to contribute to a fund to be used if any member defaults on an option obligation.
 The writer of the option maintains a margin account with a broker.
 The broker maintains a margin account with the OCC member (that clears its trades).
 The OCC member maintains a margin account with the OCC.
Exercising on Option
When an investor wishes to exercise an option:
 The investor notifies the broker handling the trade.
 The broker notifies the OCC member that clears its trades.
 The OCC member places an exercise order with the OCC.
 The OCC randomly selects a member with an outstanding short position in the same option.
 The member selects a particular investor who has written the option.
8.10 Regulation
Regulation occurs in a number of ways:
Both the exchange and the OCC have rules governing the behavior of traders.
Both federal and state regulatory authorities monitor activities.
a. The SEC regulates options markets in stocks, stock indices, currencies, and bonds at the federal level.
b. The Commodity Futures Trading Commission regulates markets for options on futures.
c. In Illinois and New York (major options markets), state laws enforce acceptable trading practices.
In general, options markets have demonstrated a willingness to regulate themselves, as no major scandals or defaults
by OCC members have occurred.
8.11 Taxation
For tax purposes, a gain or loss is recognized when:
a. the option expires unexercised, or
b. the option is sold.
Wash Sale Rule
To prevent investors from selling stock at a loss and immediately repurchasing it so that a loss can be realized for tax
purposes, tax authorities have ruled that when the repurchase is within 30 days of the sale (i.e., between 30 days before
the sale and 30 days after the sale), any loss on the sale is not deductible.
Constructive Sales
Prior to 1997, a taxpayer selling a security short while holding a long position in a substantially identical security faced
no gain or loss until the short position was closed out.
This means that short positions could be used to defer recognition of a gain for tax purposes. As a result of the Tax
Relief Act of 1997, an appreciated property is now treated as “constructively sold” when the owner:
1. Enters into a short sale of the same or substantially identical property.
2. Enters into a futures or forward contract to deliver the same or substantially identical property; or
3. Enters into one or more positions that eliminate substantially all of the loss and opportunity for gain.
8.12 Warrants, Executive Stock Options, and Convertibles

Warrants and executive stock options are call options written by a company on its own stock.
When exercised, additional shares of company stock are issued and sold to the option holder for the strike price. Thus,
exercise leads to an increased number of shares of the company’s stock that are outstanding.

Warrants:
 often come into existence as a result of a bond issue.
 are added to the bond issue to make it more attractive to investors.
Executive stock options:
 are issued to motivate company personnel to act in the best interests of the company’s shareholders.
 are usually at–the money when first issued.
 cannot be traded (unlike warrants and exchange–traded stock options).
Convertible bonds
 can be considered an embedded call option on the company’s stock.
 can be converted into equity at certain times using a predetermined exchange ratio.
 lead to more shares being issued by the company when they are exercised.
8.13 Over the Counter Markets

The over–the–counter market for options is now larger than the exchange–traded market.
In the over–the–counter market, financial institutions, corporate treasurers, and fund managers trade over the phone.
The chief disadvantage of the over–the–counter market is that option writer may default. This means that the purchaser
is subject to some credit risk. To overcome this disadvantage, market participants may require counterparties to post
collateral.

Chapter 9: Properties of Stock Options
Section Description
9.0 Introduction
9.1 Factors Affecting Option Prices
9.3 Upper and Lower Bounds for Option Prices
9.4 Put–Call Parity
9.5 Early Exercise: Calls on a Non–Dividend Paying Stock
9.6 Early Exercise: Puts on a Non–Dividend Paying Stock
9.7 Effect of Dividends
9.0 Introduction
This chapter covers stock option properties.
Different arbitrage arguments are used to explore the relationships between European option prices, American option
prices, and the underlying stock price.
The most important relationship is put–call parity, which is a relationship between European call option prices and
European put option prices.
The chapter examines American options. The author’s conclusions are:
i. it is never optimal to exercise American call options on non–dividend–paying stocks before expiration.
ii. early exercise of an American put option on such a stock can be optimal.
9.1 Factors Affecting Option Prices

Six factors affect the price of a stock option:
1. The current stock price, S0
2. The strike price, K
3. The time to expiration, T
4. The volatility of the stock price, 
5. The risk–free interest rate, r
6. The dividends expected during the life of the option
What happens to option prices when one of these factors change while all others remaining fixed?
The results are described on the following page, and then summarized in tabular form.
Figures 1 and 2 show how European call and put prices depend on the first five factors.
Here, S0 = 50, K = 50, r = 5% per annum,  = 30% per annum, T = 1 year, and there are no dividends.
The call price is 7.116 and the put price is 4.677.
1 and 2. Stock Price and Strike Price
 The payoff from a call option equals the Stock price minus the Strike price. Thus, call options become more
valuable as the stock price increases, and less valuable as the strike price increases. This effect is shown in the
table below by + and – symbols below.
 The opposite effect holds true for put options.
The effect on the price of a stock option of increasing one variable while keeping all others fixed
Variable European Call European Put American Call American Put
Stock price + – + –
Strike price – + – +
Time to expiration ? ? + +
Volatility + + + +
Risk–free rate + – + –
Dividends – + – +
+ indicates that an increase in the variable causes the option price to increase

Figure 1: Effect of changes in stock price, strike price, and expiration date on option prices when S0 = 50,
K = 50, r = 5%,  = 30%, and T = 1.
Figure 2: Changes in volatility & risk–free interest rate on option prices if S0 = 50, K = 50, r = 5%,  = 30%, & T = 1.
3. Time to Expiration
A. American options:
Both put and call options become more valuable as the time to expiration increases. Since the owner of a
longer–life option has all the exercise opportunities available to the owner of a shorter–life option (and more),
an increase in time to expiration causes the option price to increase.
B. European options:
Put and call options do not necessarily become more valuable as the time to expiration increases.
1. As stated earlier, European options can only be exercised at the maturity.
2. Consider two European call options on a stock: one with an expiration date in 1 month, and the other with an
expiration date in 2 months. Suppose a large dividend is expected in 6 weeks. Since the dividend will cause the stock
price to decline, the short–life option will be worth more than the long–life option.
4. Volatility
The owner of a call benefits from price increases, and is limited in downside risk to the price of the option.
The owner of a put benefits from price decreases and has limited downside risk in the event of price increases.
Therefore, the values of both calls and puts increase as volatility increases.

5. Risk–Free Interest Rate

Puts: The combined impact of the following two effects tends to decrease the value of a put option:
 As interest rates increase, the expected growth rate of the stock price tends to increase. However,
 The present value of future cash flows received by the holder of the option decreases.
Calls: In the case of calls, the first effect tends to increase the price (and dominates the second effect) and the second
effect tends to decrease it.
Thus, the price of a call always increases as the risk–free interest rate increases.
6. Dividends
Dividends have the effect of reducing the stock price on the ex–dividend date. Therefore, dividends have a negative
effect on the value of call options and a positive effect on the value of put options.

Assume there are some market participants, such as large investment banks, for which:
 there are no transactions costs.
 all trading profits (net of trading losses) are subject to the same tax rate.
 borrowing and lending at the risk–free interest rate is possible.
 there are no arbitrage opportunities.
S0: current stock price. ST: stock price at time T.
K: strike price of option
T: time to expiration for an option
r: risk–free rate of interest for maturity T (continuously compounded) (r > 0)
C: value of American call option to buy one share
P: value of American put option to sell one share
c: value of European call option to buy one share
p: value of European put option to sell one share
9.3 Upper and Lower Bounds for Option Prices

When option prices are above their upper (or below their lower) bounds, profitable opportunities for arbitrageurs exist.
1. Upper Bounds:
Since an American or European call option gives the holder the right to buy one share of a stock, the option can never
be worth more than the stock. Thus,
c  S0 and C  S0
Since an American or European put option gives the holder the right to sell one share for K, the option can never be
worth more than K.
p  K and P  K
For European put options, the value of the option at time T will not be worth more than K, thus its present value and
therefore,
p  K e–rT
2. Lower Bound for European Calls on Non–Dividend–Paying Stocks
The price of a European call option on a non–dividend–paying stock is: S0 – Ke–rT
Example 1:
Assume S0 = $20, K = $18; r =10%, and T = 1, then S0 – Ke–rT = $20 – $18e–0.1 = $3.71
Example 2:
Consider the situation when c = $3.00 (which is less than its theoretical minimum of $3.71).
An arbitrageur can Short the stock and then Buy the call:
i. The cash inflow equals 20.00 – 3.71 = 17.00, and its accumulated value (at the end of the year) at the risk–free rate
of return is $17e0.1 = $18.79
If the stock price is > $18, exercise the option and close out the short position. This results in a profit of 18.79 – 18 =
0.79
If the stock price is < $18 (assume $17), close out the short position and buy the stock. This results in a profit of 18.79
– 17 = 1.79
To arrive at a lower bound for a European call option, consider the two portfolios below:
Portfolio A: one European call option plus a cash amount equal to (Ke–rT )
Portfolio B: one share.

The cash amount is invested at the risk–free interest rate and it will grow to K at time T.
Q. What is Portfolio A and B worth at time T?
If ST > K, the call option is exercised at time T and portfolio A is worth ST
If ST < K, the call option expires worthless and the portfolio is worth K.
A. At time T, portfolio A is worth max (ST, K), and portfolio B is worth ST
Thus, at time T, Portfolio A is always worth at least as much as (sometimes more than) portfolio B.
At time 0, this can be represented as c + Ke–rT  S0 or c  S0 – Ke–rT
The worst case scenario for a call option is for it to expire worthless. Since its value cannot be negative, the formula
for the lower bound on a European call option is: c  max(S0 – Ke–rT , 0)
Example: Let S0 = 51, K = 50, T = 0.5, and r = 0.12.
Since the price of a European call option on a non–dividend–paying is S0 – Ke–rT, the lower bound equals
51 – 50×e–0.12×0.50 = 3.91
3. Lower Bound for European Puts on Non–Dividend–Paying Stocks
A lower bound for the a European put option on a non–dividend–paying stock is Ke–rT – S0
Example 1: Let S0 = 37, K = 40, T = 0.5, and r = 0.05.
The lower bound is 40×e–0.05×0.50– 37 = 2.01
Example 2: Consider p = $1.00 (which is less than its theoretical minimum of $2.01). An arbitrager can
 Borrow $38.00 for six months to buy both the put ($1) and the stock ($37).
 Repay 38×e0.05×0.50 = 38.96. At the end of the six months,
If the stock price < $40.00, exercise the option to sell the stock for $40.00, repay the loan. This results in a profit of
$40.00 – $38.96 = $1.04
If the stock price is > $40.00, (say $42.00), discard the option, sell the stock, and repay the loan. This results in a profit
of $42.00 – $38.96 = $3.04
For a more formal argument, consider the following two portfolios:
Portfolio C: one European put option plus one share
Portfolio D: A cash amount equal to Ke–rT
Q. What are Portfolios C and D worth at time T?
If ST < K, the option is exercised, and the portfolio is worth K.
If ST > K, the put option is worthless and portfolio C is worth ST
A. At time T portfolio C is worth max (ST, K) and portfolio D is worth K.
At time T, portfolio C is always worth at least as much as (sometimes more than) portfolio D.
At time 0, this can be represented as p + S0  Ke–rT or p  Ke–rT– S0
The worst case scenario for a put option is for it to expire worthless. Since its value cannot be negative, the formula for
the lower bound on a European put option is
p  max(Ke–rT– S0, 0)
Example: Assume S0 = 38, K = 40, T = 0.25, and r = 0.10.
Compute the lower bound for a European put option on a non–dividend–paying stock.
Step 1: Write a formula for the price European put option on a non–dividend–paying stock: Ke–rT– S0
Step 2: Determine the lower bound based on the formula p  max(Ke–rT– S0, 0).
Thus, the lower bound equals 40×e–0.10×0.25– 38 = 1.01
9.4 Put–Call Parity

1. European Options
The relationship between p and c (for a non–dividend–paying stock) is described below.
Consider the following portfolios previously described:
Portfolio A: one European call option plus cash equal to (Ke–rT)
Portfolio C: one European put option plus one share
Both are worth Max (ST, K) at expiration of the options.
Since both are European, and cannot be exercised prior to expiration, both must have the same present value. This
leads to put–call parity.
Put–call parity: p + S0 = c + Ke–rT
Based on put–call parity, the value of a European call with a certain exercise price and exercise date can be determined
from the value of a European put with the same exercise price and date, and vice versa.
If put–call parity does not hold, arbitrage opportunities exist.

For example, assume S0 = 31, K = 30; c = 3, p = 2.25; T = 0.25, and r = 0.10.

The values of portfolio A and C are computed as follows:
Portfolio A: c + Ke–rT = 3 + 30×e–0.10×0.25 = 32.26
Portfolio C: p + S0 = 2.25 + 31 = 33.25
Thus, portfolio C is overpriced relative to portfolio A.
In this case, arbitrage is accomplished by buying the call and shorting both the put and the stock.
The positive cash flow from this strategy (recognized immediately) is –3 + 2.25 + 31 = 30.25.
The accumulated value at the end of 3 months is 30.25×e0.10×0.25 = 31.02
If ST > $30 at expiration, exercise the call, and buy the stock.
If ST < $30 at expiration, exercise the put.
In either case, the investor ends up buying one share for $30, which can be used to close out the short position. The net
profit is $31.02 – $30.00 = $1.02
Table 2: Arbitrage opportunities when put–call parity does not hold. Stock price = $31; interest rate = 10%; call price
= $3. Both put and call have a strike price of $30 and 3 months to maturity.
Three–month put price = $2.25 Three–month put price = $1
Action now: Action now:
Buy call for $3 Borrow $29 for 3 months
Short put to realize $2.25 Short call to realize $3
Short the stock to realize $31 Buy put for $1
Invest $30.25 for 3 months Buy the stock for $31
Receive $31.02 from investment Call exercised: sell stock for $30
Exercise call to buy stock for $30 Use $29.73 to repay loan
Net profit = $1.02 Net profit = $0.27
Action in 3 months if ST < 30: Action in 3 months if ST < 30:
Receive $31.02 from investment Exercise put to sell stock for $3
Put exercised: buy stock for $30 Use $29.73 to repay loan
Net profit = $1.02 Net profit = $0.27
2. American Options
Although Put–call parity holds only for European options, it is possible to derive some results for American option.
Put–call parity for American option on stocks paying no dividends can be shown as:
S0 – K  C – P  S0 – Ke–rT
Example 1: Let S0 = 19, K = 20; c = 1.50; T = 5/12, and r = 0.10
Compute the price of a European put with exercise price of $20 and maturity in five months.
Step 1: Write an equation for the price of the European put using the put–call parity: p = c + Ke–rT – S0.
Step 2: Given values in the problem, solve for the price of the European put. The price of the European put with
exercise $20 and maturity in five months is p = c + Ke–rT – S0 = 1.5 + 20×e–0.10×(5/12) – 19 = 1.68.
Example 2: Determine the upper and lower bounds for the price of an American put. Using put–call parity for
American options on non–dividend paying stocks, and the data given in the previous problem, solve for the upper and
lower bounds. Since S0 – K  C – P  S0 – Ke–rT then 19 – 20  C – P  19 – 20×e–0.10×(5/12);
1  P – C  0.18. When C = 1.50, then 2.50  P  1.68; Thus, $1.68 and $2.50 are the upper and lower bounds for the
price of an American put.
9.5 Early Exercise: Calls on a Non–Dividend Paying Stock

Key result: It is never optimal to exercise an American call option on a non–dividend–paying stock before the
expiration date.
Example: Assume S0 = 50, K = 40; T = 0.833, and r = 0.10. Note: The option is deep in the money since the stock price
is $50 and the strike price is $40.
Scenario 1: If the trader plans to hold the stock for more than one month, it is better is to keep the option and exercise
it at the end of the month, rather than exercising immediately since:
i. interest is earned on the $40 for one month.
ii. the trader has insurance against a decline in the stock price below $40.
Scenario 2: What if there is belief that the stock is currently overpriced? In this case, the trader is better off selling the
option than exercising it. The price obtained should be greater than its intrinsic value of $10.

A more formal argument follows:

It has been shown that c  S0 – Ke–rT
Since C > c, it follows that C  S0 – Ke–rT
Since r > 0, C > S0 – K.
If it were optimal to exercise early, C would equal S0 – K. Thus, it can never be optimal to exercise early.
Figure 3 shows the way in which the call price varies with S0.
 It indicates that the call price is always above its intrinsic value of max(S0 – K, 0).
 As r or T or the volatility increases, the line relating the call price to the stock price moves in the direction
indicated by the arrows (i.e., farther away from the intrinsic value).
Figure 3: Variation of price of an American or European call option on a non–dividend paying stock with the stock
price, S0.
9.6 Early Exercise: Puts on a Non–Dividend Paying Stock

Key result: It can be optimal to exercise an American put option on a non–dividend–paying stock before the expiration
date. At any time during its life, a put option should always be exercised early if it is sufficiently deep in the money.
For example, assume K = $10 and ST is close to $0.
i. Immediate exercise leads to an immediate gain of $10.
ii. If not exercised immediately, the gain might be less $10, but cannot be more than $10.
iii. Receiving $10 now versus later is preferable.
Thus, it is always optimal to exercise an American put immediately when the stock price is sufficiently low. When
early exercise is optimal, the value of the option is K – S0.
Figure 4 shows the general way in which the price of an American put varies with S0.
Provided that r > 0, it is always optimal to exercise an American put immediately when the stock price is sufficiently
low. When early exercise is optimal, the value of the option is K – S0.
The curve representing the value of the put therefore merges into the put's intrinsic value, K – S0, for a sufficiently
small value of S0. This value of S0 is shown as point A. The line relating the put price to the stock price moves in the
direction indicated by the arrows when r decreases, when the volatility increases, and when T increases.
Figure 4: Variation of price of an American put option with stock price, S0.

Figure 5 shows the variation of the European put price with the stock price.

Note that point B in Figure 5, at which the price of the option is equal to its intrinsic value, must represent a higher
value of the stock price than point A in Figure 4.

Point E in Figure 5 is where S0 = 0 and the European put price is Ke–rT
Figure 5: Variation of price of a European put option with the stock price, S0.
9.7 Effect of Dividends

In the U.S., exchange–traded stock options generally have less than 8 months to maturity, and thus, dividends payable
during the life of the option are easily determinable.
Let D denote the present value of the dividends during the life of the option. Dividends are assumed to occur at the
time of its ex–dividend date (which is the approximate date that one must be a shareholder to receive the next
dividend).
1. Determining Lower Bound for Calls and Puts
Redefine portfolios A and B as follows:
Portfolio A: one European call option plus cash equal to D + Ke–rT
Portfolio B: one share
Based on logic previously used for determining the lower bound for European calls on a non–dividend paying stock, it
can be shown that c  S0 – D – Ke–rT
Redefine portfolios C and D as follows:
Portfolio C: one European put option plus one share.
Portfolio D: cash equal to D + Ke–rT
Based on logic previously used for determining the lower bound for European puts on a non–dividend paying stock, it
can be shown that p  D + Ke–rT – S0
Early Exercise
When dividends are expected, early exercise of an American call option may or may not be optimal.
 It is sometimes optimal to exercise an American call immediately prior to an ex–dividend date (since the dividend
causes the stock price to jump down, making the option less attractive).
 It is never optimal to exercise a call at other times.
Put–Call Parity
Comparing the value at option maturity of the redefined portfolios A and C shows that, with dividends, the
put–call parity result in p + S0 = c + Ke–rT becomes
p + S0 = c + D + Ke–rT = p + S0
Dividends cause S0 – K  C – P  S0 – Ke to be modified (see Problem 9.19) to
–rT
S0 – D – K  C – P  S0 – D – Ke–rT


Chapter 10: Trading Strategies Involving Options
Section Description
10.0 Introduction
10.1 Strategies Involving a Single Option and a Stock
10.2 Spreads
10.3 Combinations
10.4 Other Payoffs
10.0 Introduction
This chapter covers the range of profit patterns obtainable using options.
Although the underlying asset used is a stock, similar patterns can be obtained for other underlying assets such as
foreign currencies, stock indices, and futures contracts.
The key attraction to options is their use in creating a very wide range of payoff patterns.
In the figures below, profit will be shown as the final payoff minus the initial cost (note: it should be calculated as the
present value of the final payoff minus the initial cost).
10.1 Strategies Involving a Single Option and a Stock

Trading strategies involving a single option and the stock itself are as follows (Figures in next page):
Call Strategies.
A Covered Call is a long position in a stock plus a short position in a call option. The position “covers” a trader when
stock prices risk and a call is made on the stock.
The reverse of a writing a covered call is a short position in a stock plus a long position in a call option
Put Strategies.
A Protective Put requires buying a put option on a stock and the stock itself.
The reverse of a protective put is a short position in a put option plus a short position in the stock.
In chapter 9, Put–Call parity, p + S0 = c + Ke–rT + D, was introduced. It can be re–written as
S0 – c = – p + (Ke–rT + D)
Thus, (a long stock + a short call) = (a short put + certain amount of cash).
10.2 Spreads
A spread trading strategy:
 involves taking a position in two or more options of the same type (i.e., two or more calls or two or more puts).
 includes Bull Spreads, Bear Spreads, Box Spreads, Butterfly Spreads, Calendar Spreads, and Diagonal Spreads:
1. Bull Spreads (Buying a Call and Selling a Call):

Bull spreads can be created by buying a call option with a certain strike price (K1) and selling a call option on the same
stock with a higher strike price (K2).
Both options have the same expiration date.
A bull spread, created from calls, requires an initial investment because the value of the option sold is less than the
value of the option bought.
Total payoff is computed as the sum of the payoffs on each option and varies dependent upon the stock’s price when
the option is exercised.
Profit is calculated by subtracting the initial investment from the total payoff.
A bull spread strategy limits the trader’s upside, as well as downside, risk. Three types of bull spreads are:
1. Both calls are initially out of the money.
a. This is the most aggressive type.
b. Setup cost is small, and there is a small probability of receiving high payoff (K2 – K1).
2. One call is initially in the money; the other call is initially out of the money.
3. Both calls are initially in the money.
Spreads become more conservative as one moves from type 1 to type 3.
Assume an investor buys a call for $3 with a strike price of $30 and sells for a call for $1 with a strike price of $35.
The cost of the strategy is $3 – $1 = $2.

Figure 1: Profit patterns (a) long position in a stock combined with short position in a call; (b) short position in a stock
combined with long position in a call; (c) long position in a put combined with long position in a stock; (d) short position
in a put combined with short position in a stock.
Payoffs from a bull spread.

Stock Price Range Payoff From Long Call Option Payoff From Short Call Option Total Payoff
ST > K2 ST – K1 K2 – ST K2 – K1
K1 < ST < K2 ST – K1 0 ST – K1
ST < K1 0 0 0
The profit is therefore computed as follows:
Stock Price Range Profit
ST < 30 –2
30 < ST < 35 ST – 32
ST > 35 3
Figure 2: Profit from bull spread created using call options.
Bull spreads created by puts involve buying a put with a low strike price and selling a put with a high strike price.

Unlike bull spreads created from calls, bull spreads created from puts involve:
 a positive cash flow to the trader up front, and
 a payoff that is either negative or zero.
Figure 3: Profit from bull spread created using put options.
2. Bear Spreads: Buying a Call (with a higher strike price) and Selling a Call
An investor entering a bear spread is hoping that the stock price will fall.
A bear spread is created by buying a call with one strike price and selling a call with another strike price. However,
the strike price of the option purchased is greater than the strike price of the option sold.
It involves an initial cash inflow (since the price of the call sold is greater than the price of the call purchased).
Payoffs from a bear spread.

Stock Price Range Payoff From Long Call Option Payoff From Short Call Option Total Payoff
ST  K2 ST – K2 K1 – ST –(K2 – K1)
K1 < ST < K2 0 K1 – ST –(ST – K1 )
ST  K1 0 0 0
Assume an investor buys a call for $1 with a strike price of $35 and sell a call for $3 a call with a strike price of $30.
The investment generates $3 – $1 = $2 up front.
The profit is therefore computed as follows:
Stock Price Range Profit
ST  30 +2
30 < ST < 35 32 – ST
ST  35 –3
Bear spreads limit both the upside profit potential and the downside risk.
Bear spreads created with puts require:
i. buying a put with a high strike price and sell a put with a low strike price.
ii. an initial investment.
Figure 4: Profit from bear spread created using put options.
Figure 5: Profit from bear spread created using call options.

Box Spreads
A box spread is a combination of a bull call spread with strike prices K1 and K2 and a bear put spread with the same
two strike prices.
 Table 3 shows that the payoff from a box spread is always K2 – K1.
 The value of a box spread is always the present value of this payoff or (K2 – K1)e–rT.
 If it has a different value there is an arbitrage opportunity.
If the market price of the box spread is too low, it is profitable to buy the box; Buy a call with strike price K1, buy a put
with strike price K2, sell a call with strike price K2, and sell a put with strike price K1.
If the market price of the box spread is too high, it is profitable to sell the box. Buy a call with strike price K2, buy a
put with strike price K1, sell a call with strike price K1, and sell a put with strike price K2.
 A box–spread arbitrage only works with European options.
Table 3: Payoff from a box spread.
Stock Price Range Payoff From bull call spread Payoff From bear put spread Total Payoff
ST  K2 K2 – K1 0 K2 – K1
K1 < ST < K2 ST – K1 K2 – ST K2 – K1
ST  K1 0 K2 – K1 K2 – K1
3. Butterfly Spreads involve positions in options with three different strike prices:
i. buying a call option with a high strike price (K3) and a call with a low strike price (K1)
ii. selling 2 call options with a strike price K2 (close to stock price and halfway between K1 and K3)
A butterfly spread:
 leads to a profit if the stock price stays close to K2
 gives rise to a small loss if there is a significant stock price move in either direction.
Payoffs from a butterfly spread.
Stock Price Range Payoff From First Payoff From Second Payoff From Total
Long Call Long Call Short Calls
ST < K1 0 0 0 0
K1 < ST < K2 ST – K1 0 0 ST – K1
K2 < ST < K3 ST – K1 0 –2 (ST – K2) K3 – ST
ST > K3 ST – K1 ST – K3 –2 (ST – K2) 0
Assume a stock is currently worth $61 (and a significant price move in the next 6 months is unlikely).
The market prices of six–month calls are as follows:

Strike Price ($) Call Price ($)
55 10
60 7
65 5
To create a butterfly spread:
i. Buy one call with a $55 strike price
ii. Buy one call with a $65 strike price
iii. Sell two calls with a $60 strike price.
The cost to create the spread is $10 + $5 – (2 × $7) = $1
The maximum profit, $4, occurs when the stock price in six months is $60 (i.e. 60 – 55 – 1).
Butterfly spreads created using put options requires the investor to:
i. Buy a put with a low strike price
ii. Buy a put with a high strike price
iii. Sell two puts with a intermediate strike price.

