Gerrymandering Redux: The Relevant Market under the Draft Merger Guidelines

On July 19, 2023, the federal antitrust agencies—the US Department of Justice (DOJ) and the Federal Trade Commission (FTC)—released draft Merger Guidelines (dMGs) for public comment. Once these guidelines are finalized, they will be the official statement of US antitrust enforcement policy toward corporate mergers. This policy brief addresses how government policy delineates the relevant market in which the likely effects of proposed mergers are evaluated. It argues that the dMGs abandon critical principles, which signals a return to the gerrymandering that characterized merger cases in the 1960s.

The Invention and Centrality of the Relevant Market

It is commonly supposed that the concept of the relevant market is as old as antitrust law, but it originated in the 1948 Columbia Steel decision. The case was a merger challenge under an antitrust law not specifically directed at mergers. The decision referred to “the relevant competitive market” and observed that “it is first necessary to delimit the market in which the concerns compete and then determine the extent to which the concerns are in competition in that market.”

The decision went against the government, and that motivated Congress to enact an antitrust law that specifically targeted mergers. In 1950, Congress amended section 7 of the 1914 Clayton Act to prohibit a merger or acquisition “where in any line of commerce in any section of the country, the effect of such acquisition may be substantially to lessen competition, or to tend to create a monopoly.”

Before interpreting the new antimerger law, the Supreme Court gave meaning to the concept of the relevant market in two other cases. In a 1953 decision, the court astutely declared that “for every product, substitutes exist. But a relevant market cannot meaningfully encompass that infinite range. The circle must be drawn narrowly to exclude any other product to which, within reasonable variations in price, only a limited number of buyers will turn.”

In 1956, the court had to decide whether cellophane traded in a distinct relevant market. It properly observed that the relevant “market is composed of products that have reasonable interchangeability for the purposes for which they are produced—price, use and qualities considered” and held that “cellophane’s interchangeability with the other materials mentioned suffices to make it a part of this flexible packaging material market.”

An early decision under the amended section 7 held that

the essential issues which the Court is called upon to determine . . . are: the line or lines of commerce and the section or sections of the country in which the effects of the merger may be felt—in other words, the relevant market with respect to both products and geographic areas—and whether there is a reasonable probability that the merger may substantially lessen competition or tend to create a monopoly within the relevant market.

When the FTC contended that “market definition is not necessary in a § 7 case” a half-century later, the court held that its contention was “in contravention of the statute itself.”

The relevant market has been the main issue in merger litigation because the market determined the market shares, which determined legality. In its first case interpreting the amended section 7, the US Supreme Court declared that “the market share which companies may control by merging is one of the most important factors to be considered when determining the probable effects of the combination on effective competition in the relevant market.”

The centrality of the relevant market was cemented a year later when the court declared that

a merger which produces a firm controlling an undue percentage share of the relevant market, and results in a significant increase in the concentration of firms in that market, is so inherently likely to lessen competition substantially that it must be enjoined in the absence of evidence clearly showing that the merger is not likely to have such anticompetitive effects.

Six decades on, in 2023, the relevant market is no less central to merger assessment under antitrust law. To delineate the relevant market is to identify, in terms of products and places, the locus of active competitive forces central to assessing a merger or competitive practice. As the Supreme Court observed, the relevant market in an antitrust case is “the area of effective competition.” Competitive forces placed outside the relevant market are thereby deemed both passive and secondary. This separation of competitive forces is essential in predicting the likely competitive effects of mergers.

The legal concept of  “the relevant market” is closely related to the economic concept of “the market,” the focus of most work in microeconomics. Market-level economics, termed partial equilibrium analysis, places the analytic spotlight on one narrow sector of the economy, holding all else constant. The most familiar application is supply and demand, in which the spotlighted sector is termed the “market” and outside prices are held constant.

Partial equilibrium analysis posits a narrow scope of active competitive interaction. Prices outside the scope of the analysis are assumed to be constant; that is, competition from the outside is treated as a passive force. The best choice, generally, is to make the scope of the model congruent with the relevant market. Positing a broader interaction injects complexity into the analysis without offering insight; positing a narrower interaction both misspecifies the true competitive process and risks materially distorting the analytic conclusions.

The Dark Ages of Market Delineation

The 1960s were for market delineation much like the early 16th century was for anatomy (before Vesalius) and astronomy (before Copernicus). In Brown Shoe, the Supreme Court avoided tension with the cellophane case by asserting that a relevant market defined on the basis of reasonable interchangeability could contain “well-defined submarkets” that “constitute product markets for antitrust purposes.”

