A post in another place about Areeda’s article on essential facilities – “an epithet in need of limiting principles” – reminded me that limiting principles are not only needed to decrease the risks of over-enforcement. Areeda’s legitimate concern about nuance being lost in the application of precedent applies also to precedents which limit liability. They are relevant to under-enforcement as well.

From hard cases to bad law

Areeda wrote:

“As with most instances of judging by catch-phrase, the law evolves in three stages: (1) An extreme case arises to which a court responds. (2) The language of the response is then applied -often mechanically, sometimes cleverly- to expand the application. With too few judges experienced enough with the subject to resist, the doctrine expands to the limits of its language, with little regard to policy. (3) Such expansions ultimately become ridiculous, and the process of cutting back begins“

How do you solve a problem like predation?

Although Areeda would very likely disagree, I think you can see the same evolution in the US case law on predation. I wrote about this in the pricing chapter of Competition Law and Regulation of Technology Markets:

4.32 Current predation test In Brooke Group:⁠ [fn: Brooke Group Ltd v Brown & Williamson Tobacco Corp 509 US 209 (1993).] the Supreme Court combined the cost-based element of the Barry Wright case with the structural elements of the Matsushita case to create an overall test for predation requiring:

– prices below an appropriate measure of costs (though the court did not specify whether the appropriate measure should be average variable or average total cost); and

– a ‘dangerous probability’ of recoupment of its below-cost pricing.

4.33 Prices below costs: In applying the first of these limbs, US courts have generally adopted the Areeda Turner test for predation,⁠[fn: Philip Areeda and Donald F. Turner, ‘Predatory Pricing and Related Practices under Section 2 of the Sherman Act’, 88 Harv L Rev 697 (1975).] whereby only prices below average variable cost are seen as predatory. Although criticized,⁠[fn: See, eg, Oliver E. Williamson, ‘Predatory Pricing: A Strategic and Welfare Analysis’, 87 Yale LJ 284 (1977); Janusz A. Ordover and Robert D. Willig, ‘An Economic Definition of Predation: Pricing and Product Innovation’, 91 Yale LJ 8 (1981).] it is often seen as the best alternative to provide legal certainty and avoid enforcement errors.

4.34 Recoupment: In Brooke Group, the plaintiff (Liggett) had introduced a cut-price unbranded range of cigarettes onto the market in 1980. By 1984, unbranded cigarettes had taken 4 per cent of the market from branded cigarettes. The defendant, Brown & Williamson, had entered the cut-price segment of the market, beating the defendant’s price, sparking a price war, with allegations of predation. However, the defendant sold only 11.4 per cent of the market’s branded cigarettes. For a recoupment strategy to succeed, it would be dependent on the reactions of RJ Reynolds with about 28 per cent of the market and Philip Morris with about 40 per cent of the market. Given these facts, the Supreme Court found that recoupment was unlikely given the risks of defection during the recoupment period. Brooke Group therefore involved a relatively unconcentrated market—at least compared to simpler monopoly cases—and to argue that a predatory strategy seems doomed to failure in these circumstances seems uncontroversial.

4.35 It seems less justifiable, however, to use the case as a precedent for the difficulty of demonstrating recoupment in cases with—say—a single company with monopoly power and a market share of 80 per cent. Looking at the case against Intel, for example, put forward by the European Commission⁠ and the Federal Trade Commission,⁠ a company with a 70–80 per cent market share carried out a range of activities designed to undermine and exclude its only serious rival on the market: no concern with the need for a collusive practice or the risks of defection need reduce the likelihood of recoupment in such a case.

4.37 Recoupment as a screening device One factor that helps explain the focus of the US courts on recoupment is its role as a useful screening device, a means of quickly identifying cases that lack merit. Often it will be, ‘much easier to determine from the structure of the market that recoupment is improbable than it is to find the cost a particular producer experiences’.⁠[fn: A.A. Poultry Farms, Inc v. Rose Acre Farms, Inc 881 F.2d 1396, 1401 (7th Cir. 1989) (Easterbrook J).] This is particularly important in a system with substantial private enforcement.

