ECON 696: Managerial Economics and Strategy Lecture Notes 3: The Vertical Boundaries of the Firm In the previous

chapter, we explored the idea that firms expand horizontally, either increasing their production of one good or moving into production of other goods, and through this achieve some reduction in their average cost per item through better use of fixed inputs. In this chapter, we'll look at reasons why firms may expand vertically, upstream and downstream, and choose to produce not only their original product but also the inputs and the final product. The decision is interesting because it means a firm will stray from the usually desirable path of specialization. If a firm does choose to expand vertically, it should have a very good reason for doing so. As with economies of scope, this material does not lend itself particularly well to graphical presentation, so we will proceed with a narrative explanation. Obtaining Intermediate Inputs: The Make or Buy Decision Firms manufacture output using some intermediate inputs. These intermediate inputs may be iron ore or auto parts, unexposed film or printed photographs, sand or computer components. In each case, the firm must decide whether to make its own intermediate inputs (from still more primitive inputs) or to purchase them from other companies. There are plenty of examples given in the textbook and you should be sure you understand each of them. You should note that labor is an intermediate input into just about every production process and is different from other inputs in an important way. There are reasons why firms might want to make some inputs and buy others, and these reasons at least partially define what would be best in each situation. One very important consideration is the structure of market for the input in question. If there are a lot of firms supplying the input or if there is free entry into the input market, then it is likely that firms are producing at something near minimum average cost and the price is very near that low average cost. On the other hand, if there are a small number of firms supplying the input or if there are significant barriers to entry in the market then the price charged may be well above the average or marginal cost. Make or buy decisions should not be made without consideration of the structure of the input market.

Reasons to Buy 1. The firm might not use sufficiently large quantities to take advantage of economies of scale. This is why firms typically buy things they use only a little of, such as office supplies or coffee. It helps if markets for these inputs are fairly competitive. 2. Learning to produce the good might be so costly that the firm is willing to pay a price well above the average cost in order to avoid the cost of learning. This might be the case when the input is patented or when there is a difficult learning process involved. Boeing assembles aircraft using purchased jet engines because learning to make reliable jet engines would be unreasonably costly. 3. Special skills may be required, but the firm does not demand them in sufficient quantity to maintain a department of people with those skills. This is similar to the first reason in the list. The book gives a good example of this in discussing why British Petroleum (BP) chooses to purchase insurance against small losses. Insurance companies employ people who are experts at risk reduction, investigation of suspicious claims and litigation. It would be very costly for BP to achieve the high level of expertise it basically purchases when it pays another company for insurance. This is also why small companies hire lawyers from outside rather than maintaining in-house counsel. 4. By purchasing goods, the firm avoids monitoring (or agency) costs associated with production. That is, you don't have to hire managers and supervisors to monitor people producing that particular input. Even if the firm did have the capacity to match the efficiency of industry leaders in production of the input, it would still have the problem of carefully managing a new area of business with no immediate pressure from competitors to keep production efficient. Over a longer period of time, if they failed to keep up with leading firms in the production of that input, they might face the painful task of outsourcing and shutting down that division. Reasons not to Buy 1. Independent suppliers may not be able or may not choose to achieve the required level of timing and coordination. They may not match design specifications closely enough or they may not deliver inputs at the appropriate time, a particular problem for firms trying to practice just in time production. 2. Independent suppliers may not be as trustworthy with trade secrets as internal suppliers might be. A company protects non-patented information by not letting anyone outside of the company have access to it. Depending on the manufacturing process and the specific part involved, it may be the case that a supplier would need to know some secret information which the company would prefer to keep secret. 3. Independent suppliers might compromise on quality if the contract allows it in some way. No contract can specify every possible characteristic of a product (although some government contracts try) and there is always the danger that a supplier could take advantage of an unspecified attribute to cut her costs, supplying an unacceptable product

