The time has come to update our antitrust enforcement infrastructure.
There is a renewed focus on the effectiveness of antitrust policy, specifically merger review. Growing concerns about the effects of increased concentration within industry sectors and additional indicators of declining competition are prompting leaders to question the efficacy of current antitrust enforcement practices.
While many ideas for antitrust reform have been floated by academics and policymakers, a less discussed but highly practical area for improvement would be to initiate a program for routine evaluation of the decisions and actions of each of the antitrust agencies.
In the regulation of industries such as airlines, railroads, trucking, cable, telecom, financial markets, and others, ex-post evaluation studies—reviews of policy intervention after their implementation—have informed and improved regulation and deregulation.1 Even when studies come to divergent conclusions, analyses have the effect of narrowing differences and focusing the attention of policymakers and researchers on the most consequential issues.
In contrast, antitrust enforcement, specifically merger review, has no institutionalized method for systematic analyses of the effects and effectiveness of past policies. While progress has been made on general empiricism and economic theory—both necessary for greater understanding of the likely competitive effects of mergers—neither carries the weight of evidence from consistent and ongoing measurement of merger outcomes. Such a program would have substantial benefits to the enforcement agencies, to competition, and to consumers, and is long overdue.
Growing concern about merger enforcement
Recent empirical research has demonstrated the importance of systematic evaluation of merger enforcement decisions. For example, a recent analysis by economist John Kwoka (2015) compiled and analyzed all of the high-quality merger retrospectives in the economics literature.2 His meta-analysis found that a substantial majority of these carefully studied mergers resulted in sizeable price increases, implying that merger enforcement has too often failed to identify or failed to remedy anticompetitive mergers. And the price increases have been substantial, averaging about 10 percent after controlling for all other factors. Other retrospectives, such as those by Matthew Weinberg (2008)3 as well as Orley Ashenfelter et al. (2010)4 come to roughly similar conclusions.
A different approach is taken in a Federal Reserve Board paper by economists Bruce Blonigen and Justin Pierce (2016).5 They use non-public plant- and firm-level data from the U.S. Census Bureau as well as merger data from Thomson Reuters to study the effects of all manufacturing mergers on both price markups and productivity during the period from 1997 to 2007. This study concludes that on average mergers lead to increased markups, while leaving productivity and efficiencies largely unchanged. Although the paper does not directly study individual merger decisions by the antitrust agencies, it provides additional evidence of the adverse effects of concentration on competition.
Studies such as these have been important in focusing policy attention on the role or antitrust and identifying how it can, and needs to be, strengthened. However, most of the studies on which these meta-analyses are based have been undertaken by academics who operate under the extreme handicap of data constraints. Lacking access to data from the merging parties, these researchers make do with data from public sources such as the trade press or government reports, or which are purchased from private sources. While still informative, this results in an unavoidable emphasis on cases where more data are available.
In order to construct a large and comprehensive database of experiences from which to extract insights and improve future practice, something more is required. The Department of Justice (DOJ) and Federal Trade Commission (FTC) should immediately establish a program for systematic and continuous review of merger outcomes.
The case for ex-post evaluation of merger enforcement
A program for comprehensive and systematic ex-post evaluations of antitrust would have a unique potential to inform and improve policy actions. A more comprehensive effort would vastly increase the set of experiences and data on which to draw, and the range of questions that can be addressed. For example, do the agencies investigate many mergers that pose no competitive harm, as is often alleged, or rather, do they miss too many mergers that ultimately prove anticompetitive? When they determine a merger would be anticompetitive, do they take appropriate action, or are some policies—remedies, in particular—inadequate to preserving or restoring competition? Answers to such questions would likely improve the precision and accuracy of merger reviews and by extension, conserve on resources required to analyze future mergers.
Similarly, such an initiative would enable the enforcement agencies to better monitor sectors of the economy as a whole, rather than being limited to a case-by-case approach. Creating a robust history of price and other effects in particular industry sectors could also allow agencies to build a bette
r evidentiary basis from which to pursue challenges (as we shall illustrate momentarily). Finally, as the database of record on mergers and competitive effects, it would permit the agencies to improve their own understanding of competitive practices and the economy more generally, potentially even bringing into the field new researchers—a badly needed resource.
Proof of the value of this approach lies in the history of the FTC’s challenges to hospital mergers. In the early 2000s, after suffering a series of legal defeats of its efforts to stop such mergers, the FTC initiated a series of reviews of consummated hospital mergers.6 The intent was to step back from litigation until there was a better evidentiary basis from which to pursue challenges. That strategy in fact succeeded, with a string of successful court challenges.
