Chillin'Competition

Yesterday, the US AAG for Antitrust C. Varney murdered a widespread antitrust dogma:

“Defining markets and measuring market shares may not always be the best place to start” (see also here)

In many competition law regimes, market definition is indeed perceived as:

“a necessary precondition for the assessment of the effect of a concentration” (source: Case T-2/93, Air France v. Commission [1994] ECR I-1375, §143).

Of course, it is common in EU law to leave market definition opened, on the ground that regardless of the market’s exact boundaries, the merging parties’ market shares will not create or strenghten a dominant position.

This being said, many competition authorities nonetheless review this issue systematically, whilst they could screen anticompetitive concerns more rapidly. This burdensome analytical framework is particularly inefficient in cases where:

1. The agency can immediately ferret out risks of competitive harm, because the key economic elements of a particular theory of harm are absent. This happens, for instance, in oligopolistic dominance cases, where tacit collusion is unlikely because the market is not transparent, or because oligopolists are asymetric;
2. The agency can immediately establish that the merger raises serious doubts. This may happen, for instance, in cases where the merger involves the acquisition of a disruptive maverick player or where, thanks to sophisticated economic tools, the direct assessment of market power is possible (i.e. merger simulation on oligopolies with differentiated products).

Notwithstanding this, Varney’s assault on  market definition analysis should not be interpreted as meaning that no preliminary work on market definition is required. In both scenarios 1 and 2, the competition authority should follow a “first-look” market definition analysis so as to identify – at least approximately – those firms which are likely to compete with the merged entity. To take again the above example of a merger involving a disruptive maverick, a gross market definition is necessary to establish that the maverick firm was, prior to the merger, competing head to head with – and threatening the business interests of – the acquiring firm. Similarly, in the above mentioned oligopolistic dominance example, the competition authority needs to frame the market, if only to establish that it faces a tight oligopoly where the conditions for tacit collusion ought to be tested.

A final remark:  Varney’s statement holds the potential to lighten the evidentiary burden imposed on US antitrust authorities. Let’s just hope that the agencies will seek to use the newly freed-up administrative resources to improve their analytical framework, at the stage of the assessment of the theories of harm. This is particularly important in light of the fact that recent theories of harm in merger control are (i) increasingly intrusive: they may lead to forbid mergers with just a harmful effect on some customers (see TOTM excellent post on this); (ii) speculative: think for instance of behavioral, conduct-based, speculations in the context of vertical/conglomeral mergers; and (iii) and complex to test: think of merger simulation techniques, econometric analysis, diversion measurements, etc.

PS1: Whilst writing this post, I consulted the excellent book of Alistair Lindsay “The EC Merger Regulation: Substantive Issues”, Thomson – Sweet & Maxwell. I really recommend this book. It is the only EU merger related book that focuses only, and with great detail, on the substantive issues of EU merger analysis. My concern, however, is that the book is simply unaffordable (170£). My edition dates back to 2003. No chance I am going to change it any time soon for the 3rd edition of 2009. The bottom-line: not unlike some conference organizers, publishers should price discriminate and grant discounts to university researchers and students.

PS2: Varney’s speech was delivered in the context of the US agencies’ ongoing consultation process re. the 1992 merger guidelines.

(Image possibly subject to copyrights: source here)