An Introduction to Investment Theory

William N. Goetzmann YALE School of Management

Chapter I: Capital Markets and Investment Performance

Overview Suppose you find a great investment opportunity, but you lack the cash to take advantage of it. This is the classic problem of financing. The short answer is that you borrow -- either privately from a bank, or publicly by issuing securities. Securities are nothing more than promises of future payment. They are initially issued through financial intermediaries such as investment banks, which underwrite the offering and work to sell the securities to the public. Once they are sold, securities can often be re-sold. There is a secondary market for many corporate securities. If they meet certain regulatory requirements, they may be traded through brokers on the stock exchanges, such as the NYSE, the AMEX and NASDAQ, or on options exchanges and bond trading desks. Securities come in a bewildering variety of forms - there are more types of securities than there are breeds of cats and dogs, for instance. They range from relatively straightforward to incredibly complex. A straight bond promises to repay a loan over a fixed amount of interest over time and the principal at maturity. A share of stock, on the other hand, represents a fraction of ownership in a corporation, and a claim to future dividends. Today, much of the innovation in finance is in the development of sophisticated securities: structured notes, reverse floaters, IO's and PO's -- these are today's specialized breeds. Sources of information about securities are numerous on the world-wide web. For a start, begin with the Ohio State Financial Data Finder. All securities, from the simplest to the most complex, share some basic similarities that allow us to evaluate their usefulness from the investor's perspective. All of them are economic claims against future benefits. No one borrows money that they intend to repay immediately; the dimension of time is always present in financial instruments. Thus, a bond represents claims to a future stream of prespecified coupon payments, while a stock represents claims to uncertain future dividends and division of the corporate assets. In addition, all financial securities can be characterized by two important features: risk and return. These two key measures will be the focus of this second module.

I. Finance from the Investor's Perspective Most financial decisions you have addressed up to this point in the term have been from the perspective of the firm. Should the company undertake the construction of a new processing plant? Is it more profitable to replace an old boiler now, or wait? In this module, we will examine financial decisions from the perspective of the purchaser of corporate

securities: shareholders and bondholders who are free to buy or sell financial assets. Investors, whether they are individuals or institutions such as pension funds, mutual funds, or college endowments, hold portfolios, that is, they hold a collection of different securities. Much of the innovation in investment research over the past 40 years has been the development of a theory of portfolio management, and this module is principally an introduction to these new methods. It will answer the basic question, What rate of return will investors demand to hold a risky security in their portfolio? To answer this question, we first must consider what investors want, how we define return, and what we mean by risk.

II. Why Investors Invest What motivates a person or an organization to buy securities, rather than spending their money immediately? The most common answer is savings -- the desire to pass money from the present into the future. People and organizations anticipate future cash needs, and expect that their earnings in the future will not meet those needs. Another motivation is the desire to increase wealth, i.e. make money grow. Sometimes, the desire to become wealthy in the future can make you willing to take big risks. The purchase of a lottery ticket, for instance only increases the probability of becoming very wealthy, but sometimes a small chance at a big payoff, even if it costs a dollar or two, is better than none at all. There are other motives for investment, of course. Charity, for instance. You may be willing to invest to make something happen that might not, otherwise -- you could invest to build a museum, to finance low-income housing, or to re-claim urban neighborhoods. The dividends from these kinds of investments may not be economic, and thus they are difficult to compare and evaluate. For most investors, charitable goals aside, the key measure of benefit derived from a security is the rate of return.

III. Definition of Rates of Return The investor return is a measure of the growth in wealth resulting from that investment. This growth measure is expressed in percentage terms to make it comparable across large and small investors. We often express the percent return over a specific time interval, say, one year. For instance, the purchase of a share of stock at time t, represented as Pt will yield P t+1 in one year's time, assuming no dividends are paid. This return is calculated as: R t = [ P t+1 - Pt]/ Pt. Notice that this is algebraically the same as: Rt= [P t+1/ Pt]-1. When dividends are paid, we adjust the calculation to include the intermediate dividend payment: Rt=[ P t+1 Pt+Dt]/ Pt. While this takes care of all the explicit payments, there are other benefits that may derive from holding a stock, including the right to vote on corporate governance, tax treatment, rights offerings, and many other things. These are typically reflected in the price fluctuation of the shares.

IV. Arithmetic vs. Geometric Rates of Return There are two commonly quoted measures of average return: the geometric and the

arithmetic mean. These rarely agree with each other. Consider a two period example: P0 = $100, R1 = -50% and R2 = +100%. In this case, the arithmetic average is calculated as (10050)/2 = 25%, while the geometric average is calculated as: [(1+R1)(1+R2)]1/2-1=0%. Well, did you make money over the two periods, or not? No, you didn't, so the geometric average is closer to investment experience. On the other hand, suppose R1 and R2 were statistically representative of future returns. Then next year, you have a 50% shot at getting $200 or a 50% shot at $50. Your expected one year return is (1/2)[(200/100)-1] + (1/2)[(50/100)-1] = 25%. Since most investors have a multiple year horizon, the geometric return is useful for evaluating how much their investment will grow over the long-term. However, in many statistical models, the arithmetic rate of return is employed. For mathematical tractability, we assume a single period investor horizon.

V. Capital Market History The 1980's was one of the greatest decades for stock investors in the history of the U.S. capital markets.

(Courtesy Ibbotson Associates) We measure stock market performance by the total return to investment in the S&P 500, which is a standard index of 500 stocks, weighted by the market value of the equity of the company. Dividends paid by S&P 500 companies are assumed to be re-invested in shares of stock. This provides a measure of total investor return, before individual taxes are paid. The 1930's was one of the worst decades for U.S. stock investors.

(Courtesy Ibbotson Associates) In the 1930's stock markets crashed all over the globe. U.S. stock investors experienced a zero percent return for the eleven-year period from 12/1929 to 12/1939. U.S. Capital Markets over the Long Term: 1926 - 1995 Over the past 68 years, A stock investment in the S&P increased from $1 to $800

(Courtesy Ibbotson Associates)

VI. Risk Premium Notice in the preceding figure that a dollar invested in stock grew to $889 over the period, while a dollar invested in corporate bonds grew to $40. Why the big difference? This return

differential is commonly attributed to a difference in the risk associated with stocks as opposed to bonds. Notice that the stock line is "shakier" than the bond line. Wealth invested in stocks since 1926 was more volatile than wealth invested in bonds. Despite the higher return, the risks were higher as well. An investor typically cares about the riskiness of an investment. If, for instance, you are saving for a home purchase sometime in the next year, then you really care whether your $100,000 nest egg has a significant probability of dropping to $50,000 in twelve months. As a matter of fact, you might be willing to trade a lower rate of investment return for "insurance" that your principal will be secure. This is called risk-aversion -- and all things being equal, most investors prefer less risk to more. Summary Statistics of U.S. Investments from 1926 through March, 1995. Source:Ibbotson Associates Investment S&P total return U.S. Small Stock TR U.S. LT Govt TR U.S. LT Corp. TR U.S. 30 day T-Bills geom. mean 10.30 12.28 4.91 5.49 3.70 arith.mean 12.45 17.28 5.21 5.73 3.70 std 22.28 35.94 8.00 7.16 .96 high ret. 42.56 73.46 15.23 13.76 1.35 low ret. -29.73 -36.74 -8.41 -8.90 -0.06

The difference between the S&P total return and the U.S. 30 day T-Bill return is called the equity premium. It is the amount of return that investors demand for holding a risky security such as stocks, as opposed to a riskless security, such as T-Bills. The annual equity premium is about 9% arithmetic, and 6% geometric, over the 1926 - 1995 period.

VII. Standard Deviation as a Measure of Risk Stock returns may be riskier or more volatile, but this concept is a difficult one to express simply. To do so, we borrow a concept from statistics, called standard deviation. standard deviation is a summary measure about the average spread of observations. It is the square root of the variance, which is calculated as:

The standard deviation of one-year S&P 500 returns is about 22.28%. If S&P returns are normally distributed, this means that about 2/3 of the time we should observe an annual return within the range (12.45-22.28)= -9.93 and (12.45+22.28)= 34.73. A histogram of S&P 500 annual returns shows that returns are approximately normally distributed, or are they? A normal distribution should allow returns lower than -100%. Stocks do not. In fact, the log of the variable being normally distributed is a better approximation. However, there

is evidence to suggest that even this is not quite right. The tails of stock returns are a bit "fatter" than should be observed if returns were log-normally distributed. This lends some support to the hypothesis advanced by Benoit Mandelbrot that stock returns follow a "stable" distribution, with undefined variance. Have a look at the S&P 500 histogram yourself:

How well does standard deviation capture the notion of investor risk? It equally weights high returns with low returns. It heavily weights extreme observations. It is not concerned with the shape of the distribution. All of these are valid criticisms. However the benefits to using standard deviation are large. It is a single measure, allowing us to quantify asset returns by risk. As we will see in the next chapter, it also provides the basis for investor decisions about portfolio choice.

An Internet Scavenger Hunt See if you can:

find historical return information on Microsoft, Inc. find historical return information on the Fidelity Magellan Fund identify the top five investment banks in the U.S. find which company on the S&P 500 had the largest capital appreciation return last year find the current yields on U.S. treasury bonds

If you accept the challenge, email me with your list of http sites!

Chapter II: The Geography of the Efficient Frontier

In the previous chapter, we saw how the risk and return of investments may be characterized by measures of central tendency and measures of variation, i.e. mean and standard deviation. In fact, statistics are the foundations of modern finance, and virtually all the financial innovations of the past thirty years, broadly termed "Modern Portfolio Theory," have been based upon statistical models. Because of this, it is useful to review what a statistic is, and how it relates to the investment problem. In general, a statistic is a function that reduces a large amount of information to a small amount. For instance, the average is a single number that summarizes the typical "location" of a set of numbers. Statistics boil down a lot of information to a few useful numbers -- as such, they ignore a great deal. Before modern portfolio theory, the decision about whether to include a security in a portfolio was based principally upon fundamental analysis of the firm, its financial statements and its dividend policy. Finance professor Harry Markowitz began a revolution by suggesting that the value of a security to an investor might best be evaluated by its mean, its standard deviation, and its correlation to other securities in the portfolio. This audacious suggestion amounted to ignoring a lot of information about the firm -- its earnings, its dividend policy, its capital structure, its market, its competitors -- and calculating a few simple statistics. In this chapter, we will follow Markowitz' lead and see where the technology of modern portfolio theory takes us.