More recently, the FTC published a review of 89 divestiture and conduct remedies between 2006 and 2012.7 This is an extension of an earlier study on divestiture effectiveness that proved important in revising and strengthening that agency’s approach to merger remedies.8 There are also international examples of ex-post review of antitrust enforcement. Notably, in the United Kingdom the Competition and Markets Authority is charged as part of its antitrust mission with conducting and reporting annual ex-post evaluations of at least two of its past actions.9 To date, two reports have been completed and made public.10 These initiatives, however, do not represent comprehensive programs of data collection and analysis. Instead, they are generally ad hoc efforts motivated by specific issues and lack the ability to answer broader questions or create large databases of past experience with antitrust enforcement.
As many respected antitrust professionals have noted, this is a glaring limitation of the current enforcement infrastructure. Dennis Carlton, former chief economist at the antitrust division of the Justice Department for example, has written that “the dearth of such [quantitative] studies and measures means that there is no reliable guide for determining whether our antitrust policy is too lax in some areas and too stringent in others.”11 Similarly, William Kovacic, then-commissioner of the FTC, has urged greater attention on the evaluation of the economic effects of enforcement decisions especially by developing better quantitative measures of actual economic effects.12 Most recently, Edith Ramirez, former chairwoman of the FTC, made a speech strongly endorsing the use of retrospectives.13 The question that remains is how best to do this.
An overview of current practice
Today, mergers are reported to the agency and go through several stages, each of which generates considerable data and other information that can answer important questions such as likely competitive effects. The numbers are instructive. Each year the two agencies undertake around 50 substantial merger investigations—that is, merger proposals subject to so-called second requests for documentation and information necessary to conduct in-depth analyses.14 These are selected on the basis of preliminary investigations indicating a substantial likelihood of competitive harm. Of these 50, about three-fourths are determined likely to create some competitive harms and trigger some enforcement action. The other 20 to 25 percent are cleared: that is, approved without change.15
Most of the mergers prompting enforcement action are in fact resolved not by court challenges but rather through remedies that essentially permit the merger to go forward subject to either divestiture or conduct constraints.16 A few additional mergers are either abandoned or substantially altered in response to likely agency challenge. Actu
al court cases challenging proposed mergers are infrequent, averaging no more than about three per year.15
Unfortunately, the effectiveness of these actions is rarely reviewed after the fact. Not doing so, however, is a missed opportunity, since each of these sets of experiences can shed valuable light on important questions about policy. Mergers that trigger a second request and serious investigation are appropriate to address questions such as the extent to which enforcement agencies investigate mergers that pose no competitive harm. Similarly, examining a sample of mergers that are not investigated—all reported mergers (about 1500 per year less the 50 subject to second requests)—can cast light on whether the agencies miss too many mergers that ultimately prove anticompetitive.17 Furthermore, examining the outcomes of mergers subject to remedies could provide insights into the effectiveness of certain remedies in preserving or restoring competition.18
The way forward
Instituting a regular ex-post evaluation of merger enforcement entails two principle challenges—data collection and resources.
With respect first to data collection, in some cases this need not be especially burdensome. Certain industries have reporting requirements that automatically provide crucial data. This includes several previously regulated industries, such as airlines, that are routinely subject to antitrust scrutiny.19 In other industries, such as pharmaceuticals, mergers are quite frequent, so that there is considerable agency expertise in place—and perhaps even past retrospectives—that could be used to guide additional reviews. These considerations would significantly relieve the burden on the agency undertaking a review.
There would, of course, be less favorable circumstances. In some industries transactions are nonpublic and data would simply have to be secured from the parties directly, or possibly from a third-party provider. In the case of an industry without past retrospectives, more time and expense would necessarily be required. Even here, however, two considerations would help limit any burden.
First, there is a now-standard technique for ex-post evaluations that is relatively straightforward.20 This technique is called “difference-in-difference” and works as follows: Data on the price of the key products where the merger might increase market power would need to be collected before and after the event. If aspects of the product such as quality or costs or innovation were at issue, data on those too would need to be compiled. Then, in order to control for other possible influences on price (or quality, etc.), similar data would have to be collected on comparable products not affected by the merger—comparable in the sense of experiencing the same demand and cost forces, but not the merger. The difference in the price of the merged product would then be netted of any price change in the unaffected (but otherwise comparable) product, and the result attributed to the merger.21 This technique sharply limits data requirements as well as various pitfalls in econometrics and economic theory.