Figure 6: Profit from butterfly spread created using call options.
Figure 7: Profit from butterfly spread created using put options.
4. Calendar Spreads
This strategy involves options having the same strike price but different expiration dates.
A calendar spread is created by selling a call option and buying a longer–maturity call option with the same strike
price. It requires an initial investment.
A profit is made if the stock price at the expiration of the short–maturity option is close to the strike price of the short–
maturity option.
A loss is incurred if the stock price is significantly above or significantly below this strike price.
Types of calendar spreads (and relative strike price choices) are as follows:
A Neutral calendar spread (chosen with a strike price close to the current stock price).
A Bullish calendar spread (involves the use of a higher strike price).
A Bearish calendar spread (involves the use of a lower strike price).
Calendar spreads can be created using put options when an investor buys a long–maturity put option and sell a short–
maturity put option.
5. Diagonal Spreads
Bull, bear, and calendar spreads can all be created from a long position in one call (put) and a short position in another
call (put). In the case of bull and bear spreads, the calls have different strike prices and the same expiration date. In the
case of calendar spreads, the calls have the same strike price and different expiration dates. A diagonal spread requires
that both the expiration dates and the strike prices of the calls (puts) be different. This increases the range of profit
patterns that are possible.
10.3 Combinations
Combination strategies involve taking a position in both calls and puts on the same stock.
Four types of combinations are Straddles, Strips, Straps, and Strangles.
1. The Straddle:
The straddle involves buying a call and a put with the same strike price and expiration date.
If the stock and strike price are close at the option's expiration, a loss ensues.
If there is a large price swing in either direction, a significant profit will result.
A straddle is appropriate when there is an expectation of a large move in a stock price but the direction is not known.
Payoffs from a straddle.
Stock Price Range Payoff From Call Payoff From Put Total Payoff
ST < K 0 K – ST K – ST
ST > K ST – K 0 ST – K

Example: Assume S0  69, K  70, T  3 /12, c  4.00, p  3

1. If the stock price stays at $69, the cost is $6 (An up–front investment of $7 is required, the call expires worthless,
and the put expires at a value of $1.)
2. If the price rises to $70, a loss of $7 results (and this is the worst that can happen).
3. If the stock jumps to $90, a profit of $13 is made;
4. If the stock moves down to $55, a profit of $8 is made.
The straddle, shown below, is sometimes referred to as a bottom straddle or straddle purchase.
The dashed line shows the relationship between profit and the stock price for the individual securities constituting the
portfolio.
The solid line shows the relationship between profit and the stock price for the whole portfolio.
A top straddle or straddle write is the reverse position. It is created by selling a call and a put with the same exercise
price and expiration date. If the stock price on the expiration date is close to the strike price, a significant profit results.
2 and 3. Strips and Straps
A strip: A long position in one call and two puts with the same strike price and expiration date. The bet is that a big
stock price move occurs, but a decrease in stock price is considered more likely.
A strap: A long position in two calls and one put with the same strike price and expiration date. The bet is that a big
stock price move will occur, but a increase in stock price is considered more likely.
Profit patterns from a Strip (a) and a Strap (b)
4. Strangles (a.k.a. bottom vertical combination)

Buy a put and a call with the same expiration date and different strike prices (the call strike price, K2, is higher than the
put strike price, K1).
Similarities to a straddle:
1. The bet is that there will be a large price move but there is uncertainty whether it will be an increase or a decrease.
2. The stock price has to move farther in a strangle than in a straddle for the trader to make a profit.
3. The downside risk if the stock price ends up at a central value is less with a strangle. The farther apart the strike
prices are the less the downside risk.
Payoffs from a strangle.
Range of Stock Price Payoff From Call Payoff From Put Total Payoff
ST < K1 0 K1 – ST K1 – ST
K1 < ST < K2 0 0 0
ST > K2 ST –K2 0 ST – K2
Profit patterns from a strangle

10.4 Other Payoffs

The key point: In theory, if European options expiring at time T were available with every single possible strike price,
any payoff function could be obtained.
This can be demonstrated using butterfly spreads.
Recall that butterfly spreads are created by buying options with strike prices K1 and K3 and selling two options with
strike price K2 where K1 < K2 < K3 and K3 – K2 = K2 – K1.
Thus, the payoff pattern of a butterfly spread can be described as a spike.
As K1 and K3 move closer together, the spike becomes smaller.
Through combination of a large number of very small spikes, any payoff function can be approximated.
Payoff from a butterfly spread

Payoff
K1 K2 K3 ST


Chapter 11: Binomial Trees
Section Description
11.0 Introduction
11.1 A One–Step Binomial Model
11.2 Risk–Neutral Valuation
11.3 Two–Step Binomial Trees
11.4 A Put Example
11.5 American Options
11.6 Delta
11.7 Matching Volatility with u and d
11.8 Increasing the Number of Steps
11.9 Options on Other Assets
11.0 Introduction
The construction of a binomial tree is a useful technique to price a stock option. The binomial tree diagrams the
different paths a stock price might take over the life of an option.
11.1 A One–Step Binomial Model

Example: Valuing a European call option to buy a stock currently priced at $20 for $21 in three months. Assume that
at the end of three months the stock price will be either $22 or $18.
If the stock price is $22, the value of the option will be $1.
If the stock price is $18, the value of the option will be zero.
To value the option, a portfolio consisting of a certain number of shared and a short position in one call option can be
established so that there is no uncertainty about the value of the portfolio at the end of 3 months.
First, consider a portfolio consisting of  shares and a short position in 1 call option.
The portfolio is riskless if  is chosen so that the final value of the portfolio is the same for both alternatives.
If the stock rises to $22, the total portfolio value is 22 – 1 (cost of the option)
If the stock falls to $18, the total portfolio value is 18
Solving for the optimal % of shares, we see that 22 – 1 = 18 and thus  = 0.25 shares. Therefore,
If the stock price rises to $22, the value of the portfolio is $22 × 0.25 – 1 = $4.5
If the stock price falls to $18, the value of the portfolio is $18 × 0.25 = $4.5
Next, if the portfolio is riskless, it must earn the risk–free rate of interest.
If the risk–free rate is 12%, the present value of the portfolio is 4.5×e–0.12×0.25 = 4.367
Finally, since we know the value of the portfolio consisting of 1 call option and .25 shares of the stock equals 4.367,
we can solve for the option price. Let f = the option price.
$20 × 0.25 – f = $5 – f = $4.367.
Therefore, f = 0.633.
A generalized approach to determining option values follows:
The following notation is used.
S0 the current stock price.
f the current price of an option on the stock.
T Time until option expiration.
Assumptions:
The stock price can move up or down to a new level and end up at S0u or S0d at time T.
If the stock price moves up to S0u or down to S0d, the payoff from the option is fu or fd.
Consider a portfolio consisting of  shares of the stock and a short position in one call option.
Step 1: Calculate the value of  that makes the portfolio riskless:
If the stock price rises, the value of the portfolio is S0u – fu
f  fd
If the stock price falls the value of the portfolio is S0 d  – fd. Thus,   u
S0 u  S0 d
Step 2: Establish an equation of value which equates the cost of setting up a portfolio to the present value of a portfolio
consisting of  shares of the stock and a short position in one call option: S0  – f = (S0 u  – fu) e–rT
erT  d
Step 3: Solve for f: f = e–rT [p fu + (1 – p) fd], where p 
ud

The following example validates the results we obtained in the previous example.
e0.120.25  0.9
In the previous example, u = 1.1; d = 0.90; r = 0.12; T = 0.25; fu = 1; fu = 0. Thus, p   0.6523 , and
1.1  0.9
f = e–0.12×0.25 [0.6523×$1.0 + (1 – 0.6523) ×$0.0] = 0.633
Irrelevance of the Stock's Expected Return
The option–pricing formula, f = e–rT [p fu + (1 – p) fd], does not involve the probabilities of the stock price moving up or
down.
Although it is logical to assume that as the probability of an upward movement in the stock price increases, the value
of a call option on the stock increases. However, this is not the case. The reason is that the option is being valued in
relative and not absolute terms. Therefore, the probabilities of future up or down movements are already accounted for.

Assume that p is the probability of an up movement in the stock price and (1 – p) is then the probability of a down
movement. If the stock price moves up to S0u or down to S0 d , the payoff from the option is fu or f d .
The expected payoff from the option is [p fu + (1 – p) fd].
The expected stock price at time T, E(ST), is given by E(ST) = p×S0 u + (1 – p) S0 d = p× S0 (u – d) + S0 d
erT  d
When substituting p  in to the above equation, E(ST) = S0 e rT.
ud
This shows that the stock price grows at the risk–free rate.
Since in a risk–neutral world, individuals require no compensation for risk, the expected return on all securities is the
risk–free interest rate.
The above example demonstrates the risk– neutral valuation principle in option pricing.
The principle states that it is valid to assume the world is risk neutral when pricing options. In addition, the resulting
option prices are correct not just in a risk–neutral world, but in the real world as well.
The One–Step Binomial Example Revisited
To illustrate that risk–neutral valuation gives the same answer as no–arbitrage arguments, we consider the following
example once again.
Stock price is currently $20 and will move either up to $22 or down to $18
The option is a European call option with a strike price of $21
The risk–free interest rate is 12% per annum.
Let p be the probability of a rise in the stock price in a risk–neutral world.
1. In a risk neutral world, p must satisfy: 22 p  18(1  p )  20e0.120.25 . Thus p = 0.6523.
2. At the end of 3 months, the call option has 0.6523 probability of being worth 1 and a 0.3477 probability of being
worth zero. Its expected value is 0.6523 × 1 + 0.3477× 0 = 0.6523
3. The value of the option today is 0.6523 × e–0.12×0.25 = 0.633.
Thus, the risk–neutral valuation and the no arbitrage–arguments approach give the same answer.
Real World vs. Risk–Neutral World.
A position in a call option is riskier than a position in a stock. Thus, the discount rate to be applied to the payoff from a
call option is greater. Without knowing the option’s value, we do knot know how much greater it should be. This is
why the risk–neutral valuation solves this problem. In a risk–neutral world, the expected return on all assets (and thus
the discount rate to be applied to all expected payoffs) is the risk–free rate.
11.3 Two–Step Binomial Trees

Consider the following two step binomial tree. The goal of this analysis is to calculate the option price at the initial
node of the tree (which in this case is 1.283).
Assume we are given that u = 1.1; d = 0.90; r = 0.12; T = 0.25; S0 = 20. Further, the option under consideration has a
strike price of $21.
In the diagram above, the upper value at each node is the stock price, and the value below it is the option price. This is
true of all of Hull’s diagrams in Chapter 10 as well as the remaining diagrams in this summary.
To calculate the option value, the critical value to compute is the value of the option at node B, because at node C,
option is worth zero. Further, node C leads to either node E or node F and at both nodes the option price is zero. Since
it has been previously determined that p = 0.6523, the value of the option at node B is:
e–0.12×0.25 [0.6523×$3.2 + (1 – 0.6523) ×$0.0] = 2.0257.

Finally, given that the value of the option at node B is 2.0257 and that at node C it is zero, the value at node A is e–
0.12×0.25
[0.6523×$2.0257 + (1 – 0.6523) ×$0.0] = 1.2823.
The two step binomial tree option pricing process can be generalized as follows:
The stock price is initially S0.
24.200
3.200 = 24.2 - 21
22.000 B
2.057
20.000 A 19.800 = 20*1.10*.90
1.283 0.000 since 19.8 < 21
18.000 C
0.000 16.200 = 20*.90*.90
0.000 since 16.2 < 21
During each time step, it either moves up by a factor of u or down by a factor of d times its initial value.
The risk–free interest rate is r and the length of the time step is  t years.
By repeated application of f = e–rT [p fu + (1 – p) fd], the following equations result:
fu = e –r t [p fuu + (1 – p) fud].
fd = e –r t [p fud + (1 – p) fdd].
f = e –r t [p fu + (1 – p) fd] = e –2 r  t [p2 fuu + 2p(1 – p) fud (1 – p)2 fdd].
11.4 A Put Example

Consider a two–year European put with a strike price of $52 on a stock whose current price is $50.
Assume there are two time steps of one year, and in each time step the stock price either moves up by 20% or moves
down by 20%. Further assume that the risk–free interest rate is 5%.
Based on the above, u = 1.2; d = 0.80; r = 0.05; T = 2; S0 = 50.
To price the two–year European put option, begin with the computation of the risk–neutral probability, p.
e0.051  0.8
Using the data in the example, p   0.6282 .
1.2  0.8
The possible final stock prices are: $72 [50 × (1.2)2)], $48, and $32.
Note: See the diagram in the next section for additional information
Using the equations in section 10.3, fuu = 0; fud = 4; fdd = 20;
Finally, we compute the value of the put as follows:
f = e –2×0.05×1 [(0.6282)2 ×$0 + 2×(0.6282)×(1 – 0.6282) ×$4 + (1 – 0.6282)2 ×$20] = 4.1923.
11.5 American Options

In order to adjust the procedure described previously to account for American options, the value of the options at
intermediates nodes should be the greater of:
erT  d
1. The value determined by using f = e–rT [p fu + (1 – p) fd], where p 
ud
2. The payoff from early exercise.
72.000 = 50*1.10*1.1
0.000 since 72 > 52
60.000 B
1.4147
50.000 A 48.000 = 50*1.10*.90
5.0894 4.000 = 52 - 48
40.000 C
12.000 32.000 = 50*.90*.90
20.000 = 52 - 32
Notes:
1. At node C, the value of the option using the equation for f shown above is as 9.4636, however the payoff from early
exercise is 12 (52 – 40) and thus, the value of the option at node C is 12.00.
2. The value at the initial node is $5.0894.

11.6 Delta
The delta of a stock option is the ratio of the change in the price of the stock option to the change in the price of the
underlying stock.
Delta is the number of units of the stock that should be held in order to create a riskless hedge (the construction of a
riskless hedge is referred to as “delta hedging”).
Further, the delta of a call option is positive, while the delta of a put option is negative.
Computation of Delta:
f f
Based on the put example above,   u d  1.4147  12  0.529265 (at the end of the first step)
S uS d 60  40
0 0
04 4  20
  0.1667 or  1.000 (at the end of the second time step).
72  48 4832
In this 2–step example, we see that delta changes over time. Thus, to maintain a riskless hedge using an option and the
underlying stock, the holdings in the stock must be adjusted at the end of each step.
11.7 Matching Volatility with u and d

In this section, Hull presents a theoretical discussion of how the parameters u and d are chosen to match the volatility
of the stock price. Two main points are discussed.
First, Hull proves that the values of u and d are the same as those proposed by Cox, Ross, and Rubinstein for matching
u and d. Specifically, u  e t and d  e  t , where the volatility  of a stock price is defined so that  t is the
standard deviation of the return on the stock price in a short period of time of length t.
Secondly, Hull proves that we move from the real world to the risk–neutral world, the expected return on a stock
changes, but its volatility remains the same. Hull’s proof is an illustration of an important general result known as
Girsanov’s theorem. The parameters u and d are chosen to match the stock price volatility. This can be done by setting
ad
u  e t
, d  e  t
, a  e r t and p 
ud
Moving from one set of risk preferences to another is referred to as changing the measure. The real–world measure is
sometimes referred to as the P–measure, while the risk–neutral world measure is referred to as the Q–measure.
Example: Consider an American put option with a stock price of $50, a strike price of $52, a risk–free rate of 5%, a life
span of 2 years, and two time steps. In this case, t = 1. Suppose  = 30%.
1 1.053  0.07408
Then, u  e 0.31  1.3499 d   0.7408 , a  e 0.051  1.0513 , and p   0.5097
1.3499 1.3499  0.7408
The value of the put option is 7.43. This value is different from assuming that u = 1.2 and d = 0.8.
11.8 Increasing the Number of Steps

Suppose that there are five steps instead of two in the example we last considered. The parameters would be
t  2 / 5  0.4 , r  0.05, and   0.3. These values give
u  e0.3 0.4  1.2089, d  (1/1.2089)  0.8272, a  e0.050.4  1.0202, and
p  (1.0202  0.8272) / (1.2089  0.8272)  0.5056.
Using DerivaGern: DerivaGem, is a useful tool for becoming comfortable with binomial trees.
11.9 Options on Other Assets

Options on indices, currencies, and futures were discussed in Chapter 8 and are covered in more detail in Chapter 14.
One can construct and use binomial trees for these options in the same way as for options on stocks except that the
equations for p change.
Options on Stocks Paying a Continuous Dividend Yield
Consider the following:
 a stock pays a known dividend yield at rate q.
 the total return from dividends and capital gains in a risk–neutral world is r.
 the dividends provide a return of q.
 Capital gains provide a return of r – q.
 If the stock starts at S0 its expected value after one time step of length t must be S0e(r – q)t

(r – q)t e ( r  q ) t  d
This means that pS0u + (1 – p)S0 d = S0e so that p 
ud
As in the case of options on non–dividend–paying stocks, match volatility by setting u  e t and d  1/ u.
Options on Stock Indices
It is assumed that the valuation of an option on a stock index is similar to the valuation of an option on a stock paying a
known dividend yield.
Example: Consider the following:
A Two–step tree to value a European 6–month call option on an index when the index level is 810, strike price is 800,
risk–free rate is 5%, volatility is 20%, and dividend yield is 2%.
At each node:
Upper value = Underlying Asset Price
Lower value = Option Price
Strike price = 800
Discount factor per step = 0.9876
Time step, dt = 0.2500 years, 91.25 days
Growth factor per step, a = 1.0075
Probability of up move, p = 0.5126
Up step size, u = 1.1052
Down step size, d = 0.9048
In this case.
t  0.25, u  e0.20 0.25
 1.1052,
(0.05-0.02)0.25
d  1/ u  0.9048, ae  1.0075
p  (1.0075  0.9048) /(1.1052  0.9048)  0.5126
The value of the option is 53.39.
Options on Currencies
As discussed in Section 5.10, a foreign currency can be regarded as an asset providing a yield at the foreign risk–free
rate of interest, rf. By analogy with the stock index case we can construct a tree for options on a currency by using
ad ( r  r ) t
u  e t and d  1/ u and p  and setting a  e f
ud
 The Australian dollar is worth 0.6100 U.S. dollars and this exchange rate has a volatility of 12%.
 The Australian risk–free rate is 7% and the U.S. risk–free rate is 5%.
A three–step tree to value an American 3–month call option on a currency when the value of the currency is 0.6100,
strike price is 0.6000, risk–free rate is 5%, volatility is 12%, and foreign risk–free rate is 7%.
At each node:
Upper value = Underlying Asset Price
Lower value = Option Price
Strike price = 0.6
Discount factor per step = 0.9958
Growth factor per step, a = 0.9983
Probability of up move, p = 0.4673
Up step size, u = 1.0352
Down step size, d = 0.9660

In this case,
t  0.08333, u  e0.12 x 0.08333
 1.0352
(0.05-0.07)0.08333
d  1/ u  0.9660, a  e  0.9983
p  (0.9983  0.9660) / (1.0352  0.9660)  0.4673
Options on Futures
It costs nothing to take a long or a short position in a futures contract.
 Thus, in a risk–neutral world a futures price should have an expected growth rate of zero.
 If F0 is the initial futures price, the expected futures price at the end of one time step of length t should also equal
F0 .
1 d
This means that pF0u + (1 – p)F0 d = F0 so that p  and use
ud
ad
u  e t and d  1/ u and p  and setting a = 1
ud
 A futures price is currently 31 and has a volatility of 30%.
 The risk–free rate is 5%.
A three–step tree to value an American 9–month put option on a futures contract when the futures price is 31, strike
price is 30, risk–free rate is 5%, and volatility is 30%.
At each node: Upper value = Underlying Asset Price Lower value = Option Price
Strike price = 30 Discount factor per step = 0.9876
Growth factor per step, a = 1.000 Probability of up move, p = 0.4626
Up step size, u = 1.1618 Down step size, d = 0.8607
t  0.25, u  e0.3 0.25  1.1618

d  1/ u  1/1.1618  0.8607, a  1,
p  (1  0.8607) /(1.1618  0.8607)  0.4626

Chapter 12: Wiener Processes and Itô’s Lemma
Section Description
12.0 Introduction
12.1 The Markov Property
12.2 Continuous–Time Stochastic and Stock Processes
12.3 The Process for A Stock Price
12.4 The Parameters
12.5 Itô’s Lemma
12.6 The Lognormal Property
12.0 Introduction
A Stochastic Process:
A stochastic process is one in which the value of a variable changes over time in an uncertain way.
A stochastic process can be classified as discrete time or continuous time.
1. For a discrete–time stochastic process, the value of the variable can only change at certain fixed times.
2. For a continuous–time stochastic process, changes can take place at any time.
A stochastic variable can be classified as continuous variable or discrete variable.
1. For a continuous–variable process, the underlying variable can take any value within a certain range.
2. For a discrete–variable process, only certain discrete values are possible.
12.1 The Markov Property

A Markov Process:
A Markov process is a stochastic process where only the present value of a variable is relevant for predicting future
prices.
A Markov process describes the nature of stock prices.
Predictions are uncertain and must be expressed in terms of probability distributions.
The Markov Property:
The Markov property implies that the probability distribution of the price at any particular future time is not dependent
on the particular path followed by the price in the past.
The Markov property of stock prices is consistent with the weak form of market efficiency.
1. the present price of a stock impounds all the information contained in past prices.
2. the weak form of market efficiency appears to be true, since there is little evidence that technical analysts have
made above–average returns by interpreting historical price information.
3. competition in the marketplace that tends to ensure that weak–form market efficiency holds.
12.2 Continuous–Time Stochastic and Stock Processes

Example: Assume a variable follows a Markov stochastic process, has a current value of 10, and is subject to a change
in its value of  (0,1) during a one year period.
Note:  (0,1) indicates a standard normal probability distribution, with mean 0 and standard deviation 1.
Q. What is the probability distribution of the change in the value of the variable during two years?
A. Since the sum of two normal distributions is normal, with a mean and variance equal to the sum of the means and
variances, the change in the variable over two years is  (0, 2) .
In general, the change during any time period of length T is  (0, T ) , and the change during a very short time period
of length t is  (0, t ) .
“Uncertainty” is computed as t and is often referred to as being proportional to the square root of time.
1. A Wiener Process is a Markov stochastic process with a mean change of zero and a variance rate of 1.0 per year. It
has been used in physics to describe the motion of a particle that is subject to a large number of small molecular shocks
and is sometimes referred to as Brownian motion.
A variable z follows a Wiener process if it has the following two properties:
Property 1. The change z during a small period of time t is z   t , where  is a random drawing from a
standardized normal distribution,  (0,1) .
Property 2. The values of z for any two different short intervals of time, t, are independent.
Example: Assume a variable z follows a Wiener process, that its initial value is 25, and that time is measured in years.

At the end of year 1, the value of the variable is normally distributed with a mean of 25 and a standard deviation of 1.0.
At the end year 5, it is normally distributed with a mean of 25 and a standard deviation of 5  2.236
Based on the above, we see that the uncertainty about the value of the variable increases as the square root of how far
we are looking ahead.
Two noteworthy properties of Wiener processes, related to the t property, are:
The expected length of the path followed by z in any time interval is infinite.
The expected number of times z equals any particular value in any time interval is infinite.
2. A Generalized Wiener Process
A generalized Wiener process for a variable x can be defined in terms of dz as follows: dx = adt + bdz, where a and b
are constants.
The change in the value of x in any time interval T is normally distributed with:
mean of change in x = aT, standard deviation of change in x  b T , and variance of change in x = b2T
The generalized Wiener process has an expected drift rate (i.e., average drift per unit of time) of a and a variance rate
(i.e., variance per unit of time) of b2.
Example: Assume that the cash position of a company, initially 50 (in 000s), follows a generalized Wiener process
with a drift of 20 per year and a variance rate of 900 per year.
At the end of year 1, the cash position has a normal distribution with mean of 70 [50 + (20 × 1)] and a standard
deviation of 30 ( 900 ). At the end of 6 months, it has a normal distribution with a mean of 60 [50 + (20 × 0.50)] and
a standard deviation of 30  0.5  21.21
Notes:
The uncertainty about the cash position at some time in the future increases as the square root of how far ahead we are
looking.
The cash position can become negative (we can interpret this as a situation where the company is borrowing funds).
3. An Ito Process:
An Ito process is a generalized Wiener process where the parameters a and b are functions of the value of the
underlying variable, x, and time, t. An Ito process can be written dx = a (x, t) dt + b (x, t) dz. An Ito process assumes
that the drift and variance rate of x are equal to a(x, t) and b(x, t)2 during the time between t and t + t.
12.3 The Process for A Stock Price

Consider the stochastic process assumed for the price of a non–dividend–paying stock.
In actuality, stock prices do not follow a generalized Wiener process.
The problem: the model fails to capture a key aspect of stock prices, that the expected % return required by investors
from a stock is independent of the stock’s price.
Example: If investors require a 14% per annum expected return when the stock price is $10, then, they will also require
a 14% per annum expected return when it is $50.
Let S equal the stock price at time t, the expected drift rate in S equal to  S for some constant parameter, .
In a short interval of time, t, the expected increase in S is  St.
The parameter, , is the expected rate of return on the stock, expressed in decimal form.
The standard deviation of the change in a short period of time t should be proportional to the stock price and leads to
the model: dS   dt   dz . This is the most widely used model of stock price behavior.
S
Example: Assume a stock pays no dividends, has a volatility of 30% per annum, and provides an expected return of
15% per annum.
Thus,  = 0.15 and  = 0.30. The process for the stock price is dS  0.15dt  0.30dz
S
If S is the stock price at a particular time and S is the increase in the stock price in the next small interval of
time, S  0.15t  0.30 t , where  is a random drawing from a standardized normal distribution.
S
Consider a time interval of one week or 0.0192 year, and an initial stock price of $100. Recall
that S  0.15t  0.30 t . Then t = 0.0192, S = 100, and S  100(0.00288  0.0416 )  0.288  4.16
S
This shows that the price increase is a random drawing from a normal distribution with mean $0.288 and standard
deviation $4.16.

Discrete Time Model

The discrete–time version of the model of stock price behavior developed in the previous section is:
 S   S t   S  t
Monte Carlo Simulation
A Monte Carlo simulation of a stochastic process is a procedure for sampling random outcomes for the process.
It will be used to develop an understanding of the nature of the stock price process.
Example: the expected return from a stock is 14%, the standard deviation of the return is 20%
Let the t = 0.01 (0.01 year or 3.65 days): S  0.14  0.01S  0.2 0.01 S  0.0014 S  0.02 S
If a path for the stock price can be simulated by sampling repeatedly for  from  (0,1), then a set of results can be
determined.
Assume the initial stock price is $20. For the first period,  is sampled as 0.52.
Based on the above equation, the change during the first time period is: S  0.0014  20  0.02  20  0.52  0.236
At the beginning of the second time period the stock price is $20.236.  is sampled as 1.44.
The change during the second time period is S = 0.0014 × 20.236 + 0.02 × 20.236 × 1.44 = 0.611
The process is repeatedly performed.
Stock Price at Start of Period Random Sample for  Change in Stock Price During Period
20.000 0.52 0.236
20.236 1.44 0.611
20.847 –0.86 –0.329
20.518 1.46 0.628
By repeatedly simulating movements in the stock price, a complete probability distribution of the stock price at the end
of this time is obtained.
12.4 The Parameters

Since most investors require higher returns to take on more risk, the value of  (the expected continuously
compounded return earned by an investor per year) should depend on:
i. the risk of the return from the stock, and
ii. the level of interest rates in the economy.
However, the value of a derivative dependent on a stock is independent of .
In contrast, stock price volatility, , is extremely important, to the determination of derivatives.
As an approximation, the standard deviation of the proportional change in the stock price over a relatively long period
of time T is  T .
Thus, volatility can be interpreted as the standard deviation of the change in the stock price in 1 year.
12.5 Itô’s Lemma

The price of a stock option is a function of the underlying stock’s price and time. Let the value of a variable x follows
an Ito process: dx = a (x, t) dt + b (x, t) dz
The variable x has a drift and variance equal to a(x, t) and b(x, t)2 during the time between t and t +t.
Note that dS = S dt + S dz, with  and  constants, is a reasonable model of stock price movements.
Ito’s lemma shows that a function, G, of x and t, follows the process
 G G 1  2G 2  G 
dG   a  b  dt   b dz
 x t 2 x 2
 
 x 
2
G G 1  2G 2  G  2
It has a drift rate of a  b and a variance rate of dG    b
x t 2 x 2
 x 
Application to Forward Contracts
To illustrate Ito’s lemma, consider a forward contract on a non–dividend–paying stock.
F0  S0 erT , where F0 is the forward price at time zero, S is the spot price at time zero, and T is the time to maturity of
the forward contract.
What happens to the forward price as time passes? Define F as the forward price at a general time t and S as the stock
price at time t with (t < T).

