Book Review - The Economic Institutions Of Capitalism By Oliver Williamson

In this book, Williamson presents a refined and elaborated version of his transaction cost theory that he had first outlined in his 1975 book Markets and Hierarchies: Analysis and Anti-trust Implications. His book attempts to systematically examine those economic issues that classical economic theory simply assumes away. The classical economics believes that markets are perfect, and if they are not then the action to remove market failures needs to be initiated. Williamson, on the other hand, focuses on these economic issues that are acknowledged to be widely prevalent in any economic system. "If complexity is deep in the nature of things economic then that ought to be acknowledged rather than suppressed. An equilibrium approach to economics is thus preliminary to the study of main issues (Hayek on P8)." This book, then, is a scrutiny of such economic phenomenon as market structures, monopolies, anti-trust policies, labor policies, public utility regulation, vertical integration and other economic institutions that have traditionally been neglected by the economic theory.

His basic proposition that most of us are familiar with by now is that the transaction costs should be treated as a fundamental unit of analysis for understanding such issues. Drawing on three streams of research- economics, organization theory and contract law, he repeatedly highlights the need to consider the governance (or transaction) costs. "Rather than characterize the firm as a production function, transaction cost economics maintains that the firm is more usefully regarded as a governance structure (P13)." While his basic argument appeared sound and plausible, I got an impression that Williamson attributed more to transaction costs than it deserved. Why should we regard only governance costs? Why should we think that the firm is only a governance structure? In other words, in my view, instead of correcting an existing flaw in the theory, he seems to be, to borrow the stock market jargon, proposing an over-correction. The field would be better off considering a cost function that combines both production and governance costs or at least choosing the concept based on the specific requirements of the situation or problem at hand.

Having said that, let's now delve into the foundations of the transaction cost economics which is first three chapters in the book. Until this book, Williamson considered opportunism, bounded rationality, frequency and uncertainty to be the building blocks of TCE. However, in this book, he rightly puts forth asset specificity alongside opportunism and bounded rationality as the three legs of TCE. "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" (P42), which is assumed to be the most critical dimension of TCE (P30). Without asset specificity, markets are believed to be in a competitive world even if people are opportunistic and rationally bounded. This is because buyers and sellers can freely move between market players.

In contrast, uncertainty and frequency drop down a tad bit in the scheme of things. Now, they are supposed to be meaningful in presence of first three elements only. Conceptually, this makes a lot of sense. Take for example, if market players are uncertain about the outcomes, but they believe in the fairness of the parties to contract, the market mechanism would be adequate to deal with all the contingencies since the players would share equitably in the profits. However, we understand that such a behavioral assumption would be wrong since opportunism and bounded rationality are common behavioral traits. What intrigues me, albeit, is that if they are such common traits, then why they should even be made variables in the model. After all, a variable that doesn't vary is no variable at all. It is not surprisingly, therefore, to see most literature to refer only to asset specificity, uncertainty and frequency as the three pillars of TCE. Williamson himself seems to acknowledge this in a subsequent chapter when he mentions that "principal dimensions for describing transactions are frequency, uncertainty and asset specificity (P242)."

After outlaying his conception of economic fundamentals, Williamson proceeds on to explain the boundaries of firm, which is to say what transactions will take place in market and what within the hierarchically organized structures. In his opinion, if the expected costs or risk of transacting in a marketplace are higher than the cost of organizing the functions internally, then such transactions will take place within the firm. If we ignore his exaggerated claims, this is indeed novel and useful approach at looking the firm size and boundaries. No longer is the size of firm held irrelevant as is the case in classical economics. No longer is it believed that the firms will operate at marginal cost whether they produce internally or buy externally. It opens up a can of worms that classical economics under its perfect market and equilibrium economics assumptions puts aside as aberrations. This is a welcome change in the approach to the study of industrial economics.