I. The Risk and Return of Securities Markowitz's great insight was that the relevant information about securities can be summarized by three measures: the mean return (taken as the arithmetic mean), the standard deviation of the returns and the correlation with other assets' returns. The mean and the standard deviation can be used to plot the relative risk and return of any selection of securities. Consider six asset classes:

(Courtesy Ibbotson Associates)

This figure was constructed using historical risk and return data on Small Stocks, S&P stocks, Corporate and Government Bonds, and an international stock index called MSCI, or Morgan Stanley Capital International World Portfolio. The figure shows the difficulty an investor faces about which asset to choose. The axes plot annual standard deviation of total returns, and average annual returns over the period 1970 through 3/1995. Notice that small stocks provide the highest return, but with the highest risk. In which asset class would you choose to invest your money? Is there any single asset class that dominates the rest? Notice that an investor who prefers a low risk strategy would choose T-Bills, while an investor who does not care about risk would choose small stocks. There is no one security that is best for ALL investors.

II. Portfolios of Assets Typically, the answer to the investment problem is not the selection of one asset above all others, but the construction of a portfolio of assets, i.e. diversification across a number of different securities. The key to diversification is the correlation across securities. Recall from data analysis and statistics that the correlation coefficient is a value between -1 and 1, and measures the degree of co-movement between two random variables, in this case stock returns. It is calculated as:

Where the sigma AB is the covariance of the two securities. Here is how to use correlation in the context of portfolio construction. Consider two securities, A and B. Security A has a mean of 10% and an STD of 15%. Security B has a mean of 20% and an STD of 30%. We can calculate the standard deviation of a portfolio composed of different mixtures of A & B using this equation:

The mean return is not as complicated. It is a simple weighted average of the means of the two assets: mean p = W A RA + W B R B. Notice that a portfolio will typically have a weight of one, so usually, W A + W B = 1. What if the correlation of A&B = 0 ? Notice that a portfolio of 80% A and 20% B has a standard deviation of: sqrt(.82*.152+.22.32+2*0*.8*.2*.15*.3) = 13.4 % In other words, a mixture of 20% of the MORE RISKY SECURITY actually decreases the volatility of the portfolio! This is a remarkable result. It means you can reduce risk and increase return by diversifying across assets. What if the correlation of A&B = 1 ? In this case, the perfect correlation between the two assets means there is no diversification. The portfolio std of of the 80/20 mix is 18%. this is equal to a linear combination of the standard deviations: (.8)(.15)+(.2)(.30) = 18% What if the correlation of A&B = -1 ? This is an unusual case, because it means that when A moves up, B always moves down. Take a mixture of .665 A and (1-.665) B. sqrt(.6652*.152+(1-.665)2.32+2*0*(.665)*(1-.665)*.15*.3) = .075%, Which is very close to zero. In other words, A is nearly a perfect hedge for B. One of the few real-life negative correlations you will find is a short position in a stock offsetting the long position. In this case, since the mean returns are also the same, the expected return will be zero. These extremes of correlation values allow us to describe an envelope within which all combinations of two assets will lie, regardless of their correlations.

(Courtesy Campbell Harvey)

III. More Securities and More Diversification Now consider what will happen as you put more assets into the portfolio. Take the special case in which the correlation between all assets is zero, and all of them have the same risk. You will find that you can reduce the standard deviation of the portfolio by mixing across several assets rather than just two. Each point represents an equally-weighted combination of assets; from a single stock to two, to three, to thirty, and more. Notice that, after 30 stocks, diversification is mostly achieved. There are enormous gains to diversification beyond one or two stocks.

(Courtesy Campbell Harvey)

If you allow yourself to vary the portfolio weights, rather than keeping them equal, the benefits are even greater, however the mathematics is more challenging. You not only have to calculate the STD of the mixture between A&B, but the STD of every conceivable mixture of the securities. None-the-less, If you did so, you would find that there is a set of portfolios which provide the lowest level of risk for each level of return, and the highest level of return for each level of risk. By considering all combinations of assets, a special set of portfolios stand out -- this set is called the efficient frontier.

The efficient frontier, shown in blue, is the set of dominant portfolios, at least from the perspective of a risk averse investor. For ANY level of risk, the efficient frontier identifies a point that is the highest returning portfolio in its risk class. By the same token, for any level of return, the frontier identifies the lowest risk portfolio in that return class. Notice that it extends from the maximum return portfolio (actually a single asset) to the minimum variance portfolio. The efficient frontier has a portfolio for everyone -- there are an infinite

number of points in the set, corresponding to the infinite variation in investor preferences for risk. The area called the feasible set represents all feasible combinations of assets. There are no assets that fall outside of the feasible set.

IV. Markowitz and the First Efficient Frontier The first efficient frontier was created by Harry Markowitz, using a handful of stocks from the New York Stock Exchange. Here it is, reproduced from his book Portfolio Selection Cowles Monograph 16, Yale University Press, 1959. It has a line going to the origin, because Markowitz was interested in the effects of combining risky assets with a riskless asset: cash.

Notice, too, that it is tipped on its side. The convention of STD on the X axis is developed later.

V. An Actual Efficient Frontier Today This figure is an efficient frontier created from historical inputs for U.S. and international assets over the period 1970 through 3/1995, using the Ibbotson EnCorr Optimizer program.

(Courtesy Ibbotson Associates)

This is state-of-the-art portfolio selection technology, however it is still based upon Markowitz's original optimization program. There are some basic features to remember:

A minimum variance portfolio exists A maximum return portfolio is composed of a single asset. B,C,D & E are critical points at which one the set of assets used in the frontier changes, i.e. an asset drops out or comes in at these points. There are no assets to the northwest of the frontier. That is why we call it a frontier. It is the edge of the feasible combinations of risk and returns.

VI. The Efficient Frontier with the Riskless Asset T-Bills are often taken to be riskless assets, and their return is indicated as Rf, the risk-free rate. Once you allow the riskless asset to be combined into a portfolio, the efficient frontier can change. Since it is riskless, it has no correlation to other securities. Thus it provides no diverisfication, per se. It does provide an opportunity to have a low-risk portfolio, however. This picture is a diagram of the efficient frontier composed of ALL the risky assets in the economy, as well as the riskless asset.

(Courtesy Campbell Harvey) In this special case, the new efficient frontier is a ray, extending from Rf to the point of tangency (M) with the "risky-asset" efficient frontier, and then beyond. This line is called the Capital Market Line (CML). It is actually a set of investable portfolios, if you were able to borrow and lend at the riskless rate! All portfolios between Rf and M are portfolios composed of treasury bills and M, while all portfolios to the right of M are generated by BORROWING at the riskless rate Rf and investing the proceeds into M.

VII. Summary The Markowitz model was a brilliant innovation in the science of portfolio selection. With almost a disarming slight-of-hand, Markowitz showed us that all the information needed to choose the best portfolio for any given level of risk is contained in three simple statistics: mean, standard deviation and correlation. It suddenly appeared that you didn't even need any fundamental information about the firm! The model requires no information about dividend policy, earnings, market share, strategy, quality of management -- nothing about the myriad of things with which Wall Street analysts concern themselves! In short, Harry Markowitz fundamentally altered how investment decisions were made. Virtually every major portfolio manager today consults an optimization program. They may not follow its recommendations exactly, but they use it to evaluate basic risk and return trade-offs. Why doesn't everyone use the Markowitz model to solve their investment problems? The answer again lies in the statistics. The historical mean return may be a poor estimate of the future mean return. As you increase the number of securities, you increase the number of correlations you must estimate -- and you must estimate them CORRECTLY to obtain the right answer. In fact, with more than 1,500 stocks on the NYSE, one is certain to find correlations that are widely inaccurate. Unfortunately, the model does not deal well with incorrect inputs. That is why it is best applied to allocation decisions across asset classes, for which the number of correlations is low, and the summary statistics are well estimated.

Chapter III: Preferences and Investor Choice

The last chapter presented the Markowitz model of portfolio selection, but with one key element missing -- individual portfolio choice. The efficient frontier dominates all combinations of assets, however it still has infinitely many assets. How do you pick one portfolio out of all the rest as the perfect one for you? This turns out to be a big challenge, because it requires investors to express their preferences in risk-return space. Investors choose portfolios for a myriad of reasons, very few of which can be reduced to a twodimensional space. In fact, investors are used to having the ability the CHANGE their investment decision if it is not developing as planned. The simple Markowitz model does not allow this freedom. It is a single period model, now used widely in practice for decision-making in a multi-period world. In this chapter, we will address some of the ways that one may approximate investor preferences in mean-variance space, however these methods are only approximations.

I. Choosing A Single Portfolio How might you choose a single portfolio among all of those on the efficient frontier? One approach is to model investor preferences mathematically, using iso-utility curves. These curves express the risk-return trade-off for investors in two-dimensional space. They work exactly like lines on a topological map. They are nested lines that show the highest and lowest altitudes in the region -- except they measure altitude in units of utility (whatever that is!) instead of feet or meters. Typically, a convenient mathematical function is chosen as the basis for iso-utility curves. For instance, one could use a logarithmic function, or even part of a quadratic function to capture the essence of investor preferences. The essential feature of the function is that it must allow people to demand ever-increasing levels of return for assuming more risk.

Although the mathematics of utility functions is beyond the scope of this course, if you are interested in further investigation, I recommend visiting Campbell Harvey's Pages on Optimal Portfolios. One way to characterize differences in investor risk aversion is by the curvature of the iso-

utility lines. Below are representative curves for four different types of investors: A more risk-averse, a moderately risk-averse, a less risk-averse, and a risk-loving investor. The whole set of nested curves is omitted to keep the picture simple.

Notice that the risk-lover demands lower expected return as risk increases in order to maintain the same utility level. On the other hand, for the more risk-averse investor, as volatility increase, he or she will demand sharply higher expected returns to hold the portfolio. These different curves will result in different portfolio choices for investors. The optimization procedure simply takes the efficient frontier and finds its point of tangency with the highest iso-utility curve in the investor set. In other words, it identifies the single point that provides the investor with the highest level of utility. For risk-averse individuals, this point is unique.

