Introduction

Friedrich Hayek is one of the most important thinkers of the modern era, and among the most intriguing. He is immensely relevant to the present period. In some ways, the twenty-first century ought to belong to him.

Hayek offered enduring lessons about undesigned orders, traditions, psychology, markets, coercion, and the rule of law. He celebrated designs without designers. He emphasized how much we benefit from practices and institutions that no individual created and that we do not and cannot understand. He linked his claims about undesigned orders and dispersed knowledge with his moral and economic arguments about freedom, the rule of law, and the constitution of liberty.

Hayek is also an immensely polarizing thinker. Some people demonize him. They think of him as a satanic figure, the inspiration for all sorts of horrible things. They think that he is the principal thinker behind “neoliberalism,” which is, in their view, responsible for economic inequality, an inadequate social safety net, the excesses of the free market, and the plight of workers.

But some people worship Hayek; they consider him to be some kind of divinity. They think that if you are asked a hard question, you should ask this one instead: “What would Hayek think?” They believe that if you obtain the answer to that question, you know what everyone should think. There are Hayek loyalists, with an intense emotional commitment to Hayek. There is also a cult of Hayek, and it is not entirely different from the cults of Karl Marx, John Rawls, Michel Foucault, and Ayn Rand – and in some ways, with the cults of Bob Dylan, the Beatles, and Star Wars. If you read Hayek, you enter a whole world, one with a great deal of order and coherence and even majesty. It puts everything, or most things, in place. It is immensely repetitive, and not worse for that. (Familiar songs can sound best; they are comforting.) It is inspiring and in its way irresistible.

I understand the cult, I think, and have spent a little time in it. But while I greatly admire Hayek, I am not a member.

I received a glimpse of the demonization during an interview with The New Yorker, in which Isaac Chotiner, a journalist known for his “gotcha” interviews, noted my enthusiasm for Hayek and then asked me this: “Hayek talked about disliking ‘Near Eastern’ students because they were ‘fundamentally dishonest.’ He said that he had a ‘profound dislike for the typical Indian students at the London School of Economics, which I admit are all one type – Bengali moneylender sons.’”

My answer:

Then Chotiner’s trap was sprung: “Hayek supported the 1973 coup in Chile. There are lots of things that don’t make him seem like such a democrat or liberal, correct?”

Then me, lightly: “The criticisms are either accurate or, in some sense, fair. So when I say Hayek is a great figure or a great part of the liberal tradition, I don’t mean to endorse everything he said. I tried to be measured, but I got a flood of hate mail.”

I have learned an immense amount from Hayek specialists: a group of dedicated scholars who are, in a sense, keeping the flame alive. They have done exceedingly important work in helping people to understand Hayek’s complicated, voluminous oeuvre, to keep up with his many twists and turns, and to see how things do or do not fit together. I am not trying to do anything like that here. This is not a summary of Hayek, or anything like an intellectual biography.

My aims are narrower and more specific. I want to show that Hayek is the Apostle of Ignorance. I want to show that Hayek’s claims about our ignorance – I am tempted to say Our Ignorance – bear on current issues and debates, including uses of behavioral economics and debates about the potential of artificial intelligence and the evaluation of traditions. I want to suggest how behavioral economists and AI specialists might learn from Hayek and do their business better. Actually, I want to suggest how all of us might learn from Hayek and do our business better. With Hayek’s help, I want to show the limits of prediction. I want to say a few things about traditions and those who do not like them. My goals, in the end, are to say what I think is true, not to insist that something is true because Hayek thought it.

Hayek was not right on everything. He was full of insights, but he could be dogmatic. Some of his ideas were, I fear, a bit like slogans. Abstractions did too much work. He was too sure of himself. We could do better than he did, not only because we know so much more, but also because we can be, and should be, more pragmatic and more empirical. Still, we have a ton to learn from him. He repays rereading. Some of his writings are like poems or novels, in which you discover something new or interesting, or different, every time.

Speaking of which: What would Hayek think about something on which he did not write and about which he knew little or nothing – say, contemporary behavioral economics or artificial intelligence? This is a complicated matter. There might be no fact of the matter. We cannot read Hayek and then announce the answer. We might try to do that, but sometimes we really cannot, because Hayek did not speak to the topic.

With respect to the views of dead thinkers, answers to many particular questions are inevitably interpretive in Ronald Dworkin’s sense (Dworkin, Reference Dworkin1985): they must attempt (1) to fit the materials to be interpreted and (2) to justify them, that is, to put them in the best constructive light. It is of course true that if an approach does not fit the materials, it cannot claim to be interpretive. But with respect to many specific questions of interpretation, more than one conclusion will do well, or well enough, along the dimension of fit, and the real question is one of justification. What looks like (1), or what purports to be (1), is often (2).

That is, when a follower of Kant urges that “Kant would say x,” or that “Kantianism entails y,” the enterprise may be to make the best constructive sense of Kant and Kantianism, not to follow some edict of Kant himself. The interpreter might even be said to be authorizing a new chapter in a kind of chain novel, whose initial chapters were written by Kant, but which Kant did not complete (Dworkin, Reference Dworkin1985). Which understanding of Kant puts his views in the best constructive light? Because that is often the inevitable question, it should not be amazing to find that Jones’ preferred interpretation of Kant fits with Jones’ normative views, while Smith’s interpretation fits with Smith’s. The dispute between Jones and Smith might not be resolvable without speaking of justification – of what does, in fact, put Kant in the best constructive light. Inevitably, it is an interpretation, in which we attempt to make the best constructive sense of what he actually said and thought.

I need to say this by way of preface, because some of my discussion will invoke Hayek to cast light on issues that Hayek did not engage, and because some followers of Hayek will dispute, not unreasonably, what I have to say. To say the least, Hayek was not enthusiastic about government regulations, and I will be offering a Hayekian understanding and defense of fuel economy regulations. Modern Hayekians might reasonably cry: Foul!

So be it. The question is whether the defense works, given the truth in what Hayek said and wrote. Let’s see.

1 Ignorance

Hayek is generally known as a fierce critic of socialism and a committed advocate for free markets. So he was. His most famous book is called The Road to Serfdom (1944), and he did not approve of that road. He despised socialism. He thought that planners were doomed to fail (and to destroy freedom). His objections to planning have immense relevance and resonance today, whether we are speaking of market-friendly nations that occasionally plan (such as the United States and Germany) or market-skeptical nations that prize planning (such as China and Russia).

Both sets of nations have a great deal to learn from Hayek (Boettke, Candela, & Truitt, Reference Boettke, Candela and Truitt2024). If, for example, a nation in North America or Europe is taking steps to address climate change, it should have Hayek’s work in firm view: Planners beware. I worked in the US government on climate change issues under President Barack Obama, and I did try to keep Hayek in mind, not least in issuing energy efficiency regulations, which are a form of planning.

Even so, Hayek’s fundamental preoccupation, and what focused his attention from the 1930s to the 1990s, lay elsewhere, or more precisely, lay behind and above everything that he criticized and everything that he advocated. His preoccupation, in short, was ignorance (Caldwell, Reference Caldwell2003). Hayek was the Apostle of Ignorance. In a sense, he was a celebrant of ignorance, because he associated it with freedom. He emphasized what human beings cannot know, and what they can never know. Time and again, he drew attention to information that we do not have, and that we will never have.

In this way, he went well beyond thinkers in the Austrian School of Economics, who also focused on knowledge and ignorance, who taught him so much, and where he encamped, in one or another respect, for most of his life (Caldwell & Klausinger, Reference Caldwell and Klausinger2022). In particular, Hayek went beyond Ludwig von Mises, his brilliant, bold, iconoclastic, persistent, rigid, edgy, and cranky mentor, who first saw why socialism could not work, and who explained the impossibility of socialist planning with immense clarity. Hayek was deeper than Mises, gentler, more curious, more wide-ranging, less edgy, less harsh, more pragmatic, and somehow less ideological. He took the knowledge problem, as it is often called, from Mises (and others in the Austrian School), but he did more with it than anyone else did, and he saw it in all sorts of places. His intellectual and personal trajectories are endlessly fascinating (Caldwell & Klausinger, Reference Caldwell and Klausinger2022).

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.

Hayek admired what he saw as evolutionary thinkers, including those in the Scottish Enlightenment, who celebrated social learning, invisible hands, traditions, and undesigned orders (Ullmann-Margalit, Reference Sunstein1978; Ullmann-Margalit, Reference Sunstein and Reisch1997). Adam Ferguson, David Hume, and Adam Smith pointed to the limits of reason and to designs without designers. Carl Menger, founder of the Austrian School and one of Hayek’s inspirations, was an evolutionary thinker. Hayek loved Edmund Burke, who was no constructivist, and who embraced traditions and even prejudices, writing in his Reflections on the Revolution in France (Burke Reference Burke[1790] 2014): “You see, Sir, that in this enlightened age I am bold enough to confess that we are generally men of untaught feelings, that, instead of casting away all our old prejudices, we cherish them to a very considerable degree, and, to take more shame to ourselves, we cherish them because they are prejudices; and the longer they have lasted and the more generally they have prevailed, the more we cherish them.” Burke’s writing is unmistakably Burkean (“You see, Sir, that in this enlightened age I am bold enough to confess”). But the sentiment here is Hayekian, too (“the longer they have lasted and the more generally they have prevailed, the more we cherish them”).

