Economics, Not Politics, Should Guide Competition Policy

It is worthwhile to periodically reexamine an existing legal regime in order to determine if it is still serving the welfare of society. Antitrust law is no different, and, as new data comes to light and new theories are developed, regulators and courts should indeed consider how best to incorporate them. But not all proposals for reform are created equal. In my earlier essay, I outlined the project of one group of reformers—the neo-Brandeisians—and detailed how it was both rooted in historical misunderstandings, and, if implemented, would lead to inferior competition laws and net harm to society.

The modern antitrust regime is more than capable of adapting to new realities in the economy. Reform should focus on extending the contours of existing doctrine and not on introducing politicized enforcement that would undermine the rule of law. Mark Jamison, John McGinnis, and Hal Singer thoughtfully responded to my essay on these points, and what follows is a reflection on their thoughts and expansion on my original thesis.

Antitrust Law Is Capable of Remedying Harms in Digital Markets

Mark Jamison’s belief that modern antitrust is not capable of adequately curbing abuses in the digital economy is useful, but, I believe, actually supports my comments about the consumer welfare standard (“CWS”).

The CWS is merely a framework for directing analysis but does not prescribe particular tools with which to conduct that analysis. The metric Jamison uses to judge the efficacy of modern antitrust remains consumer welfare, even as he criticizes contemporary analytical tools for failing to capture dynamics in digital markets.

This is distinct from the neo-Brandesians who believe that economic analysis is itself flawed, and that political means of administering antitrust should be preferred. Although I disagree with Jamison that antitrust is currently incapable of handling harms that may arise in digital markets, I do agree with him that more dynamic modes of analysis should be developed and incorporated. He has been doing interesting work on this problem that I encourage readers to explore.

Taming the Uncertainty of Politicized Laws

John McGinnis offered a number of useful points. Foremost was the observation that officials tend to be willing to condemn conduct with which they are unfamiliar. Yet, it is not the judgment of courts and enforcers that are best able to handle the unimaginably vast amount of information that spontaneously orders the economy.

Thus he noted (by quoting Judge Taft) that “business justification is best measured… by the practices business traditionally employed, not by judicial determinations of what the practices should be.”

Further, consonant with my admonition that antitrust focus on problems with which it is best equipped to deal, McGinnis notes that political concerns are best dealt with in purpose-specific regulations. Income inequality, for example, has been repeatedly shown to be best handled “through progressive taxation, not regulation, including antitrust.” Thus, importing political dimensions explicitly into antitrust analysis only serves to “place[] an activist and unaccountable judiciary at the commanding heights of the economy, subordinating the market to the state.”

Hal Singer takes issue with the orientation of modern antitrust toward apolitical enforcement:

[T]he Court’s recent chipping away at antitrust protections… also reflects a political ideology. That it’s a political statement in favor of powerful economic actors does not make it any less political. It’s not helpful to accuse the New Brandeisians of politicizing antitrust, especially when your own side seeks to please its own constituency of plutocrats.

This is not a new objection. Populist critics maintain that any institutional arrangement is an expression of political preference. They are correct, but they think this observation proves more than it actually does, as they fail to distinguish between enactment of laws and enforcement of laws.

A chief objection to the populists’ project is that the expression of political preferences in the enforcement of the laws becomes unmoored from broad democratic accountability and, thus, is subject to the whims of enforcers in the executive branch. This is an entirely different matter than the negotiation necessary for the creation of public policy, which is, of course, always a political act.

McGinnis, in a separate piece, ably points out what it means to enforce laws apolitically in a liberal order:

[T]here are some hard questions around the edges of what constitutes neutrality in the application of principles, but the core is clear. As my colleague, Martin Redish… has written, “neutral principles require that whatever rationale [is selected] . . . must be applied consistently in all cases; it cannot be selectively altered in subsequent cases solely because . . . the outcome dictated by use of that principle [is seen] to be politically distasteful or offensive.” Thus, a decision in applying a principle should not depend on the identity of the parties, or even on the nature of the dispute, but rather on standards that transcend the dispute and the character of the parties.

The facile observation that all enforcement structures express the political view entailed in their enactment contributes nothing to a central question in democracies: how do you channel broad democratic political will through enforcement structures? Thus, given its proper due, the argument I and those of my philosophical ilk make is not that there should be no politics in the framing of laws—indeed its naive to assert otherwise.

 A primary goal of competition policy is to bring consumers the best goods at the lowest prices. It’s entirely irrelevant whether Amazon does this with a million third-party sellers, or closes its marketplace and manages to provide such goods on its own (as Walmart does with its brick-and-mortar stores).

