The “Small Market Problem” in EU Merger Control: Time to move on from the market definition argument

Kluwer Competition Law Blog
February 3, 2011

Please refer to this post as: , ‘The “Small Market Problem” in EU Merger Control: Time to move on from the market definition argument’, Kluwer Competition Law Blog, February 3 2011, http://competitionlawblog.kluwercompetitionlaw.com/2011/02/03/the-small-market-problem-in-eu-merger-control-time-to-move-on-from-the-market-definition-argument/


The fairness of EU competition policy has been frequently challenged in the 2000s. It has been argued that the European Commission discriminates against small market companies in its merger assessment. Discussion on the reasons of the alleged discrimination has frequently revolved around the Commission’s method of market delineation in its merger assessment, which would result in a so-called “false positive” or “type I error”. This would imply an incorrect finding of dominant position / the SIEC considerably more often and more easily in mergers between two small market companies, relative to the same sized rivals based on large national markets. Almost as often it has been argued that the method of market delineation should be changed in order to solve the alleged problem.

I have recently argued (Oinonen, M. “Does EU Merger Control Discriminate against Small Market Companies? Diagnosing the Argument with Conclusions”, Kluwer, International Competition Law Series, 2010) that this represents a common misconception of the real cause of – and particularly a viable solution to – the alleged small market problem. The rationalization behind this conclusion is wide in nature and requires both theoretical and practical discussion. In this post, I will put forward some basic thoughts on why we should move on from the market definition argument.

The discussion as to the alleged small market problem is basically analogous to a more common discussion regarding the traditionally heavy role of structural measures in the merger assessment. This has been famously presented in the US Supreme Court’s decision in Eastman Kodak Co. v. Image Technical Services, Inc., et al. in 1992: “[b]ecause market power is often inferred from market share, market definition generally determines the result of the case.” Underlying the argumentation on the small market problem is clearly also a belief that since the merger analysis often relies so heavily on the use of market shares and other structural measures of market power, market definition de facto determines the outcome of the case. This has often been found particularly problematic for the small economies where there are frequently only a few relatively big companies in small oligopolistic markets, in terms of market shares.

This argument has a tempting logic. Market delineation is of course very meaningful to the size of market shares. It is also highly possible that market shares and other concentration indices are given too much weight in the analysis, even in such a manner that ultimately they affect the result of the case too heavily.

I have recently suggested that this would indeed seem to offer one reasonable explanation to the small market problem (Oinonen, 2010). The real problem appears to relate to the false reliance in the analysis on premature conclusions of market power made on the basis of quantified market shares and concentration indices. This does not, however, make the market definition the real source of the alleged problem.

A key question often forgotten regarding the role of market delineation is that even if market shares were de facto that meaningful for the result of the merger analysis, would that make market delineation the real source of the problem? It would not. Namely, if this really was the case – i.e. if the market delineation really proved to be that decisive for the final outcome of the analysis – one should ask that what would that then tell us about the assessment of market power in general? This would mean that, contrary e.g. to the European Commission’s statement in its Horizontal Merger Guidelines, the assessment was not a genuine overall assessment in which market shares and other structural measures were meant as mere preliminary indicators of market power, and would not suffice by themselves to determine the outcome of the case. In practice, market shares would then have determined the result of the analysis – de facto, ended the analysis. Should that really be the case, what would that then tell us about the real cause of the problem?

The answer is pretty obvious. The real problem was not then, by definition, how the relevant markets were delineated, but rather, how the information drawn on the basis of the market shares in light of the relevant markets was applied in practice. I.e., how premature conclusions were drawn on the basis of the market shares and how fast such conclusions were held in the analysis? In theory at least, even large market shares could be approved in a correctly carried out overall assessment of market power. It is different if the genuine overall assessment will not de facto materialize due to the exceedingly static assessment practice bogged down by structural considerations.

Nor does the market delineation seem to offer a viable solution to the alleged problem. Let’s consider two options often brought out in the literature and the public discussion: Changing the method of market delineation so as to result in wider relevant markets in general or in small market mergers, in particular.

A change resulting in wider relevant markets across-the-board (i.e. in all mergers assessed under the ECMR) could hardly come into question, merely to try to solve a problem faced by small market mergers. Despite critical commentaries regarding some individual cases, there seem not to exist much discussion implying that the market delineation, by definition, would generally cause insuperable problems. Hardly any exceptions favouring the small markets could come into question either in EU merger policy. This seem clear already due to fact that there cannot be an ad hoc policy; nor would it probably fit in the fundamental idea of EU-wide merger policy that all companies – i.e. also the ones based on large Member States – were treated equally and given equal possibilities to benefit from the European internal market. In short, solving the alleged small market problem by interfering with the method of market delineation could hardly be a viable solution.

This approach shows how persistent adherence to the market delineation argument easily only makes us miss the truly important reasons for the alleged problem. As so often noted, market shares are not meaningless for the small market problem. They are indeed at the heart of the problem. The important point often missed here, however, is that this is not because of the market delineation, but because of how the information drawn on the quantified market shares and other concentration indices is put into practice as a part of the analysis. A key question is: Will the genuine overall assessment really materialize in practice, and if not, why is that so?