Journal of Retailing 86 (3, 2010) 227231

Transaction Cost Economics: The Origins

Oliver E. Williamson
University of California, Edgar F. Kaiser Professor Emeritus of Business, Berkeley, United States

Abstract This paper is an autobiographical recollection of how my interest in and work on transaction cost economics progressively developed. It begins with an overview of the transaction cost economics project. A sketch of my undergraduate and graduate education follows. Key events in the 1960s that set the stage for my 1971 paper on The Vertical Integration of Production are then described. This paper would turn out to be the entering wedge from which transaction cost economics would take shape and continues today, as set out in the closing pages of the article. 2010 Published by Elsevier Inc on behalf of New York University.
Keywords: Transaction cost economics; Pre-transaction cost research; Enumeration of the key events

This paper owes its origins to a request that I received from Arne Nygaard, asking me to write an introductory note for this special issue of JR. Rather than leave it at that, however, Arne went on to suggest that I reveal the story behind the success story of TCE . . .. I would like to know if there is a Newton story of falling apples (actual time and place). . .. I would like to know more about . . . Carnegie Mellon and the group of people there and the other universities where you worked.1 The answer to the rst of these is that there was no falling apple, but I did benet from a series of insights that, individually and collectively, served to unlock the gates to transaction cost economics. I begin by sketching the key moves that led into the transaction cost economics project. I then provide the details and people who participated in various stages of development. An overview As I look back on this project I sometimes think that I was predestined to do transaction cost economics. So many events lined up that it cannot be explained any other way. But of course, that is nonsense. It is not that one thing led inexorably to another but rather that my intuitions and interests and makeup were such that it was easy for me to incorporate new knowledge and insights as events unfolded. It could have been otherwise, but there was not a lot of slippage in the various turns in the road that I describe below. I

1 I have addressed these issues before, but in a less personal and less focused way than I attempt herein.

should nevertheless make clear that I owe a great deal to people who took an interest in my career and urged me to widen my horizon and/or reposition. I am especially grateful to my wife Dolores who was continuously supportive and adapted quickly, constructively, and cheerfully to new circumstances. Signicant events that led me into transaction cost economics include: (1) my undergraduate training at MIT, which provided me with analytical apparatus and grounding that would serve me well in the years ahead; (2) working for three years as a project engineer with large corporations and the U.S. government, which gave me a sense of how large bureaucracies operate, before deciding to do graduate work; (3) enrolling in the PhD program at the Graduate School of Business at Stanford only to discover, with the aid of junior faculty, that my real interests were in economics and that the Graduate School of Industrial Administration at Carnegie was, as it were, tailor made for me; (4) reveling in the excitement of interdisciplinary teaching and research at Carnegie; (5) discovering that teaching is learning in my rst job as Assistant Professor of Economics at Berkeley, where I was presented with many opportunities, as an applied microeconomist with interests in rm and market organization, to reformulate old issues constructively; (6) nding even more such opportunities when I left Berkeley to take a job as Associate Professor of Economics at the University of Pennsylvania, to include a teaching as learning experience in the newly created School of Public and Urban Policy; (7) serving as Special Economic Assistant to the Head of the Antitrust Division, where I came to appreciate that much of the economic literature that informed antitrust enforcement was wrong-headed; (8) my intuition that vertical integration could be usefully examined from a

0022-4359/$ see front matter 2010 Published by Elsevier Inc on behalf of New York University. doi:10.1016/j.jretai.2010.07.006

