US Mergers

Developments in US Competition Review of Mergers

The substantial revision of the Horizontal Merger Guidelines (2010 Guidelines) ¹ by the Antitrust Division of the Department of Justice (DoJ) and the Federal Trade Commission (FTC) after eighteen years is unquestionably the most significant development this year in United States merger review. After nearly a year's effort by a joint working group of senior economists and lawyers from both the FTC and the DoJ and numerous public workshops and multiple comment opportunities, the greatly anticipated revision was issued in final form on 19 August 2010. The new Guidelines are designed to 'better reflect the agencies' actual practices,' 'to provide more clarity and transparency,' and to 'provide businesses with an even greater understanding of how [the antitrust agencies] review transactions.' ²

Although the new Guidelines constitute a substantial revision to the 1992 Guidelines, many of the revisions are not surprising. A significant portion of the Guidelines reflect refinements and changes previously identified in the Commentary on the Horizontal Merger Guidelines, which the agencies issued jointly in 2006 to explain the basis for recent merger challenges and clearances. ³ In addition, the broad concepts of the Guidelines have not changed. Indeed, as the 19 August 2010 issuing press release stresses, the new Guidelines 'are not intended to represent a change in the direction of merger review policy.' ⁴

The critical differences between the old 1992 Guidelines and the new 2010 Guidelines are instead a shift in the relative importance of such basic analytical elements as market definition and concentration levels and in the introduction of certain new analytical tools. In general, the new Guidelines stress that merger review is a fact-based process involving analytical tools that may vary depending upon the particular characteristics of the industry, the products and the marketplace, a concept that is not particularly debatable.

The new Guidelines have generally been applauded as a necessary advancement of and contribution to transparency. Nonetheless, some of the changes and additions - such as the introduction of the concept of upward pricing pressure - are predictably generating debate and uncertainty regarding their future application. In addition, there remains uncertainty as to which analytical tools will be applied to any given merger. Finally, and most significantly, FTC Commissioner J Thomas Rosch's blunt criticisms of the Guidelines suggest some uncertainty as to whether merging parties can be confident that the Guidelines fully reflect the actual practice at the FTC. ⁵

Notable changes or additions implemented by the 2010 Guidelines and some of the debate being generated by those changes is the focus of this article.

Elimination of the mechanical five-step analytical process

The 1992 Guidelines set forth a five-step analytical process that provided for a sequential consideration of the proper definition of the relevant market and measuring that market's concentration levels and changes in those levels, the potentially adverse competitive effects, the potential for offsetting competitive effects through entry, the presence of any efficiencies and failing-firm determinations. ⁶ This step-by-step approach has been officially abandoned in the 2010 Guidelines, although each of these five elements remains a part of the Guidelines' analytical tool kit.

Instead, the 2010 Guidelines warn that 'merger analysis does not consist of uniform application of a single methodology' and is 'a fact-specific process' in which the agencies use a variety of analytical tools to determine whether a particular merger may substantially lessen competition. ⁷ As Assistant Attorney General Christine Varney explained at the final public workshop in the revision process, '[p]anelists have noted_ that far more flexibility than is indicated in the [1992] Guidelines is both the norm of actual Agency practice and, moreover, appropriate given the_ range of transactions confronted by the Agencies." ⁸

This approach provides the agencies with maximum flexibility in selecting the methodologies to use in analysing the competitive effects of mergers and in deciding whether to start their analysis with market definition, competitive effects or elsewhere. This change in approach will likely create greater uncertainty as to the specific tools and analyses that the agencies will employ in any given merger and will place a premium on good pre-filing planning by merging parties and their counsel.

Although this new flexible approach has created some initial consternation within the legal community, at the end of the day, it is doubtful that the new Guidelines significantly depart from long-standing practice. On more reflective consideration, the community of antitrust lawyers and economists has recognised that, in general, each merger is fact-specific in any event and should naturally require tools and analyses designed to get to the actual competitive impact of that particular merger. While that may appear to create uncertainty for some, it also creates opportunities for other merging parties to demonstrate that analysis of their merger should not be straightjacketed into poorly fitting, but standard, analytical tools.

