The Overlapping Markets Fallacy

Introduction

Market definition has been one of the most litigated antitrust issues for decades because it so often “determines the result of the case.” As the U.S. Department of Justice Antitrust Division put it during the Obama administration, “[f]requently, the government alleges narrow markets, the defendants describe broad markets, and the court must choose between the competing approaches.” Courts must choose between these competing approaches because, outside the per se context, “determination of the relevant market is a necessary predicate to a finding of a violation” of the antitrust laws.

Apparently no longer. In a surprising about-face, the DOJ and Federal Trade Commission now assert that courts do not need to choose between competing market definitions advanced in litigation. For example, in one recent case the FTC argued that “plaintiffs need not disprove alternative relevant markets proposed by the Defendant” because “a larger market in no way precludes the existence of a submarket . . . for antitrust purposes.” Or, to put it in economic terms, the agencies assert that it could be true both that a defendant is competitively constrained by several firms in a broader market and by only a subset of those firms in a narrower submarket. A few scholars have taken similar positions. In the past few years, both agencies have tried this tack, with mixed results. Although some courts have firmly rejected agency efforts to “have its cake and eat it too,” a few district courts have followed the suggestion that a court can avoid choosing among the litigants’ competing approaches by endorsing them both.

The new Merger Guidelines double down on this claim. They assert that “[m]ultiple overlapping markets can be appropriately defined relevant markets” and provide an example involving nested product markets from “food” all the way down to “premium low-fat chocolate chip cookies.” The Guidelines also assert that this position reflects “applicable legal precedent” and that “legal holdings reflecting the Supreme Court’s interpretation of a statute apply unless subsequently modified.” To support the overlapping markets claim, the Guidelines cite the “submarkets” dictum in the 1962 case Brown Shoe Co. v. United States. Thus, the agencies appear to view the two terms as synonymous, and we follow suit here.

As we explain in this article, overlapping markets are not the law today, nor have they been for many years. Rather, they are every bit as erroneous as the famous Cellophane fallacy. Although the agencies cite a passage from Brown Shoe in support, that discussion is a dictum; indeed, the Court did not define any overlapping markets in that case. And in subsequent cases the Court first avoided and then repudiated submarkets. Lower courts initially approached the issue inconsistently but, since the mid-1980s, have typically either rejected overlapping markets outright or avoided the issue. The Areeda and Hovenkamp treatise accurately sums up this history when it defines the term “submarket” as a concept previously “used to identify artificially narrow groupings of sales on the basis of noneconomic criteria having little to do with the ability to raise price above cost.” Ohio v. American Express Co. (Amex) encapsulated much of this history; there, the district court rejected DOJ’s proposed travel and entertainment submarket, the DOJ abandoned the overlapping submarket before the court of appeals in favor of “separate” narrow markets, and the Supreme Court held that one and only one relevant market can be defined in two-sided transaction markets.

We argue that the legal rule announced in Amex, which we dub the “One Market per Transaction” (OMT) Rule, actually applies to a far broader range of cases. As we define it, the OMT Rule requires that a factfinder assign a particular transaction of a particular product to no more than one relevant antitrust market. To riff on an example in the Merger Guidelines, if a court finds that wholesale sales of chocolate chip cookies to grocery stores compete in a relevant market for wholesale sales of desserts to grocery stores in the United States, then the OMT Rule would prohibit those sales from also being included in another relevant antitrust market for wholesale sales of chocolate chip cookies to grocery stores in the city of Baltimore.

Thus, the market definition holding in Amex articulated a standard already routinely applied, but not well understood, by lower courts. The economic intuition of the OMT Rule is simple: If a defendant’s sales to a particular customer or customer group are constrained by four competitors, then that is the market, and no smaller market can be defined around only two of those four competitors, nor a broader market around eight competitors. Or to borrow other language from Amex, when assessing the competitive effects of challenged conduct, a court “must first define the relevant market” and identify the firms that compete in it.

Therefore, under the OMT Rule, a given transaction—meaning a particular sale of a specific product or service—can be allocated to no more than one relevant antitrust market. As in Brown Shoe itself, where the Court defined separate markets for men’s, women’s, and children’s shoes, product sales may still be grouped into multiple non-overlapping relevant markets. The OMT Rule is also flexible enough to account for a range of other circumstances, including entry, exit, technological change, and price discrimination. Applying the OMT Rule is fairly straightforward in both conduct and merger cases.

Going forward, courts should formally recognize and apply the OMT Rule. Doing so would help courts simplify the market definition exercise, clarify judicial reasoning, and reduce the risk of error. The Guidelines should be revised to delete the inaccurate description of overlapping markets and, in future iterations, should adopt and apply the OMT Rule.

The OMT Rule has several implications for antitrust enforcement. First and foremost, it would reinforce and clarify the existing requirement that courts resolve disputes regarding the scope of the relevant market. Second, the OMT Rule may require litigants to approach market definition with greater rigor and consistency, as otherwise they may be tempted to use overlapping markets to bring otherwise contradictory claims, such as a monopolization claim in a narrow (sub)market and a nascent or potential competition claim in a broader one. This dynamic may be greatest when firms offer a wide range of products and services, such as is often the case with large technology (or “Big Tech”) companies. Third, it may further increase the frequency with which enforcers carve out narrower but non-overlapping markets.

The remainder of this article proceeds chronologically. Part I recounts the history of the submarkets doctrine and overlapping markets, from Brown Shoe through the agencies’ latest Guidelines. Part II describes the Amex case, which provides a clear explanation of why courts, scholars, and enforcers have heretofore rejected overlapping markets as conceptually “unsound.” Part III explains why courts can and should formally adopt and apply the OMT Rule today. Part IV identifies some potential implementation issues and enforcement implications.

Continue reading the full text of this article in PDF format.