The Theory of the Firm

The Theory of the Firm Sanghoon Lee How production and related activities are organized, and the relations between actors involved in these activities, is a subject interest in which crosses several disciplinary boundaries. Students of law, organizational behavior, management, finance, and related fields should find this compilation of economics sources on the organizational problem quite useful to their own endeavors. Economists’ newly honed methods and newly directed lines of inquiry make further contributions to the understanding of these matters. We have seen the increasing attention being paid to questions of economic organization and the nature of firms in economics. The discussion of why there are firms, what determines firm-market boundaries, and why firms are organized as they are has become a topic of broad interest. There has been the growing frequency of references to writings such as Ronald Coase’s 1937 classic and work published in the 1970s by Armen Alchian and Harold Demsetz, by Michael Jensen and William Meckling, and by Oliver Williamson. This trend has had increasing influence, as shown by its treatment in a growing formal analytical literature by such writers as Oliver Hart, Bengt Holmstrom, and Paul Milgrom. In this paper, the above mentioned writers’ theories are examined. Ronald Coase: Pioneering Role Ronald Coase’s 1937 article is called a classic “because it changed the way people think about economic organization.”(Williamson, 1991, p.3) Coase observes that the organization of economic activity in firms and markets needed to be derived whereas orthodoxy took it as a datum (determined largely by technology). What distinguishes the firm, in Coase’s view, is that when a workman moves from department Y to department X, he does not go because of a change in relative prices, but because he is ordered to do so. Outside the firm, price movements direct production, which is coordinated through a series of exchange transactions on the market. Within a firm, however, these market transactions are eliminated and in place of the complicated market structure with exchange transactions is substituted the entrepreneur-coordinator, who directs production. Thus, “the distinguishing mark of the firm is the supersession of the price mechanism.” (Coase, 1937, p.389) Evidently, within markets the price system signals resource allocation, but firms employ a different organizing principle – that of hierarchy – whereupon authority is used to effect resource reallocations. Coase insists that firm and market are alternative modes for organizing the very same transactions. 1 Coase poses both of the key questions that are at the heart of the research agenda on economic organization: “in view of the fact that it is usually argued that co-ordination will be done by the price mechanism, why is such organisation necessary?” and “Why is not all production carried on by one big firm?” Coase answers the first question by asserting that there is a cost of using the price mechanism. “The most obvious cost of “organizing” production through the price mechanism is that of discovering what the relevant prices are.” (Coase, 1937, p.390) It may be costly to determine appropriate transfer prices. Coase considers also the costs of negotiating and concluding a separate contract for each exchange transaction which takes place on a market. Therefore the hierarchy principle of the firm is necessary. Inside the firm, “the factor … agrees to obey the directions of an entrepreneur within certain limits. … Within these limits, he can therefore direct the other factors of production.” (Coase, 1937, p.391, emphasis in original) Then, what is the benefit of using the hierarchy principle, instead of using the price mechanism? According to Coase, It may be desired to make a long-term contract for the supply of some article or service. … Now, owing to the difficulty of forecasting, the longer the period of the contract is for the supply of the commodity or service, the less possible, and indeed, the less desirable it is for the person purchasing to specify what the other contracting party is expected to do. … All that is stated in the contract is the limits to what the persons supplying the commodity or service is expected to do. … When the direction of resources (within the limits of the contract) becomes dependent on the buyer in this way, that relationship which I term a “firm” may be obtained. (Coase, 1937, pp.391-392) Therefore, a firm is likely to emerge when a very short term contract would be unsatisfactory. By forming an organisation and allowing some authority to direct the resources, certain marketing costs are saved. However, Coase considers also the question of why there are any markets at all. He suggests that there are costs to coordinating internally, just as there costs to coordinating through market mechanisms. First, as a firm gets larger, there may be decreasing returns to the entrepreneur function, that is, the costs of organizing additional transactions within the firm may rise. … Secondly, it may be that as the transactions which are organized increase, the entrepreneur fails to place the factors of production in the uses where their value is greatest, that is, fails to make the best use of the factors of production. … Finally, the supply price of one or 2 more of the factors of production may rise, because the “other advantages” of a small firm are greater than those of a large firm. (Coase, 1937, pp.394-395) He concludes that at the margin, the costs of organizing within the firm will be equal either to the costs of organizing in another firm or to the costs involved in leaving the transaction to be organized by the price mechanism. That is to say, “a firm will tend to expand until the costs of organizing an extra transaction within the firm become equal to the costs of carrying out the same transaction by means of an exchange on the open market or the costs of organizing in another firm.” (Coase, 1937, p.395) Armen Alchian and Harold Demsetz: Property Rights Approach Armen Alchian and Harold Demsetz are the founding fathers of the property rights approach. This approach emphasizes the implications of property rights for risk bearing and incentive reasons. 