In this episode, Shruti speaks with Peter J. Boettke about the writings of F.A. Hayek, why artificial intelligence will not solve the knowledge problem, what many economists throughout history misunderstood about the market process, mainline vs. mainstream economics and much more. Boettke is a Distinguished University Professor of Economics and Philosophy at George Mason University, the BB&T Professor for the Study of Capitalism and the director of the F.A. Hayek Program for Advanced Study in Philosophy, Politics, and Economics at the Mercatus Center at George Mason University. He has written dozens of books, including “The Battle of Ideas: Economics and the Struggle for a Better World,” “The Economic Way of Thinking,” “Living Economics: Yesterday, Today and Tomorrow” and, most recently, “F.A. Hayek: Economics, Political Economy and Social Philosophy.”
SHRUTI RAJAGOPALAN: Welcome to Ideas of India, where we examine academic ideas that can propel India forward. My name is Shruti Rajagopalan, and I am a senior research fellow at the Mercatus Center at George Mason University.
Today my guest is Peter J. Boettke, who is a University Professor of Economics and Philosophy at George Mason University, the director of the F.A. Hayek Program for Advanced Study in Philosophy, Politics and Economics, and the BB&T Professor for the Study of Capitalism at the Mercatus Center. He is the author of dozens of books, including his most recent publication, titled “F.A. Hayek: Economics, Political Economy and Social Philosophy.” He has also co-authored and authored hundreds of papers. Personally, he has been an enormous intellectual influence as a teacher, mentor and my dissertation adviser.
We spoke about why artificial intelligence models will not solve the knowledge problem in economics, the feasibility of technosocialism, the long history of scholars understanding and misunderstanding the market process, the Vienna-Virginia tradition of economics, mainline versus mainstream economists and much more.
For a full transcript of this conversation, including helpful links of all the references mentioned, click the link in the show notes or visit mercatus.org/podcasts.
Hi, Pete, welcome to the show. It’s such a pleasure to see you.
PETER J. BOETTKE: I’m thrilled to be here with you, Shruti. This is a great opportunity.
RAJAGOPALAN: I want to start with Hayek, because I have learned so much of Hayek from you. I took the Austrian sequence with you; you were my dissertation adviser. When I have a Hayek question, it just feels about right to come to you. Recently, Daron Acemoglu said on Twitter that the insights in “The Use of Knowledge in Society,” which is Hayek’s 1945 paper, is essentially a computational critique of planning. Acemoglu argues that Hayek argued that the market system efficiently aggregates information via the price system, and that Hayek posits that central planning wouldn’t work because it would be impossible to collect and compute the right allocation of resources.
After paraphrasing Hayek in this way, Acemoglu moves on to pose this general question that, in this current world of AI and large language models and the incredible computational power that humans have acquired, what if central planning can be improved tremendously? Would Hayek then be happy with central planning? This is the premise that he started with.
I actually sent him the paper on Twitter—not just me, like 30 of us I think inundated Acemoglu with your recent paper with Rosolino Candela, “On the Feasibility of Technosocialism.” This was published recently in JEBO [the Journal of Economic Behavior and Organization]. You basically argue that there is a fundamental misunderstanding of the economic problem of society as being one of a computational nature rather than a knowledge problem, and this needs to be addressed. The market process actually does address this problem, and that’s Hayek’s main insight.
Before I get into the big debate, I wanted to frame it. But I want you to explain to most of our listeners, who may not know the main insights of Hayek’s 1937 paper—this is “Economics and Knowledge”—and Hayek’s 1945 paper, “The Use of Knowledge in Society,” and then we can go on to the rest of it.
BOETTKE: Sure. That’s great. Great context too.
‘Economics and Knowledge’
BOETTKE: I think the key insight in Hayek’s 1937 paper, “Economics and Knowledge,” is that the pure logic of choice is a necessary but not sufficient condition to explain the market order. In order to explain the market order, we have to also add in the various different institutional environments within which individuals interact with one another and which they interact with nature, and to see how it is they learn in these alternative environments, how best to coordinate their economic activity with others.
