Chapter 6. Reasons for Firm Incentives 19 have shown that people will be more cooperative with more equal shares of whatever it is that is being divided (and women are more inclined to favor “equal shares” than men). People are willing to extend favors in cooperative ventures in the knowledge that the favor will be returned. They will work harder when they believe they are not underpaid. People are more likely to cooperate with close family members and friends than far-removed strangers, and they will be less likely to cooperate with others, whether close at hand or far removed, when the cost of cooperating is high. They work harder, in other words, when they believe they are among members of their relevant “in-group.” Even training can be more effective in raising worker productivity when it is provided within in-groups, regardless of whether they come from collectivistic or individualistic societies. Why is it that people are inclined to cooperate more or less naturally? Wilson repeats a favorite example of game theorists to explain why “cooperativeness” might be partially explained as an outcome of natural selection. Consider two people in early times, Trog and Helga, who are subject to attack by sabretooth tigers. The “game” they must play in the woods is a variant of the prisoner’s dilemma game. If they both run, then the tiger will kill and eat the slowest runner. If they both stand their ground -- and cooperate in their struggle – then perhaps they can defeat the tiger. However, each has an incentive to run when the other stands his or her ground, leaving the brave soul who stands firm to be eaten. What do people do? What should they do? Better yet, what do we expect them to do -- eventually? We suspect that different twosomes caught in the woods by sabretooth tigers over the millenniums have tried a number of strategies. However, running is, over the long run, a strategy for possible extinction, given that the tiger can pick off the runners one by one. We should not be surprised that human society has come to be dominated by people who have a “natural” tendency to cooperate or who have found ways to inculcate cooperation in their members. Moreover, parents spend a lot of family resources trying to ensure that children see the benefits of cooperation, and school teachers and coaches reinforce those values with an emphasis on the benefits of sharing and doing what one is supposed to do or has agreed to do vis a vis people beyond the reach of the family. Managers do much the same. Those societies that have found ways of cooperating have prospered and survived. Those that haven’t have languished or retrogressed into economic oblivion, leaving the current generation with a disproportionate representation from groups that have been cooperative. Those who didn’t cooperate long ago when confronted with attacks by sabretooth tigers were eaten; those who did cooperate with greater frequency lived to propagate future generations. What we are saying here is that human society is complex, driven by a variety of forces -- based in both psychology and economics -- that vary in intensity with respect to one another and that are at times conflicting. However, there are evolutionary reasons, if nothing else, to expect that people who cooperate will be disproportionately represented in societies that survive. Organizations can exploit -- and, given the forces of competition, must exploit -- people’s limited but inherent desire or tendency to work together, to be a part of something that Chapter 6. Reasons for Firm Incentives 20 is bigger and better than they are. Organizations should be expected to try to reap the synergetic consequences of their individual and collective efforts. However, if that were the whole story -- if all that mattered were people’s tendencies to cooperate -- then management would hardly be a discipline worthy of much professional reflection. There would be little or no need or role for managers, other than that of cheerleader. The problem is that firms are also beset with the very incentive problems that we have stressed. The evolutionary process is far from perfect. Moreover, as evolutionary biologist Richard Hawkins has argued, we are all beset with “selfish genes” intent on using “survival machines” (living organisms such as human beings) to increase our chances (the genes’ individual chances, not so much the species’ chances) of survival. 26 “Selfish genes” are willing to cooperate, if that’s what is needed (or, rather, is what works); but the fundamental goal is survival. To the extent that Hawkins is right, what he might be saying is, in essence, that we have to work very hard to override basic, self-centered drives at the core of our being. It may well be that two people can work together “naturally,” fully capturing their synergetic potential. The same may be said of groups of three and four people, maybe ten or even thirty. The point that emerges from the “logic of collective action” is that as the group size -- team or firm -- gets progressively larger, the consequences of impaired incentives mount, giving rise to the growing prospects that people will shirk or in other ways take advantage of the fact that they and others cannot properly assess what they contribute to firm output. As we have already studied, economists concerned with the economics of politics have long recognized how the “logic of choice” within groups applies to politics. The infamous “special interest” groups, which are relatively small and have long been the whipping boys of commentators, tend to have political clout that is disproportionate to their numbers. Indeed, special interest groups often get benefits from governments, with the high costs of their programs diffused over a much larger number of a more politically latent group, the general population of voters. Mancur Olson cites farmers for being the classic case of an interest group that constitutes a minor fraction (less than three percent) of the population but that has persuaded Congress to pass a variety of programs over the years that benefit farmers and their families and impose higher prices on consumers and higher taxes on taxpayers. 