Features v. Bugs: Intel and the Relationship Between Sections 2 and 5

Cite this Article
Joshua D. Wright, Features v. Bugs: Intel and the Relationship Between Sections 2 and 5, Truth on the Market (December 16, 2009), https://truthonthemarket.com/2009/12/16/features-v-bugs-intel-and-the-relationship-between-sections-2-and-5/

There will be much to say about the Federal Trade Commission’s Intel complaint in the coming months.  And we’ve said quite a bit already.  But having just read the complaint and the statements from Chairman Leibowitz and Commissioner Rosch discussing the various rationales for making Section 5 the primary hook for this case, I wanted to share two thoughts about defenses for the move that appear in those statements.

The first comes from the joint statement:

Despite the long history of Section 5, until recently the Commission has not pursued free-standing unfair method of competition claims outside of the most well accepted areas, partly because the antitrust laws themselves have in the past proved
flexible and capable of reaching most anticompetitive conduct.  However, concern over class actions, treble damages awards, and costly jury trials have caused many courts in recent decades to limit the reach of antitrust.  The result has been that some conduct harmful to consumers may be given a “free pass” under antitrust jurisprudence, not because the conduct is benign but out of a fear that the harm might be outweighed by the collateral consequences created by private enforcement.  For this reason, we have seen an increasing amount of potentially anticompetitive conduct that is not easily reached under the antitrust laws, and it is more important than ever that the Commission actively consider whether it may be appropriate to exercise its full Congressional authority under Section 5.

Is dissatisfaction with the current state of Section 2 jurisprudence a good enough reason to warrant application of Section 5?  That’s a tricky question.   That Section 5 might apply in some situations where Section 2 does not is a perfectly reasonable proposition.  But I have a few problems with the use of this rationale here.  One is that the the Leibowitz/ Rosch story that the reduction in scope of Section 2 has nothing to do with whether conduct is anticompetitive but rather that it is all about immunizing conduct that is known to be harmful because of some sort of aversion to private enforcement is simply wrong.  And it’s wrong in an important way.  The fear that emerges out of Credit Suisse, Trinko, Brooke Group and Linkline is not merely that private actions are bad, but rather that error costs are a real problem.  In other words, the fear is that: (1) it is very difficult to determine in the first instance whether would-be exclusionary conduct is pro-competitive, anti-competitive, or competitively neutral, (2) this raises the inevitability of Type I and Type II errors, (3) a la Easterbrook, the former should be of greater concern because they create more substantial social costs (“error costs”).  Given (1)-(3), the Supreme Court has adopted liability rules that reflect the realities of the economic technology available to distinguish anticompetitive single firm conduct from pro-competitive conduct, and the asymmetrical costs of errors.

In sum, the changes in Section 2 law have not been merely out of fear of private enforcement–but rather out of fear that private enforcement in the face of Type I errors render those errors much more serious.  But make no mistake–the error cost approach is one that is concerned precisely with adopting liability rules that maximize consumer welfare.  To say otherwise is a poor description of the case law and the underlying logic.  The view implicitly adopted by the Commissioners that the antitrust laws are failing if they do not reach “most anticompetitive conduct” simply contradicts the approach taken by the Supreme Court.  The gap between actual Supreme Court interpretation of Section 2 and a hypothetical body of antitrust law that would reach all anticompetitive conduct is not one that is accidental or the product of “mere technicality.”  This is a conscious choice by the Supreme Court, based on error cost considerations that are well accepted and mainstream antitrust analysis.  This use of Section 5 cannot be said to be “gap filling.”  Instead, this invites application of Section 5 unhinged from the Section 2 principles entirely.  This, in my view, is a wrongheaded approach that is almost certain to strip away the protections for consumers embedded in the error cost approach incorporated into Section 2.  Thus, my view is that arguments that one does not like Trinko or Credit Suisse cannot and should not militate in favor of a broader scope for Section 5.

By the way, it is also worth noting that there is a striking tension between the expressed view that fear of private litigation warrants a shift toward Section 5 and the Commission’s apparent willingness to file a complaint with regard to conduct involving Nvidia for which it has not yet taken discovery.  In other words, Commissioner Rosch argues that the Commission should pursue the Intel complaint under Section 5 alone because it will avoid the social costs associated with follow-on private enforcement but should also reserve the right to proceed as if the Commission itself were a private plaintiff filing a complaint without discovery and surely without knowing whether the allegations over which it has no facts are indeed in the public interest.