De-emphasis of market definition

As further explained by AAG Varney at the final workshop, '[i]mplicit in de-emphasising the sequential nature of the Guidelines inquiry is a recognition that defining markets and measuring market shares may not always be the most effective starting point for many types of merger reviews.' ⁹ Thus, the initial guidelines proposed on 20 April 2010 and made available for public comments (Initial Proposed Guidelines) stated that market definition is merely 'one of the tools_ use[d] to assess whether a merger is likely to lessen competition.' ¹⁰

This change was misunderstood and generated extensive criticism that the DoJ and FTC were ignoring well established legal precedent that required the definition of relevant product and geographic markets in section 7 cases. ¹¹ As a result, in the final version of the Guidelines, the agencies clarified and softened their position on the role of market definition and concentration. ¹² The agencies explained that one or more relevant markets will 'normally' be identified, but that the agencies do not believe that their analysis needs to start with market definition. ¹³ They also made clear that measurement of market shares and market concentration is not 'an end in itself.' ¹⁴ To emphasise this point, the agencies noted that some of their analytical tools for determining competitive effects do not rely on market definition, ¹⁵ reminding everyone that they are still comfortable challenging a merger on a primary basis other than the merger's impact on concentration in a defined market.

As clarified, the agencies' position does not appear unreasonable and builds on the fundamental understanding that defining a relevant market and assessing shares in that market merely permits the inference of potential harm to competition from high market concentration levels or individual market shares. When direct evidence exists that certain customers are likely to be harmed by a merger, or that certain customers have already been harmed by a consummated merger, the agencies understandably believe that they do not need to prove indirectly what they can prove directly. ¹⁶ This practical approach is consistent with the 2006 Commentary on the Horizontal Merger Guidelines, which explained that the agencies apply the Guidelines framework flexibly, and the DoJ's Merger Review Process Initiative, which recognised the value of tailored second request schedules to enable the DoJ to take a 'quick look' at potentially dispositive issues. ¹⁷

The market definition section of the 2010 Guidelines also retains but clarifies in various respects the hypothetical monopolist test, which is frequently used to aid in defining relevant markets. ¹⁸ Although the Initial Proposed Guidelines suggested that the small but significant and non-transitory increase in price (SSNIP) used in the hypothetical monopolist test would be increased to 10 per cent from the original 5 per cent of the 1992 Guidelines, the final 2010 Guidelines clearly state that the agencies 'most often' use a SSNIP of 5 per cent of the price paid by consumers for products, although the agencies also made clear that a SSNIP of more or less than 5 per cent may be used in some circumstances. ¹⁹ In a new addition, the 2010 Guidelines also describe at length the kinds of historical evidence that the agencies employ to implement the hypothetical monopolist test, including, for example, customer surveys showing responses to past price changes and industry participants' behaviour in tracking and responding to one another's price changes. ²⁰

The 2010 Guideline's section on the hypothetical monopolist test explicitly includes for the first time a discussion of critical loss analysis (CLA). CLA is perceived to be a method for giving the hypothetical monopolist test a more rigorous basis. ²¹ The Guidelines introduce CLA as possibly being used 'to assess the extent to which it corroborates inferences drawn from' other evidence of customer substitution and preferences. CLA involves calculating the percentage of lost sales that will equate lost profits that result from decreased sales caused by a hypothetical monopolist's SSNIP to the increase in profits on the retained sales. ²² If the predicted or actual loss is greater than the equilibrium critical loss, then the SSNIP is unprofitable, indicating that consumers are turning to other alternatives and that the market should therefore be expanded to include those alternatives. In commenting on the proposed Guidelines, senior enforcement personnel have suggested, however, that CLA's inclusion in the Guidelines was designed to deter misuse of CLA rather than as an indication that CLA will be affirmatively used by the antitrust authorities.

