Cass Sunstein is a distinguished professor of law at Harvard, and he has occupied high positions in the Obama and Biden administrations. Although he is a left liberal rather than a classical liberal, he considers Friedrich Hayek one of his intellectual heroes. For Sunstein, Hayek’s primary insight is the importance of ignorance. Central planning cannot work, and invisible hand processes are much better than the efforts of the “constructivist rationalists,” who try to design social institutions. In Hayek for Our Time (Cambridge University Press, 2026), he says:
Hayek’s claims about ignorance organized everything, or almost everything, that he wrote. They undergirded his negative claims, that is, his objections to his assortment of bête noires: planning, designed orders, constructivism, rationalism, scientism, socialism, commands, abolition of private property, price-fixing, and hubris of all kinds. (Are all these in some sense the same? For Hayek, I think so.) They paved the way for his positive claims, that is, his claims for individualism, markets, traditions, customs, invisible hand mechanisms, undesigned orders, spontaneous orders, liberalism of a certain kind, Scottish enlightenment thinkers, the price signal, and the rule of law. They helped explain his frequent oppositions between what he liked and what he did not like. Indeed, that was a common strategy in his writing: There are the constructivists, and there are the individualists, and here is what they both think, and this is why the individualists are right. (emphasis in original)
Sunstein’s praise for Hayek puts him in a dilemma. He is one of the foremost advocates of the importance of cognitive and emotional biases in people’s decision making. In this view, people often act irrationally; because of mistakes in reasoning, they fail to achieve their own goals. He favors “nudges” to induce people to act in their own interests:
The price system is indeed a marvel, but notwithstanding Hayek’s arguments, the price system is not always entirely marvelous. In some areas, including occupational safety and energy policy, there are behavioral market failures. . . Recall that if people are suffering from present bias, unrealistic optimism, limited attention, or a problem of self-control, prices might not capture important factors. If the result is a serious welfare loss for those people, there is an argument for some kind of public response, potentially including institutions that impose mandates. When people are running high risks of mortality or otherwise ruining their own lives, it might make sense to adopt a mandate or a ban, certainly on welfare grounds.
Sunstein is an able dialectician, and he is alert to an obvious counterargument. Why assume that the experts who advise the government about these biases have got things right? Isn’t assuming this to succumb to the constructivist rationalism that Hayek warned us about? He has three defenses against this.
First, Hayek himself recognizes that people are not perfectly rational but decide in a biased way:
Hayek did not embrace the “rational actor” model that dominates Chicago School economics, and indeed he argued explicitly against that model. As he wrote, “The antirationalistic approach” that he favored, and that he wanted to distinguish from constructivism, “regards man not as a highly rational and intelligent but as a very irrational and fallible being, whose individual errors are corrected only in the course of a social process.”
Doesn’t this make room for a Hayekian behavioral economics? I do not think so, because it does not speak to the point that the supposed experts may be subject to biases of their own. As an example of such bias, Sunstein claims that car buyers ignore the clear cost-saving advantages of electric cars over gasoline-powered vehicles. Might his verdict be biased by the emotional aversion of many academics toward “gas-guzzlers”?
Second, Sunstein accurately notes that Hayek was far from being an advocate of laissez-faire capitalism. He allowed many welfare state measures, and this is an issue about which both Mises and Rothbard challenged him. Isn’t Hayek thus leaving room for “nudging”? Here Sunstein confuses two different points. The first of these, and by far the most important, concerns the implications of Hayek’s ignorance argument. If these implications are that nudging is unlikely to work, this is the salient point. The second matter, Hayek’s views on the propriety of welfare measures, is of secondary significance.
Sunstein’s third defense gets to the essence. He contends that even if you take account of the weakness of central planning and allow this as a cost, you must still assess this against the benefits of the proposed intervention. To return to the dispute over electric cars, the balance of the evidence suggests that these benefits outweigh the costs:
Such questions raise a host of empirical issues. But if consumers are not paying enough attention to savings in terms of money and time, a suitably designed fuel economy mandate might turn out to be justified, because it would produce an outcome akin to what would be produced by consumers who are at once informed and attentive. Energy efficiency requirements might be justified in similar terms, and indeed, the argument on their behalf might be stronger. If the benefits of mandates greatly exceed their costs, and if there is no significant consumer welfare loss (in the form, for example, of reductions in safety, performance, or aesthetics), then the mandates would seem to serve to correct a behavioral market failure.
Sunstein seems unaware that issues of this sort can be decided on grounds other than a cost-benefit calculation. He does not address the position that it is up to people as they actually exist to make choices, not people suitably “corrected” to make rational choices. People are free to make use of information about cognitive and emotional biases, if they wish to do so; and if the reply is that their use of this information is itself biased, that merely restates Sunstein’s view. It is not an additional point in its favor. In his comment about the behavioral economist Robert Sugden, who favors the view that people’s actual preferences should be respected, Sunstein shows that he does not grasp what is in dispute:
Sugden urges that the approach I am defending is “constructivist.” Following Hayek, Sugden rightly contrasts constructivists, like Descartes, with evolutionary thinkers, like David Hume and Adam Smith. Constructivists celebrate reason and our ability to assess institutions and codes of morality in its light. Evolutionary thinkers celebrate undesigned, unintended orders, and do not think that constructivists can understand them or improve on them. Sugden thinks that in trying to produce a kind of Hayekian behavioral economics, I have ventured into the constructivist camp, and so am not really Hayekian. Maybe so, but in my view, Sugden is engaged in name-calling here. If there is a little constructivism in my approach, and if some Hayek-risen-from-the-grave might not like it, so what?
Thomas Nagel has suggested to me that Sunstein thinks that his sort of cost-benefit utilitarianism is obviously true. Those of us who believe that people have natural rights will disagree.