The Risks of Radicalism: Exacerbating Harms from Type I Errors

By Koren W. Wong-Ervin, Anne Layne-Farrar & James Moore¹

Calls to radically change U.S. antitrust law continue to be a focus of law and policy makers. According to proponents of the proposed changes, drastic legislative amendments are necessary to remedy the (perceived) failures of current antitrust standards to prohibit anticompetitive conduct, in particular in high-technology markets. While efforts to address market failures are certainly worthy of discussion, the various legislative proposals risk serious adverse consequences, including higher prices for consumers and reduced innovation and consumer choice.

This article begins with a brief discussion of the economic basis for regulation, followed by a summary of recently proposed legislation from Senators Elizabeth Warren and Amy Klobuchar. The main part of this article is an exploration of the risks posed by these draft bills through a retrospective examination of market developments following past interventions by antitrust agencies.

I. The Economic Basis for Regulation

As an initial matter, it is worth noting that antitrust law is not regulation. Rather, it is proscriptive and not prescriptive in nature, with a default of legality. This is important because, among other things, creating ex ante regulation to prevent certain conduct risks sacrificing the efficiencies and other benefits of that conduct by imposing potentially rigid rules that lack the flexibility of existing antitrust rule of reason assessments. One of the main benefits of relying on existing antitrust laws is that they proceed primarily through fact-specific case-by-case analyses, which are more likely to maximize consumer welfare than are ex ante regulations.

That said, the theoretical basis for economic regulation rests on the idea that regulation may serve to improve the allocation of resources in a particular industry compared to the outcome in the absence of regulation. Successful identification of a market failure is a necessary but not a sufficient condition to justify regulation on economic grounds. Once a potential market failure has been identified, the proposed regulatory solution must itself survive a rigorous economic cost-benefit analysis, one that factors in the potential for imperfect regulation and unintended consequences as well as the effect of alternative solutions based on private ordering.²

When considered in this light, there are several problems with the contentions that we have a market concentration problem. For example, because economy-wide statistics inevitably aggregate economic phenomena across product and geographic lines, they can grossly overstate concentration in well-defined antitrust markets. Indeed, as then Department of Justice (“DOJ”) Chief Economist Luke Froeb and Senior Economic Counsel Greg Werden explained, the “key evidence” underlying the increasing industry concentration contention and the subsequent calls for antitrust reform is data from the U.S. Census Bureau.³ “But these data do not demonstrate increasing concentration of markets, i.e. ranges of economic activity in which competitive processes determine price and quality, and in which the impact of mergers and trade restraints are evaluated in antitrust law.”⁴ Instead, “Census data relate to aggregations of economic activity much broader than markets.”⁵

In analyzing the “issue brief” put out by the Obama Administration’s Council of Economic Advisors (CEA) (the principal study relied upon by those claiming we have a concentration problem), the DOJ economists concluded that:

[E]ven the least aggregated Census data can be over a hundred times too aggregated, yet the CEA used the most aggregated Census data. It principally cited the change in the 50-firm concentration ratio for 13 broad sectors of the U.S. economy, such as retail trade. We agree with Carl Shapiro, a member of the CEA during the Obama Administration (2011– 12), that these data are “not informative regarding the state of competition.”⁶

The economists also found that “[r]eliable data on trends in market concentration are available for only a few sectors of the economy, and for several, market concentration has not increased despite substantial merger activity.”⁷

More importantly, even if we had a concentration or competition problem, aggregate statistics are ultimately tangential, or even irrelevant, to the question of whether alleged conduct is actually anticompetitive. Indeed, there is great risk in equating concentration with harm. Fears about concentration ignore its benefits, including economies of scale, self-financing, ability to take and survive risks, and multilevel integration. As Professor Steve Berry has explained: “[P]roduct quality is going up. That’s pushing price up. That pushes margins up. The marginal cost is going down as firms get better logistics and locate closer to their customers. Marginal cost is falling. That’s efficiency. But markups go up.”⁸