Hayek contrasted evolutionary thought with constructivism, exemplified by René Descartes, Voltaire, and (alas) John Stuart Mill, which, he thought, was based on a “fatal conceit,” and which led to socialism. Constructivists (whom Hayek sometimes labeled as “rationalists”) celebrate reason and what human beings can do with it. They think that anything can be redesigned, and that with reason, we can figure out whether it should be redesigned. Thus Voltaire’s remarks, which Hayek saw as the height of folly: “if you want good laws, burn those you have and make yourselves new ones” (Hayek [1978] Reference Hayek2018, p. 4). Hayek hated that, and he thought that it captures a defining mistake.

Among other things, constructivists think that we can assess existing institutions and practices and improve on them, perhaps through planning. Constructivists assume “that all institutions which benefit humanity have in the past and ought in the future to be invented in clear awareness of the desirable effects that they produce” (Hayek, Reference Hayek1967, p. 85). What hubris, Hayek thought. (Cost-benefit analysts, be forewarned.) Constructivists adopt “the innocent sounding formula that, since man has himself created the institutions of society and civilisation, he must also be able to alter them at will so as to satisfy his desires or wishes” (Hayek [1978] Reference Hayek2018, p. 3). What foolishness, Hayek thought.

To put it mildly, Hayek did not approve of that formula. Who invented the English language, after all? Who invented German? Who invented French? He contended that, like language, “morals, law, skills and social institutions” are not a product of any “preconceived design.” He drew attention to “the proper limits of reason in the arrangement of social affairs” (Hayek, Reference Hayek1967, p. 95). Hayek claimed that we must defer “to a moral order to which we owe our existence … to a tradition which we must revere and care for … to a system which we must accept as given … to a gradually evolved set of abstract rules of which human reason can avail itself to build better than it knows … to structures based on more information than any human agency can use” (Hayek, Reference Hayek1984). Note the word choice here. “Revere and care for” – there is intense feeling there, and Hayek hopes that it is contagious.

The fox, it is said, knows many things, but the hedgehog knows one big thing. Hayek knew many things, and so he was a fox; but he was mostly a hedgehog, because there was one thing that he really, really knew.

Take, for example, his Nobel Lecture, delivered on December 11, 1974. The title of the lecture: The Pretense of Knowledge (Hayek, Reference Hayek1974; the lecture can be found in many places, including Hayek, Reference Hayek and Caldwell2014b). (Pause over the title; there we go.) Hayek objected to the “scientistic attitude,” modeled on the natural sciences, which is, in his view, exactly the wrong approach to the social sciences, which must focus on what we do not know. Prices and wages, for example, depend on “the effects of particular information possessed by every one of the participants in the market process – a sum of facts which in their totality cannot be known to the scientific observer, or to any other single brain.” Human beings have not sufficiently appreciated that fact. But that is just an illustration of the much broader problem of “the scientistic prejudice.” Consider, for example, “essentially complex phenomena,” which are all around us, and which often make predictions impossible.

Hayek offers a simple, arresting example:

It is not possible to predict the outcome of sports events involving players of approximately equal skill, because so much depends on what happens on the ground, and because we cannot know, in advance, what that will be. Man, Hayek urged, will have to learn that in “fields where essential complexity of an organized kind prevails, he cannot acquire the full knowledge which would make mastery of the events possible.” There is a characteristic claim from the Apostle of Ignorance.

In his magnificent essay on complex phenomena, written in 1964, inevitable ignorance is also the central theme (Hayek, Reference Hayek and Bunge1964, Reference Hayek and Caldwell2014b). (I will have a lot to say about that essay.) Focusing on natural selection, Hayek goes macro: Who could possibly have predicted that living species would take the form they have taken? The answer, of course, is no one, and the reason is similar to the failures of prediction in sports events. We cannot know what will happen on the ground. Hayek also goes micro: “almost any event in the course of a man’s life may have some effect on almost any of his future actions,” which “makes it impossible that we translate our theoretical knowledge into predictions of specific events.” We will see plenty of examples.

For the outcomes of economic competition, Hayek urged, “the number of distinct elements involved” are “fairly large” (an understatement!): “the overall pattern that will form itself is determined by the significantly different behavior of the different individuals so that the obstacle of obtaining the relevant data cannot be overcome by treating them as members of a statistical collective.” This is the ball game once more. (Again we will see plenty of examples.)

Hayek concluded that economic theory can describe “kinds of patterns which will appear if certain general conditions are satisfied, but can rarely if ever derive from this knowledge any predictions of specific phenomena” (Hayek, Reference Hayek1967, p. 35). If you shuffle a deck of cards, you can be confident that no card is going to appear twice. That is a pattern prediction. But you cannot know in what order the cards will fall. So too, you can predict that price controls will create shortages, but you cannot specify their size, and you cannot offer a lot of details about who will be hurt most.

Return to evolution, one of Hayek’s favorite examples (Hayek, Reference Hayek and Bunge1964). You can be clear that because of natural selection, organisms will adapt (that is a pattern prediction). But you cannot know what specific organisms will emerge. In many domains, you can say something about the form or class of outcomes, without being able to make specific predictions with any reliability. You probably cannot predict, for example, what movies will be most successful in a given year, or foresee the outcome of a presidential election five years in advance. Too many variables are in play.

2 Mistakes

Hayek was keenly interested in psychology. In some ways, that topic might have been his deepest passion. He wrote an ambitious and lengthy book on the subject, in which he emphasized that human beings do not have unmediated access to physical reality and instead see it through categories of their own making (Hayek [1952], Reference Hayek2014a). The book presaged a great deal in the way of modern understandings, and much of Hayek’s argument holds up today. Still, nothing in his work in psychology foresaw modern behavioral science, and it would be implausible to say that Hayek was a behavioral economist in the contemporary sense. Nothing in his work anticipates behavioral findings;Footnote ² he did not explore individual biases, such as present bias, optimistic bias, or availability bias. He did not have much to say about the potential relevance of behavioral findings to institutional design, to the role of government, or to the price system.

We do have to be careful here. 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” (Hayek, Reference Hayek1948, pp. 8–9). In a passage that should be more widely known, he lamented “the current misconceptions of the individualism of Adam Smith and his group in the common belief that they have invented the bogey of the ‘economic man’ and that their conclusions are vitiated by their assumption of a strictly rational behavior or generally by a false rationalistic psychology. They were,” Hayek insisted, “very far from assuming anything of this kind” (Hayek, Reference Hayek1948, p. 11).

On the contrary, “in their view man was by nature lazy and indolent, improvident and wasteful,” and “it was only by the force of circumstances that he could be made to behave economically or carefully to adjust his means to his ends. But even this would be unjust to the very complex and realistic view which these men took of human nature” (id.). Thus Hayek claimed that “we can learn more about the behavior of men” from Smith “than from most of the more pretentious modern treatises on ‘social psychology’” (id.). So, perhaps, might Hayek say about contemporary behavioral economics.

We have seen that Hayek was much more than alert to the fact that individual choosers lack important information. That problem played a defining role in his thinking. He often emphasized the extent to which each of us lives amidst, and benefits from, a set of norms, cultural understandings, traditions, and institutions that were not designed by anyone, that have been built up over time, that serve essential functions, and that we do not and cannot understand (Hayek, Reference Hayek and Caldwell2014b, p. 342):

If there is a Hayekian theory of behavioral biases, it might begin there. Call it “Undesigned Rules Neglect.” It is a serious bias, and to Hayek, the most serious of all. But if anything, Hayek is best taken to suggest not that people are knowledgeable and unbiased, but that despite their ignorance and their biases, their actual practices work well enough, or even magnificently well, as they act in accordance with rules that help us to cope with our inevitable ignorance (Caldwell, Reference Caldwell2003; Vanberg, Reference Thaler and Sunstein2017).

Hayek’s frequent emphasis on the relevance of our ignorance, and our departures from perfect rationality, to the question of institutional design is exceedingly important, and central to his thinking. Still, it should not be confused with a suggestion that choosers err when they make food choices, when they decide how to respond to safety risks, or when they select appliances or pension plans. And when Hayek spoke of the ignorance of planners, he was casting doubt on paternalism – not because choosers know best, but because planners know worst. One of his particular concerns was the process of learning and discovery that choice makes possible.

John Stuart Mill was of course the great liberal critic of paternalism, and the relationship between Hayek and Mill is both immensely complicated and fascinating (Caldwell, Reference Caldwell2008); it will be worthwhile to turn to Mill in that light. Mill’s central arguments did not involve autonomy as such, but knowledge of relevant facts, and who was most likely to have it. In On Liberty, Mill insisted:

Mill’s argument for his famous Harm Principle is emphatically epistemic. Like Hayek, Mill believed that choosers are in the best position to know what is good for them. In Mill’s view, the problem with outsiders, including but not limited to government officials, is that they lack the necessary information. Mill contended that the individual “is the person most interested in his own well-being,” and the “ordinary man or woman has means of knowledge immeasurably surpassing those that can be possessed by any one else.” When society seeks to overrule the individual’s judgment, it does so on the basis of “general presumptions,” and these “may be altogether wrong, and even if right, are as likely as not to be misapplied to individual cases.” If the goal is to ensure that people’s lives go well, Mill concludes that the best solution is to allow people to find their own path. (As discussed in more detail in Section 5, Mill probably developed these ideas together with Harriet Taylor Mill; see Mill & Mill [1859], 2026)

Hayek agreed. But Hayek did not endorse the Harm Principle. He did say that “the most conspicuous attribute of liberalism that distinguishes it as much from conservatism as from socialism is the view that moral beliefs concerning matters of conduct which do not directly interfere with the protected sphere of other persons do not justify coercion” (Hayek [1960], Reference Hayek2011, p. 524). That sounds close to Mill, and it might even be seen as an effort to update Mill and be more precise; “interfere with the protected sphere of other persons” is more careful than “harm to others.” But it goes by very quickly, and there is no sustained analysis, here or anywhere else, of Mill’s account of liberty. The most specific statements can be found in The Constitution of Liberty (Hayek [1960], Reference Hayek2011, p. 212), where Hayek notes that Mill offered “a distinction between actions that affect only the acting person and those which also affect others.” Hayek said that the distinction “has not proved very useful,” because “there is hardly any action that may not conceivably affect others.” For Hayek, “the important question is whether the actions of other people that we wish to see prevented would actually interfere with the reasonable expectations of the protected person.” That raises its own questions, of course. What are those reasonable expectations? Hayek was clear that “the pleasure or pain that may be caused by the knowledge of other people’s actions should never be regarded as a legitimate cause for coercion.”

