Development of the Theory
After receiving his Ph.D. from Carnegie-Mellon in 1963, Oliver Willamson took a position as assistant professor in the economics department at Berkeley. In his initial work he studied managers in large corporations, and the systematic exercise of their discretionary authority in ways which undermined the efficiency of the firm. The principle findings in this period were that managers, by concentrating on the number of employees reporting to them, favor firm size at the expense of profitability, and that firms with management representation on the board of directors tend to higher retained earnings than those that do not.
Despite notable exceptions (for example, Baumol, 1959), the mainstream of microeconomics research assumed that competitive pressures would eliminate all firms whose managers could not properly exploit the productive capacity under their control. The failure to do so is now called "technical inefficiency," but as late as 1966 Harvey Leibenstein was driven to call it "X-inefficiency," joking that it was a condition unknown to economic science. Perhaps because economic discourse was largely dominated by the struggle between free market and socialist economists, "allocative" inefficiency as the really significant problem: the failure to properly assign productive resources for the manufacture of the mix of goods a society requires. Leibenstein argued, however, that the losses due to allocative inefficiency in non-socialist economies were miniscule compared with those attributable to his "mystery" inefficiency. Economists, by and large, were uninterested in Leibenstein's theory (see Stigler's argument in "The Xistence of X-efficiency" that the theory is confused and unproductive.), though Louis de Alessi later demonstrated (1983) that it is squarely within the now well-established approach to property rights and incentives.
In 1965 Williamson moved to the University of Pennsylvania, and in 1966 began a year as a research economist for the Antitrust Division of the Department of Justice. His work afterward was primarily concerned with regulatory economics: barriers to entry, price regulation, antitrust defenses and enforcement. It was at this time that the transaction cost papers of R. H. Coase (1937, 1960) enter his work, in papers dealing more fundamentally with the nature of the firm; the use of transaction costs to study industrial integration would be the basis of Williamson's research for the next 30 years.
In "A dynamic stochastic theory of managerial behavior," 1967, Williamson criticized the inability of existing theories of the firm, including his own managerial discretion work, to produce testable hypotheses about the behavior of a firm and its managers. "The usual theory of the firm," he wrote, "tends to dismiss, if not ignore, the internal operations of the firm and concerns itself mainly with deducing input-output relations. If the viscera are themselves to become objects of interest, an entirely different orientation is warranted (p. 17)." The heart of the paper was the development of a simple probabilistic model for a firm whose managers preferred inefficient ways of investing money, but had to maintain profitability. With reasonable assumptions, the model showed managers cutting back and increasing efficiency in response to economic shocks. This behavior is certainly well known, and was not surprising even at the time, but Williamson was seeking a "systematic treatment of this condition as it influences conduct and performance in the large corporation (p. 11)."
In another 1967 paper, "Hierarchical control and optimum firm size," Williamson attempts to determine why firms do not grow without bound. While there had certainly been earlier treatments, Williamson argued that they were "too imprecise to permit testable implications to be derived (p. 123)." The paper features a mathematical model for a hierarchy which is headed by an omniscient executive and administered by trustworthy managers; as orders pass through a stage in the hierarchy, however, they suffer a fixed rate of lost compliance due to losses in the accuracy of the representation of the executive's wishes. The assumption of serial loss is inspired by now well-known experiments in which stories are "passed along" and then evaluated for accuracy. For the model, it is assumed that some portion of the workers' labor will be fruitlessly expended as a result of accumulated misdirection. For a given set of parameters (including appropriate spans of control, wage ratios between levels of the hierarchy, etc.), the model allows determination of the optimal number of levels in the hierarchy, trading off expansion of output against the costs of lost control. While the model could be modified to allow for enhanced communication methods, these have costs themselves, and so the same qualitative results will hold: the firm size is limited by the costs of lost control, or the costs of maintaining control. Where managers have their own goals to pursue, the proportion of wasted effort can be expected to increase, in part due to the introduction of systematic bias in the interpretation of ambiguous directives from superiors, and in part to actively opportunistic behavior. In an interesting experiment, Williamson fitted the model against data from General Motors, indicating an assumed 90% compliance rate. It should be noted that this paper preceded by several years those that would launch the economic theory of principals and agents (e.g. Spence and Zeckhauser, 1971; Alchian and Demsetz, 1972; Hurwicz, 1972; Ross, 1973). Williamson concludes that his model supports Robinson's 1934 assertion that problems of coordination ultimately account for the limits of the firm.
The problem of control loss was important to Williamson's antitrust work because of the relationship between technical inefficiency and competition. To an economist who assumed that competition eliminated technical inefficiency, any which was observed could be attributed to monopoly status. Following the organization theorists, however, Williamson accpeted that "virtually all of the interesting bureaucratic behavior observed to exist in large government bureaucracies finds its counterpart in large non-government bureaucracies as well (p. 134)." A model which characterizes the cost of bureaucracy per se is necessary if the true additional costs of monopoly-inspired inefficiency are to be measured.