II. Current Proposed Legislation and Risked Unintended Consequences

Recent draft legislation from 2020 includes a comprehensive bill by Senator Elizabeth Warren⁹ and a narrower bill focused on exclusionary conduct by Senator Amy Klobuchar.¹⁰ The Warren bill would dispense entirely with the existing case-by-case fact-specific analysis of competitive effects in favor of outright bans on so-called “mega mergers,”¹¹ and includes presumptions of illegality for “large mergers.”¹²

The Warren bill would also reintroduce long-ago abandoned presumptions of illegality for commonplace¹³ vertical restraints such as tying and bundling, even when practiced by firms without substantial market power, i.e. the ability to raise market prices above or reduce output below competitive levels for a significant period of time. The Warren bill would abandon the well-accepted definition of substantial market power and would instead interject vague notions of “fairness” to allow for a finding of “market power” based upon factors such as “directly or indirectly impos[ing] an unfair purchase or selling terms or any other unfair trading condition.”¹⁴ The Klobuchar bill would create similar presumptions of illegality based upon a market share threshold of 50 percent. In order to overcome these presumptions, both bills would require firms to accomplish the near impossible task of proving a negative in terms of harm to competition. Specifically, the Warren proposal would require firms to prove “through clear and convincing evidence that the conduct does not materially harm competition or the competitive process.” It would also prohibit courts from balancing “procompetitive efficiencies with anticompetitive impacts.”¹⁵ Similarly, the Klobuchar bill would require firms to prove that their exclusionary conduct does not present an “appreciable risk of harming competition.”¹⁶

Among other things, these bills would overturn decades of Supreme Court precedent in which the Court has held that vertical restraints must be evaluated under a full-blown, effects-based (or rule of reason) analysis.¹⁷ The burden shifting in favor of presumptions of illegality would also eliminate competition on the merits as a viable defense. In addition, the bills would require decades of judicial decisions to decide what the new language means. For example, what is “appreciable risk”? Is it negligible risk, or is the bar higher? Finally, the bills would also be contrary to the robust body of empirical evidence, including leading meta-studies from economists at the DOJ and Federal Trade Commission (“FTC”), indicating that vertical restraints and mergers are generally procompetitive or benign.¹⁸

Most importantly, the proposed legislation risks serious harm to consumers, including an increased danger of higher prices from the greater risk of type I errors (false positives). As the Supreme Court has recognized, courts face limitations in distinguishing between pro- and anticompetitive conduct in antitrust cases and has emphasized the need to avoid type I errors, particularly in monopolization cases.¹⁹ The Court has also expressed concerns, originally explained in Judge Frank Easterbrook’s seminal analysis, that the cost to consumers arising from type I errors might be greater than those attributable to type II errors (false negatives) because “the economic system corrects monopoly more readily than it corrects judicial errors.²⁰ Such imbalances are likely to be exacerbated under the two Senator-proposed bills that would abandon the rule of reason analysis required by the Supreme Court in favor of outright bans and presumptions of illegality. In other words, even under the existing case-specific analysis, there are bound to be errors (in both directions), particularly when the analysis involves predicting future market dynamics. Such errors would likely only increase under rigid prohibitions or presumptions that dispense with the fact-specific analysis required to understand whether a particular merger or type of conduct actually harms (or, in the case of mergers, is likely to harm) competition and consumers.

III. Evidence from Historical Experience

Evidence of the possible consequences of antitrust intervention — whether through merger control or conduct investigations — can be traced from the “old economy” era following passage of the Sherman Act in 1890 through to the modern digital economy. In several cases, the U.S. Antitrust Agencies brought enforcement actions to block transactions based upon predictions that the deals would entrench a market leader. Yet, evidence of what actually happened in the real world illustrates the difficulties of accurately predicting future market realities. While it is difficult to predict and model the possible consequences of a government intervention since the but-for world is unknowable, these examples do highlight the need for great caution before adopting legislation that is likely to exacerbate the possible detrimental consequences arising from false positives.