27 Political economist James Buchanan points out that honor codes, which, when they work, can be valuable to all students, tend to break down as universities grow in size. For that matter, crime, which is a violation of the cooperative tendency of a community, if not a nation, tends to rise disproportionately to the population. Buchanan’s explanation is that the probability 26 Richard Hawkins, The Selfish Gene (New York: Oxford University Press, 1989). 27 Mancur Olson, The Logic of Collective Action: Public Goods and the Theory of Groups (Cambridge, Mass.: Harvard University Press, 1965). Chapter 6. Reasons for Firm Incentives 21 of criminals being detected, arrested, and prosecuted falls with the growth in the populations of cities. 28 James Wilson also stresses that experimental evidence shows that people in small towns are, indeed, more helpful than people in larger cities, and the more densely packed the city population, the less helpful people will be. Presumably, people in smaller cities believe that their assistance is more detectable. People in larger cities are also less inclined to make eye contact with passersby and to walk faster, presumably to reduce their chances of being assaulted by people who are more likely to commit crimes. 29 In his survey of the literature on the contribution of individuals to team output, Gary Miller reports that when people think that their contribution to group goals, for example, pulling on a rope, cannot be measured, then individuals will reduce their effort. 30 When members of a team pulling on a rope were blind folded and then told that others were pulling with them, the individual members exerted 90 percent of their best individual effort when one other person was supposed to be pulling. The effort fell to 85 percent when two to six other players were pulling. The shirking that occurs in large groups is now so well documented that it has a name -- “social loafing.” A central point of this discussion is not that managers can never expect workers to cooperate. We have conceded that they will – but only to a degree, given normal circumstances. However, there are countervailing incentive forces, which, unless attention is given to the details of firm organization, can undercut the power of people’s natural tendencies to cooperate and achieve their synergetic potential. What Firms Should Do An important message of this chapter is that because people can’t have everything they want, they will do what they can to get as much as they can. “Firms” are a means by which people can get “more” of what they want than otherwise. Firms are expensive operations, by their nature. Accordingly, people would not bother organizing themselves into “firms” if there were not gains to be had by doing so. But therein lies a fundamental dilemma for managers, how can managers ensure that the gains that could be had are actually realized and are shared in some mutually agreed upon way by all of the “stakeholders” in the firm? The problem is especially difficult when everyone associated with the firm – owners, managers, line workers, buyers, and suppliers -- probably want to take a greater share of the gains than they are getting and 28 See James M. Buchanan, “Ethical Rules, Expected Values, and Large Numbers,” Ethics, vol. 76 (October 1965), pp. 1-13. From the strictly economic perspective, what is truly amazing in large cities is not how many crimes are committed, but how many people respect the property and human rights of their fellow citizens, in spite of the decreased incentives to do so. 29 Wilson, The Moral Sense, p. 49. 30 Gary J. Miller, Managerial Dilemmas: The Political Economy of Hierarchy, (New York: Cambridge University Press, 1992), chap. 9. Chapter 6. Reasons for Firm Incentives 22 contribute less in the way of work and investment than they are contributing. Managers have to find ways of overcoming the stakeholders’ inclination to “give little but take a lot.” One of the rolls of incentives is to overcome that inclination by tying how much people receive with what they give to the firm. One of the more important lessons business people learn is that efficiencies can be realized from specialization and exchange. Anyone who attempted to produce even a small fraction of what he or she consumed would be a very poor person indeed. You may recall that the late economist Leonard Reed wrote a famous article (included at the end of Chapter 1) in which he pointed out that no one person could make something even as simple as a lead pencil. 31 It takes literally thousands of people specializing in such things as the production of paint, graphite, wood products, metal, machine tools, and transportation to manufacture a pencil and make it conveniently available to consumers. No one knows enough – or can know enough – to do everything required in pencil production. Prosperity depends on our ability to become very efficient in a specialized activity and then to exchange in the market place the value we produce for a wide range of products that have been efficiently produced by other specialists. Our ability to exchange in the market place not only allows us to produce more value through specialization, it also allows us to obtain the greatest return for our specialized effort by imposing the discipline of competition on those from whom we buy. In this chapter, we extend our discussion of how transaction costs in markets can cause firms to extend the scope and scale of their operations. We are concerned with a special form of “opportunistic behavior” relating to the use of specialized plant and equipment that can cause firms to make things themselves even though outside suppliers could produce those things more efficiently. Make or Buy Decisions Much the same advantage from specialization and exchange applies to firms as well as individuals. But that comment begs an important question: Exactly what should firms make inside their organizations and what should they buy from some outside vendor? Business commentators have a habit of coming up with rules that don’t add very much to the answer. For example, one CEO deduced, “You should only do, in-house, what gives you a competitive advantage.” 