The relationship between F and S is F  Ser (T t )

We can use Ito’s lemma to determine the process for F.
F 2 F F
 e r (T t ) ,  0,   rSe r (T t )
S S 2 t
Finally, dF  (   r ) Fdt   Fdz . Like the stock price S, the forward price follows a Brownian motion. F has an
expected growth rate of  – r, rather than . The growth rate of F is the excess return of S over the risk–free rate.
12.6 The Lognormal Property

In this section, Ito’s lemma is used to derive the process followed by lnS when S follows the process described by
dS   Sdt   Sdz . Define G = lnS
G 1  2G 1 G
Because  ,  2,  0 , it can be shown that the process followed by G is
S S S 2
S t
 2 
dG      dt   dz
 2 
G = lnS follows a generalized Wiener process as  and  are constant.
It has constant drift rate [ –  2/2] and constant variance rate  2.
Derivation of the process followed by G = lnS using Ito’s lemma
The change in lnS between time zero and some future time, T, is therefore normally distributed with mean
 2    2   


 
2 
 T and variance  2
T . Thus, ln ST  ln S 0     
 
  
2  
T  ,  T 
 or


  2  
ln ST     ln S0    
  2 
  
T,  T ;

 
  2  
ln ST    m, s  where  m  ln S0      T  , and s   T  .
  2  
where ST is the stock price at a future time, S0 is the stock price at time zero, and (m, s) denotes a normal distribution
with mean m and standard deviation s.
 2 
The change in lnS between time zero and a future time, T, is normally distributed with mean     T and variance
 2 
  2      2  
  2  
 

 2T . Thus, ln ST  ln S0        T  ,  T  or ln ST    ln S0      T ,  T  
  2 
  
 
  2  
ln ST    m, s  where  m  ln S0      T  , and s   T  .
  2  
The key results are as follows: lnST is normally distributed.
The model of stock price behavior just developed implies that a stock’s price at time T, given its price today, is
lognormally distributed.
The standard deviation of the logarithm of the stock price is  T , and this is proportional to the square root of how
far ahead we are looking.

Chapter 13: The Black-Scholes Model
Section Description
13.0 Introduction
13.1 Lognormal Property of Stock Prices
13.2 The Distribution of the Rate of Return
13.3 The Expected Return
13.4 Volatility
13.5 Concepts Underlying the Black–Scholes–Merton Differential Equation
13.6 Derivation of the Black–Scholes–Merton Differential Equation
13.8 Black–Scholes Pricing Formulas
13.9 Cumulative Normal Distribution Function
13.10 Warrants and Executive Stock Options
13.11 Implied Volatilities
13.12 Dividends
13.0 Introduction
Topics:
a. Derivation of the Black–Scholes (BS) model for valuing European call and put options on a non–dividend–paying
stock, and how the BS model deals with dividend–paying stocks.
b. Volatility either estimated from historical data or implied from option prices using the model.
c. Risk–neutral valuation argument introduced in Chapter 11 can be used.
d. Pricing of American call options on dividend–paying stocks is determined.
13.1 Lognormal Property of Stock Prices

The model of stock price behavior developed in Chapter 12 assumes that percentage changes in the stock price in a
short period of time are normally distributed. Define:
 : Expected return on stock  : Volatility of the stock price
The mean of the percentage change in time t is  t the standard deviation of this percentage change is a  t so
S
that   ( t ,  t ) , where S is the change in the stock price S in time t, and  ( m, s ) denotes a normal
S
distribution with mean m and standard deviation s. From Section 12.6, it was shown that
  2  
ln ST   ln S0     T , T  , where ST is the stock price at a future time T, S0 is the stock price at time 0.
  2  
The equation above shows that lnST is normally distributed, thus ST has a lognormal distribution.
Example: Consider a stock with an initial price of $40, an expected return of 16% per annum, and a volatility of 20%
per annum. The time period is 6 months. Thus, S0 = 40;  = 0.16;  = 0.2; T = 0.50.
The probability distribution of the stock price, ST, in 6 month is:
  
0.22 
ln ST   ln 40  0.16  0.50, 0.2 0.50  , and ln ST   (3.759, 0.141) . A 95% confidence interval is computed
  2  
  
as: (3.759 – 1.96 × 0.141) < lnST < (3.759 + 1.96 × 0.141)
Thus, 32.55 < ST < 56.56 (i.e. there is a 95% probability that the stock price in six months will lie between 32.55 and
56.56).
Expected value and Variance:
The expected value of ST, E(ST), is given by E ( ST )  S0 e T .
The variance of ST, Var (ST), is given by Var ( ST )  S02 e 2 T [e T  1]
2
Example: Consider a stock with a current price of $20, an expected return of 20% per annum, and volatility of 40% per
annum. Thus, S0 = 20;  = 0.20;  = 0.40; T = 1.0 Then:
E ( ST )  20e0.21  24.43 , Var ( ST )  400e 20.21[e0.4
2 1
 1]  103.54 , and the   103.54  10.18

13.2 The Distribution of the Rate of Return

Define the continuously compounded rate of return per annum as x
It follows that ST  S0 e xT , so that x 

1 ST
ln
 2 
and it can be shown that x      ,


T S0  2 T 
 
Example: Consider a stock with an expected return of 17% per annum and a volatility of 20% per annum.
Thus,   0.17,   0.20, T  3 .
The probability distribution for the actual rate of return (continuously compounded) realized over three years is normal
 2 
with mean and standard deviation of: x    0.17 0.2 , 0.2   (0.15,0.1155)
 2 3 

Based on the above, there is a 95% chance that the actual return realized over three years lies between
–7.6% [0.15 – 1.96 × 0.1155] and +37.6% [0.15 + 1.96 × 0.1155] per annum.
13.3 The Expected Return

What Is the Expected Rate of Return?
The term t is the expected return provided by S in a very short period of time, t. However,  is not the expected
2
continuously compounded return over a long period of time. Rather, it is   per year.
2
Example: Assume the following is a sequence of returns per annum on a stock, measured using annual compounding:
15%, 20%, 30%, –20%, 25%
The expected return in one year is 0.150.20 0.30 0.20 0.25  0.14
5
However, an investor would actually earn less than 14% per annum by leaving the money invested in the stock for five
years. The estimated 5 year return with annual compounding is compute as follows:
1.15× 1.20 ×1.30 × 0.80 × 1.25 = 1.7940, and the annual return is (1.7940)1/ 5  0.124
Thus, the term expected return is ambiguous. However, throughout the rest of the chapter it will refer to  t unless
otherwise stated.
13.4 Volatility
When T is small,  T is the standard deviation of the proportional change in the stock price in time T .
Example: Assume  = 0.30 (30% per annum) and the current stock price is $50.
 The standard deviation of the proportional change in the stock price in one week is 0.30  1  0.0416
52
 A one standard deviation move in the stock price in one week is $50 × 0.0416 = $2.08.
 The standard deviation of the stock price in four weeks is ~ twice the standard deviation in one week.
Estimating Volatility from Historical Data
To estimate the volatility of a stock price empirically, we first define the following: Define:
n + 1: Number of observations
Si: Stock price at end of ith interval (i = 0, 1,…, n)
: Length of time interval in years
 Si 
Let, ui  ln   .
 Si 1 
The standard deviation of the ui is s =  ;  can be estimated as * where * s with a standard error of
*
 2n
Note: Empirical research indicates that time should be measured in trading days (days when the exchange is closed
should be ignored).
Example (see the data in table 13.1 on page 287). Table 13.1 shows a possible sequence of stock prices during 21
consecutive trading days. In this situation, 
ui  0.09531 and 
ui2  0.00326
2
 an estimate of the standard deviation of the daily return is 0.00326  0.09351  0.01216
19 380

 an estimate for the volatility per annum is  *  0.01216  0.193 (based on 252 trading days).
1/252
 the standard error of this estimate is  *  0.193  0.031 per annum.
2n 220
This analysis can be adapted to accommodate dividend–paying stocks:
 S D 
The return during a time interval that includes an ex–dividend day is given by ui  ln  i  , where D is the amount
 Si 1 
 Si 
of the dividend. The return in other time intervals is ui  ln  
 Si 1 
Trading Days vs. Calendar Days
Practitioners tend to ignore days when the exchange is closed when estimating volatility from historical data and when
calculating the life of an option. The volatility per annum is calculated from the volatility per trading day using the
formula:
volatility per annum = volatility per trading day  number of trading days per annum .
It is assumed that there are 252 trading days for stocks per year.
The life of an option is also usually measured using trading days rather than calendar days. It is calculated as T years,
trading days until option maturity
where T 
252
13.5 Concepts Underlying the Black–Scholes–Merton Differential Equation

Black–Scholes–Merton (BSM):
The BSM differential equation must be satisfied by the price of any derivative dependent on a non–dividend–paying
stock. The equation is derived in the next section.
The BSM analysis is analogous to the no–arbitrage analysis.
A riskless portfolio consisting of a position in the option and a position in the underlying stock is created.
In this situation, the return from the portfolio must be the risk–free interest rate, r.
Example: Assume that at a point in time the relationship between a small change in the stock price, S and the
resultant small change in the price of a European call option, c, is given by c = 0.40S.
This means that the slope of the line representing the relationship between c and S is 0.4.
A riskless portfolio would consist of:
1. A long position in 0.4 shares
2. A short position in one call option
The important difference between the BSM analysis and the analysis using a binomial model in Chapter 10 is that
Under BSM, the position in the stock and the derivative is riskless for only a very short period of time. To remain
riskless, it must be rebalanced frequently.
For example, the relationship between c and S might change from c = 0.40S today to c = 0.40S in two weeks.
To maintain the riskless position, an extra 0.1 shares would have to be purchased for each call option sold. The return
from the riskless portfolio in any short period of time must be r. This is the key element in the Black–Scholes analysis
and leads to their pricing formulas.
The assumptions used to derive the Black–Scholes–Merton differential equation are as follows:
1. Stock price follows the process developed in chapter 12 with  and  constant.
2. Short selling of securities with full use of proceeds is permitted.
3. No transactions costs or taxes. All securities are perfectly divisible.
4. No dividends during the life of the derivative.
5. No riskless arbitrage opportunities.
6. Security trading is continuous.
7. The risk–free rate of interest, r, is constant.

13.6 Derivation of the Black–Scholes–Merton Differential Equation

Let f be the price of a call option or other derivative of S. Here, f is a function of S and t.
The BSM differential equation is

f
t
 rS  
f
S
1
2
 2 f 
  2 S 2  2   rf
 S 
 It has many solutions, corresponding to different derivatives.
 A particular derivative is obtained when the solution depends on a boundary conditions.
Examples:
For a European call option, the key boundary condition is f = max(S – K, 0) when t = T
For a European put option, the key boundary condition is f = max(K – S, 0) when t = T

Risk–neutral valuation was introduced with the binomial model in chapter 11. It is the single most important tool for
the analysis of derivatives. It arises from one key property of the BSM differential equation. This property is that the
equation does not involve any variable that is affected by the risk preferences of investors.
The variables that do appear in the equation are the current stock price, time, stock price volatility, and the risk–free
rate of interest. All are independent of risk preferences.
The assumption that the world is risk neutral simplifies the analysis of derivatives.
Example: Consider a derivative that provides a payoff at one particular time. It can be valued using risk–neutral
valuation by using the following procedure:
1. Assume that the expected return from the underlying asset is the risk–free interest rate, r (assume  = r).
2. Calculate the expected payoff from the option at its maturity.
3. Discount the expected payoff at the risk–free interest rate.
Notes:
The risk–neutral valuation is an artificial device for obtaining solutions to the BSM differential equation.
The solutions that are obtained are valid in all worlds (not just those where investors are risk neutral).
Moving from a risk neutral world to a risk–averse world two things happen (which always offset each other).
1. The expected growth rate in the stock price changes.
2. The discount rate that must be used for any payoffs from the derivative changes.
13.8 Black–Scholes Pricing Formulas

The BSM formulas for prices at time zero of a European call option and put option on a non–dividend paying stock are
ln( S0 / K )  ( r  2 / 2)T
as follows: c = S0N(d1) – Ke–rT N(d2); p = Ke–rT N(–d2) – S0N(–d1); d1  ;
 T
ln( S0 / K )  ( r  2 / 2)T
d2   d1   T ; N(x) cumulative normal distribution function.
 T
Properties of the BS formulas:
Consider what happens when some of the parameters take extreme values.
a. When the stock price, S0, becomes very large, a call option is likely to be exercised.
i. This is similar to a forward contract with delivery price K. The call price should be c = S0 – Ke-rT
ii. When S0 becomes large, d1, and d2 become very large, and N(d1) and N(d2) are both close to 1.0.
Thus, the value of a forward contract with delivery price K can be replicated by the BSM formula above.
b. When the stock price becomes large, the price of a European put option, p, approaches zero. This is also consistent
with p = Ke–rT N(–d2) – S0N(–d1) because N(–d1) and N(–d2) are both close to zero.
c. When the volatility approaches zero, the stock will grow to S0e rT
The payoff from the call option is max(S0 erT – K, 0),and its present value is max(S0 – Ke–rT, 0)
i. Consider when S0 > Ke–rT. Therefore ln(S0 /K) + r T > 0.
As  approaches zero, d1 and d2 tend to +, and N(d1) and N(d2) are both close to 1.0. Thus,
c = S0N(d1) – Ke–rT N(d2) becomes c = S0 – Ke–rT
ii. Consider when S0 – Ke–rT. Therefore ln(S0 /K) + r T < 0.
As  approaches zero, d1 and d2 tend to –, and N(d1)and N(d2) are both close to 0.
Thus, c = S0N(d1) – Ke–rT N(d2) gives a call price of zero. The call price is therefore always max(S0 erT – K, 0) as  tends
to zero. Further, the put price is always max(K e–rT – S0, 0) as  tends to zero.

13.9 Cumulative Normal Distribution Function

Example: Assume a stock’s price six months from the expiration of an option is $42, the exercise price of the option is
$40, the risk–free interest rate is 10% per annum, and the volatility is 20% per annum. Thus,
S0 = 42; K = 40;  = 0.20; T = 0.5; r = 0.10
ln( S0 / K )  (r  2 / 2)T ln(42 / 40)  (0.1 0.22 / 2)0.50
d1    0.7693
 T 0.2 0.5
ln( S0 / K )  (r  2 / 2)T
d2   d1   T  0.7693  0.2  0.5  0.6278
 T
K e–rT = 40 e–0.05 = 38.049. For a European call, its value, c, is given by: c = 42N(0.7693) – 38.049N(0.6278)
For a European put, its value, p, is given by: p = 38.049N(–0.6278) – 42N(–0.7693)
Using the polynomial approximation: N(0.7693) = 0.7791 N(–0.7693) = 0.2209
N(0.6278) = 0.7349 N(–0.6278) = 0.2651. Thus, c = 4.76; p = 0.81
Results:
The stock price has to rise by $2.76 for the purchaser of the call to break even.
The stock price has to fall by $2.81 for the purchaser of the put to break even.
13.10 Warrants and Executive Stock Options

The BS formula can be used to value European warrants issued by a company. Consider a company with N outstanding
shares and M outstanding European warrants.
Each warrant entitles the holder to purchase M shares from the company at time T at a price of K per share.
Let: NST be the value of the company at time T. If the warrant holders exercise their options:
i. the company receives a cash inflow from the payment of the exercise price of MK
ii. the value of the company’s equity increases to NST + MK.
NST  MK
The share price immediately after exercise is .
NM
 NS  MK  N
The payoff to the warrant holder if the warrant is exercised is  T K  ST  K  . Thus, the value of
 NM  NM
N
each option is regular call options.
NM
Example:
 A company with 1 million shares worth $40 each is considering issuing 200,000 warrants giving each holder the
right to buy one share with a strike price of $60 in 5 years.
 The interest rate is 3% per annum, and the volatility is 30% per annum. The company pays no dividends.
 The value of a 5–year European call option on the stock is $7.04.
What is the cost of issues the warrants?
N = 1,000,000 and M = 200,000, so that the value of each warrant is
1,000,000/(1,000,000 + 200,000) × $7.04 = $5.87.
The total cost of the warrant issue is 200,000 × $5.87 = $1.17 million.
Assuming the market perceives no benefits from the warrant issue, expect the stock price to decline by $1.17 to
$38.83.
13.11 Implied Volatilities

Traders usually work with implied volatilities (i.e. volatilities implied by option prices observed in the market).
How implied volatilities are calculated:
Suppose the value of a call option on a non–dividend–paying stock is 1.875 when S0 = 21, K = 20, r = 0.01, and T =
0.25. The implied volatility is the value of  that, when substituted into c = S0N(d1) – Ke–rT N(d2), gives c = 1.875.
An iterative procedure can be used to find the implied . For example:
Begin by trying  = 0.20. This produces a c value of 1.76 which is too low.
Next,  = 0.30 is tried. This produces a c equal of 2.10, which is too high.
Repeating the process eventually produces the correct value of .
Finally, note that implied volatilities are used to monitor the market’s opinion about the volatility of a particular stock.

13.12 Dividends
The BSM can be modified to take account of dividends. It is assumed that the amount and timing of the dividends can
be predicted with certainty.
A dividend–paying stock can be expected to follow the stochastic process developed in Chapter 11 except when the
stock goes ex–dividend. At this point, the stock’s price goes down by an amount reflecting the dividend paid per share.
Notes:
 For tax reasons, the stock price may decrease by a smaller amount than the cash amount of the dividend.
 “Dividend” should be interpreted as the reduction in the stock price on the ex–dividend date caused by the
dividend.
Example: If a dividend of $1 per share is anticipated, and the share decreases in price by 80% of the dividend on the
ex–dividend date, the dividend should be assumed to be $0.80.
European Options:
European options can be analyzed by assuming that the stock price is the sum of 2 components:
1. a riskless component (the present value of the dividends during the life of the option discounted from the ex–
dividend dates to the present at the risk–free rate).
2. a risky component.
The BS formula is correct if S0 is equal to the risky component of the stock price and , is the volatility of the process
followed by the risky component.
European options can be used with the BSM formula provided that the stock price is reduced by the present value of
all the dividends during the life of the option.
Example: Consider a European call option on a stock paying dividends of $0.50 and having ex–dividend dates in 2
months and 5 months. Let S0 = 40; K = 40;  = 0.30; T = 0.5; r = 0.09.
The present value of the dividends is 0.50e 0.16670.09  0.50e 0.41670.09  0.9741
The option price using the BSM and the following inputs [S0 = 40 – 0.9741 = 39.0259; K = 40;  = 0.30; T = 0.5; r =
0.09 is shown below:
ln(39.0259/40)(0.09 0.32 /2)0.50
d1   0.2017 ; d2  d1   T  0.2017  0.3  0.5  0.0104
0.3 0.5
N(d1) = 0.5800; N(d2) = 0.4959
Using the equation c = S0N(d1) – Ke–rT N(d2), the call price is 39.0259 × 0.5800 – 40e–0.09×0.5 × 0.4959 = 3.67
American Options
In section 9.5, Hull states that in the absence of dividends, American options should never be exercised early.
In this section, Hull claims that when stocks pay dividends, it is optimal to exercise only at a time immediately before
the stock goes ex dividend. For example, assume that:
i. n ex–dividend dates are anticipated and that t1, t2, …tn, are moments in time immediately prior to the stock going
ex–dividend, with t1 < t2 < t3 < ... < tn.
ii. the dividends corresponding to these times are D1, D2,…, Dn respectively.
Consider the early exercise of an American call option just prior to the final ex–dividend date.
i. If the option is exercised at tn, the investor receives S(Tn) – K
ii. If the option is not exercised at tn, the stock price drops to S(Tn) – Dn
Hull goes on to show that:
 r (T  tn )
i. If Dn  K (1  e ) , it is not optimal to exercise at tn.
 r (T  tn )
ii. If Dn  K (1  e ) , it is optimal to exercise at tn.
 r ( ti 1 ti )
iii. If Di  K (1  e ) , for any i < n , it is not optimal to exercise at ti.
Hull’s final concluding statements in this section are as follows:
i. The only time that needs to be considered for the early exercise of an American call is the final ex–dividend date,
tn.
 r ( t t )  r ( T  tn )
ii. If the inequality Di  K (1  e i1 i ) holds for i = 1, 2, ... , n – 1 and inequality Dn  K (1  e ) holds, then
early exercise is never optimal.
Black’s Approximation
Black’s approximation procedure for taking account of early exercise in call options involves:
i. Calculating the prices of European options that mature at times T and tn, and then
ii. Setting the American price equal to the greater of the two.

Example: Consider the prior example of a European call option on a stock paying $0.50 and having ex–dividend dates
in 2 months and 5 months. Thus, S0  40, K  40,   0.30, T  0.5, r  0.09
Suppose the option is American rather than European. In this case, D1 = D2 = 40; S0 = 40; K = 40;  = 0.20; t1 = 1/12;
and t2 = 5/12; r = 0.10.
Between t1 and t2, K (1  e  r (t2 t1 ) )  40(1  e 0.09(0.25) )  0.89 which is greater than 0.50 and thus, the option should
not be exercised immediately before the first ex–dividend date.
Between T and t2, K (1  e  r (T t2 ) )  40(1  e 0.09(6 /125 /12) )  0.30 which is less than 0.50 and thus, the option should
be exercised immediately before the second ex–dividend date.
Now use Black’s approximation to value the option.
i. The present value of the first dividend is 0.50e 0.16670.09  0.4926
ii. The option price using the BSM and the following inputs
[S0 = 40 – 0.4926 = 39.5074; K = 40;  = 0.30; T = 5/12 = 0.41667; r = 0.09, is $3.52.
iii. Black’s approximation involves taking the greater of $3.52 and the value of the option when it can only be
exercised at the end of six months ($3.67). $3.67 is the value of the American call option.


Chapter 14: Options on Stock Indices, Currencies, and Futures
Section Description
14.1 Results for a Stock Paying a Known Dividend
14.2 Option Pricing Formulas
14.3 Options on Stock Indices
14.4 Currency Options
14.5 Futures Options
14.6 Valuation of Futures Options using Binomial Trees
14.7 Futures Price Analogy
14.8 Black’s Model for Valuing Futures Options
14.9 Futures Options vs. Spot Options
14.1 Results for a Stock Paying a Known Dividend

Consider the difference between a stock that pays no dividends and a stock that pays a dividend yield at a rate q per
annum. Hull argues that we get the same probability distribution for the stock price at time T in each of the following
two cases:
1. The stock starts at price S0 and pays a dividend yield at rate q.
2. The stock starts at price S0e–qT and pays no dividend yield.
Hull’s rule: When valuing a European option lasting for time T on a stock paying a known dividend yield at rate q,
reduce the current stock price from S0 to S0e–qT and then value the option as though the stock pays no dividends.
Application of Hull’s rule for establishing the Lower Bounds for Option Prices and Put–Call Parity
Consider the problem of determining bounds for the price of a European option on a stock providing a dividend yield
equal to q.
i. By substituting S0e–qT for S0 in c  max( S0  Ke rT ,0) , the lower bound for the European call option price, c, is
c  max( S0 e qT  Ke rT ,0)
ii. To obtain a lower bound for a European put option, replace S0 by S0e–qT in p  max( Ke rT  S0 ,0) to get
p  max( Ke rT  S0 e qT ,0)
iii. By substituting S0 by S0e–qT in p  S0  c  Ke rT , we obtain put–call parity for a stock providing a dividend yield
equal to q: p  S0 e qT  c  Ke rT
14.2 Option Pricing Formulas

By replacing S0 by S0e–qT in the BS formulas, c  S0 N ( d1 )  Ke  rT N ( d 2 ) , p  Ke rT N (  d 2 )  S0 N (  d1 ) , we obtain:
i. the price c of a European call a stock providing a dividend yield at rate q as c  S0 e qT N ( d1 )  Ke rT N ( d2 )
ii. the price p of a European put on a stock providing a dividend yield at rate q as
p  Ke rT N ( d2 )  S0 e qT N ( d1 )
iii. Note that the formulas for d1 and d2 two are changed to:
ln( S0 / K )( r  q  2 /2)T ln( S0 / K )( r  q  2 /2)T
d1  d1   d1   T
 T  T
Risk–Neutral Valuation
In a risk–neutral world, the total return on a stock must be r.
i. Since the dividends provide a return of q, the expected growth rate in the stock price must be r – q.
ii. The risk–neutral process for the stock price must be dS  ( r  q) Sdt   Sdz
To value a derivative dependent on a stock that provides a dividend yield equal to q:
i. Set the expected growth rate of the stock equal to r – q and
ii. Discount the expected payoff at rate r.
Example: To value a European call option, note that the expected stock price at time T is S0 e( r  q )T
i. The expected payoff from a call option is computed as e( r  q )T S0 N ( d1 )  KN ( d 2 )
ii. Discounting the above at rate r for time T leads to c  S0 e qT N ( d1 )  Ke rT N ( d 2 )

Binomial Trees
Here, Hull examines the effect of a dividend yield equal to q on the binomial model in Chapter 11.
i. Once again, the total return provided by a stock in a risk–neutral world must be r.
ii. Since dividends provide a return equal to q, the return in the form of capital gains must be r – q.
Thus p, the probability of an up movement in a risk–neutral world must satisfy pSu  (1  p ) Sd  e( r q ) t or
e( r  q ) t  d
p
ud
14.3 Options on Stock Indices

Options on stock indices trade in both the over–the–counter and exchange–traded markets.
i. One index option contact is on 100 times the index.
ii. Index options are settled in cash. On exercise of the option:
a. The holder of a call option receives (S – K) × 100 in cash, where S is the value of the index at the close of
trading on the day of the exercise and K is the strike price.
The holder of a put option receives (K – S) × 100 in cash.
b. The writer of the call or put option pays (S – K) × 100 or (K – S) × 100 in cash respectively.
Valuation
In valuing index futures, we assumed that the index could be treated as a security paying a known dividend yield. In
valuing index options, similar assumptions are made as follows
i. c  max( S0 e qT  Ke rT ,0) , and p  max( Ke rT  S0 e qT ,0) , provide lower bounds for European index
options.
ii. p  S0e qT  c  Ke rT is the put–call parity result for European index options
iii. c  S0 e qT N ( d1 )  Ke rT N ( d 2 ) and p  Ke rT N ( d 2 )  S0 e qT N (  d1 )
In all cases:
i. S0 is equal to the value of the index
ii.  is equal to the volatility of the index, and
iii. q is equal to the average annualized dividend yield on the index during the life of the option.
Example: Consider a European call option on the S&P 500 that is two months from maturity.
 The current value of the index is 930, the exercise price is 900, the risk–free interest rate is 8% per annum, and the
volatility of the index is 20% per annum.
 Dividend yields of 0.2% and 0.3% are expected in the first month and the second month, respectively.
Thus, S0 = 930, K = 900, r = 0.08,  = 0.2, and T = 2/12.
Compute the cost of one European call option on the S&P 500 index.
Step 1: Compute the value of q:
The total dividend yield during the option's life is 0.2 + 0.3 = 0.5%. This is 3% per annum. Hence, q = 0.03.
Step 2: Compute the value of N(d1) and N(d2):
ln(930 / 900)  (0.08  0.03  0.22 / 2)  2 /12
d1   0.5444
0.2 2 /12
d 2  0.5444  0.2 2 /12  0.4628; N ( d1 )  N (0.5444)  0.7069, N ( d 2 )  N (0.4628)  0.6782
Step 3: Using the equation at the top of the page, Compute the cost of one European call option on the S&P 500 index.
Since c  S e qT N ( d )  Ke rT N ( d ) , c  930  0.7069e0.03  2 / 12  900  0.6782e0.08  2 / 12  51.83
0 1 2
Therefore, one contract would cost $51.83 × 100 = $5,183.
Portfolio Insurance
Portfolio managers use index options to limit downside risk.
Suppose that the value of an index today is S0. Since each contract on the S&P 500 is on 100 times the index, the value
of the portfolio is protected against the possibility of the index falling below K if, for each 100S0 dollars in the
portfolio, the manager buys one put option contract with strike price K.
Example: Suppose a portfolio is worth $500,000 and the value of the index is 1,000. To protect against the value of the
portfolio dropping below $450,000 in the next three months, 5 put option contracts with a strike price of 900 can be
purchased.