The goal is, instead, to establish a framework of principles that can be consistently applied to any case that appears before a court without resorting to a mercurial list of preferences (such as which groups should be favored at any one time). This is at the very core of the consumer welfare standard.

If it appears that antitrust laws favor plutocracy, however, then we should discuss how to remedy those laws structurally such that the outcomes are more even. But what we should not do is set up a regulator with a mandate to decide which interest groups are worthy and unworthy, and skew enforcement along those preferences.

Consumer Welfare Remains the Relevant Metric

Singer framed his larger argument in a style all too common, not just to antitrust critics, but of a certain mode of policy critique. The essence of my argument was that antitrust law is good at doing certain things—using economic analysis that guides competition policy in a way that on net benefits consumers—and is not good at other things—incorporating larger social issues like income inequality or labor rights in a competition analysis.

He accused me of being conspicuously silent by not offering policy prescriptions for all of the various ills that he is concerned about. But, logically speaking, the onus for offering policy recommendations falls on those who believe they have identified an actual harm and have a good means of remedying that harm. Thus, most of his article amounts to shadow boxing with a version of my essay he wished I had written.

Singer did bring up one antitrust-relevant criticism, that was both interesting and deserving of a response:

Stout insists that Amazon competes in a larger retail market against brick-and-mortar providers, but even the Merger Guidelines contemplate price-discrimination markets, in which customers are distinguished by a certain attribute or channel. More generally, the Guidelines seek to identify the smallest set of products such that a hypothetical monopolist (or monopsonist) can exercise market power. Would Amazon, or even all e-commerce platforms, need to control brick-and-mortar outlets in order to push the price of an input below competitive levels? If not, then e-commerce constitutes a relevant market. The pandemic has made it painfully clear, if it wasn’t already, that brick-and-mortar outlets are weak substitutes to online delivery.

I followed up with him in private messages to make sure I understood his argument. I hope I do it justice here, and apologize if I do not.

Singer is posing an alternative market-definition analysis than the one I did in my essay. He is focused on the inputs to e-commerce—the supply of third-party sellers—and, by extension, the products those third-party sellers can surface on e-commerce channels. This is distinct from how I framed the problem and, indeed, how I believe critics of Amazon typically frame the problem.

This framing appears to be about monopsony, and not monopoly, strictly construed. On this account, the antitrust harm flows from the manner in which Amazon sells access to e-commerce services. Thus, the “consumers” in the analysis are merchants selecting among e-Bay.com, Amazon.com, Walmart.com, etc. for e-commerce services. If Amazon uses market power to create an exclusionary effect, the relevant market is the market for e-commerce access, and not, as I described in my essay, the market for retail goods. (Singer, privately, told me that he believes that consumers as such are still being harmed, but the argument, in my estimation, must logically proceed as I have just laid it out in order to make sense of it.)

Assuming Singer’s market, he then goes on to contemplate exclusionary harms. But the harms he conjectures are self-refuting, thanks in part to the ease of multi-homing e-commerce services. In particular, Singer worries that Amazon demotes sellers in its product search results that have insufficient inventory levels. The feared anticompetitive leveraging here would be an implicit requirement that merchants reserve all their inventory for Amazon’s platform, and starve competing platforms of that inventory.

But the feared harm doesn’t make sense on its own terms. If a merchant is being demoted for low inventory because she is selling that inventory on another platform, then the merchant clearly has a viable channel of sales outside of Amazon.com. Moreover, if anything, by demoting merchants that sell well on other platforms, Amazon only accelerates its irrelevance to third-party merchants by making other available platforms relatively more attractive.

This speculative harm is also difficult to take seriously, aside from the obvious flaw. A primary goal of competition policy is to bring consumers the best goods at the lowest prices. It’s entirely irrelevant whether Amazon does this with a million third-party sellers, or closes its marketplace and manages to provide such goods on its own (as Walmart does with its brick-and-mortar stores). New harms could arise, notably in the relationship between the large retailer and the manufacturers it works with, but whether the ecosystem of online platforms provides enough outlets for third-party merchants to proliferate is ultimately disconnected from the larger social goal of keeping economy-wide prices low and quality high.

Thus, my original framing—which requires us to look at Amazon in the context of both online and offline competition—is truly the superior framing to examine potential competition harms from the platform.

Thanks again to Jamison, McGinnis, and Singer for the spirited discussion, as well as to Law & Liberty for hosting it.