Hayek did know Mill’s work well, and had a great deal to say about it (Hayek [1951], Reference Hayek and Peart2015), as we will see in more detail in Section 5. Recall that Hayek never claimed that individuals are unerring in their judgments about what is best for them. Far more cautiously, and characteristically, he said that “nobody can know who knows best,” and “the only way we can find out is through a social process” (Hayek, 1948). It is speculative to wonder, but as a restriction on government action, the Harm Principle might sometimes have seemed, to Hayek, to be a bit too stringent. Is it so clear that he would have opposed motorcycle helmet laws? Laws requiring people to buckle their seatbelts? At the same time, Hayek clearly thought that the Harm Principle is too permissive. If you do not laugh at my jokes, or if you stop reading this Element, have you not harmed me? If you decide to stay home rather than to go to the local restaurant, have you not harmed that restaurant? We have evidence that Hayek did not embrace the Harm Principle for just this reason. Recall the passage just quoted from The Constitution of Liberty, and in his seminar at the University of Chicago, Hayek worried over the potential permissiveness of the Harm Principle.

However that may be, Hayek’s arguments against coercion, and Mill’s defense of the Harm Principle, have been put under considerable pressure by behavioral findings. Emphasizing the occasional human propensity to blunder, and the incentives of sellers to exploit those blunders, some people have questioned whether the price system is always so marvelous (Akerlof & Shiller, Reference Akerlof and Shiller2015), and have been asking whether new institutional arrangements, involving mandates and bans, have a fresh justification (Conly, Reference Conly2013; Bubb & Pildes, Reference Bubb and Pildes2014).

Some of the answers seem profoundly anti-Hayekian; because coercion is involved, they might well be taken, on Hayekian premises, to be indefensible abridgements of liberty. But the motivation for those answers is clear: If we know that people’s choices lead them in the wrong direction, why should we insist on freedom of choice? In the face of human errors, is it not odd, or even perverse, to insist on that form of freedom? Is it not especially odd to do so if we know that in many contexts, people choose wrongly, thus injuring their future selves? Ought not institutional design to take behavioral biases into account?

3 Hayekian Fuel Economy Standards

I now turn to a case study involving fuel economy mandates. I explore the possibility of rejecting the price signal and defending such mandates, as opposed to economic incentives, by reference not only to externalities but also to behavioral market failures, captured in insufficient consumer attention, ex ante, to economic savings (reduced costs of operating a car) and time savings (fewer trips to the gas station). The more general goal is to ask whether expert institutions that impose such mandates may reduce “internalities,” understood as the costs that choosers impose on their future selves (Allcott & Sunstein, Reference Allcott and Sunstein2015). Note well: Hayek did not trust experts much, so we have to approach that question with fear and trembling.

Nonetheless, I intend the issue of fuel economy mandates to be exemplary. A similar analysis might be made of cigarette taxes, taxes on sugar-sweetened beverages (Allcott, Lockwood, & Taubinsky, Reference Allcott, Lockwood and Taubinsky2019), savings policy, mask mandates, and many other problems. Admittedly, this is a bit of a venture, and perhaps an adventure. After sketching the argument, I will explore whether, on broadly Hayekian grounds, the argument might be found convincing. As we will see, the verdict will be ambivalent. (I hope you won’t mind.)

Here is my starting point: In principle, fuel economy mandates, perhaps imposed by institutions charged with environmental protection, might simultaneously reduce internalities and externalities. On imaginable assumptions, such mandates might turn out to have higher net benefits than carbon taxes, because the former, unlike the latter, deliver consumer savings.Footnote ⁷ To say the least, this is not a conventional view, because fuel economy standards are a highly inefficient response to the externalities produced by motor vehicles, especially when compared to optimal corrective taxes (Karplus et al., 2013).

In the end, everything turns on whether the imaginable assumptions turn out to be true. What behavioral economists consider to be errors might be viewed, by some neo-Hayekians (but not, I speculate, Hayek himself!), as not errors at all. My goal is not to run the numbers or to reach a final conclusion, but to make three more general points, which are that (1) mandates might turn out to be justified on welfare grounds, (2) the standard economic preference for economic incentives misses something of considerable importance, and (3) a Hayekian approach allows us to make some progress in assessing (1) and (2).

Most motor vehicles emit pollution, including greenhouse gases, and the use of gasoline increases national dependence on foreign oil. Recall Hayek’s acknowledgement: “Nor can certain harmful effects of deforestation, of some methods of farming, or of the smoke and noise of factories, be confined to the owner of the property in question or to those who are willing to submit to the damage for an agreed compensation. In such instances we must find some substitute for the regulation by the price mechanism” (Hayek [1944], Reference Hayek2007, p. 87). With respect to greenhouse gases, we are dealing with something akin to “the smoke and noise of factories.”

But what is the right substitute? This is a question of institutional design. On broadly Hayekian grounds, some kind of cap-and-trade system or corrective tax is the best response, designed to ensure that drivers internalize the social costs of their activity. The reason is largely epistemic: as compared to regulatory mandates, cap-and-trade systems and corrective taxes allow for competition and leave the private sector with a great deal of flexibility and a capacity for plenty of innovation (Potts, Reference Pichert and Katsikopoulos2018). They can find their own means to the chosen goal, which means that economic incentives of some kind, and not mandates, are the appropriate instrument (Stewart & Wiener, Reference Smithson2003). They can compete over means, and competition is a discovery procedure (Hayek, Reference Hayek and Caldwell2014b). For any given reduction in pollution levels, incentives impose a lower cost (Ackerman & Stewart, Reference Ackerman and Stewart1985). They preserve the price signal and the marvel that is the market. They allow competition to serve as a discovery procedure. The choice between cap-and-trade and carbon taxes raises a host of important questions, but they are not at issue here.Footnote ⁸

For obvious reasons, a great deal of recent analysis has been focused on greenhouse gas emissions and how best to reduce them (Nordhaus, Reference Nordhaus2015). In principle, regulatory institutions have a host of options. They might create subsidies (say, for electric cars). They might use nudges (say, by providing information about greenhouse gas emissions on fuel economy labels) (Sunstein & Reisch, Reference Sunstein and Reisch2014). They might impose regulatory mandates (say, with fuel economy and energy efficiency standards). Careful analysis suggests that carbon taxes can produce reductions in greenhouse gas emissions at a small fraction of the cost of fuel economy mandates (Karplus et al., 2013; Davis & Knittel, Reference Knittel2019; Knittel, Reference Knittel2019). On one account, “a fuel economy standard is shown to be at least six to fourteen times less cost effective than a price instrument (fuel tax) when targeting an identical reduction in cumulative gasoline use” (Karplus et al., Reference Karplus, Paltsev, Babiker and Reilly2013).

These are points about how best to reduce externalities, and on Hayekian grounds, they seem decisive. But behaviorally informed regulators focus on consumer welfare, not only externalities. They are concerned about a different kind of market failure. They speculate that at the time of purchase, many consumers might not give sufficient attention to the costs of driving a car (Bubb & Pildes, Reference Bubb and Pildes2014). Even if they try, they might not have a sufficient understanding of those costs, because it is not simple to translate differences in miles per gallon (MPG) into economic and environmental consequences (Larrick & Soll, Reference Larrick and Soll2008). An obvious response, preserving freedom of choice and in a more Hayekian spirit, would be disclosure, in the form of a fuel economy label that would correct that kind of behavioral market failure (EPA, 2015). In principle, such a label, if behaviorally informed, should solve the problem, and it avoids coercion. Consider Hayek’s words on the latter topic, again sounding quite Kantian (Hayek [1960], Reference Hayek2011, p. 71):

Labels do not do that. They do not coerce, and they also do not discriminate (a particular problem for Hayek, who opposed “commands,” which do discriminate). They allow for the use of dispersed information. They do not tinker with prices or quantities. In short: labels should be used to promote consumer welfare, by increasing the likelihood that consumers will make optimal choices, and corrective taxes should be used to respond to externalities. This might even be seen as a broadly Hayekian approach to the problem.

Such a label might, for example, draw consumers’ attention to the long-term costs or savings associated with motor vehicles. Focused on internalities, the existing label in the United States does exactly that; it specifies annual fuel costs and also five-year costs (or savings) compared to the average vehicle. Use of such a label has evident appeal, and it might well be right. But there are concerns. Once more, one set of them is unquestionably Hayekian: Who is producing the label, what do they know, and will companies commit fraud? We need to have some confidence that policymakers know what they are doing. There have been incidents of fraud, above all from Volkswagen, and we might fear that they are more pervasive than we know.