One example is the FTC’s 2000 decision to block the merger of Milnot Holding and H.J. Heinz on the grounds that it would have resulted in a reduction of the number of competing firms from three to two. A retrospective analysis conducted by an FTC economist revealed that, in the eight years following the FTC’s decision to block the merger, market concentration and the prices offered by the market leader significantly increased even though the merger had been blocked.²¹ One possible explanation (predicted by the parties) is that the merger would have created a more effective second competitor to take on the industry leader. In other words, the FTC’s intervention deprived the market of a strong #2 competitor, which could have disciplined prices to the benefit of consumers. For example, the merging parties had argued that the greater market penetration enabled by the combination of the parties’ brands was necessary to recoup investments in advertising that are essential to launching new products.²² Without this market penetration, Heinz predicted that it would be unable to chip away at Gerber’s significant market share.

An example discussed in a theoretical paper written by former FTC Chairman Tim Muris and former FTC General Counsel Jonathan Neuchterlein is the DOJ’s 1944 case against the Great Atlantic and Pacific Tea Company (more commonly known as A&P), whose executives faced charges for criminal violations of the Sherman Act.²³ The district court convicted A&P on two core theories of liability: (1) that A&P engaged in predatory low-cost pricing to drive out its rivals²⁴; and (2) that the company obtained an unlawful efficiency advantage through vertical integration.²⁵ According to the district court, these practices resulted in “unreasonable advantages” over competing grocery stores and higher costs to those stores.²⁶ As Muris & Neuchterlein explain, the ultimate result of this prosecution may have been to discourage low-cost distribution practices and low pricing due to the risk of criminal exposure — to the detriment of consumers.²⁷ These concerns regarding enforcement actions remain today. For example, in the modern economy, grocers like Wal-Mart, Costco, and Sam’s Club have pursued vertical low-cost, vertical integration — in other words, much of the same conduct that A&P was accused of many decades ago. Yet the Klobuchar bill would eliminate the Supreme Court’s Brooke Group test for predatory pricing — the very test that makes prosecutions like that against A&P more difficult and that otherwise limits overenforcement against low-cost pricing.

There is also the FTC’s 2004 decision to challenge Blockbuster Video’s proposed acquisition of Hollywood Video, which the parties sought at least in part to address new competition, including from Netflix.²⁸ Contrary to the FTC’s prediction that the merger would have entrenched Blockbuster as the market leader, Blockbuster went bankrupt within five years of the parties’ decision to abandon the deal in response to the FTC suit to block it.²⁹ Retrospective analyses of Blockbuster’s fall reveal that Blockbuster was unable to adapt to the advantages of the new Netflix business model, which included lower operational costs without brick-and-mortar retail stores, lower distribution costs associated with direct-to-customer DVD shipping, a greater variety of movie selections, and the rise of online streaming following the growth of broadband access.³⁰ The FTC’s analysis failed to give sufficient weight to these market developments.

Again, while it difficult to predict and model the possible consequences of government intervention, according to the merging parties “[a]n acquisition of Hollywood would [have] allow[ed] Blockbuster to immediately accelerate its plans to bring its expanded array of offerings to more consumers through an accretive acquisition of stores.”³¹ In other words, the merger could have enabled Blockbuster to continue expanding consumer choices by enabling it to immediately expand retail locations without additional capital expenditures. Another possible result of allowing the merger is that it would have given Blockbuster the tools to keep up with Netflix. This would have given consumers one more online option and kept mail-order video services — which Blockbuster was in the process of launching³² — and potentially streaming services, more competitive.