1 Their interest in property rights led them in a natural way to investigate the operation of the firm and other organizations. In the early 1970’s they published a landmark paper on the theory of the firm. This paper emphasizes the way in which the firm can internalize externalities associated with incentive problems. The difficulties of monitoring in the presence of shirking are emphasized, and the article represents the beginning of the modern literature on incentive contracts and moral hazard. Let us examine closely their famous 1972 paper. In the paper, Alchian and Demsetz begin by asserting that there is no difference between the firm and the market. The firm is assumed to have no power of fiat, no authority, no disciplinary action any different from market contracting. Therefore, there is no “relationship between a grocer and his employee different from that between a grocer and his customers.” (Alchian and Demsetz, 1972, p.777) “The employee “orders” the owner of the team to pay him money in the same sense that the employer directs the team member to perform certain acts. The employee can terminate the contract as readily as can the employer, and long-term contracts, therefore, are not an essential attribute of the firm. Nor are “authoritarian,” “dictational,” or “fiat” attributes relevant to the conception of the firm or its efficiency.” (Alchian and Demsetz, 1972, p.783) 1 On this view, Jensen and Meckling (1976) show how costs arise when the management and ownership functions are separated. The reason is that it is impossible at zero cost for “owners” (or principals) to ensure that “managers” (or agents) act in the owners’ interests. These costs, known as agency costs, consist of monitoring expenditures by the principals, bonding payments by the agents, and residual losses. The authors distinguish between pecuniary and nonpecuniary returns and show how managers have incentives to expand larger amounts of resources on non-pecuniary returns when their equity or pecuniary interests in the firm decrease. Organizations characterized by the separation of ownership from control exist because, just as there are costs to equity financing, there are costs to debt financing. 3 Alchian and Demsetz define the ownership of the classical capitalist firm as the “entire bundle of rights: 1) to be a residual claimant; 2) to observe input behavior; 3) to be the central party common to all contracts with inputs; 4) to alter the membership of the team; and 5) to sell these rights” (Alchian and Demsetz, 1972, p.783). They consider the team production, team organization, difficulty in metering outputs, and the problem of shirking and show how team production results in organizations characterized as the classical firm. The economic organization makes better use of their comparative advantages to the extent that it facilitates the payment of rewards in accord with productivity. The problem with team production is that the activities of workers are not perfectly separable. With team production it is difficult, solely by observing total output, to either define or determine each individual’s contribution to this output of the cooperating inputs. Hence, it is not possible to compensate workers based on their marginal inputs. Moreover, workers have an incentive to shirk, or free-ride, on the actions of the other workers by reducing the effort they offer to the team. The reason is that the cost of shirking is not born solely by the individual through a reduction in his compensation. Market contracting for individual contributions is not possible. Market competition, in principle, could monitor some team production. … But completely effective control cannot be expected from individualized market competition for two reasons. First, for this competition to be completely effective, new challengers for team membership must know where, and to what extent, shirking is a serious problem, i.e., know they can increase net output as compared with the inputs they replace. … But, by definition, the detection of shirking by observing team output is costly for team production. Secondly, assume the presence of detection costs, and assume that in order to secure a place on the team a new input owner must accept a smaller share of rewards (or a promise to produce more). Then his incentive to shirk would still be at least as great as the incentives of the inputs replaced, because he still bears less than the entire reduction in team output for which he is responsible. (Alchian and Demsetz, 1972, p.781) The costs of metering or ascertaining the marginal products of the team’s members call for new organizations and procedures. One method of reducing shirking is for someone to specialize as a monitor to check the input performance of team members. The monitor, by virtue of monitoring many inputs, acquires special superior information about their productive talents. But who will monitor the monitor? Another constraint can be imposed on the monitor: give him title to the net earnings of the team, net of payments to other inputs. Specialization in monitoring plus reliance on a residual claimant status will reduce shirking. 4 Therefore an organization in which workers are paid a wage by a employer who oversees or monitors the actions of the other workers overcomes the free-riding problem if the employer has claims on the residuals of production. Alchian (1965, 1969) explains the property rights in some aspects. He defines the system of property rights as “a method of assigning to particular individuals the “authority” to select, for specific goods, any use from a nonprohibited class of uses.”(Alchian, 1965, p.130) According to Alchian (1965), behavior under each organization is different, not because the objectives sought by organizations are different, but, instead, because the costs-rewards system impinging on the members of the organization are different. He compares public ownership with private ownership to provide the theorem: “Under public ownership the costs of any decision or choice are less fully thrust upon the selector than under private property.” (Alchian, 1965, p.146, emphasis in original) If public ownership in some government activity were converted to private property, the method of achieving the government objectives would be changed. Alchian (1969) reviews the separation of ownership and control which Berls and Means (1933) pointed out. “What must be meant by those who speak of separation of control from ownership is a reduced ability of the owners to revoke and reassign delegations of decision-making authority that will affect value. … In this sense, a manager’s deviations from stockholders’ interests are less likely to be “policed.””(Alchian, 1969, p.230) Alchian, however, argues that the fact that the dispersed ownership has increased certainly does not imply that managerial activity in these situations will be less consistent with the shareholders’ interests. He illustrates the theme of separation of ownership and control in large dispersed corporations with public utilities and transportation companies, as well as with unregulated firms. He concluded that “in corporations which are not profit-seeking or are not privately owned … behavior will deviate from that in a profit-seeking corporation.” (Alchian, 1969, p.246) Accordingly, something can be deduced and verified about the relationship between behavior and the types of property rights. Demsetz (1991, 1995) also defines characteristics of firm as specialization, continuity of association, and reliance on direction and proposes the property rights approach. The firm properly viewed is a “nexus” of contracts. Our interest might center on explaining (1) the persistence of certain types of contracts that are found in this nexus, (2) the variation observed in other types of contracts that are “more or less” included in this nexus, and (3) the (horizontal and vertical) scope of activities covered by these contracts. … Past and current interest in the existence, the internal organization, and the vertical and horizontal scope of “the firm” fit comfortably into the three areas of inquiry listed above, and it is in these that I wish to show that information cost has relevance that 5 extends beyond its significance in transaction cost and moral hazard problems. (Demsetz, 1991, pp.169-170) For Demsetz, markets and firms