I am delighted to announce that this month’s guest article is authored by Thomas B. Nachbar, Professor of Law at the University of Virginia School of Law. Tom is making the point that primary implications behind recent proposed antitrust reforms are not economic, but political. Tom, thank you very much! All the best, Thibault Schrepel
An America Fit for the Digital Age?
Commentators have made much sport, here and elsewhere, of evaluating recent proposed antitrust reforms for their likely economic effects. As commentators like George Priest, Doug Melamed and Nicolas Petit, and others have pointed out there is much to be concerned about in the economic understandings underlying recent proposals. I think many of these concerns are valid or, at least, that they haven’t been adequately responded to by reformers, and I look forward to seeing those responses in the days to come.
My point here is to suggest that many of the criticisms of reform might be missing the central point of reforms: that their primary implications are not economic, they are political. The effects we should be worried about are not the effects on efficiency or output but on the relative power of regulators and markets.
Although dominated by economic analysis (some would say since Justice Brandeis’s opinion in Chicago Board of Trade), antitrust law has always been about allocating control over markets between the public and the private. Antitrust laws are forms of commercial regulation that have always had a political cast.
That is a message that has been vigorously taken up of late, albeit in a very limited form. Many of the concerns of antitrust reformers are political, but they nevertheless remain rooted firmly in the economics of markets. Floor speeches and books by senators highlight the danger of political influence wielded by firms that have grown too large. These concerns, which echo those of Robert Pitofsky and others, and even Brandeis himself, seek to directly connect economic power to political power, set of concerns Arthur Austin captured in the term “societal antitrust.” Antitrust laws, under this view, maintain a balance of political power by preventing excessive concentration of political power. Those who would include concerns over the interests of labor in antitrust analysis similar seek to expand the relevant considerations, a point that Richard Epstein has commented on in his own recent contribution to Concurrentialiste.
While I applaud the extension of the analysis beyond economics alone, even that wider aperture misses much of the political dimension of antitrust. For one thing, U.S. antitrust law has never been particularly well-suited to prevent the accumulation of wealth. Even Learned Hand’s approach to Section 2 liability in Alcoa would have allowed the tremendous accumulation of wealth in Alcoa if it occurred “by virtue of  superior skill, foresight and industry.” For another, it’s not at all clear that such concerns can effectively be stated in economic terms. Leading antitrust economist Kenneth Elzinga, among others, has amply pointed out the ways that “equitable” goals like wealth redistribution are a poor fit for antitrust. But my point is not Elzinga’s (or Epstein’s on labor) that antitrust analysis cannot successfully be expanded to include these broader considerations, although I think their points are well-taken.
My point is that antitrust itself should primarily be viewed through the lens of control, not economics. Again, I am not necessarily speaking of the “usual trade-off between the State and the Market” described by Alexandre de Streel in his own contribution to Concurrentialiste. That tension certainly exists, and will always exist. What I think is best demonstrated, though, by the current reform movement, is not only a reconsideration of the balance of power between private and public actors (which I take to be de Streel’s concern) but also the forms of control available to each. The variety of recent reforms and proposals for reform U.S., the E.U., and China provide a prime example.
Although U.S. antitrust reformers are quicker to look at the E.U. for examples, China might be a better model. I say that not because of China’s record of confusing the line between economic and political control, as demonstrated by China’s seemingly arbitrary enforcement attitude toward regulating its own Internet industry, but because the risk of similar arbitrary enforcement seems so unlikely in liberal constitutional governments like those in the E.U., the U.K., and the U.S. One can be concerned about the possibility that Lina Khan has specifically targeted Amazon, a concern understandably fueled by some of the political rhetoric underling the societal antitrust movement, but no one seriously thinks the FTC is going to investigate Amazon’s business practices because of some message contained in season 3 of The Boys. The relative comfort we have with the FTC’s regulatory stance is not the product of U.S. antitrust law (although U.S. antitrust law itself respects those concerns); it’s the product of larger features of the U.S. legal landscape, most notably the First Amendment but also a set of regulatory practices that have grown up in the U.S. legal context. The First Amendment would indeed protect Amazon from interference over its production of content, but it’s not clear it would need to because the culture that dominates U.S. competition authorities would likely keep them from going in that direction in the first place. Some competition-related rhetoric does raise problems of the overlap between competition and speech regulation. The concerns voiced in the Republican Staff Report over bias at large tech platforms sound in content regulation, but their they are notable in part for their failure to gain traction, and given their asymmetrical stance (with a concern over bias in a particular partisan direction), it seems hard to imagine them ever doing so.