As another example of the difficulties of predicting future market realities, particularly in fast-moving technology markets, consider the FTC’s 2000 decision to impose remedies as a condition of allowing AOL’s merger with Time Warner. Contrary to the FTC’s prediction that the merger would enable AOL to become the leading provider of broadband Internet access,³³ AOL quickly faded with the rise of other broadband providers.³⁴ The FTC presumed, based on historical experience, that AOL would carry its dominance in dial-up forward to broadband. The cable companies, however, surpassed AOL by building powerful broadband networks while AOL continued to rely on outdated dial-up technologies. By the time AOL introduced its own broadband service at a premium, it was too late.³⁵

Yet another example can be seen in the DOJ’s 2000 decision to condition AT&T’s (which owned Excite@Home, one of the largest providers of broadband Internet access) acquisition of MediaOne on its divestiture of MediaOne’s interests in Road Runner, a web-portal provider that competed with Excite@Home.³⁶ Following the DOJ’s remedy, Excite@Home declare bankruptcy despite the DOJ’s predictions.³⁷ In other words, the DOJ’s prediction that the combined company’s interests in the two Internet portals would give the company “undue leverage in its dealings with broadband content providers”³⁸ failed to appreciate the nature of dynamic markets in which new entrants can, and often do, overtake incumbents by rendering once-dominant business models (in this case, the walled-in environment model of AOL, Road Runner, and Excite@Home) antiquated.³⁹

IV. Conclusion

The examples discussed in this article illustrate that, even under existing fact-specific effects-based analyses, errors occur. Adopting legislation that would prevent antitrust enforcers from conducting careful analysis would deprive the agencies of the ability to minimize the false positives that risk harm to consumers and competition.

Click here for a PDF version of the article

1 Koren W. Wong-Ervin is a Partner at Axinn, Veltrop, & Harkrider LLP (Axinn) and former Counsel for IP & International Antitrust and Attorney Advisor to Commissioner Joshua D. Wright at the U.S. Federal Trade Commission. Anne Layne-Farrar is a Vice President at Charles River Associates (CRA) and an Adjunct Professor at Northwestern University Pritzker School of Law. Jimmy Moore is an Associate at Axinn. The authors thank Greg Werden for his insightful comments. The views expressed here are the authors’ alone and do not necessarily represent the views of Axinn, CRA, or any of their clients. Axinn represents a number of clients, including Google, that may have an interest in the subject matters discussed in this article. While CRA may have clients with an interest in the subject covered here, Dr. Layne-Farrar is not representing any of them.

2 Koren W. Wong-Ervin et al., Comment of the Global Antitrust Institute, George Mason University School of Law, on the European Commission’s Public Consultation on the Regulatory Environment for Platforms 9-10 (December 29, 2015), http://masonlec.org/site/rte_uploads/files/GAI_Comment%20on%20EC%20Platform%20Consultation_12-29-15_FINAL.pdf. See also Harold Demsetz, Information and Efficiency: Another Viewpoint, 12 J. L. & ECON. 1, 1-22 (1969).

3 Gregory J. Werden & Luke M. Froeb, “Don’t Panic: A Guide to Claims of Increasing Concentration,” Antitrust Magazine, Forthcoming; Vanderbilt Owen Graduate School of Management, 1 (2018).

4 Id. at 1.

5 Id. at 1.

6 Id. at 1.

7 Id. at 1.

8 Transcript at 7, The Current Landscape of Competition and Privacy Law and Policy, Fed. Trade Comm’n Hearings on Competition and Consumer Protection in the 21st Century (September 13, 2018). In addition, there is very little empirical basis to presume any systematic relationship between market structure, competition, and innovation. While there is credible causal evidence that market incentives matter, the empirical literature attempting to link market structure — typically measured by the number of firms or market shares in broadly defined markets — and product market competition to innovation are based on cross-section analyses that do not produce casual inference and as a whole yield inconclusive results.

9 Anti-Monopoly and Competition Restoration Act, 116^th Cong. (2020). Draft Bill on file with authors. The draft bill addresses a wide variety of issues related to merger control, conduct issues, and agency enforcement authority, including: banning all “mega mergers”; imposing stringent additional procedural requirements on “large mergers”; shifting the (heightened) burden of proof as to all other mergers on the merging parties; narrowing consideration of merger efficiencies; requiring notifications to state attorneys general; mandating review, and potentially unwinding, of all “mega mergers” from the last twenty years; expanding individual liability, particularly for CEOs; severely increasing the categories of conduct constituting restraints of trade or monopoly practices; expanding the FTC’s remedial authorities; and overturning Supreme Court precedents that heightened standards for pleadings and class certification.

