Mr. Raphael and Ms. Granger concurred that market definition plays a critical role in antitrust matters. Mr. Raphael said that, under current law, a market generally needs to be defined to analyze whether the plaintiff theory in a conduct-related matter makes any sense. Ms. Granger mentioned that defining an antitrust market is critical for merger review. For instance, the relevant line of commerce needs to be defined to assess the potential competitive effect of a merger. She said that part of the frustration of defining a market is that, to some extent, it is inherently speculative.
Ms. Blackburn stated that from a merger enforcement perspective, the agencies are directed by the Clayton Act to ask whether there are lines of commerce and sections of the country where the transaction may have an anticompetitive effect. She highlighted that market definition can flow from direct evidence of where the competition is occurring.
The conversation moved next to the tools used to define a market. Ms. Blackburn started by defining the Hypothetical Monopolist Test (HMT) as a thought experiment that asks whether the candidate market is too narrow and whether there are reasonable substitutes outside the proposed geographic area or proposed set of products consumers can turn to. She emphasized that a good tool needs to fit its purpose, so markets cannot necessarily be defined for all mergers or conduct cases using the HMT. There are some markets, like nascent product markets and labor markets, where it may be difficult to implement the HMT. In those situations, markets could be defined using qualitative evidence of competition, per the Brown Shoe practical indicia. Related to this point, Ms. Blackburn said that agencies and courts pay attention to how parties see their markets.
Mr. Zaslavsky asked Ms. Chipty if she could comment on the need for new tools for more complex situations. Ms. Chipty said that new tools do not seem to be needed. She criticized the tendency to affirm that we need new tools if there are new problems. Her view is that HMT is way more robust than some practitioners think: it is a concept of limiting principle that asks how much substitution there is in the candidate market. Such substitution may be assessed based on the attributes that drive consumer choices and how consumer choices change in response to changes in those attributes.
There was live debate among panelists on this topic. Mr. Raphael raised the issue of the relevance of the SNIPP test and related quantitative analyses for defining markets. He said that the SNIPP test is a tool that some economists suggest is not the product of academic economic scholarship. He stated that the SNIPP test potentially privileges the role of economists and that lawyers and parties sometimes undervalue core business documents by relying on them to analyze the conduct but not for market definition.
Following up on Mr. Raphael’s point, Ms. Granger reinforced that market definition is necessary under the law and that we must be aware that different tools produce different outcomes. In that regard, she said that we shouldn't expect everyone to agree on the market definition methodology: the analysis is fact-specific to the case. Market definition analysis is still necessary in conjunction with other evidence. Regarding the relevance of documents, Ms. Chipty agreed that documents are important. Still, a framework is needed to weigh their content: just because a document mentions a competitor does not necessarily mean that competitor is relevant to understanding the competitive effect of any given conduct.
The conversation moved then to the potential tensions surrounding market definition. Mr. Zaslavsky asked the panelists about the tension between market definition being increasingly important and the increasing number of matters where applying the standard market definition framework is difficult. Mr. Raphael commented that market definition remains critical because the law makes it critical and will remain important. He agreed that there are novel cases where courts, attorneys, and economists are struggling with the role of market definition. Mr. Raphael suggested that one might consider whether, just as plaintiffs can go straight for a direct effect approach in these situations, the defendants can move straight to direct competition arguments without going through the formal market definition exercise.
Ms. Chipty was dubious about skipping the formal market definition exercise. She stated that the new merger guidelines mention the word market more than five hundred times, including concepts like market participants, market power, and relevant market. She pointed out that it is difficult to pin down any of those items without getting a properly defined market. Ms. Chipty insisted that practitioners need to get creative to narrow down the market and how to use direct evidence. Ms. Blackburn stated that, in her opinion, the definition of the market is equally important as it has always been. It is a central concept for enforcement actions. She also agreed that practitioners need to be more flexible when defining a market. Regarding the tools, she said you can use a variety of tools. The important thing is that the tool chosen is suited to answering the questions that judges care about regarding market definition.
Finally, Ms. Granger raised the point that from a practitioner’s standpoint, the timing of market definition is changing, too. Parties are having to take a position on market definition earlier in cases. Importantly, the set of tools for market definition is different earlier than later in the case. For example, in early stages, parties can use publicly available and in-house data only, but in later stages parties can use other information obtained in discovery.