"Vertical integration: market failure considerations," written in 1971, solidified his approach to the problem of economically desirable mergers. While standard explanations relied on technological considerations like economies of scale or process interdependency, Williamson argued that such approaches were inadequate. After all, he argues, an appropriately written contract should be able to permit separate firms working in common to exploit all of the same economies offered as reasons for integration. The fact that such contracts are not common suggests that some more fundamental reason exists for the creation and merger of firms. The real problem, Williamson claims, is the combination of uncertainty and bounded rationality, our limited ability to encompass all the knowledge relevant to the solution of a problem. If there were no uncertainty, a once-and-for-all contract could be written between, say,an iron smelter and a steel maker, and they could then build appropriate adjacent facilities and engage in efficient production, unencumbered by the need to legally merge the ownership of their firms. If there were no limits to our cognitive powers, the two could write an exhaustive contract which specified the terms of exhange as a function of all possible states of nature. Because there are both, contracts must account for a range of possibilities often too great to specify explicitly. If perfect cooperation could be counted on, the two parties could conduct business smoothly even without a complete contract. Because it cannot, the possibility for a prisoners' dilemma ensues: even if both sides were to benefit from cooperation, distrust might interfere.
Even if once-for-all and long-term contracts are so difficult to emply, there is still the possibility of short-term contracting, which allows for renegotiation in the face of changing circumstances. Of course, it also allows for the costs of negotiation to be incurred frequently; these include the direct cost in labor, but also the opportunity cost of operating sub-optimally until a new agreement is reached. Moreover--and this becomes very important in the later theory--short-term contracts can make suppliers reluctant to invest in relationship-specific assets (like a steel mill next to someone else's blast furnace but without its own), leading to underinvestment. If the exchange requires such investments, they may also act as a barrier to entry and put the two firms in a bilateral monopoly situation--essentially where we were with an incomplete long-term contract.
In short, market arrangements falter when a complete contract cannot be written, and particularly when there is a "small-numbers" situation, i.e. an imperfect market. Firms, on the other hand, offer an opportunity to promote cooperation, in that both entrepreneurs are now rewarded in proportion to the joint profit, rather than those of a single party to the exchange. When the true state of the world is ambigous, separate firms may hold (or at least claim to hold) different opinions; in a single firm, access to the private information of both firms is easier, if not perfect, reducing the likelihood of strategic misrepresentation and improving the chance that the right actions will be taken. The firm also has access to a broader range of control mechanisms than do firms in the marketplace, which Williamson sees as an advantage in coordinating action; while this is likely true, those mechanisms have been rejected in the common law, and an investigation of when stronnger mechanisms are actually beneficial seems warranted. While it may be that the conditions which require integration also require the full control apparatus of hierarchy, this has been treated as a conclusion though introduced as an assumption, even though the analysis has, in recent years, been extended to progressively less hierarchical forms of organization.
In his Strategy and Structure (1962), Alfred Chandler found that diversification, as in product lines or operating regions, had led American firms to break their hierarchies into divisions. In "Managerial discretion, organizational forms, and the multidivisional hypothesis," another of Williamson's 1971 publications, he argued that the success of the multidivisional form followed from its superior control properties. By limiting the size of individual bureaucracies, the opportunity for miscommunication was reduced; by defining the scope of each division's action, the responsibilities of managers can be more clearly defined. With closer communication and less opportunity to pursue personal goals, managers of multidivisional firms ran their shops more efficiently than they would if the same firms were run as monolithic bureaucracies.
"Markets and hierarchies: some elementary considerations," published in 1973, extends the analysis of internal organizations (firms). First comes a brief discussion of peer groups, intended "to avoid imputing benefits to hierarchy that can be had, inns (firms). First comes a brief discussion of peer groups, intended "to avoid imputing benefits to hierarchy that can be had, in some degree, by simple nonhierarchical associations." Recall our discussion of the contractual difficulties facing separate but interdependent iron and steel producers; we noted that uniting ownership in a single firm helped to align the interests of the principals. Peer groups, which are nonhierarchical arrangements for asset and income sharing, do allow for aligning the interests of principals, but do not permit the auditing and control mechanisms of hierarchies with superior-subordinate relations. Because the firm's revenues can often be traced to workgroups but not individual's, unscrupulous or underperforming members of a peer group can benefit at the expense of others; because the group is composed of equals, identifying and penalizing such parties is more difficult than in a hierarchy.
By monitoring workers and assigning reward and promotion in response to observed performance, a firm can relax its screening procedures for new members (since it can let in nearly anyone, and reward them only if they perform) and reduce incentives for "free riding" (since underperformance will be met with underpayment). Those who perform well and in good faith can be induced to enter at a low wage by the promise that they will later be rewarded for their performance, while those who underperform or shirk are less capable of exploiting the firm than they would be in a peer group. This analysis of employment and promotion is strongly reminiscent of the economic approach to insurance sales, where the seller seeks to assess the buyer's true riskiness in the face of measurement difficulties and possible deception ("moral hazard"). The crucial aspect of this analysis is that it is not in general straightforward to relate an individual's actions directly to the income of the firm; if it were, the subjective effort estimations of managers would be unnecessary.
