Static vs. Dynamic Antitrust: A Reply

June 2019

Static vs. Dynamic Antitrust: A Reply By John M. Yun (George Mason University)¹

Sanjukta Paul and Hal Singer (P&S) responded to my recent CPI Comment, “News Media Cartels are Bad News for Consumers,”² which addressed the likely harm that would arise from the proposed Journalism Competition and Preservation Act, if passed. P&S criticize my conclusion that the legislation would likely harm competition and consumers on a variety of grounds. In this short reply, I address their central points and demonstrate that our disagreement largely, but not completely, arises from P&S restricting their analysis to a static view of markets and antitrust.³

First, P&S assert that “[h]istorically, a rigid stance against horizontal coordination beyond firm boundaries is relatively new.”⁴ P&S assert further that legislators at the time of the Sherman Act, and even Senator Sherman himself, all contemplated and supported some forms of coordination. Even taking these assertions at face value, this appeal to static, historical statements taken in the late 1800s is immaterial to the economic analysis and the fundamental question of whether or not legalizing a news media cartel — that is legislatively allowed to nakedly fix prices and exclude organizations and sites that are not deemed “news content creators” — would be beneficial to society.⁵

Further, this plea to legislative intent is in line with the latest version of old arguments to overturn the current consumer welfare standard in antitrust law. The reality, however, is that there was no uniform approach to antitrust since its inception, and characterizations to the contrary involve a selective view of the complexity and lack of coherence in early antitrust jurisprudence.⁶ What is clear, however, is that the antitrust laws have, by design, always had an evolutionary character that recognizes the need to adjust to new learnings.⁷ Further, the notion that Senator Sherman and his contemporaries passed the Sherman Act so that conglomerates and multi-billion dollar operations such as the News Corp, AT&T, Viacom, the Walt Disney Corporation, the Fox Corporation, The New York Times Company, Bloomberg, and the Gannett Company could collude and fix prices along with the nearly 2,000 newspapers that comprise the News Media Alliance⁸ stretches the bounds of credulity.⁹

Second, P&S argue that “[i]t is unclear why we ought to allow Google and Facebook to serve as de facto market coordinators for huge swaths of socially significant economic activity while condemning limited, countervailing coordination rights for newspapers.”¹⁰ This statement builds to the final punchline:

Facebook’s algorithm perversely rewards click-worthy (but often unreliable) stories by moving them to the top of users’ news feed, and co-mingles sponsored content or ads alongside user-generated content in its news feed, thereby equating the quality of legitimate news and potentially fake news (not all sponsored content is fake news). This sort of anticompetitive conduct is tolerable for Professor Yun because it occurs inside the (colossal) firm boundaries of the tech platforms.¹¹

The argument appears to be that, because Google and Facebook are engaging in conduct P&S deem anticompetitive in regard to its platform design, the antitrust laws should allow “countervailing coordination rights for newspapers.” Moreover, P&S assert curiously that “[t]his sort of anticompetitive conduct is tolerable for Professor Yun.” One problem with this line of argument is that P&S nowhere establish beyond naked assertions and theorizing that Google and Facebook’s conduct with respect to their news feeds result in any market failure, much less the specific market failure — harm to the competitive process — with which antitrust is concerned.¹² Further, P&S offer up the puzzling claim that I “tolerate,” if not condone, anticompetitive conduct as long as “it occurs inside” a “colossal” tech firm, i.e. that unilateral conduct by large tech platforms is per se legal. They do so, by necessity, without reference to anything I wrote in the CPI Column or have written elsewhere. This mischaracterization is, at best, an unfortunate oversight. Finally, in an odd and ironic twist of antitrust logic, P&S would condemn as per se illegal (or at least without requiring even a scintilla of evidence of anticompetitive effects) Google and Facebook’s conduct — when from an economic perspective it should properly be assessed under a rule of reason — yet they advocate for a per se legality (via legislation) for a naked restraint on price (and also allow for naked exclusion) by an industry-wide cartel.¹³

As a brief, but relevant, aside, if the proposed coordination were indeed “limited,” an exemption from the law as it exists is absolutely unnecessary to allow news organizations to collaborate. News organizations are currently permitted to collaborate, and the FTC and DOJ’s Antitrust Guidelines for Collaborations Among Competitors provides them with guidance about danger zones and safe harbors.¹⁴ Thus, a legislative request for an exemption to allow collaboration under a procompetitive rationale is disingenuous because procompetitive collaborations are already allowed. An exemption will therefore have the effect of immunizing anticompetitive collaborations. To this point, the FTC is occasionally asked to comment on proposed bills to exempt collaborations from federal antitrust laws. In one instance, the FTC stated the following in response to proposed legislation that would immunize coordination among health care providers in Alabama — including on price:

First, the antitrust laws permit health care collaborations that do not harm consumers. As the FTC and its staff have consistently explained, many competitor collaborations – including health care provider collaborations and mergers – can be efficient and procompetitive, and are therefore lawful. Second, because the antitrust laws already permit procompetitive health care collaborations, the Bill’s purported ‘immunization’ provision would foster anticompetitive mergers, collective negotiations, and other conduct that would not pass muster under the antitrust laws. Hence, the antitrust immunity contemplated by the Bill would likely increase health care costs, diminish incentives to improve quality, and decrease access to health care services for Alabama consumers.¹⁵

Third, P&S argue that I have understated “the dominance of Google and Facebook” and that I have overstated “the countervailing power of even well-established newspapers like the New York Times.” Competition policy is ill-served by prejudging the outcome of the analysis by resorting to mere labels rather than by a careful examination of the facts. As stated in my original article, “an examination of publicly available data indicates that Google and Facebook do not account for the majority of traffic to news sites…In order to be a ‘gateway,’ a platform must be responsible for the overwhelming majority of traffic to a website, and there must be no viable alternative outlets.”¹⁶ Certainly, it can be true that specific sites might receive a great deal of referral traffic from Google and Facebook while others receive very little. Take for instance, the New York Times. P&S state: “As Professor Yun acknowledges, the New York Times depends on social media and search engines for an impressive 11 and 30 percent of its traffic, respectively.” Another way of stating this is to say that the New York Times receives 59 percent of its traffic from non-social media and non-search engines. Just for the sake of argument, let us assume that Google and Facebook account for 80 percent of all social media and search engine traffic to the New York Times, this means that it receives 67 percent of its traffic from non-Google and non-Facebook sources. This hardly qualifies as “dominance” that would justify the legalization of an industry-wide price fixing cartel to offset this power — particularly if we consider that the most common search terms are queries like the “nytimes,” which means users are specifically looking for the site.¹⁷ Further, the implied presumption is that, if news sites lost some of their traffic referrals from Google and Facebook, that they would not be able to replace that traffic. The entire basis of dynamic analysis in antitrust investigations is premised on how market participants would respond to changes in the market along with repositioning and entry. While it might entirely be true that much of the lost traffic could not be viably replaced, it is ultimately is an open question — not a conclusion.

Finally, P&S state that legalizing a media cartel would simply “affect the distribution of ad revenue between original publishers and the dominant platforms; it would not affect the price to consume news.”¹⁸ Again, this is a highly static perception of markets with the implication that Google and Facebook would not change their behavior in face of a news media cartel attempting to fix prices and exclude sites that are not deemed “news content creators.” The thought that only the distribution of ad revenue would change is a prediction that is unlikely to hold. Further, while it might not affect the nominal “price to consume news,” which presumably refers to the fact that Google and Facebook are “zero-price” platforms — it ignores the non-pecuniary costs to consumers — including convenience, speed, innovation, and other metrics of quality. Further, it ignores the potential price change to consume media on the news sites themselves. Perhaps the current price on those sites will not change (mostly, a mix of ad and subscription-based monetization) but perhaps it will. Consequently, there is no assurance that allowing a cartel to fix the price and set the terms of trade for an entire industry will not impact price and quality. Additionally, the claim by P&S that platforms such as Google and Facebook “gobble up ad revenue and add to their already enormous profits, which has no discernible consumer benefit”¹⁹ is a bold statement given the level of innovation and spending on R&D at these platforms.²⁰

In sum, the impact of the bill is to legalize an industry-wide media cartel, which can collectively fix prices and can exclude organizations that are not deemed “news content creators.” The likely result is higher content costs for these platforms, as well as provisions that will stifle the ability and freedom to innovate. In turn, this could negatively impact quality for the users of these platforms. Thus, the conclusion from the original article still holds: “There are very good reasons why antitrust jurisprudence reserves per se condemnation to the most egregious anticompetitive acts including the formation of cartels,”²¹ where the purpose is to engage in a naked restraint on price and competition.

Click here for a PDF version of the article

1 Associate Professor of Law and Director of Economic Education, Global Antitrust Institute, Antonin Scalia Law School, George Mason University.

2 John M. Yun, “News Media Cartels are Bad News for Consumers,” CPI’s North American Column, April 2019.

3 Sanjukta Paul & Hal Singer, “Countervailing Coordination Rights in the News Sector are Good for the Public (A Response to Professor Yun),” CPI’s North American Column, June 2019.

4 Supra note 3, at 3.

5 The bill states that the price fixing must be related to “the quality, accuracy, attribution or branding, and interoperability of news.” As noted in the original column, “the plasticity of this clause suggests there is no real hinderance to price fixing” (supra note 2, at 3). The bill is also exclusive to “news content creators,” which again suffers from plasticity and will likely be used to nakedly exclude certain sites from both joining the cartel and from enjoying the benefits of the cartel (via the bill’s nominal non-discrimination clause). See supra note 2, at 3.