are not substitutes; if there is a relationship between the two, it is a complementary relationship. What is needed is a concept of the firm in which production is exclusively for sale to those who are formally outside the firm. … Markets are conceptualized arrangements for matching bids and asks, for exchanging entitlements, and for revelation of the prices that accomplish these exchanges, but the activities that accomplish these things are housed in firms. These firms may be brokers, members of organized exchanges, contract lawyers, financial news gatherers and promulgators, etc. All are specialized producers of exchange services who may or may not rely heavily on managed coordination in the production of these services. (Demsetz, 1995, p.9) Unlike transaction cost economics, the bottom line of Demsetz’s theory is that firms exist because producing for others, as compared to self-sufficiency, is efficient; this efficiency is due to economies of scale, to specialized activity, and to the prevalence of low, not high, transaction costs. On this view, a firm may exist if the efforts of some are managed by the direction of others, but this interpersonal direction is not necessary and is not the essence of the firm. Demsetz (1995) introduces the importance of information costs in transacting and managing production and exchange. He notes that the relevant question is not whether management cost is more or less than transaction cost, but whether the sum of management and transaction cost incurred through in-house production is more less than the sum of management and transaction cost incurred through purchase across markets. Internal versus external costs of production are also important to consider. Corporate charters, longterm associations, and the conscious direction of resources identify firm-like organization. Demsetz concludes that the vertical boundaries of a firm are determined by the economics of conservation of expenditures on knowledge. Alchian’s work on the theory of the firm continued into the late 1970’s as his work with Robert Crawford and Benjamin Klein studied the way in which specific investment creates hold up problems that can be solved in part by vertical integration among business units. Armen Alchian, Robert Crawford and Benjamin Klein: Hold-Up Problem 6 The 1978 article by Benjamin Klein, Robert Crawford and Armen Alchian describe the effects of post-contractual opportunism on market contracting and organizational form. Real world example of Fisher Body and GM is given in this article. In 1920s Fisher Body made car bodies for GM. GM was upset at the prices Fisher was charging and wanted a plant built adjacent to the GM plants. Fisher refused, GM eventually bought Fisher out. It is Coase’s insight that transaction, coordinating, and contracting costs must be used to explain the degree of vertical integration. Thus if transaction costs are high, the firm will vertically integrate. This article extends this by looking at the possibility of post-contractual opportunistic behavior. They state that postcontractual opportunistic behavior, which is defined as unanticipated non-fulfillment of a contract, is particularly bad where assets are specialized. When long-term investments in assets are specific, the quasirents generated may be appropriated ex post by contracting agents. Quasi-rents are defined as the difference in the value of the asset in its current and next best use. If quasi-rents are significant, then there is an incentive for investing agents to vertically integrate in order to reduce the risks of the appropriation of quasirents by opportunistic individuals. To sum up, “as assets become more specific and more appropriable quasi rents are created (and therefore the possible gains form opportunistic behavior increases), the costs of contracting will generally increase more than the costs of vertical integration.”(Klein, Crawford and Alchian, 1978, p.298) More generally, if an asset has a substantial portion of quasi-rents which are strongly dependent upon some other particular assets, both assets will tend to be owned by one party. In other words, problems of incomplete contracting are often relieved by unified ownership. Klein, Crawford and Alchian (1978, p.326) argue that Coase’s distinction between firm and market is too simplistic. Many long-term contractual relationships (such as franchising) blur the line between the market and the firm. Therefore, they suggest that the economic rationale for different types of particular contractual relationships in particular situations should be examined, and consider the firm as a particular kind or set of interrelated contracts. The conventional sharp distinction between markets and firms may have little general analytical importance. In this sense, the pertinent economic question is “What kinds of contracts are used for what kinds of activities, and why?” In the 1988 paper, Klein reexamines the General Motors-Fisher Body case. He suggests that the primary costs saved by vertical integration are not associated with the incompleteness of formal contracting, but rather the costs associated with contractually induced hold-ups. The analysis indicates that hold-up potentials are created not solely from the existence of firm-specific investments, but also from the existence of the rigidly set long-term contract terms that are used in the presence of specific investments. Vertical integration, by shifting ownership of the firm’s organizational asset, creates a degree of flexibility and avoids this 7 contractually created hold-up potential, thereby resulting in significant transaction cost savings. (Klein, 1988, p.214) To solve the hold-up problem, agents must consider the relative advantages and disadvantages of long-term contracting or vertical integration and will have to decide whether to use a long-term contract or vertical integration to solve the hold-up problem. Whether agents adopt vertical integration “depends upon the magnitude of these specific investments, combined with the ability to write long-term contracts that flexibly track market conditions without creating an alternative hold-up potential. Since the ability to write and use long-term contracts depends, in part, upon the underlying market uncertainty and on the level of transactor reputations, these factors will also influence the likelihood of vertical integration.”(Klein, 1988, p.223) Klein shows how vertical integration in the General Motors-Fisher Body case mitigates the hold-up problem created by long-term contracting and concludes that “the long-term, fixed price formula, exclusive dealing contract” that Fisher Body and GM has negotiated was not only defective but created an enormous potential for fisher to hold up General Motors. According to Klein (1988, p.223), this analysis of the motivation for vertical integration is consistent with the fundamental point recognized by Coase that a transaction within the firm is something that is inherently different from a transaction in the marketplace. But the view of the firm as a “nexus of contracts” that has developed in reaction to Coase’s fundamental distinction between the firm and the market is incomplete and misleading. The classic transaction cost problem was posed by Ronald Coase in 1937: When do firms produce to their own needs (integrate backward, forward, or laterally) and when do they procure in the market? He argued that transaction cost differences between markets and hierarchies were principally responsible for the decision to use markets for some transactions and hierarchical forms of organization for others. This approach to the economic organization of technologically separable stages of production was successively worked up by Williamson and by Klein, Crawford, and Alchian (1978). Oliver Williamson: Transaction Cost Economics In the traditional economic thought, the firm is characterized as a production function. But “transaction cost economics maintains that the firm is (for many purposes at least) more usefully regarded as a governance structure.” (Williamson, 1985, p.13) Transaction cost economics relies on and develops the propositions below. 