It would also be possible to wonder whether a label will be sufficiently effective; this is an empirical question, not resolvable in the abstract. Perhaps some or many consumers will pay too little attention to the label, and hence will not purchase cars that would save them a significant amount of money (Bubb & Pildes, Reference Bubb and Pildes2014). And if some or many consumers are genuinely inattentive to the costs of operating a vehicle (at the time of purchase), then it is possible that fuel economy standards, with a level of stringency that would be difficult to defend on standard economic grounds, might turn out to be justified.

In support of that argument, let us go directly to the merits, and put Hayek to one side for a moment. It would be useful for public institutions to focus directly on two kinds of consumer savings from fuel economy standards, involving internalities rather than externalities: money and time. In fact, the vast majority of the quantified benefits from recent fuel economy standards come not from environmental improvements, but from money saved at the pump; turned into monetary equivalents, the time savings, coming from the fact that people do not have to go to gas stations, are also significant. Under the Obama Administration, at least, the Department of Transportation found consumer savings of about $529 billion; time savings of $15 billion; energy security benefits of $25 billion; carbon dioxide emissions reductions benefits of $49 billion; other air pollution benefits of about $14 billion; and less than $1 billion from reduced fatalities (NHTSA, 2012). The total projected benefits were $633 billion over fifteen years, of which a remarkable 84 percent come from savings at the pump, and no less than 86 percent from those savings along with time savings (Bento et al., 2020). Pause over that if you would. Only about 14 percent of the total savings come from reducing externalities! That should make contemporary Hayekians pretty nervous. We will see why in a moment.

In its own rulemaking, the Trump Administration rethought those numbers by reference to recent workFootnote ⁹ raising questions about whether consumers are insufficiently attentive to the economic savings, but it projected the consumer savings to be in the same general vicinity and actually even higher. The Biden Administration produced numbers that are broadly similar to those in predecessor administrations, in the sense that once again, the strong majority of the monetized benefits come from consumer savings.Footnote ¹⁰ The Trump Administration has reoriented its analysis on the authority of recent work (Busse et al., Reference Busse, Knittel and Zettelmeyer2013; Sallee et al., Reference Salganik, Lundberg and Kindel2016; Allcott & Knittel, Reference Knittel2019), raising questions about whether consumers are insufficiently attentive to the economic savings.

Hayekians will have many questions and doubts here. Can we trust any of these numbers? Are they not speculative in the extreme? Perhaps public officials – planners if not socialists – just do not know enough to get the numbers close to right. A further problem, and an even more fundamental one, is that on standard economic grounds, which modern Hayekians should be inclined to support, it is not at all clear that these consumer benefits are entitled to count in the analysis. The reason is that they are purely private savings and do not involve externalities in any way (Gayer & Viscusi, Reference Gayer and Viscusi2013). Shouldn’t government simply let people buy the cars they want? In deciding which cars to buy, consumers can certainly take into account the private savings from fuel-efficient cars. If they choose not to buy such cars, it might be because they do not value fuel efficiency as compared to other vehicle attributes (such as safety, aesthetics, and performance). Where is the problem? Where is the market failure? Isn’t the government engaged in pure paternalism, a form of coercion?

It is true that the paternalism and the coercion could be worse. With a fuel economy standard, the government does not tell people: You must buy this car and not that one. Any such standard allows plenty of room for freedom of choice, for competition, and for innovation. Still, the whole fleet is required to become more fuel-efficient, which means that we are not simply relying on the market. And a main reason that we are not doing that is that we think that consumers are not making the right choices on their own.

If the problem lies in a lack of information, the standard economic prescription is the same as the behaviorally informed one: Fix the label and provide that information so that consumers can easily understand it. We can understand this prescription in Hayekian terms as a plea for an educative nudge (Thaler & Sunstein, Reference Sunstein2021), rather than a more aggressive kind of intervention. And indeed, educative nudges are a key tool in the toolbox of the behaviorally informed policymaker. They might be seen as part of the same general program favored by those who prefer “boosts,” understood as efforts to inform and educate people so as to promote their own agency (Hertwig, Reference Hertwig2017). Such efforts might be seen as Hayekian in spirit.

We have seen, however, that even with the best fuel economy label in the world, consumers might turn out to be insufficiently attentive to the benefit of improved fuel economy at the time of purchase, not because they have made a rational judgment that they are outweighed by other factors, but simply because they focus on other variables, such as performance, size, and cost (Gabaix & Laibson, Reference Gabaix and Laibson2006). The problem may be one of insufficient attention (Gabaix, Reference Gabaix2017). A behavioral hunch, discussed below, is that automobile purchasers do not give adequate consideration to economic savings.Footnote ¹¹ Apart from savings, there is the question of time: How many consumers think about time savings when they are deciding whether to buy a fuel-efficient vehicle?

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 (Sunstein, Reference Sunstein2019b). Energy efficiency requirements might be justified in similar terms, and indeed, the argument on their behalf might be stronger.Footnote ¹² 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. And indeed, the US Government has so argued (EPA, 2010). The central conundrum has been referred to as the Energy Paradox in this setting (and in several others). In short, the problem is that consumers appear not to purchase products that are in their economic self-interest. There are strong theoretical reasons why this might be so:

• – Consumers might be myopic and hence undervalue the long-term.

• – Consumers might lack information or a full appreciation of information even when it is presented.

• – Consumers might be especially averse to the short-term losses associated with the higher prices of energy-efficient products relative to the uncertain future fuel savings, even if the expected present value of those fuel savings exceeds the cost (the behavioral phenomenon of “loss aversion”).

• – Even if consumers have relevant knowledge, the benefits of energy-efficient vehicles might not be sufficiently salient to them at the time of purchase, and the lack of salience might lead consumers to neglect an attribute that it would be in their economic interest to consider.

• – In the case of vehicle fuel efficiency, and perhaps as a result of one or more of the foregoing factors, consumers may have relatively few choices to purchase vehicles with greater fuel economy once other characteristics, such as vehicle class, are chosen.

Of course, public institutions should be cautious before accepting a behavioral argument on behalf of mandates or bans. Hayek himself was highly skeptical about reliance on the wisdom of experts: “One thing, in fact, which the work on this book has taught me is that our freedom is threatened in many fields because of the fact that we are much too ready to leave the decision to the expert or to accept too uncritically his opinion about a problem of which he knows intimately only one little aspect” (Hayek [1960], Reference Hayek2011, p. 50; see also Scheall, Butos, & McQuade, Reference Scheall, Butos and McQuade2019). On this ground, there is a fair Hayekian objection to the behavioral defense of fuel economy mandates.

Behavioral biases have to be demonstrated, not simply asserted, and Hayekians might be skeptical here; as I have noted, important research suggests that consumers do pay a lot of attention to the benefits of fuel-efficient vehicles.Footnote ¹³ Some of that research finds that with changes in gas prices, consumers adjust their purchasing decisions, strongly suggesting that in choosing among vehicles, consumers are highly attentive to fuel economy (Busse, Knittel, & Zettelmeyer, Reference Busse, Knittel and Zettelmeyer2013; Sallee, West, & Fan, Reference Salganik, Lundberg and Kindel2016). Other research points in the same direction. It finds that when aggressive steps are taken to inform consumers of fuel economy, they do not choose different vehicles, which suggests that a lack of information, and perhaps a lack of salience, are not causal factors here (Allcott & Knittel, Reference Knittel2019).

On the other hand, some evidence cuts the other way. A large-scale study of actual behavior finds that after a significant correction of erroneously stated MPG measures, consumers were relatively unresponsive (Gillingham, Houde, & van Benthem, Reference Gillingham, Houde and van Benthem2019). As Gillingham et al. write, “Using the implied changes in willingness-to-pay, we find that consumers act myopically: consumers are indifferent between $1 in discounted fuel costs and 15–38 cents in the vehicle purchase price when discounting at 4%.” Puzzlingly, many consumers do not buy hybrid vehicles even in circumstances in which it would seem rational for them to do so (Duncan, Reference Duncan, Ku, Julian and Carley2019). According to the leading study, a significant number of consumers choose standard vehicles even when it would be in their economic interest to choose a hybrid vehicle, and even when it is difficult to identify some other feature of the standard vehicle that would justify their choosing it.

It is also possible to think that even if consumers are responsive to changes in gasoline prices, they are still myopic with respect to choices of vehicles that have technological advances. Graham puts it crisply (Graham, Wiener, & Robinson, Reference Graham, Wiener and Robinson2019):

More broadly, the government’s numbers, finding no significant consumer welfare loss from fuel economy standards, are consistent with the suggestion that consumers are suffering from some kind of behavioral bias. At the same time, the government’s numbers, projecting costs and benefits, might be wrong, and a faithful Hayekian would suspect so (Gayer & Viscusi, Reference Gayer and Viscusi2013). Engineering estimates might overlook some losses that consumers will actually experience (a Hayekian worry). No one doubts that consumers have highly diverse preferences with respect to vehicles, and even though they are not mere defaults, fuel economy standards should be designed to preserve a wide space for freedom of choice (Greenstone, Sunstein, & Ori, Reference Greenstone, Sunstein and Ori2020). Appropriate standards ensure that such space is maintained. Economic incentives have inherent advantages on this count.