The problem with applying this methodology to identifying optimal portfolios is that it is difficult to figure out the risk-aversion of individuals or institutions. Just like mapping an unknown terrain, the asset allocator must try to map the clients preference structure -- never knowing whether it is even consistent from one day to the next!

II. Another Approach: Preferences about Distributions

The Markowitz model is an elegant way to describe differences in distributions of returns among portfolios. One approach to the portfolio selection problem is to choose investment policies based upon the probability mass in the lower left-hand tail. This is called the shortfall criterion. It's simplicity has great appeal. It does not require a complete topological mapping of investor preferences. Instead it only requires the investor to specify a floor return, below which he or she wants to avoid falling. The short-fall approach chooses a portfolio on the efficient frontier that minimizes the probability of the return dropping below that floor. Suppose, for instance, your specify a floor return level equal to the riskless rate, Rf. For every portfolio on the frontier, you calculate the ratio:

Notice that the shortfall criterion is like a t-statistic, where the higher the value, the greater the probability. The portfolio that has the highest probability of exceeding Rf is the one for which this value is maximized. In fact, the similarity to a t-statistic extends even further, as we will see. Another useful thing is that it turns out that it is quite simple to find the portfolio that maximizes the probability of exceeding the floor. You can do it graphically!

Identify the floor return level on the Y axis. Then find the point of tangency to the efficient frontier. In the figure, for instance, the tangency point minimizes the probability of having a return that drops below R floor. One particular floor value is of interest -- that is the floor given by the riskless rate, Rf. The slope of the short-fall line when Rf is the floor is called the Sharpe Ratio. The portfolio with the maximum Sharpe Ratio is the one portfolio in the economy that minimizes the probability of dropping below treasury bills. By the same token, it is the one portfolio in the economy that has the maximum probability of providing an equity premium! That is, if you must bet on one portfolio to beat t-bills in the future, the

tangency portfolio found via the Sharpe Ratio would be it. The "safety-first" approach is a versatile one. In the above example, we maximized probability of exceeding a floor by maximizing the slope, identifying a point of tangency. You can also find portfolios by other methods. For instance, you can check the feasibility of a desired floor and probability of exceeding that floor by fixing the Y intercept and fixing the slope. Either the ray will pass through the feasible set, or it will not. If it does not, then there is no portfolio that meets the criteria you specified. If it does, then there are a number of such portfolios, and typically the one with the highest expected return is the one to choose.

Another approach is to find a floor that meets your probability needs. In other words, you ask "Which floor return may I specify that will give me a 90% confidence level that I will exceed it?" This is equivalent to setting the slope equal to the t-statistic value matching that probability level. Since this is equivalent to a one-tailed test, you would set the slope to 1.28 (i.e. the quantile of the normal distribution that gives you 90% to the left, or 10% in the right side of the distribution. For a 95% chance, you would choose a slope of 1.644. For a 99% chance you would choose a slope of 2.32. Once you choose the slope, then move the line vertically until it becomes a tangent. This will give you both a floor and a portfolio choice.

III. A Note on Value at Risk The safety first approach can be used to calculate the value-at-risk of the portfolio. Valueat-risk is an increasingly popular measure of the potential for loss over a given time horizon. It is applied in the banking industry to calculate capital requirements, and it is applied in the investment industry as a risk control for portfolios of securities. Consider the problem of estimating how big a loss your portfolio could experience over the next month. If the distribution of portfolio returns is normal, then a three standard deviation drop is possible, but not very likely. Typically, the estimate of the maximum expected loss is defined for a given time horizon and a given confidence interval. Consider the type of loss that occurs once in twenty months. If you know the mean and standard deviation of the portfolio, and you specify the confidence interval as a 5% event (1 in twenty months) or a 1% event (1 in a hundred months) it is straightforward to calculate the "Value at Risk." Let Rp be the portfolio return and STDp be the portfolio standard deviation. Let T be the tstatistic associated with the confidence interval. T of 1.64 corresponds to a one in 20 month event. Let Rvar be the unknown negative return portfolio return that we expect to occur one in twenty times.

The equation for the line is: Rp = Rvar + T*STDp and thus, Rvar = Rp - T*STDp. Rvar multiplied times the value of the assets in the portfolio is the Value at Risk. Suppose you are considering the VAR of a $100 million pension portfolio over the monthly horizon. It is composed of 60% stocks and 40% bonds, and you are interested in the 95% confidence interval. Let us assume that the monthly expected stock return is 1% and the expected bond return is

.7%, and their standard deviations are 5% and 3% respectively. Assume that the correlation between the two asset classes is .5. First we calculate the mean and standard deviation of the portfolio:

Rp = (.6)*(.01) + (.4)(.007) = .0088 STDp = sqrt[ .6^2*.05^2 + .4^2*.03^2 + 2*.5*.6*.4*.05*.03] = .038 Then, Rvar = .0088 - 1.64*.038 = -.054
Thus, the monthly value-at-risk of the portfolio is ($100 million)(.054) = $5.4 million. Note that, despite the terminology, this does not really mean that $94.6 is not at risk. The analysis only means that you expect a loss at least as large as $5.4 million one month out of 20. This approach to calculating value-at-risk depends on key assumptions. First, returns must be close to normally distributed. This condition is often violated when derivatives are in the portfolio. Second, historically estimated return distributions and correlations must be representative of future return distributions and correlations. Estimation error can be a big problem when you have statistics on a large number of separate asset classes to consider. Third, returns are not assumed to be auto-correlated. When there are positive trends in the data, losses should be expected to mount up from month to month. In summary, value at risk is becoming pervasive in the financial industry as a summary measure of risk. While it has certain drawbacks, its major advantage is that it is a probability-based approach that can be viewed as a simple extension of safety-first portfolio selection models.

IV. Conclusion Creating an efficient frontier from historical or forecast statistics about asset returns is inherently uncertain due to errors in statistical inputs. This uncertainty is minor when compared to the problem of projecting investor preferences into mean-standard deviation space. Economists know relatively little about human preferences, especially when they are confined to a single-period model. We know people prefer more to less, and we know most people avoid risk when they are not compensated for holding it. Beyond that is guess-work. We don't even know if they are consistent, through time, in their choices. The theoretical approach to the portfolio selection problem relies upon specifying a utility function for the investor, using that to identify indifference curves, and then finding the highest attainable utility level in the feasible set. This turns out to be a tangency point. In practice, it is difficult to estimate a utility function, and even more difficult to explain it back to the investor.

An alternative to utility curve estimation is the "safety-first" technology, which is motivated by a simple question about preferences. What is your "floor" return? If you can pick a floor, you can pick a portfolio. In addition, you can identify a probability of exceeding that floor, by observing the slope of the tangency line. Safety-first also lets you find optimal portfolios by picking a floor and a probability, as well as simply picking a probability. Value at risk is becoming increasingly popular method of risk measurement and control. It is a simple extension of the safety-first technology, when the assets comprising the portfolio have normally distributed returns.

IV. Epilogue Notice that the introduction of a genuine risk-free security simplifies the portfolio problem for all investors in the world. Their optimal choice is reduced to the problem of choosing proportions of the riskless asset and the risky portfolio T (tangency). MRA (More Risk Averse) investors will hold a mix of tangency portfolio and T-bills, LRA (Less Risk Averse) investors will borrow at the riskless rate and invest the proceeds in the tangency portfolio.

If we could only figure out what the tangency portfolio is composed of, we could solve everyone's investment decision with the same product! What do you think T is composed of? The answer is in the next chapter. For more information about the utility approach to risk, see the excellent write-up by Campbell Harvey on Optimal Portfolios.. For a comprehensive hyper-text book on investment decision-making, see William Sharpe's Macro-Investment Analysis

Chapter IV: The Portfolio Approach to Risk

I. The Quest For the Tangency Portfolio In the 1960's financial researchers working with Harry Markowitz's mean-variance model of portfolio construction made a remarkable discovery that would change investment theory and practice in the United States and the world. The discovery was based upon an idealized model of the markets, in which all the world's risky assets were included in the investor opportunity set and one riskless asset existed, allowing both more and less risk averse investors to find their optimal portfolio along the tangency ray.

Assuming that investors could borrow and lend at the riskless rate, this simple diagram suggested that everyone in the world would want to hold precisely the same portfolio of risky assets! That portfolio, identified at the point of tangency, represents some portfolio mix of the world's assets. Identify it, and the world will beat a path to your door. The tangency portfolio soon became the centerpiece of a classical model in finance. The associated argument about investor choice is called the "Two Fund Separation Theorem" because it argues that all investors will make their choice between two funds: the risky tangency portfolio and the riskless "fund". Identifying this tangency portfolio is harder than it looks. Recall that a major difficulty in estimating an efficient frontier accurately is that errors grow as the number of assets increase. You cannot just dump all the means, std's and correlations for the world's assets into an optimizer and turn the crank. If you did, you would get a nonsensical answer. Sadly enough, empirical research was not the answer, due to statistical estimation problems. The answer to the question came from theory. Financial economist William Sharpe is one of the creators of the "Capital Asset Pricing Model," a theory which began as a quest to identify the tangency portfolio. Since that time, it has developed into much, much more. In fact, the CAPM, as it is called, is the predominant model used for estimating equity risk and return.

II. The Capital Asset Pricing Model Because the CAPM is a theory, we must assume for argument that ... 1. All assets in the world are traded 2. All assets are infinitely divisible 3. All investors in the world collectively hold all assets 4. For every borrower, there is a lender 5. There is a riskless security in the world 6. All investors borrow and lend at the riskless rate 7. Everyone agrees on the inputs to the Mean-STD picture 8. Preferences are well-described by simple utility functions 9. Security distributions are normal, or at least well described by two parameters 10. There are only two periods of time in our world

This is a long list of requirements, and together they describe the capitalist's ideal world. Everything may be bought and sold in perfectly liquid fractional amounts -- even human capital! There is a perfect, safe haven for risk-averse investors i.e. the riskless asset. This means that everyone is an equally good credit risk! No one has any informational advantage in the CAPM world. Everyone has already generously shared all of their knowledge about the future risk and return of the securities, so no one disagrees about expected returns. All customer preferences are an open book -- risk attitudes are well described by a simple utility function. There is no mystery about the shape of the future return distributions. Last but not least, decisions are not complicated by the ability to change your mind through time. You invest irrevocably at one point, and reap the rewards of your investment in the next period -- at which time you and the investment problem cease to exist. Terminal wealth is measured at that time. I.e. he who dies with the most toys wins! The technical name for this setting is "A frictionless one-period, multi-asset economy with no asymmetric information." The CAPM argues that these assumptions imply that the tangency portfolio will be a value-weighted mix of all the assets in the world The proof is actually an elegant equilibrium argument. It begins with the assertion that all risky assets in the world may be regarded as "slices" of a global wealth portfolio. We may graphically represent this as a large, square "cake," sliced horizontally in varying widths. The widths are proportional to the size of each company. Size in this case is determined by the number of shares times the price per share.