If China is notable as a counter-example on regulatory practice, the E.U. is providing an example in regulatory form. Sen. Amy Klobuchar’s Competition and Antitrust Law Enforcement Act, introduced in February of 2021, followed a more traditional approach to U.S. antitrust law by seeking to alter the general legal standards applied in antitrust cases, with some specific references to platforms (including some that I’ve suggested are extremely problematic). Such congressional alterations have been a feature of U.S. antitrust law since the Clayton Act in 1914. The Clayton Act changed the legal standard applicable to mergers (which had been actionable under the Sherman Act) and also established restrictions on specific types of agreements, such as exclusive dealing, but it did so by incorporating principles from antitrust law. The Clayton Act highlighted certain transactions, but it did not fundamentally change the U.S. approach to antitrust law, which was to evaluate most transactions by their effect on “competition.” That has led to frustration by many, who lament the courts’ failure to adopt a view of competition (or harm to competition) that comports with their own.
More recent examples, such as the American Choice and Innovation Online Act, seem to be following the approach of the EU Digital Markets Act, which provides both more specific criteria for coverage (such as revenue of market capitalization thresholds) along with a much more detailed list of prohibited practices. It is significant enough for U.S. regulators to look on the E.U. for suggestions on competition law given the different traditions underlying each, but what is more important than the substantive implications are those for their form: the conduct covered by the DMA is highly specific. U.S. antitrust law has largely eschewed such a specific approach to general competition law.
This shift toward greater specificity, which reflects a broader distinction in law between rules and standards, can be looked at a number of different ways. Many would look at it as a shift to “ex ante” rules and away from “ex post” decisionmaking, and it could even lead to more predictable outcomes. This is purportedly the thinking behind the U.S. DoJ/FTC Joint Merger Guidelines, which, when coupled with pre-merger review, provide a guide to companies in planning mergers. When it comes to the DMA, and any potential U.S. variant, though, given how long it takes to adjudicated a case in the E.U. (or the U.S.), the new rules are unlikely to lead to any kind of reliable specificity anytime soon.
Instead, a more helpful way to think about increased specificity is not in terms of reliability but rather as an attempt shift the locus of decision-making authority. Highly specified rules shift decision-making authority from rule appliers (like judges) to rule-makers.
The identity of rule makers, though, varies, and that is where things really get interesting for antitrust. In the E.U., specific rules shift authority from courts to the Commission, since, while legislation is adopted by act of the European Parliament and the Council, it is on recommendation of the Commission. In the U.S., the allocation of regulatory authority is more fluid. Congress can shift authority to itself, or it might shift authority to an administrative agency (like the Federal Trade Commission) either through the power to initiate actions to enforce a clear rule or by giving the agency itself the power to promulgate its own rules. Unsurprisingly, the FTC Chair Lina Khan and previous FTC Commissioner Rohit Chopra have previously argued for just such rulemaking authority for the FTC. Some would skip the substantive question entirely and just focus on changing the judges themselves.
I don’t think any of this is news to anyone; the moves in the European Commission, Congress, and the FTC have been watched by many. My point is that they should be viewed not for their likely effects on the economy (which is unknowable at this point) but for their implications for regulation.
