This paper, available here, argues that legal requirements and economic reasoning are not aligned as regards predatory pricing. Predation is not a strategy predominately used by ex ante dominant firms, but rather a strategy to gain ex post dominance. Consequently, the current legal practice in Europe and other jurisdictions, which requires ex ante dominance to pursue predatory pricing, makes the prosecution of predatory pricing virtually impossible because it overlooks the basic economic rationale for predatory pricing. This inconsistency has become even more severe because the adoption of a “more economic approach”: in fact, the more accurate the economic assessment is, the less probable is a conviction of harmful predation under the current legal framework.
The authors suggest prohibiting predatory pricing independently from other exclusionary abuses. Instead, predatory pricing should be subject to the same analytical framework as mergers, where a similar economic and business logic applies. Since recoupment of predation is akin to the unilateral effects arising from the merger between a predator and its prey, competition authorities can use their well-established tools and methods from merger control to assess the harm of predatory pricing.
The paper is structured as follows:
- Section 2 summarises the main economic theories of predation.
Modern economic theories commonly analyse predatory pricing as a dynamic process, particularly as intertemporal price discrimination. During a first period, a firm (the predator) offers very low prices to foreclose actual competitors from the market or to prevent potential competitors from entering the market. In a second period, the preying firm recoups its losses from the first period by extracting monopoly rents. A common denominator of all rational models of predation is the importance of asymmetric and imperfect information. Another thing that such models have in common is that they do not require ex ante dominance for rational and harmful predation to occur. It suffices that the market believes that a firm is strong enough to conduct a successful predatory strategy and enjoy monopoly profits in the future.
For predation to be successful – and have detrimental effects on consumer welfare – what matters is ex post rather than ex ante dominance. However, currently predation is only illegal for dominant firms, even though it is difficult to find circumstances in which predation is a rational strategy for a dominant firm. Dominance renders predatory strategies unnecessary, as a dominant firm can already enforce prices above the competitive level and the marginal gains from additional market power are small. The most likely predation scenario by a dominant firm might occur in an industry with strong economies of scale and/or strong network effects. There, a dominant incumbent can pursue a predatory pricing strategy to prevent a more efficient competitor from growing and achieving sufficient economies of scale. Another scenario in which the dominance condition does not destroy incentives for a rational predation strategy is predation occurring in another market than the one where exclusion is envisaged. A reputation for toughness will affect most business segments of a firm and deter not only imminent but also future market entry. A similar argument applies to low-cost signals, especially if there are common costs across different markets.
The problem, however, is that all these scenarios can be subject to traditional market foreclosure analysis. As a result, it is unclear when predatory pricing analysis can be useful from an antitrust standpoint when applied to ex ante dominant firms.
Section 3 presents the legal framework applicable to predation and explains why it contradicts economic thinking on the topic.
In the United States, predatory pricing has been of concern for almost a century. Section 2 of the Sherman Antitrust prohibits monopolisation and attempted monopolisation, including predatory pricing. Before 1975, when an article by Areeda and Turner introduced a “meaningful and workable test” based on average variable cost (AVC), the conviction rate for predatory pricing was above 75%. A second significant change was brought in by the Brooke case in 1993, which created a requirement of proof that a predator could recoup its losses due to increased market power. Such proof turned out to be very difficult to establish, and the outcome is that not a single successful predatory pricing case has been brought since then.
In Europe, predatory pricing will only infringe competition law if a firm engages in it to protect or strengthen a pre-existing dominant position. In addition, European antitrust enforcement has undergone a move towards a “more economic approach”, with the goal of substituting formalistic per se rules for an effects-based approach rooted in economic theory. Whereas, in the past, the legal framework focused on cost-price comparisons and required the proof of an exclusionary strategy (i.e. intent), under the effects-based approach the crucial thing is to show that the predation strategy entails a sacrifice for the preying firm. The problem with applying the ‘more economic approach’ to predatory pricing is that, as soon as a firm has taken up a dominant position, the economic rationale for pursuing a predatory pricing strategy tends to disappear. In practice, the European Commission’s enforcement practice has restricted itself to the two very narrow scenarios where it will be rational for a dominant company to engage in predation: either a dominant firm leverages its market power from one market (where it is dominant) to another market; or an incumbent firm deters market entry. While a credit to the European authorities’ enforcement practice, the problem is that this leaves any other category of harmful predation beyond the reach of European antitrust policy, when successful predation by an ex ante dominant firm might be less harmful to competition and consumers than predation by a non-dominant firm. The latter may permanently alter the market structure and increase prices substantially, whereas the former only results in negligible gains of market power. In other words, the pricing conducts with the strongest economic effect on competition cannot be pursued, which leads to under-enforcement.