The central question of the paper is: "What organizational relations are to be expected if a set of ... work groups ... is engaged in recurring exchange of a small numbers sort for which successive adaptations are required?" Recalling the difficulties of market contracting in this case, Williamson suggests a consideration of possible ways to collect these workgroups in a single organization. One choice is to choose a single manager who will thereafter manage all of the workgroups; citing bounded rationality and the span of control literature, Williamson dismisses this solution. Another possibility is the use of internal contracting: a new workgroup takes on the administration of the facility, providing inputs and disposing of outputs, while the existing workgroups contract with one another and the "capitalist" or facility owner. This system was in fact in common use in the United States in the late nineteenth century, but suffered from coordination problems (for example, excessive inventories), externalities (for example, equipment was overutilized by contractors who did not own it), and the bargaining difficulties inherent in bilateral monopoly situations. Complex (non-flat) hierarchies, Williamson argued, developed to address these problems between workgroups, just as simple hierarchies addressed them within workgroups.
All of the principle elements of the theory had been laid out by this point, and it remained only to bring them together systematically. Markets and Hierarchies was the first coherent statement of what Williamson initially called "the markets and hierarchies approach," but has since been accepted as the modern statement of Coase's transaction cost economics (except, perhaps, by Coase, who admits its impact without accepting its conclusions). The individual economic exchange, or transaction, is taken as the basic unit of analysis; the methodology is to consider different institutions for administering exchange relations, or governance structures, and determining what characteristics make transactions better suited for one kind of governance structure than another. Because all governance structures have limitations and costs, the strategy is comparative: given a choice between structures, which is better. In his 1975 book, Williamson limits his attention to choosing between free markets and hierarchical firms, leaving for later intermediate forms of organization like joint ventures and strategic alliances. Intermediate structures are messier and harder to understand than markets and hierarchies (even at the level of simply defining them), and it is often wise to solve a simple problem before moving on to a harder one. To the extent that other governance structures truly are intermediate, an understanding of their comparative strengths is likely to benefit from the analysis of the extreme forms. Several authors have attempted to demonstrate the existence of non-intermediate "third forms," involving governments ( ) or EDI networks ( ), for example, but none have really influenced the central literature.
Economists had already considered the problem of "market failure", the case where free markets do not operate smoothly, and we have discussed Williamson's approach to the problem (uncertainty and bounded rationality are necessary to interfere with long-term contracting, and opportunism coupled with a lack of alternate exchange partners interfere with short-term contracting. This is essentially the statement of Coase's conclusions: hierarchies economize on transaction cosrarchies economize on transaction costs. Observed trends in integration behavior suggest conditions of market failure. Williamson's theory considers the symmetric problem of "organizational failure," the case where hierarchies do not operate smoothly. Trends toward outsourcing, divestment, or decentralization suggest conditions of organizational failure. It is important to realize that failure is relative, in that a given governance structure will only be abandoned if another will perform better. In a 1989 book review ("Markets or governments"), Williamson approvingly cites Coase (1964): "the main question ... is how alternative arrangements will work out in practice ... Until we realize that we are choosing between social arrangements that are all more or less failures, we are not likely to make much headway."
As we mentioned briefly in our discussion of contracting difficulties, Williamson emphasizes the importance of relationship-specific investments as a cause of market failure. An asset is relationship-specific to the extent that its value diminishes if the relationship ends. Customized equipment is an example, but so too are immovable investments in a partner's plant, as in the iron-steel case discussed above. Because custom or specific assets can be far more efficient than generic ones, it is often economically desirable to employ them; to the extent that their use introduces significant transaction costs, however, they may require integrated ownership or prove undesirable. The claim that asset specificity induces vertical integration is one of the most important refutable hypotheses offered by the theory, and the evidence for and against it is being gathered and debated even now. Coase, by the way, claims to have abandoned the notion as intuitively problematic in 1932, though he has not produced a refutation.
The most important development of the next ten years was the exploration of the contract law approach to organization. Classical contract law, "sharp in by clear agreement, sharp out by clear performance," is suited to perfect markets. Neoclassical contract law, which relieves parties from strict literal interpretation, is better suited to conditions of uncertainty. Under forbearance law, however, the courts refuse to consider disputes internal to the firm (with exceptions for civil rights, etc.), removing exchange from contract law considerations ("Comparative economic organization...", 1991). Intermediate between neoclassical contracting and integration is "relational contracting," typically unenforceable in the courts and dependent on arbitration and enforcement by private parties. Use of the contracting approach allows some consideration of intermediate governance structures, and an analysis of private ordering seems vital to the understanding of strategic alliances and "virtual" corporations..