The court went on to declare that “the boundaries of such a submarket may be determined by examining such practical indicia as industry or public recognition of the submarket as a separate economic entity, the product’s peculiar characteristics and uses, unique production facilities, distinct customers, distinct prices, sensitivity to price changes, and specialized vendors.”

Over the next few years, the Supreme Court accepted any market definition the government contrived. When two banks proposed to merge, the relevant market was the “cluster of products … and services denoted by the term ‘commercial banking.’” It did not matter that the products and services were not interchangeable with each other or that nonbanks provided most of them.

When a manufacturer of metal containers (cans) merged with a manufacturer of glass containers (bottles and jars), the relevant market was defined as “the glass and metal container industries” to “conform to competitive reality.” It sufficed that there was, or might be, some competition in some uses. Dissenting Justice Harlan insisted that this market was “unrelated to any reality whatsoever.”

The Supreme Court’s disdain for principle was clear in two decisions of June 13, 1966. In a case involving alarm services, the court defined the relevant market as “accredited central station [alarm] service.” (Central station alarms transmit a signal to a manned monitoring station.) The court did so although burglar and fire alarms were not interchangeable and although alarms with and without central station monitoring were interchangeable. Dissenting Justice Fortas lambasted the “gerrymandered market definition.”

In a beer case, the Supreme Court could have accepted the government’s evidence for a three-state geographic market. Instead, the court declared that the relevant market need not be defined “by metes and bounds” and that “the failure of the Government to prove by an army of expert witnesses what constitutes a relevant ‘economic’ or ‘geographic’ market is not an adequate ground on which to dismiss” the case. Dissenting Justice Harlan observed that “the possibilities for ‘gerrymandering’ are limitless.”

The Supreme Court has had little to say about market delineation since 1966 and has not considered the merits of a merger case since 1975. Since the Supreme Court has had no occasion to overrule the 1960s market delineation precedents, they technically remain good law.

Bringing Order To Chaos

After the election of Ronald Reagan, Stanford law professor William Baxter took charge of the DOJ’s Antitrust Division. Issuing new Merger Guidelines (MGs) was a priority, and Baxter did so on June 14, 1982. He pledged not to gerrymander markets, and the 1982 MGs asserted four key market delineation principles, all of which have been reasserted by subsequent guidelines until now. The 1982 MGs tried to bury Brown Shoe’s “practical indicia.” Like the Scarecrow in The Wizard of Oz, the “practical indicia” lacked a brain and operated as a checklist.

A brain was supplied by the hypothetical monopolist test (HMT), which holds that a candidate market is in fact a market only if a hypothetical monopolist controlling it would raise prices significantly. The HMT was developed by the DOJ in 1978 to analyze competition in the coal industry. The test explained why it made sense to define relatively narrow coal markets that focused on points of production. It has been universally embraced by courts.

As initially formulated, the HMT defined only markets focused on suppliers. The 1982 MGs, however, also articulated the concept of price discrimination markets, which were focused on consumers. The second market delineation principle articulated by the 1982 MGs was that the geographic dimensions of markets are delineated around points of production, but when price discrimination is possible, additional relevant markets can be delineated, and they are points of consumption.

In a relevant market focused on suppliers, the competitors are those with production facilities in the defined area, and their market shares are determined by the capacities or outputs of those facilities. In a relevant market focused on customers, the process of identifying competitors and assigning them shares is more complex. A competitor not selling in the defined area nevertheless is a significant competitor there if it would divert product there in the event that prices in the area rose.

The 1982 MGs prevented gerrymandering by combining the first two principles with two others to generate a unique relevant market for every starting point. The shape of the relevant market was controlled by the circle principle, which holds that if product (or area) A is included in a market, then product (or area) B must be included if B is at least as good a substitute for the starting product as A. The minimum size of the circle is controlled by the HMT, but, as a general rule, an infinite number of concentric circles pass the HMT. The smallest market principle holds that the relevant market generally is the smallest circle that passes the HMT.

In the 1982 MGs, the application of the HMT was explained, for pedagogic purposes, as an iterative procedure. The initial candidate market for any merger consisted of a product of one of the merging firms and a place at which this product was produced. A candidate market sequence was formed by adding additional product-place pairs one at a time; each new pair was the next-best substitute for the starting product and place. The process stopped when a candidate market passed the HMT.

The 1982 MGs were revised slightly in 1984. Among the amendments was a concise statement of the iterative process:

The Department will begin with each product (narrowly defined) produced or sold by each merging firm and ask what would happen if a hypothetical monopolist of that product imposed a “small but significant and nontransitory” increase in price. If the price increase would cause so many buyers to shift to other products that a hypothetical monopolist would not find it profitable to impose such an increase in price, then the Department will add to the product group the product that is the next-best substitute for the merging firm’s product and ask the same question again. This process will continue until a group of products is identified for which a hypothetical monopolist could profitably impose a “small but significant and nontransitory” increase in price. The Department generally will consider the relevant product market to be the smallest group of products that satisfies this test.