10 Klobuchar Introduces Legislation to Deter Anticompetitive Abuses, Klobuchar.Senate.Gov (March 10, 2020), https://www.klobuchar.senate.gov/public/index.cfm/news-releases?ID=E59886E1-12EE-48A5-94F5-044658A75513. Anticompetitive Exclusionary Conduct Prevention Act of 2020, S. 3426, 116th Cong. § 4 (2020). Senator Klobuchar has also previously introduced a bill that would make unlawful any merger that results in more than de minimis competitive effects and would shift the burden to the merging parties to show that the merger would be procompetitive. See Consolidation Prevention and Competition Promotion Act of 2019, S. 307, 116th Cong. § 3 (2019).

11 “Mega mergers” are defined as mergers in which either the acquiring entity or acquired entity has annual revenue of more than $40 billion; both entities have annual revenue of more than $15 billion; the post-merger entity would have market shares of greater than 45 percent as a seller or 25 percent as a buyer; and the transaction would result in fewer than four significant competitors in the marketplace. The bill carves out an exception for certain transactions that meet the “failing firm” standard.

12 “Large mergers” are defined as mergers in which either the acquiring entity or acquired entity has annual revenue of $5 billion to $40 billion; both entities have annual revenue of $1 billion to $15 billion; the post-merger entity would have market shares of 10 percent to 45 percent as a seller or 10 percent to 25 percent as a buyer; the transaction would result in fewer than five significant competitors in the marketplace; and during the preceding seven-year period, either party has been found to have violated the antitrust laws.

13 Vertical restraints are ubiquitous and widely used by a variety of firms, including those without substantial market power. Given that the potential to harm competition and generate anticompetitive effects arises only when vertical restraints are practiced by a firm with substantial market power, the fact that such conduct is commonplace even for firms without the ability to exclude is a strong indication that there are legitimate business justifications for the conduct that have nothing to do with seeking to exclude competitors.

14 Anti-Monopoly and Competition Restoration Act, 116^th Cong. § 6(a) (2020).

15 Anti-Monopoly and Competition Restoration Act, 116^th Cong. § 6(a) (2020).

16 Anticompetitive Exclusionary Conduct Prevention Act of 2020, S. 3426, 116th Cong. § 4 (2020).

17 See, e.g. Continental T.V., Inc. v. GTE Sylvania, Inc., 433 U.S. 36, 54–55 (1977) (concluding that vertical territorial restrictions reduce intrabrand competition but “promote interbrand competition by allowing the manufacturer to achieve certain efficiencies in the distribution of its products”); State Oil Co. v. Khan, 522 U.S. 3, 18 (1997) (holding that maximum resale price maintenance agreements are subject to the rule of reason, not per se rule, because “the per se rule [as applied to maximum RPM agreements] could in fact exacerbate problems related to the unrestrained exercise of market power by monopolist dealers” and thus “may actually harm consumers and manufacturers”); Leegin Creative Leather Prods., Inc. v. PSKS, Inc., 551 U.S. 877, 890 (2007) (holding that minimum RPM agreements are also subject to the rule of reason due to promotion of interbrand competition by “encourag[ing] retailers to invest in tangible or intangible services or promotional efforts that aid the manufacturer’s position as against rival manufacturers” and by potentially “giv[ing] consumers more options so that they can choose among low-price, low-service brands; high-price, high-service brands; and brands that fall in between”); Jefferson Parish Hosp. Dist. No. 2. v. Hyde, 466 U.S. 2, 15–16 (1984) (limiting per se treatment in tying cases to those in which there is an element of “forcing” and a substantial volume of commerce is foreclosed).

18 For a summary of the leading meta-studies indicating that verticals are generally procompetitive or benign, see Koren W. Wong-Ervin’s 2019 article at https://www.americanbar.org/content/dam/aba/publishing/antitrust_source/2018-2019/atsource-february2019/feb19_wong_ervin_2_18f.pdf.