8 1. The transaction is the basic unit of analysis. 2. Any problem that can be posed directly or indirectly as a contracting problem is usefully investigated in transaction cost economizing terms. 3. Transaction cost economies are realized by assigning transactions (which differ in their attributes) to governance structures (which are the organizational frameworks within which the integrity of a contractual relation is decided) in a discriminating way. Accordingly: a. The defining attributes of transactions need to be identified. b. The incentive and adaptive attributes of alternative governance structures need to be described. 4. Although marginal analysis is sometimes employed, implementing transaction cost economics mainly involves a comparative institutional assessment of discrete institutional alternatives – of which classical market contracting is located at one extreme; centralized, hierarchical organization is located at the other; and mixed modes of firm and market organization are located in between. 5. Any attempt to deal seriously with the study of economic organization must come to terms with the combined ramifications of bounded rationality and opportunism in conjunction with a condition of asset specificity. (Williamson, 1985, pp.41-42, emphasis in original) Williamson (1991) deals with efforts to operationalize the Coasian insight that transaction cost differences are mainly responsible for market, hierarchy, and hybrid modes of organization. The operationalization of transaction cost economics entails three key moves: the basic unit of analysis needs to be declared, the rudimentary attributes of human nature need to be identified, and the intertemporal process transformations of organization need to be described. Williamson’s response to the first of these is to make the transaction the basic unit of analysis, whereupon attention is directed to the dimensionalization of transactions. The principal dimensions on which transaction cost economics presently relies for purposes of describing transactions are “the frequency with which transactions recur, the uncertainty to which transactions are subject, and the degree of asset specificity on which they rely.” (Williamson, 1993, p.16) Although all are important, many of the refutable implications of transaction cost economics turn presently on this last. Transaction cost economics employs two critical behavioral assumptions. “The first is a cognitive assumption: human agents are assumed to be “intendedly rational, but only limitedly so” … which condition is commonly referred to as bounded rationality. … The second behavioral assumption is that human agents 9 are given to opportunism.” (Williamson, 1991, p.92) “The main ramification of bounded rationality, for purposes of studying economic organisation, is that all complex contracts are unavoidably incomplete. … Opportunism refers to self-interestedness with guile, whereupon economic agents will sometimes say one thing and do another.” (Williamson, 1993, p.11) Moreover, differences between societies (over time and over space) in socialization affect both bounded rationality and opportunism. “Embeddedness attributes of at least five kinds have a bearing on the latter: culture, politics, regulation, professionalisation, and networks.” (Williamson, 1993, p.15) Transaction cost economics appeals for predictive purposes. A predictive theory of economic organization will align transactions (which differ in their attributes) with governance structures (which differ in their costs and competencies) in a discriminating (mainly, transaction cost-economizing) way. The first application of this approach is that of vertical integration. The study of labor organization thus turns out to have numerous parallels with, rather than be sharply different from the study of intermediate product markets. “Addressing the economics of organization in transaction cost terms discloses that hierarchy also serves efficiency purposes and furthermore permits a variety of predictive statements regarding the organization of work to be advanced.” (Williamson, 1985, p.13) A recent application of transaction cost economics involves corporate finance. The object is to align investment projects (which differ in their attributes) with financial instruments (where debt and equity are viewed as alternative governance structures). Williamson (1988) examines corporate finance through the lens of transaction cost economics and proposes a combined treatment of corporate finance and corporate governance. The conventional view is that the cost of capital is independent of the choice of financial instruments (the theorem of Modigliani and Miller). Transaction cost economics maintains that the asset attributes of investment projects differ and that efficiency purposes are served by aligning projects with the governance structure competencies of debt and equity in a discriminating way. Debt and equity are treated not mainly as alternative financial instruments, but rather as alternative governance structures. Transaction cost approach “maintains that some projects are easy to finance by debt and ought to be financed by debt. These are projects for which physical asset specificity is low to moderate. As asset specificity become great, however, the preemptive claims of the bondholders against the investment afford limited protection – because the assets in question have limited redeployability. Not only does the cost of debt financing therefore increase, but the benefits of closer oversight also grow. The upshot is that equity finance, which affords more intrusive oversight and involvement through the board of directors (and, in publicly held firms, permits share ownership to be concentrated), is the preferred financial instrument for projects where asset specificity is great.” (Williamson, 1988, p.589) That is, transaction cost reasoning supports the use of debt (rules) to finance redeployable assts, while non-redeployable assets are financed by equity (discretion). 