I think it’s a great thing to start with the ’37 paper because that’s where Hayek really starts to articulate his particular understanding of what it is that prices do in an economy. In my own book on Hayek, I try to tell the evolution of Hayekian ideas by an arc of his career. I start with his own beginning, which was on the imputation problem, but carry from that imputation problem, basically, how is it that I value a hog? How the value of a hog is derived from the value of the bacon that consumers use, and that spreads back through the structure of production such that farmers want to maintain the hog, and all those things like that.
That’s how you get the coordination of economic activities through time, and what is the role of prices in the coordination problem because prices are these guides to future action. What is the role of profit-and-loss accounting in being able to lure us into ventures and discipline us from the sad adventures, if we make mistakes or whatever, and that selection process of what’s going on. Hayek’s early part of his career, which includes not only his business cycle theory, which is important to stress—his business cycle theory is a relative price story of the business cycle. It’s a microfoundational story of the business cycle. It’s a price-theoretical rendering of the business cycle.
Then his critique of socialism, which he’s building on Mises, is again a price-theoretical explanation. I think if you look at his essay “A Trend of Economic Thinking” in 1931, you could infer from that that his argument is as follows: If you knew economics, you would know socialism doesn’t work because all good trained economists understand the role of property, prices and profit and loss. That’s 1931. At the same time, he’s very successful in communicating originally his business cycle to the young graduate students and great excitement to the community at the time in the 1930s and his move from Vienna to LSE.
By the mid-1930s, there starts to be rumbling that people are saying, “Oh, I can use the very same neoclassical tools to prove that I could plan the economy.” I think Hayek starts thinking to himself after the “Collectivist Economic Planning” book, which is 1935, “What’s going on here?” These are brilliant people like Abba Lerner, his student, and he’s like, “What’s going on? Why is this happening?” He comes up in the second arc of his career, which is a twofold move, I argue.
One is, is that he thinks there’s some deep philosophical mistake that economists have made, which misconstrues the nature of their science. Then there’s also, because they misconstrued their science, there’s a loss of the institutional foundations. If you go all the way back to Hume, Hume’s argument was that society, to get off the ground, needed to have stability of possession, transference by consent and the keeping of promises. You needed to have property, contract and consent, which is an institutional foundation.
The Importance of Institutions
BOETTKE: There’s a brilliant book by Lionel Robbins that was published in 1952, but originally written in 1939 before the war, but they had paper shortages and everything like that, so it couldn’t come out. It’s on the theory of economic policy among the classical economists. What Robbins argues in that is that the economist of the classical school—Smith, Hume, all the way up to Mill—they never argued about economics in a vacuum. Economics was always embedded within a set of institutions. What he postulates is that economic science grew up alongside of the evolution of liberal institutions, or property, contract and the rule of law.
They’re embedded in one another. Many of us, like you yourself, who’s an expert in property rights economics and everything like that, you’ll know that Alchian begins by saying, let’s begin the analysis in exchange, in production. Or in university economics, we’re going to begin in a world where we have clearly defined and enforced property rights. That’s the basis of exchange. To a large extent, when you treat something as given, it’s easy to then treat it as forgotten. Hayek’s like, “How could you forget property, contract and consent? We need to bring that back into economics.”
In the 1940s, what he does, starting in the 1930s and into the 1940s, he does a dual move, which is one in the philosophy of science, and then the other one is into the restatement of the liberal principles of political economy and justice. That defines his career. In my reading, Hayek never left economics. My view is he’s the very first new institutional economist, which is he’s saying, “Hey, wait a minute, guys, time out. We can’t do economics as if it’s in a vacuum. To try to do economics with omniscient agents and benevolent agents is the same as trying to do economics without scarcity.” If you said to me, what’s the economics of heaven? I would say there is no economics of heaven because there’s no scarcity, therefore there’s no tradeoffs. When we postulate economic actors that are omniscient and benevolent, we’re doing the same thing. Hayek is saying, “No, we can’t do that.”
Now, back to your question. 1937 is his first foray into this. Now, his teacher, mentor, “teacher” in quotes, but mentor, Mises, who is largely responsible for his career—what Mises did was he divided the methodology into theory and into empirics. As a result, there’s a tough understanding, I would argue, about what it is that the market is, because is the market the same as the derivation of the pure logic of choice, or is the market something different? Hayek reaches back intellectually, I would argue, to the original masters of Austrian economics, which are both Böhm-Bawerk and, more importantly, Menger.
