A Proposal to Invigorate Ex Post Merger Policy

By Menesh S. Patel¹

Introduction

Consider a hypothetical merger subject to Hart-Scott-Rodino. Suppose that the merger receives a second request, but the reviewing agency ultimately determines that the transaction poses no meaningful competitive risk and so permits the transaction to go forward (i.e. clears the transaction). Now suppose that at some point after clearance and merger consummation, the agencies receive credible and competent evidence that the merger, contrary to their ex ante predictions, is generating substantial harm in a well-defined relevant market.

What ought the agencies do in this circumstance? Should the merger be immune from agency ex post review and challenge because the agencies only get one bite at the proverbial apple, and the transaction cleared Hart-Scott-Rodino during that first bite? Or should the agencies act in fidelity to their mission of protecting competition and accordingly commence an investigation and eventual suit challenging the merger, despite the agencies themselves previously permitting the merger to go forward and the merger having already been consummated?

In this short article, I describe the justifications and contours of a policy recommendation that I set out in greater detail elsewhere,² calling on the agencies to increase, in a principled fashion, ex post challenges to mergers that they previously reviewed and cleared through the Hart-Scott-Rodino process. As described below, the specific recommended shift in antitrust policy would advance the objectives of antitrust by better ensuring that mergers subject to Hart-Scott-Rodino do not go on to ultimately inflict competitive harm, and does so in a manner that minimizes the potential costs associated with this expansion to ex post merger policy.

II. The theoretical and evidentiary justification for additional agency challenges to previously reviewed and cleared mergers

Merger analysis is a complex, resource intensive exercise that requires the reviewing agency to make an informed prediction about whether the merger will likely generate substantial competitive harm. If the transaction is subject to the reporting requirements of Hart-Scott-Rodino (“HSR”), the reviewing agency will have the benefit of a wide array of material relevant to the transaction obtained through the HSR process. As part of that process, the agencies’ economists will seek to assess the merger’s expected competitive effects through the application of econometric and other analysis. Those econometric tools have broadened over the years and now extend far beyond the simple HHI market screens found in the Horizontal Merger Guidelines. Agency economists have at their disposal a battery of powerful statistical methodologies capable of assessing a merger’s expected competitive effects, such as merger simulations, estimation of diversion ratios and GUPPI (Gross Upward Pricing Pressure Index), and natural experiments. Agency economists deploy those tools with great sophistication and care.³

Nonetheless, as with any predictive exercise, the agencies’ ex ante predictions may prove to be incorrect ex post. A merger predicted to generate no price increase, for instance, may ultimately turn out to generate substantial price increases. This could be the case for a variety of reasons, even if the parties complied fully with their HSR production obligations: for example, the agencies may have access to just a portion of the information relevant to the competitive effects analysis; time and resource limitations may preclude the agencies from robustly analyzing the information they do obtain; and regardless of the rigor of the agencies’ competitive effects analysis, some inherent uncertainty will remain.

The prospect of mergers being deemed ex ante unproblematic but ultimately generating competitive harm is not merely theoretical. There is mounting evidence that, despite their best efforts, the agencies’ ex ante merger predictions can deviate from realized competitive effects. The evidence comes in the form of merger retrospectives, which are empirical studies that use pre- and post-merger data to evaluate how a given merger actually affected competition related variables, ordinarily price. There is a large body of merger retrospectives, and their interest among academics and antitrust policymakers is growing.⁴

Many retrospectives conclude that the evaluated mergers, including mergers that were subject to the HSR process, ultimately generated competitive harm. For instance, a recent study by Nathan Miller and Matthew Weinberg evaluated the realized competitive effects of the 2007 combination of Miller and Molson Coors,⁵ which was the subject of intense antitrust scrutiny by the Antitrust Division. The agency undertook an eight-month investigation of the transaction and conducted a competitive effects analysis based on extensive information obtained from the merging parties, competitors, and downstream purchasers.⁶ The agency ultimately cleared the transaction without conditions, in part because of a prediction that the transaction would generate verifiable and merger-specific efficiencies in the form of production cost savings, which would translate to lower consumer prices. ⁷ That ex ante prediction, however, failed to manifest. Miller and Weinberg’s retrospective of the transaction showed that it actually increased prices in the relevant market by 6 percent.⁸