Revised Herfindahl-Hirschman Index (HHI) levels

The agencies have also updated the market concentration levels that they deem likely to warrant further scrutiny or challenge of a merger. The concentration levels in the existing 1992 Guidelines were recognised as 'inaccurately describ[ing] the Agencies' enforcement policy' because 'the Guidelines overstate the importance of HHIs in merger analysis' and 'HHI thresholds set forth in the Guidelines no longer capture agency practice or economic learning about the kinds of mergers that are most likely to lead to consumer harm.' ²³ The new proposed Guidelines include revised HHI thresholds to more accurately express how the agencies use HHIs, as set forth in the following table.

These changes in HHI thresholds were broadly welcomed and generally viewed as more consistent with actual agency practice with respect to most industries. The 19 August 2010 press release made clear, however, that the new thresholds will not apply to bank mergers. Instead, the 1995 Bank Merger Competitive Review Guidelines, ²⁸ originally developed jointly by the DoJ and the federal banking authorities, will continue to govern the review of bank mergers. The Bank Merger Guidelines will maintain the modified 1992 Guideline thresholds that have been used in bank mergers for many?years. No announcements, regarding reactions to the 2010 Guidelines, have been made by other agencies, eg, the Federal Energy Regulatory Commission, which have developed their own merger analyses based on the 1992 Guideline's HHI thresholds.

Evidence of adverse competitive effects

The 2010 Guidelines introduce a new section on 'Evidence of Adverse Competitive Effects,' ²⁹ which identifies an uncontroversial, non-exhaustive list of five types of evidence and three sources of evidence that the agencies, in their experience, have found informative in predicting the likely competitive effects of a merger. Consistent with the Guidelines's de-emphasis of market definition and shares, only two of the types of evidence relate to market shares and concentration. The proposed Guidelines also describe the most common sources of useful evidence as data and testimony provided by the merging parties and information provided by customers. Indeed, contemporaneous documents from the files of the merging parties and the views of customers are considered by the agencies to be extremely important sources of information. The proposed Guidelines expressly state what has been widely understood: that the agencies are, in contrast, sceptical of competitors' views.

The inclusion of this new section emphasises the fact-specific nature of merger analysis and is consistent with the agencies' officially making competitive effects the 'central question,'29 although this has been the unofficial case for some time. According to the 2010 Guidelines, the agencies will 'consider any reasonably available and reliable evidence' to address this central question. ³⁰

Focus has shifted to competitive effects

Expanded discussion of unilateral effects

In keeping with the new Guideline's focus on competitive effects, the Guidelines provide an expanded discussion of how the agencies evaluate unilateral competitive effects. ³¹ The concept of adverse unilateral effects was first explicitly introduced in the 1992 Guidelines, but this is an area where there have been significant developments in agency practice and understanding since 1992. The agencies routinely use several important tools when assessing unilateral effects that were not even mentioned in the 1992 Guidelines. These include diversion ratios, price-cost margins, win-loss report data, customer switching patterns, and the views of competitors, customers, and industry observers in cases of mergers of firms selling differentiated products.³²

The new Guidelines have given the agencies an opportunity to lay out in detail their views as to how unilateral competitive effects can arise and how they will be analysed. The agencies seized this opportunity in some respects, but failed to do so in others. They did identify and discuss in detail four circumstances in which unilateral competitive effects can arise:?differentiated products subject to price increases; products sold through individual negotiations or auctions that may facilitate price increases; homogeneous products subject to capacity reductions; and the potential curtailment of innovation. ³³ The 2010 Guidelines explicitly warn, however, that this list of four scenarios is not exhaustive and that unilateral effects can arise in many other ways, including through exclusionary conduct. ³⁴

On the other hand, the agencies were not as thorough as many practitioners had hoped in explaining with some precision the ways in which they use various analytical tools and how they handle certain unresolved issues regarding those tools. In particular, the less than detailed discussion of the agencies' use of diversion ratios and the 'value of diverted sales,' which is also called 'upward pricing pressure' (UPP), have generated a good deal of comment and debate.

