Part 1 of this short blog series on “Nudging Antitrust,” focused on defining Commissioner Rosch’s recently articulated vision of behavioral economics as it relates to antitrust and competition policy and its differences with more “conventional” economic approaches that are bound by the rationality assumption. By the way, one should note that these more conventional approaches include both Chicago and game-theoretic Post-Chicago approaches, though the Commissioner reserves most of his ire for the former. Today, in Part 2, I’ll turn to Commissioner Rosch’s vision for antitrust policy informed by a particular version of behavioral economics focuses on the Commissioners case against the critiques of behavioral economics as he interprets them.
Here, again, the analysis is replete with mischaracterizations about his target: “conventional economic theory.” Rosch first offers four critiques of behavioral economics he finds unconvincing and why conventional economists make them: (1) neoclassical economists are clinging to the rationality assumption despite the fact that their position is “fundamentally at war with the position of behavioral economists that buyers and sellers do not always behave rationally,” (2) “many neoclassical economists (and their clients) yearn for certainty and predictability,” yet behavioral economics highlights “the ways in which modern antitrust laws’ pursuit of predictability may be costing us too much in the form of aggressive antitrust law enforcement;” (3) “neoclassical economic models are sometimes offered as a substitute for empirical evidence of the effects that a practice or transaction may have instead of simply corroborating that empirical evidence,” (4) and that conventional economists are simply collectively behaving as an incumbent would protecting its market share against a new potential entrant. The Commissioner then turns to two critiques that he concludes have some merit: (1) that behavioral economics lacks an organizing principle, and (2) that behavioral economics is too subjective to be useful for policy analysis in a world where regulators are also afflicted with behavioral biases.
For an analysis of these claims and further discussion, read below the fold.
The first three criticisms are empirical in nature. Rosch implicitly assumes that behavioral models have actually been shown in antitrust-relevant policy contexts to systematically outperform “conventional” models. Rosch asserts that conventional models do not predict predation or anticompetitive mergers or cartels in certain settings (“These individual biases can manifest themselves in firm behavior in several ways that U.S. antitrust law does not predict” and “What behavioral economics has done – by identifying ways in which the assumption of rationality may miss the mark – is to highlight the ways in which modern antitrust laws’ pursuit of predictability may be costing us too much in the form of aggressive antitrust law enforcement.”). The irony here is that Commissioner Rosch consistently offers up behavioral economics in the context of some empirical claim about its ability to reliably identify irrational instances of anticompetitive behavior and contribute to antitrust policy by outperforming rational choice models. This while complaining that neoclassical economists point to models without evidence.
The day may come when behavioral economics can live up to this claim and the claims made on its behalf in the behavioral law and economics movement more generally — but in the antitrust context, it has not done so. What is the empirical evidence concerning behavioral theories about cartels relative to conventional economics? In what contexts is this evidence offered up? Exclusive dealing? Resale price maintenance? Mergers? While a blog post is not the appropriate forum for a full-scale review of the empirical literature on these matters, consider the excellent paper from a group of highly respected antitrust economists on behavioral economics and merger policy, which concludes:
Proponents of behavioral antitrust suggest readjusting substantive rules of law, redefining burdens in litigation, and reshaping agency assessments, all on the basis of arguments involving departures from profit maximization. To the extent such departures are mistakes, proponents of behavioral antitrust propose to inject paternalism into competition policy, but that is antithetical to the fundamental idea of competition policy. To the extent these departures result from pursuit of non-profit objectives, proponents might identify good reasons for concern about particular forms of anticompetitive conduct, but they offer nothing to improve the identification of anticompetitive conduct. No adjustment in merger assessments should be made on the basis that firms sometimes merge because of mistaken expectations or because managers might pursue objectives other than profit. What matters in merger control are the likely effects of the mergers, not the motivations for mergers.
Agencies and courts embraced neoclassical economics and the assumption of profit maximization in competition cases because they provide organizing principles for establishing the basic contours of the law and a lens for examining the evidence in particular cases. Psychology can inform economics in important ways but cannot take the place of economics in competition policy.