A common initial reaction to the HMT was that it would lead to excessively broad relevant markets. This reaction had no basis in economics, and experience proved it wrong. A typical argument made by merging firms is that the relevant market is much broader than is alleged by the government, and courts have cited the HMT and the smallest market principle in rejecting such contentions.

The 1992 Horizontal Merger Guidelines, released jointly by the DOJ and the FTC, made an important clarification about how the HMT works. The hypothetical monopolist controls multiple prices, and it could maximize its profits by raising each price by a different amount. The 1992 MGs clarified that the HMT is passed only if the hypothetical monopolist would significantly increase the price of one of the merging firms’ products.

Until the dMGs were released, the DOJ and the FTC operated under the 2010 Horizontal Merger Guidelines. These guidelines dropped the pedagogic device of the iterative approach but made no other material changes. They stated the HMT as follows:

The test requires that a hypothetical profit-maximizing firm, not subject to price regulation, that was the only present and future seller of those products (“hypothetical monopolist”) likely would impose at least a small but significant and non-transitory increase in price (“SSNIP”) on at least one product in the market, including at least one product sold by one of the merging firms.

They explicitly stated the circle principle:

When applying the hypothetical monopolist test to define a market around a product offered by one of the merging firms, if the market includes a second product, the Agencies will normally also include a third product if that third product is a closer substitute for the first product than is the second product.

And they stated the smallest market principle: “Because the relative competitive significance of more distant substitutes is apt to be overstated by their share of sales, when the Agencies rely on market shares and concentration, they usually do so in the smallest relevant market satisfying the hypothetical monopolist test.”

Back To the Dark Ages

The dMGs modify the formal statement of the HMT from the 2010 Horizontal Merger Guidelines in several ways:

The HMT asks whether a hypothetical profit-maximizing firm, not prevented by regulation from worsening terms, that was the only present and future seller of a group of products (“hypothetical monopolist”) likely would undertake at least a small but significant and non-transitory increase in price (“SSNIP”) or other worsening of terms (“SSNIPT”) for at least one product in the group.

For present purposes, the critical alteration is in the final prepositional phrase. Under the dMGs, the HMT can be passed even though the hypothetical monopolist would not increase the price of anything sold by the merging firms. This defeats the entire purpose of the HMT because it means that merging every incumbent in a market passing the HMT could have no anticompetitive effect that has anything to do with the merger.

The alteration would be of little consequence if the circle principle were observed, because the hypothetical monopolist nearly always raises the price of the starting product at least as much as it raises any other price. But there is no hint of the circle principle in the dMGs, and when the circle  principle is not observed, the alteration telegraphs the intention to delineate markets that pass the HMT because of the inclusion of products that are not close substitutes for the products of merging firms. If the HMT were made to work that way, it would become arbitrary and capricious.

The dMGs also affirmatively disclaim the smallest market principle. The text asserts, “Market definition ensures that antitrust markets are sufficiently broad, but it does not lead to a single relevant market.” And footnote 86 adds, “Multiple overlapping markets can be appropriately defined relevant markets. For example, a merger to monopoly for food worldwide would lessen competition in well-defined relevant markets for, among others, food, baked goods, cookies, low-fat cookies, and premium low-fat chocolate chip cookies.”

But each of these product groups cannot be “the area of effective competition.” The definite article makes that impossible. The dMGs allude to the 1953 dictum that the relevant market excludes the products and areas “to which, within reasonable variations in price, only a limited number of buyers will turn,” and yet the DOJ and the FTC insist that all food could be a relevant market.

It is pointless for the agencies to explain how they would analyze the most far-fetched hypothetical imaginable. And if two cookie companies were to merge, the agencies obviously would not define the relevant market as all food or all baked goods, but the dMGs do not say what they would do. And if the cookie merger got to court, the agencies would need some principle to counter the defense contention that the relevant market is all baked goods.

The dMGs herald a return to the dark ages of merger litigation, with litigants trying to maximize or minimize the apparent significance of the pending merger by gerrymandering the relevant market. The Supreme Court once stood willing to go along with whatever the government proposed, but trial courts today are not so obliging. The agencies have eschewed principles to which trial judges were highly receptive.