There is perhaps no better example of this than the European Commission’s proposed Digital Services Act. Although similarly named to the Digital Markets Act, promulgated by the same Commissioner (Margrethe Vestager, whose portfolio includes competition), and although the European Commission’s website describes both has having the same two main goals, the DSA is a very different animal from the DMA.
Unlike the DMA, the DSA is primarily concerned with the flow of information, including “illegal content” (content “not in compliance with Union law or the law of a Member State”) and, with regard to “very large online platforms,” the risk that they will be used either to suppress the exercise of fundamental rights or be manipulated,
including by means of inauthentic use or automated exploitation of the service, with an actual or foreseeable negative effect on the protection of public health, minors, civic discourse, or actual or foreseeable effects related to electoral processes and public security.
This is not your grandfather’s competition law; there is good reason why Margrethe Vestager’s portfolio has expanded beyond competition to building “A Europe Fit for the Digital Age.”
The instinctive American response to regulation of information flows is to fall back on the First Amendment. We are, after all, the country that applied the First Amendment to the placement of newsracks. Tempting as it is to emphasize American exceptionalism by playing the First Amendment card, I want to move beyond that response to a different analogy: broadcast media regulation by the Federal Communications Commission.
Before the FCC became the world’s largest auction house (in 2021, Christie’s reported $7.1 billion in sales; a single spectrum auction in 2021 netted the FCC $21.8 billion), it regularly reviewed the content of American broadcasters, not only for decency, but also for “fairness,” defined to include the obligation to provide coverage that fairly reflects differing viewpoints. In 1987, the FCC abandoned the pursuit of fairness, but it continued regulating broadcast licensees under the “public interest” standard codified in the Communications Act, including inquiry into such things as the degree to which broadcast stations were serving local, or children’s, interests.
Again, my point is not that such regulation is or should be constitutionally problematic, it is that it was the implementation of a “regulatory” approach to communications, one that is reflected in the DSA. That is, while I think it is hard to imagine either European or American regulators discriminating on the basis of content (as Chinese regulators almost certainly do), it is rather easy to imagine the adoption of a similarly “regulatory” approach (which is to say, persistent regulatory oversight by a body with attenuated political accountability, as was the FCC) to Internet platforms in both the U.S. and in Europe.
It is through the lens of regulation, not the specific competition policies at issue, that we should be looking at the various reforms being proposed in Congress and the EU. The most convincing argument in favor of permissive antitrust rules has always been that most anticompetitive practices are self-correcting. That is so because forms of control available to firms largely must operate through the market, and so the market provides a correcting force to many harmful practices. The key to dealing with harmful conduct by firms is to identify those (such as cartel) that are not self-correcting through market forces, and antitrust should be (and is) most concerned with those. The answer by antitrust skeptics to the populist cry to curb the power of large firms is that the ultimate exercise of power occurs not in firms but in markets. Thus, contra Senator Klobuchar, Alexandre de Steel correctly identifies the struggle as “between the State and the Market,” not “between the State and the Firm.” But the forms of government control are different and require a different correcting force, and it has been skepticism over the ability of government to either identity or correct its errors – not the substance of any particular antitrust rule – that has dominated U.S. skepticism over robust antitrust enforcement since the 1970s.
While the many, particular antitrust reforms currently considered by Congress or the European Commission embody a variety of contestable economic propositions, it is a mistake to focus overly on their substance. Some of the proposals may very well make economic sense; most of the assertions of harms to markets as dynamic as digital markets, or features of such markets as intangible and contingent as “innovation,” are both unverifiable and unfalsifiable. Such proposals should be evaluated less for their likely economic effects and more for the ways in which they shift control, either between markets and government or among governmental actors. (They should also be studied for the ways in which they make it easier or harder or correct errors in the exercise of that control.) Antitrust has always been more about allocating control between markets and government than about efficiency or particular economic outcomes. In considering the current antitrust reform movement, we ignore that feature of antitrust to our peril.
Thomas B. Nachbar
Citation: Thomas B. Nachbar, An America Fit for the Digital Age?, CONCURRENTIALISTE, May 2, 2022