Hayek as a New Brandeisian? The Need to Distinguish Theory from Practice in Hayekian Competition Policy

October 2018

Hayek as a New Brandeisian? The Need to Distinguish Theory from Practice in Hayekian Competition Policy

By Joseph V. Coniglio (Wilson Sonsini Goodrich & Rosati)¹

Introduction

The New Brandeis Movement (“NBM”) initially appeared to constitute a broad policy platform aimed at replacing antitrust law’s consumer welfare standard with a structural paradigm that would reanimate the purported political content of the Sherman Act.²  After receiving criticism from many in the antitrust bar, the NBM appears—at least as a practical matter—to have both abandoned a political reinterpretation of U.S. antitrust law and emphasized Section 5 of the FTC Act, which is already generally understood to reach at least some business conduct that is beyond the scope of the Sherman Act, as a vehicle for its enforcement program.³  Still, a structural theory of antitrust divorced from the consumer welfare standard—and thus distinct from the Harvard School—appears to remain central to the NBM’s competition policy.⁴

Separated from both broader political goals and the consumer welfare standard, however, the question arises as to what rationale the NBM would now employ for its structure and process paradigm, which ultimately appears to reduce to a “big is bad” approach to antitrust that would condemn at least some increases in market concentration without any presumption or case-specific evidence evincing a reduction in consumer welfare.⁵  To answer this question, it may be worth recalling what is perhaps the NBM’s most sacred text.  Here, concentrated markets were suggested to be inherently problematic in light of Friedrich Hayek’s famous knowledge problem.⁶  As the movement’s founder Barry Lynn there stated, “Hayek’s vision provides us with perhaps the single most eloquent and concise depiction of how to structure a political economy” distinct from “some central planning authority, public or private…”^<7/sup>

By indicting concentration on Hayekian grounds, the NBM has availed itself of an argument for its structural paradigm that appears to have gone largely unaddressed by commentators.  Such a reliance on Hayek, however, is entirely misplaced when the full implications of his theory of knowledge and competition policy are considered.  While it is true that, in theory, Hayek recognizes the benefits of decentralization, he is nonetheless clear that, with respect to implementing practical policy, monopoly is not necessarily less desirable than less concentrated market structures.  For Hayek, while antitrust law can play an important role in the economy, it should not view monopoly as inconsistent with the discovery process of competition.  On the contrary, the NBM’s structural paradigm is itself in tension with Hayek’s understanding of the rule of law, which is an essential underpinning of the very Hayekian competitive order the NBM claims to champion.

The Theoretical Benefits of Decentralization

For Hayek, the economy is not a static phenomenon whereby planners—be they firms in a decentralized market, monopolists, or governments—are able to maximize welfare using a given understanding about the supply and demand for various commodities.  Rather, the economy is a dynamic process whereby economic actors adapt to changing circumstances.  Accordingly, the core problem faced by planners seeking to bring about a functioning economy is not gathering the general scientific knowledge sought after by technocrats, but rather “the knowledge of the particular circumstances of time and place.”⁸  It is this latter knowledge, however, that cannot be accumulated by centralized private or government planners, but is instead dispersed amongst the various actors that make up the economy as a whole.  For this reason, for Hayek, a rational economic order is not achievable by “a central board which, after integrating all knowledge, issues its orders,” but most be achieved “by some form of decentralization.”⁹

The NBM appears to contrast Hayekian political economy with contemporary mainstream competition policy in two ways.  First, instead of some measure of welfare maximization, the goal of political economy is “to structure society as a whole to be able to change as circumstances change.”¹⁰  Second, business conduct that has a centralizing effect on economic planning by increasing market concentration is treated by the NBM as inherently undesirable irrespective of whether it will likely reduce consumer welfare.