which, nevertheless, conforms to the specifications of the contract. The costs of writing very complete contracts and enforcing those contracts can be prohibitively high. These issues fall under the heading of transactions costs and will be covered in the next chapter. Fallacious Reasons Not to Buy 1. Producing a product yourself protects you from increases in the market price. While it is true that prices fluctuate from time to time, even in competitive markets, producing an input and using it internally disguises rather than lowers the cost of using that input when its market price is very high. If a firm produces its own electricity, for example, and the price of electricity rises dramatically (as happened in December of 2000 in the western U.S.) the opportunity cost to a firm of using its own electricity is the price for which it might have been sold. Internal analyses stating that the cost of using that electricity was the generating cost would simply be incorrect. As mentioned in the text, there are financial instruments that allow firms to protect themselves from fluctuations in input prices and these are almost certainly better approaches to the problem. 2. By producing an input yourself, you keep for yourself that profit that you supplier would have earned. In fact, if you are purchasing inputs in a market that is fairly competitive, firms are probably only earning a fair rate of return on their capital. The best you could really hope to do by producing the input yourself is to achieve the same fair rate of return on your investment, and that would happen only if you could achieve a level of efficiency equal to that of established and time-tested firms in the industry. On the other hand, if you are purchasing inputs from a firm with some significant monopoly power that is charging a price much higher than their average cost, there is probably some barrier to entry that would prevent you from producing the input. Transactions Costs and the Hold-up Problem There are two important concepts discussed in this section. The first is transactions costs, or the costs of making and enforcing a contract with a supplier. These costs can be significant, and a lot of effort goes into reducing them, not only on the part of individual businesses, but also by people who write contract law. When transactions costs are high, it can become almost impossible to conduct business. The second is the hold-up problem. A hold-up problem occurs when a seller has only one potential buyer or a buyer has only one potential seller. In each case, one party may be at an extreme disadvantage, leaving the other with the potential to extract all the benefit from their trade. Economic agents generally try to avoid these situations. The relationship between transactions costs and the hold-up problem is that potential hold-up problems increase the transactions costs involved in a relationship between a

supplier and a customer. If there is the potential to be held up, very complex contracts may be written in an attempt to avoid such an outcome. Easy Buying: Spot Markets If a firm needs to purchase a very standard, readily available good, it may simply choose to purchase it in a spot market from a seller which will deliver the good or service more or less immediately in exchange for immediate payment. If the good is widely used (such as office supplies, gasoline or building materials) then there is no need to do anything more than go to one of many sellers and make the purchase at the prevailing market price. Buying standard products in a spot market is no problem. Problems arise when the desired input is not an item typically produced. It may be something which no other company uses (a special part), a service provided in a remote location (helicopter transport to an oil field in Alaska) or something to be provided in unusually large quantities (labor for a large factory). Some examples are: a brake which has to be specially manufactured for a new model of automobile an operating system written for a particular type of computer processor an aluminum smelter buying an extremely large amount of electricity In each of these cases, the firm will have to find a supplier willing to undertake costly provision of these inputs. Because the product is not something which the supplier usually produces, the description of the item or service, a schedule for its delivery and the specifics of payment all have to be spelled out in a contract. Contracts and Their Characteristics Any trade that is not an immediate exchange of goods or services for money is governed by some sort of contract, either explicit or implied. The contract describes to a lesser or greater extent what is to be delivered and how the supplier is to be compensated. Contracts have a number of problems. Contracts are costly to write. The more complicated the product described, the arrangements established or the contingencies anticipated the greater will be the transactions costs involved, and these may be sufficiently large to make the transaction unprofitable. Contracts may be costly to enforce. If there is some dispute between the parties, they will engage in negotiation and, sometimes, litigation, both costly propositions. In addition, one party may have to find evidence that the other side violated the contract, another potential cost.