Moving to the next topic, Mr. Zaslavsky asked the panelists what tools attorneys and economists should use when the standard approach is challenging. Ms. Chipty took the lead in addressing this topic. She was clear that the best way to determine whether the conduct at issue is anticompetitive is by doing a direct effect analysis. For example, in a merger case, the best approach is to conduct a head-to-head competition analysis. The problem is that sometimes, conducting a direct evidence analysis is not feasible. She said attorneys and economists should build indirect evidence in those cases. For example, diversion analysis is a way to assess the extent of head-to-head competition. In complex situations, economists should think hard about the fundamentals of demand for the products at stake and then try to find a natural experiment where one of those product attributes changes. For instance, the entry or exit of a competitor or the introduction of a new product may shed light on the extent to which the pre-existing or the remaining market participants are a competitive constraint for the parties. Mr. Raphael followed Ms. Chipty's indirect evidence approach by saying that the doctrine privileges direct evidence and deemphasizes the potential of indirect evidence to shed light on parties’ conduct, competitive constraints, and the reality of market competition.
Regarding using natural experiments to get evidence of the indirect effects, Ms. Blackburn emphasized that it is a clever idea and natural experiments can be useful evidence of real-world behavior. She followed up with some examples of natural experiments. Using a hospital going out-of-network with a particular payor would help to identify substitution and diversion by patients. Similarly, exit events like companies closing stores can also help to get diversion ratios and would shed some light on the product market and the geographic market. She raised some caveats about using natural experiments in non-traditional markets, like labor markets. For example, there are multiple reasons why people choose a job, and a specific natural experiment would yield information on a subset of such reasons. The final comment on this topic was from Ms. Granger, who shared that there seems to be less dissatisfaction with the market definition concept and more dissatisfaction with the tools available to identify the relevant market. She noted that antitrust authorities in other jurisdictions sometimes use surveys as a tool for market definition, but U.S. agencies eschew this tool.
The session’s next topic was whether we need a broader toolbox to define markets. Ms. Raphael began by saying there is much work to be done regarding the quality dimension. He raised the issue that quality seems to be an unexplored dimension of competition in general, even in traditional market contexts. There is a need to measure quality objectively, and economists may help to think about how to do that.
The conversation then pivoted to substantive legal rules on market definition and how they are more important in some contexts than others. Ms. Granger started by saying that the problem is that not everyone agrees on when market definition is important. However, she thinks the debate is more about the tools for market definition than the need to define markets. She reinforced the point that defining markets is necessary given the law that exists today. Ms. Granger said, though, that we need to better acknowledge the limitations of the tools we are using. Mr. Raphael said that given the importance of market definition to antitrust analysis, the antitrust bar might consider whether we could have a Markman hearing-type process for market definition. Ms. Blackburn was skeptical that Ms. Raphael's proposal would be workable in practice; for example, key witnesses might have to testify twice: once to talk about market definition and again to talk about competitive effects. Similarly, Ms. Chipty pointed out that market definition relates to the conduct at stake, so markets shouldn't be defined in a vacuum.
At the end of the session, the panel answered questions from the public. The first question was whether practitioners correctly think about innovation markets and nascent competition. Mr. Raphael addressed the question by saying that the key question is whether there is a competitive constraint, and we are interested in whether firms are doing something that will inhibit competition in the relevant activity in the future. Defining the market in this scenario can be challenging. Ms. Chipty commented that, in a way, there are some potential products that we cannot even imagine, so it is a really hard question how to deal with innovation markets and nascent competition. Mr. Zaslavsky mentioned he sees a trend toward more flexibility in cases involving markets where we don’t have quantitative evidence. There is a challenge of being flexible and rigorous at the same time. He mentioned that the 5th Circuit’s decision on Illumina-Grail provides an example of judges giving plaintiffs flexibility to define a market without performing a SSNIP test.
The last question was about the challenges of using documents to define labor markets. Ms. Blackburn repeated her previous point that people take jobs for multiple reasons beyond employers competing for them on economic terms, and this is an important distinction from markets for products and services. As such, enforcers need to look at multiple sides of the picture and have a sense of what evidence and documents judges find more relevant and persuasive. Mr. Raphael said that the labor market definition is particularly challenging for less skilled areas because the opportunities for those workers would be extremely broad.