If the index drops to 880 in three months:

i. The portfolio will be worth $440,000 (5 × 100 × $880), but
ii. The payoff from the options will be 5 × (900 – 880) × 100 = $10,000, bringing the total value of the portfolio up to
the insured value of $450,000.
When the Portfolio's Beta Is Not 1.0
In the example above, it was assumed that the portfolio was well diversified and had a beta of 1.0.
If the portfolio's returns are not expected to equal those of an index, the CAPM can be used.
Under CAPM, the expected excess return of a portfolio over the risk–free interest rate equals beta times the excess
return of a market index over the risk–free interest rate.
Example: Assume you are given the following information:
 A $500,000 portfolio has a beta of 2.0 (see line 7 below).
 The current risk–free interest rate is 12% per annum.
 The dividend yield on both the portfolio and the index is expected to be 4% per annum.
 The S&P 500 index is currently 1,000.
 If the value of the index in the next 3 months proves to be 1,040, what is the value of the portfolio?
1. Value of index in three months 1,040
2. Return from change in index 40/1,000, or 4% per three months
3. Dividends from index 0.25 × 4 = 1% per three months
4 = 2. + 3. 4. Total return from index 4 + 1 = 5% per three months
5. Risk–free interest rate 0.25 × 12 = 3% per three months
6 = 4. – 5. 6. Excess return of index over risk–free interest rate 5 – 3 = 2% per three months
7. Excess return of portfolio over risk–free interest rate 2.0 × 2 = 4% per three months
8 = 5. + 7. 8. Return from portfolio 3 + 4 = 7% per three months
9. Dividends from portfolio 0.25 × 4 = 1% per three months
10 = 8. – 9. 10. Increase in value of portfolio 7 – 1 = 6% per three months
11. Value of portfolio 5000,000 × 1.06 = $530,000
Relationship between value of index and value of portfolio for beta = 2.0
Value of index Value of portfolio
in three months in three months ($)
1,080 570,000
1,040 530,000
1,000 490,000
960 450,000
920 410,000
880 370,000
Suppose a portfolio manager is looking to ensure that the value of the portfolio does not drop below $450,000.
Based on the table above, the appropriate strike price for the put options purchased is 960.
Since 100S0 = $100,000 and beta = 2.0, two put contracts are required for each $100,000 in the portfolio. Since the
portfolio is worth $500,000, 10 contracts should be purchased.
Note: If the value of the index falls to 880, the table above indicates that the value of the portfolio is 370,000.
However, the put options pay off (960 – 880) × 10 × 100 = $80,000, which restores the portfolio to the insured level.
Note: There are two reasons why the cost of hedging increases as the beta of a portfolio increases:
1. more put options are required, and
2. the options have a higher strike price.
14.4 Currency Options

Like options on stock indices, European and American options on foreign currencies are actively traded in both the
over–the–counter and exchange–traded markets. To hedge a foreign exchange exposure, foreign currency options are
an alternative to forward contracts.

i. A company knowing it will receive sterling at a certain time in the future can hedge its risk by buying put options
on sterling that mature at that time. The benefits to a company using this strategy are:
a. a guaranty that the value of the sterling will not be less than the exercise price
b. rendering any favorable exchange–rate movements to the company.
ii. A company knowing that it will pay sterling at a certain time in the future can hedge by buying calls on sterling
that mature at that time. The benefits to a company employing this strategy are:
a. a guaranty that the cost of the sterling will not exceed a certain amount
b. rendering favorable exchange–rate movements to the company.
Q. What is the difference when a forward contract is used to hedge vs. an option approach to hedge?
A1. A forward contract locks in the exchange rate for a future transaction, while an option provides a type of
insurance.
A2. A forward transaction costs nothing to enter into. An options require a premium to be paid up front.
Valuation
Define S0 as the spot exchange rate (the value of one unit of the foreign currency measured in the domestic currency).
rf is the foreign risk–free interest rate, and is the yield received by the owner of foreign currency.
Therefore,
i. the European call price, c, and put price, p, are therefore given by
r T r T
c  S0 e f N ( d1 )  Ke rT N ( d 2 ); p  Ke rT N (  d 2 )  S0 e f N (  d1 )
where S0 is the value of the exchange rate at time zero, and
In( S0 / K )  ( r  rf   2 / 2)T In( S0 / K )  ( r  rf   2 / 2)T
d1  ; d2   d1   T
 T  T
ii. both the domestic interest rate, r, and the foreign interest rate, rf , are the rates for maturity T.
Note: Put and call options on a currency are symmetrical in that
i. a put option to sell XA units of currency A for XB units of currency B is the same as
ii. a call option to buy XB units of currency B for XA units of currency A.
Example. Consider a 4 month European call option on the British pound.
i. Suppose the current exchange rate is 1.600, the strike price 1.600, and the risk free rate in the U.S. is 8% per
annum, the risk–free interest rate in Britain is 11 % per annum, and the option price is 4.3 cents.
ii. Thus, S0 = 1.6, K = 1.6, r = 0.08, rf = 0.11, T = 4/12, and c = 0.043.
Calculate the implied volatility.
Note: The implied volatility can be calculated iteratively.
i. A volatility of 20% gives an option price of 0.0639
ii. A volatility of 10% gives an option price of 0.0285.
iii. The iterative procedure ultimately results in an implied volatility of 14.1%.
Valuation in terms of the forward price:

( r  r )T
The forward rate, F0, for a maturity T is given as F0  S0 e f
Knowing this, The BSM formulas can be simplified to
c  e rT [ F0 N ( d1 )  KN ( d 2 )]; and p  e rT [ KN (  d 2 )  F0 N (  d1 )] where
In( F0 / K )   2T / 2 In( F0 / K )   2T / 2
d1  , and d 2   d1   T
 T  T
Note: For the above call and put equations to apply, the maturities of the forward contract and the option must be the
same.
14.5 Futures Options

Options on futures contracts (futures options) trade on different exchanges.
i. Are American–style options and require the delivery of an underlying futures contract when exercised.
ii. If a call futures option is exercised, the holder acquires a long position in the underlying futures contract + cash
equal to the most recent settlement futures price – the strike price.
iii. If a put futures option is exercised, the holder acquires a short position in the underlying futures contract + cash
equal to the strike price – the most recent settlement futures price.
iv. the effective payoff from a call futures option = the futures price (at the time of exercise) – the strike price; the
effective payoff from a put futures option = the strike price – the futures price at the time of exercise.

Example: Suppose it is 8/15 and an investor has 1 Sept. futures call option contract on copper with a strike price of 70
cents/ pound. 1 futures contract is on 25,000 pounds of copper.
Assume that the futures price of copper for delivery in Sept is 81 cents, and at the close of 8/14 trading it was 80 cents.
If the option is exercised, the investor receives:
Cash = 25,000 × (80 – 70) cents = $2,500 + a long position to buy 25,000 pounds of copper in Sept.
i. If the position in the futures contract was closed out immediately, the investor would have a $2,500 cash payoff +
25,000 × (81 – 80) cents = $250
ii. The total payoff from exercising the option on 8/15 = $2,750 = 25,000(F – K), where F is the futures price at the
time of exercise and K is the strike price.
Example: An investor has one Dec. futures put option on corn with a strike price of 200 cents per bushel.
One futures contract is on 5,000 bushels of corn. Assume the current futures price of corn for delivery in Dec. = 180
cents, and the most recent settlement price is 179 cents.
If the option is exercised, the investor receives 5,000 × (200 – 179) cents = $1,050 + a short position in a futures
contract to sell 5,000 bushels of corn in Dec.
If the position in the contract is closed, the investor has 1,050 cash payoff – 5,000 × (180 – 179) cents (i.e. $50). The
net payoff from exercise = $1,000 = 5,000(K – F), where F is the futures price at the time of exercise and K is the
strike price.
Quotes
Most futures options are American, and are referred to by the month in which the underlying futures contract matures
(not by the expiration month of the option).
The maturity date of a futures option contract is usually on (or a few days before), the earliest delivery date of the
underlying futures contract.
Options on Interest Rate Futures
The most actively traded futures options in the U.S. are those on T–bond futures, T–note futures, and Eurodollar
futures.
i. A T–bond futures option is an option to enter a T–bond futures contract. one T–bond futures contract is for the
delivery of Treasury bonds with a $100,000 face value. The price of a T–bond futures option is quoted as a percent
of the face value of the underlying T–bonds to the nearest 1/64 of 1%.
The table below gives the price of the April call futures option on Treasury bonds as 2–11 (2 11/64%) of the debt
principal when the strike price is 104 (implying that one contract costs $2,171.87). Quotes for options on T–notes
are similar.
INTEREST RATE: T–Bonds (CBT)

$100,000; points– 64ths of 100%
STRIKE CALLS–SETTLE PUTS–SETTLE
PRICE Apr May Jun Apr May Jun
104 2–11 2–35 2–61 0–03 0–28 0–54
ii. An option on Eurodollar futures is an option to enter into a Eurodollar futures contract. Recall from chapter 5,
when the Eurodollar futures quote changes by 1 basis point (0.01), there is a gain or loss on a Eurodollar futures
contract of $25. Similarly, in the pricing of options on Eurodollar futures, 1 bp = $25.
In the shortest maturity contract, prices are quoted to the nearest ¼ of a bp. For the next 2 mos., they are quoted to
the nearest ½ bp. The WSJ quote for the CME Eurodollar futures contract shown below should be multiplied by 10
to get the CME quote in bp. For example, the 5.92 quote for the CME March call futures option when the strike
price is 94.50 in indicates that the CME quote is 59.25 bp. and 1 contract costs 59.25 × $25 = $1,481.25.
Eurodollar (CME)$ million; pts of 100%

STRIKE CALLS–SETTLE PUTS–SETTLE
PRICE Mar Apr May Mar Apr
9450 5.92 10.30 ––––– 0.00 0.00
iii. Interest rate futures contracts work in the same way as other futures contracts.
The payoff from a call is max (F – K, 0), where F is the futures price at the time of exercise and K is the strike price.
Further, the option holder obtains a long position in the futures contract at exercise and the option writer obtains a
short position (in addition to the cash payoff).

Since interest rate futures prices increase when bond prices increase (i.e., when rates fall), and vice versa, investors
who thinks that short–term interest rates will rise speculate by buying put options on Eurodollar futures (an investor
who thinks rates will fall speculates by buying call options on Eurodollar futures).
An investor who thinks that long–term interest rates will rise speculates by buying put options on T– note futures or T–
bond futures, and an investor who thinks they will fall can speculate by buying call
options on these securities.
Example: Suppose that it is Feb and the futures price for the Jun Eurodollar contract is
33 93.82 (which corresponds to a 3 mo. Eurodollar interest rate of 6.18% per annum).
The price of a call option on the contract with a strike price of 94.00 is quoted as 0.20.
The option is attractive to an investor who feels that interest rates are likely to fall.
Assume short–term interest rates fall about 100 bps over the next 3 mos., and
the investor exercises the call when the Eurodollar futures price is 94.78 (which
corresponds to a 3 mo Eurodollar interest rate of 5.22% per annum.)
The payoff = 25 × (94.78–94) = $1,950. The cost of the contract = 20 × 25 = $500.
The profit = $1,450.
Example: Suppose that it is Aug and the futures price for the Dec. T–bond contract is 96–09 (or 96 9/32 = 96.28125).
The yield on long–term government bonds is about 6.4% per annum.
If an investor feels that the yield will fall by Dec., buy Dec. calls with a strike price of 98. Assume that the price of
these calls is 1–04 (1 4/64 = 1.0625% of the principal).
If long–term rates fall to 6% per annum, and the T–bond futures price rises to 100–00, the net profit per $100 of bond
futures = 100.00 – 98.00 – 1.0625 = 0.9375
Since 1 option contract is for the purchase/sale of instruments with a face value of $100,000, the profit = $937.50 per
option contract purchased.
Reasons for the Popularity of Futures Options (vs options on the underlying asset).
1. The main reason is that a futures contract is (generally) more liquid and easier to trade.
2. A futures price is known immediately from trading on the futures exchange (vs. the spot price not being so readily
available).
Example: T–bonds.
i. The market for T–bond futures is more active than the market for any particular T–bond.
ii. A T–bond futures price is known immediately from trading (vs. the current market price of a bond known only by
contacting one or more dealers.
iii. Investors would rather take delivery of a T–bond futures contract than T–bonds.
For commodities futures, it is easier and more convenient to make or take delivery (e.g. of a live–hogs futures
contract than it is to make or take delivery of hogs).
iv. Exercising a futures option usually does not lead to delivery of the underlying asset, because the underlying
futures contract is closed out prior to delivery.
v. Futures options are eventually settled in cash, which is appealing to many investors (with limited capital) who may
find it difficult to come up with the funds to buy the underlying asset when an option is exercised.
vi. Futures and futures options are traded in pits side by side in the same exchange, which aids hedging, arbitrage, and
speculation, making the markets more efficient.
vii. Futures options tend to entail lower transactions costs than spot options in many situations.
Put–Call Parity for European futures options
Assumption: there is no difference between the payoffs from futures and forward contracts. See Chapter 8 for the
derivation of the put–call parity relationship for European stock options.
Consider European call and put futures options, both with strike price K and time to expiration T.
Form two portfolios:
Portfolio A: a European call futures option + cash equal to Ke–rT. In this case, cash is invested at the risk–free rate, r,
and grows to K at time T. Let FT be the futures price at maturity of the option.
If FT > K, the call option is exercised and the portfolio is worth FT.
If FT < K, the call is not exercised and portfolio A is worth K.
The value of portfolio A at time T is therefore the max (FT , K)
Portfolio B: a European put futures option + a long futures contract + cash equal to F0e–rT. In this case, cash is invested
at the risk–free rate and grows to F0 at time T.
The put option has a payoff of max(K – FT , 0). The futures contract has a payoff of FT – F0.

The value of the at time T is F0 + (FT – F0) + max(K – FT, 0) = max(FT, K)

Because the 2 portfolios are equal at time T and since no early exercise opportunities exist, they must be worth the
same today. Portfolio A today = c + Ke–rT, (c is the price of the call futures option). Portfolio B today = p + F0e–rT, (p is
the price of the put futures option). Thus, c + Ke–rT = p + F0e –rT.
Example: Assume the price of a European call option on silver futures for delivery in 6 mos. is $0.56 per ounce, and
the exercise price is $8.50.
Assume the silver futures price for delivery in 6 mos. is $8.00 and the risk–free interest rate for an investment that
matures in 6 mos. is 10% per annum.
Thus, the price of a European put option on silver futures with the same maturity and exercise date as the call option is
0.56  8.50e0.10.5  8.00e0.10.5  1.04
For American futures the put–call parity relationship is F0 e rT  K  C  P  F0  Ke rT
14.6 Valuation of Futures Options using Binomial Trees

The key difference between futures options and stock options: there are no up–front costs when a futures contract is
entered into. See chapter 10 since this section uses a binomial tree approach for valuation.
Suppose the current futures price is 30 and is expected to move to 33 or 28 over the next month.
Consider a 1 mo. call option on the futures with a strike price of 29 (ignore daily settlement).
If the futures price moves to 33, the payoff from the option is 4 and the value of the futures contract is 3.
If the futures price moves to 28, the payoff from the option is 0 and the value of the futures contract is –2.
Futures prices movements
To set up a riskless hedge, construct a portfolio consisting of a short position in 1 option contract and a long
position in  futures contracts.
If the futures price moves to 33 , the value of the portfolio is 3 –4
If the futures prices moves to 28 , the value of the portfolio is – 2 .
The portfolio is riskless when 3 – 4 =–2.  = 0.8.
For this value of , we the portfolio = 3 × 0.8 – 4 = –1.6 in one month. Assuming a risk–free interest rate of 6%,
the value of the portfolio today must be –1.6 e – 0.06 × 0.08333 = –1.592
The portfolio consists of 1 short option and A futures contracts.
Because the value of the futures contract today is 0, the value of the option today must be 1.592 .
A Generalization
To generalize, consider a futures price that starts at F 0 and is anticipated to rise to F 0u or move down to F 0d over
the time period T.
Consider a derivative maturing at the end of the time period, and we suppose that its payoff is fu, if the futures price
moves up and fd if it moves down.
The riskless portfolio consists of a short position in 1 option combined with a long position in  futures contracts,
f  fd
where  = u
F0 u  F0 d
The value of the portfolio at the end of the time period is always ( F0 u  F0 )   f u
Let the risk–free interest rate be r. The value of the portfolio today is ( F0 u  F0 )  fu  e rT
Another expression for the present value of the portfolio is –f (f is the value of the option today).
Thus,  f  ( F0 u  F0 )  fu  e rT
Substituting for  and simplifying , the equation is f  e rT  pf u  1  p  f d  , where p  1  d .

ud
Assuming u = 1.1 , d = 0.9333, r =0.06, T=0.08333, fu=4, and f d = 0:
1  0.9333
p  0.4 , and f = e–0.06 × 0.08333[0.4 × 4 + 0.6 × 0] = 1.592 , which agrees with the answer obtained
1.1  0.9333
earlier for this example.
14.7 Futures Price Analogy

Hull makes the following analogy of futures options to the analysis of options on a stock paying a dividend yield:
“A futures price behaves in the same way as a stock paying a dividend yield at the domestic risk–free interest
rate, r ”.

Further, for all future prices, the expected growth rate in a futures price in a risk–neutral world is zero.
This applies in the world where interest rates are stochastic as well as where they are constant.
i. ( F ) , where
Because the expected growth rate of the futures price is zero, F0  E T
FT is the futures price at the maturity of the contract, and
F0 is the futures price at time zero, and
denotes the expected value in a risk–neutral world.
E
ii. Because FT = ST, where ST is the spot price at time T, F0  E (F )
T
Thus, for all assets the futures price equals the expected future spot price in a risk–neutral world.
14.8 Black’s Model for Valuing Futures Options

European futures options can be valued by extending the results discussed thus far.
i. It is assumed that futures prices have the same lognormal property assumed for stock prices.
The European call price and put price for a futures option are given by:
c  e rT [ F0 N ( d1 )  KN ( d 2 )], and p  e rT [ KN (  d 2 )  F0 N (  d1 )] , where
In( F / 0 K )   2T / 2 In( F0 / K )   2T / 2
d1  ; and d 2   d1   T , and  is the volatility of the futures price.
 T  T
ii. When the cost of carry and the convenience yield are functions only of time, it can be shown that the volatility of
the futures price is the same as the volatility of the underlying asset.
Note: Black's model does not require the options contract and the futures contract to mature at the same time.
Example. Consider a European put futures option on crude oil.
The time to the option's maturity is four months, the current futures price is $20, the exercise price is $20, the risk–free
interest rate is 9% per annum, and the volatility of the futures price is 25% per annum.
Thus, F0 = 20, K = 20, r = 0.09, T = 4/12,  = 0.25, and ln(F0/K) = 0.
 T  T
Step 1: Find N(–d1) & N(–d2); d1   0.07216, d 2    0.07216 , and N(–d1) = 0.4712, N(–d2) = 0.5288
2 2
Step 2: Using the equation for the price of a European put on a futures options, p  e rT [ KN (  d 2 )  F0 N (  d1 )] ,
compute the price of the put: p  e0.09 x 4 /12 (20  0.5288  20  0.4712)  1.12 or $1.12.
14.9 Futures Options vs. Spot Options

Compare options on futures and options on spot when they have the same strike price and time to maturity.
An option on spot or spot option is an option to buy or sell the underlying asset in the spot market.
 The payoff from a European spot call option with strike price K is max(ST – K, 0), where ST is the spot price at the
option's maturity.
 The payoff from a European futures call option with the same strike price is max(FT – K, 0), where FT is the
futures price at the option's maturity.
1. If the European futures option matures at the same time as the futures contract:
FT = ST and the two options are in theory equivalent.
2. If the European call futures option matures before the futures contract, it is worth:
a. more than the corresponding spot option in a normal market (futures prices are higher than spot prices) and
b. less than the corresponding spot option in an inverted market (where futures prices are lower than spot prices).
Similarly,
1. A European futures put option equals its spot option counterpart when the futures option matures at the same time
as the futures contract.
2. If the European put futures option matures before the futures contract, it is worth:
a. less than the corresponding spot option in a normal market and
b. more than the corresponding spot option in an inverted market.
Results for American Options
 Traded futures options are usually American.
 Since r is positive, there is always some chance that it will be optimal to exercise an American futures option early,
and thus American futures options are worth more than their European counterparts.

It is not generally true that an American futures option equals the corresponding American spot option when the
futures and options contracts have the same maturity.
Example: Suppose a market exists with futures prices consistently higher than spot prices prior to maturity (e.g. as is
the case with most stock indices, gold, silver, low–interest currencies, etc.)
 An American call futures option must be worth more than the corresponding American spot call option. In some
situations the futures option will be exercised early, which will provide a greater profit to the holder. Thus, an
American put futures option must be worth less than the corresponding American spot put option.
 In an inverted market, with futures prices lower than spot prices, as is the case with high–interest currencies and
some commodities, the reverse must be true.
American call futures options are worth less than the corresponding American spot call option, whereas American put
futures options are worth more than the corresponding American spot put option.
The differences just described hold true when the futures contract expires later than the options contract as well as
when the two expire at the same time.
In fact, the differences tend to be greater the later the futures contract expires.


Chapter 15: The Greek Letters
Section Description
15.1 The Details of Selling a Hypothetical Call Option
15.2 Naked and Covered Positions
15.3 A Stop–Loss Strategy
15.4 Delta Hedging
15.5 Theta
15.6 Gamma
15.7 Relationship among Delta, Theta, and Gamma
15.8 Vega
15.9 RHO
15.10 The Realities of Hedging
15.11 Scenario Analysis
15.12 Portfolio Insurance
15.13 Stock Market Volatility
Introduction
Financial institutions manage the risk they incur when selling options or other derivatives that have been tailored to
meet the needs of their clients. The exposure created cannot be neutralized by buying on an exchange the same option
as it has sold. Since the option has been tailored to the needs of a client, it does not correspond to the standard products
traded by exchanges. Thus, hedging the exposure is difficult.
Alternative approaches to solving this problem are discussed in this chapter. Topics covered include:
 The “Greeks” measure a different dimension of the risk in an option position.
 The creation of synthetic options. Options do not necessarily need to be purchased, but can be created
synthetically. This process is very closely related to the hedging of options.
A trader’s goal is to manage the Greeks so that all risks are acceptable.
15.1 The Details of Selling a Hypothetical Call Option

For the next few sections, consider the following viewpoint: A financial institution with trade executed: The sale of a
European call option on 100,000 shares of a non–dividend–paying stock for $300,000.
Assumptions: The stock price is $49, the strike price is $50, the risk–free interest rate is 5% per annum, the stock
price volatility is 20% per annum, the time to maturity is 20 weeks (0.3846 years), and the expected return from the
stock is 13% per annum [S0 = 49;  = 0.13;  = 0.20; T = 0.3846; r = 0.05; K = 50]
The problem: How to hedge the exposure created by the sale of a call option.
15.2 Naked and Covered Positions

Two possible strategies to hedge the risk created by the sale of the call option are described as follows:
Strategy 1. Adopt a naked position (i.e. do nothing).
Q. When does this strategy works well?
A. When the stock price is below $50 at the end of the 20 weeks.
Q. Why is this the case?
A. Because the option expires worthless and the seller of the call makes a profit of $300,000.
Q. What is the cost to the institution if the call is exercised?
A. 100,000 × [Stock price – strike price]. For example, assume the stock price is $60 after 20 weeks. The cost of
satisfying the option is 100,000 × [$60 – $50] = $1,000,000.
Strategy 2. Adopt a covered position (e.g. buy 100,000 shares as soon as the option has been sold).
Q. When does this strategy works well?
A. When the stock price is above $50 at the end of the 20 weeks.
Q. Why is this the case?
A. Because the institution can deliver the stock if the option is exercised.
Q. What is the institution’s exposure to loss?
A. A drop in the price of the stock. For example, if the stock price drops to $40, the institution loses
100,000 × [$49 – $40] = $900,000 on its stock position.