A second helpful factor in easing the agencies’ burden is that the standards for such an evaluation study would not be those of academic publishing. Rather, these studies would be oriented toward internal use, requiring fewer of the bells and whistles that characterize published work.
Of course, internal work might be subject to concern by outside stakeholders that the antitrust enforcement agencies are biased toward results that show the agencies succeeding in their mission. To counter this concern, studies could undergo outside peer review by an academic panel or an independent agency such as the Government Accountability Office.22 Alternatively, the agencies could publicize the data they collect and employ, enabling outside reviewers to replicate and thereby validate the basic results of the internal analysis.
As noted, there will be some cases where data must be compelled. There are at least two methods by which this may be done. First, the FTC has so-called 6(b) authority under which it can require a reasonable amount of data from firms for research purposes.23 This authority is well established and regularly used, including in its own studies of remedies. Alone, it should prove sufficient for after-the-fact analysis of many mergers. The Justice Department lamentably has no equivalent authority but should investigate whether it might be able to employ some alternative method for compelling data for purposes of evaluating its policies. Secondly, both agencies could adopt a standard practice of requiring post-merger data in the substantial share of cases resolved through remedies and consent orders. Such a provision is necessary to evaluate the adequacy of the remedies imposed in future years, and as such, should be routinely included in such orders.
The second major roadblock toward instituting an ex-post review of antitrust enforcement is resources. While a permanent data collection and analysis effort may seem like a luxury to agencies that are already resource constrained, its creation would be a vital asset with outsized benefits for future policy making. Existing analytic infrastructure within enforcement agencies however, makes pinpointing the requisite investments particularly challenging. The FTC, for example, already has as many as 50 economists within the Bureau of Economics reviewing mergers and/or assisting antitrust investigations but has no formal robust data collection division or monitoring function.24 Similarly, while the DOJ has an Economic Analysis Group that analyzes the potential competitive effects of proposed mergers, it lacks any explicit mandate or system to collect and analyze outcomes and trends data.25
By contrast, other federal data collection efforts offer guidance for what would be required. The Federal Reserve, for example, has a large research staff which uses macroeconomic, financial market, and banking data to help calibrate monetary and regulatory policy.26 A separate example is the Office of Financial Research (OFR). Created under the Dodd–Frank Wall Street Reform and Consumer Protection Act, OFR was established to be the principal data collector for the Financial Stability Oversight Council and its participating regulators. OFR monitors and evaluates the impacts of financial regulations, policies, and enforcement practice.27 These models provide a sense of the likely scope and scale of the required investment.
The time has come to update our antitrust enforcement infrastructure. Numerous policymakers, academics, and institutions agree that we currently have an inadequate understanding of the effects of mergers and the effectiveness of merger policy. Similarly, several academic studies, largely limited to public data, have shown that merger enforcement has been far too permissive, allowing mergers that often prove anticompetitive. The consequences of permissiveness, such as increasing market power and consumer harm, may have broader implications for the economy overall, including effects on innovation and inequality. Regrettably, while other regulatory bodies have processes to monitor effectiveness and guide future decision-making, the same cannot be said of the antitrust enforcement practices at the DOJ and FTC. We are long past due for the creation of a comprehensive, systematic program of retrospectives of mergers and merger policy.
(Note: John E. Kwoka is the Neal F. Finnegan Distinguished Professor of Economics at Northeastern University, where he teaches and conducts research in the areas of industrial organization, antitrust, and regulatory economics.His recent research has focused on the effectiveness of merger policy in the U.S. Marc Jarsulic is the Vice President for Economic Policy at the Center for American Progress. He has worked on economic policy matters as deputy staff director and chief economist at the Joint Economic Committee, as chief economist at the Senate Banking Committee, and as chief economist at Better Markets.)