We have to remember, in both Böhm-Bawerk and Menger, they make a three-part distinction: pure theory, which is the rationation of the individual chooser. Then what they do is they develop the situational logic of how markets and how different rules that govern the market will dictate the different ways in which the outcomes come about. And then empirical analysis, historical analysis. There’s this really interesting area, which Menger calls it empirical realistic theory, or Böhm-Bawerk just calls it applied theory of price, which is going to be this institutionally contingent theory.
Depending on the various institutional arrangements, we’re going to have different manifestations of the logic of choice. The logic of choice, the pure rationation, diminishing returns and diminishing marginal utility and whatnot, that’s still there. I can get a demand curve to slope downward, but I can’t necessarily get the interaction to the market unless I fill it with this institutional context of property, contract and consent. Most of the time I just assume those things. So therefore, that’s where the laws of supply and demand work, and there’s no intervention, and I get a notion of an unhampered market economy.
Hayek was saying, “Look, we need to look at how those institutions impact the way in which actors in the economy learn.” That’s the key thing. What he’s arguing is that the pure logic of choices is necessary but not sufficient. This causes some Austrian types to have agita because they think he’s deviating from Mises. Chris Coyne and I wrote a paper about this thing years ago. But the ’37 paper, pure logic of choice, the situational logic of various different institutional contexts. You combine the pure logic of choice with the situational analysis of institutions, and that forms your framework from which you then do empirical analysis.
Institutional Variation and Learning Characteristics
BOETTKE: That’s Hayek’s big ’37 piece. What his argument is, is that our justification of how markets operate is in applied theory, not in the pure theory. That means that we have to have institutions. Now, what’s the big thing for Hayek? The big issue for Hayek is that what makes institutional variation important is not the incentives that it structures. That’s incentive institutionalism. Again, Hayek does not deny that, but that’s not his point. His point is that different institutions have different learning characteristics to them. One way to think about it is a classroom, and certain classrooms are more conducive to learning and other classrooms are less conducive to learning.
What Hayek was trying to say, in essence, is that an unhampered market economy, one built on property, contract and consent, is actually a really, really effective learning mechanism for economic actors. They are going to be hit with stark signals. It’s not only that they respond quickly, but they respond correctly to the adaptation. I have a postcard that Hayek signed to somebody—to Pascal Salin, actually, is who he sent it to. It says the following thing: “What makes the market process important is not that it adapts to change, but that it adapts to the constant need to readapt to change, because we’re in this world of ceaseless change.”
Now, we’re talking about economic activity through time, not any static state of affairs. We’re not trying to look at the characteristic of a price that would generate an equilibrium. That’s the idea of prices as a sufficient statistic. We are looking at prices as guides to our future activity. Now, it’s true that prices economize. He does make that point, but what they economize on is the amount of knowledge that we do to plan for our future, not the sufficient statistic, tell me what an equilibrium condition is. If I stand at standard price theory—price equals marginal cost—it tells me that the full opportunity costs are taken into account, I’m going to produce at the minimum point on the average cost curve. I’ve got all these conditions going in.
To Hayek, those optimality conditions are by-products of the market, not assumptioned into the market. In that ’37 piece, one of the things that he leverages is an argument by Morgenstern on the paradox of perfect foresight. His argument is that you can’t have a market process theory in which the agents have perfect knowledge because if they had perfect knowledge, they would in fact be trapped in an impasse. Morgenstern uses the Holmes-Moriarty example. Holmes knows Moriarty is going to be on a train to Dover, but Moriarty knows that Holmes knows, and then you’re stuck in this impasse.
Hayek’s point is not so much how do I solve the paradox of perfect foresight. It’s that the paradox of perfect foresight is going to leave me unable to act, and so in the world of action, we are never having perfect knowledge. The perfect knowledge outcome is the by-product when all of the process has, in fact, ceased because it’s completed its job.
Just to relate again back to broader people outside of Hayek—and I recommend to all of your listeners here to go online and look at the interview in the Hayek tapes with Armen Alchian. It’s brilliant. I’m going to invoke Armen Alchian. When Armen Alchian published the selection of his papers with Liberty Fund, the title of that collection was called “Economic Forces at Work.” What I love about that is, it’s not economic forces after they’ve worked, which is what we learn in textbooks. When we learn what the equilibrium conditions are, those are the outcome after all the mechanisms of work.