Conclusions:
 Neither a naked position nor a covered position provides a satisfactory hedge.
 A perfect hedge would ensure that the cost is always $240,000 (the price computed using the Black–Scholes
formula).
 Under a perfect hedge, the standard deviation of the cost of writing the option and hedging it is zero.
15.3 A Stop–Loss Strategy

A Stop–Loss strategy is one type of hedge. Assume you have written a European call option with strike price, K, to
buy one unit of a stock.
a. The hedge involves buying the stock as soon as its price rises above K, and selling as soon as it falls below K.
b. The objective is to hold a naked position whenever the stock price is less than K and a covered position whenever
the stock price is greater than K.
The stop–loss strategy can be demonstrated graphically as follows:
Stock Price S(t)
Buy the stock at time t1, and at time t3, and
deliver it at time T
Sell the stock at time t2 and at t4

K
t1 t2 t3 t4 Time (t)
One might think that the total cost, Q, of writing and hedging the option is Q  Max [ S0  K , 0] , since all purchases
and sales made, subsequent to time 0, are at price K. If this were correct, the hedging scheme would work perfectly in
the absence of transactions costs. Further, the cost of the option would always be less than its Black–Scholes price (the
Black–Scholes Model can be used to determine its theoretical price). Therefore, one could earn riskless profits by
writing options and hedging them.
Two reasons why equation Q  Max [ S0  K , 0] is incorrect.
1. The cash flows to the hedger occur at different times and must be discounted.
2. Purchases and sales cannot be made at exactly the same price, K.
Since we cannot assume that both purchases and sales are made at the same price, purchases must be made at must
be made at K +  and sales must be made at K –  . Thus, every purchase and subsequent sale involves a cost
(apart from transaction costs) of 2 . Assuming stock prices change continuously,  can be made arbitrarily
small, but this is offset by the increasing frequency of trading.
Conclusions
The stop–loss strategy does not work particularly well as a hedging scheme, particularly in the case of an
out–of–the–money option.
 If the stock price never reaches the strike price, K , the hedging scheme costs nothing.
 If the path of the stock price crosses the strike price level many times, the hedging scheme becomes quite
expensive.
To assess the overall performance of the scheme, Monte–Carlo simulation can be used.
The table below shows the performance of stop–loss strategy.
t (weeks) 5 4 2 1 0.5 0.25
Hedge performance 1.02 0.93 0.82 0.77 0.76 0.76
 The hedge performance measure is the ratio of the standard deviation of the cost of writing the option
and hedging it to the Black–Scholes price of the option.
 A performance measure below 0.70 was impossible to attain (regardless of how small t is).

15.4 Delta Hedging

The delta of an option was initially introduced in chapter 10, and is further defined below.
Symbol: 
Formula: c
 . Delta is the partial derivative of the call price with respect to underlying asset price
S
(this assumes that volatility is constant and there are no transaction costs).
Defined: The delta of an option is defined as the rate of change of the option price with respect to the
price of the underlying asset.
A graphical representation of how delta is determined is shown below:
Q: If the delta of a call option on a stock is 0.6, what does this mean?
A: When the stock prices changes by a small amount, the option price changes by about 60% of that amount.
A Delta Hedging Example:
Givens: (1) An investor has sold 20 call options contracts (each option contract controls 100 shares).
(2) The stock price is $100 and the option price is $10.
By doing nothing, the investor is subject to incurring a loss should the stock price rise above $100.
To “delta hedge” this position, buy 0.60 × 2,000 = 1,200 shares.
Why? Because the gain (loss) on the option position can be offset by the gain (loss) on the stock position.
For example, if the stock price rises by $1.00, the investor’s realizes (if the total position is terminated):
a gain of $1,200 ($1.00 × 1,200 shares) on the shares purchased, and a loss of $1,200 ($0.60 × 2,000 shares) on the
options written. The option prices rises and falls by its delta (in this case 0.60 × $1.00 = $0.60)
In summary:
The delta of the investor’s option position is 0.60 × (–2,000) = –1,200.
The delta of the investor’s long position is 1.0 × (1200) = 1,200 (since the delta of the stock is 1.0).
The delta of the investor’s overall position is = 0 (a zero delta position is a.k.a. “delta neutral”).
Rebalancing:
Since delta changes, the delta hedged (or delta neutral) position will remain in tact for only a short time period.
Therefore, the hedge must be adjusted or rebalanced.
For example, if the stock price rises to $110, then delta rises from 0.60 to 0.65. To maintain the hedge, an extra 100
shares must be purchased [0.05 (e.g. 0.65 – 0.60) × 2,000 shares = 100 shares].
Definitions:
i. a dynamic hedging scheme: one that requires the hedge position to be adjusted periodically.
ii. a static hedging scheme: one that once set up, is never adjusted (a.k.a. hedge–and–forget schemes).
Relationship to Black–Scholes analysis:
Black and Scholes showed that one can set up a riskless portfolio consisting of a position in a derivative on a stock
and a position in the stock. Essentially, they set up a delta–neutral position and argue that the return on the position
over a short period of time equals the risk–free interest rate.
Delta of European Calls and Puts:
1. For a European call option on a non–dividend paying stock:
ln( S0 / K )  ( r  2 / 2)T
a. The Black–Scholes formula shows that (call) = N(d1), where d1 
 T
b. Delta hedging for a short position in a European call option requires keeping a long position of N(d1) shares
at any given time.
c. Delta hedging for a long position in a European call option involves maintaining a short position of N(d1)
shares at any given time.

2. A European put option on a non–dividend paying stock:

a. The Black–Scholes formula shows that (put) = N(d1) – 1.
b. Delta hedging for a long position put option requires a long position in the underlying stock 0.
c. Delta hedging for a short position put option requires a short position in the underlying stock 0.
Delta of Other European Options:
A. For European call options on a stock index paying a dividend yield q:  = e–qT N(d1) where
ln( S0 / K )( r q 2 /2)T
d1 
 T
B. For European put options on the stock index,  = e–qT [N(d1) – 1],
 rf T ln( S0 / K )( r r f  2 /2)T
C. For European call options on a currency:   e N ( d1 ), where d1  and rf is the foreign
 T
risk–free interest rate.
r T
D. For European put options on a currency:   e f [ N ( d1 )  1]
ln( F0 / K )  2T /2
E. For European futures call options:  = e–rT N(d1), where d1  and r is the domestic risk–free
 T
interest rate.
F. European futures put options:  = e–rT [N(d1) – 1]
Example 1:
A U.S. bank has sold six–month put options on £1 million with a strike price of 1.6000 and wishes to make its
portfolio delta neutral.
Assume the current exchange rate = 1.6200, the riskfree interest rate in the United Kingdom = 13%, the risk–free
interest rate in the U.S. = 10%, and the volatility of sterling is 15%.
Thus, [S0 = 1.62; rf = 0.13;  = 0.15; T = 0.50; r = 0.10; K = 1.60].
Question: Delta hedging requires what type of position in sterling to be set up initially to make the portfolio delta
neutral?
r T ln( S0 / K )( r rf  2 /2)T
The delta of a put option on a currency is   e f [ N (d1 )  1] , where d1 
 T
ln(1.62/1.6) (0.10 0.130.0225/2)0.50 0.0124 0.009375
d1    0.0287 and N(d1) = 05115.
0.15 0.50 0.106066
Thus, the delta of the put option is   e0.130.50 [0.5115  1]  0.458
a. This means that when the exchange rate increases by S, the price of the put goes down by 45.8% of S.
b. The delta of the bank's total short position is –1,000,000 × –0.458 = +458,000.
c. To make the position delta neutral, we must therefore add a short sterling position of £458,000 to the option
position. The short sterling position has a delta of –458,000 and neutralizes the delta of the option position.
As is true with all delta hedging, the short position must be changed as time passes.
Delta of Forward Contracts:
Delta can be applied to financial instruments other than options.
Consider the case of a forward contract on a non–dividend paying stock.
The value of a forward contract is, S0 – Ke–rt where K is the delivery price and T is the forward contract's time to
maturity. Since the value of a forward contract on a non–dividend paying stock changes by S when the stock itself
changes by S, we conclude that the delta of a forward contract on one share of a non–dividend paying stock is
always 1.0. Thus,
i. a short forward contract on one share can be hedged by purchasing one share.
ii. a long forward contract on one share can be hedged by shorting one share.
Because delta is always 1.0, no changes need to be made to the position in the stock during the life of the contract.
For an asset providing a dividend yield at rate q, the forward contract's delta is e–qT.
In the case of a stock index, q is set equal to the dividend yield on the index. For a currency, it is set equal to the
foreign risk–free rate, rf.

Delta of a Futures Contract

If the underlying asset is a non–dividend–paying stock, the futures price, F0, is given by: F0 = S0erT.
 When the stock price increases by S, the futures price increases by S erT
 Thus, the delta of the futures contract is erT.
Sometimes a futures contract is used to achieve a delta–neutral position.
Define:
T : Maturity of futures contract.
H A : Required position in the asset for delta hedging.
H F : Alternative required position in futures contracts for delta hedging.
a. If the underlying asset is a non–dividend–paying stock, HF = HA×e–rT.
b. When the underlying asset pays a dividend yield q: HF = HA×e–(r – q)T
c. For a stock index we set q equal to the dividend yield on the index; for a currency we set it equal to the foreign risk–
 ( r  r f )T
free rate, rf, so that H F  H A  e
Example 2: Consider the option in Example 1 (see the previous page), where hedging using currency requires a short
position of £458,000.
Suppose that the bank decides to hedge using nine–month currency futures contracts. In this case, T = 0.75 and
 ( r  r f )T
e  e  (0.100.13)0.75  1.0228 , so that the short position in currency futures required for delta hedging is
 ( r  rf ) T
HF  HA e = £458,000 1.0288  £468,442.
Because each futures contract is for the purchase or sale of £62,500, this means that (to the nearest whole number) 7
contracts should be shorted (seven is the nearest whole number to 468,442/62,500).
Dynamic Aspects of Delta Hedging
Tables 2 and 3 provide two simulations of the operation of delta hedging for the example in section 1
a. The hedge is assumed to be adjusted or rebalanced weekly.
b. The initial value of delta is calculated from [S0 = 49;  = 0.20; T = 0.3846 (20 weeks); r = 0.05; K = 50].
ln(49 / 50)  (0.05 0.04 / 2)20 0.00672037
and is equal to N(d1). d1    0.0541
0.20 20 0.12403
Thus, N(d1) = N(0.0541) = 0.522 (from the N(x) table).
This means that as soon as the option is written, $2,557,800 must be borrowed to buy 52,200 shares at a price of $49.
Because the interest rate is 5%, interest costs totaling $2,500 ~ [(0.05 × 2,557,800) ÷ 52] are incurred in the first week.
Notes about Table 2: Simulation of delta hedging. Option closes in the money
 Delta is recomputed at the end of the first week to be 0.458 (using S1 = 48.125;  = 0.20; T = 19 weeks; r = 0.05;
K = 50)
 6,400 shares must be sold to maintain the hedge, cash of $308,000 is received and the cumulative borrowings at
the end of week one are reduced to $2,252,300.
 Toward the end of the life of the option, it is apparent that the option will be exercised and delta approaches 1.0.
 By week 20, the hedger has a fully covered position. The hedger receives $5,000,000 (100,000 × $50) for the stock
held, so the total cost of writing the option and hedging it is $263,300 (5,263,300 – 5,000,000).
Notes about Table 3 “Option Closes out of the Money, Cost to the Option Writer is $256,600”:
 Since the option closes out of the money, delta approaches zero.
 By week 20 the hedger has a naked position and has incurred costs totaling $256,600.
Comparing the results of Tables 2 and 3 to the results of the Black–Scholes model:
a. When discounting the option costs to the beginning of the 20 week period, the costs are close to but not exactly the
same as the Black–Scholes price of $240,000. If the hedging scheme worked perfectly, the cost of hedging after
discounting would be exactly equal to the Black–Scholes theoretical price.
b. The reason for the variation in the cost of delta hedging is that the hedge is rebalanced only once a week. The more
frequently rebalancing takes place, the greater the reduction in the uncertainty in the cost of hedging.

Dynamic Aspects of Delta Hedging:

Simulation of Delta Hedging; Option Closes in the Money
Cost of Option to the Writer = $263,400
Cost of Shares Cumulative
Shares purchased Interest Cost
Stock purchased (sold) Cost (incl. Int)
Week Price Delta (sold) (000s) (000s) (000s)
(1) (2) (3) (4) (5) (6)
0 49.000 0.522 52,200 $2,557.8 2.5 $2,557.8
1 48.125 0.458 (6,400) -$308.0 2.2 $2,252.3
2 47.375 0.400 (5,800) -$274.8 1.9 $1,979.7
3 50.250 0.596 19,600 $984.9 2.9 $2,966.5
4 51.340 0.693 9,700 $498.0 3.3 $3,467.4
5 53.125 0.774 8,100 $430.3 3.8 $3,901.0
6 53.000 0.771 (300) -$15.9 3.7 $3,888.9
7 51.875 0.706 (6,500) -$337.2 3.4 $3,555.4
8 51.375 0.674 (3,200) -$164.4 3.3 $3,394.4
9 53.000 0.787 11,300 $598.9 3.8 $3,996.6
10 49.875 0.550 (23,700) -$1,182.0 2.7 $2,818.4
11 48.500 0.413 (13,700) -$664.5 2.1 $2,156.7
12 49.875 0.542 12,900 $643.4 2.7 $2,802.1
13 50.375 0.591 4,900 $246.8 2.9 $3,051.7
14 52.125 0.768 17,700 $922.6 3.8 $3,977.2
15 51.875 0.759 (900) -$46.7 3.8 $3,934.3
16 52.875 0.865 10,600 $560.5 4.3 $4,498.6
17 54.875 0.978 11,300 $620.1 4.9 $5,123.0
18 54.625 0.990 1,200 $65.6 5 $5,193.4
19 55.875 1.000 1,000 $55.9 5.1 $5,254.3
20 57.250 1.000 - $0.0 $5,259.4
5,263.3 (per the text)
Total 100,000
Column (1) given.
Column (2) Computed using the N(d1)
Column (3): Week 0; (2) * 100,000; Weeks 1 - 20: [(2i) - (2i-1)] * 100,000
Column (4) = Column (1) * Column (3)/1,000
Column (5) given.
Column (6): Week 0; (4); Weeks 1 - 20: [(6i-1) + (5) + (4)]
Table 4: Performance of Delta Hedging
Time Between Hedge Rebalancing ( in Weeks): 5 4 2 1 0.5 0.25
Performance measure: 0.43 0.39 0.26 0.19 0.14 0.09
Delta hedging is a great improvement over the stop–loss strategy. Unlike the stop–loss strategy, the performance of
delta hedging gets steadily better as the hedge is rebalanced more frequently.
Delta hedging attempts to keep the total wealth of the financial institution as unchanged as possible.
Consider the following (values drawn from table 14.2):
Value of Shares +2,557,800 +4,171,000
I l l l I
Time (weeks): 0 1 2 ... 8 9
Value of Option +240,000 +414,500
Cumulative Cost –2,557,800 –4,000,500
During the 9 week period, the overall wealth of the financial institution changed as follows:
Value of shares owned at week nine: $4,171,000 (78,700 × $53)
Cost of shares purchased at week zero – 2,557,800
Value of option at week nine – 414,500
Value of option at week zero + 240,000
Difference in cumulative cost (week 9 – week 0) –1,442,700 (4,000,500 – 2,557,800)
–$ 4,000
Delta of a Portfolio

is the delta of an options portfolio, dependent on a single asset whose price is S, where  is the value of the
S
portfolio. The delta of the portfolio can be calculated from the deltas of the individual options in the portfolio. If a
portfolio consists of an amount, wi of option i (1  i  n), the delta of the portfolio is given by:

n
   wi  i , where i is the delta of the ith option.
i 1
Example: Suppose a US-company has the following three positions in options on the Australian dollar:
Position Type and No. of options Strike Price Expiration Delta of each option
Long 100,000 Call Option 0.55 3 months 0.533
Short 200,000 Call Option 0.56 5 months 0.468
Short 50,000 Put Option 0.56 2 months –0.508
The delta of the whole portfolio is
n
   wi  i = [0.533 × 100,000 – 200,000 × 0.468 – 50,000 × (–0.508)] = –14,900
i 1
Different approaches to making the portfolio delta neutral.
1. Enter into a long position of 14,900 Australian dollars.
2. Enter into a long position in Australian dollar forward contracts.
Suppose that the risk–free rate of interest is 8% in Australia and 5% in the U.S. (r = 0.05 and rf = 0.08.)
r T
a. the delta of the forward contract maturing at time T on one Australian dollar is e f , e 0.080.5  0.9608
b. The long position in Australian dollar forward contracts for delta neutrality is 14,900/0.9608 = 15,508.
3. Enter into a long position in Australian dollar futures contracts.
 ( r  r f )T
Using the equation H F  H A  e , the long position in Australian dollar futures contracts for delta neutrality is
–(0.05 – 0.08)0.50
HF = 14,900×e = 15,125.
15.5 Theta
Definition: The theta of a portfolio of derivatives, :
a. is the rate of change of the portfolio value with respect to the passage of time (or with respect to a decrease in the
times to maturity of the derivatives in the portfolio).
b. is also referred to as the time decay of the portfolio.
S N ( d1 )
A. For a European call option on a non–dividend–paying stock,    0  rKe rT N ( d 2 ) , where:
2 T
ln( S0 / K )  ( r   2 / 2)T 1  x2 /2
d1  , d 2  d1   T and N ( x )  e
 T 2
S N ( d1 )
B. For a European put option on the stock,    0  rKe rT N (  d 2 )
2 T
C. For a European call option on a stock index paying a dividend at rate q,
S N ( d1 ) e qT ln( S0 / K )  ( r  q   2 / 2)T
 0  qS0 N ( d1 )e qT  rKe rT N ( d 2 ) , where d1 
2 T  T
S N ( d1 ) e  qT
D. For a European put option on the stock index    0  qS0 N (  d1 )e qT  rKe rT N (  d 2 )
2 T
Notes:
 With q equal to rf , equations C. and D. above give thetas for European call and put options on currencies.
 With q equal to r and S0 equal to F0, equations C. and D. give thetas for European futures options.
 In the formulas above, time is measured in years. However, theta is usually quoted in days. To obtain the theta
with time measured in days, the results of using the formulas must be divided by 252 (the number of trading
days in one year).
More about Theta:
a. Theta is usually negative for an option (since as time passes, the option becomes less valuable).
b. Theta is not the same type of hedge parameter as delta. Since there is no uncertainty about the passage of time, it
does not make any sense to hedge against the effect of the passage of time.
c. Theta is a proxy for gamma in a delta–neutral portfolio.

15.6 Gamma
Definition: The gamma, , of a portfolio of derivatives on an underlying asset:
a. is the rate of change of the portfolio's delta with respect to the price of the underlying asset.
b. is the second partial derivative of the portfolio value with respect to the asset price:    2  / S 2 .
Gamma measures the curvature of the relationship between the option price and the stock price.
Example:
When the stock price moves from S to S', delta hedging assumes that the option price moves from C to C’ when in
actual fact it moves from C to C”. The difference between C’ and C” leads to a hedging error. The error depends on
the curvature of the relationship between the option price and the stock price. Gamma measures this curvature.
C’’
C’
S S’
Recall that Theta is a proxy for gamma in a delta–neutral portfolio.
For a delta–neutral portfolio,    t  (1 / 2) S 2 (14.6).
Symbol Description
 price change of the portfolio.
S price change of an underlying asset during a small interval of time,  t.
Example: Suppose that the gamma of a delta–neutral portfolio of options on an asset is – 10,000. Equation (14.6)
shows that if a change of +2 or –2 in the price of the asset occurs over a short period of time, there is an unexpected
decrease in the value of the portfolio of approximately 0.5 × –10,000 × 22 = –$20,000.
The relationship between  and S:
 When gamma is positive, theta tends to be negative.
 When gamma is negative, theta tends to be positive.
Making a Portfolio Gamma Neutral
A position in either the underlying asset itself or a forward contract on the underlying asset has zero gamma and cannot
be used to change the gamma of a portfolio.
To adjust the gamma of a portfolio, we must take a position in an instrument such as an option that is not linearly
dependent on the underlying asset.
Example: Suppose that a delta–neutral portfolio has a gamma equal to  and a traded option has a gamma equal to
T.
If the number of traded options added to the portfolio is wT, the gamma of the portfolio becomes wT T   . Hence,
 
the position in the traded option necessary to make the portfolio gamma neutral is , since  T    0 .
T T
Example: Suppose that a portfolio is delta neutral and has a gamma of –3,000. The delta and gamma of a particular
traded call option are 0.62 and 1.50, respectively.
3,000
The portfolio can be made gamma neutral by including in the portfolio a long position of  2,000 in the
1.5
traded option.
However, the delta of the portfolio will then change from zero to 2,000 × 0.62 = 1,240. Thus, 1,240 of the
underlying asset must be sold from the portfolio to keep it delta neutral.
Notes:
 Delta neutrality provides protection against relatively small stock price moves between rebalancing.
 Gamma neutrality provides protection against larger movements in this stock price between hedge rebalancing.

Calculation of Gamma
N ( d1 )
The gamma of a European call or put option on a non–dividend–paying stock is   , where
S0 T
ln( S0 / K )  ( r   2 / 2)T 1  x2 / 2
d1  , d 2  d1   T and N ( x )  e
 T 2
For a European call or put option on a stock index paying a continuous dividend at rate q .
N (d1 )e  qT

S0 T
Notes: This formula gives the gamma for:
a. a European option on a currency when q is put equal to the foreign risk–free rate, and
b. a European futures option with q = r and S 0 = F 0.
15.7 Relationship among Delta, Theta, and Gamma

  2
 ;  ;  2 .
t S S
 It follows that:   rS   (1/ 2) 2 S 2   r
 For a delta–neutral portfolio,  = 0 and   (1/ 2) 2 S 2   r
This shows that when  is large and positive, gamma tends to be large and negative, and vice versa. This is consistent
with the way in which Figure 15.8 has been drawn and explains why theta can be regarded as a proxy for gamma.
15.8 Vega
The vega of a portfolio of derivatives, V , is the rate of change of the value of the portfolio with respect to the

volatility of the underlying asset: V 

If vega is high in absolute terms, the portfolio’s value is very sensitive to small changes in volatility (and vice
versa).
A position in the underlying asset or in a forward contract has zero vega.
The vega of a portfolio changes by adding a position in a traded option.
If V is the vega of the portfolio and VT is the vega of a traded option, a position of V / VT in the traded option
makes the portfolio instantaneously vega neutral.
A portfolio that is gamma neutral will not, in general, be vega neutral, and vice versa 0.
For a portfolio to be both gamma and vega neutral, at least two traded derivatives dependent on the underlying
asset must usually be used.
Example 6: You are given a portfolio that is delta neutral. In addition, you are given the following:
Symbol Description
  5, 000 The gamma of the portfolio is –5,000
V  8, 000 The vega of the portfolio is –8,000
t1  0.5 The gamma of traded option 1 is 0.5
t 2  0.8 The gamma of traded option 2 is 0.8
Vt1  2.0 The vega of traded option 1 is 2.0
Vt 2  1.2 The vega of traded option 2 is 1.2
 t1  0.6 The delta of traded option 1 is 0.6
 t 2  0.5 The delta of traded option 2 is 0.5
wi The amount of traded option i.
V 8, 000
1. To make the portfolio vega neutral, take a long position of w1    4, 000 in traded options.
Vt1 2

2. To maintain delta neutrality, 2,400 units of the asset must be sold, since delta increases to 2,400.
Note: The gamma of the portfolio changes from –5,000 to –3,000.
3. To make the portfolio gamma and vega neutral, we need to determine the w1 and w2 to be included in the
portfolio:
–5,000 + 0.5w1 + 0.8w2 = 0
–8,000 + 2.0w1 + 1.2w2 = 0
The solution is w1 = 400, w2 = 6,000.
4. The delta of the portfolio after adding the positions in the two traded options is 400 × 0.6 + 6,000 × 0.5 =
3,240. Hence 3,240 units of the asset would have to be sold to maintain delta neutrality.
Formulas for Vega:
1. A European call or put option on a non–dividend–paying stock: V  S0 T N ( d1 ) , where
ln( S0 / K )  ( r   2 / 2)T 1  x2 / 2
d1  , d 2  d1   T and N ( x )  e
 T 2
2. A European call or put option on a stock or stock index paying a continuous dividend yield at rate q,
ln( S0 / K )  ( r  q   2 / 2)T
V  S0 T N ( d1 )e qT , where d1 
 T
The above equation gives the vega for:
a. a European currency option with q equal to rf
b. a European futures option with q equal to r and S0 equal to F0
Notes:
 The vega of a long position in a call or put option is always positive.
 Gamma neutrality protects against large changes in the price of the underlying asset between hedge rebalancing.
 Vega neutrality protects against variations in .
15.9 RHO
The rho of a portfolio of derivatives measures the sensitivity of the portfolio value to interest rates.
rho = /r
A. For a European call option on a non–dividend–paying stock, rho = KTe–rTN(d2)
B. For a European put option on the stock, rho = –KTe–rTN(–d2), where d 2  d1   T and
ln( S0 / K )  ( r   2 / 2)T
d1  .
 T
These formulas apply to European call and put options on stocks and stock indices paying a dividend yield at rate q.
In the case of currency options, there are two rhos corresponding to the two interest rates.
1. The rho corresponding to the domestic interest rate is given by previous formulas.
2. The rho corresponding to the foreign interest rate for a European call on a currency is given by
r T r T
rho  Te f S0 N (d1 ) , and, for a European put it is rho  Te f S0 N ( d1 )
Section 15.5 – 15.9 Review
Given the following: S0 = 305; K = 300; q = 0.03; r = 0.08;  = 0.25; T = 1/3; compute the theta, gamma, vega, and rho
of a four Month European Put Option on a Stock Index:
Preliminary Calculations: compute d1 , d2, N’(d1), N’(–d1) and N’(–d2)
ln( S0 / K )  ( r  q   2 / 2)T ln(305/ 300)  (0.08  0.03  (0.25)2 / 2)0.3
d1  . d1   0.30215 ; d 2  d1   T  0.1578
 T 0.25 0.3
 (0.30215) 2
1
N (0.30215)  e 2  0.3811 . N ( 0.30215)  0.3974 and N ( 0.1578)  0.4370
2  3.1415
 S N ( d1 ) e qt 1 2x
2
1. Theta.   0  qS0 N (  d1 )e qT  rKe rT N (  d 2 ); N ( x )  e

2 T 2

 (0.30215)2
1
305  e 2
 0.25  e0.03(0.3)
 2  3.1415  0.03  305  0.3794  e0.03(0.3)  0.08(300)e0.08(3)  0.4370
2 0.3
 24.915  3.4369  10.212  18.14
When time is measured in days, theta is –18.14/252 = – 0.072.
(0.30215)2
1 
 e9.03(0.3)
e 2
N '( d1 )e qT 2  3.14159 0.3811  0.99
2. Gamma.      0.0085705
S0 T 305  0.25  0.3 44.022
An increase of 1.0 (from 305 to 306) in the index increases the delta of the option by 0.00857.
2
(0.30215 )
1 
3. Vega. V  S0 T N ( d1 )e  305 0.3  (
 qT
e 2
)  e0.03(0.3) = 176.0918 × 0.381 × 0.99 = 66.44
2  3.1415
For a 1% increase in  (from 25% to 26%), the option increase by 0.6644 (= 0.01 × 66.44).
4. RHO   KTe T  N (  d 2 )  300(0.3)e0.08(0.3)  N ( 0.69784)
 97.368  0.4370  42.54
For a 1% increase in rf (0.08  0.09) , the option decreases by (0.01)(42.54)  0.4254
For more information, see examples 15.3, 15.5, 15.7, and 15.8.
15.10 The Realities of Hedging

Traders would rebalance their portfolios frequently to maintain a zero delta, a zero gamma, a zero vega, etc. However
in practice, this is not possible. Assume a large portfolio is dependent on a single underlying asset:
a. traders can zero out delta at least once a day by trading the underlying asset.
b. traders cannot so easily zero out gamma and vega because it is difficult to find options or other non linear
derivatives that can be traded in the volume required. Thus, gamma and vega are monitored, and corrective action
is taken when they get too large in either a positive or negative direction.
Economies of scale exist for options traders.
a. Given a portfolio of several hundred options on the S&P 500, rebalancing (either by trading the stocks
underlying the index or by trading index futures) is worthwhile since the cost of rebalancing can be recovered
by the profit on many different trades.
b. However, maintaining delta neutrality for an individual option (e.g. on the S&P 500) by trading daily would be
prohibitively expensive.
The majority of institutional trades are sales of call and put options, which have negative gammas and negative
vegas.
 With the passage of time, both the gamma and vega of a institution's portfolio tend to become progressively
more negative, and thus traders are looking for ways to buy options (i.e., acquire positive gamma and vega).
 However, gammas and vegas become small (and thus, of little consequence) when they become deep in or out
of the money and this mitigates the traders' problems referenced above.
15.11 Scenario Analysis

Scenario analysis involves calculating the gain or loss on a portfolio:
a. over a specified period ( usually depended upon the liquidity of the instrument), and
b. under a variety of different scenarios (dependent upon the volatility of the underlying variable).
Consider table 5 below: Profit or Loss Realized in two weeks (in millions)
Exchange Rate
Volatility 0.94 0.96 >>> 1.0 >>> 1.04 1.06
8% +102 +55 +6 –34 –80
10% .80 +40 +2 –38 –85
12% +60 +215 –2 –42 –90
Although the greatest loss in such a table usually corresponds to one of the corners, this is not always the case (e.g.
under a reverse butterfly spread, the greatest loss occurs when the exchange rate does not move).