32 Okay, but why would anyone get a competitive advantage by doing anything inside, given that such a move reduces, to one degree or another, the advantage of buying from 31 See Leonard Reed, “I Pencil,” The Freeman, December 1958: pp. 32-37. 32 Al Dunlap and Bob Andelman, Mean Business: How I Save Bad Companies and Make Good Companies Great (New York: Times Books, 1996), p. 55. Chapter 6. Reasons for Firm Incentives 23 the cheapest outside competitor? Answers have varied over time (although the one we intend to stress relates to incentives). At one time, the answer to the make-or-buy problem would have focused on technological considerations: Firms often produce more than one product because of what economists call “economies of scope,” a situation where the skills developed in the production of one product lower the cost of producing other products. 33 But even firms with diverse product lines are actually quite specialized in that they purchase most of the inputs they use in the market rather than produce them in-house. General Motors, for example, does not produce its own steel, tires, plastic, or carpeting. Instead, it is cheaper for General Motors, and the other automobile manufacturers, to purchase these products from firms that specialize in them and to concentrate on the assembly of automobiles. 34 Neither do restaurants typically, grow their own vegetables, raise their own beef, catch their own fish, or produce their own toothpicks. Given the advantages of specialization in productive activities and buying most of the needed inputs in the market place, a reasonable question is why firms do as much as they do? Why don’t firms buy almost all the inputs they need, as they need them, from others and use them to add value in very specialized ways? Instead of having employees in the typical sense, for example, a firm could hire workers on an hourly or daily basis at a market-determined wage reflecting their alternative value at the time. Instead of owning and maintaining a fleet of trucks, a transport company could rent trucks paying only for the time they are in use. Loading and unloading the trucks could be contracted out to firms that specialize in loading and unloading trucks. The transport firm would specialize in actually transporting products. Similarly, the paper work required for such things as internal control, payroll, and taxes could be contracted out to those who specialize in providing these services. Indeed, taken to the limit there would cease to be firms as we typically think of them. Rather there would be only individual resource owners all operating as independent contractors, with each buying (or renting) everything they need to add value in a very specialized way and then, after the value is added, selling to another individual who adds more value until a good or service is finally sold to the final consumer. This extreme form of specialization and reliance on market exchange is clearly not what we observe in the economy. There are limits to the efficiency to be realized from further 33 For example, a firm that has the equipment necessary to produce one type of electrical appliance may find that this equipment can be fully utilized if also used to produce other types of electrical appliances. 34 Historically, automobile manufacturers did produce quite a lot of their parts in-house for reasons that will be explained later in this chapter. But the trend has been to rely more on outside suppliers, with the lowest cost manufacturers leading this trend. For example, Chrysler, the lowest-cost American producer, was producing only 30 percent of its parts in-house in the mid 1990s, versus 50 percent for Ford (the second lowest-cost American producer) and 70 percent for General Motors. Toyota produces only 25 percent of its parts in-house. See John A. Byrne, “Has Outsourcing Gone Too Far?” Business Week, April 1, 1996, p. 27. Chapter 6. Reasons for Firm Incentives 24 specialization, and as a manager it is useful to understand the cause of these limits and what it implies about the advantages of producing in-house rather than buying in the market. The problem with total reliance on the market should now be familiar: there are significant costs -- transaction costs -- associated with making market exchanges. You have to identify those who are able and willing to enter into a transaction, negotiate the specific terms of the transaction and how those terms might change under changing circumstances, draw up a contract that reflects as accurately as possible the agreed upon terms, arrange to monitor the performance of the other party to make sure the terms of the agreement are kept, and be prepared to resolve conflicts that arise between the agreement and the performance. Because of these transaction costs, it is often better for some individual or some group of individuals to directly manage the use of a variety of resources in a productive enterprise that we call a firm. Transaction costs are lower, for example, when owners of labor become employees of the firm by entering into long-run agreements to perform tasks, that are not