That’s in fact what equilibrium means. We’ve in fact worked through the process, and we’ve settled in. To use another example, Machlup used to make a joke which he says, “Why do we draw a supply and demand curve if all we cared about was the P and Q vector that clears the market?”
RAJAGOPALAN: The kid at the back of the class can see. They can’t see it if it’s just a point on the blackboard.
BOETTKE: It’s just a point. They can’t see it. The kid in the back has to see it. But what we really want to explain is all the exchange behavior that goes on prior to that, which, of course, as you know as a student of Buchanan as well, that Buchanan’s big issue is that economics is about exchange and the institutions within which exchange takes place. All of the new institutional economics that comes about in 1950 to 1980—property rights economics, law and economics, public choice economics, entrepreneurial economics—all of that stuff derives, in many ways, from that ’37 paper, because the ’37 paper laid the space open for institutional analysis, but institutional analysis that doesn’t deny the marginalist principles.
Prior to the ’37 paper, to say “I was an institutionalist” would be to say that there is no logic of choice because as the institutions change, the people change and therefore the laws of economics change. I’ll use one last analogy in the hope of clarifying this position, and it comes from Mill, actually, but it’s really discussed quite well in several papers by Dan Hausman, the economic methodologist, the philosopher. He uses the example of a harbor and the law of tides. The law of tides is a physical law built on the idea of gravitational pull as the earth rotates around and what it does to the ocean levels and everything like that. Those are exact laws. We know what those laws tell us.
But when we build a harbor, what we do is we actually intervene into the nature, and we structure various different things with the way we build the harbor. The way we build that harbor is going to affect the way in which the law of the tides manifests itself. That’s like what’s going on in the market because the institutional rules under which the market operates are not going to violate the idea that individuals are striving to do the best that they can given their circumstances. What it does is it changes what it means for them to strive.
What Hayek is saying is that not only is it their incentive to strive, but it also is how do they know what the right thing to do is to strive. That is dependent on these signals. That paper, the ’37 paper, gives rise to contextual economics, the idea that economics always takes place in an applied sense, in a context.
‘The Use of Knowledge in Society’
BOETTKE: Now we get to the ’45 paper, and he stresses that the data of the market—and this is going to get to your Acemoglu point—the data of the market is the data of the context. That’s why the way he describes context is that it’s the knowledge of time and place of the actor. Not anyone can have it. The problem in the market isn’t treating given data and then processing it computationally. The problem is that you have to have the context; otherwise, that data doesn’t exist.
In the 1945 paper, “The Use of Knowledge in Society,” Hayek is trying to explain to his audience—as he says, he wants to use the term “marvel” to shock them out of their complacency. I think it’s important to remember that this was in the American Economic Review, not in Time Magazine. He’s not talking to the average person.
He’s arguing that professional economists had become complacent about the functioning of the price system. He wants to explain the functional significance of property, prices and profit and loss for the coordination of economic activity through time. He focuses on the power of relative prices to guide our decisions. The idea that the individual is the man on the spot that has to actually act on those prices, interpret those prices and then behave in accordance to those prices. It’s that idea of the price system as a whole, and the telecommunication aspect of that, that is crucial to his insights.
In the 1945 paper, the reason why he calls it “The Use of Knowledge in Society” as opposed to “The Use of Information in Society,” is that it’s not just little bits and pieces of knowledge. It’s knowledge that is not, in fact, given to us. It is generated through the very process of exchange and production activity that gives rise to the knowledge that we utilize. It’s only knowledge within that context. Outside of that context, it doesn’t exist. It has to actually be context dependent.
That’s a crucial aspect of his story. The socialist calculation debate was initiated in 1920 by Ludwig von Mises, and it was directed at people who believed that the experience of World War I planning could then be used in peacetime to be able to then organize the economy in disruptive times and to improve a better world. They wanted to rationalize the economy. They called it the rationalization of production.
Their goal was to rationalize production so that they would deliver mankind from the kingdom of necessity, a world of scarcity, to the kingdom of freedom, into a burst of productivity in which they would be able to actually achieve all the wonders of the market without any of the disruptions of the anarchy of production, so to speak. Ludwig von Mises raised the challenge that in their effort to achieve that, their goal was to abolish private property in the means of production. And Mises argued, once they abolished the means of production, they would, in fact, eliminate the very means by which they could rationalize production.
