“You must be the best judge of your own happiness.”
Jane Austen said that, in Emma, but the statement is also a keystone principle of modern microeconomic theory, and it provides the epistemic foundation that makes benefit-cost analysis possible. The only way to know people’s preferences is observe the choices that they themselves freely make; all inferences about the “public” interest must begin there.
Behavioral economists have poked holes in this principle by demonstrating in laboratory studies, as well as in the field, that people will often make irrational or inconsistent choices, so that simple models of rational individual preferences are not always good at predicting the actual behavior of real people. What influence, if any, should these findings have on public policy? Very little, I will argue.
Ten days after taking the oath of office in 2009, President Obama issued a memorandum (link is no longer available) that among other things sought to “clarify the role of the behavioral sciences in formulating regulatory policy.” He was widely expected to follow this up with an Executive Order incorporating the behavioral economics perspective into federal regulatory analysis. The expected Executive Order did not emerge until 2015, and by then it included only mild exhortations for agencies to deploy the “Nudge” concept that had been popularized by Cass Sunstein, President Obama’s Administrator of the Office of Information and Regulatory Affairs.
Nonetheless, regulatory agencies have increasingly used consumer irrationality to justify regulatory interventions, even where there is no apparent market failure. They attribute economic benefits, amounting to many billions of dollars, to regulatory actions that give consumers nothing new, but simply deprive them of their preferred choices. How exactly is that beneficial? Regulators even make the presumption that they are far better at judging businesses’ interests than are the businesses themselves. Ted Gayer and Kip Viscusi challenge this form of regulatory analysis:
How can it be that consumers are leaving billions of potential economic gains on the table? . . Moreover, how can it also be the case that firms seeking to earn profits are likewise ignoring highly attractive opportunities to save money? . . Rather than accept the implications that consumers and firms are acting so starkly against their economic interest, a more plausible explanation is that there is something incorrect in the assumptions being made in the regulatory impact analyses.
In previous posts I have argued that benefit-cost analysis, applied to regulation, should be viewed less as a tool to inform the regulators, and more as a test to see whether the regulators are acting as faithful agents of the public’s interest. From this perspective, it becomes clear that behavioral economics can be permitted only a limited role in justifying regulation. When a government agency proposes to use force against its own citizens, we cannot accept the explanation that “The public doesn’t know what’s good for them.”
Last year in the Journal of Policy Analysis and Management, Susan Dudley and I engaged in a Point/Counterpoint exchange with Hunt Allcott and Cass Sunstein on the question of how observable anomalies in consumer decision making – “internalities” in their parlance – should be used in Regulatory Impact Analysis. Despite agreement among all four authors on the general principles of applying benefit-cost analysis to regulation, we sharply disagreed about the treatment of internalities.
Allcott and Sunstein argue “In markets where there are internalities, numerous papers have shown theoretically that taxes or other forms of government intervention can increase welfare.” They see behavioral insights as providing a natural extension of conventional welfare economics:
Notice here the direct analogy between internalities and externalities: . . . this model simply restates the standard Pigouvian model of externality taxation. It is thus useful to think of internalities as “externalities that individuals impose on themselves.
In contrast, Dudley and Mannix refuse to surrender consumers’ sovereignty over their own welfare.
“[C]hoice architecture” cannot produce benefits by destroying choice. . . Allowing regulators to control consumers “for their own good” – based on some deficiency in the consumers themselves rather than any failure in the marketplace – is to abandon any serious attempt to keep regulatory policy grounded in an objective notion of the public good.
Any truthful analysis of benefits and costs will tell us what consumers think, not what the regulator thinks consumers should think. We do not allow the government to change the results of elections because of some theory of irrational and biased voters; neither should we allow it to distort consumers’ revealed preferences in an economic analysis.
The economics literature is filled with proposals for “nudges” to encourage individuals to make better (i.e., more accurately self-interested) choices – a kind of soft paternalism that sounds harmless enough. In the hands of regulators, however, behavioral economics has quickly evolved into a ready-made excuse for a hard, and even oppressive, paternalism. I would urge the incoming administration to apply a more traditional and rigorous form of benefit-cost analysis: one that evaluates regulators’ conformance to the public’s preferences, rather than the other way around.
As I commented when President Obama first raised the prospect of behaviorally-informed regulation:
Ultimately, we insist that our regulators start from a presumption of rationality for the same reason that we insist that our criminal courts start from a presumption of innocence: not because the assumption is necessarily true, but because a government that proceeds from the opposite assumption is inevitably tyrannical.