228

O.E. Williamson / Journal of Retailing 86 (3, 2010) 227231

lens of contract perspective in a combined economics and organization theory way; and (9) an appreciation that a wide range of other economic and organizational phenomena could be interpreted as variations on a few key transaction cost economizing themes. From engineering to business to economics As indicated, my engineering training at MIT provided me with a good toolkit and grounding where by the latter I mean that although hypothetical ideals are often a good place to start it is necessary thereafter to make provision for friction, resistance, turbulence, and human nature as we know it.2 (It was vital for me and, I conjecture, for many other applied microeconomists, to be grounded in this way.) I was accepted into the PhD program at the Graduate School of Business at Stanford in 1958 and took a series of MBA courses in the rst year, one of which was economics as taught by James Howell. Jim had recently received his PhD from Yale, then worked with Aaron Gordon, at Berkeley, on the Gordon-Howell report of the status of U.S. business schools,3 and had accepted an assistant professor appointment in the business school at Stanford. Howell was a very lucid, no-nonsense teacher and I found the subject of economics fascinating. After one of my visits to his ofce to discuss nuances he pronounced that I had very good intuitions and should be taking courses in the economics department. I followed his advice and took a microtheory class with Kenneth Arrow. Needless to say, that was an exhilarating experience. That is the only class that I took with Ken, but it had immediate and lasting effects. He remembers me as asking good questions and, to my very great pleasure, describes me as one of his students. This class with Ken settled the idea that I should take more economics, which I was able to do because the second year MBA program was very exible and allowed me to take courses in economics, mathematics, and statistics outside of the Business School in my second year.4 Because of an ofce crowding condition, three PhD students, myself being one, shared an ofce with another new faculty appointment, Charles Bonini, who had just nished his PhD at Carnegie. Chuck would ask me about my interests periodically, I would describe them, and he would tell me You ought to be at Carnegie. I knew nothing of the PhD program in the Graduate School of Business at Carnegie, but I eventually inquired and quickly recognized that this was a leading-edge program from which I was sure that I would benet. Having applied for and received a three-year portable fellowship from the Ford Foundation, and with the agreement of my wife to leave California for Pennsylvania, Dolores and I and our two children arrived in Pittsburgh in September 1960. We discovered, to our surprise and delight, that Pittsburgh was a very nice city.

Enumeration of the key events The overarching insight that I learned at Carnegie was the Carnegie Triple: be disciplined; be interdisciplinary; have an active mind. Jacques Dreze speaks for me and many others in his statement that Never since have I experienced such intellectual excitement (1995, p. 123). Highlights, at Carnegie and later, include the following: 1. The playful remark by James March in his organization theory class that large corporations are managed so as to maximize slack, rather than prots. This immediately registered on me as a research opportunity. 2. The Selling Expense as a Barrier to Entry paper that I prepared for Herbert Simons class on Mathematical Social Science was mainly an exercise in applied price theory, for which I was a bit apologetic. Simon saw it differently: Anything that advances our understanding of complex phenomena is to be valued. I was relieved by his pluralistic view of social science. And I was pleased when (a revised version of) this term paper was published in the February 1963 issue of the Quarterly Journal of Economics. 3. Although he did not put it this way, Jack Muth had a way of pushing the logic to completion that I had not previously witnessed. Pushing the logic would sometimes reveal that behavior that many public policy analysts, myself included, thought of as objectionable was altogether predictable. If deviations from normative public policy are not contrived but are the natural result of the circumstances, then understanding rather than misconstruing the origins of the regularities would be vital to the making of public policy. 4. Allan Meltzer encouraged me to be my own man at a critical juncture in my career where there was a divide in the road. Things could have gone otherwise. 5. Milton Friedmans critique of the simple Keynesian model, as based on empirical work that he had done jointly with Meisselman, was contested at the National Bureau of Economic Research conference on money-macro at Carnegie in 1962. No ones position was changed by the arguments that ensued, but Friedman had the data and empirical analysis to support his position. The graduate students in attendance concluded that when he says, she says impasses arise, he/she who has the data has the advantage.5 6. I read Alfred Chandlers book, Strategy and Structure sometime in the mid-1960s and was taken aback. In addition to competition in the product market and competition in the capital market arguments on which I had previously relied6 as checks upon managerial discretion, such discretion could also be held in check by internal organization.
5 Friedman later advised me (by email, on February 6, 2006) that I believe in every area where I have had some inuence that it has occurred less because of the pure analysis than it has because of the empirical evidence that I have been able to organize. 6 Managerial discretion is projected to be greater if competition in product and capital markets is weak.

The quoted phrase is from Frank Knight (1965, p. 270). This study was sponsored by the Ford Foundation. 4 I spent most of the second year in the classrooms of Herman Chernoff, Bernard Haley, Emanuel Parzen, Melvin Reder, and Hirofumi Uzawa.
3