The agencies' confirmation that they do use, and plan to continue to use, diversion ratios generated concerns that diversion ratios would be estimated using market shares, particularly given that the agencies did not discuss the evidence from which they would calculate diversion ratios. ³⁵ Commenters pointed out that the use of market shares requires assumptions that are unlikely to be reliable in many circumstances and may suggest patterns of substitution that are not realistic. ³⁶ The agencies did not revise the discussion of diversion ratios in response to these comments, but the agencies' view - that market definition and market share calculation can often be delayed until after an analysis of competitive effects - suggests that the agencies should avoid using market shares to estimate diversion ratios.

The Guidelines also include the relatively new UPP concept for the first time, but leave many questions unanswered about the precise manner for applying the concept. UPP examines post-merger changes in pricing incentives and was principally developed by Carl Shapiro and Joseph Farrell, who are currently in charge of economics at the DoJ and FTC, respectively. The agencies view it as an alternative to market definition and concentration analysis for differentiated products and a more reliable alternative than HHI levels for diagnosing differentiated product mergers. ³⁷ UPP essentially quantifies the concept that the merger of two differentiated products that are close substitutes will increase incentives to raise the price of one product because a significant portion of the lost sales will be internalised post-merger.

Comments on the Guideline's addition of UPP to merger analysis were mixed, as reflected, for example, in the public writings of three noted antitrust economists, each of whom is a former deputy assistant attorney general for economics at the DoJ. In his comments on the Guidelines, Professor Robert Willig described UPP as 'a potentially powerful new tool,' 'well founded in economics' and 'a welcome addition to merger analysis." ³⁸ Professor Willig nonetheless urged that analysis of the relevant market be undertaken in conjunction with UPP analysis and expressed concern that the Guidelines did not discuss any of the substantial issues regarding how to calibrate, calculate or interpret UPP in practice. ³⁹

Similarly, Professor Dennis Carlton submitted comments that also raised numerous questions about the problems of measuring UPP in practice and that pointed out that UPP is derived from a static Bertrand model of competition involving only price rivalry, which may not accurately fit all industries and markets. Professor Carlton demonstrates that, in certain circumstances, the direction of the price pressure can be unclear and may be both 'downward' and 'upward.' In his view, these limitations require that UPP be used and interpreted with great caution, especially given that UPP at most indicates the presence of upward pressure on prices and is never capable of predicting the magnitude of any possible price increase. ⁴⁰

On a similar theme, another former deputy assistant attorney general for economics, Daniel L Rubinfeld, points out, in a recently published article authored with Roy J Epstein, that UPP is just one special case of merger simulation where diversion between the products of the merging firms is assumed to be just uni-directional. ⁴¹ In other words, if the merging firms each have one product, customers divert from product 1 to product 2 but not vice versa in response to potential post-merger changes in positioning or other attributes of the merging parties' products. More importantly, Professor Rubinfeld points out that, in many cases, calculating UPP may not be appreciably easier than undertaking a merger simulation and that a merger simulation may use a more robust model and will at least have the advantage of predicting the magnitude of the projected price increase in addition to the incentive to increase prices. ⁴²

Some of the controversy regarding unanswered questions about the agencies' utilisation of diversion ratios and UPP are derived from their direct or indirect and controversial reliance upon price-cost margins. The use of margins in evaluating pricing incentives and pressure as well as in section 2.2.1 on sources of evidence and in the application of the hypothetical monopolist test generated a number of comments regarding the difficulty in accurately measuring margins and in accurately interpreting their implications. These concerns were heightened by the Initial Proposed Guidelines' strong statements about the inferences to be drawn from high margins. ⁴³ The final Guidelines did not diminish this reliance on margins, as noted in Commissioner Rosch's statement on release of the Guidelines, ⁴⁴ although the agencies did moderate their statements regarding inferences to be drawn from high margins to recognise that high margins can be consistent with earning competitive returns. ⁴⁵