Glossary of Transaction Cost Economics
Asset Specificity: An asset is specific to a given exchange relation (or transaction) to the extent that it cannot be redeployed for use in another context without appreciable loss in productive value. Asset specificity has come to be the central concept in Williamson's transaction cost theory: it is claimed that transaction-specific assets introduce contractual difficulties which are the principle reason for replacing market contracts with employment contracts. As a result, the basic explanation for the existence of hierarchy is the presence of a significant economic advantage to the use of specific, rather than generic, assets. Williamson distinguishes four majon distinguishes four major types of asset specificity: site specificity, physical asset specificity, human asset specificity, and asset dedication (see below, under asset specificity, site, etc.). Parties to a contract may be reluctant to invest in economically desirable specific assets, due to concern about recovering the costs of the nonredeployable asset, whether because of demand uncertainty or contract termination; by removing contractual concerns, integration can realize production efficiency gains. If a contracter does make significant transaction-specific investments, such that it is a highly efficient exchange partner, they may act as a barrier to entry by raising the entry costs of potential rivals (they must match the specific investment, or operate at tighter margins). As a result, asset specificity can turn a competitive market into a small numbers exchange situation, leading to potential market failure; Williamson calls this process the fundamental transformation. Asset specificity only leads to these results under the theory's behavioral assumptions and in the face of uncertainty.
Asset specificity, dedication: Dedicated assets are those otherwise ordinary to the contracter's business, but acquired for a specific relation. An example is capacity expansion for the sake of a particular buyer. This kind of "weak" specificity does not typically lead to integration, though risk-sharing contractual arrangements may arise.
Asset specificity, human: Human asset specificity arises whenever there is significant learning-by-doing on the job: workers from outside the organization are not good substitutes for those inside the firm. It is generally equivalent to a worker's idiosyncratic knowledge of a transaction. To the extent that the firm needs continuing access to this knowledge, and not just a distinct product of its application (like a patent or an expert system), employment contracts are likely to supplant market (consulting) contracts. It is important to note that knowledge of the corporation as a culture, which can facilitate communication and efficient exchange, may produce human asset specificity; transactions which benefit greatly from this effect are likely to undergo the fundamental transformation and experience market failure.
Asset specificity, physical: Physical asset specificity obtains when an asset is less valuable when used in any other transaction than in the one it was intended for. If the asset is mobile and redeployable, contractual difficulties can be avoided by quasi-integration, in which one firm owns the asset used by another, and can redeploy it to a new partner. This is generally equivalent to nonfungibility.
Asset specificity, site: Site specificity obtains when an asset is useful only in its current location, as when it is immobile or dependent on another specific asset. An example would be a brake-parts factory located on the grounds of an auto manufacturer, or a coil-rolling facility attached to a steel plant. The bilateral monopoly relationship which ensues (an extreme case of small-numbers exchange) generally leads to integration.
Behavioral assumptions: Williamson's theory assumes that agents act with bounded rationality, and that they are in general self-interest seeking. Moreover, at least some agents practice opportunism, acting with dishonesty and guile to the detriment of others, and detecting these agents is sufficiently difficult that every contractual relationship requires safeguards.
Bounded rationality: Boundedly rational behavior is intended to be rational, in the sense of calculatedly maximizing personal utility, but is constrained by limited cognitive or computational abilities .
Capital market, internal: A smoothly functioning capital market is vital for commerce, and there may be circumstances in which an internal capital market can operate more efficienctly than the external one. For example, a capitalist inside the firm's general office may be better able to understand the business of a workgroup proposing a project, or to assess the true risk level and reward potential, than an outside investor safeguarding against opportunism. Although it is conceivably feasible to sell equity positions in individual corporate projects, financial practice suggests that there are transaction cost reasoe are transaction cost reasons not to. Investors either buy into the entire firm (and hence its entire portfolio of projects) or demand that the project be spun off into a distinct business. The internal capital market, however, is able to consider individual projects suggested by workgroups which have been integrated, presumably for some other purpose. See conglomeration and corporation, M-form.
Conglomeration: Growth of a firm, by entry, merger, or acquisition, into an unrelated area of business, is called conglomeration, and the resulting firm a conglomerate. Williamson attributes the growth of conglomerates in the 1970's to regulatory pressure: firms were afraid to integrate for fear of antitrust prosecution. The M-form corporation (see Corporation, M-form) provided a governance structure which could be used for the effective management of diverse but commonly owned enterprises. While there are no technological reasons (economies of scale and scope) for joint ownership of unrelated businesses, there may be transaction cost benefits, like the creation of an efficient capital market, internal to the firm, perhaps exploiting information which an outside investor, safeguarding against opportunism, cannot accept at face value.
Contracts: Contracts embody the expectations and conditions of an exchange or exchange relationship, and range from the automatic and impersonal ("placing an order") to the formal expression of a strategic alliance. Classical, neoclassical, and relational contracting are governance structures for handling progressively greater interdependence of the parties under progressively more complex and uncertain circumstances.
Contracts, classical: Classical contracts are described as "sharp in by clear agreement; sharp out by clear performance." They assume that the identity of the parties involved is irrelevant; essentially, the object of exchange is treated as a standard commodity. There is an emphasis on the letter of the agreement, and remedies are clearly and narrowly defined. The result is that the consequences of fulfillment or of breach are well understood from the beginning. Classical contracts are associated with transactions in perfect markets (where identity is insignificant) or in the absence of uncertainty or bounded rationality (where complete specification of alternatives is possible).
