It takes a little unpacking to understand the impact of the 2010 Guidelines on the exercise of geographic market definition. While the 2010 Guidelines offer a variety of changes and clarifications about how to define a geographic market, they also mark a departure from the historic focus on first defining a market, and then assessing changes in concentration. The 2010 Guidelines emphasize that “merger analysis does not consist of uniform application of a single methodology.” At one level, nothing new is being said here. From the beginning, the DOJ, and then the FTC, make clear that government enforcement is not tied to the Guidelines in some wooden and mechanical manner. But at another level, the 2010 Guidelines represent a paradigm shift: The analytical center stage is now devoted to attempting to measure directly the economic consequences of horizontal mergers rather than inferring the consequences from implied changes in market structure.Footnote 27 To the extent that this was a protocol already being followed within the enforcement agencies, the protocol is now explicit and public.
The 2010 Guidelines incorporate analytical tools that the Agencies will use to assess competitive effects, indicate an openness to more diverse forms of evidence, and remove the algorithmic description of the HMT. These changes move the Guidelines’ focus away from a one-size-fits-all analytic approach and towards an acknowledgement of the appropriateness of using a variety of methods to measure the price changes that consumers may face as a result of a merger. In this shift, both product and geographic market definition, previously considered the first steps of any merger analysis, are now viewed as secondary to an understanding of the competitive effects of a merger. Discussion of these competitive effects occupies the first section of the 2010 Guidelines, with market definition contingent upon the identified competitive effects. Indeed, the 2010 Guidelines state, “The Agencies’ analysis need not start with market definition. Some of the analytical tools used by the Agencies to assess competitive effects do not rely on or require market definition” (DOJ and FTC 2010, p. 7).Footnote 28
The 2010 Guidelines also describe competitive effects and market definition as jointly determined: “Evidence of competitive effects can inform market definition, just as market definition can be informative regarding competitive effects” (DOJ and FTC 2010, p. 7). Taken together, these statements suggest a diminished role for market definition by the enforcement authorities, as it is explicitly no longer the required first step to assessing the impact of a given merger.Footnote 29 After 2010, the (sometimes perplexing) central question is: How does one assess the effect on price and output of a potential merger without defining a market?
The 2010 Guidelines provide several categories of evidence that the Agencies consider to be informative of potential adverse competitive effects. These include (1) “historical events, or ‘natural experiments;’” (2) whether there is substantial head-to-head competition between the merging parties; and (3) whether one of the merging firms is a “disruptor” that threatens to displace incumbents (DOJ and FTC 2010, p. 3). Geographic concerns are explicitly addressed in the historical events category, where the Agencies provide as an example that using geographical variation of competition between the merging parties “may be informative of post-merger prices” (DOJ and FTC 2010, p. 3).
Even with the 2010 Guidelines’ de-emphasis of the primary role of market definition in merger enforcement, this version still offers additional details as to how the agencies would define markets in those circumstances when it is appropriate or necessary to do so. For the first time since 1982, the HHI thresholds for enforcement policy are relaxed, as well as the doctrinal reliance on market shares (DOJ and FTC 2010, pp. 1, 19). Also, while the 2010 Guidelines preserve the demand substitution focus of market definition as implemented with the HMT (and explained it in greater detail than previous versions), this version does not describe any algorithm to determine the smallest product and geographic markets over which a hypothetical monopolist could exert market power, and no longer indicates the order in which products or geographic areas should be tested. Instead, the discussion of the HMT is centered on a few examples, tendered to help a reader understand the Agencies’ thinking when considering evidence in market definition. For example, when defining product markets the 2010 version states that “[g]roups of products may satisfy the hypothetical monopolist test without including the full range of substitutes from which customers choose” and illustrates this principle with an example.Footnote 30
When defining geographic markets, the 2010 version recognizes the importance of the location of both producers and customers. While the DOJ may have in practice been considering both producer and customer-based approaches in the past when assessing geographic markets (e.g., depending on the potential for price discrimination), the 2010 Guidelines are the first to emphasize these options in detail. The 2010 version states, “The arena of competition affected by the merger may be geographically bounded if geography limits some customers’ willingness or ability to substitute to some products, or some suppliers’ willingness or ability to serve some customers.”