While some retrospectives conclude otherwise, when viewed in aggregate the merger retrospectives that have been conducted to date indicate that a meaningful fraction of mergers inflict consumer harm. For instance, a recent and well-known meta-study by John Kwoka aggregated the findings of a group of retrospectives that collectively analyzed the competitive effects of mergers on 119 products.⁹ The meta-study determined that the underlying mergers resulted in higher prices for more than 60 percent of the products at issue, with the associated mergers causing the price of the 119 products to increase by an average of 4.31 percent,¹⁰ which included transactions that were subject to the Hart-Scott-Rodino process. Other meta-studies similarly show that a non-negligible percentage of mergers, including previously reviewed and cleared mergers, go on to generate substantial competitive harm.¹¹ None of this is especially surprising given the considerable uncertainty associated with merger review.

Because of the theoretically expected and empirically demonstrated gap between ex ante predictions and actual competitive outcomes, agency challenges to previously reviewed and cleared mergers can correct merger-related impairments to competition and enhance consumer welfare. Furthermore, under the express terms of the HSR Act, the agencies are not legally precluded from mounting an ex post merger challenge despite previously clearing the merger.¹² Nonetheless, while the agencies challenge with some regularity consummated mergers that were not subject to HSR,¹³ it is extremely rare for them to mount ex post challenges to mergers that were subject to HSR,¹⁴ despite the clear theoretical, evidentiary, and legal justification for doing so.

III. The costs of additional ex post agency challenges to previously reviewed and cleared mergers and the appropriate policy response

Additional agency challenges to previously reviewed and cleared mergers have the prospect of generating substantial benefits for consumers and for competition. That fact alone, however, does not imply that the optimal policy response is for the agencies to simply magnify the number of ex post challenges to HSR-reportable transactions. That unnuanced policy response fails to accommodate the significant costs that could accompany that expansion in after-the-fact merger review. Because of those potential costs, the appropriate policy response instead is a principled increase in agency challenges to previously reviewed and cleared mergers along the lines described below.

There are various potential costs associated with additional ex post merger challenges. First, because an agency challenge to a previously reviewed and cleared merger almost always will be directed at a consummated merger,¹⁵ these ex post merger challenges will face the same known difficulties as challenges to consummated mergers generally do.¹⁶ Chief among these difficulties is the availability of a workable and corrective remedy. Once the parties have sufficiently integrated their assets and operations, a challenge to a consummated merger faces the potentially intractable problem of devising an implementable remedy that effectively restores competition to the relevant market. A structural remedy such as a breakup may be infeasible or unable to restore competition because of the difficulty in separating the parties’ commingled assets and operations in a manner that enables the divested unit to sufficiently infuse competition into the target market. Clear evidence of this so-called unscrambling of the eggs problem can be found in pre-HSR agency challenges, which ordinarily involved challenges to consummated mergers and which generated remedial outcomes that seemingly failed to alleviate the competitive harm in the affected markets.¹⁷

There are a number of other potential costs associated with additional ex post merger challenges to HSR-reportable mergers, such as disruption to the finality of the merger review process, diversion of agency resources away from ex ante enforcement, and diminished agency incentives for vigorous ex ante enforcement.¹⁸ While the agencies should amplify their challenges to previously reviewed and cleared mergers, they should do so in a manner that not only seeks to achieve the potential benefits but also respects the potential costs associated with that shift in merger policy.

In particular, as discussed in the more detailed article on which this piece is based, the appropriate policy response involves the agencies increasing the extent to which they challenge previously reviewed and cleared mergers, but challenging a given merger ex post only if both of the following two limiting conditions are satisfied:

Requirement one (the evidence requirement): the preponderance of the agencies’ evidence shows that the merger has or is likely to substantially lessen competition.