15.12 Portfolio Insurance

A portfolio manager is often interested in acquiring a put option on her portfolio. This provides protection against
market declines while preserving the potential for a gain if the market does well.
One approach (discussed in Chapter 13) is to buy put options on a market index such as the S&P 500. An alternative is
to create the option synthetically.
Creating Options Synthetically
Creating options synthetically requires:
a. taking a varying position in the underlying asset (or futures on the underlying asset) so that the delta of the
position is made equal to the delta of the required option.
b. that the position taken is the reverse of that necessary to hedge it (since hedging involves the creation of an equal
and opposite option synthetically).
Two reasons why creating put options might be more attractive than buying the put options:
1. Options markets do not always have the liquidity to absorb large institutional trades
2. Fund managers often require strike prices and exercise dates different from those available in the market.
A synthetic option can be created from:
a. trading the portfolio or
b. trading in index futures contracts.
Example: Creating a synthetic put option by trading the portfolio.
ln( S0 / K )( r q 2 /2)T
Given: the delta of a European put is  = e–qT [N(d1) – 1], where d1 
 T
–qT
At any given time, a proportion, e [1 – N(d1)], of stock is sold and the proceeds invested in riskless assets.
Actions required by the portfolio manager:
Value of the original portfolio: The delta of the put becomes: The proportion of the portfolio sold:
Declines More negative Must be Increased
Increases Less Negative Must be Decreased
The cost of insurance arises because the portfolio manager is always selling after a decline in the market and buying
after a rise in the market.
Using this strategy to create portfolio insurance means that, at any given time, funds are divided between the stock
portfolio on which insurance is required and riskless assets. As the value of the stock portfolio increases, riskless assets
are sold and the position in the stock portfolio is increased. As the value of the stock portfolio declines, the position in
the stock portfolio is decreased and riskless assets are purchased.
Example 9
Given: A 6 month European put option, with a strike price of $87 million, is purchased to protect a portfolio worth
$90 million against market downturns. Assume: S0 = 90; K = 87; q = 0.03; r = 0.09;  = 0.25; T = 0.5.
Compute: (a) the proportion of the portfolio that should be sold initially; (b) the delta of the portfolio and the
additional proportion of the portfolio that should be sold if the portfolio drops to $88 million.
a. First compute N(d1). d1  ln(90/87)(0.090.03(0.25) / 2)0.5  0.4499 . N(0.4499) = 0.6736
2
0.25 0.50
Thus, the proportion of the portfolio to be initially sold is e0.030.50 [0.6736  1]  0.3215 = 32.15%).
b. To compute the delta of the portfolio if the portfolio drops to $88 million, again compute N(d1):
ln(88/87)(0.090.03(0.252 )/ 2)0.50
d1   0.3222; N (0.3222)  0.6255
0.25 0.50
Thus, the delta of the portfolio if the portfolio drops to $88 million equals e0.030.5 [0.6255  1.0]  0.3689
The additional proportion of the portfolio that should be sold if the portfolio drops to $88 million is
0.3689 – 0.3215 = 4.74%.
Use of Index Futures
It may be preferable to use index futures to create portfolio insurance rather than using the underlying stocks,
because the transactions costs associated with trades in index futures are generally less than those associated
with the corresponding trades in the underlying stocks.
When using this method, the stock portfolio is kept intact and index futures contracts are shorted.
Equations:
A. The dollar amount of futures contracts shorted as a proportion of the value of the portfolio should be
e qT e ( r  q )T * [1  N ( d1 )]  eq ( T *T ) e rT * [1  N ( d1 )] , where T* is the maturity date of the futures contract.

B. If the portfolio is worth K1 times the index and each index futures contract is on K2 times the index, then the
number of futures contracts shorted at any given time should be
eq (T *T ) e rT *[1  N ( d1 )]( K1 / K 2 )
Example 10
Suppose that in the previous example futures contracts on the S&P 500 maturing in nine months are used to create the
option synthetically. In this case, initially T = 0.5, T* = 0.75,
Assume: S0 = 900; K = 870; q = 0.03; r = 0.09;  = 0.25; T = 0.5; K1 = 100,000; K2 = 250; T* = 0.75;
Question: a. compute the proportion of the portfolio to be sold to create a six month put option.
b. compute the number of futures contracts that should be shorted if nine–month futures contracts on the S&P 500
are used.
2
ln(900/870)(0.090.03 0.25 )0.50 0.03390.045625
a. First compute N(d1). d1  2   0.449; N (0.449)  0.6736 .
0.25 0.50 0.1767
Thus, the proportion of the portfolio that should be initially sold is e0.030.50 [1  0.6736]  0.322
b. To compute the number of futures contracts that should be shorted, use the formula:
e0.03(0.750.50) e0.090.75 [1  0.6736]  (100,000 / 250)  123
Frequency of Rebalancing and October 19, 1987
Important notes:
a. Knowing the frequency of portfolio adjustment or rebalancing is important when creating put options
synthetically.
b. Creating put options on the index synthetically does not work well if the volatility of the index changes rapidly
or if the index exhibits large jumps. After the stock market crash of 10/19/87, portfolio managers who insured
themselves by:
i. buying put options survived this crash well.
ii. creating put options synthetically found that they were unable to sell either stocks or index futures fast
enough to protect their position.
15.13 Stock Market Volatility

Portfolio insurance schemes have the potential to increase volatility.
When the market declines, these schemes cause portfolio managers either to:
a. sell stock, which is likely to drive down the market index, or
b. sell index futures contracts, which is liable to drive down futures prices.
Similar actions (but occurring in the opposite direction) occur when the market rises.
Individual investor portfolio insurance schemes (e.g. entering a market when it rises) are also at work.
Finally, if portfolio insurance trades are a very small fraction of all trades, there is likely to be little effect on the
market.


Chapter 18: Value at Risk
Section Description
18.1 The VaR Measure
18.2 Historical Simulation
18.3 Model Building Approach
18.4 Linear Model
18.5 Quadratic Model
18.6 Monte Carlo Simulation
18.7 Comparison of Approaches
18.8 Stress Testing and Back Testing
18.9 Principal Components Analysis
18.10 Appendix 18A: Cash Flow Mapping
The risk measures discussed in Chapter 15, The Greek Letters, provide valuable information for a company’s trader,
but are of limited use to senior management. VaR attempts to provide a single number summarizing the total risk in a
portfolio of financial assets. Two approaches in calculating VaR are the historical simulation approach and the model–
building approach. Both are widely used and there is no agreement on which of the two is better.
18.1 The VaR Measure

VaR is aimed at making a statement of the following form:
“We are X percent certain that we will not lose more than V dollars in the next N days.”
The variable V is the VaR of the portfolio. It is a function of two parameters:
N = the time horizon (N days).
X = the confidence level.
VaR is easy to understand. It answers a simple question: “How bad can things get?”
The portfolios in diagrams 1 and 2 below have the same VaR, but the portfolio in diagram 2 is much riskier because
potential losses are much larger.
The computation of VaR from a probability distribution of changes in portfolio value, where confidence level is X%
(diagrams 1 and 2 respectively).
Conditional VaR (C–VaR).
C–VaR asks the question “If things do get bad, how much can we expect to lose?”
C–VaR is the expected loss during an N–day period conditional that we are in the (100 – X)% left tail of the
distribution.
Given X = 99 and N = 10, C–VaR is the average amount lost over a 10–day period assuming that a 1% worst–case
event occurs.
In spite of its weaknesses, VaR (not C–VaR) is the most popular measure of risk among both regulators and senior
management.
The Time Horizon (the N–Day VaR)
The formula for an N–day VaR is as follows: N day Var = 1  day VaR  N .
 In practice, analysts set N equal to 1, since there was not enough data to estimate the behavior of market variables
over periods of time longer than one day.
 The formula is exactly true when the changes in the value of the portfolio on successive days have independent
identical normal distributions with mean zero; otherwise it’s an approximation.

18.2 Historical Simulation

Example: Assume we wish to calculate VaR for a portfolio using a one–day time horizon, a 99% confidence level, and
500 days of data.
Step 1: Identify the market variables (e.g. exchange rates, equity prices, interest rates, etc.) affecting the portfolio.
Step 2: Collect data on the movements in these market variables over the most recent 500 days. This provides us
with 500 alternative scenarios for what can happen between today and tomorrow.
Table 1: Data for VaR historical simulation calculation
Day Market Variable 1 Market Variable 2  Market Variable N
0 20.33 0.1132  65. 37
1 20.78 0.1159  64.91
2 21.44 0.1162  65.02
    
499 25.75 0.1323  61.99
500 25.85 0.1343  62.10
Step 3: Generate Scenarios
Table 2: Scenarios generated for tomorrow (Day 501) using data in above
Scenario Market Market  Market Portfolio
Number Variable 1 Variable 2 Variable N value
($ millions)
1 26.42 0.1375  61.66 23.71
2 26.67 0.1346  62.21 23.12
     
499 25.88 0.1354  61.87 23.63
500 25.95 01363  62.21 22.87
Table 2 shows the values of the market variables tomorrow if their percentage changes between today and tomorrow
are the same as they were between Day i – 1 and Day i for 1  i  500.
If we define vi as the value of a market variable on Day i and suppose that today is Day m, them the ith scenario
vi
assumes that the value of the market variable tomorrow will be vm  .
vi -1
Example: In table 1, m = 500. The value for variable 1 today is v500, is 25.85.
The values for variable 1 at time (0) and time (1) are v0 = 20.33 and v1 = 20.78.
20.78
Thus, the value of the first market variable in the first scenario is 25.85   26.42
20.33
Step 4: Compute the change in the value of the portfolio between today and tomorrow for all scenarios.
The final column of Table 2 shows the value of the portfolio tomorrow for each of the 500 scenarios.
Assume the value of the portfolio today is $23.50 million. Then the change in the value of the portfolio between today
and tomorrow, for all scenarios can be computed:
 For Scenario 1, it is +$210,000 (23.71M – 23.5M).
 For Scenario 2, it is –$380,000 (23.12M – 23.5M), etc.
Step 5: To compute the one–day 99% VaR, rank the changes in value and select the fifth–worst loss.
Final notes: Each day the VaR estimate is updated using the most recent 500 days of data. Thus, on Day 502 we use
data from Day 2 to Day 502 to determine VaR, and so on.
18.3 Model Building Approach

Before discussing the details of the approach, one issue concerning the units for measuring volatility will be described
below.
Daily Volatilities
Units for measuring volatility:
a. In option pricing, the volatility of an asset is usually quoted as a “volatility per year”.
b. In VaR calculations, the volatility of an asset is usually quoted as a “volatility per day”.
The Relationship between the volatility per year (used in option pricing) and the volatility per day (used in
VaR calculations).  yr   day  252 , so daily volatility is about 6% ( 1  252 ) of annual volatility.

Assume there are 252 trading days in a year.

Symbol Description
yr the volatility per year of a certain asset.
the volatility per day of a certain asset. This is approximately equal to the standard deviation of the
day asset’s daily return.
for the purposes of calculating VaR we assume exact equality.
 yr   day  252 , so daily volatility is about 6% ( 1  252 ) of annual volatility.
A. How VaR can be calculated in the single–asset case.
Given: A portfolio consists of $10 million in shares of IBM. The volatility of IBM is 2% per day.
Determine: A 99% confidence level for losses over 10 days (i.e. N = 10 and X = 99) (See C. below)
Notes:
 the standard deviation of daily changes in the value of the position is 2% × $10 million ($200,000).
 the standard deviation of a 10 day period change in the value of the position equals
10 × (the change over a 1 day period).
Thus, the standard deviation of the change in the value of the IBM portfolio over a 10–day period is $632,456
( 10  200, 000 ), assuming that the changes on successive days are independent.
B. Expected values:
1. In VaR calculations, it is assumed that the expected change in the price of a market variable over the time period is
zero.
2. Example: Given IBM has an expected return of 13% per annum:
a. For a 1–day period, the expected return is 0.13  252 (about 0.05%) where as the standard deviation of the
return of 2%.
b. Over a 10–day period, the expected return is 0.13  25.2 (about 0.5%) where as the standard deviation of the
return of 2 × 10 (about 6.3%).
C. Computation of the 10–day, 99% VaR:
1. The change in the value of the portfolio of IBM shares over a 10–day period has a standard deviation of $632,456
and (at least approximately) a mean of zero.
2. Assuming normally distributed changes and based on the Table for N(X), N(–2.33) = 0.01. This means:
 there is a 1% probability that a normally distributed variable will decrease in value by more than 2.33 standard
deviations.
 we are 99% certain that a normally distributed variable will not decrease in value by more than 2.33 standard
deviations.
The 10–day, 99% VaR for this portfolio is 2.33 × 632,456 = $1,473,621.
Next, assume a portfolio $5 million in AT&T, which has a daily volatility of 1% (16% per year).
 the standard deviation of the change in the value of the portfolio per 10 days is:
5,000,000 × 0.01 × 10 = 158,114.
 The 10–day, 99% VaR is 2.33 × 158,114 = $368,405 (assuming the change is normally distributed).
D. A Two–Asset Case:
Given: a portfolio consisting of $10 million of IBM shares and $5 million of AT&T shares.
Assume the returns on the two shares have a bivariate normal distribution with a correlation of 0.30.
Recall: if two variables X and Y have standard deviations equal to X and Y with the coefficient of correlation
between them being equal to , the standard deviation of X + Y is given by:
 X Y   X2   Y2  2    X  Y
Applying this to the two asset portfolio results in the following:
a. The standard deviation of the change in the value of the portfolio consisting of both stocks over a 1–day period
is 200,0002 50,0002  20.30200,00050,000  220,227
b. The 1–day 99% VaR for the portfolio is 2.33 × 220,227 = $513,129
c. The 10–day 99% VaR for the portfolio is 10 × $513,129 = $1,622,657
E. The Benefits of Diversification:
Less than perfect correlation leads to some of the risk being “diversified away”. In the example above:
 The10–day, 99% VaR for the portfolio of IBM shares is $1,473,621.

 The 10–day, 99% VaR for the portfolio of AT&T shares is $368,405.
 The 10–day, 99% VaR for the portfolio of both IBM and AT&T shares is $1,622,657.
Thus, the benefit of diversification equals (1,473,621 + 368,405) – 1,622,657 = $219,369
Perfect correlation leads to no risk being “diversified away”.
18.4 Linear Model

The example above illustrates that VaR calculations are straightforward when:
1. Considering a portfolio of assets; and
2. Changes in the values of asset prices have a multivariate normal distribution.
To generalize: Symbol Description
P Value of a portfolio.
n Number of assets in the portfolio.
i Amount invested in asset i, (1  i  n).
xi Return on asset i in one day.
a. the dollar change in the value of our investment in asset i in one day is i × xi.
n
b. P (the dollar change in the portfolio value in one day) = 
i 1
i  xi .
Example: In the previous two asset portfolio, 1 = 10, 2 = 5, and the dollar change in the portfolio value in one day is
equal to P = 10×x1 + 5×x2.
a. To calculate VaR we, therefore, need to calculate only the mean and standard deviation of P. Assume that the
expected value of each xi is zero. This implies that the mean of P is zero.
b. To calculate the standard deviation of P, let:
1. i represent the daily volatility of the ith asset, which is the s.d. of xi
2. ij represent the coefficient of correlation between returns on asset i and asset j (xi and xj).
n n
The variance of P, P, is given by:  P2  
i 1 j 1
i   i   j   i   j .
n n n
This equation can also be written as  P2  i2  i2  2   ij i  j  i  j
i 1 i 1 j i
Note: The standard deviation of the change over N days is  P  N .

The 99% VaR for an N–day time horizon is 2.33 ×  P  N .
Based on the previous example: 1 = 0.02, 1 = 0.01 and 12 = 0.30, and since 1 = 10, 2 = 5, then
 P2  102  0.022  52  0.012  2 10  5  0.3  0.02  0.01  0.0485 , and thus,
a. The standard deviation of the change in the portfolio value per day (in millions of dollars) is
 P  0.0485  0.220
b. The 10–day 99% VaR is 2.33 × 0.220 × 10 = $1.623 million.
Note: This agrees with the calculation in the previous section.
Handling Interest Rates
Because it is impossible to define a separate market variable for every single bond price or interest rate to which a
company is exposed, simplifications are necessary. One possibility is to assume that only parallel shifts in the yield
curve occur. In this case, it is only necessary to define only one market variable: the size of the parallel shift.
Approach 1: Duration: Assume: All shifts in a zero–coupon yield curve are small parallel shifts.
If so, use the approximate duration relationship: P = –DPy
Symbol Description
P Value of a portfolio, dependent on interest rates.
D the modified duration of the portfolio.
y the size of the parallel shift in the yield curve in one day.
P the resultant change in the portfolio value.
This approach does not usually give enough accuracy. Thus, the procedure usually followed is to choose as market
variables the prices of zero–coupon bonds with standard maturities:

Approach 2: Cash Flow Mapping

Given the limitation associated with approach 1, the procedure usually followed is to:
a. choose as market variables the prices of zero–coupon bonds with standard maturities, such as 1 month, 3 months, 6
months, 1 year, 2 years, 5 years, 7 years, 10 years, and 30 years.
b. take the cash flows from instruments in the portfolio and map them into cash flows occurring on the standard
maturity dates described above for the purposes of calculating VaR.
See Appendix 18-A for more information on cash flow mapping.
Applications of the Linear Model
A. The simplest application of the linear model is to a portfolio with no derivatives consisting of positions in stocks,
bonds, foreign exchange, and commodities.
The change in the value of the portfolio is linearly related to changes in the value of the underlying market
variables (stock prices, zero–coupon bond prices, exchange rates, and commodity prices).
B. The linear model can also accommodate a portfolio having derivatives. The examples described below can be
reduced to creating bond portfolios handled in the way described in the previous section.
Example 1. A forward contract to buy a foreign currency (the exchange of a foreign zero–coupon bond maturing at
time T, for a domestic zero–coupon bond maturing at time T).
When calculating VaR:
 the forward contract is treated as a long position in the foreign bond combined with a short position in the
domestic bond.
 Each bond can be handled in the way described in the previous section.
Example 2: An interest rate swaps (the exchange of a floating–rate bond for a fixed–rate bond).
 The fixed–rate bond is a regular coupon–bearing bond.
 The floating–rate bond is worth par just after the next payment date, and can be regarded as a zero–coupon bond
with a maturity date equal to the next reset date.
The Linear Model and Options
The linear model is only an approximation when the portfolio contains options.
Example: Consider a portfolio consisting of options on a single stock.
Symbol Description
P Value of a portfolio, dependent on interest rates.
S A single stock with current price S.
 the rate of change of the value of the portfolio with S.
P the dollar change in the portfolio in one day.
S the dollar change in the stock price in one day.
Note: It is only approximately true that  = (P /S), and thus P =  S.
Next, define x as the proportional change in the stock price in one day so that  x = (S /S).
Again, an approximate relationship between P and x is P = S  x.
Given a position in several underlying market variables that includes options, we derive an approximately linear
relationship between P and xi’s as
n
P   Si i xi (here Si is the value of the ith market variable and  i is the delta of the portfolio with respect to the
i 1
ith market variable).
n
This corresponds to P     x
i 1
i i with i = Si i
n n n
Example: Use  P2  i2  i2  2   ij i  j  i  j to calculate the standard deviation of P.
i 1 i 1 j i
Given: A portfolio consists of options on IBM and AT&T.
The options on IBM have a delta of 1,000 and the options on AT&T have a delta of 20,000.
n
The IBM share price is $120 and the AT&T share price is $30. Based on P   S  x , it is approximately true that
i 1
i i i
P  120 1,000  x1  30  20,000  x2  120,000  x1  600,000  x2 , where

x1 and x2 are the proportional changes in the prices of IBM and AT&T in one day and P is the resultant change in
the value of the portfolio.
Assuming that the daily volatility of IBM is 2% and the daily volatility of AT&T is 1% and the correlation between the
daily changes is 0.3,
a. the standard deviation of P (in thousands of dollars) is:
 P  (1200.02)2 (6000.01)2 20.31200.026000.01  7.099
b. the 5–day 95% value at risk is 1.65  5  7,099  26,193 . Note N (–1.65) = 0.05
18.5 Quadratic Model

When a portfolio includes options, the linear model does not take account of the gamma of the portfolio.
Definitions:
a. Delta is the rate of change of the portfolio value with respect to an underlying market variable.
b. Gamma is the rate of change of the delta with respect to the market variable.
It measures the curvature of the relationship between the portfolio value and an underlying market variable.
Refer to Figure 16.3 Probability distribution for value of portfolio (a) positive gamma, (b) negative gamma (see page
356).
When gamma is positive, the probability distribution of P tends to be positively skewed (and vice versa). Figures
16.4 and 16.5 illustrate the reason for this result.
Figure 16.4 Translation of normal probability distribution for asset price into probability distribution for a long call on
an asset (see page 357).
 A long call is an example of an option position with positive gamma.
 In this case, when the probability distribution for the price of the underlying asset is normal, the probability
distribution for the option price is positively skewed.
Figure 16.5 Translation of normal probability distribution for asset price into probability distribution for a short call on
an asset (see page 357).
 A short call position has negative gamma.
 In this case, when the probability distribution for the price of the underlying asset is normal, the probability
distribution for the option price is negatively skewed.
Impact on VaR computations when a Portfolio includes Options
a. The VaR for a portfolio is critically dependent on the left tail of the probability distribution of P.
 When using a 99% confidence level, the VaR is the value in the left tail below which there is only 1% of the
distribution.
 As previously mentioned, a positive gamma portfolio (e.g. a long call) tends to have a thinner left tail than the
normal distribution.
If we assume the distribution is normal, we will tend to calculate a VaR that is too high. Similarly, a negative
gamma portfolio tends to have a fatter left tail than the normal distribution. If we assume the distribution is normal
we will tend to calculate a VaR that is too low.
b. A more accurate estimate of VaR (than given by the linear model) can be obtained by using both delta and gamma
measures to relate P to the xi’s.
Example: A portfolio consists of a single asset whose price is S. Let the delta of a portfolio equal  and its gamma
equal .
Using the Taylor series expansion, an improvement over the approximation in the equation P =  S is
1 S 1
P  S    S  . Since x  , we see that P  Sx  S 2  x  .
2 2
2 S 2
18.6 Monte Carlo Simulation

Monte Carlo simulation can also be used to generate the probability distribution for  P.
Example: To calculate a one–day VaR for a portfolio, perform the following steps:
Step 1. Value the portfolio today using the current values of market variables.
Step 2. Sample once from the multivariate normal probability distribution of the xi’s
Step 3. Use the values of the sampled xi’s to determine the value of each market variable at the end of one day.
Step 4. Revalue the portfolio at the end of the day.

Step 5. Subtract the value calculated in Step 1 from the value in Step 4 to determine a sample P.
Step 6. Repeat Steps 2 – 5 many times to build up a probability distribution for P.
Example: Assume 5,000 different sample values of  P are computed as described above.
The 1–day 99% VaR is the value of P for the 50th (0.01 × 5,000) worst outcome.
The 1–day 95% VaR is the value of P for the 250th (0.05 × 5,000) worst outcome.
The N–day VaR is usually assumed to be the 1–day VaR multiplied by N
The drawback of Monte Carlo simulation is that it tends to be slow because a company’s complete portfolio has to be
revalued many times. A partial remedy to this problem is to use the partial simulation approach.
n n n
By assuming that P   Si i xi   Si S j ij xi x j describes the relationship between P and the xi’s
i 1 i 1 j 1
Steps 3 and 4 can be eliminated which avoids the need for a complete revaluation of the portfolio.
18.7 Comparison of Approaches

Two methods for estimating VaR have been discussed: the historical simulation approach and the model–building
approach.
Advantage and Disadvantages to using the Model Building Approach:
Advantages Disadvantages
Results can be produced 1. It assumes that the market variables have a multivariate normal distribution. In
quickly. practice, daily changes in market variables often have distributions that are quite
different from normal.
2. It tends to give poor results for low–delta portfolios
Advantage and Disadvantages to using the Historical Simulation Approach:
Advantages Disadvantages
1. Historical data determines the joint probability It is computationally slow and does not easily allow
distribution of the market variables. volatility updating schemes to be used.
2. It avoids the need for cash–flow mapping.
18.8 Stress Testing and Back Testing

STRESS TESTING
In addition to calculating a VaR, companies may perform a stress test of their portfolio.
1. Stress testing involves estimating how the portfolio would have performed under extreme market moves.
For example, to test the impact of an extreme movement in U.S. equity prices, a company might set the
proportional changes in all market variables equal to:
i. those that existed on October 19, 1987 (when the S&P 500 moved by 22.3 standard deviations), or
ii. those that existed on January 8, 1988 (when the S&P 500 moved by 6.8 standard deviations).
2. Stress testing can take into account extreme events that do occur from time to time but that are virtually impossible
according to the probability distributions assumed for market variables.
For example, consider a 5–standard–deviation daily move in a market variable.
This happens about once every 7,000 years under a normal distribution but, in practice, it is not uncommon to see
a 5–standard–deviation daily move once or twice every 10 years.
BACK TESTING
1. Back testing involves testing how well the VaR estimates would have performed in the past.
2. For example, assume a 1–day 99% VaR is being computed. Back–test by determining how often the loss in a day
exceeded the 1–day 99% VaR.
i. If this happened 1% of the days, the methodology for calculating VaR may be appropriate.
ii. If this happened on 10% of days, the methodology is suspect.
18.9 Principal Components Analysis (PC)

PC analysis is used to handle risk arising from groups of highly correlated market variables.
The analysis takes historical data on movements in the market variables and attempts to define a set of components or
factors that explain the movements.
For example, consider 10 U.S. Treasury rates with maturities between three months and 30 years.
 The tables below show results using 1,543 daily observations between 1989 and 1995.