Here is the equilibrium part of the argument: Assume that all investors in the world collectively hold all the assets in the world, and that, for every borrower at the riskless rate there is a lender. This last condition is needed so that we can claim that the positions in the riskless asset "net-out" across all investors. From the two-fund separation picture above, we already know that all investors will hold the same portfolio of risky assets, i.e. that the weights for each risky asset j will be the same across all investor portfolios. This knowledge allows us to cut the cake in another direction: vertically. As with companies, we vary the width of the slice according to the wealth of the individual.

Notice that each vertical "slice" is a portfolio, and the weights are given by the relative asset values of the companies. We can calculate what the weights are exactly:

weight on asset i = [price i x shares i] / world wealth

Each investor's portfolio weight is exactly proportional to the percentage that the firm represents of the world's assets. There you have it: the tangency portfolio is a capitalweighted portfolio of all the world's assets.

III. Investment Implications The CAPM tells us that all investors will want to hold "capital-weighted" portfolios of global wealth. In the 1960's when the CAPM was developed, this solution looked a lot like a portfolio that was already familiar to many people: the S&P 500. The S&P 500 is a capital-weighted portfolio of most of the U.S.'s largest stocks. At that time, the U.S. was the world's largest market, and thus, it seemed to be a fair approximation to the "cake." Amazingly, the answer was right under our noses -- the tangency portfolio must be something like the S&P 500! Not co-incidentally, widespread use of index funds began about this time. Index funds are mutual funds and/or money managers who simply match the performance of the S&P. Many institutions and individuals discovered the virtues of indexing. Trading costs were minimal in this strategy: capital-weighted portfolios automatically adjust to changes in value when stocks grow, so that investors need not change their weights all the time -- it is a "buy-and-hold" portfolio. There was also little evidence at the time that active portfolio management beat the S&P index -- so why not?

IV. Is the CAPM true? Any theory is only strictly valid if its assumptions are true. There are a few nettlesome issues that call into question the validity of the CAPM:

Is the world in equilibrium? Do you hold the value-weighted world wealth portfolio? Can you even come close? What about "human capital?"

While these problems may violate the letter of the law, perhaps the spirit of the CAPM is correct. That is, the theory may me a good prescription for investment policy. It tells investors to choose a very reasonable, diversified and low cost portfolio. It also moves them into global assets, i.e. towards investments that are not too correlated with their personal human capital. In fact, even if the CAPM is approximately correct, it will have a major impact upon how investors regard individual securities. Why?

V. Portfolio Risk Suppose you were a CAPM-style investor holding the world wealth portfolio, and someone offered you another stock to invest in. What rate of return would you demand to hold this stock? The answer before the CAPM might have depended upon the standard deviation of a stock's returns. After the CAPM, it is clear that you care about the effect of this stock on the TANGENCY portfolio. The diagram shows that the introduction of asset A into the portfolio will move the tangency portfolio from T(1) to T(2).

The extent of this movement determines the price you are willing to pay (alternately, the return you demand) for holding asset A. The lower the average correlation A has with the rest of the assets in the portfolio, the more the frontier, and hence T, will move to the left. This is good news for the investor -- if A moves your portfolio left, you will demand lower expected return because it improves your portfolio risk-return profile. This is why the CAPM is called the "Capital Asset Pricing Model." It explains relative security prices in terms of a security's contribution to the risk of the whole portfolio, not its individual standard deviation.

VI. Conclusion The CAPM is a theoretical solution to the identity of the tangency portfolio. It uses some ideal assumptions about the economy to argue that the capital weighted world wealth portfolio is the tangency portfolio, and that every investor will hold this same portfolio of risky assets. Even though it is clear they do not, the CAPM is still a very useful tool. It has been taken as a prescription for the investment portfolio, as well as a tool for estimating an expected rate of return. In the next chapter, we will take a look at the second of these two uses.

Chapter V: Further Explorations of the Capital Asset Pricing Model

I. Risk-Return Tradeoff: A Technical Aside Recall from last chapter that, when investors are well-diversified, they evaluate the attractiveness of a security based upon its contribution to portfolio risk, rather than its volatility per se. The intuition is that an asset with a low correlation to the tangency portfolio is desirable, because it shifts the frontier to the left.

This institution was formalized by Stephen Ross in an article called Finance, published in The New Palgrave. It is a simple argument that shows the theoretical basis for the "pricing" part of the Capital Asset Pricing Model. Here goes: Suppose you are an investor who holds the market portfolio m and you are considering the purchase of a quantity dx of asset A, by financing it via borrowing at the riskless rate. This augments the return of the market portfolio by the quantity:

dEm = [E A - Rf]dx
Where d symbolizes a small quantity change. This investment also augments the variance of the market portfolio. The variance of the market portfolio after adding the new asset is:

v + dv = v + 2dx cov(A,m) + (dx)2 var(a)

The change in the variance is then:

dv = 2 dx cov(A,m) + (dx)2 var(A)

For small dx's this is approximately:

dv = 2 dx cov(A,m)
This gives us the risk-return tradeoff to investing in a small quantity of A:

Risk-Return Tradeoff for A = dEm/dv = [E A - Rf]dx / 2 dx cov(A,m) Risk-Return Tradeoff for A = dEm/dv = [E A - Rf]/ 2 cov(A,m)
Now, if the expected return of asset A is in equilibrium, then an investor should be indifferent between augmenting his or her portfolio with a quantity of A and simply levering up the existing market portfolio position. If this were NOT the case, then either the investor would not be willing to hold A, or A would dominate the portfolio entirely. We can calculate the same Risk-Return Tradeoff for buying dx quantity of the market portfolio P instead of security A.

Risk-Return Tradeoff for P = dEm/dv = [E m - Rf]/ 2 var(m)

The equations are almost the same, except that the cov(A,m) is replaced with var(m). This is because the covariance of any security with itself is the variance of the security. These RiskReward Tradeoffs must be equal:

[E A - Rf]/ 2 cov(A,m) = [E m - Rf]/ 2 var(m) Thus, [E A - Rf] = [cov(A,m)/var(m)][E m - Rf]

The value cov(A,m)/var(m) is also known as the of A with respect to m. is a famous statistic in finance. It is functionally related to the correlation and the covariance between the security and the market portfolio in the following way:

II. A Model of Expected Returns In the preceding example, notice that we used the expression expected returns. That is, we found an equation that related the expected future return of asset A (in excess of the riskless rate) to the expected future return of the market (in excess of the riskless rate). This expected return is the return that investors will demand when asset prices are in the equilibrium described by the CAPM. For any asset i, the CAPM argues that the appropriate rate at which to discount the cashflows of the firm is that same rate that investors demand to include the security in their portfolio:

One surprising thing about this equation is what is not in it. There is no measure of the security's own standard deviation. The CAPM says that you do not care about the volatility of the security. You only care about its beta with respect to the market portfolio! Risk is now re-defined as the quantity of exposure the security has to fluctuations in the market portfolio.

III. The Security Market Line The CAPM equation describes a linear relationship between risk and return. Risk, in this case, is measured by beta. We may plot this line in mean and space:

One remarkable fact that comes from the linearity of this equation is that we can obtain the beta of a portfolio of assets by simply multiplying the betas of the assets by their portfolio weights. For instance the beta of a 50/50 portfolio of two assets, one with a beta of .8 and the other with a beta of 1 is .9. Easy! The line also extends out infinitely to the right, implying that you can borrow infinite amounts to lever up your portfolio. Why is the line straight? Well, suppose it curved, as the blue line does in the figure below. The figure shows what could happen. An investor could borrow at the riskless rate and invest in the market portfolio. Any investment of this type would provide a higher expected return than a security which lies on the curved line below. In other words, the investor could receive a higher expected return for the same level of systematic risk. In fact, if the security on the curve could be sold short, then the investor could take the proceeds from the short sale and enter into the levered market position -- generating an arbitrage in expectation.

IV. Expectations vs. Realizations It is important to stress that the vertical dimension in the security market line picture is expected return. Things rarely turn out the way you expect. However, the CAPM equation also tells us about the realized rate of return. Since the realization is just the expectation plus random error, we can write:

R i = Rf + i [ Rm - Rf ] + ei
This is useful, because it tells us that when we look at past returns, they will typically deviate from the security market line -- not because the CAPM is wrong, but because random error will push the returns off the line. Notice that the realized R m does not have to behave as expected, either. So, even the slope of the security market line will deviate from the average equity risk premium. Sometimes it will even be negative!

V. An Example The appeal of the CAPM is clear -- it radically simplifies an inherently complex and troublesome problem. The question of the appropriate discount rate becomes virtually a back-of-the-envelope calculation! In fact, if you know a security's beta, estimating the discount rate is a snap: multiply beta times the expected risk premia of the market portfolio over the riskless rate. For example, suppose you are a banker considering a private equity investment in a company with a new drug process. The process is inherently risky, i.e. the standard deviation of the project is 75% per year. The beta of the project is .5. The Rf = 5% and the E[Rm] = 13.5%. What is the required rate of return on the project?

Theory tells us that the answer does not depend upon the volatility associated with the returns. Instead we use the beta of the project.

E[Rdrug]= 5% + (.5)(13.5% - 5%) = 9.25%

This is the required rate of return on the project. The answer would not change if the range of outcome next year broadened or narrowed. The is the only relevant piece information -- now all that remains is to estimate it!