always spelled out clearly in advance, under the direction of managers in return for a fixed wage or salary. A market transaction is not needed every time it is desirable to alter what a worker does. Employment contracts typically allow managers wide discretionary authority to re-deploy workers as circumstances change without having to incur further transaction costs. Furthermore, with a uniform employment contract with a large number of workers, a manager can direct productive interactions between these workers that might otherwise require negotiated agreements between each pair of workers. As an example, ten workers could be hired with ten transactions, each negotiated through a relatively simple and uniform employment agreement. If those ten workers were independent contractors who had to interact with each other in ways that employees of a firm often do, they might well have to negotiate the terms of that interaction in 45 separate agreements. 35 In general, the higher the cost of transacting through markets, the more a firm will make for itself with its own employees rather than buy from other firms. The reason restaurants don’t make their own toothpicks is that the cost of transactions is extremely low in the case of toothpicks. It is hard to imagine the transaction costs of acquiring toothpicks ever getting so high that restaurants would make their own. But one might have thought the same about beef until McDonalds opened an outlet in Moscow. Because of the primitive nature of markets in Russia when McDonalds opened its first Moscow outlet (before the collapse of the Soviet Union), relying on outside suppliers for beef of a specified quality was highly risky. Because of the high transaction costs, McDonalds raised it own cattle to supply much of its beef requirements for its Moscow restaurant. 35 In general N people can pair off in [(N-1)xN]/2 different ways. So ten people can pair off in [9x10]/2 = 45 different ways. The difference between the number of people (number of contracts required in an employment relationship) and the number of pairs of people (the number of contracts that could be required otherwise) increases as the number of people increases. For example, with 100 people, the number of possible pairs is 4,950. And the number of separate contracts could be larger than the number of pairs of people if they also grouped into teams with different teams having to negotiate with one another. Chapter 6. Reasons for Firm Incentives 25 Negotiating an agreement between two parties can be costly, but the most costly part of a transaction often involves attempts to avoid opportunistic behavior by the parties after the agreement has been reached. Agreements commonly call for one or both parties to make investments in expensive plant and equipment that are highly specific to a particular productive activity. Once the investment is made, it has little, if any value in alternative activities. Investments in highly specific capital are often very risky, and therefore unattractive, even though the cost of the capital is less than it is worth. The problem is that once someone commits to an investment in specific capital to provide a service to another party, it is very tempting for that other party to take advantage of the investor’s inflexibility by paying less than the original agreement called for. 36 There are so-called “quasi rents” that are appropriable, or that can be taken by another party through unscrupulous, opportunistic dealing. 37 The desire to avoid this risk of opportunistic behavior can be a major factor in a firm’s decision to make rather than buy what it needs. Consider an example of a pipeline to transport natural gas to an electric generating plant. Such a pipeline is very expensive to construct, but assume that it lowers the cost of producing electricity by more than enough to provide an attractive return on the investment. To be more specific, assume that the cost of constructing the pipeline is $1 billion. Assuming an interest rate of 10 percent, the annual capital cost of the pipeline is $100 million. 38 Further assume that the annual cost of maintaining and operating the pipeline is $25 million. Obviously it would not pay investors to build the pipeline for less than a $125 million annual payment, but it would be attractive to build it for any annual payment greater than that. 39 Finally, assume that if the pipeline is constructed it will lower the cost of producing electricity by $150 million dollars a year. The pipeline costs less than it saves and is clearly a good investment for the economy. But would you invest your money to build it? 36 Similarly, a firm that invests in a facility that, because of its location, is dependent on a particular supplier for an important input may find that the supplier demands a higher price than agreed upon after the facility is built. 37 For those knowledgeable in economic jargon, appropriable “quasi rents” are not the same thing as “monopoly rents” (or monopoly profits achieved by charging higher than competitive prices because of barriers to entry). Appropriable quasi rents are the differences between the purchase and subsequent selling price of an asset, when the selling price is lower than the purchase price simply because of the limited resale market for the asset. See Benjamin Klein, Robert Crawford, and Armen Alchian, “Vertical Integration, Appropriable Rents, and the Competitive Contracting Process,” Journal of Law and Economics (October 1978): pp. 297-326. 38 Technically this assumes that the pipeline lasts forever. While this assumption is obviously wrong, it doesn’t alter the cost figure much, if the pipeline lasts a long time. The assumption helps us simplify the example without distorting the main point. 