A real question, of course, is the magnitude of net benefits from alternative approaches. As we have seen, Hayek at one point urged that it is both appropriate and necessary to ask that question. If the consumer savings are assumed to be very large, fuel economy standards are likely to have correspondingly large net benefits. To give a very rough, intuitive sense of how to think about the comparative question, let us suppose that the US government imposed an optimal carbon tax. Simply for purposes of analysis, suppose that it is $50 per ton, understood to capture the social cost of carbon.Footnote ¹⁴ Suppose that in relevant sectors, including transportation, a certain number of emitters decide to reduce their emissions, on the grounds that the cost of reducing them is (on average) $Y, which is lower than $50. The net benefit of the carbon tax would be $50 minus Y, multiplied by the tons of carbon emissions that are eliminated. It is imaginable that the resulting figure would be very high. But it is not necessarily higher than the net benefits of well-designed fuel economy standards.

With these qualifications, an argument for fuel economy standards, made by reference to behavioral market failures and to internalities in particular, is at least plausible. In this context, nudges (in the form of an improved fuel economy label) and mandates (in the form of standards) might march hand-in-hand. It is true that if the goal is only to reduce externalities, a carbon tax is far better than a regulatory mandate. It is emphatically true that for Hayekian reasons, the best approach to internalities should not be coercion but instead appropriate disclosure, designed to promote salience and to overcome limited attention. Even if we trust the empirical findings, we might not trust regulatory institutions and their capacity to assess them (Schubert, Reference Scheall, Butos and McQuade2017). But with an understanding of behavioral findings, a regulatory approach, promoting consumer welfare as well as reducing externalities, might turn out to have higher net benefits than the standard economic remedy of corrective taxes and disclosure.

With respect to both content and institutional design, everything turns on what the evidence shows, and on the particular numbers. But in principle, regulation of other features of motor vehicles could also be justified in behavioral terms; certain safety equipment might not be sufficiently salient to consumers at the time of purchase, and some such equipment might fall in the category of experience goods (Sunstein, Reference Sunstein2019b). Credit markets can be analyzed similarly (Sarin, Reference Salganik, Lundberg and Kindel2019). The broadest point is that while a presumption in favor of freedom of choice makes a great deal of sense (Glaeser, Reference Glaeser2006), it is only a presumption. It might be overcome, especially when it can be shown that behavioral biases are having significant effects.

How can that be done? What are the appropriate institutional arrangements? Can an approach be adopted that might fairly claim to be Hayekian? Maybe not. In view of Hayek’s skepticism about coercion, planner ignorance, and top-down expertise, the answer might ultimately turn out to be “no” (Schubert, Reference Scheall, Butos and McQuade2017). That would be a reasonable view; it is probably right, at least if we are trying to stay close to what Hayek actually wrote.

But return to the five questions in Section 2. At least in the first instance, and possibly in the last, behaviorally informed policy ought to be based not on the preferences and values of social planners, but on learning from the choices of informed and unbiased choosers. It might well turn out to be possible to identify those choices. Once we have done so, we might be on the road toward identifying appropriate interventions, whether they involve nudges, taxes, subsidies, or mandates. True, it would be extravagant to claim that those interventions, defended by reference to people’s choices under epistemically favorable conditions, are Hayekian. But it may not be extravagant to say that they are in Hayek’s general spirit, and respectful of his most fundamental concerns.

4 The AI Calculation Debate and the Socialist Calculation Debate Are the Same Thing

Can socialism work?

The Socialist Calculation Debate is about that question (Boettke, Candela, & Truitt, Reference Boettke, Candela and Truitt2024). Very simply, and once upon a time: Some people, including Otto Neurath and Abba Lerner, argued that central planners could dispense with the price signal and set prices by trial-and-error. Private property, in their view, was not necessary. Austrian economists, including and perhaps above all, Hayek, disagreed. They insisted that socialist planners cannot possibly know what they need to know in order to make socialism work. They cannot rationally set prices and wages for next year or even next month. They are irremediably ignorant. Their knowledge is hopelessly inadequate. Any judgments on their paper will be confounded by circumstances.

These claims bear on numerous questions. Can planners, socialist or not, make a system of rent control work? Can planners, socialist or not, choose maximum hours laws for truck drivers or airline pilots? Can they decide to favor electric motor vehicles, and if so, exactly how? Can they have some success in deciding what energy sources to promote over the coming decades? Some people think that the answers to these questions are “yes,” or something like that. Hayek thought that the answer was clearly no. The calculations are simply not possible. Hayek won the Socialist Calculation Debate. Actually, it was a smashing victory (Boettke, Candela, & Truitt, Reference Boettke, Candela and Truitt2024).

I have referred to a paper Hayek published in 1964 (and in various places, including Hayek, Reference Hayek and Bunge1964 and Hayek, Reference Hayek1967), in which he explored the limits of prediction, with reference to the theory of evolution. Hayek emphasizes that Darwin’s theory points to a process or mechanism that need not have produced the same organisms that we observe on earth. The theory of evolution describes “a range of possibilities,” one that is extremely wide. And even if we knew (as we do not) everything about (1) the mechanism of mutation, (2) the circumstances in which particular mutations would appear, and (3) the precise advantages that any mutation would confer, we still would not be able “to explain why the existing species or organisms have the particular structures which they possess, nor to predict what new forms will spring from them” (Hayek, Reference Hayek1967, p. 33). This is a striking claim, and it is not intuitive.

Hayek explains that the reason for our ignorance is “the actual impossibility of ascertaining the particular circumstances which, in the course of two billion years, have decided the emergence of the existing forms, or even those which, during the next few hundred years, will determine the selection of the types which will survive” (Hayek, Reference Hayek1967, p. 33). The number of relevant facts is simply too large. It is not possible to insert them into some formula that could then spit out some predictions. Note that Hayek’s claim is very strong. He refers to an “actual impossibility,” thus suggesting that the problem cannot be solved.

In Hayek’s account, complex social phenomena have the same characteristics. In the social domain, “individual events regularly depend on so many concrete circumstances that we shall never in fact be in a position to ascertain them all; and that in consequence the ideal of prediction and control must largely remain beyond our reach.” To drive the point home, Hayek observes that “almost any event in the course of a man’s life may have some effect on almost any of his future actions,” which “makes it impossible” for us to “translate our theoretical knowledge into predictions of specific events.” Hayek acknowledges that the advances of science have produced a great deal of exuberance, but he is exuberant in his own way in offering this conclusion: “It is high time, however, that we take our ignorance more seriously” (Hayek, Reference Hayek1967, pp. 34, 39).

I am about to flip a coin. Can you predict whether it will be heads or tails?

You might have gotten it right. But if so, you were lucky. Whether it would come up heads or tails depends on an assortment of factors that could not be identified before I flipped the coin.

In principle, of course, reliable predictions could be made; if you knew everything relevant to whether a coin will come up heads or tails, you could of course make an accurate prediction. So let’s not talk about randomness. The problem is the absence of essential information. It is our ignorance. No wonder that the Apostle of Ignorance drew attention to that problem.

5 Hayek, Mill, and Tradition

How should we think about customs and traditions? Some people think that we should be skeptical of them. Economists worry that traditions can be a product of informational cascades, in which people follow others not because they know that others are acting sensibly, but simply because the actions of others seem to carry information (Sunstein, Reference Sunstein2019a). Cascade effects can lead people to do stupid things. We might buy products that do us no good. We might make foolish investments. Informational cascades might lead rational people to perpetuate stupid traditions (Ullmann-Margalit, Reference Sunstein and Reisch1997). Behavioral economists much worry over traditions and conformity effects. They emphasize that if people are boundedly rational, they might end up participating in the creation of shared practices, or traditions, that might not help anyone, or that might hurt a lot of people. People might be scared of things that are not dangerous. “Availability cascades” can lead to fears that are widely shared but baseless (Kuran & Sunstein, Reference Kuran and Sunstein1999).

But Hayek cherished traditions, as undesigned orders from which we greatly benefit. Of course, he accepted the idea that we might be able to improve on them, so long as we approach them with humility. But still, he cherished them. How should we think about that? Hayek did not have a simple answer to that question.

Hayek and John Stuart Mill help to define the liberal tradition, but in both temperament and orientation, they could not be further apart. Mill was a progressive, a social reformer, an optimist about change, in some ways a radical. In some ways, he was a rationalist or a constructivist, of the sort that Hayek despised. Mill believed that, properly understood, liberalism calls for significant revisions in the existing economic order, which he saw as palpably unjust: “The most powerful of all the determining circumstances is birth. The great majority are what they were born to be” (Mill, Reference Mill1879, p. 31).

Hayek was not exactly a conservative; in fact, he was sharply critical of conservatism on the ground that it was largely oppositional and did not offer an affirmative position. He did not think that conservatives adequately understood the nature and importance of undesigned orders, the problems with scientism, and what follows from an appreciation of the knowledge problem. We have seen that Hayek did not like coercion, including coercion of the sort that conservatives favor. Still, he is a great hero to many conservatives. One reason is that he generally venerated traditions and longstanding practices, seeing them as embodying the views and knowledge of countless people over long periods (Caldwell, Reference Caldwell2003).

Hayek greatly admired Burke, who (it will be recalled) attacked the idea that self-styled reformers, equipped with an abstract theory, should feel free to override social practices that have stood the test of time. Hayek also admired Mandeville, who spoke of law this way (Mandeville, Reference Mandeville and Kaye1924, ii, p. 142): “We often ascribe to the excellency of man’s genius, and the depth of his penetration, what is in reality owing to the length of time, and the experience of many generations.” To be sure, Hayek did not think that traditions can never be changed. But he liked incremental change rather than wholesale change, and he insisted that traditions often capture the wisdom of many people, extending over generations. They are undesigned orders, and as we have seen, he liked undesigned orders a lot.