Is there a single instance in which behavioral economics has been shown to offer a clear and testable prediction about firm behavior that deviates from conventional theory predictions, been subjected to empirical testing, and proved to have greater predictive power? I do not contend that this will not, or cannot, happen. Perhaps it will; perhaps it won’t. In either event, however, empirical superiority must be earned – not simply asserted.
Commissioner Rosch’s fourth claim, that “rational choice” economists reject behavioral economics to protect their turf, seems far-fetched at best. Those who select into economics at a professional level tend to take seriously things like the scientific method. In the world of economics, as Stigler put it, “it takes a theory to beat a theory.” The rules of engagement in the war of economic theory are clear: the theory with the greatest predictive power wins. The null hypothesis in Rosch’s example should be that economists criticizing the application of behavioral economics to antitrust as it has been articulated (notice, once again, not behavioral economics per se) do so because they believe other theoretical models have better explanatory power and therefore will result in better policy. Is there any evidence whatsoever to reject that hypothesis? Further, taking the claim seriously, is it even clear that increasing the relevance of behavioral economics into antitrust analysis hurts the incumbent economists? I doubt it. Confusion about how to conduct that analysis, when and where it applies, and how to beat it in court all point toward a greater role for “conventional economists” in terms of a pure public choice story in the same way that most economists’ private benefits are greater under the rule of reason than per se rules.
One point merits a brief aside. While Commissioner Rosch likes to frame the conventional economic theory world as exclusively Chicago, it is of course, a much broader set. Chicagoans and Post-Chicagoans in the world of antitrust economists agree on quite a bit in terms of an analytical framework. And in the modern world of professional antitrust economists who are potentially, as Rosch puts it, “losing their monopoly on economic thought when it comes to antitrust,” Rosch’s unfounded accusation includes Post-Chicagoans too. To the extent that the policy proposals proffered by proponents of behavioral antitrust are almost exclusively more interventionist than existing policy, Post-Chicago economics are the closest substitutes and therefore have the most to lose if the behavioral antitrust approach gains traction. This leads to an apparent puzzle: Where are the Post-Chicagoan economists in criticizing the behavioral approach? To defend their theories? To demand evidence? Of course, one answer is that there are a bunch of them in the Obama administration and without the currently luxury of participating in these debates. But here are a few related observations on the topic. After the Kodak decision, the Post-Chicago approach appeared to be mounting a serious challenge to the Chicago/Harvard stranglehold of Sherman Act jurisprudence. But, as we now know, the challenge was not overwhelmingly successful in US federal courts (though it has been made some impact in the courts, much more in the US antitrust agencies, and even greater impact overseas). Interventionist-minded law professors and regulators have thus far looked for alternative intellectual firepower to fuel arguments for additional intervention. Behavioral economics is therefore more likely to be used as a substitute for Post-Chicago economic arguments not because it offers superior analytics in the antitrust setting, but because interventionists motivated by ideological priors needed fresh economic blood. Again, the sloppy use of behavioral concepts and arguments does nothing but bolster the impression that these arguments are not being offered because regulators are studiously applying a serious consideration of the economics, but because they’ve the opportunity to take the economic arguments where they would like to go – i.e. more intervention into the market (with corresponding losses in consumer welfare). My prediction, then, is that there is a serious Post-Chicago backlash against the behavioral approach a few years off in the distance, if not sooner.
As for the last two critiques, which Rosch recognizes as having some merit, he offers the following. First, Rosch contends that the critique that behavioral economics does not have an overarching principle has only limited force. Why?
Given the extent to which behavioral economics has questioned the assumption of rationality that underlies neoclassical analysis, rejecting behavioral economics whole cloth for lack of an organizing principle is arguably just another way of saying that an answer is better than no answer, however wrong that answer may be.
