This paper,available here , is not technically about merger control, but it is as relevant here as in any other competition topic – and it fits nicely with wider discussions of market power and market entry, which, as we have seen in past weeks, are common in merger control.

While the usefulness of, and methodologies concerning market definition would seem to be well established, in practice both are actively questioned. Some have even argued that market definition is unnecessary in competition law. While this argument is not new, Louis Kaplow has recently advanced this thesis with a particularly pointed argument that: (1) market definition serves no role except to produce market shares, (2) market shares are poor measures of market power, and (3) antitrust would be better served by ignoring market shares and trying to assess market power from estimates of residual-demand curves and the like instead.

The goal of this paper is to trace the internal logic of market definition, and to argue that the logic inherent to market definition narrows and focuses the competition inquiry. There are at least three good reasons for courts and practitioners to properly define and interpret relevant markets in antitrust cases. First, the claim that market definition can be entirely replaced by things like econometric estimates of residual demand curves is disputed, to say the least. Second, regardless of the academic debate, courts have long relied on market definition in antitrust cases. So long as binding precedent continues to invoke relevant markets, it will remain important to define them. Third, market definition plays an important practical role in antitrust analysis. In the review of merger notifications, for example, market definition helps clarify the analysis by imposing analytic discipline, providing a logical way to organise information, and focusing the scope of further competitive effects analysis.  Market definition also supports structural inferences about competitive effects and provides important context for other evidentiary considerations (such as the possibility of entry or exit).

However, to determine when and how market definition is useful, one needs first to establish when the use of market definition is unhelpful, if not outright fallacious. The paper is structured around these potential fallacies.

Section I deals with the natural market fallacy.

The natural market fallacy is the mistaken belief that the boundaries of relevant markets should conform to lay intuition, conventional language or mere factual observation. However, economic markets are not tangible things, but analytical concepts, and it is mistaken to reify them. Though the mapping is admittedly crude, it is useful to distinguish two types of natural market concepts: those that define markets by the observable characteristics of products and producers, and those that define markets by observed substitutability. Both concepts are problematic.

The first approach equates market definition with the identification of observable product characteristics and lay recognition of industry lines. At a high level, these efforts seek to find characteristics of products and producers that will indicate how a market should be categorised. It is not difficult to mould the peculiar product characteristics test into a rough approximation of how substitutable one product may be for another, and courts and scholars have similarly reinterpreted the practical indicia as factors relevant to assessing closeness of competition. In practice, however, most cases adopting this test focused instead on intuitive delimitations based on product characteristics, in a way that has no basis on economics.

At the same time, standards for defining relevant markets have always included various approximations to the economic idea of substitutability. Examples are efforts to identify markets by reference to the cross-elasticity of demand or the reasonable interchangeability of use between products. At a high level, the idea is that, somewhere in the field of increasingly distant competing products, substitutability becomes too weak to warrant inclusion in the relevant market. There are admirable qualities to this substitutability-based approach to defining markets, but it too rests on what is an essentially a naturalistic concept of a market. The cross-elasticity and reasonable interchangeability standards articulate the same approach to market definition: identify the boundaries of a relevant market by drawing a line where the substitutability of products and producers becomes too attenuated. However, neither standard even attempts to articulate is where the cut-off lies: how small must be the cross-elasticity of demand, and how poor must be the interchangeability of use, for the edge of a relevant market to be reached. It is at this point that the naturalistic foundations of substitutability-based markets become clear. This approach embraces the naturalistic fallacy of seeking some market that exists independent of any given inquiry or investigation— a group of products or producers that are each other’s closest substitutes in some absolute or fixed sense of the term. However, perfect gaps in competition are rare; a given product usually faces competition from products of varying degrees of substitutability at varying price points.

The problem with the naturalistic fallacy is that markets are not “definite economic [entities], the existence of which has merely to be recognized by the investigator”, but analytical tools which “may and should be used with all degrees of inclusiveness” in the process of studying a problem. From this perspective, markets are simply a lens for focusing analysis on the problem at hand—and thus the market can and must always be moulded to reflect whatever is the problem at hand. There are no right or wrong markets in this approach, only markets of various utility in studying a particular problem or question.

This suggests two fundamental properties of antitrust market definition. First, there is no such thing as an economically interesting natural antitrust market; there is not any real market waiting to be found. Market definitions are not true (false), correct (incorrect), or real (unreal), but merely appropriate (inappropriate). Appropriateness is a relative or contextual quality; and the context in market definition is the theory of harm. It follows that, as purely analytical constructs, the definition of relevant markets will always depend upon the nature of the problem to be analysed.

Section II considers the independent market fallacy.