VI. How Do You Estimate ? may be all we need, but it is not immediately clear how it should be estimated. What we really need is a quantitative estimate of how the future return changes in response to future changes in the world market portfolio. Good Luck! It is tough to even guess the empirical composition of the market portfolio, let alone estimate a beta. In practice (although it is not theoretically justified) analysts typically use the S&P 500 equity risk premium in this calculation. To estimate beta, regress the security returns for the past several periods (usually 60 months) on the market returns. The slope in this regression is an estimate of .

Notice that this shows concretely that empirical property of as it measures the co-movement of the security with the market. Unfortunately, since the S&P 500 is not the world market portfolio, we are somewhat in the dark about how well this beta measures the true systematic risk.

VII. Assessing the CAPM The CAPM is a classical model in finance. It is an equilibrium argument that, if true, answers most important investment questions. It tells us where to invest, how to invest and what discount rate to use for project cash flows. Not only that, it is a disarmingly simple

model. The expected return of a security depends upon a simple statistic: . The relationship between risk and return is linear. Calculation of portfolio risk is trivial. At the same time, the CAPM is revolutionary. It tells us that the variance of a project is NOT a factor in determining the appropriate, risk-adjusted discount rate. It turns financial research from roll-up-your-sleeves fundamental analysis into a statistics problem. In short, the CAPM turned Wall Street on its head.

VIII. Conclusion. Is the CAPM True? Here comes the bad news. Despite twenty years of attempts to verify or refute the Capital Asset Pricing Model, there is no consensus on its legitimacy. There are a few hints that the model is incorrect. For starters, we all hold different portfolios. Therefore, it cannot be exactly true. Researchers have focused upon the more interesting issue of whether rates of return depend upon and whether the elegant, linear form of the model holds for stocks. What they have found is that real markets typically deviate broadly from the exact model. While there are long periods in U.S. Capital market history when realized returns are positively related to betas, there are also long periods when they are not. Among the most forceful arguments against the CAPM advanced in recent times is a study by Eugene Fama and Kenneth French. These authors found that beta did a relatively poor job at explaining differences in the actual returns of portfolios of U.S. stocks. Instead, Fama and French noted that there were other variables besides beta with respect to the market that explained returns. Some of these were "fundamental" ratios long used by financial analysts in the preCAPM era such as Book to Market Ratio and Earnings Price Ratio. Another was simply the relative size of the company. The evidence against the CAPM continues to grow and despite its elegance, most researchers have turned to more more complex, but more powerful models.

Chapter VI: The Arbitrage Pricing Theory

I. Holding the Security Market Line No matter how theoretically appealing it may be, even the most ardent supporters of the Capital Asset Pricing Model admit the model does not quite fit reality. It is difficult to test the CAPM without data on the global wealth portfolio, and the S&P just won't do. We know that some of the most obvious implications of the CAPM are violated -- for instance, we all hold different portfolios. We are still in the dark about the more fundamental implications, such as the question of whether only systematic risk is priced. In the 1970's, financial researchers took a different approach to the issue of identifying a discount rate for securities. This time, the security market line was the motivation for further theory. Consider this -- even if the CAPM is untrue, the security market line STILL remains an appealing diagram. The SML diagram contains the seeds to a different asset pricing model, called the Arbitrage Pricing Theory. The APT was developed by Stephen Ross. Like the CAPM, it argues that discount rates are based upon the systematic risk exposure of the

security, as opposed to the total risk. Unlike the CAPM, it does not require that all investors behave alike, nor does it claim that the capital-weighted market portfolio i.e. the tangency portfolio, it the only risky asset that will be held.

II. Who Put the 'A' in the APT? Consider a world where investors are broadly diversified, but there may be multiple sources of risk in the economy. Instead of everyone caring solely about the market portfolio, investors actually care about lots of things, including shifts in stock index levels, interest rates, inflation, changes in GNP or other broad macro-economic factors that are difficult to purge from your portfolio through diversification. For now, focus on one of these factors -the S&P 500. There is no need to presume that this or any factor matches the world wealth portfolio -- it is just one source of risk that people care about.

Suppose, for argument's sake, that security A plotted off the S&P 500 security market line. The CAPM says that it cannot, but what if it did? If everyone realized that A's expected return was higher than B's, then many of them would try to exploit such an opportunity. If A lies above the SML (whether in one dimension or several!) then this implies that A is underpriced given its beta. Investors will notice this, and will buy A. They may finance this purchase by selling (i.e. shorting) B, a portfolio with the same systematic risk. In our example, the purchase of A by investors will drive up the price of A, reducing its expected return, and force it into the neighborhood of the security market line. In other words, deviations from linear pricing will be met swiftly by "arbitrage." In fact, arbitrage in this context is a slight misnomer, because this investment strategy involves some risk. It is more properly terms an "arbitrage in expectations" because the investor is locking in a positive EXPECTED payoff, not a positive GUARANTEED payoff.

II.1 An Aside on Short-selling Short-selling is a procedure that allows you to profit when the price of a security declines. In effect, it allows you to take a negative position in the security -- just the opposite of a long position i.e. holding the security. To short a stock, you must borrow a share from someone who holds it (typically via your broker) and then promise to return the share of stock upon demand. Then you sell the share of stock. This activity has two effects. First, you get money from the sale of the share of stock. Second, you incur an obligation to return a share of the same stock in the future. If the stock price drops, you can fulfill your obligation by buying a share on the market for less that the price at which you shorted it. The more the price drops the more you profit. Of course, if the price rises, you lose.

III. Arbitrage in Expected Returns: An Example Suppose you observed the following conditions: 1. Risk-free bonds may be purchased at a cost of $100 (or in fractions if required). They are known to pay off in one year $110 with certainty. 2. All investors can borrow and lend at the riskless rate. 3. Shares of the market portfolio may be purchased for $100 each. They are expected to pay off $120 at the end of the year, but there is uncertainty involved. Shares of the market may be purchased and shorted without transactions costs. 4. Shares of asset A may be purchased for $100 and they are expected by everyone to be worth $150 at the end of the year. Asset A has a beta of 1.3. As with the market, shares of A may be shorted and purchased without transactions costs. How would an investor proceed in an expectations arbitrage? First, calculate the expected return of asset A, under linear pricing model assumptions: E[Ra = R f + a(E[Rm] - R f) where Rf = the riskless rate E[Rm] = expected return on the market E[Ra] = expected return on asset A This tells us that everyone SHOULD expect a share of A to be worth, at the end of the period: $100 x [(1+ .10 + 1.3 x .10] = $123 The expected return in this case yields an expected future value which is lower than $150. This is a logical inconsistency if the index model were true. In practical terms it is underpriced. To exploit this underpricing we take the following actions:

ACTION 1) Buy one share of asset A, costing money 2) Short 1.3 shares of market portfolio, generating cash proceeds 3) Buy .3 bonds, costing money NET POSITION

POSITION -$100 $130 -$30 $0

SYSTEMATIC RISK

()
1.3 -1.3 0 0.0

Now, what happens at the end of the period? The market has a realization, different from its expectation and asset A has a realization different from its expectation. This may be expressed as: Rm = E[Rm] + em Ra = E[Ra] + eA When things turn out exactly as expected, em and eA both equal zero. Thus, action (1) yields $150, action (2) yields -$156 and action (3) yields $33. This is a net gain of 183-156 = $27. Now suppose the returns did not occur as expected, i.e. the errors were not zero, nor were they equal. You would receive: $27 + $100 + eA - $130 em Sometimes this is negative, sometimes this is positive. It has a variance, and thus is risky. In other words, the "A" in the APT is not true arbitrage, but arbitrage in expectations.

IV. The Arbitrage Pricing Theory Argument The APT argument is best understood from the arbitrage in expectations example presented above. To achieve "arbitrage" pricing, we must assume that:

There exist some important systematic risks driving security returns in a linear fashion Investors perceive these risks and can estimate the sensitivity of the security to them Some investors are risk-takers in the economy These investors can and will exploit differences in expected return by undertaking risk arbitrage

Then:

Expected returns will be determined such that the expected returns of securities in the economy plot on or close to the security market line in as many dimensions of risk as there are factors.

V. The World of the APT The APT gives up the notion that there is one right portfolio for everyone in the world, and it replaces it with an explanatory model of what drives asset returns. The world of the APT is not some ideal, knife-edge equilibrium in which all investors are stuck in the same portfolio. It is a world with many possible sources of risk and uncertainty. More formally, it is based upon the assumption that there are a few major macro-economic factors that influence security returns. No matter how thoroughly you diversify, you can't avoid these factors, although you can tilt your portfolio away from them. The APT claims that investors will "price" these factors precisely because they are sources of risk that can't be diversified away. That is, they will demand compensation in terms of expected return for holding securities exposed to these risks. Just like the CAPM, this exposure is measured by a factor beta. It is tempting to see the APT as a behavioral model. It describes a world in which investors behave intelligently by diversifying, but they may chose their own systematic profile of risk and return by selecting a portfolio with its own peculiar array of betas. While formal proofs of the APT rely upon static equilibrium arguments, the spirit of the APT is an active one. It allows a world where occasional mispricings occur. Investors constantly seek information about these mispricings and exploit them as they find them. It allows for an industry of information collectors, risk arbitrageurs and speculators. It allows for different types of investors as well as evolving types of risks. In other words it describes a world somewhat closer to the world in which we live.

VI. Applying the APT Finding Factors How do we apply the APT? One difficulty with the model it is generality. We have left the simple world of the CAPM. We no longer know exactly what sources of systematic risk people truly care about. On the other hand, reading the financial section of the newspaper we can get idea. The Wall Street Journal for instance, regularly reports on surprises in interest rates, surprises in GNP, surprises in inflation and changes in the stock market indices. All of these are candidates for APT factors. Indeed, we may not actually need to identify the economy's risk factors. We only need to find a collection of things that together are good proxies for them. After the theoretical development of the APT, Chen, Roll and Ross set out on a quest for the factors. They found that a collection of four or five macro-economic series' that explained security returns fairly well. These factors turned out to be surprises in inflation, Surprises in GNP, surprises in investor confidence (measured by the corporate bond premium) and shifts in the yield curve. In general these do as good a job at explaining returns as the S&P index. Of course, no one really knows if these are the "true" factors. As the APT continues to be used in practice, other variables are likely to be

used. Once factors are chosen, only the unanticipated portion of the factor is used for estimating the APT model. As with the CAPM, we usually regress historical security returns on the factor to estimate 's. These 's are used in a model of expected returns to estimate the discount rate. Building portfolios The APT is a useful tool for building portfolios adapted to particular needs. For example, suppose a major oil company wanted to create a pension fund portfolio that was insulated against shock to oil prices. The APT allows the manager select a diversified portfolio of stocks that has low exposure to inflation shocks (oil prices are correlated to inflation). If the CAPM is a "one size fits all" model of investing, the APT is a "tailor-made suit." In the APT world, people can and do have different tastes and care more or less about specific factors. Sensitivity analysis With the APT we can model the effects of different economic scenarios on the investment portfolio. Once factor betas are estimated, we can describe the expected change in security returns with respect to changes in that factor. How will my portfolio perform in a recession? Am I exposed to shifts in the yield curve? These are typical questions addressed by APT analysis.