The 2010 Guidelines describe that one must decide whether to start with either the supplier location or the customer location depending on the relevant economic conditions, especially the potential for price discrimination. This language can only elevate the validity to the outside world of geographic market definition exercises that were centered on customer location from the beginning of the analysis. Prior versions of the Guidelines described geographic market definition as an exercise to be conducted around product flows and/or producer locations, with the implications of price discrimination and customer location receiving only passing mention.
As might be obvious by now, the approach to geographic market definition can be quite different, depending on where one starts. In regard to supplier location based markets: “[T]he Agencies normally define geographic markets based on the location of suppliers” if the product is generally being picked up by the customer at the supplier location and there is no “price discrimination based on customer location.” Once the appropriateness of a supplier-based geographic market is determined, this version sheds light upon the ambiguity in implementation that was introduced in earlier versions, by specifying that “sales made by suppliers located in the geographic market are counted, regardless of the location of the customer making the purchase” (DOJ and FTC 2010, p. 14).
In contrast, for customer-based markets: “[I]f price discrimination based on customer location is feasible as is often the case when delivered pricing is commonly used in the industry, then the Agencies may define geographic markets based on the locations of customers” (DOJ and FTC 2010, p. 13). The 2010 Guidelines make clear that in this exercise the geographic markets “encompass the region into which sales are made. Competitors in the market are firms that sell to customers in the specified region. Some suppliers that sell into the relevant market may be located outside the boundaries of the geographic market” (DOJ and FTC 2010, p. 14). To clarify the customer location methodology, the 2010 version provides three examples, which describe examples of when and how a market may be defined around the locations of customers, as well as which sales should be included in the geographic market when defined in this manner.Footnote 31
The presence of price discrimination (or not) always has been an analytical key to determining whether one should begin with supplier location or customer location; these are two very different approaches to geographic market definition. The 2010 Guidelines offer general guidance with regard to how to determine whether price discrimination is in fact present: “For price discrimination to be feasible, two conditions typically must be met: differential pricing and limited arbitrage” (DOJ and FTC 2010, p. 6). In practice, there remains disagreement about how to show whether or not these conditions are present, and hence there can remain disagreement about the appropriate way to approach geographic market definition in a given matter.
When defining geographic markets, the 2010 Guidelines also provide a list of relevant evidence that the Agencies will consider, which varies to some extent from previous versions (DOJ and FTC 2010, p. 14). Six factors exemplify the 2010 Guidelines’ shift towards including analysis for mergers that may facilitate price discrimination. These are:
How customers have shifted purchases in the past between different geographic locations in response to relative changes in price or other terms and conditions;
The cost and difficulty of transporting the product (or the cost and difficulty of a customer traveling to a seller’s location), in relation to its price;
Whether suppliers need a presence near customers to provide service or support;
Evidence on whether sellers base business decisions on the prospect of customers switching between geographic locations in response to relative changes in price or other competitive variables;
The costs and delays of switching from suppliers in the candidate geographic market to suppliers outside the candidate geographic market; and
The influence of downstream competition faced by customers in their output markets (DOJ and FTC 2010, p. 14).
The 2010 Guidelines broadly preserve the definition of market participants as well as elements of the demand-side focus that was in the 1992 and 1997 versions; but this version makes no distinction between “committed” and “uncommitted” entrants. The 2010 version instead distinguishes between firms that could provide rapid supply responses following any post-merger price changes “without incurring significant sunk costs” and firms for which “entry would take place more slowly” (DOJ and FTC 2010, pp. 15–16). The former firms were considered market participants, while the latter firms’ ability to restrict a merger’s adverse effects was measured based on the “timeliness, likelihood, and sufficiency” of entry (DOJ and FTC 2010, p. 28).
While the 2010 Guidelines diminish the primary role of market definition by focusing on competitive effects and remove market definition as a necessary first step in every case, defining markets remains crucial to many cases. In regards to geographic market definition, the 2010 Guidelines adopt a less algorithmic method that spells out two very different approaches to assessing a geographic market and its participants, with the opportunity for—and proof of—price discrimination at its core.