Next, Williamson moves on to the main theme of the book: providing alternative explanations vertical integration, mergers and monopolies, and joining issues with anti-trust enforcements. He believes that vertical integration results not because of technological determinism or a desire for monopolistic power but from a pragmatic desire to economize on transaction costs. In the similar vein, he contends that non-standard contracting practices such as long-term contracts are not monopolistic practices, but perfectly justifiable attempts at minimizing transaction costs. Further, he attributes such decisions "to a condition of asset specificity (P86)" since asset specificity in conjunction with uncertainty "makes it more imperative to organize transactions within the governance structure that have the capacity to work things out (P79)." The author makes a persuasive case for five out of six hypotheses on the boundaries of firm. However, his fifth hypothesis that claims that "firms will never integrate for production reasons alone" seems a little far-stretched. The fact that some firms organize for efficiency reasons doesn't and can't automatically preclude the fact that some firms organize for monopolistic or technological reasons. Once again, the author's case would have been better served by refraining from such overstatements.

Next, Williamson turns his attention to analysis of such arrangements as can neither be classified as market contracts nor as hierarchical structures, but fall somewhere in between. Also known as hybrid structures, these include credible commitments, joint ventures, relational contracting, hostage models, reciprocal arrangements, and network relationships. His main claim is that even when such arrangements appear to be exercise of monopolistic power, they may be justifiable from transaction cost perspective. "A comparative institutional assessment of contractual alternatives discloses that efficiency purposes are often served by hostages and it is in the mutual interest of the parties to achieve that result. Not only can producers be induced to invest in the mutual interest of the parties to invest in the most efficient technology, but buyers can be induced to take delivery whenever demand realizations exceed marginal cost." Interesting proposition, but it doesn't explain the impact on the hostage (e.g. P&G) if the monopoly (e.g. Wal-Mart) decides to dump it! His second main claim derives from Coase's 1960 article on problem of social cost. Recall Coase's claim that when people are left to bargain among themselves, most economic externalities can be better resolved than when courts or other non-market interventions take place. Williamson develops on this proposition and claims that parties to a contract don't normally take recourse to courts, but try to use "private ordering" to resolve their disputes. I would presume this would chiefly be out of concern for future business relations.

Let's wrap up this review with a summary of strengths and weaknesses. For the strengths, I will let the Williamson speak for himself. To quote him,
"As compared with other approaches to the study of economic organizations, transaction cost economics (1) is more micro-analytic (2) is more self-conscious about its behavioral assumptions (3) introduces and develops the economic importance of asset specificity (4) relies more on comparative institutional analysis (5) regards the business firm as a governance structure rather than as a production function and (6) places greater weight on the ex-post institutions of the contract, with special emphasis on private ordering as compared with court ordering."
-Williamson, P387

While the theory is conceptually persuasive and logically sound, a principal weakness of transaction cost analysis lies in its post-facto nature of analysis. Notwithstanding Williamson's superb efforts, it has been rather difficult to define it in a way that it can be measured and tested. The theory in its current formulation continues to be plagued with a criticism that it's tautological in nature, after all ex-post facto any system can be shown to be economizing on transaction cost or at least that it will be eventually replaced if it doesn't. Therefore, transaction cost economics needs to find variables with predictive powers.

Williamson mentions three limitations of his work- its crude form, instrumentalism, and incompleteness. To me, these appeared more to be challenges for future research rather than any weaknesses in the theory. Besides occasional excessive enthusiasm and exaggerations and the difficulty in operationalization of the concept, a major challenge in reading this book is to be prepared to learn a new language! Williamson's choice of words lives a reader with no less an impression.

Overall, Williamson does a superb job in developing the transaction cost economics that had first appeared in Coase's 1937 article 'nature of firm', but had been left untouched until this work because of difficulties in operationalization and empirical testing. Williamson succeeded in overcoming most of these challenges and it is for the future researchers to meet the rest.

Reference:

Williamson, Oliver. The Economic Institutions of Capitalism. 1st. New York: The Free Press, 1985.