VII. Conclusion: APT as a Model of Expected Returns The APT has a number of benefits. First, it is not as a restrictive as the CAPM in its requirement about individual portfolios. It is also less restrictive with respect to the information structure it allows. The APT is a world of arbitrageurs and vendors of information. It also allows multiple sources of risk, indeed these provide an explanation of what moves stock returns. The benefits also come with drawbacks. The APT demands that investors perceive the risk sources, and that they can reasonably estimate factor sensitivities. In fact, even professionals and academics can't agree on the identity of the risk factors, and the more betas you have to estimate, the more statistical noise you must live with.

Chapter VII: Where Do Betas Come From?

I. Beta Risk In the previous chapter, we focussed on the Arbitrage Pricing Theory as an alternative model to the classical CAPM. In fact, the CAPM is not inconsistent with the APT. Although its intellectual foundations differ, the two theories are basically arguments that the expected return of a security (i.e. the appropriate discount rate for its cash flows!) is a linear function of systematic risk. The major difference in practice between the CAPM and the APT is that the CAPM uses one risk variable, the market portfolio, while the APT uses several. The APT factors are typically macro-economic - they are related broadly to the

economy. None the less, these factors will also affect the market portfolio. Thus, when you use the CAPM, the one single factor will reflect the variation in the APT factors. So far, we have used beta as a way of calculating expected returns, but in fact, it is also a risk measure. What kind of risk does it measure, exactly? It certainly does not measure exposure to diversifiable risk, since CAPM and APT assume that investors are diversified. Instead, it captures systematic risk -- risk common to the entire economic system, the market. Macro-economists call this business cycle risk, and have noted that major industrial economies have historically fluctuated through periods of boom to periods of bust. Stock prices are barometers of expectations about these cycles. In fact the first widely quoted index, the Dow, Jones Average, was used by its creator to try and identify peaks and troughs in the market. These are called "Bull" and "Bear" markets, and there are of interest to all investors because it is virtually impossible to avoid broad market trends. When the market crashes, as it did in the early 1930's and in the early 1970's, for instance, few stocks are unaffected, however some stocks are more severely hurt by a crash than others. Stocks that drop dramatically when the market falls are those with high betas. The good news is that these same high-beta stocks recover more quickly when the market changes from a "Bear" to a "Bull." Betas tend to be related to industry. High-technology, for instance, is a high-beta industry. The food industry is a low beta industry. The expectations about future cash flows of high technology are high when the economy is in a period of expansion, growth and development but low when it contracts. On the other hand, the food industry is relatively isolated from broad market booms and busts because people always need to eat! The relationship between the returns of the security and the market factor can be seen by plotting market returns on one axis and the returns for one company over the same period on the other.

For virtually all stocks in the economy, this relation will be positive. Most securities have some degree of positive exposure to bull and bear market cycles. In fact, we quantify this exposure by the slope of the regression line estimated from this graph. The steeper the slope, the more systematic risk, the shallower the slope, the less exposed the company is to the market factor. In fact, the coefficient quantifies the expected return for the stock, depending upon the actual return of the market. For instance, consider a company with a beta of 1.5. If the market return is 20 percentage points over the T-bill rate in one year, then we expect the stock return to be 30 percentage points over T-bills in that year. In practice, this is the way that a company's beta is measured, and consequently, this is how the discount rate is estimated. The analyst obtains historical returns on the company, and compares them via linear regression to the market factor, or macro-economic factors if APT is used. It is important to note that a low beta does not mean that the standard deviation of the company's return is low. Even though the relationship to the market index may be almost flat, the variation in company returns can be large. Thus, volatile securities can and often do have low market betas. Although linear regression is an easy statistical tool to use to estimate betas, remember that it is only provides an estimate. In fact, the validity of beta will depend upon several things. Betas can and do change over time, as companies change their business. The regression assumes that betas are fixed over the estimation period. This is why analysis use a limited time period, say, five years, to obtain beta. Any longer interval may make this assumption grossly invalid. Second, since beta is a coefficient from a regression, it is only as valid as the data used to calculate it. Second, you may not have the right regressors in your model. This is an APT vs. a CAPM issue. The factor (e.g. S&P 500) may not completely capture systematic risk exposure. This is not uncommon. Suppose a company is very exposed to

interest rate risk, but has a moderately low S&P 500 beta. If interest rate risk demands market compensation by portfolio investors, then you will be underestimating the systematic risk of the company if you only use the S&P 500 when calculating the expected return and discount rate via a single-factor model. There are finally some crucial data issues. You might not have enough accurate data for beta estimation. When you only have a few points, the slope in the regression has a high standard error and you cannot trust it. When the t-statistic of beta is low, sometimes you cannot even reject the hypothesis that beta is different from zero, even though economic reasoning suggests that the firm is exposed to business cycle risk, or factor risk. In that case, trust economics, not statistics. The most extreme case of not enough data is when you have to estimate the beta of a firm that is not publicly traded In that case, what do you do? Analysts usually rely upon "comparable" firm's betas. That is, they look for betas of firms in the same industry, and assume that the systematic risk exposures are the same throughout the industry. There is one other crucial thing to control for when using comparable betas, and that is, leverage.

II. Financial Leverage and Beta Even firms within one industry have different levels of debt, and increasing debt increases leverage. Increasing leverage increases beta. Recall, in the APT arbitrage in expectations example, that we could "synthesize" a security with a beta of 1.3 by borrowing 30% of our wealth, and investing the total in an asset with a beta of one. We moved out the security market line by borrowing. Suppose, for instance, that investor A hold a portfolio of $100 invested in an S&P 500 index trust. In order to increase his expected return, investor B, who also has $100, borrows an additional $30 for one year at 0% interest, and invests $130 in the S&P 500 index trust. What will happen if the S&P goes up by next year? A will have $110, for a gain of 10%, while B will have $143 - $30, leaving a gain of 13%! What will happen if the market drops by 10% next year? A will have $90, a loss of -10%, while B will have a net loss of $87, a 13% loss. B's leverage increased his exposure to market risk. Leverage can be used by corporations as well as individuals to increase their expected returns, and in fact, this is exactly what some firms do. Even if they are in a low-beta business, such as a utility, they can increase expected return through leverage.

III. Leverage and the Cable T.V. Industry The cable television industry is a utility. If we could observe an unlevered cable company, it would undoubtedly have a low beta. Good television reception is like food, people can't seem to live without it, even in a recession. Thus, it is not as cyclical as some other businesses. Empirical research has shown that the beta of the average all-equity cable TV company (called an asset beta) is .67, but most firms borrow more that their total equity value! Thus, the beta of their equity (that is, the beta measured by regression of stock returns on the market) is greater than one: 1.85. This increases the average expected return in the industry from 11.39 to 21.41.

How do we calculate the Asset beta? For that matter, suppose I had a company whose leverage differed from the average? This is fairly straightforward to do. To determine your firm's beta from industry norms, assume that the company is a portfolio of two securities, a debt security and an equity security. In fact, from the perspective of an investor who can potentially buy up all outstanding stocks and bonds of the company, it is. Thus, the beta of the company is the weighted average of the betas of the two parts. It is typical to assume that the beta of debt is zero, which vastly simplifies estimation and calculation, but is only approximately correct.

To apply this approach: 1. find the average industry beta of equity (by regression or reference book)

2. find the average industry leverage :

o o o o o

i.e. weight on debt & weight on equity weight on debt is: D/(D+E) weight on equity is: E/(E+D) where E = market value of equity (shares x price/share) and D = market value of debt (Face value is usual approximation)

3. Find the "unlevered" asset beta of industry, assuming beta of debt = 0

4. Lever up to your own firm's debt level.

IV. Capital Budgeting Applications of Levered Betas: Discounting Cash Flows Leverage can have a huge effect on financial decisions. For instance, suppose that you ignored the effect of leverage in the cable television industry. You might draw the mistaken conclusion that cable t.v. assets are unusually sensitive to business cycle fluctuations, when in fact they are relatively stable. This has immediate implications for investment decisions.

Example 1: Project Valuation Suppose you are an analyst working for AT&T, the telephone company. The company has a large cash "war chest" for investment in new opportunities and it is considering a move into providing local cable television service. It is currently evaluating the profitability of bidding on the franchise for the borough of Queens, in New York City. You have been asked to evaluate an all-equity investment in this new cable system. Based upon estimates subscriber rates, you calculate that the project will have net cash flows of $234 million / year, and for simplicity, assume that this can be considered a perpetuity, with no future growth or decline in cash flows. What is your estimate of the value of the project? You have the cash flows, but you need the discount rate. Obviously, since there is no

existing Queens cable company you cannot observe the historical beta, but you can use the industry norms. In this case, the cable T.V. asset beta = .67. Assume the current riskless rate is: rf = 5.5, and you take the equity premium to be the long-term historical average: ERP = 8.5. Project Value = Annual net cash flow / CAPM expected return Project Value = $234/(.055 + .67*.085) = $2.09 Billion

Example 2: Merger and Acquisition Application Another way firms move into a new industry is through acquisition -- that is to buy the outstanding stock of another firm. Suppose you are an AT&T analyst, and were considering the purchase of a successful cable television company Cable Vision, a firm with $3.6 billion in outstanding equity (that is, the price per share times the number of shares). Assume you know that you have estimated Cable Vision's beta as 2, and that the earnings are $150/year. Also, assume that Paul Kagan the media industry security analyst estimates the growth in earnings to by 18% per year. To begin your analysis, assume earnings are a good estimate of net cash flow. Use analysts ' forecast of earnings growth to apply perpetuity model with growth: Assume beta = 2 Earnings = $150 million Growth = 18% Rf = 5.5 ERP = 8.5 Estimate R from equity beta: R = .055 + 2*.085 = .225 P = 150/ (.225 - .18) = 3.33 Billion. In other words, maybe Cable Vision is slightly overpriced.