 The first column shows the maturities of the rates, and the remaining 10 columns show the ten factors (or principal
components) describing the rate.
Factor Loadings (interest rate moves for a particular factor) for U.S. Treasury Data
PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10
3 mo. 0.21 –0.57 0.50 0.47 –0.39 –0.02 0.01 0.00 0.01 0.00
6 mo. 0.26 –0.49 0.23 –0.37 0.70 0.01 –0.04 –0.02 –0.01 0.00
12 mo. 0.32 –0.32 –0.37 –0.58 –0.52 –0.23 –0.04 –0.05 0.00 0.01
2 yr. 0.35 –0.10 –0.38 0.17 0.04 0.59 0.56 0.12 –0.12 –0.05
3 yr. 0.36 0.02 –0.30 0.27 0.07 0.24 –0.79 0.00 –0.09 –0.00
4 yr. 0.36 0.14 –0.12 0.25 0.16 –0.63 0.15 0.55 –0.14 –0.08
5 yr. 0.36 0.17 –0.04 0.14 0.08 –0.10 0.09 –0.26 0.71 0.48
7 yr. 0.34 0.27 0.15 0.01 0.00 –0.12 0.13 –0.54 0.00 –0.68
10 yr. 0.31 0.30 0.28 –0.10 –0.06 0.01 0.03 –0.23 –0.63 0.52
30 yr. 0.25 0.33 0.46 –0.34 –0.18 0.33 –0.09 0.52 0.26 –0.13
The factor loadings have the property that the sum of their squares for each factor is 1.0.
Factor score defined: The factor score for a day is defined as the quantity of a particular factor in the interest rate
changes on a particular day.
Standard Deviation of Factor Scores (measured in basis points (bps)) listed in order of importance
PC1 PC2 PC3 PC4 PC5 PC6 PC7 PC8 PC9 PC10
17.49 6.05 3.10 2.17 1.97 1.69 1.27 1.24 0.80 0.79
Interpretation and analysis of Principle Component (PC) Factors shown above
The first factor in the first table (column PC1) corresponds to a roughly parallel shift in the yield curve.
Example: For 1 unit of factor 1, the 3–month rate increases by 0.21 basis points, the 6–month rate increases by 0.26
basis points, and so on.
The second factor (shown in the column labeled PC2) corresponds to a “twist” or “steepening” of the yield curve.
Example: Rates between 3 months and 2 years move in one direction; rates between 3 years and 30 years move in the
other direction.
The third factor corresponds to a “bowing” of the yield curve.
Example: Rates at the short end and long end of the yield curve move in one direction; rates in the middle move in the
other direction.
Interest rate changes, for any given day, can be expressed as a linear sum of the factors by solving a set of ten
simultaneous equations (note there are 10 rates and 10 factors).
Notes on the Standard Deviation of Factor Scores:
 The importance of a factor is measured by the standard deviation (SD) of its factor score.
 A 1 SD move in the first factor corresponds to a 3.67 (0.21 × 17.49) bps move in the 3–month rate, a 4.55 (0.26 ×
17.49) bps move in the 6–month rate, and so on.
The technical details of how the factors are determined are discussed by the authors. It is sufficient for us to note that
the factors are chosen so that the factor scores are uncorrelated.
For example, it is important to note that the first factor score (amount of parallel shift) is uncorrelated with the second
factor score (amount of twist) across the 1,543 days.
 The variances of the factor scores add up to the total variance of the data. Thus, the total variance of the original
data is 17.492 + 6.052 + 3.102 + ... + 0.792 = 367.9
Most of the risk in interest rate moves is accounted for by the first two or three factors:
17.49 2
the first factor accounts for  83.1% of the variation in the original data.
367.9
17.49 2  6.052
the first two factors account for  93.1% of the variation in the data.
367.9
Thus, we can relate the risks in a portfolio of interest rate dependent instruments to movements in these factors (instead
of considering all ten interest rates).
Using Principal Components Analysis to Calculate VaR
Example: Assume we have a portfolio, with exposures shown in the table below, to interest rate moves.
Change in Portfolio Value for a 1 Basis Point Rate Move ($ millions)
1 yr rate 2 yr rate 3 yr rate 4 yr rate 5 yr rate
+10 +4 –8 –7 +2

The values in the table above can be interpreted as follows:

 A 1-basis change in the one–year rate causes the portfolio value to increase by $10 million
 A 1-basis point change in the two–year rate causes it to increase by $4 million; and so on.
Since over 90% of the uncertainty in interest rate moves are determined by the first two factors, we can use these
factors to model interest rate moves.
Note: The data in the table below is an excerpt from the data on the previous page in your manual.
Factor Loadings (interest rate moves for a particular factor) for U.S. Treasury Data
PC1 PC2 PC3
3 mo. 0.21 –0.57 0.50
6 mo. 0.26 –0.49 0.23
12 mo. 0.32 –0.32 –0.37
2 yr. 0.35 –0.10 –0.38
3 yr. 0.36 0.02 –0.30
4 yr. 0.36 0.14 –0.12
5 yr. 0.36 0.17 –0.40
Using the data in the tables above, our exposure (measured in millions of dollars per factor score basis point) to the
first and second factors can be computed as follows:
exposure to the first factor equals 10× (0.32) + 4 × (0.35) – 8 × (0.36) – 7 × (0.36) + 2 × (0.36) = –0.08
exposure to the second factor equals 10 × (–0.32) + 4 × (–0.10) – 8 × (0.02) – 7 × (0.14) + 2 × (0.17) = –4.40
Let f1 and f2 be the factor scores (measured in basis points). The approximate change in the portfolio value is given by
P = –0.08×f1 – 4.40×f2. Since the factor scores are uncorrelated and using the standard deviations previously given,
we can compute the 1–day 99% VaR
The standard deviation of P  0.082  17.492 + 4.402  6.052  26.66
The 1–day 99% VaR is 26.66 × 2.33 = 62.12.
Comments on the 1–day 99% VaR computation:
a. There is little exposure to the first factor (–0.08) and significant exposure to the second factor (–4.40).
b. Using only a one factor would significantly understate VaR.
c. The duration–based analysis (described in section 16.4 of the manual) would also significantly understate VaR as
it considers only parallel shifts in the yield curve.
Other Uses of Principal Components Analysis:
A financial institution may have exposures to a number of different stock indices. By identifying factors that describe
movements in the indices, a principal components analysis can be used to replace the market indices in a VaR analysis.
Final Notes:
 The effectiveness of a principal components analysis depends on how closely correlated the group of market
variables are to one another.
 VaR relates actual changes in a portfolio to percentage changes in market variables (the xi’s).
 Principal components analysis is carried out on percentage changes in market variables rather than on their actual
changes.
18.10 Appendix 18A: Cash Flow Mapping

A procedure for mapping cash flows to standard maturity dates is shown below. Consider a portfolio consisting of a
single bond of $1million maturing in 0.8 years, having a coupon of 10% per annum payable semiannually.
a. the coupon payments of $50,000 in 0.3 years and 0.8 years, and a principal payment of $1million in 0.8 years.
b. the bond can be ‘“reconstructed’ using a 0.3–year zero–coupon bond with a principal of $50,000 and a “position”
in a 0.8–year zero–coupon bond with a principal of $1,050,000.
How to adjust to non–typical time periods:
a. The position in the 0.3–year zero–coupon bond is mapped into an equivalent position in three–month and six–
month zero–coupon bonds.
b. The position in the 0.8–year zero–coupon bond is mapped into an equivalent position in six–month and one–year
zero–coupon bonds.
The result is that the position in the 0.8–year coupon–bearing bond is, for VaR purposes, regarded as a position in
zero–coupon bonds having maturities of three months, six months, and one year.
Use as market variables the prices of zero–coupon bonds with standard maturities.

a. Choose one month, three months, one year, two years, five years, seven years, ten years, and thirty years
maturities.
b. Regard each T–Bond as a portfolio of its constituent zero–coupon bonds. The position in each of the zero–coupon
bonds is then mapped into an equivalent position in the adjacent standard–maturity zero–coupon bonds.
The Mapping Procedure
Assume that $1,050,000 will be received in 0.8 years. Further assume we are given the following data:
Maturity 3–month 6–month 1 –year
Zero rate 5.50 6.00 7.00
Bond price volatility (% per day) 0.06 0.10 0.20
Correlation between 3–month 6–month 1–year

daily returns bond bond bond
3–month bond 1.0 0.9 0.6
6–month bond 0.9 1.0 0.7
1–year bond 0.6 0.7 1.0
Step 1: Interpolate between the 6–month rate of 6.0% and the 1–year rate of 7.0%:
0.40 × (6.0%) + 0.60 × (7.0%) = 6.6%
1,050,000
The present value of the $1,050,000 cash flow to be received in 0.8 year is  997,662 .
1.0660.80
Step 2: Interpolate between the 0.1% volatility for the 6–month bond and the 0.2% volatility for the 1–year bond.
0.40 × (0.001) + 0.60 × (0.002) = 0.0016 = 0.16% volatility for the 0.8–year bond.
Step 3: Allocate  of the present value (997,622) to the six–month bond and (1 –  ) of the present value to the 1–
n n n
year bond. Using the equation  P2  i2  i2  2   ij  i   j  i  j , and matching variances
i 1 i 1 j i
we obtain 0.0016  0.001   0.002 (1   )  2  0.70  0.001  0.002   (1   )

2 2 2 2
Solving this quadratic equation, one obtains  = 0.3203 (thus, 32.03% of $997,622 should be allocated to a
6–month zero–coupon bond and 67.97% of the value should be allocated to a 1–year zero–coupon bond.)
Step 4: Replace the 0.8–year bond worth $997,662 with the following:
a. a 6-month bond worth 997,662 × 0.3203 = $319,589, and a
b. one–year bond worth 997,662 × 0.6797 = $678, 073
Step 5: For the $50,000 cash flow received at time 0.3 years, a similar mapping to a 3 month and 6 month security
can take place.
The results of the calculations are shown below.
$50,000 received $1,050,000 received
in 0.3 years in 0.8 years Total
Position in 3–month bond ($) 37,397 37,397
Position in 6–month bond ($) 11,793 319,589 331,382
Position in 1–year bond ($) 678,074 678,074
Notes:
a. Carry out similar calculations for the $50,000 cash flow. It can be mapped into a position worth $37,397 in a
three–month bond, a position worth $11,793 in a six–month bond
b. Use the following to determine the variance of the change in the price of the 0.8–year bond with n = 3, 1 =
37,397, 2 = 331,382, 3 = 678,074,  1 = 0.0006,  2 = 0.001,  3 = 0.002, 12= 0.9, 13 = 0.6, and  13 = 0.70. The
variance is 2,628,518. The standard deviation of the change in the price of the bond is 2,628,519  1,621 . The
10–day 99% VaR is 1,621 × 10 × 2.33 = 11,946 or about $11,950.
Advantages of cash flow mapping:
1. it preserves both the value and the variance of the cash flow.
2. the weights assigned to two adjacent zero–coupon bonds are always positive.

Chapter 20: Credit Risk
Section Description
20.0 Introduction
20.1 Credit Ratings
20.2 Historical Default Probabilities
20.3 Recovery Rates
20.4 Estimating Default Probabilities From Bond Prices
20.5 Comparison Of Default Probability Estimates
20.6 Using Equity Prices To Estimate Default Probabilities
20.7 Credit Risk In Derivatives Transactions
20.8 Credit Risk Migration
20.9 Default Correlation
20.10 Credit VaR
20.0 INTRODUCTION
Significant credit risk for banks and other financial institutions arise from defaults by borrowers, counterparties in
derivatives transactions, and bond issuers.
This chapter focuses on the following key issues associated with credit risk.
a. Different approaches to estimating the probability of default are discussed.
b. The key difference between risk–neutral and real–world probabilities of default is explained.
c. The nature of the credit risk in over–the–counter derivatives transactions and the clauses derivatives dealers write
into their contracts to reduce credit risk.
d. Default correlation, Gaussian copula models, and the estimation of credit value at risk.
20.1 CREDIT RATINGS

Rating agencies rate the creditworthiness of corporate bonds.
Rating Agency Investment Grade Non–investment Grade Finer detail measures
Moody’s Aaa, Aa, A, Baa Ba, B, and Caa. Aa1, Aa2, and Aa3, etc
S&P AAA, AA, A, BBB BB, B, and CCC AA+, AA, and AA–, etc.
Notes: Aaa bonds are considered to have almost no chance of defaulting.
Only Aaa category for Moody's and the AAA category for S&P are not subdivided.
20.2 HISTORICAL DEFAULT PROBABILITIES

Example:
a. A Baa bond has a 0.20% chance of defaulting by the end of the 1st year, a 0.57% chance of defaulting by the end
of the 2nd year, etc.
b. The probability of a bond defaulting during a particular year can be calculated from the table as follows: the
probability of a Baa–rated bond defaulting during the 2nd year is 0.57 – 0.20 = 0.27%.
 For investment grade bonds the probability of default in a year tends to be an increasing function of time. Time
must elapse for the fortunes of the issuer to decline to such an extent that a default happens.
 For bonds with a poor credit rating, the reverse is often true (e.g. the probabilities that a Caa–rated bond will
default during years 1, 2…5 are 23.65%, 13.55%,…, 5.27%, respectively). The next year or two may be
critical. If the issuer survives this period, its financial health is likely to have improved.
Default probabilities can be estimated from historical data.
Table 1 Average cumulative default rates: 1970–2003. (Source: Moody's)
1 2 3 4 5 7 10 15 20
Aaa 0.00 0.00 0.00 0.04 0.12 0.29 0.62 1.21 1.55
Aa 0.02 0.03 0.06 0.15 0.24 0.43 0.68 0.08 2.70
A 0.02 0.09 0.23 0.38 0.54 0.91 1.59 2.94 5.24
Baa 0.20 0.57 1.03 1.62 2.16 3.24 5.10 9.12 12.59
Ba 1.26 3.48 6.00 8.59 11.17 15.44 21.01 30.88 38.56
B 6.21 13.76 20.65 26.66 31.99 40.79 50.02 59.21 60.73
Caa 23.65 37.20 48.02 55.56 60.83 69.36 77.91 80.23 80.23

Default Intensities
From Table 1:
 The unconditional probability of a Caa bond defaulting during the 3rd year is 48.02 – 37.20 = 10.82%
 The probability that the Caa–rated bond will survive until the end of year 2 is 100 – 37.20 = 62.80%.
 The probability that it will default during the third year conditional on no earlier default is therefore
0.1082/0.6280, or 17.23%. Conditional default probabilities are referred to as default intensities or hazard rates.
The 17.23% just calculated is for a 1–year time period. Consider a short time period of length t.
 The default intensity (t) at time t is defined so that (t) t is the probability of default between time t and t + t,
conditional on no earlier default.
 If V(t) is the cumulative probability of the company surviving to time t (i.e., no default by time t), then V(t + t) –
V(t) = –(t) V(t) t
t
 Taking limits
dV (t )   (t )V (t ) from which we get V (t )  1  e   ( ) d
0
dt
t
Define Q(t) as the probability of default by time t. It follows that Q (t )  1  e 0

  ( ) d
 , or Q (t )  1  e  ( t )t eq (20.1)
where  (t )t is the average default intensity between time 0 and time t.
20.3 RECOVERY RATES

When bankruptcy occurs, assets are sold by the liquidator and the proceeds are used to payoff debt holder claims (as
far as possible and in order of entitlement).
Table 2 Percent recovered on corporate bonds as a percentage of face value (Moody's)
Average Recovery Rate (%)
Senior secured 51.6
Senior unsecured 36.1
Senior subordinated 32.5
Subordinated 31.1
Junior subordinated 24.5
Note: senior secured debt holders received an average of 51.6 cents per dollar of face value while junior subordinated
debt holders received an average of only 24.5 cents per dollar of face value.
In the U.S., the bondholder would make a claim for the bond's face value plus accrued interest.
Recovery rates are significantly negatively correlated with default rates.
Moody's looked at average recovery rates and average default rates each year between 1982 and 2003. It found that the
following relationship provides a good fit to the data:
Average recovery rate = 50.3 – 6.3 × Average default rate, where the average recovery rate and the average default rate
are measured as percentages.
20.4 ESTIMATING DEFAULT PROBABILLITIES FROM BOND PRICES

The probability of company default can be estimated from the prices of bonds it has issued. Assume that that the only
reason a corporate bond sells for less than a similar risk–free bond is the possibility of default.
An approximate calculation:
Suppose that a bond yields 200 basis points more than a similar risk–free bond. The holder of a corporate bond must
be expecting to lose 200 basis points (or 2% per year) from defaults.
 The expected recovery rate in the event of a default is 40%.
 An estimate of the probability of default per year conditional on no earlier default 0.02/(1 – 0.4) = 3.33%.
s
 In general, h  , (eq 20.2) where h is the default intensity per year, s is the spread of the corporate bond yield
1 R
over the risk–free rate, and R is the expected recovery rate.
A More Exact Calculation
Suppose a corporate bond lasts for 5 years, provides a coupon 6% per annum (paid semiannually) and that the yield on
the corporate bond is 7% per annum (with continuous compounding). The yield on a similar risk–free bond is 5%
(with continuous compounding).
 The yields imply that the price of the bond is 95.34 and the price of the risk–free bond is 104.09.
 The expected loss from default over the 5–year life of the bond is therefore 104.09 – 95.34 = $8.75.

Suppose that the probability of default per year (assumed in this example to be the same each year) is Q.
Table 3 below calculates the expected loss from default in terms of Q assuming:
 Defaults can happen at times 0.5, 1.5, 2.5, 3.5, and 4.5 years (immediately before coupon payment dates).
 Risk–free rates for all maturities are 5% (with continuous compounding).
Table 3: Calculation of loss from default on a bond in terms of the default probabilities per year, Q.
Notional principal = $100.
Time Default Recovery Risk–free Loss given Discount PV of Expected
(years) probability amount ($) value ($) default ($) factor Loss ($)
0.5 Q 40 106.73 66.73 0.9753 65.08Q
1.5 Q 40 105.97 65.97 0.9277 61.20Q
2.5 Q 40 105.17 65.17 0.8825 57.52Q
3.5 Q 40 104.34 64.34 0.8395 54.01 Q
4.5 Q 40 103.46 63.46 0.7985 50.67Q
Total 288.48Q
Consider the 3.5 year row in the table above.
 The expected value of the risk–free bond at time 3.5 years (calculated using forward interest rates) is
3 + 3e–0.05×0.5 + 3e–0.05×1.0 + 103e–0.05×1.5 = 104.34
 The amount recovered if there is a default is 40, so that the loss given default is 104.34 – 40 = $64.34. The present
value of this loss is 54.01. The expected loss is therefore 54.01Q.
 The total expected loss is 288.48Q.
 Thus, 288.48Q = $8.75; Q = 3.03%.
The calculations just performed assume that:
 the default probability is the same in each year and
 defaults take place at just one time during the year.
Now, extend the calculations to assume:
 that defaults can take place more frequently.
 a constant default intensity or a particular pattern for the variation of default probabilities with time, instead of
assuming a constant unconditional probability of default.
With several bonds, several parameter estimates can be computed to determine the term structure of default
probabilities. Example: Suppose we have bonds maturing in 3, 5, 7, and 10 years.
 Use the first bond to estimate a default probability per year for the first 3 years,
 Use the second bond to estimate default probability per year for years 4 and 5,
 Use the third bond to estimate a default probability for years 6 and 7, and
 Use the fourth bond to estimate a default probability for years 8, 9, and 10 (see problems 20.15 and 20.27).
This is analogous to the bootstrap procedure in Section 4.5 for calculating a zero–coupon yield curve.
The Risk–Free Rate
When bond prices are used to estimate default probabilities, what is the meaning of the terms “risk–free rate” and
“risk–free bond”? The answer differs, as explained below:
 the spread s is the excess of the corporate bond yield over the yield on a similar T–bond (see eq. 20.2).
 the risk–free value of the bond must be calculated using the risk–free rate (see table 3).
 the benchmark risk–free rate usually used in quoting corporate bond yields is the yield on similar Treasury bonds
(e.g. a bond trader might quote the yield on a particular corporate bond as being a spread of 250 basis points over
Treasuries).
 traders use LIBOR/swap rates as proxies for risk–free rates when valuing derivatives (see section 4.1). traders also
use LIBOR/swap rates as risk–free rates when calculating default probabilities. Example: the spread s in eq. 20.2 is
the spread of the bond yield over the LIBOR/swap rate when they determine default probabilities from bond prices
 the risk–free discount rates used in the calculations in Table 3 are LIBOR/swap zero rates.
Credit default swaps can be used to imply the risk–free rate assumed by traders.
 The rate used is approximately equal to the LIBOR/swap rate minus 10 basis points on average, and is plausible.
 The credit risk in a swap rate is the credit risk from making a series of 6–month loans to AA–rated counterparties
and 10 basis points is a reasonable default risk premium for a AA–rated 6–month instrument (see Section 7.5).
Asset Swaps
Traders often use asset swap spreads to extract default probabilities from bond prices.
Asset swap spreads provide a direct estimate of the spread of bond yields over the LIBOR/swap curve.

Consider an asset swap spread for a particular bond as 150 basis points. There are three possible situations:
1. The bond sells for its par value of 100. The swap involves one side (company A) paying the coupon on the bond and
the other side (company B) paying LIBOR plus 150 basis points.
2. The bond sells below its par value (e.g. for 95). The swap is structured so that in addition to the coupons, company
A pays $5 per $100 of notional principal at the outset.
3. The underlying bond sells above par (e.g. for 108). Company B then makes a payment of $8 per $100 of principal at
the outset.
Thus, the present value of the asset swap spread is the amount by which the price of the corporate bond is exceeded by
the price of a similar risk–free bond where the risk–free rate is assumed to be given by the LIBOR/swap curve (see
Problem 20.24).
Consider the example in Table 3 where the LIBOR/swap zero curve is flat at 5%.
Suppose it is known that the asset swap spread is 150 basis points (instead of knowing the bond's price).
Thus, the amount by which the value of the risk–free bond exceeds the value of the corporate bond is the present value
of 150 basis points per year for 5 years.
 Assuming semiannual payments, this is $6.55 per $100 of principal.
 The total loss in Table 3, in this case, would be set equal to $6.55.
 The default probability per year is Q = 6.55/288.48 = 2.27%.

Default probabilities estimated from historical data are much less than those derived from bond prices.
Table 4 shows, for companies that start with a particular rating, the average annual default intensity over 7 years
calculated from (a) historical data and (b) bond prices.
The calculation of default intensities using historical data are based on eq (20.1) and Table 1.
1
From eq (20.1), we have  (7)  ln[1  Q (7)] , where  (t ) is the average default intensity (or hazard rate) by time t
7
and Q(t) is the cumulative probability of default by time t.
The values of Q(7) are taken directly from Table 1. Consider an A–rated company.
 The value of Q(7) is 0.0091.
1
 The average 7–year default intensity is  (7)  ln[0.9909]  0.0013
7
The calculations using bond prices are based on eq (20.2) and bond yields published by Merrill Lynch.
The results shown are averages between December 1996 and July 2004. The recovery rate is assumed to be 40% and
the risk–free interest rate is assumed to be the 7–year swap rate minus 10 basis points.
Example:
 For A–rated bonds, the average Merrill Lynch yield was 6.274%.
 The average swap rate was 5.605%, so the average risk–free rate was 5.605% – 0.10% = 5.505%.
Thus, the average 7–year default probability is (0.06274 – 0.05505)/(1.0 – 0.4) = 0.0128 = 1.28%.
Table 4 shows:
 the ratio of the default probability (backed out of bond prices) to default probability (calculated from historical
data) declines as the credit quality declines with the ratio very high for investment grade companies.
 the difference between the two default probabilities increases as credit quality declines.
Table 4: Seven–year average default intensities (% per annum)

Historical Default intensity
Rating default intensity from bonds Ratio Difference
Aaa 0.04 0.67 16.80 0.63
Aa 0.06 0.78 13.00 0.72
A 0.13 1.28 9.80 1.15
Baa 0.47 2.38 5.10 1.91
Ba 2.47 5.07 2.10 2.67
B 7.69 9.02 1.20 1.53
Caa 16.90 21.30 1.30 4.40
Table 5 provides another way of looking at these results.

It shows the excess return over the risk–free rate (still assumed to be the 7–year swap rate minus 10 basis points)
earned by investors in bonds with different credit rating.
Table 5: Expected excess return on bonds (basis points)
Rating Bond yield spread over Spread of risk–free rate Spread for historical Expected excess
Treasuries over Treasuries defaults return
Aaa 83 43 2 38
Aa 90 43 4 43
A 120 43 8 69
Baa 186 43 28 115
Ba 347 43 144 160
B 585 43 449 93
Caa 1,321 43 1,014 264
Consider again an A–rated bond.
 The average spread over Treasuries is 120 basis points.
 Of this, 43 basis points are accounted for by the average spread between 7–year Treasuries and our proxy for the
risk–free rate.
 A spread of 8 basis points is necessary to cover expected defaults. (This equals the real–world probability of
default from Table 4 times 1 minus the assumed recovery rate of 0.4.)
Thus, an expected excess return (after expected defaults have been taken into account) equals 69 basis points.
Tables 4 and 5 show that a large percentage difference between default probability estimates translates into a small
(but significant) expected excess return on the bond.
 For Aaa–rated bonds, the ratio of the two default probabilities is 16.8, but the expected excess return is only 38
basis points.
 The expected return tends to increase as credit quality declines.
Real–World vs. Risk–Neutral Probabilities
Default probabilities implied from bond yields are risk–neutral probabilities of default.
Consider the calculations of default probabilities in Table 3.
 The calculations assume that expected default losses can be discounted at the risk–free rate.
 The risk–neutral valuation principle shows that this is valid provided the expected losses are calculated in a risk–
neutral world (i.e. the default probability Q in Table 3 must be a risk–neutral probability).
By contrast, the default probabilities implied from historical data are real–world default probabilities (a.k.a. physical
probabilities). The expected excess return in Table 5 arises from the difference between real–world and risk–neutral
default probabilities.
If there were no expected excess return, then the real–world and risk–neutral default probabilities would be the same,
and vice versa. Why are there big differences between real–world and risk–neutral default probabilities? Stated
differently, why do corporate bond traders earn more than the risk–free rate on average?
Potential reasons include:
2. Corporate bonds are relatively illiquid and bond traders demand an extra return in compensation.
3. Subjective bond trader default probabilities may be higher than those in Table 1 (e.g. bond traders may be allowing
for depression scenarios worse than seen during 1970 to 2003).
4. Bonds do not default independently of each other.
 This is the most important reason for the results in Tables 4 and 5.
 There are periods of time when default rates are very low and periods when they are very high.
 This gives rise to systematic risk (i.e., risk that cannot be diversified away) and bond traders should require an
expected excess return for bearing the risk.
 The variation in yearly default rates may be due to overall economic conditions or because a default by one
company has a ripple effect resulting in defaults by other companies (the latter referred to as credit contagion).
5. Bond returns are highly skewed with limited upside.
 Thus, it is more difficult to diversify risks in a bond portfolio than in an equity portfolio.
 Since a large number of different bonds must be held, and since many bond portfolios are not fully diversified,
bond traders may require an extra return for bearing unsystematic risk in addition to the systematic risk
mentioned above.