Contracts, neoclassical: In the presence of uncertainty and bounded rationality, it can be impossible to write contracts which describe all possible contingencies. Rather than abandon exchange altogether, or internalize the transaction, the practice of writing incomplete contracts has developed. Such contracts are generally of long term, and include specification of mechanisms for amending the agreement as conditions change, including the involvement of third-party arbitration. An important difference between private arbitration and court litigation is the intention to preserve the exchange relationship: arbitrators are working for a solution that is best for the relationship, while jurists are working for a just allocation of the rights and responsibilities remaining in a professional relationship unlikely to survive litigation. Arbitration offers opportunity for more interactive or cooperative approaches than litigation, and abitrators can be chosen who have a better business understanding of the transaction in question than could readily be transmitted to a judge. The involvement of a neutral third party is a mechanism which promotes identifying the private information of the parties and acting on it without opportunism.
Contracts, relational: Complex, long-term exchange relations are sometimes impossible without a presumption of fundamentally cooperative intent foreign to the notion of litigation. The governance structures which have come to be called relational contracts typically include extra-legal cultural elements (professional societies, for example) to foster cooperation, and include commitments not enforceable in the courts. Reciprocity and risk-sharing (or risk-matching) as tokens of good faith are common.
Corporation, H- or Holding form: The H-form corporation is a holding company, a legal structure providing for the common ownership of firms, but decentralized management. H-form corporations do not exploit the opportunities for coordination and control offered by common ownership, and can therefore fail to c and can therefore fail to correct for market failure. Chandler (1962) documents the transition of large H-form companies, prominently General Motors, to M-form corporations.
Corporation, M- or Multidivisional form: The multidivisional firm is organized into distinct and separate hierarchies, but is distinguished from H-form corporations by the general office, which practices strategic planning, capital allocation, and control for all divisions. Divisionalization helps to avoid the organizational failure problems of the U-form, while the central planning and control mechanisms compensate for market failure in ways the H-form does not. The active asset allocation function of the general office effectively creates an internal capital market (see capital market, internal).
Corporation, U- or Unitary form: The U-form corporation is constructed as a single, monolithic hierarchy encompassing all the functions of the firm. With growth, the U-form corporation eventually suffers organizational failure. Chandler (1962) documents the transition of the large U-form corporations, prominently du Pont, by M-form corporations.
Dedicated assets: See asset specificity, dedication.
Efficient boundaries: The efficient boundaries problem is the transaction cost formulation of the make-or-buy decision: given a network of related transactions, which of them should be organized together into a firm to minimize production and transaction costs; that is, where should the boundaries of the firm be drawn to maximize efficiency?
Fundamental transformation: The smooth functioning of the market depends on the ability of a buyer to obtain noncollusive bids from qualified suppliers; where this is not the case, market failure is likely. If the winner of a supply contract invests heavily in relationship-specific assets (see asset specificity) which makes the supplier an especially efficient partner for the buyer, it will be at a competitive advantage compared to other potential suppliers. This first-mover advantage can act as a deterrent to other suppliers, and small-numbers exchange can emerge after the fact, even if the initial bidding was highly competitive. Williamson calls this transformation from a competitive to a small numbers situation as a result of specific investment the fundamental transformation.
Governance structure: A governance structure is a set of rules and institutions for adminstering an economic exchange relationship. The free market/common law is an example, as is common ownership with hierarchy. The task of transaction cost economics is to determine which governance structure, among a list of those available, is most efficient (production and transaction cost minimizing) for governing a given transaction.
Incentivity: Williamson uses "incentivity" to describe the effectiveness of the incentives a worker is offered for performing. The free market offers high incentivity, in that an entrepreneur has direct rights (residual claimancy) to the profit streams that arise his actions; in a small-numbers exchange relation, however, entrepreneurs have market power and are tempted to act opportunistically by the high reward rate available to them. The salaries and bonuses associated with hierarchy are less directly linked to any particular act of business acumen, and thus have relatively low incentivity. On the other hand, because a manager will not be able to claim the immediate gains of noncooperation, he is more likely to cooperate. Cost-plus supply contracts, typical of internal supply, also offer low incentivity to managers, who will not be able to claim all the benefits of redusing supply cost;.the result is that they are less likely to achieve minimum cost supply.
Integration: When a firm enters a related business, as a new entrant or by acquiring an existing operation, or when firms in related businesses merge, the act is called integration. Integration with a buyer or supplier is called vertical, integration with a firm at the same stage of production is called horizontal. The same act between unrelated firms is called conglomeration. Williamson suggests that integration helps to coordinate the interests of workers in the integratts of workers in the integrated enterprises, in that rewards now depend on the joint profit, avoiding the prisoners' dilemma complications. Integration in the absence of evident technological efficiencies was traditionally viewed with the suspicion of antisocial intent, and analyzed in an antitrust framework, but Williamson argued that technological factors are not themselves sufficient reason for any act of integration, the real cause being transaction cost reduction. That being the case, "antisocial" integration may serve a genuine efficiency purpose, and the appropriate regulatory action cannot be determined without a consideration of transaction costs.
Internalization: When an exchange relation is removed from the marketplace and subordinated to the administration of the firm, it is said to have been internalized. For the sake of exposition, Williamson generally assumes that all transactions are initially in the market, and so internalization frequently appears synonymous with integration.