10 Although transaction cost approach is underdeveloped in process respects, process arguments nevertheless play a prominent role. The Fundamental Transformation “is one illustration of the proposition that process matters. Briefly, the argument is that it does not suffice to demonstrate that a condition of large numbers competition obtains at the outset. It is also necessary to examine whether this continues or if, by reason of transaction specific investments and incomplete contracting, a condition of bilateral trading evolves thereafter.” (Williamson, 1991, p.94, emphasis in original) Transaction cost economics maintains that “a condition of large numbers bidding at the outset does not necessarily imply that a large numbers bidding condition will obtain thereafter. … Accordingly, what was a large numbers bidding condition at the outset is effectively transformed into one of bilateral supply thereafter.” (Williamson, 1991, p.99, emphasis in original) Victor Goldberg: A Relational View of Contract Goldberg (1976) presents a conceptual framework offering a different perspective on regulation and institutional choice. The emphasis is on aspects of contractual complexity typically glossed over in economic analyses of regulation. The paradigmatic contract of economic theory is a discrete transaction conveying a well-defined object in exchange for cash. “Economic theory generally ignores the complexity of contractual arrangement by implicitly assuming that most exchange takes place in the discrete transaction form.” (Goldberg, 1976, p.445) This discrete transactional mold is apt to be inappropriate for representing relations which are to take place over a long period of time and in which the parties will have to deal with each other regularly over a wide range of issues. While the parties might want to go into considerable detail at the formation stage concerning the rights and obligations of each party given various contingencies, it will often prove too costly to specify the precise terms of the contract and it will be desirable instead to use rough formulae or mutual agreement to adjust the contract to current situations. As the relational aspects of the contract become more significant, emphasis will shift from a detailed specification of the terms of the agreement to a more general statement of the process of adjusting the terms of the agreement over time – the establishment, in effect, of a “constitution” governing the ongoing relationship. (Goldberg, 1976, p.428) Thus, “transactions, however well defined they might appear, are nested in a complex shifting pattern of contractual jurisdictions which, taken together, establish the rights and obligations of the 11 respective parties and the roles of the agents.” (Goldberg, 1976, p.429) In effect, Goldberg (1976) characterizes natural monopoly industries not by their alleged decreasing average costs, but by the features which make long-term relationships between consumers and producers desirable and which further make it extremely difficult to determine at the outset the specific terms of that relationship. The longer the anticipated relation and the more complexity and uncertainty entailed in the relation, the less significance will be placed on the price and quantity aspects at the formation stage. “The emphasis will instead be on establishing rules to govern the relationship: rules determining the appropriate length of the relationship: rules determining the process of adjustment to unexpected factors that arise in the course of the relationship: and rules concerning the termination of that relationship.”(Goldberg, 1976, p.432) A second problem concerns the reliance of individuals on agents (for gathering information, making decisions, negotiating contracts, adjusting the terms of ongoing relationships, and so on). It is convenient to distinguish the discrete transaction of traditional theory from contracts which exhibit one or the other (or both) of these elements – an ongoing relationship and agency. Such contracts will be called “administered contracts.” The administered contracts approach provides a different perspective for examining regulatory institutions. Regulation can be viewed as an implicit administered contract in which both elements are significant. The justification of regulation is seen to rest not on narrow natural monopoly grounds; rather it rests on the long-term relational matters. Victor Goldberg and John Erickson’s detailed study of petroleum coke contracts (1987) describes the variety and complexity of long-term agreements. They examine the price and quantity provisions of contracts for petroleum coke between eight oil refineries and the Great Lakes Carbon Corporation. Their analysis begins by tracing the duration and minimum quantity requirements of petroleum coke contracts to (i) the site specificity of production and processing facilities and (ii) the need, owing to petroleum coke’s bulky and hazardous nature, for its timely removal from the refinery. They then turn to price adjustment provisions, which they interpret as operating principally to mitigate relational frictions rather than to attenuate moral hazard: since quantity restrictions left the parties little discretion over quantity, Goldberg and Erickson (1987) reject incentive alignment as the motive for price adjustment and attribute the inclusion of price adjustment provisions instead to the goal of reducing precontract search and postagreement jockeying over the terms of the trade. An aspect of contracting stressed by Goldberg and Erickson is the interaction among contract terms that makes interpretation of particular provisions dependent on the rest of the agreement – detail that is often accessible only through intensive case analysis. Goldberg and Erickson show that the contracts have transaction cost minimizing features that guard parties from the hazards of transaction-specific investments. The contracts are long-term because of the transaction-specific relationship between the sellers and buyers of petroleum coke. Moreover, the contracts 12 frequently use non-linear prices as a means of protecting the seller’s reliance on the buyer to purchase given quantities of output. The contracts support the “relational” view of contracting. Long-term contracts enable economic actors to coordinate behavior, while stopping short of full, formal vertical integration. These contracts reflect the problems posed by the human characteristics of the trasactors (their limited ability to process information, their propensity for strategic or opportunistic behavior); by the physical characteristics of the subject matter of the transaction (that is, problems with defining and monitoring quality, the production technology, the ease of access to alternative buyers or sellers, the relationspecific nature of the physical plant, and so forth), and by the reliability of external enforcement (legal enforcement or good will). These contracts embody the particular solutions developed by the contracting parties. Our working assumption here is that the solutions are correct – that is, that the contracts are “efficient” adaptations. (Goldberg and Erickson, 1987, p.370) Goldberg and Erickson (1987) explain the structure of the petroleum coke contracts without invoking power, anticompetitive behavior, or risk preferences. Instead, they focus on long-term reliance, short-term coordination, the costs of acquiring information, and other “relational” concepts. Sanford Grossman and Oliver Hart: Ownership Approach In the 1986 paper, Sanford Grossman and Oliver Hart emphasize that contractual rights can be of two types: specific rights and residual rights. When it is costly to list all specific rights over assets in the contract, it may be optimal to let one