In a provocative lecture in 1983, Hayek offered, as his title, a sentence from David Hume: “The rules of morality are not the conclusions of our reason” (Hayek, Reference Hayek1983). By this, Hayek did not mean that the rules of morality are a product of passion rather than reason. Instead, he meant that the rules of morality are a product of an evolutionary process that we should regard with reverence. “The traditional rules of morality are,” Hayek wrote, “an indispensable condition of the very existence of present mankind which we cannot alter at will to please our tastes, and which we can at most endeavor gradually to develop or improve within a framework that is given to us” (p. 1). Hayek distinguished between David Hume’s view, which he shared, and the utilitarian view, “which is precisely the kind of rationalist explanation of morals” that ought to be opposed (p. 2). What is crucial is that “man has neither designed nor does understand his own morals” (p. 4). Hayek singles out “the rules which define the institutions of individual property and those of the family” (p. 5).

Those rules, Hayek thought, are a product of “selective evolution.” That means that “the greater part of present humanity owes their very lives to the observation of traditional rules – rules which they did not like because they consisted of restraints on their innate instincts, and were beyond their capacity intellectually to justify” (p. 7). Hayek did not like, and meant to oppose, the views of the Enlightenment rationalists, who sought “precisely to free us from all such traditional beliefs” (p. 7). In his view, the rationalists “imagine that they can invent for us a better moral which will secure for us a more pleasant, more beautiful, and more just world,” but they “ignore how much we owe to traditional morals” (p. 8). Hayek rejected the belief “that it was man’s intellect which enabled him to design his morals, and therefore also, where the results do not satisfy his desires, to replace them by better ones” (p. 10). This is the fatal conceit.

Still, Hayek did not favor morals legislation. He insisted that “where private practices cannot affect anybody but the voluntary adult actors, the mere dislike of what is being done by others, or even the knowledge that others harm themselves by what they do, provides no legitimate ground for coercion” (Hayek [1960], Reference Hayek2011, p. 212). That sounds a lot like Mill in On Liberty. As we have seen, Hayek thought that Mill’s Harm Principle was too permissive. But he stood with Mill in rejecting the view that traditions, as such, are an adequate basis for coercion. Indeed, and strikingly, he thought that “the most conspicuous instance of” the illegitimate use of coercion “is that of the treatment of homosexuality.” In his account, “private practice among adults, however abhorrent it may be to the majority, is not a proper subject for coercive action for a state whose object is to minimize coercion.” That is a striking claim, especially when we consider that it was made in 1960. We might note that Hayek did not give a careful argument on behalf of this conclusion, as he did for free markets. (Was he a pro tanto rationalist?) Still, it is what he thought.

Mill had an abstract theory, one based on a conception of liberty from both government and oppressive social customs, and he thought that society could be evaluated by reference to it. Thus the beginning of The Subjection of Women (Mill, Reference Mill1869, p. 1):

To that extent, Mill was indeed a rationalist in Hayek’s sense. And consider these words, which Hayek could never have written (Mill, Reference Mill1869, pp. 43–44): “Stupidity is much the same all the world over. A stupid person’s notions and feelings may confidently be inferred from those which prevail in the circle by which the person is surrounded. Not so with those whose opinions and feelings are an emanation from their own nature and faculties.”

More rationalism, no doubt about it. Understanding Hayek and Mill, and what unifies and separates them, is fundamental to understanding the liberal tradition, and what contemporary Hayekianism might look like.

Against this background, there is every reason to be intrigued by a book with the title Hayek on Mill (Hayek [1951], Reference Hayek and Peart2015). Hayek died in 1992, but the University of Chicago Press is continuing to publish a multivolume edition of his collected works. Readers have been discovering essays by Hayek that were never published, were not easily available, or were not widely known. What would Hayek have to say about a great champion of liberty, in some ways his intellectual ancestor, a sharp critic of traditions who ended up embracing socialism? What would Hayek, one of the great twentieth-century liberals, say about Mill, the greatest nineteenth-century liberal?

How stunning, then, to find that Hayek offered only a few snippets on that question. His treatment, first published in 1951, grew out of an enormous, uncharacteristic, and somewhat obsessive undertaking. Startlingly, Hayek assembled what remains of the correspondence between Mill and his eventual wife, Harriet Taylor (one or the other destroyed numerous letters, probably including the most interesting), and used it as the basis for a narrative account of their mysterious love affair.

The book raises mysteries of its own. For all his greatness, Hayek was not exactly a romantic writer. Of course he celebrated the operations of free markets, but he did not display a lot of interest in the full range of human emotions. Was he passionate? In a way, about markets and the rule of law (as he understood them). Some liberals, including Mill, have a romantic streak in their writing; Hayek did not. How was it, exactly, that Hayek, of all people, became captivated by the story of John Stuart Mill and Harriet Taylor?

Here is one answer: Hayek’s personal life was not free of storms, and some of the storms overlapped with Mill’s, in the sense that he had a passionate romantic relationship, eventually leading to a second marriage of his own (Ebenstein, Reference Ebenstein2018; Caldwell & Klausinger, Reference Caldwell and Klausinger2022). A more academic answer is that he had to explain to himself and others why Mill – one of the few thinkers he had to regard as an intellectual equal or superior – moved away from what Hayek celebrated as classical liberalism, which for Hayek was focused on limited government and protection of free markets. But Hayek’s interest in the romance itself outpaced his interest in the evolution of Mill’s thinking (perhaps because of the beauty and great delicacy of the correspondence). Still, Mill was a romantic liberal, seeing liberalism itself as a kind of romance; Hayek was anything but that, and he saw liberalism as a marvel, but also as anything but a kind of romance.

Does Mill’s relationship with Harriet Taylor have anything to do with liberalism, liberty, and economic theory? I think so (Mill & Mill [1859], 2026). One of the lessons we can draw from Hayek’s work of excavation is that Mill’s distinctive form of liberalism, with its emphasis on individual freedom from the confining effect of social norms, owed a large debt to his relationship with Taylor. As we shall see, Hayek himself missed the connection entirely, because his own preoccupations lay elsewhere. Still, Hayek was focused on something that did indeed divide him from Mill, and it had to do with the right attitude toward customs and traditions. Mill and Hayek were, in a way, intellectual siblings, but Hayek’s focus on undesigned orders, and the dispersed knowledge they reflect, was not by any means Mill’s emphasis. That focus was Hayek’s favorite thing; Mill’s favorite thing, I think, was individual liberty as he understood it. On Hayek’s grounds, Mill suffered from a kind of hubris. In the end, Mill was a kind of planner, and Hayek could not abide that.

Hayek begins with one of his central puzzles, and it involves Taylor rather than Mill: “The literary portrait which in the Autobiography John Stuart Mill has drawn for us of the woman who ultimately became his wife creates a strong wish to know more about her” (Hayek [1951], Reference Hayek and Peart2015, p. 9). Mill’s own account suggests that she must have been “one of the most remarkable women who ever lived.” Hayek quotes a very long passage from Mill himself (Mill, Reference Mill1873, pp. 186–187, as cited in Hayek [1951], Reference Hayek and Peart2015):

Mill had a lot more to say about Harriet Taylor (Reference Mill1859b, p. v): “Were I [but] capable of interpreting to the world one half the great thoughts and noble feelings which are buried in her grave, I should be the medium of a greater benefit to it, than is ever likely to arise from anything that I can write, unprompted and unassisted by her all but unrivalled wisdom.”

It is somehow right, and also deeply moving, that an extraordinary edition of On Liberty, published in 2026, finally credits Harriet Taylor as the book’s coauthor (Mill & Mill [1859], 2026). Mill would have loved that – Hayek not so much. One of Hayek’s projects is to discover whether Mill’s account of his wife was “sheer delusion” (Hayek [1951], Reference Hayek and Peart2015, p. 24).

Mill and Taylor met at a dinner in 1830, when she was just twenty-two, a mother of two boys, and married for four years to John Taylor, eleven years older than she and a junior partner in a family firm of wholesale druggists. Thomas Carlyle called him “an innocent dull good man” (p. 20). An acquaintance describes her, at the time, as “possessed of a beauty and grace quite unique of their kind,” with “large dark eyes, not soft or sleepy, but with a look of quiet command in them” (Hayek [1951], Reference Hayek and Peart2015, p. 21). (The best words here are “grace” and “quiet command.”) She wrote poetry and was soon to produce a number of essays on social usages and conventions, including one that strongly prefigured Mill’s attacks, or their attacks, on conformity, decades later, in On Liberty (Reference Mill, Mill, Turner, Jacobs, McCabe, Osburg, Shefczyk and Schmidt-Petri1859) (p. 22).

For his part, Mill was nothing like the dry, somewhat desiccated old man depicted in photographs. Twenty-four at the time, he must have cut a dashing figure, having been described by Carlyle as “a slender, rather tall and elegant youth,” who was “remarkably gifted with precision of utterance, enthusiastic, yet lucid, calm” (p. 27). At the same time, his emotional state was not good. In a forlorn letter to a friend, written a year before meeting Taylor, he referred to “the comparative loneliness of my probable future lot,” and contended poignantly that there was “now no human being … who acknowledges a common object with me” (p. 33–34).