Second, Commissioner Rosch’s separate statement makes a similar error in one of the four factors he argues militates in favor of Section 5 over Section 2.  Specifically, Commissioner Rosch asserts that Section 5 as a stand alone violation makes more sense when harm to competition is difficult to distinguish from harm to competitors:

Under those unique circumstances, the oft-repeated admonition that the Sherman and Clayton Acts protect competition, not competitors, and the federal courts’ attendant disinclination to protect competitors in cases brought under those statutes, do not fit well. If the firm with monopoly or near-monopoly power (here, allegedly Intel) engages in an exclusionary and unjustifiable course of conduct that hurts its only competitor in the CPU markets (here, allegedly AMD) or its only two competitors in the computer graphics product markets (here, allegedly AMD and Nvidia), given the uncommonly high entry barriers, that exclusionary conduct harms competition too, by inhibiting those rivals from constraining the exercise of monopoly power.

Again, Commissioner Rosch implicitly adopts the view that there is no “consumer welfare” based content to the old admonition about protecting competition rather than consumers.  More generally, however, the argument just doesn’t make economic sense.  If what Rosch means is that in markets with only one competitor harm to that competitor is more likely to lead to harm to competition (or is even synonymous with harm to competition), then that makes Section 2 more appropriate!  In other words, in such cases there will always be harm to competition so there is no need to expand Section 5 to reach the conduct.  Instead, Commissioner Rosch apparently is arguing that plaintiffs should be able to avoid the rigorous proof standards of Section 2 in cases where it is really hard to distinguish harm to individual competitors generated from vigorous competition from harm to the competitive process.  This turns antitrust on its head.  It is not gap-filling. And it is not merely extending Section 5 to situations of anti-competitive conduct known to be outside the scope of Section 2.

Rather, and this is the important part, this is the use of Section 5 to evade the protections afforded consumers by Section 2.  One of those, and perhaps the most important, is that plaintiffs be forced to demonstrate that (given the lack of empirical support for many of the vertical exclusionary theories in the literature) the conduct at issue has actually harmed consumers.  Using Section 5 to evade this fundamental feature of Section 2 may make the litigation easier to win, but it certainly doesn’t make it in the public interest.  The fact that in duopolies it might be more difficult to show competitive harm is no excuse to lighten the plaintiff’s burden by throwing Section 2 out in favor of Section 5.  Commissioner Rosch seems to be adopting the position that in competition in markets with few firms, we can presume that harm to competitors is a sufficient condition for harm to competition.

This is the European approach.   At best.  Despite attempts to suggest there is nothing to see here, there really is.  This is not a subtle shift in single firm antitrust jurisprudence.   This is a debate over whether the error  cost protections in Section 2 law are features or bugs, with the Commission taking the (indefensible) latter view.

Note that in May 2009 I wrote the following:

I’m quoted in the WSJ as saying that I believe it is much more likely that the US gets involved in the Intel litigation than was the case two weeks ago.  Its hard to avoid that conclusion after reading the combination of statements from the FTC on the repudiation of the Section 2 Report, the new life of Section 5, as well as the competitive pressures placed on that agency from the DOJ’s new agenda and the EU fines.

The problem is that the content of the Section 2 Report was not just policy statements from the Bush administration political appointees about what the Section 2 should be.  It was a serious project with engagement from DOJ and FTC appointees, staffers, the academic community, and business representatives to summarize the existing law and existing evidence as well as generate some guidance on best practices where available.  Turns out that with two years to work on the project and that breadth of resources and diversity of viewpoints, the Section 2 Report really does accurately state the law with respect to exclusive dealing, predatory pricing, loyalty rebates, and such.  And that law isn’t going anywhere.  Perhaps the mission of the new DOJ and FTC will be to change the law?  Or perhaps the FTC will avoid the unfavorable Section 2 law by substituting Section 5 for cases like Intel where they are unlikely to win under a Section 2 theory.  But the Supreme Court and the federal case law under Section 2 remain substantial obstacles to convergence that extends beyond the hallways of the agencies.

So far so good.  I’ll stick to my guns.  The FTC will lose under any elements of the case brought under Section 2.  To the extent that an appropriate interpretation of the requirements of Section 5 are at least informed by the requirements of Section 2, I think the FTC will also ultimately lose on its Section 5 claims (on appeal, of course).  But the Section 5 case is obviously much harder to predict.  Two things are certain.  The FTC will expend millions of dollars in resources litigating this case with Intel.  Some pro-competitive conduct will be chilled as a result of the action.  Taken in light of traditional market performance indicators like price and output in the microprocessor market, as well as AMD’s financial performance during the time period of the alleged misconduct, today’s announcement is neither a bright day for the FTC, nor consumers.