The independent market fallacy is the common misconception that relevant markets exist independently of a theory of anticompetitive injury. This idea long applied in merger review, where market definition was seen as a discrete, initial step in the process. Instead, market definition is useful mainly as providing an outer bound on the concerns sufficient to bring about the injury in question: a market defined as a group of transactions in which the contemplated injury could occur, at least under assumptions favourable to the theory of harm. The Hypothetical Monopolist Test (HMT) provides a good example of this. The typical articulation of the HMT (focusing on a 5% price increase for a non-transitory period of time) can only produce relevant markets in which coordination among the set of producers could result in something like a near-term price increase of 5%. Whether the merger would bring about this result is a question requiring further analysis of market structure and incentives. The relevant market identifies the suspect participants and helps contextualise the further analysis needed to answer this dispositive question.

In other words, an analytically useful relevant market cannot be defined except by reference to some potential anticompetitive injury: the relevant market must be customised to the theory of harm. For example, when calculating the HMT, the base price and the hypothesised price increase should be specified to reflect the specifics of any competitive concern under investigation. On what basis should the hypothetical price increase should be chosen in a proper application of the HMT? On the basis of the theory of anticompetitive injury in question. At a minimum, this means that we ought never to choose a hypothetical price increase larger than what we worry might be the actual competitive effect. If the theory of anticompetitive injury is a large price increase, maybe 10–20%, then a relevant market defined by the HMT with a 5% hypothesized price increase is likewise inapt to the problem at hand. The same goes for the base price. Is the proper HMT base price in merger cases the current price—even if it reflects the ongoing exercise of market power—or is it an estimate of the competitive price? The answer is—once again—that the proper choice of base price depends on the theory of harm. If the concern is that a merger will allow the remaining firms to elevate price above the current level, then the current price is the appropriate baseline against which to define the market. However, if the concern is that a merger will stabilise or entrench already existing price elevation, then some measure of competitive pricing but for the ongoing cause of price elevation is the appropriate baseline to use in defining the relevant market.

Section III deals with the single market fallacy.

The single market fallacy is the pervasive mistake of assuming that antitrust cases only involve a single relevant market. Few ideas in antitrust are as settled as the tacit assumption that there can be only a single relevant market for a given inquiry. This is reflected in numerous practices. Often, antitrust trials play out with the plaintiff and defendant each putting forth arguments for what the relevant market should be, and with courts taking on the role of arbiter in deciding what market definition will prevail. Closely related is the habit of scholars to see market definition as an exercise in selecting the best relevant market for a case from among the possible alternatives.

However, since a relevant market is nothing but an analytical tool used to study a given theory of harm, and since there are many possible theories of harm incipient in any interesting fact-pattern, there will typically be many helpful relevant markets in any given case. As already discussed, the proper definition of a relevant market depends on the specific features of a theory of harm. Given that there may be many theories of harm in any given trial or investigation, there may also be many helpful relevant markets. One may have to pursue market analyses for vertical and horizontal theories of harm, or, in in merger cases, for unilateral and coordinated theories of anticompetitive harm. A question that might follow from the previous discussion is how an antitrust violation should be defined in the absence of a singular relevant market concept. What if the analysis suggests that anticompetitive injury has occurred, or is likely to occur, in some relevant markets but not in others? The logical answer is that injury in any properly defined relevant market is sufficient to establish a violation.

In short, market definition is merely an analytical tool used to facilitate the evaluation of specific theories of anticompetitive harm. The theory-dependence of markets obviates the idea that market definition should be the strict starting point of antitrust analysis. The ultimate question should take precedence: whether antitrust injury has occurred or is likely to occur. Since all that a relevant market does is identify a group of transactions in which anticompetitive injury is possible—at least under assumptions favourable to the theory of harm—allegation of a relevant market adds nothing to the allegation of anticompetitive injury itself. In other words, while market definition can be a useful step in antitrust analysis, it is not a necessary step, as is made clear by the rules governing per se infringements.


This is a thought-provoking paper, to the point where I am not sure what to think about it – maybe because it is very US-focused. On the one hand, I agree with the authors: absent legal requirements, market definition is not necessary to identify anticompetitive effects, even if it can be a very useful analytical tool to that end. On the other hand, I find it hard to accept that market definition has no autonomy whatsoever from the applicable theory of harm, as the authors seem to argue. For example, we may need to establish dominance or market power; and even though this is naturally related to the ability to engage in anticompetitive harm, dominance also operates as a self-standing screen that determines which unilateral actions are cognisable under antitrust. As a rule, dominance or monopolisation are evaluated by reference to specific markets, but I fail to see how potential theories of harm must play a role in defining these markets. Maybe I am falling into one of the fallacies identified by the authors; then again, maybe the authors dismiss too easily the need to conceptualise markets independently from the analysis of anticompetitive effects if they are to serve a useful analytical role.  

Author Socials A weekly email with competition/antitrust updates. All opinions are mine

What do you think?

Note: Your email address will not be published