Example 3: P/E Ratios as Approximate Discount Rates There are many reasons why the perpetuity model is not an exact formula for corporate valuation. First, it assumes no uncertainty about future cash flows or future discount rates. Second, it is a highly stylized model of future cash flows. Third, it requires estimation of inputs that cannot always be correctly estimated, e.g. the growth of earnings. None-the-less, P/E ratios must have some relationship to discount rates. Note that if:

This Ri is calculated from earnings and price, not from beta, so it is an independent check on the level of systematic risk. Ri may be calculated and compared to CAPM/APT discount rate. E.G. If: Cablevision P/E = 3,600/150= 24 then Rp/e = 1/24+ .18 = .042 + .18 = .221 Matching the CAPM/APT discount rate pretty well.

Example 4: Project Choice The firm itself may be thought of as a portfolio of projects, each with a project (i.e. asset) beta. In this setting, cash flows from each project should be discounted at the rate appropriate to that project. This is important, because the wrong discount rate may result in an incorrect capital budgeting decision. For instance, what if you discounted every project at the company cost of capital?

You will reject some worthwhile projects with low betas, and you accept high beta projects that make the firm riskier. You will end up selecting for exposure to systematic risk. What if you take projects below your company cost of capital? Doesn't this mean that you will be borrowing at a higher rate than your projects are yielding? No! Accepting lower beta projects will lower the expected return of the firm and thus lower the financing costs proportionally.

V. Conclusion CAPM betas and APT factor loadings are more than inputs to estimates of expected returns. They are measures of the systematic risk of the company or the portfolio. Both asset pricing models are linear, which implies that the betas measure the amount that actual returns for a security are expected to change when the market (or macro-economic factor) changes. In practice, betas are estimated with historical data, using regression techniques. When historical data is not available, industry comparables are used, and adjustments are made for leverage. Firms may use leverage to adjust their expected return and systematic risk exposure just as investors do. The beta of the underlying asset held by the firm may be much lower than the observed beta of the stock of the company, if the company is highly levered. We used the Cable T.V. industry to explore how to lever and unlever the beta of companies. This method can be applied to a number of corporate finance problems, including decisions about investment and acquisition.

Chapter VIII: Information and the Efficiency of the Capital Markets

I. A World of Arbitrageurs Recall the original motivation for the APT: because active investors search for opportunities to exploit arbitrage in expectations, they will not allow securities to plot far from the security market line (or plane). Thus, the security market line must be approximately correct, for if it were not, the opportunities to make money would be huge.

Opportunities for exploiting this "Arbitrage in Expectations" are likely to be fairy rare. However, when they arise, or are discovered through research, investors will seek to exploit them. Enormous potential rewards to arbitrage in expectations will motivate firms to do research about expected returns, and research about betas. This research will include analysis about cash flows and discount rates, as well as idiosyncratic information such as the health of the CEO, the relative merits of the product and so on. Arbitrageurs will seek out any information that will change their expectations about future returns sufficiently to allow profitable exploitation through buying or shorting the security. When such opportunities arise, there is a powerful motivation to seize the chance quickly. There is no telling how fast other arbitrageurs can find out what you have when they start buying, the price of an underpriced asset will rise, and the chance to make money will go away -- at least in theory. The issue we address in this chapter is "How well does this arbitrage work, and how quickly do mispricings go away?"

II. Mispricings Created by the Arrival of Information One answer to these questions is provided by looking at how market participants react to news that is suddenly revealed to the public. Major corporate events can immediately change expected future cash flows. For instance, a major disaster such as the Bophol chemical spill immediately drove down the Union Carbide stock price. In fact, prices react within a matter of minutes to such news, and the reaction is over within the day! In empirical "event studies" which focus on corporate news releases, there is little evidence that you can make money by investing on yesterday's news. This means, for instance, that when you read in the Wall Street Journal that a company announced the discovery of a new cure for the common cold at a news conference yesterday, arbitrageurs have already bought the shares, and driven the price up.

Often there is major news about the discount rate used to discount the future cash flows in valuation. For instance, when the Federal Reserve cuts the discount rate, we expect the net present value of corporate securities to increase -- that is stocks should jump. When discount rate changes are announced, stock prices react that day, and not the next day. This empirical evidence strongly indicates that, at least in the highly liquid, openinformation economy of the U.S. capital markets, stock prices are Efficient.

III. Watch My (Invisible) Hand The market is said to be efficient if it rapidly and completely impounds all relevant information into asset prices. This is nothing more than saying that Adam Smith's "invisible hand" of the market place works quickly! If prices are unfair (i.e. the asset is overpriced) then arbitrageurs will short the asset, until reduced demand for purchasing it caused the price to fall. The opposite, of course is true for underpriced securities. This informational efficiency is different than, say, the economic concept of Pareto efficiency, which concerns allocation. To say that a market is efficient is to make a statement about the speed at which new information filters into the price. The degree of market efficiency depends upon a number of things. First, an illiquid market is not likely to be efficient. Speed of reaction clearly depends upon the ability of the participants to buy and sell the securities. Second -and this is psychological -- it depends upon the existence of at least some investors with cool-headed, careful economic judgment. Suppose investors quickly hear about the latest disaster, such as the collapse of Barings, but the could not reasonably evaluate its effect on the net present value of the company. An inefficient market may over-react to bad or good news -- not because information is slow to arrive, but because careful economic reasoning does not prevail. Of course, if you can keep your head while those around you are losing theirs, you may make a profit in this type of inefficient market. There is some evidence that the U.S. stock market over-reacts to bad and good news, however it is still not clear whether this pattern may be exploited by a clever arbitrageur. In fact, if there were clear evidence on this point, I am sure the opportunity to exploit the passions of fellow investors would soon disappear! The greatest investor of all at exploiting market over-reaction was Nathan Rothschild, the London banker. He was known to have the most sophisticated information network in Europe, and everyone knew he would have the latest news about the outcome of the battle of Waterloo. Would he buy or sell Bank of England securities? One day Rothschild came out and quietly sold. Suddenly, astute investors got wind of this, and reacted with a flurry, dumping everything they owned. Rothschild then quietly bought in the panic. He made a killing! By the 18th century, financial news traveled fast -- in Rothschild's case, it may have even traveled by carrier pigeon. Studies of the efficiency of stock prices in this era indicate that when prices moved on the Amsterdam Stock Exchange on Monday, by Thursday they would move on the London Exchange -- this is about the time it took for a fast messenger to travel the distance from city to city, crossing the English Channel.

IV. Benefits of an Efficient Market

So far, arbitrageurs sound like vultures waiting to swoop in for the kill. They take risks to exploit new information at the expense of the less informed. The costs seem to be rewarding opportunism at the expense of other investors. Are there any benefits to having a market operate efficiently? Arguments in favor of efficient capital markets are: (1) the market price will not stray too far from the true economic price if you allow arbitrageurs to exploit deviations. This will avoid sudden, nasty crashes in the future. (2) An efficient market increases liquidity, because people believe the price incorporates all public information, and thus they are less concerned about paying way too much. If only the market for television sets were as efficient as the market for stocks! A lot less comparison shopping would be needed. (3) Arbitrageurs provide liquidity to investors who need to sell or buy securities for purposes other than "betting" on changes in expected returns. Currently, China is seeking to limit access to global financial information in Shanghai (site of its major stock exchange). The government wishes to keep certain kinds of information from market participants. Is this desirable? Will this be possible? V. Market Efficiency & Corporate Managers Market efficiency has implications for corporate managers as well as for investors. This takes a lot of the "gamesmanship" out of corporate management. If a market is efficient, it is difficult to fool the public for long and by very much. for instance, only genuine "news" can move the stock price. It is hard to pump-up the stock price by claims that are not verifiable by investors. "Fake" news will not move the price -- or if it does, the price will quickly revert to the pre-announcement value when the news proves hollow. Publicly available information is probably impounded in the price already. This is hard for some managers to believe. An example is Sears' attempt to sell the Sears Tower in Chicago in the late 1980's. The company believed that, since it carried the property on its balance sheet at greatly depreciated values, the public did not credit the company with the full market price of the building and thus Sears stock was underpriced. This proved to be false -- in fact, it seems that Sears was overestimating the value of the building and the stock price was relatively efficient! Another lesson: accounting tricks don't fool anybody. Don't worry about timing accounting charges and don't worry about whether information is revealed in the footnotes or in the statements. An efficient market will quickly figure out the meaning of the information, once it is made public.