To eliminate any doubt that the agencies were choosing to go back to the dark ages, the dMGs affirmatively state that “a relevant market can be identified from evidence on observed market characteristics (‘practical indicia’),” then quote the full list of indicia from Brown Shoe. Indeed, the dMGs mention the “practical indicia” before they mention the HMT. And the dMGs offer nothing to make the “practical indicia” more than a meaningless checklist.

Magic Market Delineation

The “practical indicia” and the HMT are just two of four “tools” the dMGs state that the agencies use “to demonstrate the validity of a candidate antitrust market.” The first “tool” listed is this:

Direct evidence of substantial competition between the merging parties can demonstrate that a relevant market exists in which the merger may substantially lessen competition and can be sufficient to identify the line of commerce and section of the country affected by a merger, even if the precise metes and bounds of the market are not specified.

If the merging parties competed significantly only with each other, that would be highly significant, but that is not what the dMGs posit. Nor do the dMGs posit that the merging parties produce a homogeneous product for which there is no close substitute. The discussion of “substantial competition between the merging parties” elsewhere in the dMGs indicates that the agencies have differentiated products in mind. But substantial competition between two differentiated products supports no logical inferences about the nature or extent of competition with other differentiated products. Anything is possible.

Either the agencies are claiming that they can perform magic or they are concealing a plan to declare that the merging parties are the only competitors in some relevant market, the dimensions of which are immaterial. In extraordinary circumstances, such a declaration might be sensible, but it would be laughable in the general run of cases presenting “substantial competition between the merging parties.”

The second tool listed is this:

Direct evidence of the exercise of market power can demonstrate a relevant market in which that power exists. This evidence can be valuable when assessing the risk that a dominant position may be entrenched, maintained, or extended, since the same evidence identifies market power and the rough contours of the relevant market.

The agencies might envision that an actual monopolist actually raising prices effectively performs the HMT. While this is true in theory, translating this idea into practice presents difficulties. In literally every case, direct evidence establishes that both merging parties charge prices exceeding short-run marginal cost by at least the threshold used in applying the HMT. They are exercising significant market power, but their significant price-cost margins provide no clue about even the “rough contours of the relevant market.”

The description of the second tool suggests that it is applicable only if one of the merging parties has a dominant position. But a dominant market position requires a dominant market share, which presupposes a relevant market.

Generalizing the HMT

The dMGs change the HMT from a test that asks whether a hypothetical monopolist would raise price significantly to a test that asks whether a hypothetical monopolist would worsen “terms along any dimension of competition, including price (SSNIP), but also other terms (broadly defined) such as quality, service, capacity investment, choice of product variety or features, or innovative effort.” This generalization achieves nothing and is impractical.

The HMT is a straightforward application of the economic theory of monopoly. To the extent that it is possible to approximate the demand curve faced by a hypothetical monopolist, the theory of monopoly permits the calculation of a monopoly price that can then be compared with the prevailing price. That comparison is the HMT.

Price is not the only dimension of competition, but “merger enforcement, like other areas of antitrust, is directed at market power,” and market power is power over price. During four decades of applying the HMT, no case has come to light in which the test yielded the wrong answer because it focused exclusively on price. Even when current competition does not focus on price, market power created by a merger would be exercised by charging a higher price for something.

When innovation drives competition, the concern is with the creation and exercise of market power associated with the goods and services that are the fruits of the innovation. The delineation of the relevant market should focus on the potential to exercise market power over goods or services. When rivals compete to create next-generation products, the delineation of the relevant market should focus on customer demand, just as with existing products.

In zero-price markets, antitrust analysis should follow the money. Any product or service given away is supported by an ancillary revenue source. This phenomenon was observed a century ago with radio, which quickly hit on advertising as the revenue source. The proper application of the HMT recognizes that market power must be exercised through the ancillary revenue source and that exercising market power feeds back to the free service. This feedback can be positive, of course, because charging more for advertising means selling less and reducing the nonmonetary cost advertising imposes on users of the free service.

In generalizing the HMT, the DOJ and the FTC might intend to omit the market-power analysis at the heart of the HMT. They might just posit an arbitrary change in some dimension of competition and assert that customers will not switch in response. But such a demonstration should be deemed insufficient to carry the burden of proof.


Economists are constantly refining their tools for evaluating the likely competitive effects of mergers, but market shares and market concentration continue to play significant roles, especially in litigation. Moreover, economic modeling demands the separation between active and passive competitive forces that is done in delineating the relevant market. It is imperative that sound principles govern market delineation, yet the DOJ and the FTC are abandoning the principles they long applied. Courts should just say no.

Citations and endnotes are not included in the web version of this product. For complete citations and endnotes, please refer to the downloadable PDF at the top of the webpage.