Should real–world or risk–neutral default probabilities be used in the analysis of credit risk?
The answer depends on the purpose of the analysis.
 When valuing credit derivatives or estimating the impact of default risk on the pricing of instruments, use risk–
neutral default probabilities.
The analysis calculates the present value of expected future cash flows and almost invariably (implicitly or
explicitly) involves using risk–neutral valuation.
 When carrying out scenario analyses to calculate potential future losses from defaults, use real–world default
probabilities.
20.6 USING EQUITY PRICES TO ESTIMATE DEFAULT PROBABILITIES

The problem with relying on a company’s credit rating to estimate its probability of default is that credit ratings are
revised infrequently.
Some argue that equity prices provide more up–to–date information for estimating default probabilities.
Merton’s model assumes a company's equity is an option on the company’s assets. Suppose a firm has one zero–
coupon bond outstanding and that the bond matures at time T. Define:
V0: Value of company's assets today D: Amount of debt interest and principal due to be repaid at time T
VT: Value of company's assets at time T  V: Volatility of assets (assumed constant)
E0: Value of company's equity today  E: Instantaneous volatility of equity
ET: Value of company's equity at time T
If VT < D, the company is likely to default on the debt at time T. The value of the equity is zero.
If VT > D, the company is likely repay the debt at time T. The value of the equity at this time is VT – D.
Thus, the value of the firm's equity at time T is ET = max(VT – D, 0)
The above equation shows that the firm’s equity is a call option on firm’s assets with a strike price equal to the
repayment on the debt.
ln V0 / D  ( r   V2 / 2)T
The Black–Scholes formula values a firm’s equity as E0 = V0 N(d1) – De–rt N(d2) where d1 
V T
and d 2  d1   V T
The value of the debt today is V0 – E0.
The risk–neutral probability that the company will default on the debt is N(–d2). To calculate this, we need to
determine V0 and  V.
 neither of these are directly observable.
 however, E0 is observable if the company is publicly traded.
This means that E0 = V0N(d1) – De–rt N(d2) provides one condition that must be satisfied by V0 and  V.
E
We can estimate  E using It ô 's lemma:  E E0   V V0  N (d1 ) V V0
V
This provides another equation that must be satisfied by V0 and  V. Thus, E0 = V0N(d1) – De–rt N(d2) and  E E0 = N(d1)
 V V0 provide simultaneous equations that can be solved for V0 and  V.
Example 1. The value of a company's equity is $3 million. The volatility of its equity is 80%.
The debt to be paid in one year is $10 million. The risk–free rate is 5% per annum. Thus, E0 = 3,  E = 0.80, r = 0.05, T
= 1, D = 10.
a. Solving E0 = V0N(d1) – De–rt N(d2) and  E E0 = N(d1)  V V0 yields V0 = 12.40 and  V = 0.2123.
b. The parameter d2 is 1.1408, so that the probability of default is N(–d2) = 0.127 = 12.7%.
c. The market value of the debt is V0 – E0 = 12.40 – 3.0 = 9.40.
d. The present value of the promised payment on the debt is 10e–0.05×1 = 9.51.
e. The expected loss on the debt is therefore (9.51 – 9.40)/9.51 or about 1.2% of its no–default value.
Comparing this with the probability of default gives the expected recovery in the event of a default as (12.7 –
1.2)/12.7, or about 91%.
Q: How well do the default probabilities produced by Merton's model relate to actual default experience?
A: Merton's model produces a good ranking of default probabilities (risk–neutral or real–world). Thus, a monotonic
transformation can be used to convert the probability of default output from Merton's model into a good estimate of
either the real–world or risk–neutral default probability.

20.7 CREDIT RISK IN DERIVATIVES TRANSACTIONS

The credit exposure on a derivatives transaction is more complicated than that on a loan because the claim that will be
made in the event of a default is more uncertain.
Consider a financial institution that has one derivative contract outstanding with a counterparty. Consider three
possible situations:
1. The contract is always a liability to the financial institution (e.g. a short option position). There is no credit risk to
the financial institution. If the counterparty goes bankrupt, there will be no loss or gain to the financial institution.
2. The contract is always an asset to the financial institution (e.g. a long option position). There is always credit risk
to the financial institution. If the counterparty goes bankrupt, the financial institution has to make a claim against
the assets of the counterparty and may receive some percentage of the value of the derivative.
3. The contract can become either an asset or a liability to the financial institution (e.g. a forward contract). There
may or may not be credit risk to the financial institution. If the counterparty defaults when the value of the
derivative is:
a. positive to the financial institution, a claim will be made against the assets of the counterparty and a loss is
likely to be experienced.
b. negative to the financial institution, no loss is made because the derivative will be retained, closed out, or sold
to a third party.
Adjusting Derivatives' Valuations for Counterparty Default Risk
How should the end–user of derivatives adjust the value of a derivative to allow for counterparty credit risk?
Let f0 be the value of a derivative today, and assume that defaults can take place at times t1, t2,,,,, tn and that the value
of the derivative to the financial institution (assuming no defaults) at time ti is fi .
Let the risk–neutral probability of default at time ti be qi and the expected recovery rate as R.
 The financial institution's potential loss at time ti is max(fi, 0).
 The expected recovery in the event of a default is R* max(fi, 0).
 Assume the recovery rate and the probability of default is independent of the value of the derivative.
The risk–neutral expected loss from default at time ti is q (1  R ) Eˆ [max( f ,0)] , where Ê is the expected value in a
i i
risk–neutral world.
n
On a present value basis, the cost of defaults equal u v
i 1
i i , where ui = qi(1 – R) and vi is the value today of an
instrument that pays off the exposure on the derivative under consideration at time ti.
Consider the three categories of derivatives mentioned earlier.
n
For the first category, the value of fi is always negative, the total expected loss from defaults, u v
i 1
i i  0 , and the
financial institution needs to make no adjustments for the cost of defaults.

For the second category, fi is always positive, max(fi, 0) is always equal to fi .
Since vi is the present value of fi, it always equals f0.
The expected loss from default is f0 × the total probability of default during the life of the derivative × (1 – R).
Example 2: Consider a 2–year over–the–counter option with a value (assuming no defaults) of $3. Suppose:
 the company selling the option has a 2-year risk–neutral default probability of 4%
 the recovery in the event of a default is 25%.
The expected cost of defaults is $3 × 0.04 × (1 – 0.25) = $0.09.
The buyer of the option should pay only $3.00 – $0.09 = $2.91.
For the third category of derivatives, the sign of fi is uncertain.
 vi is a call option on fi with a strike price of zero.
 vi may be calculated by simulating the underlying market variables over the life of the derivative. Sometimes
approximate analytic calculations are possible (see, e.g., problems 20.17 and 20.18).
The above analyses assume that the probability of default is independent of the value of the derivative.
 This is reasonable when the derivative is a small part of the portfolio of the counterparty or when the counterparty
is using the derivative for hedging purposes.
 If a party wants to enter into a large derivatives transaction for speculative purposes , and when the transaction has
a large negative value for the counterparty (and a large positive value for the financial institution), the chance of
counterparty declaring bankruptcy may be much higher than when the situation is the other way round.

20.8 CREDIT RISK MIGRATION

In many cases, the analysis presented thus far overstates the credit risk in a derivatives transaction. This is because a
number of clauses that derivatives dealers include in their contracts to mitigate credit risk.
Netting: Netting complicates the estimation of the losses in the event of a counterparty default. Netting:
 is a clause in most contracts.
 states that if a counterparty defaults on one contract, it must default on all outstanding contracts with the financial
institution.
Netting can substantially reduce credit risk for a financial institution. Suppose:
 a financial institution has 3 contracts outstanding with a counterparty.
 the contracts are worth +$10 million, +$30 million, and –$25 million to the financial institution.
 the counterparty runs into financial difficulties and defaults on its outstanding obligations.
Without netting, the counterparty would default on the 1st two contracts and retain the 3rd for a loss to the financial
institution of $40 million.
With netting, it must default on all 3 contracts, resulting in a loss to the financial institution of $15 million. (Note: if
the third contract had been worth –$45 million to the financial institution, the counterparty would choose not to default
and there would be no loss to the financial institution.)
Consider the following:
 A financial institution has a portfolio of N derivative contracts with a particular counterparty.
 The no–default value of the i th contract is Vi and the recovery made in the event of default is R.
Note: Derivatives rank as unsecured claims in the event of default and the claim made in the event of default is the no–
default value at the time of default.
N
Without netting, the financial institution loses (1  R )  max(V ,0) .
i 1
i
Its loss is the payoff from a portfolio of call options on the contract values where each option has a strike price of zero.
 N 
With netting, the financial institution loses (1  R )max  Vi ,0 
 i 1 
Its loss is the payoff from a single option on the portfolio of contract values with a strike price of zero. Further, the
value of an option on a portfolio is never greater than, and is often considerably less than, the value of the
corresponding portfolio of options.
Reducing Exposure to Credit Risk
Two ways financial institutions reduce potential losses in the event of a default.
1. Collateralization. Collateralization mitigates credit risks.
Consider that a company and a financial institution have entered into a number of derivatives contracts.
 A collateralization agreement specifies that the contracts be marked to market using a pre–agreed formula.
 If the total value of the contracts to the financial institution is above a certain threshold level on a certain day, it
can ask the company to post collateral.
The amount of collateral posted, when added to collateral already posted by the company, is equal to the difference
between the value of the contract to the financial institution and the threshold level.
 When the contract moves in favor of the company so that the difference between value of the contract to the
financial institution and the threshold level is less than the total margin already posted, the company can reclaim
margin.
 In the event of a default by the company, the financial institution can seize the collateral. If the company does not
post collateral as required, the financial institution can close out the contracts.
Example: Assume the threshold level for the company is $10 million and contract is marked to market daily for the
purposes of collateralization.
 If one day the value of the contract to financial institution is $10.5 million, it can ask for $0.5 million of collateral.
 If the next day the value of the contract rises to $11.4 million it can ask for a further $0.9 million of collateral.
 If the value of the contract falls to $10.9 million on the following day, the company can ask for $0.5 million of the
collateral to be returned.
The threshold (i.e. the $10 million) can be regarded as a line of credit that the financial institution is prepared to grant
to the company.
 The threshold amount is not subject to protection.
 Even when the threshold is zero, the protection is not total.

When a company gets into financial difficulties, it is likely to stop responding to requests to post collateral. By the
time the counterparty exercises its right to close out contracts, their value may have moved further in its favor.
2. Use of downgrade triggers.
 Are clauses in contracts with counterparties that state that if the credit rating of the counterparty falls below a
certain level, then the contract is closed out with one side paying an amount to the other side.
 Lead to a significant reduction in credit risk, but do not completely eliminate all credit risk.
If there is a large downward fall in the credit rating of the counterparty (e.g. from A to default) in a short period of
time, the financial institution will still suffer a credit loss.
20.9 DEFAULT CORRELATION

“Default correlation” is the tendency for two companies to default at about the same time.
Reasons why default correlations exist.
1. Two companies in the same industry or geographic region tend to be affected similarly by external events and may
experience financial difficulties at the same time.
2. Economic conditions cause average default rates to be higher in some years than in others.
Default correlation means that credit risk cannot be completely diversified away and is the major reason why risk–
neutral default probabilities are greater than real–world default probabilities.
Modeling Default Correlation
Two types of models of default correlation:
2. Reduced–form models. Reduced models:
 assume that the hazard rates for different companies follow stochastic processes and are correlated with
macroeconomic variables. If the hazard rate for company A is high, there is a tendency for the hazard rate for
company B to be high.
 are mathematically attractive and reflect the fact that economic cycles generate default correlations.
 can be made consistent with either historical default probabilities or the risk–neutral probabilities of default
backed out from corporate bond prices.
Disadvantage: the range of default correlations that can be achieved is limited.
Even when there is a perfect correlation between two hazard rates, the corresponding correlation between defaults
is usually very low.
However, the problem can be solved by extending the model so that the hazard rate exhibits large jumps.
2. Structural models.
 These models are based on the idea similar that a company will default if the value of its assets is below a
certain level.
 Default correlation between companies A and B is introduced into the model by assuming that the stochastic
process followed by the assets of company A is correlated with the stochastic process followed by the assets of
company B.
Advantage over reduced–form models: Correlation can be made as high as desired.
Disadvantage relative to reduced–form models: they are liable to be computationally quite slow.
The Gaussian Copula Model for Time to Default
The Gaussian copula model for the time to default is a popular and practical reduced form default correlation model.
This model:
 quantifies the correlation between the times to default for two different companies.
 implicitly assumes that all companies will default eventually
 can be used in conjunction with either real–world or risk–neutral default probabilities.
The left tail of the real–world probability distribution for the time to default of a company can be estimated from data
produced by rating agencies such as that in Table 1.
The left tail of the risk–neutral probability distribution of the time to default can be estimated from bond prices using
the approach in Section 20.4.
Define t1 as the time to default of company 1 and t2 as the time to default of company 2.
The probability distribution of a company's time to default is not approximately normal, and thus a Gaussian copula
model can be used. Transform t1 and t2 into new variables x1 and x2 using x1  N 1 [Q1 (t1 )], x2  N 1 [Q2 (t2 )]
where Q1 and Q2 are the cumulative probability distributions for t1 and t2 respectively, and N –1 is the inverse of the
cumulative normal distribution (u = N –1(v) when v = N (u)).

 These are “percentile–to–percentile” transformations. For example, the 5–percentile point in the probability
distribution for t1 is transformed to x1 = –1.645, which is the 5–percentile point in the standard normal distribution.
x1 and x2 have normal distributions with mean 0, and standard deviation 1.0.
Assume that the joint distribution of x1 and x2 is bivariate normal with correlation  12.
 This assumption is referred to as using a Gaussian copula.
 The assumption is convenient because it means that the joint probability distribution of t1 and t2 is fully defined by
the cumulative default probability distributions Q1 and Q2 for t1 and t2, together with a single correlation parameter
 12.
The attraction of the Gaussian copula model is that it can be extended to many companies.
Suppose that we are considering n companies and that ti is the time to default of the ith company.
Transform each ti into a new variable, xi that has a standard normal distribution.
The transformation is the percentile–to–percentile transformation xi = N –1[Qi (ti)], where
where Qi is the cumulative probability distribution for ti. We then assume that the xi are multivariate normal. The
default correlation between ti and tj is measured as the correlation between xi and xj. This is referred to as the copula
correlation.
The Gaussian copula approach:
 is a useful way representing the correlation structure between variables that are not normally distributed.
 allows the correlation structure of the variables to be estimated separately from their marginal (unconditional)
distributions.
 assumes that after a transformation is applied to each variable they are multivariate normal.
Example 3 – Use of a Gaussian copula approach
Suppose we wish to simulate defaults during the next five years in n companies.
 For each company the cumulative probability of a default during the next 1, 2, 3, 4, and 5 years is 1 %, 3%, 6%,
10%, and 15%, respectively.
 Sample from a multivariate normal distribution to obtain xi (1  i  n).
 Convert the xi to ti, a time to default.
Determine when defaults take place within a year. When the sample from the normal distribution is:
 less than N – 1(0.01) = –2.33, a default takes place within the first year,
 between –2.33 and N – 1(0.03) = –1.88, a default takes place during the second year;
 between –1.88 and N – 1(0.06) = –1.55, a default takes place during the third year;
 between –1.55 and N – 1(0.03) = –1.28, a default takes place during the fourth year;
 between –1.28 and N – 1(0.03) = –1.04, a default takes place during the fifth year.
 greater than –1.04, there is no default.
Using Factors to Define the Correlation Structure
To avoid defining a different correlation between xi and xj for each pair of companies i and j in the Gaussian copula
model, a one–factor model is often used.
The assumption is that xi  ai M  1  ai2 Zi
M is a common factor affecting defaults for all companies and Zi is a factor affecting only company i.
The variables M and the Zi have independent standard normal distributions.
The ai are constant parameters between –1 and +1.
The correlation between xi and xj is ai aj.
Suppose that the probability that company i will default by a particular time T is Qi(T).
Under the Gaussian copula model, a default happens when N ( xi ) Qi (T ) or xi  N 1 [Qi (T )]
N 1 [Qi (T )]  ai M
From equation (20.6), this condition is xi  ai M  1  ai2 Z i  N 1[Qi (T )] , or Z i 
1  ai2
 N 1 [Q (T )]  a M 
Conditional on the value of the factor M, the probability of default is therefore Qi (T M )  N  i i
 Eq
 1  a 2 
 i 
(20.7)
A particular case of the one–factor Gaussian model is where the probability distributions of default are the same for all
i and the correlations between xi and xj is the same for all i and j.

Suppose that Qi(T) = Q(T) for all i and that the common correlation is , so that ai   for all i.
 N 1 [Q (T )]   M 
Eq (20.7) becomes Qi (T M )  N  i
 Eq (20.8)
 1 2 
 
Binomial Correlation Measures
Consider two companies A and B.
A default correlation measure is the coefficient of correlation between:
1. A variable that equals 1 if company A defaults between times 0 and T and 0 otherwise; and
2. A variable that equals 1 if company B defaults between times 0 and T and 0 otherwise
PAB (T )  QA (T )QB (T )
The measure is  AB (T ) = , where
[QA (T )  QA (T )2 ][QB (T )  QB (T )2 ]
PAB(T) is the joint probability of A and B defaulting between time zero and time T.
QA(T) is the cumulative probability that company A will default by time T, and
QB(T) is the cumulative probability that company B will default by time T.
AB(T) typically increases as T increases.
Relationship between Measures
Define: M(a, b; ) as the probability in a standardized bivariate normal distribution that the first variable is less than a
and the second variable is less than b if the coefficient of correlation between the variables is .
Let AB be the default correlation between A and B in the Gaussian copula model.
M ( xA (T ), xB (T );  AB )  QA (T )QB (T )
It follows that PAB (T )  M ( x A (T ), xB (T );  AB ) so that  AB (T ) 
[QA (T )  QA (T )2 ][QB (T )  QB (T )2 ]
This shows that, if QA(T) and QB(T) are known, AB(T) can be calculated from AB and vice versa.
AB is usually markedly greater than AB(T).
Example 4 Suppose the probability of company A defaulting in a one–year period is 1% and company B defaulting in
a one–year period is 1%. Further, let AB = 0.20.
Compute AB (1)
M ( x A (1), xB (1);  AB )  QA (1)QB (1)
Step 1: Write an equation to determine AB (1).  AB (1)  , where
[QA (1)  QA (1)2 ][ QB (1)  QB (1)2 ]
Qi (ti ) as the cumulative probability distribution of ti and ui (ti )  N -1 [Qi (ti )] for 1  i  N .
Step 2: Using the equations in Step 1, compute AB (1)
Compute xA (1) and xB (1); xA (1) = xB (1) = N – 1(0.01) = –2.326
Since AB is 0.20, then M ( x A (1), x B (1),  AB ) = 0.000337 ; Thus, AB (1) = 0.024

Credit VaR is defined similarly to Market VaR (described in Chapter 18).
Example: A credit VaR with a confidence level of 99.9% and a 1–year time horizon is the credit loss that we are 99.9%
confident will not be exceeded over 1 year.
Consider a bank with a very large portfolio of similar loans.
 Assume the probability of default is the same for each loan and the correlation between each pair of loans is the
same.
 When the Gaussian copula model for time to default is used, the right–hand side of equation (20.8) is
approximately equal to the percentage of defaults by time T as a function of M.
M has a standard normal distribution.
We are X% certain that its value will be greater than N –1(1 – X) = – N –1(X).
Thus, we are X % certain that the percentage of losses over T years will be less than V(X, T), where
 N 1 [Q (T )   N 1 ( X ) 
V ( X ,T )  N   , and where
 1 
 
Q(T) is the probability of default by time T and
 is the copula correlation between any pair of loans.

A rough estimate of the credit VaR when an X % confidence level is used and the time horizon is T is therefore L(1–
R)V(X, T), where L is the size of the loan portfolio and R is the recovery rate.
The contribution of a loan of size Li to the credit VaR is Li(1–R)V(X, T).
Example 4: Consider the following:
 A bank has a total of $100 million of retail exposures.
 The 1–year probability of default averages 2% and the recovery rate averages 60%.
 The copula correlation parameter is estimated as 0.1. Therefore,
 N 1 [(0.02)  0.10 N 1 (0.999) 
V (0.999,1)  N    0.0128 . This shows that the 99.9% worst case default rate is
 1  0.10 
12.8%. The 1–year 99.9% credit VaR is therefore $100 million × 0.128 × (1 – 0.6) or $5.13 million.
CreditMetrics
One method for calculating credit VaR is CreditMetrics
 a probability distribution of credit losses is estimated by simulating credit rating changes for each counterparty.
 the probability distribution of losses over a one–year period can be determined as follows.
a. sample to determine credit rating changes of all counterparties throughout the year for each simulation trial.
b. re–value outstanding contracts to determine the total of credit losses from defaults and credit rating changes
 a probability distribution for credit losses is obtained after a large number of simulation trials are conducted.
Comments:
 This approach is computationally time intensive.
 It has the advantage that credit losses are defined as those arising from credit downgrades and defaults.
 The impact of credit mitigation clauses such as those described Section 20.–8 can be approximately incorporated
into the analysis.
Table 6 is typical of the historical data provided by rating agencies on credit rating changes and could be used as a
basis for a CreditMetrics Monte Carlo simulation.
 It shows the percentage probability of a bond moving from one rating category to another in 1–year.
 For example, a bond that starts with an A credit rating has a 91.84% chance of still having an A rating at the end of
1 year. It has a 0.02% chance of defaulting during the year, a 0.13% chance of dropping to B, and so on.
Table 6 One–year ratings transition matrix (probabilities expressed as percentages).

Rating at year–end
Initial Rating Aaa Aa A Baa Ba B Caa Default
Aaa 92.18 7.06 0.73 0.00 0.02 0.00 0.00 0.00
Aa 1.17 90.85 7.63 0.26 0.07 0.01 0.00 0.02
A 0.05 2.39 91.84 5.07 0.50 0.13 0.01 0.02
Baa 0.05 0.24 5.20 88.48 4.88 0.80 0.16 0.18
Ba 0.01 0.05 0.50 5.45 85.13 7.05 0.55 1.27
B 0.01 0.03 0.13 0.43 6.52 83.21 3.04 6.64
Caa 0.00 0.00 0.00 0.58 1.74 4.18 67.99 25.50
Default 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100.00
In sampling to determine credit losses, the credit rating changes for different counterparties should not be assumed to
be independent.
A Gaussian copula model can be used to construct a joint probability distribution of rating changes similarly to the
way it is used to describe the joint probability distribution of times to default.
The copula correlation between the rating transitions for two companies is usually set equal to the correlation between
their equity returns using a factor model similar to that in Section 20.9.
Example: Suppose the following:
 we wish to simulate the rating change of a Aaa and a Baa company over a one–year period using the transition
matrix in Table 6.
 the correlation between the equities of the two companies is 0.2.
 sample two variables xA and xB from normal distributions so that their correlation is 0.2 (on each simulation trial).
 the variable xA determines the new rating of the Aaa company and the variable xB determines the rating of the Baa
company.

From the normal distribution table of values, we determine the following:

N –1(0.9218) = 1.4173 N –1(0.9218 + 0.0706) = 2.4276 N –1(0.9218 + 0.0706 + 0.0073) = 3.4319
The Aaa company:
 stays Aaa–rated if xA < 1.4173
 becomes Aa–rated if 1.4173  xA  2.4276, and
 becomes A–rated if 2.4276  xA  3.4319, and so on.
Similarly,
N –1(0.000) = –3.2905 N –1(0.0005 + 0.0024) = –2.7589 N –1(0.0005 + 0.0024 + 0.0520) = –1.5991
The Baa company:
 becomes Aaa–rated if xA  –3.2905,
 becomes Aa–rated if –3.2905  xB  –2.7589, and
 becomes A–rated if –2.7589  xB  –1.5991, and so on.
The appealing feature of the Gaussian copula approach is that it allows us to use the properties of multivariate normal
distribution to jointly sample the credit migration of many companies.
Losses from credit rating changes are determined by valuing the contract at the end of the year with:
(i) the credit rating at the beginning of the year and
(ii) the credit rating at the end of the year.
The credit loss is the difference between the two.


Delgado Francisco - Derivados Financieros (2010)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Delgado Francisco - Derivados Financieros (2010)

Uploaded by

Copyright:

Available Formats

Derivados Avanzados Diciembre 2010 Página: 1/358

FUTURES AND OPTIONS MARKETS

1 Introduction to Futures and Options

3 Structure of Futures Prices

10 Structure of Option Prices: Arbitrage Relations

FUTURES AND OPTIONS MARKETS

1 Introduction to Futures and Options

2 Functioning of Futures and Option Markets

3 Structure of Futures Prices

PART 2 – FUTURES. Continuation

4 Hedging with Futures Contracts

5 Risk and Return in Futures Markets

6 Futures Contracts on Physical Commodities

PART 2 – FUTURES. Continuation

7 Stock Index Futures Contracts

8 Interest Rate Futures Contracts

PART 2 – FUTURES. Continuation

9 Currency Futures Contracts

10 Structure of Option Prices: Arbitrage Relations

PART 3 – OPTIONS. Continuation

12 Option Trading Strategies

PART 3 – OPTIONS. Continuation

13 Common Stock Option Contracts

14 Stock Index and Stock Index Futures Option Contracts

15 Interest Rate and Interest Rate Futures Option Contracts

PART 3 – OPTIONS. Continuation

16 Other Option Contracts

1 INTRODUCTION TO FUTURES AND OPTIONS

1.1 WHAT ARE FUTURES CONTRACTS?

1.2 WHAT ARE OPTIONS?

1.3 ECONOMIC PURPOSE OF FUTURES AND OPTIONS

1.1 WHAT ARE FUTURES CONTRACTS?

FIGURE 1.2 Profit Diagrams for Futures Position Held to Expiration

Profit (b) Short Futures Position

1.2 WHAT ARE OPTIONS?

Profit (b) Short Call

Profit (b) Short Put

1.3 ECONOMIC PURPOSE OF FUTURES AND OPTIONS

2 FUNCTIONING OF FUTURES AND OPTION MARKETS

2.1 EXCHANGE MARKETS VERSUS OTC MARKETS

2.2 CONTRACT DESIGN

2.3 TRADING PROCEDURES

2.4 THE CLEARING HOUSE

2 FUNCTIONING OF FUTURES AND OPTION MARKETS

2.1 EXCHANGE MARKETS VERSUS OTC MARKETS

2.2 CONTRACT DESIGN

Method of Contract Settlement at Maturity

Grade of Underlying Commodity

Time of Contract Settlement at Maturity

2.3 TRADING PROCEDURES

2.4 THE CLEARING HOUSE

Margin or Performance Bond

Trading and Settlement Example

TABLE 2.1 Example of trading in futures and associated cash flows.

TABLE 2.2 Example of trading in options and associated cash flows.

4. COBERTURA CON CONTRATOS DE FUTUROS

4.1 UNA COBERTURA TRADICIONAL

4.2 RIESGO DE LA BASE

4.3 COBERTURA DE MERCADERÍAS NO-DISPONIBLES (NON-DELIVARABLE)

4.4 COBERTURA ÓPTIMA – ENFOQUE MEDIA/VARIACIÓN

4.5 COBERTURA ÓPTIMA – ENFOQUE DE REGRESIÓN DE MÍNIMOS CUADRADOS

4.6 CALCULANDO EL RATIO DE COBERTURA

4. COBERTURA CON CONTRATOS DE FUTUROS