6 See, e.g. Richard Hofstadter, The Paranoid Style in American Politics and Other Essays (1965) at 199-200. For a walk-through of the various eras of antitrust enforcement from an economic perspective, see William E. Kovacic & Carl Shapiro (2000), “Antitrust Policy: A Century of Economic and Legal Thinking,” Journal of Economic Perspectives 14, pp. 43-60.  See also Douglas H. Ginsburg (2014), “Bork’s ‘Legislative Intent’ and the Courts,” Antitrust Law Journal 79, pp. 941-951; William H. Page (2008), “The Ideological Origins and Evolution of U.S. Antitrust Law,” Issues in Competition Law and Policy 1, ABA Section of Antitrust Law, pp. 1-17.

7 See, e.g. Justice O’Connor, writing for a unanimous court in State Oil Co. v. Khan (1997) in overturning the per se condemnation of maximum resale price maintenance (“the general presumption that legislative changes should be left to Congress has less force with respect to the Sherman Act in light of the accepted view that Congress ‘expected the courts to give shape to the statute’s broad mandate by drawing on common-law tradition,’” p. 284).

8 See https://www.newsmediaalliance.org/about-us/ (“Our members represent nearly 2,000 diverse news organizations in the United States—from the largest news groups and international outlets to hyperlocal news sources, from digital-only and digital-first to print news—we represent all news media content creators.”).

9 In a related point, P&S later argue that antitrust already allows for an exemption for collective wage-setting; consequently, it appears the argument would go, we should allow for a news media cartel that fixes price. Again, equating collective wage-setting with allowing naked price fixing by conglomerates and a news media organization that claims they “represent all news media content creators” is simply not a credible claim. Further, there are more surgical means to support small businesses including direct subsidies. For instance, the federal government has a small business administration to help small businesses. The government could do something similar with small news organizations. This approach would have the benefit of being targeted precisely to the problem identified and would be harder for large media conglomerates to co-opt.

10 Supra note 3, at 3.

11 Supra note 3, at 3.

12 For more on this point, see Seth B. Sacher & John M. Yun (2017), “Fake News is Not an Antitrust Problem,” Competition Policy International, Antitrust Chronicle, pp. 28-35.

13 Of course, there is a larger debate regarding whether or not antitrust should allow counter coordination to mitigate perceived market power. See Sanjukta Paul, “Antitrust as Allocator of Coordination Rights,” UCLA Law Review, forthcoming, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3337861. (Although, this debate goes further back and is often discussed in the context of bilateral monopolies. See, e.g. Roger D. Blair & Christina DePasquale, 2014, “Bilateral Monopoly: Economic Analysis and Antitrust Policy,” The Oxford Handbook of International Antitrust Economics: Volume I, Edited by Roger D. Blair & D. Daniel Sokol.) While this debate is beyond the scope of this reply, it is important to consider that what antitrust uniformly condemns as per se illegal, as it relates to price, is naked restraints where the price fixing is not ancillary to some larger efficiency justification. See BMI v. CBS, 441 U.S. 1 (1979). Procompetitive coordination is allowed, which I discuss in the following paragraph. Finally, a fundamental premise of the counter coordination argument, in the context of bilateral monopolies, is that there is a monopsony purchaser of inputs. As discussed infra, the evidence clearly demonstrates that this is not the case here.

14 See https://www.ftc.gov/tips-advice/competition-guidance/guide-antitrust-laws/dealings-competitors.

15 See https://www.ftc.gov/policy/advocacy/advocacy-filings/2016/05/ftc-staff-comment-alabama-state-senate-regarding-hb-241-sb.

16 Supra note 2, at 3-4. See also Hunt Allcott & Matthew Gentzkow (2017), “Social Media and Fake News in the 2016 Election,” Journal of Economic Perspectives 31, pp. 221-236 at 222, who find top U.S. news sites received only 10.1 percent of their traffic from all social media sites and 30.6 percent from all search engines.

17 According to P&S, the New York Times is at the upper bound of dependency on Google and Facebook. See supra note 3, at 4 (“Because the New York Times is larger than most news publishers, it follows that the dependence of typical news outlet on Facebook and Google is even greater.”).

18 Supra note 3, at 4.

19 Supra note 3, at 4.

20 Of the top 20 companies in 2018, Google and Facebook were number two and fourteen, respectively, in R&D spending. Further, among the top 20, both Google and Facebook are in the top ten in R&D spending as a percentage of revenue. See https://www.strategy-business.com/feature/What-the-Top-Innovators-Get-Right. See also Reinhardt Krause, “Apple R&D Spending Rises But Still Trails Google, Facebook, Amazon,” May 10, 2016, https://www.investors.com/news/technology/apple-rd-spending-rises-but-still-trails-facebook-google-amazon.

21 Supra note 1, at 5.