O.E. Williamson / Journal of Retailing 86 (3, 2010) 227231

229

Specically, although Chandler never put it this way, the move from a functionally organized and centralized (unitary form) corporation to a divisionalized corporation where the central ofce had a planning and internal resource allocation capability (multidivisional form) could also attenuate managerial discretion. In that event, management was both the problem and (within limits) the solution. 7. My experience as Special Economic Assistant to the Head of the Antitrust Division in 196667 convinced me that, as able and conscientious as the leadership and staff of the Antitrust Division was, signicant parts of antitrust enforcement were wrong-headed. The major problem was this: except as nonstandard and unfamiliar contracting practices and organizational structures had a technological justication, all were presumed to be anticompetitive.7 (As the Head of the Antitrust Division pronounced, I approach territorial and customer restrictions not hospitably in the common law tradition, but inhospitably in the tradition of antitrust law.)8 I knew from my training at Carnegie that the prevailing thinking (Posner 1977, p. 1) upon which antitrust enforcement then rested was seriously truncated. Yet others were unpersuaded. 8. Two consulting assignments that I had in the late 1960s and early 1970s would prove to be important. The rst was my work on New York Mayor John Lindsays CATV task in 196970. The second was a consulting assignment that my Penn colleague Almarin Phillips and I took with one of the major petroleum companies after the rst oil crisis. The Federal Trade Commission, which had been told, on short notice, to undertake an antitrust inquiry, had quickly produced a laundry list of objectionable practices. Although most did not withstand scrutiny, there was one for which there was no good explanation: exchange agreements. The common justication was that exchange agreements saved on cross-hauling expenses. When I observed, however, that this could be realized by having each company report surpluses and decits to a central market, this was met by silence. The main response of the managers, engineers, and lawyers was that they did not like the question. That was disconcerting. 9. When Coase (1960) and Arrow (1969) pushed the thenstandard assumption of zero transaction costs to completion Coase with respect to externalities and Arrow with respect to vertical integration both reached a similar result: externalities and vertical integration both vanished when the logic of zero transaction costs was taken to completion. Since reality told us that externalities and vertical integration were part of the economics landscape, that was disconcerting. Evidently the standard assumption by economists that transaction costs were zero would have
7 This was because antitrust enforcement relied on applied price theory and the structure-conduct-performance paradigm, neither of which made provision for the importance of organization (broadly construed to include both the mechanisms of interrm contract and internal organization). 8 The quotation appears in the 1968 N.Y. State Bar Association, Antitrust Symposium, p. 29.

to go. Upon opening the black box of rm and market organization, however, positive transaction costs were everywhere. That was chaos. What to do? 10. It has been my experience as an academic that teaching is learning.9 This is a common theme in Section 4, which follows. The transaction cost economics project Pre-transaction cost research My response to Jim Marchs remark that managers maximize slack was to interpret this as expense preference and embed it in a managerial utility function that had three arguments: expenditures for staff, expenditures for emoluments, and discretionary investment. The utility function where these three favored types of expenditures appeared was then maximized subject to a minimum prot constraint. My dissertation, The Economics of Discretionary Behavior: Managerial Objectives in a Theory of the Firm, was completed in the spring of 1963 and, as a winner of the Ford Foundation dissertation competition, was published by Prentice Hall in 1964.10 Post-dissertation research of mine that served to credentialize me as a journeyman economist included Peak Load Pricing and Optimal Capacity under Indivisibility Constraints (1966), Social Choice: A Probabilistic Approach (with Thomas Sargent, 1967), Hierarchical Control and Optimum Firm Size (1967), Economies as an Antitrust Defense: The Welfare TradeOffs (1968b), and Wage Rates as a Barrier to Entry: The Pennington Case in Perspective (1968a).11 But my applied micro research would soon undergo what for me was an irreversible change. Transaction cost economics Given my dissatisfaction with antitrust enforcement on vertical integration/vertical market restrictions, I resolved to examine the economic literature on these topics when I returned to teaching at the University of Pennsylvania in 1967. Working off of the premise that teaching is learning, I organized a graduate

9 There is, of course, a tradeoff between teaching multiple sections of one course (with one preparation) and teaching multiple courses. I nevertheless found that teaching new courses usually presented me with new research opportunities and I was disappointed when a new course did not turn out this way. Indeed, when I had an opportunity to put two years of research back-to-back (the rst at the Behavioral Sciences Center at Stanford; the second at the Center for Advanced Study at Princeton), I declined the latter. 10 This formulation was inspired in part by William Baumols sales maximization hypothesis (1959), as was Robin Marriss growth maximization model (1964). 11 The rst of these had its origins in my undergraduate IO class in the spring of 1964; the second was a product of teaching a graduate class on The Pricing of Public Services in the spring of 1965; the third was my rst effort to solve the puzzle of rm size (as posed by Frank Knight and Ronald Coase); the fourth originated when I was with the Antitrust Division; and the fth originated in an undergraduate IO class that I taught in 1964 when an outside speaker disputed the possibility that wage rates would be used as a barrier to entry.

