Seminar 3

Seminar 3: Oliver Williamson

In this seminar we look at the way in which Oliver Wiliamson built on the work of Ronald Coase, to look more closely at the role of transaction costs. This has led to the growth of what is referred to as ‘new institutional economics’, which we consider here. The economics underpinning this approach depart from neo-classical economics and what we consider here are the policy implications that result from this departure.


Williamson ‘Transaction Cost Economics: The Governance of Contractual Relations’ (1979) 22(2) Journal of Law and Economics 233

Williamson ‘The New Institutional Economics: Taking Stock, Looking Ahead,” (2000) 38(3) Journal of Economic Literature 595

Further Reading

Williamson’s Nobel Prize lecture:

6 comments on “Seminar 3

  1. Irina Domurath says:

    I must admit that I found Williamson’s 2000 paper very difficult to read and understand. From what I understand, Williamson’s insight gives a new twist to what we have learned so far about the Chicago School in general and Coase in particular.

    Williamson provides for an analysis of governance of contractual relations through the lense of transaction costs. He explicitly grounds his research on Coase’s revelation of the importance of transaction costs and concludes that special governance structures replace the standard market-oriented contractual exchange when the so-called transactions-specific value is great, namely in long-term and relation-specific contracts, for example in the fields of commercial, labour and family law. In his 1979 articles he draws up two insightful schedules matching common transactions with respective governance structures, according to which only non-specific transactions should be entirely subject to market governance (classical contracting). In contrast, idiosyncratic or mixed transactions are subject to neoclassical (supported by third-party assistance through arbitration) and relation (built upon trust) contracting.

    It becomes clear that the transaction-cost-grounded approach bears relevance for the important role of institutions in contract-governance. However, I had a problem in comprehending the institutional focus when solving the issue of transaction costs, given that there are other market participants than corporations, firms, and institutions, which have a role to play when placing individual choices. Moreover, I found it unclear whether and to what extent Williamson relates to public policy-making. Finally, even though Williamson does mention the “human dimension” in decision-making processes, I do not see his approach as going far beyond the conventional approach in economics based on rational behaviour.

  2. Tim says:

    Irina mentioned the issue to what extent Williamson’s analysis is helpful to analyse public policy-making. I had the idea that his analysis might be helpful to look at certain problems that occur when authorities have to cooperate (and fail to properly do so). More in particular, I was thinking about the performance of unlawful state aid recovery by Member States if this ordered by the Commission (which is highly problematic), which requires from both side transaction-specific investments while such transactions occur frequently. According to Williamson’s model, both parties should vertically integrate. This is not the case as to state aid recovery, and may be Williamson insights offer a way to approach this problem from a different angle.

    Also, when Williamson analyses which governance structure is appropriate to govern contractual relationships, he seems to focus on the interests of the parties involved. He does not seem to look at the externalities. However, for instance vertical integration may have implications for competition, producing consequences affecting. I missed this aspect of externalities in his paper, as e.g. competition law might render it impossible for parties to apply the governance structure that suits them the best from a cost economizing point of view.

  3. Argyri says:

    Apologies for the delayed response!
    I found Tim’s point on competition law rules possibly being an obstacle to parties agreeing on a cooperation governance structure that suits them very interesting and look forward to discussing further in class. If I understood correctly especially from the first article (1979), Williamson sees a spectrum of optimal organization for commercial transactions, some closer to the market (thus with more competition rules) and some less. This is to be specified according to the degree in which 1.uncertainty, 2. frequency of exchange and 3. transaction specific investments appear in each case (applying this logic he sees, for example, a necessity in regulation of natural monopoly (p. 257)). Thus, maybe competition law does not concern him at all in some cases, and concerns him more in cases of nonspecific transactions, which he thinks are efficiently organized by markets.
    Last, I wanted to add the following to the discussion:
    After reading the materials, I was thinking of the impact that this theory of economics of organization (transaction cost economics) has, or must have, on the design of Alternative Dispute Settlement mechanisms. I understood that a starting point of the theory is that adjudication in case of breach of contract is the least optimal solution and thus what is best is finding the optimal organization of the relationship in order for it to be flexible and the parties to be able to trust the continuity of the contract. I am wondering, however, whether very cheap adjudication could also be a solution. Or is the reputation loss caused by an adjudication (or maybe the loss of a presupposed trust between the parties, that presumably existed before they reached to court) the biggest cost of all, that makes adjudication in any case worth avoiding?

  4. Marios Iacovides says:

    I found Williamson’s analysis in “Transaction-Costs Economic” to offer an interesting way to look at and revisit old cases but also explain current phenomena.

    For example, his views on vertical integration and the explanation of why it occurs sheds new light on Commercial Solvents. Did the Commission really look at all on whether it made sense for the supplier of the chemical to vertically integrate and stop supplying Zoja? Who broke the trust required for a long-standing idiosyncratic relationship first: Zoja who had turned to other suppliers in order to get the chemical, or Commercial Solvents who refused to supply a long-standing customer?

