Introduction
Reduced emphasis on market definition and market share

Reduced focus on predicted price increases
Shift from 'coordinated effects' to 'unilateral effects'
Simplified definitions of 'entry'
Little change in discussion of efficiencies
Monopsony power
Partial acquisitions
Comment


Introduction

The Department of Justice Antitrust Division and the Federal Trade Commission have just released for public comment a new draft of the Horizontal Merger Guidelines - the first major revision since 1992. The draft will disappoint those who anticipated major changes in direction and reassure those who feared them.

The new guidelines continue to rely on economic factors and do not revive the populist policies of the 1960s and early 1970s. However, they also emphasize (Section 1) that:

  • "merger analysis is necessarily predictive";
  • absolute precision in merger analysis is thus impossible; and
  • statistical calculations are not the only sources of economic wisdom. The agencies will "consider many sources of evidence" (Section 2.2).

This candid acknowledgement of a more holistic, indeed subjective, approach will not surprise experienced counsel, who have long recognized that the technical economic inquiries articulated in the 1992 guidelines did not necessarily reflect actual practice. The judicial response is more difficult to predict, however, in light of increasing insistence on rigorous economic proof in some recent decisions.

This update examines some significant changes in more detail.

Reduced emphasis on market definition and market share

Successive previous versions of the Horizontal Merger Guidelines, published in 1982, 1984 and 1992, have progressively given less weight to market concentration statistics. However, these guidelines continued to assume that the first step in merger analysis would be the definition of a 'relevant market'. In the 1992 guidelines the appropriate calculations were discussed immediately after the introductory paragraph (Section 1). However, in the new draft guidelines these statistical matters are deferred until Section 4, after an extended discussion of other sources of information about economic consequences, including the informed views of merger parties, customers and analysts (Section 2).

Moreover, Section 4 states at the outset: "Market definition is not an end in itself: it is one of the tools." Competitive analysis "need not start with market definition". Product and geographic markets may occupy a spectrum of both close and more distant substitutes. "[D]efining a market to include some substitutes and exclude others is inevitably a simplification."

If market definition is an imprecise inquiry, it makes sense to reduce further the emphasis on market definition and market shares. The new guidelines raise the Herfindahl- Hirschmann Index (HHI) thresholds between 'unconcentrated', 'moderately concentrated' and 'highly concentrated' markets to levels more consistent with agency practice, but note that the presumptions from increased concentration enhance the likelihood of serious agency review.

A reduced emphasis on markets and shares may meet with some judicial resistance. However, counsel should welcome this more realistic approach. If generally accepted, it could have a salutary impact on the way that mergers are investigated and litigated. Today, vast time and resources can be consumed in arguments over market definition and market share calculations, which thereafter may play only a limited role in the final decision.

Reduced focus on predicted price increases

The traditional focus on likely price increases has been modified. When markets are defined, for example, the agencies will still apply the so-called 'SSNIP' test (Section 4.1.1), which defines a 'market' as the smallest product and geographic space within which a hypothetical monopolist "likely would improve at least a small but significant and non-transitory increase in price". But if it is not always necessary to define markets rigorously, these hypothetical predictions may also be unnecessary.

Another shift flows from the articulation of a "unifying theme" that mergers should not be permitted to create, enhance or entrench market power (Section 1). The word 'entrench', which does not appear in the 1992 version, signals that an anti-competitive effect can be found – even in the absence of a likely price increase – if the merger would make it easier for the surviving company simply to maintain already high prices that might otherwise be expected to fall (Section 4.1.2).

The extended discussion of 'non-price' effects also modifies the prior guidelines' almost exclusive concern about likely price increases, which again was not always consistent with agency practice. Injury from non-price effects was briefly acknowledged in the 1992 guidelines (Section 0.1 n 6), but is given greater emphasis in Section 6.4 of the new guidelines, titled "Innovation and Product Variety". It is also noteworthy that proof of adverse effects on innovation do not depend on definition of an 'innovation market', which is a difficult and controversial concept.

Shift from 'coordinated effects' to 'unilateral effects'

The 1982 and 1984 guidelines focused exclusively on the increased likelihood of either explicit or implicit cooperation among competitors after a merger. Concern about adverse 'unilateral effects' resulting from the post-merger conduct of a single firm was first introduced in the 1992 version, in a section that followed discussion of coordinated effects (Sections 2.1 and 2.2). The new guidelines discuss unilateral effects first and at greater length (Sections 6 and 7), because these are more often applied by the agencies today.