230

O.E. Williamson / Journal of Retailing 86 (3, 2010) 227231

student seminar in which the students and I went methodically through the literature. Although there were good and thoughtful price theoretic treatments of these issues, this literature could neither explain the wide degree to which vertical integration was practiced in the United States, Europe, and other progressive countries nor justify the inhospitality tradition in antitrust. That, however, was a negative conclusion. The challenge was to work up a constructive response. Another fortuitous development would facilitate that: the University of Pennsylvania hired Jules Margolis as the rst dean of the new School of Public and Urban Policy. Jules had an inspired vision for an interdisciplinary PhD program at SPUP and asked me to participate by teaching a two semester sequence in organization theory. Although I had never taught an organization theory course before, I viewed the offer as an opportunity to bring economics and organization theory together on the possibility that a combined effort would lead to a deeper understanding of vertical organization (and, with luck, to economic organization more generally). My paper on The Vertical Integration of Production: Market Failure Considerations took shape during my teaching of organization theory during the spring and fall of 1970, was presented at the annual meeting of the American Economic Association in January 1971, and was published in the Papers and Proceedings of the AEA in the May 1971 issue of the American Economic Review. This paper differed from orthodoxy in the following respects: (1) the orthodox lens of choice is supplanted by the lens of contract/governance (which leads into an examination of phenomena in microanalytic detail); (2) the assumption of hyperrationality is supplanted by bounded rationality (on which account all complex contracts are incomplete); (3) a narrow view of simple self-interest seeking is expanded to include strategic behavior during the contract implementation interval; (4) the standard assumption of zero transaction costs is supplanted by a focus on positive transaction cost differences, conditional on the attributes of transactions, as between markets and hierarchies; (5) adaptation (of autonomous and coordinated kinds, which correspond, roughly, to market and hierarchy) is taken to be the main problem of organization; and (6) much of the comparative institutional action is shown to reside in the condition of bilateral dependency (which could vary from slight to great, depending on whether the investments made in supporting assets were generic or specic) and disturbances to which adaptations were needed. This reformulation revealed that the decision to outsource a technologically separable good or service or to integrate (produce to ones own needs) turned on transaction cost differences. The seeds of a predictive theory of rm and market resided therein. That was a lot of novelty to pack into one paper and understandably took a while to register. I nevertheless knew that this was an important paper going so far as to tell Dan McFadden in the spring of 1971 that I felt that I was the rst person in the world to understand vertical integration. But it did not instantly register on me that this approach had ramications for the study of economic organization more generally. As, however, the class and I continued to wrestle with the problems of economic organization I recognized that, directly

or indirectly, we were continuously appealing to several recurring themes. So I gave the class the assignment (and took it myself) to work through the basic regularities over the weekend. What appears in Markets and Hierarchies as The Organizational Failures Framework (1975, chap. 2) had its origins in this class. Transaction cost economics got underway as I and others discovered that any issue that arises as or can be reformulated as a contracting problem can be examined to advantage in transaction cost economizing terms. A huge number of economic phenomena including vertical integration, vertical market restrictions, franchising, regulation and deregulation, labor market organization, the organization of work, corporate nance and corporate governance, family rms, multinational rms, and the economics of trust qualify. The fact that certain regularities keep recurring is responsive to Friedmans observation that a fundamental hypothesis in science is that appearances are deceptive and that there is a way of looking at or interpreting or organizing the evidence that will reveal supercially disconnected and diverse phenomena to be manifestations of a more fundamental and relatively simple structure (1953, p. 33). This would be repeated, moreover, in the business schools, where applications of transaction cost economics to marketing, strategy, organizational behavior, nance, operations management, and accounting would be made. Applications within the contiguous social sciences, and law were also numerous and growing. My experience with the CATV industry, as acquired from working on the New York City project that I mentioned earlier, permitted me to address the question of Franchise Bidding for Natural Monopoly in General and with Respect to CATV (1976) in a more nuanced way than it had been addressed previously (Posner 1972). Likewise the puzzle of exchange agreements in the petroleum industry that I had rst encountered in the early 1970s was one that I could interpret in credible contracting terms in Credible Commitments: Using Hostages to Support Exchange (1983) once I read the incriminating memoranda on such agreements, as reported in the study on The State of Competition in the Canadian Petroleum Industry (Quebec 1981).12 Language that can (and was) interpreted as evidence of anticompetitive purpose and effect when examined through the orthodox lens of applied price theory could and often did take on an altogether different interpretation when examined through the lens of contract/governance. This is not to say that the latter lens is right and the earlier lens is obsolete. Public policy is poorly served, however, by exclusive reliance on the former. Knee-jerk monopoly reasoning is bankrupt. My position is that transaction cost economics is one of the informative lenses with which to study complex economic organization. The 35 years from Markets and Hierarchies (1975) to

Robert J. Bertrand, Q. C., Director of Investigation and Research, Combines Investigation Act, coordinated the eight-volume study, The State of Competition in the Canadian Petroleum Industry (Quebec, 1981).