4.122 EU and US law on recoupment The absence under EU law of the need for a recoupment analysis contrasts markedly with the need, under US law, for a ‘dangerous probability’ of recoupment. The gap between EU and US could be narrowed, if the original Brooke Group rationale of recoupment as a simple screen for unmeritorious cases were brought more to the fore: there seems little scope for that, however.

… particularly in high fixed costs industries

So I argue in the book that there is nothing wrong with a recoupment test in itself, but there is a risk that an overly formulaic application of the rule can lead to under-enforcement. This is particularly the case when you combine overly formulaic ideas of recoupment with overly formulaic applications of the relevant cost test…

4.123 Difficulties of the AAC/AVC test in certain cost structures In the Article 102 Enforcement Priorities Guidance,⁠1 the Commission refers to pricing below average avoidable costs in relation both to predation⁠2 and to rebates.⁠3 However, firms cannot survive on a market if they only recover their marginal costs. The Department of Justice’s now withdrawn report on s 2 of the Sherman Act recognized that marginal cost pricing would be unlikely in high fixed cost industries: ‘Depending on the size of the firm’s fixed costs, even a significant margin between price and short-run marginal cost may be insufficient to earn even a normal return.’ It did so in a section highlighting that pricing above marginal costs should not, without more, be an indicator of market power,⁠4 but the point remains true as a description of the economic realities in high fixed cost industries.

1 Guidance on the Commission’s enforcement priorities in applying Article 82 of the Treaty [now Article 102 TFEU] to abusive exclusionary conduct by dominant undertakings, 24 February 2009, 2009 OJ C 45, p 7.
2 At paragraph 64: ‘The Commission will take AAC as the appropriate starting point for assessing whether the dominant undertaking incurred or is incurring avoidable losses. If a dominant undertaking charges a price below AAC for all or part of its output, it is not recovering the costs that could have been avoided by not producing that output: it is incurring a loss that could have been avoided. Pricing below AAC will thus in most cases be viewed by the Commission as a clear indication of sacrifice.’
3 At paragraph 44: ‘W here the effective price is below AAC, as a general rule the rebate scheme is capable of foreclosing even equally efficient competitors.’
4 US Dept of Justice, ‘Competition and monopoly: single-firm conduct under section 2 of the Sherman Act’ (2008) (now withdrawn), 29: ‘Indeed, a firm should not be found to possess monopoly power simply because it prices in excess of short-run marginal cost and hence has a high price-cost margin (footnote: See Mar. 7 Hr’g Tr., supra note 6, at 13–14 (Nelson); id. at 97 (Katz); see also CARLTON & PERLOFF, supra note 8, at 93 (distinguishing monopoly from market power on the basis that more than just a competitive profit is earned when a firm with monopoly power optimally sets its price above its short-run marginal cost).)’

4.126 In high fixed and low variable cost industries, the use of average avoidable costs by itself will operate to favour dominant companies at the expense of their competitors and, ultimately, of consumers.⁠ [fn: The problems with ignoring fixed costs are discussed in H. Hovenkamp, The Antitrust Enterprise (Harvard University Press, 2006).]

When I drafted the book, I put Hovenkamp’s criticisms rather mildly. In the Antitrust Enterprise he said:

“the AVC test works very poorly in industries that have high fixed costs and relatively low marginal costs, such as airlines and public utilities, and perhaps some markets that have a large intellectual property component in their value…
[In such industries, e.g. airlines] …the AVC test amounts to a virtual license to engage in predatory pricing.”

Though I perhaps mischaracterised the criticism slightly in the footnote as Hovenkamp focusses more on the problem of ignoring opportunity costs.

So I think – although as I say I’m sure that Areeda wouldn’t – that current US law on predation provides a very good example of Areeda’s concern quoted at the beginning: “(1) An extreme case arises to which a court responds. (2) The language of the response is then applied -often mechanically, sometimes cleverly- to expand the application. With too few judges experienced enough with the subject to resist, the doctrine expands to the limits of its language, with little regard to policy. (3) Such expansions ultimately become ridiculous, and the process of cutting back begins.”

More work required

Though notwithstanding Hovenkamp’s highlighting of the problems, I’m not sure that the cutting back has yet truly begun.