The subtle ability throughout the speech to shift theoretical assertion into empirical claim is impressive. “Given the extent to which ….” Really? What exactly is “the extent to which” behavioral economics has questioned this assumption in an antitrust relevant context? Most of the experimental and field evidence (but not all) applies to individuals rather than firms. There is, of course, evidence from the field and some study of firm decision-making. Recall, however, that the primary argument Rosch offers about firm irrationality concerns agency costs! The actual evidence on what we know about behavioral economics and irrational firm behavior is nowhere close to sufficient to alter current policy, and even the threat of doing so imposes substantial uncertainty costs on firms. I will discuss this point in much greater detail in the forthcoming paper. The real crux of the “lack of organizing principle objection,” (see, e.g. this essay by Judge Douglas Ginsburg and Derek Moore) is that behavioral economics suffers from a lack of clear and testable predictions about antitrust-relevant policy consequences (Will prices go up? Will output go down? What will be the effect of the merger?) and therefore offers little consistent guidance to judges who must apply economic analysis in real cases. Given this objection, it is not clear what to make about Rosch’s assertion that rejecting behavioral economics “whole cloth” for this reason is like saying a wrong answer is better than no answer.
No. This is wrong. No, it is not like that at all. Again, the Commissioner has subtly snuck in some empirical judgments that find no support in the existing literature. The analogy suggests that conventional economic theory provides an answer, but the wrong answer, and behavioral economics, despite any shortcomings, can be demonstrated based upon theory and evidence to improve competition policy relative to policy based on conventional alternatives. To beat the same old drum: show me. The evidence is not there. And unsurprisingly, it is not provided in any of the Commissioner’s speeches on the topic. The refrain may be tiresome, but when billions of dollars worth of consumer welfare are on the line, it bears reprise.
Lastly, Commissioner Rosch takes on the tried-and-true objection to behavioral economics-based policy that regulators are also humans and suffer from cognitive biases. While many have made this point, it is frequently associated with Judge Posner. The response? It deserves to be quoted in its entirety:
My problem with this criticism is that it ignores the fact that, unlike human beings who make decisions in a vacuum, government regulators have the ability to study over time how individuals behave in certain settings (i.e., whether certain default rules provide adequate disclosure to help them make the most informed decision). Thus, if and to the extent that government regulators are mindful of the human failings discussed above, and their rules are preceded by rigorous and objective tests, it is arguable that they are less likely to get things wrong than one would predict. Of course, it may be the case that the concern with behavioral economics is less that regulators are imperfect and more than they are subject to political biases and that behavioral economics is simply liberalism masquerading as economic thinking.24 My response to that is that political capture is everywhere in Washington and that to the extent behavioral economics supports “hands on” regulation it is no more political than neoclassical economics which generally supports “hands off” regulation. On a more serious note, perhaps the best way behavioral economics could counter this critique over the long run would be to identify ways in which the insights from behavioral economics suggest regulation that one would not expect from a “left-wing” legal theory.
“We’re the government and we’re here to help.” Sigh. But seriously, human beings making decisions “in a vacuum?” It is individuals and firms who are making decisions insulated from market forces that create profit-motive and other incentives to learn about irrationality and get decisions right — not regulators? The response to the argument that behavioral economics is simply liberalism masquerading as economic thinking (by the way, the argument is not that, it is that antitrust policy based on behavioral economics has not yet proven to be any more than simply interventionism masquerading as economic thinking — but I quibble) is weak. Commissioner Rosch’s response seems to be that what antitrust is about is a battle between two ideological regimes dressing up policy preferences in the economic armor of their choices — behavioral economics for the lefties and Chicago School for the righties — and then having at the political capture game. Of course, we can and should do better. Indeed, as the Supreme Court has noted for more than thirty years now, American antitrust actually does better. Economics is not merely an ideological anchor for right-wing financial policies. Instead, economic analysis is ensconced at the heart of American antitrust policy specifically because it provides the best, most predictive testable implications for maximizing consumer welfare. There are serious arguments to be had about what behavioral economic theories actually say about firms, competition, and market outcomes, and what how, if at all, we should modify our current presumptions in antitrust enforcement in antitrust. We must ask certain questions beforehand, however. What do behavioral theories predict? What is the empirical evidence to test these predictions? How robust is it? Could a shift to a behavioral approach be justified in an evidence-based antitrust regime? Consumers deserve better.
In Part III, I’ll turn to Commissioner Rosch’s account of the policy implications of behavioral antitrust.