Requirement two (the remedies requirement): the agencies reasonably believe there is a remedy that would correct the merger’s competitive harm.

It is helpful to provide a few brief observations about these two limiting conditions. First, they serve as valuable filters separating the types of mergers the agencies may challenge ex post and those they should not, as they focus attention on the most important aspects of the competitive effects analysis: first, based on the available evidence, the challenge is directed at an anticompetitive merger and second, there plausibly exists a remedy that could correct the merger’s competitive harm. The first requirement is especially relevant in light of recent popular discussions of antitrust, which sometimes seek to use antitrust as a means to prohibit mere bigness. A merger that simply generated a large market participant but that generated no competitive harm should not be the subject of an ex post challenge. Additionally, insisting on evidence of competitive harm mitigates the risk of political misuse of antitrust, which is a growing concern.¹⁹

The second requirement is especially relevant to mergers that cleared the HSR process and were consummated years ago. While there may be some remedy that alleviates the merger’s competitive harm, in the case of a long-consummated merger the appropriate remedy may not be a breakup, for that may not be practical or effective because of the parties’ complete or near complete integration. Instead, in that circumstance, an appropriate remedy may be conduct-based. A breakup also may not be corrective because of the nature of the competitive harm. For example, if the underlying competitive harm is diminished consumer privacy, a breakup may not be an appropriate vehicle for rectifying that impairment to consumer welfare.

Second, the two requirements are necessary conditions rather than sufficient ones. Therefore, even if they are met, the agencies in their discretion may elect not to challenge the merger ex post. This could be the case for a number of reasons. For instance, the agencies may choose to only focus on those anticompetitive mergers that have the highest expected net benefit from ex post challenge. Because they are necessary conditions, the two limiting requirements continue to provide the agencies with the discretion they need to conduct effective merger policy in the face of finite — and clearly insufficient — enforcement resources.

Next, the two limiting conditions above should not be construed as exhaustive. If it is determined that additional necessary conditions should be imposed on the exercise of the agencies’ ex post enforcement authority, the set of limiting conditions should be expanded to encompass those other conditions. A public comment period would be especially valuable in identifying any such other conditions and to gather valuable input concerning the issue of ex post challenges to HSR-reportable transactions more generally.

Finally, there is inherent value in the agencies clearly specifying the circumstances that must be met before they challenge a previously reviewed and cleared merger. Because of the current rarity of such after-the-fact merger challenges, once a merger has cleared the HSR process the merging parties can be almost certain that their transaction will not be subject to further agency challenge.²⁰  One potential objection to amplified agency ex post challenges to previously reviewed and cleared mergers is that this policy shift would disrupt this near-finality of the merger process and therefore diminish merging parties’ incentives to integrate and achieve planned efficiencies, to the ultimate detriment of consumers. By clearly articulating the preconditions to any ex post merger challenge, the agencies would mitigate merging parties’ uncertainties about whether their transactions would be subject to a later challenge despite clearing the HSR process.

IV. Conclusion

One of the known virtues of the Hart-Scott-Rodino Act is that it successfully established a structured mechanism through which the agencies could evaluate a merger prior to consummation for its expected competitive effects. Modern developments in econometrics have increased the precision of the agencies’ ex ante competitive effects predictions, and the agencies’ economists deftly apply those econometric tools. But merger review is a complex and uncertain exercise. As is theoretically expected and is empirically demonstrated by merger retrospectives, the agencies’ ex ante predictions may prove to be wrong in some circumstances. A merger that is expected to be competitively benign may turn out to be competitively malignant, even significantly so. What is proper antitrust policy in this circumstance? This short article, and the lengthier piece on which it is based, argue that the appropriate policy is not for the agencies to take no action, as is effectively the case now. Instead, the appropriate antitrust policy is for the agencies to challenge an anticompetitive merger ex post, even if the agencies previously reviewed and cleared the transaction pursuant to Hart-Scott-Rodino, so long as they do so in a principled manner that respects the limiting conditions discussed above.