Few courts have relied solely on a unilateral effects theory, and it remains to be seen how courts will embrace the new tools and methodologies advanced in the Guidelines. The first court decision involving UPP since the Initial Proposed Guidelines were issued denied the government leave to amend the complaint to add the UPP theory and held that the government still had to define an appropriate market, as required by long-established antitrust jurisprudence. ⁴⁶ This suggests that the DoJ and the FTC will need to continue to define and defend relevant markets in court for the foreseeable future. ⁴⁷

Updated discussion of coordinated effects

The core goal of coordinated effects analysis of mergers remains the same as in the 1992 Guidelines: determining whether the merger will increase the likelihood that firms in the market can engage in certain profitable conduct as a result of the firms accommodating one another. The new Guidelines do describe more extensively the factors on which the DoJ and FTC will focus in evaluating whether the merger will change how firms interact with one another and the degree to which firms can predict whether one another's responses will be accommodating to price increases or other such actions.

Predictably, the Guidelines indicate that the agencies are likely to challenge a merger on a coordinated effects theory if the market is moderately or highly concentrated and if the market shows some vulnerability to coordinated conduct due to product homogeneity, pricing transparency and other familiar industry characteristics. Less predictably, the Guidelines also say that the agencies must have 'a credible basis on which to conclude that the merger may enhance' the market's vulnerability to coordinated conduct. The Initial Proposed Guidelines required that the agencies have a 'plausible theory' as to how adverse coordinated effects would in fact be facilitated by the merger. The change from 'plausible theory' to 'credible basis' appears to make the Guidelines more rigorous, but whether this 'credible basis' requirement will be a meaningful restraint on the agencies will become apparent only with their application of the new Guidelines to actual transactions and some testing of that application in the courts.

Continued tough standards on entry and efficiencies

The 2010 Guidelines retain the essential concepts regarding the impact of potential entry on merger analysis: it must be timely, likely and sufficient in its magnitude, character and scope to deter or counteract the likely adverse effects of the merger. The 2010 Guidelines also stake out three positions that confirm that the agencies continue to be tough on merging parties who rely on entry arguments.

First, the agencies have eliminated the two year time period for entry. They now require entry that is 'rapid enough' to render unprofitable any anticompetitive conduct encouraged by the merger. Merging parties should anticipate that 'rapid enough' is far less than two years.

Second, the agencies report that a successful entry theory will depend heavily upon presenting evidence of actual successful entry in the past.

Third, and most controversially, the Initial Proposed Guidelines stated that the agencies will 'normally' consider entry to be sufficient to offset harmful competitive effects only if it replicates 'at least the scale and strength of one of the merging firms.' ⁴⁸ The response to this position, which suggested that entry by even a series of smaller firms would not be viewed as adequate to replace a single merging firm of substantial size, was predictable. Numerous commenters were quick to criticise this position as unduly restrictive. For example, Professor Willig described this position as 'highly inappropriate and inconsistent with foundational principles' because 'it would be an extremely unusual competitive effect of concern that would eliminate the entire beneficial activities of one of the merging firms.' ⁴⁹ Thus, Professor Willig argued that the Initial Proposed Guideline's position was 'implausible' because it required in effect that demand for output in the relevant market be substantially and immediately greater post-merger than pre-merger to sustain the existing merged parties and the scale required of the new entrant. ⁵⁰

The final Guidelines did respond to this criticism. Although the Guidelines retained the concept of replicating the scale and strength of one of the merging firms and make clear that such entry would be deemed to satisfy the requirement that entry be sufficient to deter or counteract the competitive effects of the merger. The Guidelines now also reflect that smaller scale entry 'may be sufficient' if such smaller scale firms do not face 'significant competitive disadvantage'. ⁵¹ In the end, this leaves entry analysis essentially where it was under the 1992 Guidelines.