In other words, the designers of a political economy should aim to devise institutions that enable people to gather and process and transmit useful information to one another and through society itself, and to keep these institutions small enough and diverse enough to react swiftly to that useful information.  Rather than attempt to locate thinking in some central planning authority, public or private, reason and the liberty to use it should be located as much as possible at the bottom, in the individuals who run the local machine shop, the local store, the local farm, the local bank.¹¹

Interpreted this way, Hayek’s knowledge problem could provide the NBM with a prima facie justification for its structural paradigm.  On this view, to solve the central economic problem of adaptation—not technocratic welfare maximizing—the ideal political economy should be decentralized to “[e]nsure that the knowledge of the particular circumstances of time and place will be promptly used.”¹²  Regardless of its effects on consumer welfare, so the NBM’s argument would go, business conduct that results in more concentrated markets harms competition by hindering the discovery of the dispersed economic knowledge that, through the market, would otherwise be better communicated throughout the economy to achieve a more rational economic order.

Perfect Competition Is Not Competition Policy

Hayek’s understanding of competition as a discovery procedure differs from the static neoclassical model of perfect competition, where complete information, no barriers to entry, and a large number of buyers and sellers for a homogenous product are assumed.  For Hayek, the almost invariable failure of conditions of complete information to obtain does not indicate a problem with competition—rather, it is “the main task which only the process of competition can solve.”¹³  Similarly, with respect to decentralization, the fact that it is “no[t] possible that every commodity or service that is significantly different from others should be produced by a large number of producers” does not imply competition isn’t working.  By contrast, fewer producers of a commodity or service can signal the existence of cost advantages or a highly valued product that spurs competitive entry.  Simply put, the conditions for perfect competition “do[] not provide [] a valid test which can meaningfully be applied to the achievements of practical policy.”¹⁴

While Hayek views decentralized market structures as a desirable means of communicating dispersed economic knowledge in theory, it therefore does not follow that he would support the NBM’s structuralist competition policy.  That is, notwithstanding his theory of economic knowledge, Hayek is clear that, as a matter of practical policy, monopoly is not, by itself, a competitive problem for regulators to solve:

So long as any producer is in a monopoly position because he can produce at costs lower than anybody else can, and sells at prices which are lower than anybody else can, and sells at prices which are lower than those which anybody else can, that is all we can hope to achieve—even though we can in theory conceive of a better use of resources which, however, we have no way of realizing.¹⁵

The attempt to use Hayek to justify structuralism, therefore, fundamentally fails to grasp the full implications of his theory of knowledge.  Just as Hayek envisions competition as a discovery process to communicate economic knowledge that cannot be centralized in government planners or monopolists, so too would Lynn’s “designers of a political economy” have insufficient knowledge to divine the levels of market concentration sufficient to promote decentralization as an end in and of itself.  As Hayek states, not only is “there no possible measure or standard by which we can decide whether a particular enterprise is too large” but such a program makes the crucial mistake of “produc[ing] essentially anti-liberal conclusions drawn from liberal premises.”¹⁶  Accordingly, commentators have recognized that Hayek’s own views on antitrust would not justify condemning a dominant firm’s conduct simply because it increased market concentration, but because it increased its market power through exclusionary behavior such as predation or price discrimination.¹⁷

The Rule of Law

Even if, as a matter of economic policy, the NBM’s structural paradigm were supported by Hayek’s theory of knowledge, that would still not be enough for it to constitute a legitimate ground for government intervention.  For Hayek, although some interventions will be justified in reference to economic expediency, others “run counter to the very principle on which a free system rests and which must therefore all together be excluded if such a system is to work.”¹⁸  As such, it is the “rule of law [that] provides the criterion which enables us to distinguish between those measures which are and those which are not compatible with a free system.”¹⁹  Importantly, the rule of law does not exclude all “general regulations of economic activity,” but only those that involve some form of arbitrary discrimination.²⁰