party purchase all residual rights. Ownership is the purchase of these residual rights. Imagine that firm 1 is an electricity generating plant that is located next to a coal mine in order to use the mine’s coal to make electricity. Let Фi(q1, q2) represent the quality of the coal delivered. Suppose that the boiler firm 1 installs to burn coal does not function well if the coal supplied is impure. Ex ante there may be many potential impurities, and it may be impossible to allow for each of these in the contract. Ex post, however, it may be clear what the relevant impurity is – high ash content, say. Our supposition is that, if firm 1 owns firm 2, it can, ex post, exercise its rights of control over firm 2’s assets to direct that 13 the coal should be taken from a deposit with low ash content (i.e., firm 1 chooses a subvector of q2). In contrast, if firm 2 owns firm 1, it can exercise its rights of control over firm 1’s assets to direct that the boiler should be modified to accept coal with high ash content. An alternative to ownership in this example is a contract that gives firm 1, say, the specific right to direct the areas of the mine in which coal is dug out. This would clearly be reasonable for any one particular right of control. However, we have in mind a situation in which there are many aspects of a firm’s operations, each of which may be important in a different contingency, and thus the costs of assigning specific rights of control ex ante are much higher than the costs of assigning generalized control. (Grossman & Hart, 1986, p.699-670) When residual rights are purchased by one party, they are lost by a second party, and this inevitably creates distortions. That is, integration shifts the incentives for opportunistic and distortionary behavior, but it does not remove these incentives. Firm 1 purchases firm 2 when firm 1’s control increases the productivity of its management more than the loss of control decrease the productivity of firm 2’s management. Grossman and Hart argue that there is often a low-cost alternative to contracts that allocate all specific rights of control. In particular, when it is too costly for one party to specify a long list of the particular rights it desires over another party’s assets, then it may be optimal for the first party to purchase all rights except those specifically mentioned in the contracts. Vertical integration is the purchase of the assets of a supplier (or of a purchaser) for the purpose of acquiring the residual rights of control. Grossman and Hart’s model emphasizes the distortions, due to contractual incompleteness, that can prevent a party from getting the ex post return required to compensate for his ex ante investment. To the extent that the marginal and average values of investment move together, the allocation of ownership rights, by changing the average investment return, will affect the level of investment. We have seen that, if firm i owns firm j, firm i will use its residual rights of control to obtain a large share of the ex post surplus, and this will cause firm i to overinvest and firm j to underinvest. Under nonintegration, on the other hand, the ex post surplus will be divided more evenly, and so each firm will invest to a moderate extent. Integration is therefore optimal / when one firm’s investment decision is particularly important relative to the other firm’s, whereas nonintegration is desirable when both investment decisions are “somewhat” important. (Grossman & Hart, 1986, pp.716-717) 14 Grossman and Hart’s 1986 paper is the first fully formal effort to model incomplete contracting. A central result of the paper is that ownership matters. “Incompleteness of contracts opens the door to a theory of ownership.” (Hart, 1988, p.141) “The impossibility of ex ante bargaining over all aspects of the product to be delivered, that is, the incompleteness of the contract, is the source of our conclusion than the distribution of property rights has efficiency consequences.” (Grossman & Hart, 1986, p.718) The theory focuses attention entirely on the ownership of physical assets and uses the concept of residual rights of control to explain both the costs and the benefits of vertical integration. Grossman and Hart (1986) embraces the transaction cost economics argument that all complex contracts are unavoidably incomplete, as a consequence of which it is impossible to concentrate all of the relevant contracting action in the ex ante incentive alignment. The basic setup entails incomplete contracts (by reason of bounded rationality) and contractual hazard (which is due to asset specificity and opportunism). The trade off, however, with which Grossman and Hart are concerned arises entirely because of ex ante investment distortions (which vary with ownership). Grossman and Hart develop a model in which both ex ante alignment and ex post adaptation differences between market and hierarchical modes of organization are recognized. “Through their influence on the distribution of ex post surplus, ownership rights will affect ex ante investment decisions.” (Grossman & Hart, 1986, p.696) Oliver Hart’s 1988 paper illustrate the way residual rights of control can explain asset ownership with a few examples in which vertical integration overcomes contractual difficulties when investments in assets are relationship-specific. In those cases, “it is desirable for incentive reasons for some part of the overall return stream to be borne by one party. This can be achieved by transferring one asset’s returns to that party. … We saw, however, that if the transfer is attempted without a / corresponding change in ownership or control rights, it will not be fully effective: some of the returns will be diverted by the owner, and the incentive effect will be diminished. Thus to resolve incentive problems, it is necessary not only to assign the various parts of the return scheme to the difference managers efficiently, but also to allocate ownership and control rights to support this assignment.” (Hart, 1988, pp.149-150) The idea that ownership is linked with residual rights of control forms the basis of a theory of integration developed in Grossman and Hart (1986). Hart (1988) also argues that in a world of incomplete contracts there is an optimal allocation of residual rights of control; to the extent that ownership goes together with residual rights of control, there is therefore an optimal allocation of asset ownership. Hart’s emphasis on how integration changes control over physical assets is seen to be in contrast to Coase’s 1937 paper which focuses on the way integration changes an ordinary contractual relationship into one where an employee accepts the authority of an employer (within limits). However, Hart points out that these approaches are not contradictory. “Authority and residual rights of control are very close and there is no reason why our analysis of the costs and benefits of allocating residual rights of control could not be extended to cover human, as well as physical, asset. In fact, 15 residual rights of control over employees and over physical assets are likely to be related.” (Hart, 1988, p.151) In the 1995 book, Hart proposes two ideas: “The first is that contracts are incomplete. The second is that, because of this, the ex post allocation of power (or