The paper – which can be downloaded here – seeks to deal with a significant challenge for competition law, which has become more prominent with the proliferation of high tech markets: several practices, and particularly non-price strategies, fall outside the scope of competition law because mechanisms for assessing their legality are not adequate. The author’s ambition is to contribute to the literature by advancing a new test, called the “enhanced no economic sense” test, to be applied to non-price strategies.
The paper proceeds in three steps:
The first part presents the enhanced no-economic sense test. This test is based on the simple idea that a practice should be regarded as anti-competitive if it only makes sense from an economic point of view because of its tendency to eliminate or to restrict competition. Unlike the ‘profit sacrifice test’, the no-economic sense test allows for the condemnation of practices that do not lead to the infringer. The test follows four steps:
- Step 1: Does the practice implemented by the dominant company tend to reduce or eliminate competition? If the answer is negative, the practice is legal. If the answer is positive, the analysis moves on the second step.
- Step 2: Does the practice provide a benefit to the dominant firm solely because of its tendency to reduce or eliminate competition? If the answer is positive, the practice must be condemned. If the answer is negative, the analysis moves on the third step.
- Step 3: If the judge suspects that the practice may have both pro- and anti-competitive effects, he must determine whether it is possible to distinguish between the different elements of a conduct and to identify their economic justifications. If the answer is negative, the practice is deemed to be legal as a whole. If the answer is positive, the analysis moves on the last step.
- Step 4: The parts of a conduct that make economic sense for reasons that are not anti-competitive must be allowed. The parts of a conduct that only tend to reduce or eliminate competition must be condemned. The penalties should be proportional to the intensity/number of anti-competitive practices.
The second part of this section deals with a number of criticisms directed at the no-economic sense test. A particularly important criticism is that this test is not able to evaluate hybrid practices that produce both positive and negative effects on competition – since such practices will theoretically make some economic sense. The author thinks this is an advantage of the test, because it errs on the side of promoting innovations and innovative business practices. In any event, the author purports to apply an improved version of the no economic sense test, which would enhance legal certainty while minimising legal errors. As regard software, where code is required for each product change, the test could be applied by identifying the purpose and effects of every update of a product, and, more specifically, of each element of each update. This approach could be extended to other divisible technological developments. For indivisible developments, anticompetitive effects would be left undisturbed as a result of the development also having a legitimate pro-competitive goal.
Part two revisits a number of important predatory innovation cases through the prism of the enhanced no-economic sense test. This includes the notorious US Eastman Kodak case (1979); the 2001 US Microsoft case; the 2004 EU Microsoft case; and the US Apple iPod cases (2010-2014). The basic finding of this section is that the proposed legal test would have led to clearer analysis in these cases and, potentially, to different outcomes through a differentiation of those (separable) practices that were anticompetitive from those that were procompetitive.
The last part of the paper presents the author’s conclusion that the enhanced non-economic sense test would improve the quality of the law as regards the analysis of non-price strategies; that it would create a uniform rule of law and thus enhance legal coherence; and that it would create a legal test which would be easier to implement and that would ultimately increase consumer welfare by encouraging companies to continue to innovate.
Comment: This is an ambitious piece – perhaps too ambitious. I agree with the author that our tests for non-price infringements are patchy at best. Hence, I think it’s inherently suspect to purport to have developed an approach that works for all non-price practices while also avoiding errors. And the fact is that the proposed test does not do these things – or is only able to achieve these goals by tacitly advocating a view of competition law where only the most egregious anticompetitive practices are prohibited, without discussing the various controversial normative positions that underpin such an approach (e.g. regarding the goals of competition law, regarding the inability of courts to obtain perfect information, regarding the balance between Type I and Type II errors, etc.). I also don’t see how the proposed test would apply beyond specific markets – namely those concerned with software – and incremental innovations.
But the proposed approach may still prove valuable. It is true that we could do with better tests, and that taking into account the separability of the various elements of a business practice may be a better way to approach competition analysis than just assessing the practice in the round ex post to determine whether to fine a company. Such an approach could potentially limit the impact of false positives on market innovations, and may lead to clearer guidance to businesses on which practices are prohibited under competition law.