In his autobiography, Mill insisted that it was not until years after meeting Taylor that their relationship “became at all intimate or confidential” (Reference Mill1873) (p. 37). Hardly. Referring to an article published in mid-1831, Taylor’s closest friend pointedly wrote her, “Did you or Mill do it?” (p. 37). In the same year, a letter from a mutual friend, written to John Taylor, spoke mysteriously of the need for a “reconciliation” between Mr. Taylor and Mill (p. 38). In 1832, Mrs. Taylor wrote to Mill that they must not meet again, to which Mill responded in French: “Sa route et la mienne sont séparées, elle l’a dit: mais elles peuvent, elles doivent, se rencontrer. A quelqu’ époque, dans quelqu’ endroit, que ce puisse être, elle me trouvera toujours ce que j’ai été, ce que je suis encore” (p. 39). (“Her path and mine are separate, she said so: but they can, they must, come together. At whatever time, in whatever place that might be, she will find me forever as I was, as I am still.” If you have ever been in a real romance, those words will speak directly to you.) A few weeks later, their relationship resumed.

By 1832, the two had embarked on some kind of love affair. Taylor wrote Mill: “Far from being unhappy or even low this morning, I feel as tho’ you had never loved me half so well as last night” (Hayek [1951], Reference Hayek and Peart2015, p. 48). And later, in response to an apparent confession from Mill (p. 49):

The Taylors agreed to a separation, and Mill and Harriet were able to spend time together. To a close friend, Mill wrote a rapturous letter (Hayek [1951], Reference Hayek and Peart2015, p. 51): “I am astonished when I think how much has been restrained, how much untold, unshewn and uncommunicated till now … . Not a day has passed without removing some real & serious obstacle to happiness … . There will never again I believe be any obstacle to our being together entirely.”

Taylor wrote in a similar spirit, stating that “there has been so much more pain than I thought I was capable of, but also O how much more happiness” (p. 51).

Hayek writes that in the middle of the 1830s, Mill and Taylor did not try to conceal their intimacy but confronted with a great deal of malicious gossip, they withdrew almost entirely from social life (2015). In 1834, Carlyle wrote of the rumor that Mill had “fallen desperately in love with some young philosophic beauty” and been “lost to all his friends and to himself” (pp. 54–55). Enraged by the gossip, Mill cut off a number of his friendships. He wrote to a friend: “… what ought to be so much easier to me than to her, is in reality more difficult – costs harder struggle – to part company with the opinion of the world, and with my former mode of doing good in it” (p. 51). You might reasonably speculate that Mill’s eventual writing on conformity, and on the importance of individuality, had some connection with his personal experiences.

It is not entirely clear what happened between Mill and Taylor from the middle 1830s to the late 1840s. What little remains of their correspondence shows a degree of agitation within and between them. Taylor to Mill: “I don’t know why I was so low when you went this morning. I was so low – I could not bear your going my darling one; yet I should be well enough accustomed to it by now” (p. 93). The two were alert to the reactions of others; Mill seemed especially sensitive on that count. Taylor to Mill, perhaps teasingly: “I was not quite wrong in thinking you feared opinions. – I never supposed you dreaded the opinions of fools but only of those who are otherwise wise & good but have not your opinions about Moralities” (p. 94). But there can be little doubt about the intensity of their relationship. Taylor to Mill: “When I think that I shall not hold your hand until Tuesday the time is so long & my hand so useless. Adieu my delight” (p. 103).

In 1848, Taylor returned to London after traveling with Mill, only to discover that her husband had fallen gravely ill with cancer. For a period of two months, she dedicated herself entirely to caring for him and saw Mill not at all, restricting herself to correspondence, some of it angry, even bitter, and full of dignity (Hayek [1951], Reference Hayek and Peart2015, p. 152): “You talk of my writing to you ‘at some odd time when a change of subject of thought may be rather a relief than otherwise’! odd time! Indeed you must be ignorant profoundly of all that friendship or anxiety means when you can use such pitiful narrow hearted expressions.”

And as her husband neared the end: “The sadness & horror of Nature’s daily doings exceed a million fold all the attempts of Poets! There is nothing on earth I would not do for him & there is nothing on earth which can be done. Do not write” (p. 159). (Those who read Taylor’s letters might be excused for falling in a little in love with her.)

In 1851, two years after her husband’s death, she and Mill were married. The event must have been joyful, a kind of completion, but as Hayek reports, “the marriage led to the most painful episode in Mill’s life, his complete break with his mother and her other children” (p. 167). The occasion for the break is yet another mystery. It must have had something to do with his reaction to their disapproval or their sense of scandal. But who knows? Somewhat unkindly, Hayek describes it as “almost as unintelligible to his relations as to us” (pp. 167–68). Twenty years after the break, his sister Harriet expressed genuine bafflement, reporting that “up to the time of his marriage he had been everything to us,” and “it was a frightful blow to lose him at once and for ever, without [one] word of explanation, – only in evident anger” (p. 168). Anger at what, exactly?

The marriage itself was quiet, productive, and supremely happy, but both husband and wife suffered from a series of illnesses. In 1854, Mill believed himself to be gravely sick. He wrote (p. 199): “The only change I find in myself from a near view of probable death is that it makes me instinctively conservative. It makes me feel, not as I am accustomed – oh, for something better! – but oh, that we could be going on as we were before. Oh, that those I love could be spared the shock of a great change!”

But he recovered well, and it was Taylor who became desperately ill four years later. Mill wrote an appeal to a doctor: “I implore you to come immediately. I need hardly say that any expense whatever will not count for a feather in the balance” (p. 252). He never came (perhaps because the letter arrived too late).

Shortly after her death, Mill wrote: “It is doubtful if I shall ever be fit for anything public or private, again. The spring of my life is broken. But I shall best fulfil her wishes by not giving up the attempt to do something useful” (Hayek [1951], Reference Hayek and Peart2015, pp. 253–54). On Liberty was published in 1859 and dedicated “to the beloved and deplored memory of her who was the inspirer, and in part the author, of all that is best in my writings” (p. 256). Mill lived to 1873, and many of his greatest works, some perhaps coauthored (Mill & Mill [1859], 2026), appeared after Taylor’s death.

Remarkably, Hayek ends his book saying very little about Taylor’s influence on Mill’s thought. But in fragments of the book, and in other essays he wrote, we can uncover Hayek’s views on the mysteries with which he began. Hayek agrees that Taylor’s “influence on his thought and outlook, whatever her capacities may have been, were quite as great as Mill asserts” (p. 373). At the same time, Hayek concludes, “they acted in a way somewhat different from what is commonly believed. Far from it having been the sentimental it was the rationalist element in Mill’s thought which was strengthened by her influence” (p. 374).

Recall that Hayek saw rationalism, constructivism, and scientism as essentially the same thing, and it was what he spent a lifetime arguing against. In the end, he thought, Mill was on the wrong side, and it was Taylor who put him there. Recall that in Hayek’s view, there is an enduring opposition between true individualism, fundamental to liberalism as he saw it, and rationalism, with the latter associated with socialism and illiberalism. By rationalism, Hayek meant to refer to what I have been emphasizing throughout: the hubristic view that with the aid of reason, human beings can reassess and plan a social order, subjecting it to the control of individual human reason, rather than relying on free markets, spontaneous orders, evolution, traditions, and the workings of the invisible hand as described by Scottish Enlightenment thinkers such as Adam Smith, a founding individualist.Footnote ²²

Rationalists, in Hayek’s account, end up as collectivists. They foolishly think that human beings can effectively design rules and institutions, that “fatal conceit” that “always tends to develop into the opposite of individualism, namely, socialism or collectivism” (Hayek Reference Hayek and Peart[1951], 2015).Footnote ²³ The fatal conceit leads to illiberalism and to an obliteration of free markets. More fundamentally, it is based on what Hayek saw as the defining error of the rationalists. Mill was a liberal, and a friend of freedom, but in some respects, Hayek thought, Mill joined their ranks.

As we have seen, Hayek did not believe in complete laissez-faire. He favored a guaranteed minimum for the poor (“some minimum of food, shelter, and clothing, sufficient to preserve health and the capacity to work, can be assured to everybody” ([1944], Reference Hayek2007, p. 148)) and even a comprehensive system of social insurance. His own conception of liberalism allowed all that. Nonetheless, he insisted that rationalism is both arrogant and dangerous, assuming that human beings know what they cannot know, and he believed that Taylor moved Mill in its direction.

To be sure, Hayek was complicated here. He celebrated traditions as undesigned orders, and he defended traditional morals. Still, he did not think that traditions should be frozen in amber. He was open to changes and improvements. Nor did he think that traditional morality was a legitimate basis for coercion. Recall that with respect to the force of law, he took an even stronger pro-liberty line than Mill did, seeing the Harm Principle as insufficiently constraining. It would be fair to wonder whether Hayek fully sorted out his enthusiasm for traditions, and for traditional morals, alongside his opposition to coercion. Perhaps the best way to do that is to say that traditional morals should be celebrated in the civil sphere, but should not be enforced by the state. That is a coherent view, and it does seem to be Hayek’s own. He wrote, “Perhaps one of the most important characteristics that distinguish a free from an unfree society is indeed that, in matters of conduct that do not directly affect the protected sphere of others, the rules which are in fact observed by most are of a voluntary character and not enforced by coercion” (Hayek, 1960, p. 213).

At the same time, the prohibition on coercion “does not necessarily mean that in a free society such conduct should also be exempt from the pressure of opinion or disapproval” (id.). Here Hayek distinguished his own view from Mill’s, noting that Mill “directed his heaviest attack against such ‘moral coercion’” (p. 214). In Hayek’s view, Mill “probably overstated the case for liberty.” This is a remarkable claim, for Hayek was not wont to say that other people, or that anyone, overstated that case. Still, Hayek was right to separate himself from Mill here. Mill was keenly aware that traditions can be oppressive and stupid, and that they can stifle individuality, even if they are not enforced by law. Hayek did not have a lot to say about that. I think it is fair to say that where Hayek gave pride of place to undesigned orders, Mill put freedom at the center.