19 Pac. Bell Tel. Co. v. LinkLine Commc’ns, Inc., 555 U.S. 438, 451 (2009) (“To avoid chilling aggressive price competition, we have carefully limited the circumstances under which plaintiffs can state a Sherman Act claim by alleging that prices are too low.”); Credit Suisse Sec. (USA) LLC v. Billing, 551 U.S. 264, 283 (2007) (“[W]here the threat of antitrust lawsuits, through error and disincentive, could seriously alter underwriter conduct in undesirable ways, to allow an antitrust lawsuit would threaten serious harm to the efficient functioning of the securities markets.”); Verizon Commc’ns Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 U.S. 398, 414 (2004) (“Mistaken inferences and the resulting false condemnations are especially costly, because they chill the very conduct the antitrust laws are designed to protect.” (internal quotations omitted)).

20 Frank H. Easterbrook, The Limits of Antitrust, 63 Tex. L. Rev. 1, 15 (1984)

21 Viola Chen, The Evolution of the Baby Food Industry 2000-2008, Working Paper No. 297 (2008) (finding that Gerber’s prices of the proposed target firm rose by 3 percent in the years following the merger even after adjusting for inflation).

22 See Fed. Trade Comm’n v. H.J. Heinz Co., 246 F.3d 708, 722–24 (D.C. Cir. 2001).

23 See generally Timothy J. Muris & Jonathan E. Nuechterlein, Antitrust in the Internet Era: The Legacy of United States v. A&P (George Mason Law & Economics, Research Paper No. 18-15, 2018).

24 The decision predated the current predatory pricing test established in Brooke Group by several decades.

25 United States v. N.Y. Great Atl. & Pac. Tea Co., No. 67 F. Supp. 626 (E.D. Ill.), aff’d, 173 F.2d 79 (7th Cir. 1949).

26 See, e.g. id. at 642, 658.

27 See, e.g. Muris & Neuchterlein, supra note 22; see also Marc Levinson, The Great A&P and the Struggle for Small Business in America (2011). As Muris and Neuchterlein explain, there can be a variety of explanations for A&P’s fate, with one plausible one that, in addition to whatever chilling effects the DOJ’s actions might have had, A&P (unlike its rivals) was distracted for a lengthy period in defending against the prosecution and other government reviews. See id. at 18–19.

28 Mike Musgrove, Blockbuster Ends Effort to Acquire Competitor, Wash. Post (March 26, 2005).

29 Michelle Lou and Saeed Ahmed, There’s Now Only One Blockbuster Left on the Planet, CNN (March 6, 2019).

30 Greg Satell, A Look Back at Why Blockbuster Really Failed and Why It Didn’t Have To, Forbes (September 5, 2014).

31 Press Release, Blockbuster to Launch Cash Tender Offer for Hollywood Entertainment, Blockbuster Inc. (December 28, 2004).

32 Martin Peers & Nick Wingfield, Blockbuster Set to Offer Movies by Mail, Wall St. J. (February 11, 2004).

33 Compl. ¶ 8, In the Matter of America Online, Inc. and Time Warner Inc., FTC Dkt. No. C-3989 (December 14, 2000).

34 Rob Lever, Verizon Buys Faded Internet Pioneer AOL for $4.4 Bn, Yahoo News (May 12, 2015).

35 Thomas W. Hazlett, A Lesson for Today’s Tech Trustbusters, Wall St. J. (January 9, 2020); Saul Hansell, As Broadband Gains, the Internet’s Snails, Like AOL, Fall Back, N.Y. Times (February 3, 2003).

36 Press Release, Justice Department Requires AT&T to Divest Media One’s Interest in Road Runner Broadband Internet Access Service, U.S. Dep’t of Justice (May 25, 2000).

37 Matthew Fordahl, Excite@Home Files Bankruptcy Papers, L.A. Times (September 29, 2001).

38 See supra note 36.

39 Bob Fernandez, A Clue in 2000? Ruling, Phila. Inquirer (September 21, 2014).