Market failure: When an unregulated marketplace is unable to provide goods or services efficiently (with respect to some omplementable alternative), we say market failure has taken place. In the transaction cost literature, small-numbers exchange is generally the cause of market failure. Market failure can lead to the introduction of nonmarket exchange mechanisms like government regulation (see antitrust) or integration/internalization.
Opportunism: A self-interest seeking economic agent is said to be opportunistic if he pursues his goals with calculated dishonesty, or guile.
OrgabOrganizational failure: Although hierarchical organization offers advantages over markets in some situations, it is not universally superior (if it were, central planning would likely have demonstrated more laudable performance). When an existing organization is unable to offer goods and services as efficiently as a market would, we say that organizational failure has taken place. Organizational failure can lead to divestment and "spinning-off."
Organizational failures framework: Williamson considered his emphasis on the symmetry of market failure and organizational failure to be a significant conceptual advance for the analysis of efficient boundaries. When describing his methodology in the early 1970's, he used the unfortunate term "organizational failures framework" to indicate that his approach was based on this symmetry. Hence, chapter four of Markets and Hierarchies, which is about market failure, has the confusing title "The Organizational Failures Framework." The phrase should probably have been "institutional failures framework," but has in any rate been dropped from the literature in favor of "transaction cost economics."
Prisoners' dilemma: The prisoners' dilemma is one of the classic problems in game theory, and is essential to the literature on cooperation and opportunism (see Axelrod, 1984). Two criminals have been charged and arrested, and are being held in separate cells. If both keep quiet, they will receive 1 year sentences on a minor charge; if one of them testifies against the other, he will go free but the other gets 5 years; if, however, both of them testify, they both get 3 years. Assume one criminal believes his partner will testify against him with probability p: then his expected punishment when keeping silent is p*5 + (1-p)*1 = 1 + 4p, while his expected punishment if testifying is p*3 + (1-p)*0 = 3p < 1 + 4p. No matter what p is, then, he will testify. But going through the same calculation causes the partner to testify as well, and they both spend 3 years in jail--even though cooperation would have made them both better off. Williamson doesn't actually talk much about prisoners' dilemmas, but his discussions of negotiation in a small-numbers exchange situation resolve around the same issues: the way the fear of opportunism, even in its absence, can lead parties to a decision to choose a solution which is bad for both of them.
Quasi-integration: Quasi-integration is when a firm buys an asset specific to an exchange and loans it to a contractual partner, taking it away if the partner's contract is not renewed. This practice is very common in the metal-working industry, where custom tools and dies are used by suppliers but owned by buyers. Quasi-integration preserves the viability of the marketplace by avoiding the fundamental transformation.
Self-interest seeking: An economic agent is said to be self-interest seeking if has goals of his own which he actively pursues. When commands or contracts are ambiguous, the agent will naturally exercise his discretion in ways consonant with his self-interests; if the agent is also opportunistic, he may pursue his goals even when they conflict with his promises.
Small-numbers exchange: When there are few parties available to fill one side of an exchange, whether supplier or buyer, Williamson calls the situation small-numbers exchange. The condition ge. The condition thus includes both oligopoly and oligopsony. Small-numbers exchange with self-interest seeking agents can lead to market failure, in ways well-described in the prisoners' dilemma and antitrust literatures, and market failure can lead to integration. Williamson also argues that asset specificity can create a small-numbers exchange situation through what he calls the fundamental transformation, by creating a barrier to entry.
Technological separability: Two processes are said by Williamson to be technologically separable if all coordination issues between them could be solved (perhaps not efficiently) by keeping a suitably large buffer inventory between them. In such cases, each process could feasibly be operated in a separate firm. The efficient boundaries problem is to determine when the integration of two such independent firms is economically desirable. Alchian and Demsetz (1972) held that technological inseparabilities were the cause of hierachical organization, but Williamson argues that only very simple organizations could be explained this way. Alchian (1982) has since agreed with Williamson that asset specificity is the real determinant of hierarchy.
Transaction: A transaction is an economic exchange, the transfer of some good or service between technologically separable activities. The transaction is regarded by transaction cost economics (as by the institutional economics of John R. Commons, 1934) as the fundamental unit of investigation. The rules and institutions governing a particular transaction constitute its governance structure, and the task of transaction cost economics is to determine, for a particuor a particular transaction, the most efficient (production and transaction cost minimizing) governance structure from a list of those available. While the transaction is in progress, there is a relationship between the parties, and the literature sometimes considers the exchange and the exchange relation simultaneously as "the transaction," particularly when discussing recurrent transactions between parties in a bilateral monopoly.
Transaction costs: Transaction costs are the costs of administering an exchange relationship. These include the costs of negotiating, drafting, and monitoring contracts; the costs of settling disputes and enforcing settlements; and the opportunity costs associated with administering a contract inefficiently until a new agreement is recognized as necessary and then reached.
Oliver E. Williamson's Publications
1963. "Managerial discretion and business behavior," American Economic Review 53(5).