Contracts cannot specify every aspect of the item or service to be delivered, so they are incomplete. That is, there will be some characteristics of the product that will not be specifically described in the contract. These unspecified attributes may be exploited by the supplier in an attempt to increase her profits. A disk drive manufacturer may supply a computer assembler with a certain number of drives with the specified capacity and reliability, but which operate more slowly than the assembler would have liked. If the speed of the drives is not specified in the contract, the drives would meet the specifications of the contract and be less costly for the drive manufacturer to supply. Contracts cannot anticipate every possible state of the world. Things may happen which were totally unexpected and not discussed in the contract. If a power company agrees to supply a computer server farm with electricity at a certain price, but then finds that the price of electricity on the markets rises far above the contract price, it may find it financially impossible to supply the server farm with the quantity of power it demands. If the contract does not specify what is to be done in this situation, each side faces some risk. Incentives in Contracts: When Should Payment Be Tied to Outcomes? One important thing to consider in a contract is that it establishes a relationship which will generate some profit and then specifies how that profit will be divided. One of the parties in a contract will have more opportunity to influence how profitable the relationship is, and that party's payment should be most closely tied to the total profit generated. For example, employees whose performance is critical to a company's profitability should receive a greater portion of their compensation in the form of incentives related to profits while employees whose performance is less critical may receive a larger portion of their pay as a straight salary, perhaps with a small incentive package. As another example, consider an arrangement between an artist and an art dealer. The artist will put some amount of effort into producing each work and the quality of the work and its selling price will be related to this amount of effort. Similarly, the art dealer will put some amount of effort into selling that work of art, and if the dealer puts in a higher level of effort, the work is likely to sell more quickly and for a higher price. If it is believed that the quality of the art produced is more important in driving sales then the artist's compensation should be closely tied to the selling price and the dealer should receive more of a flat fee. If it is believed that the effort put forth by the dealer is more important than the actual quality of the work, then the artist should receive more of a flat fee for each piece produced and the dealer should be compensated based on the selling price. Relationship Specific Capital and the Holdup Problem Contracts are made when it is impossible (or very costly) to purchase an input in a spot market. This usually means that the supplier is doing something special for its customer,

perhaps involving some sort of specialized investment. Similarly, the firm purchasing the input is likely to have made some specialized investment of its own in preparation to use the input. Depending on the situation, the value of these investments may be zero if the transaction described in the contract falls through. Once a firm has made a specialized investment, it is potentially exposed to the holdup problem. This problem occurs when, seeing the other's investment completed, one side demands (usually with an excuse) re-negotiation of the contract to give them more favorable terms. The side which has already made the investment may face the choice of submitting to the re-negotiation or finding itself with millions of dollars worth of useless equipment. There are numerous examples of these types of relationships. One of the most extreme examples of relationship specific capital can occur in the oil industry. Consider that for oil to get to market, there may need to be a working oil field, a pipeline to transport the crude oil and a refinery to process it into useable products. Without the other two, each of these investments is of little or no value. Now, consider that three different companies (a drilling, a piping and a refining company) each agrees to build their part of the system in exchange for a percentage of the final revenues. If two of the three made the investments before the third, the third company would be in a position to hold up the other two. That is, they could refuse to make the investment and build their part of the system unless the other two allowed them to have a larger part of the revenues. There are a number of coal-burning power plants located at the mouths of coal mines. Rather than transport the coal to a plant many miles away, the plants burn the coal right at the mine mouth and simply transport the electricity. The problem is that if a electrical builds the plant at the mine mouth, the mining company may choose to hold them up for a larger share of the revenue from the final electricity sales before developing the mine. Without an operating adjacent coal mine, the power plant would be of little value. Alternately, if the mine started operations in anticipation of the power plant being built, the power company could then hold up the miners. Auto companies redesign their cars every year, and occasionally new parts are needed. An auto company may finish a design and then go to outside suppliers to make brakes for their new car. The brakes would be specific to that car, and the cars could not be sold without the brakes. A brake manufacturing firm would be contracted to supply brakes for the new cars, the auto maker would then set up a factory to produce the cars and/or the brake manufacturer would set up their machinery to produce the specified brakes. In this case, once one of the companies has set up for production, it is exposed to the hold up problem. If the brake manufacturer sets up and produces several thousand sets of brakes before the auto maker has its production line established, the auto maker could demand that the contract be renegotiated with a lower price. If the auto maker sets up the production line and starts making automobiles before the brake manufacturer has started making brakes, the brake manufacturer could demand a re-negotiation giving them a