control) matters.” (Hart, 1995, p.3) These two ideas – contractual incompleteness and power – can be used to understand a number of economic institutions and arrangements. In an incomplete contracting world, because of the difficulty of specifying quality and quantity very far in advance, both parties recognize that any long-term contract is incomplete and subject to renegotiation. Even if renegotiation proceeds smoothly – that is, if problems due to haggling and asymmetric information do not arise – the division of the gains from trade will depend on the ex post bargaining strengths of the parties rather than on what is specified in the initial contract or on economic efficiency. Why does ownership of physical or nonhuman assets matter? The answer is that ownership is a source of power when contracts are incomplete. To understand this, note that an incomplete contract will have gaps, missing provisions, or ambiguities, and so situations will occur in which some aspects of the uses of nonhuman assets are not specified. (Hart, 1995, p.29) Given that a contract will not specify all aspects of asset usage in every contingency, who has the right to decide about missing usages? According to the property rights approach, it is the owner of the asset in question who has this right. That is, the owner of an asset has ‘residual control rights’ over that asset: the right to decide all usages of the asset in any way not inconsistent with a prior contract, custom, or law. Hart focuses on how ownership affects the incentives of top managers. However, workers’ incentives will also be affected by changes in ownership and this can influence the costs and benefits of integration. “The main lesson to be drawn form the example is that changes in control affect the incentives of those who do not have control rights either before or after the change. … Thus, the example also confirms the thrust of Proposition 2(D), that highly complementary assets should be owned together. In fact, the example again strengthens Proposition 2(D), in the sense that it shows that an additional benefit of concentrating ownership of highly complementary assets is that worker incentives are improved. … In other words, the example shows how control over nonhuman assets can lead to authority over human assets, in that employees will tend to carry out actions or investments that are of interest to their boss.” (Hart, 1995, pp.60-61) Hart (1995) develops a theory of the firm based on the idea that firm boundaries are chosen to allocate power optimally among the various parties to a transaction. One implication of the theory is that a merger between firms with highly complementary assets is value-enhancing, and a merger between firms with independent assets is value-reducing. The reason is as follows. If two highly complementary firms have 16 different owners, then neither owner has real power since neither can do anything without the other. It is then better to give all the power to one of the owners through a merger. On the other hand, if two firms with independent assets merge, then the acquiring firm’s owner gains little useful power, since the acquired firm’s owner loses useful power, since she no longer has authority over the assets she works with. In this case, it is better to divide the power between the owners by keeping the firms separate. Previous chapters have shown that an agent should own an asset if he has important assetspecific investment decisions to make or if he has human capital essential for the asset’s use. This chapter considers what happens if the agent is wealth-constrained and cannot buy the asset outright. Under these conditions, the agent will have to raise funds form an outside investor (or form a set of investors) to purchase the asset. This creates a new agency problem. The agent may behave opportunistically, thus depriving the investor of an adequate return on her investment. This chapter studies how the investor can protect herself against such opportunistic behaviour. … It will be shown that in some cases it is better for the investor to enter into a financial contract with the agent of the following form. The agent borrows money from the investor and promises to make certain repayments. If he makes them, he retains control of the asset. If he does not make them, control shifts to the investor. In other words, the theory of incomplete contracts and property rights – when extended to the case of limited wealth – can explain the use of debt financing. (Hart, 1995, p.95) Hart (1995) also analyses the financial structure of a closely held company and a public company. It is argued that debt can play an important role in constraining managers or owners of both kinds of company. He shows that the view of debt as a constraining mechanism can explain the types of debt a company issues (how senior the debt is, whether it can be postponed) and that the possibility of takeovers can explain why many companies bundle votes and dividend claims together – that is, why they adopt a one share-one vote rule. “In the Hart-Moore model … debt forced an entrepreneur to pay out funds to investors rather than to himself. In the various models … debt forced the sale of assets and limited a manager’s ability to make unprofitable, but power-enhancing, investments.” (Hart, 1995, p.156) Bengt Holmstrom and Paul Milgrom: Multi-task Agency Theory The 1991 article by Bengt Holmstorm and Paul Milgrom is concerned with measurement problems that arise in conjunction with the employment relation. 17 In the standard economic treatment of the principal-agent problem, compensation systems serve the dual function of allocating risks and rewarding productive work. It remains a puzzle for this theory that employment contracts so often specify fixed wages and more generally that incentives within firms appear to be so muted, especially compared to those of the market. Also, the models have remained too intractable to effectively address broader organizational issues such as asset ownership, job design, and allocation of authority. Contrary to the usual principal-agent setup, in which the task is strictly one-dimensional, Holmstrom and Milgrom examine multidimensionality of two kinds: agents may be asked to perform several tasks (for example, teaching and research) or to perform a single task that has multiple attributes (for example, quantity and quality). The distinguishing mark of our model is that the principal either has several different tasks for the agent or agents to perform, or the agent’s single task has several dimensions to it. … Multidimensional tasks are ubiquitous in the world of business. As simple examples, production workers may be responsible for producing a high volume of good quality output, or they may be required both to produce output and to care for the machines they use. … In general, when there are multiple tasks, incentive pay serves not only to allocate risks and to motivate hard work, it also serves to direct the allocation of the agents’ attention among their various duties. This represents the firs fundamental difference between the multidimensional theory and the more common one-dimensional principalagent models. (Holmstrom and Milgrom, 1991, p.25, emphasis in original) Differential ease of measurement is characteristic of the