VI. Three Degrees of Efficiency What kind of information is impounded in the stock price? It turns out that there are lots of different levels of market efficiency, depending upon the source or the information being impounded. The best way to illustrate this is by example. Suppose you had a hyper-efficient market that impounded All private information. This means that even a personal note passed between the CEO and the CFO regarding a major financial decision would suddenly impact the stock price! If so, this is called Strong-Form Efficiency. Few people believe that the market is strong-form efficient, but it is nice to have this benchmark! How about all public information? That is, all information available in annual reports,

news clippings, gossip columns and so on? If the market price impounds all of this information the market is called Semi-Strong Form Efficient. Most people believe that the U.S. equity markets by and large reflect publicly available information. But consider this -is information I put on the Internet public? Are government files available under the freedom of information act public? There must be subtle shades of semi-strong market efficiency, but they are not typically differentiated. Each new piece of information an analyst gathers should be carefully considered with regard to whether it is already impounded in the stock price. The easier it was to get, the more likely it is to have already been traded upon. The final form of market efficiency is Weak Form Efficiency A weak-form efficient market is one in which past security prices are impounded into current prices. Since past prices are deemed public information, weak form efficiency implies semi-strong form efficiency and semi-strong form efficiency implies strong form efficiency. Weak form efficiency implies that you can't make excess profits by trading on past trends. Funny, a lot of people do just that. They are called technical analysts, or chartists. What would you do if you noticed that every time the market went up by 1%, the next day on average, it went up again by 1/2 %? What would you do if you noticed that every time the market went down by 1%, the next day on average, it went down again by 1/2 %? If your answer is that you would buy on an up day and sell on a down day, you have the makings of an active technical trader! Academics have been testing trading rules like this for forty years, and traders have been exploiting them for even longer. The concept behind the simple rule described above is momentum. Although it is a widely used concept for technical investing, there is no evidence that any short-term market-timing rule actually makes money. The reason for this is the following: What if everyone followed the same strategy? Wouldn't the opportunity go away? It turns out that daily to stocks are returns are weakly autocorrelated (i.e. they have momentum, but the costs of exploiting this pattern are high. You have to buy and sell stocks every day, and in doing so, you have to pay brokerage fees. Thus, while major patterns in stock prices should not exist, weak patterns that are too costly to arbitrage may persist. If these simple trends are arbitraged away, then the market will follow a random walk, i.e. past deviation from expected returns tell you nothing about future deviations from expected returns. Charles Henry Dow, Wall Street Journal founder and creator of the Dow index (in 1896) noted that his stock price index alternated between bull and bear markets. He hoped to time this market, and his dream has long outlived him. The financial press (and web) are rife with market timing newsletters, purporting to find long-term patterns in the stock market. Recent research has suggested that markets may actually follow to multiple-year cycles. But can you make money? It is still hard to tell! Current hedge fund operators and high-tech investors use sophisticated "artificial intelligence" and "neural nets" to find patterns. Some make money. Is it luck? If it works, why do they offer you the opportunity to buy their software?

VII. Evidence For Market Efficiency

A simple test for Strong Form Efficiency is based upon price changes close to an event. Acts of nature may move prices, but if private information release does not, then we know that the information is already in the stock price. For example, consider a merger between two firms. Normally, a merger or an acquisition is known about by an "inner circle" of lawyers and investment bankers and firm managers before the public release of the information. When these insiders violate the law by trading on this private information, they may make money. They also make it to the SEC's wall of shame.

Unfortunately, stock prices typically move up before a merger, indicating that someone is acting dishonestly. The early move indicates that the market has a tendency towards strong-form efficiency, i.e. even private information is incorporated into prices. However, the public announcement of a merger is typically met with a large price response, suggesting that the market it not strong-form efficient. Leakage, even if illegal, does occur, but it is not fully impounded in stock price. By the way, until recently, insider trading was legal in Switzerland. Is the stock market semi-strong form efficient? The most obvious indication that the market is not always and everywhere semi-strong form efficient is that money managers frequently use public information to take positions in stocks. While there is no evidence that they beat the market on a risk-adjusted basis, it is hard to believe that an entire industry of information production and analysis is for naught. It seems likely that there is value to publicly available information, however there are probably degrees to which information really is public knowledge. What is surprising is that recent studies have shown some evidence that excess returns can be made by trading upon very public information. These tests usually take the form of "backtesting" trading strategies. That is, you play a "what-if" game with past stock prices, and pretend you followed some rule, using information available only at the time

of the pretend trade. One common rule that seems to perform well historically is to buy stocks when the dividend yield is high. This apparently has made money in the past, even though the information about which stocks have high yields and which have low yields is widely available. Another rule that generates positive excess returns in back-tests is to buy stocks when the earnings announcement is higher than expected. This seems simple, since current announcements and even forecasts are widely available as well. Does this mean that it is easy to become rich on Wall Street? Hardly! The profitability of these simple trading rules depends upon the liquidity of the stocks involved, and trading costs ("frictions"). Sometimes the costs outweigh the benefits. While many investment managers explain that they pursue a strategy of buying "Value" stocks (such as low P/E firms) few of these managers have consistently superior track records. The assumption of semi-strong form efficiency is a good first approximation for a market with as many sharp traders and with as much publicly available information as the U.S. equity market. Is the stock market weak form efficient? Weak form efficiency should be the simplest type of efficiency to prove, and for a time it was widely accepted that the U.S. stock market was at least weak form efficient. Recall that weak form efficiency only requires that you cannot make money using past price history of a stock (or index) to make excess profits. Recall the intuition that, if people know the price will rise tomorrow, then they will bid the price up today in order to capture the profit. Researchers have been testing weak form efficiency using daily information since the 1950's and typically they have found some daily price patterns, e.g. momentum. However, it appears difficult to exploit these short-term patterns to make money. Interestingly, as you increase the horizon of the return, there seems to be evidence of profits through trading. Buying stocks that went down over the last two weeks and shorting those that went up appears to have been profitable. When you really increase the horizons, stock returns look even more predictable. Eugene Fama and Ken French for instance, found some evidence that 4 year returns tend to revert towards the mean. Unfortunately, this is a difficult rule to trade on with any confidence, since the cycles are so long -- in fact, they are as long as the patterns conjectured by Charles Henry Dow some 100 years ago! Does this all lend credence to the chartists, who look for cryptic patterns in security prices -- perhaps. But in all likelihood there is no easy money in charting, either. Prices for widely trades securities are pretty close to a random walk, and if they were not, then they would quickly become so, as arbitrageurs moved in to buy the stock when it is underpriced and short it when it is overpriced. But who knows. Maybe a retired rocket scientist playing around with fractal geometry and artificial intelligence will hit upon something -- of course if he or she did, it wouldn't become common knowledge, at least for a while!

VIII. Conclusion

The efficient market theory is a good first approximation for characterizing how prices is a liquid and free market react to the disclosure of information. In a word, "Quickly!" If they did not, then the market is lacking in the opportunism we have come to expect from an economy with arbitrageurs constantly collecting, processing and trading upon information about individual firms. The fact that information is impounded quickly in stock prices and that windows of investment opportunity are fleeting is one of the best arguments for keeping the markets free of excessive trading costs, and for removing the penalties for honest speculation. Speculators keep market prices close to economic values, and this is good, not bad.

Acquira Co.

1995 William N. Goetzmann

Yale School of Management

Belva Broadie works for a boutique investment banking firm, Acquira, Inc. which services a select number of corporate clients. Acquira's specialty is in identifying potential candidates for acquisition by their clients. Belva recently received a call from a client, Hilbert Stacey, chief financial officer of Lehigh Drug company, a $4 billion pharmaceutical manufacturing firm located in Pennsylvania. Stacey would like Acquira to identify two potential candidates for merger with Lehigh. Lehigh is an all-equity firm with no debt on the balance sheet, and wishe to expand within the health care industry through merger. They wanted to look at firms in the medical supplies and equipment industry, as well as firms in the health insurance industry.

Belva's preliminary scan of candidates yielded one firm in each industry that she thought would interest Stacey. The first, DB Supplies, a $1 billion company, manufactures a variety of devices used by hospitals. Her researched showed that DB has a price/earnings ratio of 20 and a debt to market value of equity of 20% . The consensus among analysts following the firm is that the growth rate in future cash flows should be 15%. The second firm, Consolidated Healthco, is also a $1 billion company. Its P/E ratio is 30, and Debt to Equity ratio is 100%. The consensus among analysts following the firm was that the growth rate in future cash flows should be 20%.

In order to prepare a more thorough analysis of the acquisition candidate, Belva collected information about the long-term pattern of returns of all three firms, as well as information about the market. The quarterly time series of total returns for the last few years for each firm plus the S&P 500 are shown in Table I.

Assignment: This assignment is due on October 17 at 5:00 PM. There are no exceptions to the deadline, since answer sheets will be immediately available outside my office at 223, 52 Hillhouse. The analysis can be done as a group, however each student must write up and submit the results individually. 1) Calculate the historical quarterly mean, and standard deviation for each stock and the market. Convert these numbers to annualized values by multiplying the quarterly mean returns by four, and the quarterly standard deviations by the square root of four.

2) Calculate the correlations among each of the four series, based upon quarterly data.

3) Estimate the beta of each of the three companies with respect to the S&P 500, based upon quarterly data.

4) Assume that the current riskless rate is 5.5%, and that the equity risk premium is 8%. Calculate the expected return for each of the three companies, based upon the Capital Asset Pricing Model.

5) Compare the actual returns over the period to the expected returns based upon the CAPM. Assume the t-bill rate was unchanged over the period. The difference between the actual return and the CAPM expected return is called Jensen's alpha. Are the alphas for each firm positive or negative? Is this a violation of the CAPM? Why or why not?

6) Assume there are no taxes. Estimate the asset beta of each of the three firms 7) Assume that Lehigh merges with DB.

What would be the combined standard deviation of the new firm? What would be the beta of the new firm? What would be the new expected return of the new firm?

What is the expected Sharpe ratio of the new firm?

8) Assume Lehigh merges with Consolidated. What would be the combined standard deviation of the new firm? What would be the beta of the new firm? What would be the new expected return of the new firm? What is the expected Sharpe ratio of the new firm?

9) Assume that a single-stage discounted cash flow model under certainty fairly represents the relationship between current price, earnings, discount rates and growth for each firm. That is: using analysts' forecasts of the growth rate in earnings, gi, compared the discount rates implied by the P/E ratios for DB and Consolidated to those implied by the CAPM. Does either firm plot above or below the security market line?

10) What inferences do you draw from your analysis. In particular, Which merger increases the probability of the new firm achieving a return in excess of treasury bills? Is either firm a clear bargain? What recommendation will you make to Stacey regarding the effect of either acquisition upon Lehigh? Assuming the dividend growth model is approximately correct, and the CAPM discount rate is the appropriate one, what price would you suggest that Stacey pay for DB? What price would you suggest Lehigh pay for pay for Consolidated?

Table 1: Quarterly Total Returns

Quarter 1991.1 -.010 Lehigh .088 DB Consolidated .432 .100 S & P 500

1991.2 1991.3 1991.4 1992.1 1992.2 1992.3 1992.4 1993.1 1993.2 1993.3 1993.4 1994.1 1994.2 1994.3

.002 .013 .033 -.121 .035 -.120 -.074 -.113 -.046 .174 .117 -.167 .000 .288

-.109 .041 -.001 .013 .118 -.009 .009 -.104 .026 .079 -.025 .052 .095 .126

.276 .274 .301 -.081 .203 .134 .237 -.163 .170 .107 .224 -.029 .097 .154

.042 .019 .051 .021 .031 -.006 .055 .010 .026 .051 .036 -.057 .024 .039