39 The 10 percent interest rate is assumed to be an investor’s opportunity cost of capital investment. So any return greater than 10 percent is sufficient to make an investment attractive. It is assumed that the annual $25 million for maintaining and operating the pipeline includes all opportunity costs (if the payments to compensate the investor for maintenance and operation costs are made as these costs are incurred, then the costs for these items are not affected by the interest rate). Chapter 6. Reasons for Firm Incentives 26 Any price between $125 and $150 million a year would be attractive to both investors in the pipeline and the electric generating plant that would use it. If, for example, the generating plant agrees to pay investors $137.5 million each year to build and operate the pipeline, both parties would realize annual profits of $12.5 million from the project. But the investors would be taking a serious risk because of the lack of flexibility after the pipeline is built. The main problem is that a pipeline is a dedicated investment, meaning there is a big difference in the return needed to make the pipeline worth building and the return needed to make it worth operating after it is built. While it takes at least $125 million per year to motivate building the pipeline, once it has been built it will pay to maintain and operate it for anything more than $25 million. Why? Because that is all it takes to operate the line. The pipeline investment itself is a sunk cost, literally and figuratively, not to be recaptured once it has been made. So after investors have made the commitment to construct the pipeline, the generating plant would be in a position to capture almost the entire value of initial pipeline investment by repudiating the original agreement and offering to pay only slightly more than $25 million per year. 40 Of course, our example is much too extreme. The generating plant is not likely to risk its reputation by blatantly repudiating a contract. And even if it did, the pipeline investors would have legal recourse with a good chance of recovering much, if not all, of their loss. Furthermore, as the example is constructed, the generating plant has more to lose from opportunistic behavior by the pipeline owners than vice versa. If the pipeline refuses service to the plant, the cost of producing electricity increases by $150 million per year. So the pipeline owners could act opportunistically by threatening to cut off the supply of natural gas unless they receive an annual payment of almost $150 million per year. But our main point dare not be overlooked and should be taken seriously by cost minimizing and profit maximizing business people: Anytime a transaction requires a large investment in dedicated capital, there is the potential for costly problems in negotiating and enforcing agreements. True, opportunistic behavior (actions taken as a consequence of an investment that has been made and cannot be recaptured) will seldom be as blatant as in the above example where it is clear that a lower price is a violation of the contract. But in actual contracts involving long-term capital commitments, unforeseen changes in circumstances (higher costs, interrupted supplies, stricter government regulations, etc.) can justify changes in prices, or other terms of the contract. Typically contracts will attempt to anticipate some of these changes and incorporate them into the agreed upon terms, but it is impossible to anticipate and specify 40 Economists refer to this as capturing all the quasi rents from the investment. To elaborate on what we have already said about quasi rents, rent is any amount in excess of what it takes to motivate the supply of a good or service before any investment has been made. In the case of the pipeline, anything in addition to $125 million a year is rent. On the other hand, a quasi rent is any amount in excess of what it takes to motivate the supply of a good or service after the required investment is made. In the pipeline example, anything in excess of $25 million a year is quasi rent. So once the investor has committed to the pipeline, any offer over $25 million a year will motivate the supply of pipeline service and allow the generating plant to capture almost all of the quasi rent. Chapter 6. Reasons for Firm Incentives 27 appropriate responses to all possible changes in relevant conditions. Therefore, there will usually be ambiguities in long-term contractual arrangements that open the door for opportunistic behavior of the type just discussed, and that can be resolved only through protracted and expensive legal action. So committing to investments in dedicated capital carries great risk of opportunistic behavior without some assurance that such behavior will not pay. One way to obtain this assurance is for the investment to be made by the same firm that will be using the output it produces. Alternatively, the firm that makes the investment in the specific capital can merge with the firm that depends on the output from that investment. The early history of the automobile industry provides an example of a merger between two companies that can be explained by the advantages of producing rather than buying when dedicated capital investment is involved. 