1964. The Economics of Discretionary Behavior: Managerial Objectives in a Theory of the Firm, Englewood Cliffs, NJ: Prentice-Hall.
1965. "A dynamic theory of interfirm behavior," Quarterly Journal of Economics 79:580-607.
1965. "Innovation and market structure," Journal of Political Economy 73:67-73.
1966. "Peak load pricing and optimal capacity under indivisibility constraints," American Economic Review 56(4)
1967. "A dynamic stochastic theory of managerial behavior," in Almarin Phillips and Oliver E. Williamson (eds.), Prices: Issues in Theory, Practice, and Public Policy. Philadelphia: University of Pennsylvania Press.
1967. "The economics of defense contracting: incentives and performance," in Roland N. McKean (ed.), Issues in Defense Economics, New York: National Bureau of Economic Research.
1967. "Hierarchical control and optimum firm size," Journal of Political Economy 75:123-138.
1967. With D. G. Olson and A. Ralston. "Externalities, insurance, and disability analysis," Economica N.S. 34:235-253.
1967. With T. J. A. Sargent. "A probabilistic approach," Economics Journal 77(308).
1968. "Wage rates as a barrier to entry: the Pennington case in perspective," Quarterly Journal of Economics 82:85-116.
1968. "Economies as an antitrust defense: the welfare tradeoffs," American Economic Review 58:18-35.
1970. Corporate Control and Business Behavior: An Inquiry into the Effects of Organization Form on Enterprise Behavior, Englewood Cliffs, NJ: Prentice-Hall.
1970. "Administrative decision making and pricing: externality and compensation analysis applied," in Julius Margolis (ed.), The Analysis of Public Output, New York: National Bureau of Economic Research.
1971. "The vertical integration of production: market failure considerations," American Economic Review 61: 112-123.
1971. "Managerial discretion, organization forms and the multi-divisional hypothesis," in Robin Marris and Adrian Wood (eds.), The Corporate Economy. Cambridge, MA: Harvard University Press.
1971. With D.F. Turner. "Market structure in relation to technical and organizational innovation," in J.B. Heath (ed.), International Conference on Monopolies, Mergers and Restrictive Practices. London: HMSO.
1972. "Antitrust enforcement and the modern corporation," in Victor R. Fuchs (ed.), Policy Issues and Research Opportunities in Industrial Organization. New York: National Bureau of Economic Research.
1972. "Dominant firms and the monopoly problem: market failure considerations," Harvard Law Review 85:1512-1531.
1972. With N. Bhargava. "Assessing and classifying the internal structure and control apparatus of the modern corporation," in Keith Cowling (ed.), Market Structure and Corporate Behavior Theory and Empirical Analysis of the Firm. London: Gray-Mills Publishing Ltd.
1973. "Some notes on the economics of atmosphere," Discussion Paper #29, Fels Center of Government, University of Pennsylvania, Philadelphia.
1973. "Markets and hierarchies: some elementary considerations," American Economic Review 63: 316-325.
1973. "Limits of vertical integration and firm size," in Aubrey Silbertson and Francis Seton, Industrial Management: East and West, New York: Praeger.
1974. "The economics of antitrust: transaction cost considerations," University of Pennsylvania Law Review 122: 1439-1496.
1974. "Patent and antitrust law: book review," Yale Law Journal 83:647-671.
1975. With M. L. Wachter and J. E. Harris. "Understanding the employment relquote Understanding the employment relation: the analysis of idiosyncratic exchange," Bell Journal of Economics 6:250-280.
1975. Markets and Hierarchies: Analysis and Antitrust Implications: A Study in the Economics of Internal Organization, New York: Free Press.
1976. "Economics of internal organization: exit and voice in relation to markets and hierarchies," American Economic Review 66:369-377.
1976. "Franchise bidding for natural monopolies--in general and with respect to CATV," Bell Journal of Economics 7:73-104.
1977. "Predatory pricing: a strategic and welfare analysis," Yale Law Journal 87:284-340.
1978. With Michael Wachter. "Obligational markets and the mechanics of inflation," Bell Journal of Economics 9:549-571.
1979. "Transaction-cost economics: the governance of contractual relations," Journal of Law and Economics 22:3-61.
1979. "Review of Bork, The Antitrust Paradox: A Policy at War With Itself," University of Chicago Law Review 46: 526-531.
1979. "Assessing vertical market restrictions," University of Pennsylvania Law Review 127: 953-993.
1979. "On the governance of the modern corporation," Hofstra Law Review 8: 63-78.
1980. "Emergence of the visible hand: implications for industrial organization," in Alfred D. Chandler, Jr. and Herman Daems (eds.), Managerial Hierarchies. Cambridge, MA: Harvard University Press.
1980. "The organization of work," Journal of Economic Behavior and Organization 1:5-38.
1981. "Cost escalation and contracting," Discussion Paper #95, Center for the Study of Organizational Innovation, University of Pennsylvania, Philadelphia.
1981. "The economics of organization: the transaction cost approach," American Journal of Sociology 87:548-577.
1981. "The modern corporation: origins, evolution, attributes," Journal of Economic Literature 19:1537-1568.