It is fair to wonder whether Hayek saw, clearly enough, that traditions may be a product of conformity pressures or cascade effects, leading to terrible things (Ullmann-Margalit, Reference Sunstein and Reisch1997; Sunstein, Reference Sugden2020). Many liberals follow Mill. They might respect traditions, but they insist on critical assessment of them, and they believe that traditions can be unjust or foolish. Perhaps some people have not participated enough or at all, in their creation, and not sufficiently in their perpetuation. (Consider Mill’s view on sex equality. Or not just people; consider animal welfare, which greatly concerned Mill, and which did not seem to concern Hayek much.) Perhaps traditions were created by some kind of accident, and then got entrenched. Perhaps they are a product of some kind of cascade, informational or reputational (Kuran & Sunstein, Reference Kuran and Sunstein1999). There are, after all, multiple equilibria. If we converged on one practice, well, maybe it is an accident (a Hayekian point), and maybe we should not be so wedded to it.

In a brilliant essay, Edna Ullmann-Margalit, evidently an admirer of Hayek, nonetheless urges against him (Ullmann-Margalit, Reference Sunstein and Reisch1997),

She adds that we “need to subject each existing social institution to a critical examination, free of presuppositions, in order to see whether it was indeed of such positive social function and of such pedigree as to make it worthy of respect and preservation.” That is a rationalist claim, and Ullmann-Margalit urges that Hayek was wrong not to be open to it.

In my view, Hayek was obtuse in his conclusions about Taylor’s influence on Mill. There is a real irony here. The Apostle of Ignorance studied Mill and Taylor for a very long time, and he remained ignorant of something essential to his whole project. He did not get Mill or Taylor quite right (Mill & Mill [1859], 2026).

To be sure, Mill cared deeply about social justice, and he came to embrace what he described as a form of socialism, above all because of the unfairness of “the present economic order of society” (Mill, 1879). But his complex writing on that topic should hardly be seen as an endorsement of centralized government planning. Mill did not quite know what Hayek knew, or quite see what Hayek saw, and Hayek’s lifelong obsessions were not Mill’s cup of tea. If he were to time travel and read Hayek, I think he would nod with admiration and appreciation – but he would also have been a bit annoyed.

Where Taylor most influenced Mill was on topics that were not the subject of Hayek’s main focus. Hayek missed all this. Mill’s The Subjection of Women was profoundly affected by Taylor’s views as expressed in her 1851 essay, The Enfranchisement of Women.Footnote ²⁴ Taylor sketched many of Mill’s central arguments, and others that were more radical still, including an explanation of why married women should work outside the home (Mill [1851], Reference Mill1868, p. 13): “A woman who contributes materially to the support of the family, cannot be treated in the same contemptuously tyrannical manner as one who, however she may toil as a domestic drudge, is a dependent on the man for subsistence.”

On Liberty, now thought to be written by both Mill and Taylor (Mill & Mill Reference Mill, Mill, Turner, Jacobs, McCabe, Osburg, Shefczyk and Schmidt-Petri[1859], 2026), is widely taken to be an argument for limited government, and so it is. It is also an attack on conformity and tradition, and this is indeed a topic that separated Hayek and Mill, and where Taylor really did have an influence. More precisely, Mill and Taylor were collaborators. It is crucial to recall here that in contending that people may be restrained only to prevent “harm to others,” Mill was, and Mill and Taylor were, speaking of the effects of social norms and conventions, not merely of government (Reference Mill, Mill, Turner, Jacobs, McCabe, Osburg, Shefczyk and Schmidt-Petri1859, p. 17). Much of Mill’s attack was on the oppressive quality of public opinion.

Thus Mill writes, “the tyranny of the majority was at first, and is still vulgarly, held in dread, chiefly as operating through the acts of the public authorities.” But if society “issues wrong mandates instead of right, or any mandates at all in things with which it ought not to meddle, it practices a social tyranny more formidable than many kinds of political oppression, since, though not usually upheld by such extreme penalties, it leaves fewer means of escape, penetrating much more deeply into the details of life, and enslaving the soul itself.” In Mill’s view, “there needs protection” not only against the state but “also against the tyranny of the prevailing opinion and feelings; against the tendency of society to impose, by other means than civil penalties, its own ideas and practices as rules of conduct on those who dissent from them.” As we have seen, Hayek did not endorse these points; recall his striking view that Mill “probably overstated the case for liberty.”

Taylor herself had made similar arguments more than two decades earlier, and it is hard to mistake the connections among her youthful views, Mill’s own painful experiences in the 1830s, and their, and his, passionate arguments against the tyranny of custom. Far from valorizing traditions, Mill’s particular case for liberty emphasized the immense importance of allowing experiments of living. In his view, “the worth of different modes of life should be proved practically, when any one thinks fit to try them. It is desirable, in short, that in things which do not primarily concern others, individuality should assert itself” (Mill, Reference Mill1859b, p. 105). That idea is the beating heart of Mill’s conception of liberalism.

Here we can find the sharpest of the divergences between two of the great figures in the liberal tradition. Enthusiastic about individualism, Hayek also prized traditions and customs. He saw them with a kind of reverence, as a product of dispersed knowledge. Here, by contrast, are Mill’s words (Reference Mill1859b, p. 105): “Where, not the person’s own character, but the traditions or customs of other people are the rule of conduct, there is wanting one of the principal ingredients of human happiness, and quite the chief ingredient of individual and social progress.”

This is a timeless claim, to be sure, and it is certainly not Hayekian. It is a form of rationalism (a point for Hayek’s characterization of Mill here). It is also intensely autobiographical. Mill and Taylor embarked for many years on a kind of “experiment of living” that was designed to promote their own happiness despite being roundly condemned by “the traditions or customs of other people.” But their individuality asserted itself. The worth of their different modes of life was proved practically. There are reasons to think that Mill’s arguments against traditions have a lot of force, and that Hayek’s praise of invisible hand mechanisms was a bit sloppy (Ullmann-Margalit, Reference Sunstein and Reisch1997).

All this leaves the mystery with which Hayek started: Who was Harriet Taylor? Hayek’s own verdict was clear. She “was an unusual person. But the picture Mill has given us of her is throughout determined by his own character and tells us probably more of him than of her” (Hayek Reference Hayek and Peart[1951], 2015, p. 312). To Hayek, Mill was in the grip of a delusion. Thus Hayek’s conclusion (p. 311): “Behind the hard shell of complete self-control and strictly rational behavior there was [in Mill] a core of a very soft and almost feminine sensitivity, a craving for a strong person on which he could lean, and on whom he could concentrate all his affection and admiration.”

Though fascinated by the scandal and the romance, Hayek rendered a cold verdict on Taylor herself. Perhaps he was right. But no one can fail to be moved by Mill’s competing verdict, his inscription on her grave: “She was the sole earthly delight of those who had the happiness to belong to her …. Were there but a few hearts and intellects like hers this earth would already become the hoped-for heaven.” We can bow humbly in Mill’s direction, and ultimately accept his objections to the pressures of conformity (Sunstein, Reference Sunstein2019a), without denigrating Hayek’s arguments in favor of traditions as undesigned orders that often (not always!) serve us well.

Epilogue

Hayek’s work is like an unimaginably large house, full of rooms, large and small. You enter one of them, the living room downstairs, and you learn about constructivism and how wrong and dangerous it is. You leave and enter another, right next to it, and you learn about the varieties of liberalism. Then you exit and walk down the corridor, and there’s a new room, a brighter one, different from the rest, and you learn something about the human mind and how it creates categories. You decide to go upstairs, and there’s a small room immediately on the left, and you enter, and now you get a new understanding of the rule of law. In the massive attic, you find out, in the first room you enter, what is wrong with progressive taxation.

Right next to it, with large, comfortable chairs, is a room focused on individualism. In a big room in the large basement, you learn something new about majority rule. In the den on the first floor, there’s Hayek himself, telling you, with warmth and precision, about Adam Smith, Bernard Mandeville, Ludwig von Mises, Joseph Schumpeter, John Stuart Mill, John Maynard Keynes, his cousin Ludwig Wittgenstein – and oh, about capitalism too.

In every one of the countless rooms, Hayek will tell you something about human ignorance, at least if you stay there long enough. Initially, you might find that topic a bit discouraging, verging on dispiriting, but in Hayek’s hands, it is exhilarating, even thrilling. How and when does innovation work? Why are free markets such a marvel? Why is today so unfathomably different from yesterday? How is competition a discovery procedure? What’s wrong with experts? Why don’t we know what our world will be like a decade from now? Why do planners get it wrong, even if their motivations are good? Why is freedom so important? Why should we celebrate the rule of law?

Hayek’s answers are enduring. Everyone should know them. On the socialist calculation debate, Hayek was right, and his views bear on many calculation debates, including the AI calculation debate. To be sure, behavioral economists have taught us many things that Hayek did not and could not know (even if he did know more than you might suspect, in part from reading Adam Smith, that early behavioral economist). But it would be possible to accept much of what Hayek did know, including about human ignorance, and to use behavioral findings in a way that takes Hayek on board. It is tempting to dismiss traditions. In my view, traditions deserve to be scrutinized. (Hurray for Mill; hurray for a dose of rationalism.) But Hayek’s work on undesigned orders suggests the immense importance of approaching longstanding practices with humility and respect.

“It is high time,” Hayek wrote, “that we take our ignorance more seriously.” Hayek showed that doing that is a kind of liberation for thinkers of all varieties, and even more importantly for the human spirit.