The treatment of efficiencies in the new Guidelines is more detailed than in the 1992 Guidelines but does not make any dramatic changes. The efficiencies section does contain a footnote that hints that the agencies might recognise efficiencies based on reductions in fixed costs. The agencies have historically declined to do so even though the 2006 Commentary on the Horizontal Merger Guidelines indicated that fixed cost efficiencies would be considered at least as an 'exception[ ] to the general rule.' ⁵² The footnote in the new Guidelines expresses the view that efficiencies with respect to fixed costs will not benefit customers in the short term but may do so in the longer term in particular circumstances such as improving product innovation. The footnote also warns, however, that any such efficiencies, even if recognised will be given less weight because they are more distant in time and more uncertain.

***

The 2010 Guidelines are a clear improvement over continued agency practice pursuant to eighteen year old Guidelines that were no longer being followed. The new Guidelines also do much to describe actual agency thinking and process in numerous areas.

Nonetheless, there is a nagging sense that the final Guidelines have not achieved their explicit goal of clarity and transparency as to all aspects of the review of mergers and acquisitions by the agencies. Certain concepts - most particularly UPP - are clearly not fully developed and, therefore, necessarily cannot further clarity and transparency at this time. It is also unsettling, given the stated goal for these Guideline revisions, that the final Guidelines could be issued in the face of Commissioner Rosch's simultaneous and blunt statements that they 'do not describe the way that the Bureau of Competition and enforcement staff at the Commission proceed today.' ⁵³ Hopefully, the DoJ, the FTC and Commissioner Rosch will continue their desire for transparency and clarity and will through speeches, investigation-closing statements and other opportunities continue to explain the Guidelines and the actual agency practice to insure that merging parties have the actual guidance they need.

Notes

1

. US Department of Justice, Antitrust Division, and Federal Trade Commission, The Horizontal Merger Guidelines (19 August 2010) (www.justice.gov/atr/public/guidelines/hmg-2010.html).

2

. Press Release, Department of Justice, Department of Justice and Federal Trade Commission Issue Revised Horizontal Merger Guidelines (19 August 2010) (www.justice.gov/atr/public/press_releases/2010/261642.htm) (19 August 2010 Press Release).

3

. FTC Seeks Public Comments On Revised Merger Guidelines, 98 Antitrust & Trade Regulation Report 478 (23 April 2010).

4

. 19 August 2010 Press Release at 3.

5

. Statement of Commissioner J Thomas Rosch on the Release of the 2010 Horizontal Merger Guidelines (19 August 2010) (www.ftc.gov/os/2010/08/100819hmgrosch.pdf) (Rosch Statement) at 1 and 3.

6

. See generally US Department of Justice, Antitrust Division, and Federal Trade Commission, Horizontal Merger Guidelines (1992, revised 1997) (www.usdoj.gov/atr/public/guidelines/hmg.htm).

7

. 2010 Guidelines at 1.

8

. Christine A Varney, Assistant Attorney General, Antitrust Division, US Department of Justice, Remarks as Prepared for the Horizontal Merger Guidelines Review Project's Final Workshop (26 January 2010) (www.justice.gov/atr/public/speeches/254577.htm).

9

. Id.

10

. US Department of Justice, Antitrust Division, and Federal Trade Commission, Horizontal Merger Guidelines for Public Comment: Released on 20 April 2010, at section 4 (www.ftc.gov/os/2010/04/100420hmg.pdf).

11

. Eg, Comments of the ABA Section of Antitrust Law, HMG Revision Project - Comment Project No. P092900 (4 June 2010) (www.ftc.gov/os/comments/hmgrevisedguides/548050-00026.html) (ABA Comments) at 6-8.

12

. 2010 Guidelines at Section 4.

13

. Id.

14

. Id.

15

. Id.