- See, for example, the large literatures cited in W. Kip Viscusi, John M. Vernon and Joseph E. Harrington, Economics of Regulation and Antitrust, Fourth Edition, The MIT Press, Cambridge, 2005.[↩]
- John Kwoka, Mergers, Merger Control, and Remedies: A Retrospective Analysis of U.S. Policy, MIT Press, 2015.[↩]
- Matthew Weinberg M. (2008), The Price Effects of Horizontal Mergers, Journal of Competition Law and Economics, vol. 4 (2), 433-447, available here.[↩]
- Orley Ashenfelter, Daniel Hoskens, and Matthew Weinberg, “Did Robert Bork Understate the Competitive Impact of Mergers? Evidence from Consummated Mergers,” Journal of Law and Economics, August 2014, available here.[↩]
- Bruce Blonigen, and Justin R. Pierce. “Evidence for the Effects of Mergers on Market Power and Efficiency,” Finance and Economics Discussion Series 2016-082. Washington: Board of Governors of the Federal Reserve System, available here.[↩]
- Edith Ramirez, “Retrospectives at the FTC: Promoting an Antitrust Agenda,” ABA Retrospective Analysis of Agency Determinations in Merger Transactions Symposium, George Washington University Law School, Washington, DC, June 28, 2013, available here.[↩]
- Federal Trade Commission Bureaus of Competition and Economics, “The FTC’s Merger Remedies 2006-2012,” January 2017, available here.[↩]
- William J. Baer, “A Study of the Commission’s Divestiture Process” (Federal Trade Commission, 1999), available here.[↩]
- United Kingdom Competition and Markets Authority, “CMA Impact Assessment 2015/16” July 14, 2016, available here.[↩]
- Ibid; United Kingdom Competition and Markets Authority, “CMA Impact Assessment 2014/15” July 21, 2015, available here.[↩]
- Dennis Carlton, “Why We Need to Measure the Effect of Merger Policy and How To Do It,” Competition Policy International, 2009, available here. It should be noted that Carlton emphasizes the need for additional data for a complete analysis of agency decision-making—data on market definition, and information on internal agency predictions at the point in time where a decision regarding possible challenge is made. Merger studies address the former question, though often accepting established market definitions, whereas the latter are nonpublic.[↩]
- William Kovacic, “Assessing the Quality of Competition Policy: The Case of Horizontal Merger Enforcement,” Competition Policy International, April 2009, available here. See also William Kovacic, “Using Ex Post Evaluations to Improve the Performance of Competition Policy Authorities,” Journal of Corporation Law, Vol. 31, No. 2, p. 503, 2006, available here.[↩]
- Edith Ramirez, “Retrospectives at the FTC: Promoting an Antitrust Agenda,” available here.[↩]
- Between FY 2006 and 2015 for example, there was on average 47 investigations in which second requests were issued. See Federal Trade Commission and Department of Justice: Hart-Scott-Rodino Annual Report: Fiscal Year 2015, August 2016, available here.[↩]
- John Kwoka, Mergers, Merger Control, and Remedies: A Retrospective Analysis of U.S. Policy. Table 2.1.[↩][↩]
- For a discussion of remedy types, see John Kwoka, Mergers, Merger Control, and Remedies: A Retrospective Analysis of U.S. Policy.[↩]
- See Federal Trade Commission and Department of Justice: Hart-Scott-Rodino Annual Report: Fiscal Year 2015.[↩]
- If the question concerns mergers overall, of course, none of these suffices: the pre-merger notification requirement covers only mergers with a value in excess of about $80 million. Those total around 1500 per year at present, the vast majority of which are competitively harmless. See Federal Trade Commission and Department of Justice: Hart-Scott-Rodino Annual Report: Fiscal Year 2015.[↩]
- Most commercial airlines in the U.S. are required to report on a quarterly basis a ten percent sample of tickets. Information on passenger count, prices, and much else is compiled and released by the Department of Transportation. See United States Department of Transportation, Bureau of Transportation Statistics, available here.[↩]
- See OECD Reference Guide on ex-post Evaluation of Competition Agencies Enforcement Decisions, April 2016. available here.[↩]
- Thus, if the pre-merger price of the product were 100 and its post-merger price rises to 110, this ten percent price increase would have to be adjusted for other factors. If the prices of comparable products not affected by the merger rose by three percent during the same time period, one can conclude that the merger is responsible for a seven percent price increase. This technique is known as difference in differences.[↩]
- The OECD now organizes an annual program for review of evaluations undertaken by various competition authorities. See OECD, “Workshop on ex-post evaluation of enforcement decisions by competition authorities” available here.[↩]
- Federal Trade Commission, “A Brief Overview of the Federal Trade Commission’s Investigative and Law Enforcement Authority” July 2008, available here.[↩]
- Federal Trade Commission, “About the Bureau of Economics,” available here.[↩]
- The United States Department of Justice, “Economic Analysis Group”, available here.[↩]
- Board of Governors of the Federal Reserve System, Economic Research & Data, available here.[↩]
- Office of Financial Research, “Strategic Plan Fiscal Years 2015-2019” available here.[↩]