multitask problem and more complicated tradeoffs arise than appear in the standard principal-agent setup. “In general, in multitask principal-agent problems, job design is an important instrument for the control of incentives. In the standard model, when each agent can engage in only one task, the grouping of tasks into jobs is not a relevant issue.” (Holmstrom and Milgrom, 1991, pp.25-6, emphasis in original) Given a highly incomplete set of performance measures and a highly complex set of potential responses from the agent, how can the agent be motivated to act in the social interest? Holmstrom and Milgrom’s approach emphasizes that incentive problems must be analyzed in totality; one cannot make correct inferences about the proper incentives for an activity by studying the attributes of that activity alone. In this approach, it is important that “how cross-sectional variations in the parameters that determine the optimal design of jobs, the optimal intensity of incentives, and the optimal allocation of ownership lead to covariations among endogenous variables that are similar to the patterns we find in actual firms.”(Holmstrom and Milgrom, 1991, p.50) 18 In their 1994 article, Holmstrom and Milgrom explore the twin hypotheses (i) that highperformance incentives, worker ownership of assets, and worker freedom from direct controls are complementary instruments for motivating workers, and (ii) that such instruments can be expected to covary positively in cross-sectional data. The intuitive idea is that increasing the incentive for just one task could cause a worker to devote too much effort to that one task while neglecting other aspects of the job, and that increasing incentives for all of the agent’s activities avoids that cost. Asset ownership, contingent rewards, and job restrictions, all influence different dimensions of the worker’s task portfolio. If, as suggested, there is a desire to keep the various incentives in balance, then one would expect that in an optimal system, the three instruments would have to be similarly balanced. Weak incentives for maintaining asset values should go with weak incentives for narrowly measured performance and significant restrictions on worker freedom (excluding an activity is the same as setting its incentive to zero). These, of course, are the attributes that typically characterize employment. (Holmstrom and Milgrom, 1994, p.973) Holmstrom and Milgrom (1994) try to explain why the attributes of an employment relationship differ in so many ways from the attributes of a contractor relationship. “The fact that employment and contracting are multifaceted relationships, each characterized by its own distinct set of attributes, ranks as one of the most significant regularities to be explained by a theory of the firm.”(Holmstrom and Milgrom, 1994, p.988) They treat the attributes of a contracting relationship just the same as if they were inputs to production. It then explores how changes in a parameter affecting one attribute affect choices of other attributes. In their application, an increase in the cost of measuring sales performance acts like an increased input price, leading to the substitution of salary for commissions and to a complementary increase in limitations on outside activities and the frequency of job rotation. They also relate the conclusions to empirical evidence, particularly that on the organization, compensation, and management of sales forces. 19 References Alchian, Armen A., 1965, “Some Economics of Property Rights,” Il Politico, 30, no. 4, reprinted in Alchian, 1977, pp.127-149. Alchian, Armen A., 1969, “Corporate Management and Property Rights,” Economic Policy and the Regulation of corporate Securities, ed. by Henry Manne (Washington: American Enterprise Institute), reprinted in Alchian, 1977. Alchian, Armen A., 1977, Economic Forces at Work, Liberty Press. Alchian, Armen A. and Harold Demsetz, 1972, “Production, Information Costs, and Economic Organization,” American Economic Review, 62, no.5, 777-95, reprinted in Mark Casson (1996). Berls, A. and Means, G., 1933, The Modern Corporation and Private Property, New York: Macmillan. Coase, R. H., 1937, “The Nature of the Firm,” Economica, New Series, IV, November, pp.386-405, reprinted in Mark Casson, 1996. Demsetz, Harold, 1991, “The Theory of the Firm Revisited,” in Williamson and Winter (eds.), 1991, pp.159178. Demsetz, Harold, 1995, The Economics of the Business Firm, Cambridge University Press. Goldberg, Victor P., 1976, “Regulation and Administered contracts,” Bell Journal of Economics, Vol. 7, Issue 2, pp.426-448, reprinted in Williamson and Masten(eds.), 1999. Goldberg, Victor P. and John R. Erickson, 1987, “Quantity and Price Adjustment in Long-Term Contracts: A case Study of Petroleum Coke,” Journal of Law & Economics, Vol. 30, Issue 2, pp.369-98, reprinted in Williamson and Masten(eds.), 1999. Grossman, Sanford J. and Oliver D. Hart, 1986, ‘The Costs and Benefits of Ownership: A Theory of Vertical and Lateral Integration’, Journal of Political Economy, 94 (4), August, 691-719, reprinted in Mark Casson (1996). Hart, Oliver D., 1988, “Incomplete Contracts and the Theory of the Firm”, Journal of Law, Economics and Organization, Vol. 4, Issue 1, pp.119-39, reprinted in Williamson and Winter (eds.), 1991. Hart, Oliver D., 1995, Firms, Contracts, and Financial Structure, Clarendon Press. Holmstrom, Bengt and Milgrom, Paul, 1991, “Multitask Principal-Agent Analyses: Incentive Contracts, Asset Ownership, and Job Design,” Journal of Law, Economics and Organization, Vol.7, Special Issue, pp.24-52, reprinted in Williamson and Masten (eds.), 1999. Holmstrom, Bengt and Milgrom, Paul, 1994, “The Firm as an Incentive System,” American Economic Review, Vol.84, No.4, pp.972-991. Jensen, Michael C. and Meckling, William H., 1976, “Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure,” Journal of Financial Economics, Vol.3, No.4, pp.305-360. 20 Klein, Benjamin, 1988, “Vertical Integration as Organizational Ownership: The Fisher Body—General Motors Relationship Revisited,” Journal of Law, Economics and Organization, Vol.4, Issue 1, reprinted in Williamson and Winter (eds.), 1991, pp.213-226. Klein, Benjamin, Robert Crawford and Armen Alchian, 1978, “Vertical Integration, Appropriable Rents, and the Competitive Contracting Process,” Journal of Law and Economics, Vol.21, Issue 2, pp.297-326, reprinted in Williamson and Masten(eds.), 1999. Mark Casson (ed.), 1996, The Theory of the Firm, Edward Elgar Publishing. Williamson, Oliver E., 1985, The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting, The Free Press. Williamson, Oliver E., 1988, “Corporate Finance and Corporate Governance,” Journal of Finance, Vol.43, No.3, pp.567-591. Williamson, Oliver E., 1991, “The Logic of Economic Organization,” in Williamson and Winter (eds.), 1991, pp.90-116. Williamson, Oliver E., 1993, “The Economic Analysis of Institutions and Organizations – in general and with respect to country studies,” Working Paper, Organisation for Economic Co-operation and Development. Williamson, Oliver E. and Scott E. Masten (eds.), 1999, The Economics of Transaction Costs, Edward Elgar Publishing. Williamson, Oliver E. and Sidney G. Winter (eds.), 1991, The Nature of the Firm- Origins, Evolution, and Development, Oxford University Press. 21