Section 4 explains why predation is akin to a merger.
Economically, mergers and predation are substitute strategies to gain a dominant position, even if an apparent willingness to avoid price wars and prefer takeovers instead may induce additional entry into the market. Hence, reputational considerations can make predation a superior strategy to merging, since it creates externalities in other markets and over a longer period. Mergers and predation can also complement each other if predatory pricing improves the condition of a takeover bid and deters future entry.
As regards anticompetitive effects, the structural effects of mergers and predation are very similar. Ex post, the merged entity as well as the predator may be able to raise prices unilaterally. Hence, successful predation results in “unilateral effects” akin to those in a merger. Like a merger, successful predation changes the market structure as the market shares of one firm are acquired by another. Recoupment is only possible if the predator, in the absence of the competitive constraint of the prey, can charge excessive prices. Since predation and mergers produce essentially the same market outcome, there is no reason to treat these two strategies differently.
Section 5 concludes by proposing a feasible framework to address predation.
A better “more economic approach” to predation would consist in aligning it with merger control and dropping the ex ante dominance requirement. In merger control, the likelihood of an impediment to competition has to be established on the basis of a forward-looking economic reasoning that takes into account all potential effects of the transaction. The adoption of the SIEC-test (“significant impediment of effective competition”) decoupled merger control from dominance, even though in most (but not all) cases a SIEC is equivalent to acquiring a dominant position. Predatory pricing should be analysed along the same lines. Like in merger control, competition authorities should conduct an ex ante evaluation of potential restrictions of competition in predation cases. This requires estimations regarding future market structures and events, such as the possibility of market entry, the role of switching costs, sunk costs and network effects—all well-known concepts from merger control. Hence, competition authorities could use their well-established tools and methods from merger control to assess the harm of predatory pricing. While this would move away from the current rule-based approach to predatory pricing in Europe, it need not detract from legal certainty. Firms and authorities are accustomed to assess antitrust concerns in the case of a merger: methods and principles are tailored to identify harmful concentration in a market, but also to prevent excessive intervention by the authority. There is no reason why companies would be unable to apply the same principles to predation.
In assessing predatory pricing, the relevant question is whether successful predation can eliminate competition when the predator takes over a prey’s market share. If yes, recoupment is possible and the negative effect for competition is apparent—irrespective of ex ante dominance. However, abandoning the dominance condition might cause excessive intervention, since dominance also serves as a screening device. Hence, there is a need for an alternative criterion. The solution to this could also be borrowed from merger control, where intervention thresholds based on turnover and market shares are common. Finally, the fine distinction between competition on the merits and harmful predatory pricing requires that serious consideration be given to efficiency considerations, as is again common in merger control. Together with the traditional criteria of potential foreclosure and sacrifice, these three elements from merger control create a feasible framework to pursue predatory pricing in line with modern economics.
I think this a solid paper that questions the logic of the current rules governing predatory practices. I am unable to comment on how solid the argument is from an economic standpoint, but it speaks to concerns about the absence of a ‘monopolisation’ infringement in Europe akin to that in place in the US.
The only quibble I have with the paper is with the argument that moving to a more economic approach as regards predatory pricing does not reduce legal certainty. I can see two ways this argument might legitimately be made. First, calculating costs is such an uncertain endeavour that assessing effects will not be any harder (i.e. an empirical argument). Second, moving away from a per se rule based on dominance and costs towards an effects-based approach is justified by the risks of over- and under-enforcement the authors identify (i.e. a normative argument). However, to say that legal certainty is not affected because principles common to merger control would apply, as the authors do, is to confuse things.
Moving from a rule to a standard – which is what the authors’ approach entails – means that how the law is applied in specific cases becomes more uncertain; the fact that the methods to apply the standard are well-known does not change this. This can be shown by how contentious the analysis of problematic mergers can become, and how long it takes to evaluate them properly. Reduced legal certainty may well be justified, in this context as in others; but it is important to acknowledge when such reductions occur, to seriously discuss whether they are justified.