For Hayek, the rule of law thus requires proscribing certain classes of government intervention regardless of their economic consequences.  Controls of prices and quantities represent a paradigmatic case of government action that violates the rule of law.  By fixing prices at levels different than what would be obtained in absence of the government regulation, an inequality between supply and demand will result.  In light of this inequality, for the control to work, the government will ultimately have to determine who will transact and who will not.²¹  However, in Hayek’s view, making such a judgment “would necessarily be discretionary and must consist of ad hoc decisions that discriminate between persons on essentially arbitrary grounds,” with the legal implication that “price and quantity controls must be altogether excluded in a free system.”²²

The NBM’s structural paradigm similarly offends the rule of law in a way that, on Hayekian grounds, would make it incompatible with the free market system.  In treating market structure as dispositive, the NBM would make market definition paramount so as to determine which firms should be considered when evaluating concentration levels.  However, deciding what degree of substitutability should be used would, just as in the case of price controls, be an arbitrary choice that had the effect of discriminating between which firms were included in the market.  By contrast, when market definition is used within a consumer welfare frame, such as in contemporary merger analysis, the arbitrariness of any such judgment is overcome by the ultimate requirement of a connection to consumer harm—even in the case where concentration thresholds are used as presumptions for anticompetitive welfare effects, as opposed to bigness per se.

Conclusion

Hayek’s theory of knowledge simply does not support the NBM’s structural paradigm.  For the same reason that decentralization is in theory preferable to centralized decision-making for communicating dispersed economic knowledge, there is no basis to believe as a policy matter that regulators have the requisite amount of information to identify what level of market concentration impairs competition.  And because of the arbitrary and discriminatory way in which NBM’s structural paradigm would be applied, its policy prescriptions offend the very rule of law principles that Hayek saw as fundamental to the free market system.  Indeed, as a final point, in light of the NBM’s criticism of the Chicago School’s purported appeal to “natural forces” of the market,²³ its proponents should similarly consider whether Hayek’s view about the inability of knowledge of the particular circumstances to be centralized may itself be made unempirical by technological advances concerning artificial intelligence, big data, and the internet of things.

Click here for a PDF version of the article

¹ Associate, Wilson Sonsini Goodrich & Rosati. The author thanks Dan Graulich and Blair Kuykendall for helpful comments.  All views are those of the author.

² Khan, Amazon’s Antitrust Paradox, 126 Yale L.J. 710, 743 (2017).

³ Open Markets Institute, Comment, The Overlooked Vitality of Antitrust Doctrine (2018).

⁴ Khan, The New Brandeis Movement: America’s Antitrust Debate, 9(3) JECLAP 131 (2018).

⁵ See Coniglio, Why The ‘New Brandeis Movement’ Gets Antitrust Wrong, Law360 (April 24, 2018).

⁶ Lynn, Cornered: The New Monopoly Capitalism and the Economics of Destruction 249 (2008).

⁷ Id. at 250.

⁸ Hayek, The Use of Knowledge in Society, in Individualism and Economic Order 80 (1948).

⁹ Id. at 84.

¹⁰ Lynn, Cornered: The New Monopoly Capitalism and the Economics of Destruction 249 (2008).

¹¹ Id. at 250.

¹² Hayek, The Use of Knowledge in Society, in Individualism and Economic Order 84 (1948).

¹³ Hayek, The Meaning of Knowledge in Society, in Individualism and Economic Order 97 (1948).

¹⁴ Hayek, Law Legislation and Liberty, The Political Order of a Free People 67 (1979)

¹⁵ Id. at 73.

¹⁶ Id. at 77.

¹⁷ See generally Paul, Hayek on Monopoly and Antitrust in the Crucible of United States v. Microsoft, 1 NYU J. L. & Liberty 167 (2005).

¹⁸ Hayek, The Constitution of Liberty 331 (R. Homowy ed., 1960).

¹⁹ Id.

²⁰ Id. at 334-36.

²¹ Id. at 337.

²² Id.

²³ Khan, The New Brandeis Movement: America’s Antitrust Debate, 9(3) JECLAP 131, 132 (2018).