Unilateral effects
Greater emphasis on unilateral effects can further reduce the need for market definition or concentration calculations, because these inquiries may not be necessary (Section 6.1). This theory of competitive harm is relatively easy to articulate: if the producer of one product acquires a close substitute product, it may have an incentive to increase prices on one of them because some of the resulting lost sales will be captured by the substitute. A prediction of whether this may actually happen, however, requires estimates of:

  • sales lost by the product subject to a hypothetical price increase;
  • the percentage of these lost sales likely to be captured by the newly acquired product; and
  • the future incremental profit margins on both.

Confidence in these predictions depends on the availability of data.

The new guidelines do not endorse any specific technique for these calculations and acknowledge that there may be a number of alternative "economic models" and other "available and reliable information" that will be relevant to the analysis. Thus, the agencies have a "tool box" of alternative analytical techniques, and will choose the appropriate tool based on the facts of each transaction and the nature of the data available. It is possible that resources saved by reducing the emphasis on market definition and market shares will be diverted in some cases to disputes over alternative calculations of unilateral effects.

Coordinated effects
Reduced reliance on market definition and market concentration statistics will diminish the relative importance of coordinated effects. However, the new guidelines do provide one useful caveat: "[t]he higher the HHI level, the greater is the likelihood that the Agencies will request additional information" (Section 5.3). This statement seems to recognize the concept of a "sliding scale", which holds that the burdens of defence are roughly keyed to the strength of the prosecutor's case - another example of a more flexible and more realistic approach.

There also is implicit recognition of the fact that coordination need not be perfect in order to be harmful. The 1992 guidelines emphasized that successful coordination depends on the ability to detect and punish deviations (Sections 2.1 and 2.12). The new version emphasizes various other factors that may enhance or inhibit coordination (Sections 7.1 and 7.2), and thus recognizes that the effect can be present in varying degrees.

Simplified definitions of 'entry'

The new guidelines eliminate the semantically confusing distinction between 'uncommitted' and 'committed' entrants in the 1992 version. So-called 'rapid entrants' can be included in the initial market definition (Section 5.1), without the unrealistic expectation that their entry (or exit) would be virtually costless.

In addition, the "actual history of entry" is "given substantial weight" (Section 9). This recognizes current practice and, if used with discretion, could be a less burdensome and controversial inquiry than calculations of what potential entrants might do in hypothetical circumstances. However, lack of past entry has little predictive value if the industry has changed fundamentally. Moreover, absence of entry at pre-merger prices does not preclude the real possibility that entry will discipline price increases post-merger.

The three significant elements of "timeliness, likelihood, and sufficiency of the entry" are also cast in less technical terms, and the arbitrary two-year test for timeliness has been eliminated. Since the old two-year test might also be viewed as sufficient proof of timeliness, its elimination could favour prosecutors in some circumstances. Overall, however, it makes better sense to weigh the three elements together rather than specify individual hurdles.

Little change in discussion of efficiencies

The close correspondence between efficiencies language in this version and the 1992 version is itself noteworthy, in light of heightened scepticism about business acumen after the 2008 crash. There is some greater willingness to credit fixed cost savings. Conversely, predicted lower prices are discounted if accompanied by "reductions in product quality or variety" (Section 10).

Monopsony power

The draft guidelines include a discussion of the potential issues raised by mergers of competing buyers, which can enhance market power on the buying side of the market. The draft states that the agencies use "essentially the [same] framework" for evaluating such mergers. Monopsony issues have arisen most often in transactions involving health insurers and in agriculture.

Partial acquisitions

Although partial acquisitions have been the subject of enforcement actions at both agencies, the draft guidelines for the first time discuss the potential competitive effects of such acquisitions, which include:

  • the ability for a minority owner to influence the competitive conduct of the target firm;
  • reduced incentives to compete because of profits gained through the minority interest; and
  • the exchange of competitively sensitive information, which may facilitate coordination.

Comment

The new draft guidelines provide a more accurate description of actual agency practice in recent years. They do not evidence an intention to make radical changes in merger analysis. The increased emphasis on economic evidence that cannot readily be captured by statistics may look like a retreat from purely objective agency determinations. However, some subjective judgements are inevitable and have always been present – and dramatic changes are tempered by the influence of long-term career staff and by the oversight of courts, which tend to rely on precedent. The avoidance of lengthy upfront disputes over market definition, with attendant calculations, might appear to favour the agencies, and greater flexibility could have an adverse effect on predictability. However, the more holistic approach also creates opportunities for defence counsel with good stories to tell. The future level of merger activity is likely to be driven, as it has been, by general economic conditions and by government regulations other than antitrust.

For further information please contact Janet L McDavid or Thomas Leary at Hogan & Hartson LLP by telephone (+1 202 637 5600) or by fax (+1 202 637 5910) or by email (jlmcdavid@hhlaw.com or tbleary@hhlaw.com).