Click here for a PDF version of the article

1 Acting (Tenure Track) Professor of Law, UC Davis School of Law. J.D., Stanford Law School, Ph.D. (Economics), University of Wisconsin-Madison.

2 See Menesh S. Patel, Merger Breakups, 2020 Wisc. L. Rev. 975 (2020).

3 For an example of the econometric analysis used in a well-known merger challenge see Orley Ashenfelter, David Ashmore, Jonathan B. Baker, Suzanne Gleason & Daniel S. Hosken, Empirical Methods in Merger Analysis: Econometric Analysis of Pricing in FTC v. Staples, 13 Int. J. of the Economics of Business 265 (2006).

4 See, e.g. Federal Trade Commission, FTC Hearing #13: Merger Retrospectives (Apr. 12, 2019), available at https://www.ftc.gov/news-events/events-calendar/ftc-hearing-14-merger-retrospectives (FTC hearing on merger retrospectives).

5 Nathan H. Miller & Matthew C. Weinberg, Understanding the Price Effects of the MillerCoors Joint Venture, 85 Econometrica 1763 (2017).

6 Department of Justice, Statement of the Department of Justice’s Antitrust Division on its Decision to Close its Investigation of the Joint Venture Between SABMiller and Molson Coors Brewing Company (June 5, 2008), available at https://www.justice.gov/archive/atr/public/press_releases/2008/233845.pdf.

7 Id.

8 See Miller & Weinberg, supra note 5, at 1764. See also Orley Ashenfelter & Daniel Hosken, The Effect of Mergers on Consumer Prices: Evidence From Five Selected Case Studies, NBER Working Paper 13859 (2008), available at https://www.nber.org/papers/w13859 (evaluating price effects of five consumer products mergers, each subject to Hart-Scott-Rodino, and finding significant price increases for groups of products in four of the five studied mergers).

9 John Kwoka, Mergers, Merger Control, and Remedies (2015).

10 See id. at 94. The standard error of that estimate is not provided. See id.

11 See, e.g. Orley Ashenfelter, Daniel Hosken & Matthew Weinberg, Did Robert Bork Understate the Competitive Impact of Mergers? Evidence From Consummated Mergers, 57 J. L. & Econ. S67 (2014) (surveying 49 studies and finding that of those 49 studies, 36 found evidence of merger-induced price increases).

12 See 15 U.S.C. § 18a(i)(1).

13 See, e.g. Investigation of Consummated and Non-Notifiable Mergers, submitted by the United States to the Organisation for Economic Co-operation and Development (Feb. 25, 2014), available at https://www.justice.gov/atr/file/788016/download, at 4-5 (providing examples of recent agency challenges to consummated mergers not subject to HSR).

14 For discussion of the few agency challenges to HSR reportable transactions, see Patel, supra note 2.

15 There is the possibility that the agencies challenge a transaction after clearing it under HSR but before the parties consummate the deal.

16 For a comprehensive analysis of agency challenges to consummated mergers, see Scott A. Sher, Closed But Not Forgotten: Government Review of Consummated Mergers Under Section 7 of the Clayton Act, 45 Santa Clara L. Rev. 41 (2004).

17 See, e.g. Kenneth G. Elzinga, The Antimerger Law: Pyrrhic Victories?, 12 J. L. & Econ. 43 (1969) (evaluating a set of pre-HSR merger challenges and finding only a few resulted in successful remedial relief).

18 For discussion of these and other potential costs, see Patel, supra note 2.

19 See, e.g. Spencer Weber Waller & Jacob E. Morse, The Political Misuse of Antitrust: Doing the Right Thing for the Wrong Reason, Antitrust Chronicle (July 2020).

20 Of course, the merged entity’s conduct could still subject it to antitrust liability, such as Section Two liability.