12

O.E. Williamson / Journal of Retailing 86 (3, 2010) 227231

231

2010 reveal that transaction cost economics has not only had wide reach but is an empirical success story: empirical tests of the predictions of transaction cost economics numbered over 800 as of 2006 and have been broadly corroborative (Macher and Richman 2008). Indeed, despite what almost 30 years ago may have appeared to be insurmountable obstacles to acquiring the relevant data [which are often primary data of a microanalytic kind], today transaction cost economics stands on a remarkably broad empirical foundation (Geyskens, Steenkamp, and Kumar 2006, p. 531). There is no gainsaying that transaction cost economics has been much more inuential because of the empirical work that it has engendered (Whinston 2001). This project is not, moreover, completed. Transaction cost economics faces a challenging future in conceptual, theoretical, empirical, and public policy respects as the baton is passed to a new generation. References
Arrow, Kenneth J. (1969), The Organization of Economic Activity: Issues Pertinent to the Choice of Market Versus Nonmarket Allocation, in The Analysis and Evaluation of Public Expenditure: The PPB System. Vol. 1, U.S. Joint Economic Committee, 91st Congress, 1st Session, Washington, DC: U.S. Government Printing Ofce, 5973. Baumol, William J. (1959), Business Behavior, Value and Growth, New York: Macmillan. Chandler, Alfred D. Jr. (1962), Strategy and Structure, Cambridge, MA: MIT Press. Coase, Ronald H. (1960), The Problem of Social Cost, Journal of Law and Economics, 3 (1), 144. Dreze, Jacques (1995), Forty Years of Public Economics A Personal Perspective, Journal of Economic Perspectives, 9 (2), 11130. Friedman, Milton (1953), Essays in Positive Economics, Chicago: University of Chicago Press. Geyskens, Inge, Jan-Benedict E.M. Steenkamp and Nirmalya Kumar (2006), Make, Buy, or Ally: A Meta-analysis of Transaction Cost Theory, Academy of Management Journal, 49 (3), 51943.

Knight, Frank H (1965), Risk, Uncertainty, and Prot, New York: Harper & Row, Publishers, Inc. Macher, Jeffrey T. and Barak D. Richman (2008), Transaction Cost Economics: An Assessment of Empirical Research in the Social Sciences, Business and Politics, 10 (1), 163. Marris, Robin (1964), The Economic Theory of Managerial Capitalism, New York: Free Press. Posner, Richard A. (1972), The Appropriate Scope of Regulation in the Cable Television Industry, The Bell Journal of Economics and Management Science, 3 (1), 98129. (1977), Economic Analysis of Law, 2nd ed. Boston: Little, Brown. Whinston, Michael (2001), Assessing Property Rights and Transaction-Cost Theories of the Firm, American Economic Review, 91 (2), 1848. Williamson, Oliver E. (1963), Selling Expense as a Barrier to Entry, Quarterly Journal of Economics, 77, 11228. (1964), The Economics of Discretionary Behavior: Managerial Objectives in a Theory of the Firm, in Ford Foundation Doctoral Dissertation Series Englewood Cliffs, New Jersey: Prentice Hall. (1966), Peak-Load Pricing and Optimal Capacity under Indivisibility Constraints, The American Economic Review, 56 (4), 81027. (1967), Hierarchical Control and Optimum Firm Size, Journal of Political Economy, 75 (April), 12338. (1968a), Wage Rates as a Barrier to Entry: The Pennington Case in Perspective, Quarterly Journal of Economics, 82 (February), 85116. (1968b), Economies as an Antitrust Defense: The Welfare Tradeoffs, American Economic Review, 58 (March), 1835. (1971), The Vertical Integration of Production: Market Failure Considerations, American Economic Review, 61 (2), 11223. (1975), Markets and Hierarchies: Analysis and Antitrust Implications, New York: Free Press. (1976), Franchise Bidding for Natural MonopoliesIn General and with Respect to CATV, Bell Journal of Economics, 7 (1), 73104. (1983), Credible Commitments: Using Hostages To Support Exchange, American Economic Review, 73 (4), 51940. Williamson, Oliver E. and Thomas Sargent (1967), Social Choice: A Probabilistic Approach, The Economic Journal, 77 (308), 797813.