    His application of commercial transaction governance structures (albeit modified) to other types of transactions, such as labour markets, was also interesting. Cases such as Viking and Laval came to mind. What happens when a model which has worked without problems for so long suddenly gets a shock? Unions take control of arbitration so that group interests can supersede individual interests and opportunism, and courts should as far as possible stay out of disputes on these matters, but what happens when different group interests collide? In Laval it was the Latvian workers’ interests versus the Swedish workers’ ones, a collective agreement set against another collective agreement. Could the ECJ really have stayed out of the dispute even though Swedish law did not regulate the matter by law but let it to the unions to de facto regulate through negotiation and collective bargaining?

    Finally, Williamson’s comment that “A transaction-cost justification for regulating activities for which transaction-specific investments are lacking (for example trucking) is not apparent. The possibility that politics is the driving consideration in such industries warrants consideration” reminded me of the recent discussion in Greece with regard to requirements from the IMF to open up various trades (such as trucking). Since there could not be any transaction-cost justifications for regulating such activities (e.g. restricting the number of licenses to drive trucks, allowing truckers to sell their licenses to others so that the licenses themselves became traded for astronomical sums), Greece really had no other way of explaining the regulations beyond mere populism.

  5. Haukur says:

    I found his ‘four levels of social analysis’ quite interesting. It sort of explains why certain policy choices at the lower levels can be successful while others cannot due to pressure from higher ranking norms. Seen from this perspective it can be said that customs and traditions at the highest level influence to some extent the possible policy equilibriums at the lower levels. Interestingly he points out that the top level norms are remarkably resistant to change which in turn guarantees relative stability of the lower level equilibriums. For anyone that follows politics this might sound familiar. Politicians routinely promise change, while the result once elected is often more of the same. The reason for this could be that there is not that much that can be changed. Most things are already fixed in equilibriums determined by the higher institutional structure of society.

    Regarding the transaction cost article I think that perhaps a practical application of the analysis there could be relevant in the field of public procurement. Sometimes it could be better for the public authorities to build up expertise themselves instead contracting private firms, and the other way around.

  6. giorgiomonti says:

    I will have a go at responding quickly to some of the points you made here, hopefully complementing what we discuss in class.

    On Irina’s point about what’s unorthodox here. (a) it is clear that the human person that Williamson defines is not ‘homo economicus’ but is a tad closer to reality. he draws heavily on two scholars: Richard Dawkins and Herbert Simon to suggest certain characteristics of human behaviour that can be useful to model behaviour more accurately: bounded rationality, opportunism and conscious foresight. If you try and run Becker’s crime and punishment theory or the Rustichini example (seminar 1 readings) with this changed perspective, we might not get the same answers: some of the human attributes thwart the deterrence model approach; (b) traditional L&E does not look at transaction costs closely, but assumes them away. However, this underplays a relevant factor in deciding on what is efficient, so to try and claim that Copyright is efficient (Posner, seminar 1) without considering transaction costs involved in what is sometimes a property and sometimes a liability regime is not complete. In this way it corrects, but does not revolutionise conventional L&E.

    On the public policy dimension. This has been one of my difficulies with the work. Recall that Williamson worked at the Department of Justice Antitrust Division in the late 1960s. His most influential antitrust article was a claim that we should consider allowing mergers to monopoly because… you guessed it, the merger might lower transaction costs and so the firm may be more efficient. But in the two papers I selected today there is a bit of an ambiguity:he is looking at the way to economise production and transaction costs, but seemingly unconcerned about externalities. This might be a weakness, not least as Argyri pointed out, what if the most effective organisational makeup is anticompetitive?

    On applications. There is a very ambitious statement in the 1979 paper that any issue that can be analysied contractually may be delat with using TCE. Similar sentiments appear in the 2000 paper. Is this overly ambitious? Tim’s idea of thinking how to best design state aid recovery by the EU versus Member States is one example. Recall Becker: you need a structure where the probability of getting caught multiplied by the loss incurred is greater than be expected gain. From this perpsective you can criticise state aid law on both fronts: (1) low probability and (2) no loss to the state (they recover thier money (with interest). But TCE teaches you different: even if you raise the probability and increase the penalty you have major transaction costs before you get this right. so then is this an argument for an even closer union – a centralised budget? Haukur’s application to public procurement is also on point: there may be certain forms of recurrent purchasing for which outsourcing is inefficient from a TCE perspective. what is the policy implication here then? if we find that certain actors do things that raise their transaction costs, do we ban them from doing so? Mario’s point on Viking is worth noting too, indeed labour relationsis discussed in the 1979 article. Williamson would opt for non intervention when collective bargaining allows to absorb transaction costs better than the courts. But is this plea for non intervention based on TCE a weak one?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s