higher price for the brakes. In each case, if one side refuses to renegotiate, they may be stuck with millions of dollars worth of capital investments and inventory which are, for the moment, worthless. To put a different sort of a spin on this type of example, consider that firms that train their workers give them skills which they may then use to earn higher wages elsewhere. Newly trained employees may hold up their current employer, demanding more pay to keep them from taking their newly acquired skills elsewhere. To prevent this, employers often train employees only in firm specific skills rather than skills they may employ elsewhere. Solutions to Contracting Problems Despite the previous discussion about problems with contracts and relationship specific capital, these are used very successfully by thousands of firms. Only a handful of contracts wind up going so wrong as to be written up as cases in business school publications. What makes contracts work so well in spite of their many potential problems? First, most business relationships are ongoing, or at least hold the possibility of being ongoing if the current transaction is satisfactory. A supplier may have the option to save a little money by skimping on an unspecified attribute, but it knows that to do so could eliminate the possibility of any future orders. A construction company working even just once for a business will have its reputation to worry about. Cutting corners while still meeting the terms of the contract may damage that reputation. Any time a firm attaches some value to future transactions or to its reputation and knows that a trading partner can damage that reputation, the problems described in this chapter are unlikely to occur. However, if you are dealing with a firm you will never contact again and you have no power over their reputation, you should be very cautious in writing the contract and monitoring their behavior. Second, standardized contracts and a stable legal structure make litigation outcomes relatively easy to predict. When contracts go wrong in the U.S., attorneys usually have a good idea how any lawsuits are likely to turn out and pursue the much less costly avenue of negotiation and settlement. In new areas of business or in countries with newly formed or rapidly evolving legal systems, this reliability may be absent and the expected cost of making a contract will be higher. If you are operating in a relatively unstable legal environment, any contracts made should be undertaken with great care. It may be the case that firms in countries characterized by an unstable legal system (especially Russia and eastern Europe) either vertically integrate or engage only in lines of business where they can purchase all necessary inputs in spot markets. Third, many firms will have arrangements with a number of suppliers for any one input, leaving them less exposed to any potential holdup problems. Knowing that a firm has other suppliers to which it can turn, any supplier will be unlikely to try to holdup the firm and will make every attempt to deliver the input under the conditions and for the price

specified in the contract. A firm with only one supplier of a critical input leaves itself dangerously exposed to that supplier's whims. Finally, many of the problems can be solved by avoiding contracts entirely and simply expanding the firm vertically to include all activities involving relationship-specific capital.

Summary This chapter has been a first look at the decision of a firm to vertically integrate, the socalled "make or buy" decision. Firms use inputs to create their outputs. They may find it advantageous to buy these inputs if: they will use relatively small quantities of the input the input is particularly difficult to produce there is some special expertise associated with the input which the firm lacks making the input would involve potentially large monitoring or agency costs Firms my find it advantageous to make input if: there are potential coordination problems with suppliers buying inputs would involve sharing trade secrets which the firm wants to protect the input is so complex that it can't be sufficiently well described in a contract There are some incorrect arguments about why inputs should be made rather than bought. It is fallacious to say that producing an input yourself protects you from price fluctuations you necessarily avoid paying for a supplying firm's profits by producing an input By purchasing inputs in markets, firms avoid the problems of vertical expansion, most notably getting into activities in which they have no efficiency advantage. Contracts allow firms to specialize in the activity at which they are most efficient, very often despite incomplete contracts, relationship specific capital and various other potential hazards. While these hazards sometimes manifest themselves, given a predictable legal environment and the potential for long-term relationships these arrangements usually prove to be mutually beneficial. Under less stable or less predictable conditions more caution in contracting is required and vertical integration may be necessary.