16

. Christine A Varney, Assistant Attorney General, Antitrust Division, US Department of Justice, Remarks as Prepared for the Horizontal Merger Guidelines Review Project's Final Workshop (26 January 2010) (www.justice.gov/atr/public/speeches/254577.htm).

17

. Id.

18

. The Hypothetical Monopolist Test is a conceptual technique for assisting in market definition. It examines whether a hypothetical monopolist in the product market as defined can profitably impose a small but significant and non-transitory increase in price. If the answer is yes, and the price increase would be profitable for the hypothetical monopolist, then the market is appropriately defined; if the answer is no, then the relevant product market is defined too narrowly and must be expanded. The same approach is also used for geographic markets.

19

. 2010 Guidelines at section 4.1.2.

20

. 2010 Guidelines at section 4.1.3.

21

. Id.

22

. The formula for calculating the critical loss is S / (M + S), where S is the amount of the SSNIP and M is the pre-merger gross margin.

23

. Christine A Varney, Assistant Attorney General, Antitrust Division, US Department of Justice, Remarks as Prepared for the Horizontal Merger Guidelines Review Project's Final Workshop (26 January 2010) (www.justice.gov/atr/public/speeches/254577.htm). HHI levels are calculated by the agencies to measure market concentration levels. They are calculated by summing the squares of the individual market shares of all the firms in the relevant market.

24

. An increase in excess of 100 points 'potentially raise significant competitive concerns and often warrant scrutiny.' (2010 Guidelines at section 5.3.)

25

. An increase greater than 100 creates a rebuttable presumption of competitive harm. (2010 Guidelines at section 5.3.)

26

. Increases between 100 and 200 'potentially raise significant competitive concerns and often warrant scrutiny.' An increase greater than 200 creates a rebuttable presumption of harm to competition. (2010 Guidelines at section 5.3.)

27

. www.justice.gov/atr/public/guidelines/6472.htm

28

. 2010 Guidelines at section 2.

29

. Id.

30

. Id.

31

. 2010 Guidelines at section?6.

32

. Id.

33

. Id.

34

. Id.

35

. Elizabeth M Bailey, Gregory K Leonard and Lawrence Wu, Comments on the 2010 Proposed Horizontal Merger Guidelines (3 June 2010) (www.ftc.gov/os/comments/hmgrevisedguides/548050-00012.pdf) at p2.

36

. Id.

37

. 2010 Guidelines at section 6.1.

38

. Robert Willig, Public Comments on the 2010 Draft Horizontal Merger Guidelines (www.ftc.gov/os/comments/hmgrevisedguides/548050-00015.html) (Willig Comments) at 3-4.

39

. Id.

40

. Generally, Dennis W Carlton, Comment on Department of Justice and Federal Trade Commission's Proposed Horizontal Merger Guidelines (4 June 2010) (www.ftc.gov/os/comments/hmgrevisedguides/548050-00034.pdf).

41

. Daniel L Rubinfeld and Roy J Epstein, Understanding UPP, The BE Journal of Theoretical Economics, volume 10, issue 1, article 21, at 6-9.

42

. Id.

43

. Eg, ABA Comments at 17.

44

. Rosch Statement at 2.

45

. 2010 Guidelines at n.3.

46

. City of New York v Group Health Inc, No. 06 Civ. 13122 (RJS), 2010 WL 2132246 (SDNY 11 May 2010).

47

. UPP has also been criticised as almost always predicting a price increase even when market concentration would suggest otherwise. See Gopal Das Varma, 'Will Use of the Upward Pricing Pressure Test Lead to an Increase in the Level of Merger Enforcement?', 24 Antitrust 27 (2009).

48

. Initial Proposed Guidelines at section 9.3.

49

. Willig Comments at 5.

50

. Id.

51

. 2010 Guidelines at section 9.3.

52

. Federal Trade Commission and US Department of Justice, Commentary on the Horizontal Merger Guidelines (2006) at 57-58.

53

. Rosch Statement at 1 and 3.