1981. With William G. Ouchi. "The markets and hierarchies program of research: origins, implications, prospects," in William Joyce and Andrew van de Ven (eds.), Organizational Design. New York: Wiley.
1982. "Antitrust enforcement: where it has been; where it is going," in John Craven (ed.), Industrial Organization, Antitrust, and Public Policy. Boston: Kluwer-Nijhoff Publishing.
1983. "Organizational innovation: the transaction cost approach," in J. Ronen (ed.), Entrepreneurship. Lexington, MA: Heath Lexington.
1983. "Credible commitments: using hostages to support exchange," American Economic Review 73:519-540.
1983. "Organization form, residual claimants, and corporate control," Journal of Law and Economics 26:351-366.
1984. "The economics of governance: framework and implications," Journal of Theoretical Economics 140:195-223.
1984. "Corporate governance," Yale Law Journal 93.
1984. "Perspectives on the modern corporation," Quarterly Review of Economics and Business 24:64-71.
1985. The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting, New York: Free Press.
1986. With Michael Riordan. "Asset specificity and economic organization," International Journal of Industrial Organization.
1986. Economic Organization: Firms, Markets and Policy Control, New York: New York University Press.
1987. Antitrust Economics: Mergers, Contracting and Strategic Behavior. New York: Blackwell.
1988. "Breach of trust in hostile takeovers: comment," in A. Auerbach (ed.), Corporate Takeovers, Chicago: University of Chicago Press.
1988. "Corporate finance and corporate governance," Journal of Finance July:567-591.
1988. "The economics and sociology of organization: promoting a dialogue," in George Farkas and Paula England (eds.), Industries, Firms, and Jobs. New York: Plenum.
1988. "The logic of economic organization," Journal of Law, Economics, and Organization 4:65-93.
1988. "Technology and transaction cost economics: a reply," Journal of Economic Behavior and Organization 10:355-364.
1989. "The firm, the market,and the law: book review," California Law Review 77:223-231.
1989. "Markets or governments: choosing between imperfect alternatives: book review," Journal of Economic Literature 27:92-94.
1989. "Transaction cost economics," in R. Schmalensee and R. Willig (eds.), Handbook of Industrial Organization, Vol. 1. Amsterdam: North-Holland.
1990. "Political institutions: the neglected side of the story," Journal of Law, Economics, and Organization 6:262-266.
1990. "Chester Barnard and the incipient science of organization," in Oliver E. Williamson (ed.), Organization Theory: From Chester Barnard to the Present and Beyond. New York: Oxford University Press.
1990. With Dennis W. Carlton. "Vertical integration and market foreclosure: comments and discussion," Brookings Papers on Economic Activity Annual, p.277(10).
1991. "Strategizing, economizing, and economic organization," Strategic Management Journal 12:75-94.
1991. "Comparative economic organization: the analysis of discrete structural alternatives," Administrative Science Quarterly 36:269-296.
1991. "Markets, hierarchies, and the modern corporation: an unfolding perspective," Journal of Economic Behavior and Organization 17:335-352.
1991. "Economic institutions: spontaneous and intentional governance," Journal of Law, Economics, and Organization.
1993. "Calculativeness, trust, and economic organization," Journal of Law, Economics, and Organization 36:453-486.
1993. "Contested exchange versus the governance of contractual relations," Journal of Economic Perspectives 7:103(6).
References
Alchian, Armen A. 1982. "First national maintenance vs. national labor relations board," unpublished manuscript. Cited in Williamson's Economic Institutions of Capitalism, p. 53.
Alchian, Armen A. and Harold Demsetz. 1972. "Production, information costs, and economic organization," American Economic Review 62:777-795.
Baumol, W.J. 1959. Business Behavior, Value, and Growth. New York: .
Chandler, Alfred D., Jr. 1962. Strategy and Structure: Chapters in the History of the American Industrial Enterprise, Cambridge, MA: MIT Press.
Coase, Ronald H. 1937. "The nature of the firm," Economica N.S. 4:386-405.
-- . 1960. "The problem of social cost," Journal of Law and Economics 3:1-44.
-- . 1964. "The regulated industries: discussion," American Economic Review 54:194-197.
Commons, John R. 1934. Institutional Economics. New York: Macmillan.
De Alessi, Louis. 1983. "Property rights, transaction costs, and X-efficiency: an essay in economic theory," American Economic Review 73:64-81.
Hurwicz, Leonid. 1972. "On informationally decentralized systems," in C.B. McGuire and R. Radner (eds.), Decision and Organization, Amsterdam: North-Holland.
Leibenstein, Harvey. 1966. "Allocative efficiency vs. `X-efficiency'," American Economic Review 56:392-415.
Robinson, E.A.G. 1934. "The problem of management and the size of firms," Economic Journal 44:240-254.
Ross, S. 1973. "The economic theory of agency: the principal's problem," American Economic Review 63:134-139.
Spence, A.M. and Richard Zeckhauser. 1971. "Insurance, information, and individual action," Journal of Economic Literature 21:981-990.
Stigler, George J. 1976. "The Xistence of X-efficiency," American Economic Review 66:213-216.