41 In 1919, General Motors entered into a long-term contract with Fisher Body for the purchase of closed metal car bodies. This contract required that Fisher Body invest in expensive stamping machines and dies specifically designed to produce the bodies demanded by GM. This put Fisher Body in a vulnerable position, given that once the investment was made GM could have threatened to buy from someone else unless Fisher Body reduced prices substantially. This problem was anticipated, which explains why the contract required that GM buy all of the closed metal bodies from Fisher and specified the price as equal to Fisher’s variable cost plus 17.6 percent. However, while these contractual terms protected Fisher against opportunistic behavior on the part of GM, they created an unanticipated opportunity for Fisher to take advantage of GM. The demand for closed metal bodies increased rapidly during the early 1920s (in part because of increased auto sales, but also from a dramatic shift from open wooden bodies to closed metal bodies). The increased production lowered Fisher’s production costs, and indeed made it possible for Fisher to lower its costs significantly more than it did. Evidence suggests that Fisher took advantage of the 17.6 percent “price add-on” by keeping its variable costs (particularly labor costs), and therefore the price charged GM, higher than necessary. General Motors was aware of this “over charge” and requested that Fisher build a new auto body plant next to GM’s assembly plant. This would have eliminated the costs of transporting the auto bodies (a variable cost that came with the 17.6 percent add-on) and reduced GM’s price. Fisher refused to make the move, however, possibly because of concerns that such a dedicated investment to GM requirements would be exploited by GM. As a result of the potential haggling, threats and counter-threats, GM bought Fisher Body in 1926 and the two companies merged. GM could buy Fisher simply because their tenuous dealings, with accompanying transaction costs, were depressing both companies’ market value. GM could pay a premium for Fisher simply because of the anticipated transaction cost savings. 41 The following discussion of the relationship between General Motors and Fisher Body is taken from Klein, Crawford, and Alchian, “Vertical Integration, Appropriable Rents, and the Competitive Contracting Process,” Journal of Law and Economics (October 1978): pp. 308-310. Chapter 6. Reasons for Firm Incentives 28 In an ideal world without transaction costs, General Motors would have bought auto bodies from specialists subject to the constant discipline of market competition. In the real world of transaction costs, GM made the auto bodies itself. The construction of electric generating plants next to coalmines provides another example of the potential benefits to a firm for producing an input rather than buying it when highly specific capital is involved. There is an obvious advantage in “mine-mouth” arrangements from reducing the cost of transporting coal to the generating plant. But if the mine and the generating plant are separately owned, the potential for opportunistic behavior exists after the costly investments are made. The mine owner, for example, could take advantage of the fact that the generating plant is far removed from a rail line connecting it to other coal supplies by increasing the price of coal. To avoid such risks, common ownership of both the mine and the generating plant is much more likely in the case of “mine-mouth” generating plants than in the case of generating plants that can rely on alternative sources of coal. And, when ownership is separate in a “mine-mouth” arrangement, the terms of exchange between the generating plant and mine are typically spelled out in very detailed and long-term contracts that cover a wide range of future contingencies. 42 There are other ways a firm can benefit from the advantages of buying an input rather than producing it while reducing the risks of being “held-up” by a supplier who uses specialized equipment to produce a crucial input. It can make sense for the firm to buy the specialized equipment and then rent it to the supplier. If the supplier attempts to take advantage of the crucial nature of the input, the firm can move the specialized equipment to another supplier rather than be forced to pay a higher than expected price for the input. This is exactly the arrangement that automobile companies have with some of their suppliers. Ford, for example, buys components from many small and specialized companies, but commonly owns the specialized equipment needed and rents it to the contracting firms. 43 Firms are also aware that those who supply them with services are reluctant to commit themselves to costly capital investments that, once made, leave them vulnerable to hold-up (demands that the terms and conditions of the relationship be changed after an investment that cannot be recaptured has been made). In such case the firm that provides the capital equipment 42 For a detailed discussion of the mine-mouth arrangements, see Paul Joskow, “Vertical Integration and Long-Term Contracts: The Case of Coal-Burning Electric Generating Plants,” Journal of Law, Economics, and Organization (Spring 1985): pp. 33-80. 43 The Ford example is discussed on pages 245-46 of Robert Cooter and Thomas Ulen, Law and Economics (Glenview, Illinois: Scott, Foresman and Company, 1988). Also, Alex Taylor III, op cit., discusses this strategy by automobile companies as a way of reducing the number of suppliers they depend on (therefore reducing transaction cost) without increasing their vulnerability to hold-up. On p. 54 he states, “Even now some manufacturers pay for the suppliers’ equipment so if production falters, they can yank out the machinery and install it in someone else’s factory.” These arrangements also have advantages from the small contracting companies’ perspective, since they provide a signal to the auto companies that the contractors will play straight with them. The advantage of a business being able to commit itself to honest dealing is discussed later in the book. . For example, Chrysler, the lowest-cost American producer, was producing only 30 percent of its parts in-house in the mid 1990s, versus 50 percent for Ford. their contribution to group goals, for example, pulling on a rope, cannot be measured, then individuals will reduce their effort. 30 When members of a team