That’s why it’s called the socialist calculation debate, because they were worried about rational economic calculation. What rational economic calculation is, is the ability to sort, from the array of technologically feasible projects, those which are economically viable. An example is, if I’m choosing a railroad track, should I build that railroad track out of platinum or out of steel? You might say right off the bat, “Oh, platinum’s too expensive.” I’ll ask you, “Why?” You’ll say, “Oh because the market price of platinum is too high.” That’s the point. Without private property, we wouldn’t know that market price of platinum.
We would normally say, “Well, why don’t we pick platinum because it’s a technologically superior metal to steel in producing the railroad?” The problem is that that would end up by producing less with more. We would be wasting resources because platinum could be used in alternative uses in much more effective ways. Without that guiding role of the price system, you’re not going to be able to know that outcome. And so, as a result, you’re going to end up by having systemic waste, and therefore, the least thing that you’ll do is have a rationalized production process. Instead, you’ll be trapped and snared in an irrational process that in fact makes us poorer rather than delivers us from our sad state.
Hayek and Mises did mention the incentive problem. Both of them go out of their way to say socialist managers will not face the same incentives. That’s the oldest argument in intellectual history, goes all the way back to Aristotle’s critique of Plato. They also do argue that there is a computational problem, which Plato identified, which is that in a world of 700 goods and 100 people or whatever, you would still have to solve a system of simultaneous equations.
Barone makes this argument as well. It’s fascinating in both Barone and Pareto, they argue that you cannot answer that question, it’s not computable, but the market does it every day. People forget this because they think that they somehow proved in advance that Mises was wrong, but they don’t. Smith actually says it in the famous passage about the common woolen coat. He says the number of exchanges would be beyond our ability to compute, to be able for us to get the common woolen coat. The way I think of what people are doing there is just trying to explain the amazing complexity that’s involved in producing the simplest products.
Here I’m standing at an iPhone. Just think about all the component parts all throughout the world, all the knowledge that’s embedded in that, and how you gather and put that all together. That problem is not the significant problem. The significant problem is the generation of the knowledge that’s in the context that tells people that I should do this kind of chip rather than that kind of chip. I only get that if I’m in that context. If I’m not in that context, that never exists. It’s not that the knowledge is difficult to obtain; it doesn’t exist. It’s not even there. That’s what Hayek is trying to get at in “The Use of Knowledge in Society.”
But along the way, while he’s doing it, he uses these various other things: prices economizing the amount of information we use. Prices are these summary devices. They are an incentive wrapped in a signal. All those kinds of arguments are in there as well, and so it depends on how people read it. The reason why someone as brilliant as Acemoglu will focus on that issue is because it’s that issue that the mechanism design people picked up on. Again, there’s an interlock here. They took—Herwitz and all of them took Hayek’s challenge seriously, and they said, “I have to prove that I can actually generate a process which will be informationally efficient.”
That’s all this computable stuff. Then you get to Lange, who in ’67 says, “What’s the problem? I now have a supercomputer.” Keep in mind that Lange is saying that we already have Kantorovich developing linear programming. We have various people who have tried to do this, even Leontief’s input-output tables. A lot of people in the planning field have already been trying to answer Hayek’s challenge.
Coordination vs. Computation
RAJAGOPALAN: Before we get to the computational question, to me it seems like there are two, three different aspects to this. The first part of Hayek’s insight in ’37 and ’45 which is lost is the institutional context that you’ve pointed out. The second part which I think has been lost is subjectivism. People will have a subjective interpretation of the facts before them, and they will have subjective expectations that they form about the future, even if we impose this bizarre uniform incentive structure.
These days it’s a lot in vogue to say that, “Oh, everyone is individual, everyone is diverse. We need to treat each person exactly like the special snowflake that they are.” It’s odd that Hayek did that way back 100 years ago, and he’s like, “Yes, each person is special because they have very specific knowledge, but that knowledge comes from their subjective view of the world.” This of course is the Mengerian tradition. That’s one part to me that seems to be lost.
The other part is this conflation between the coordinating aspects of the price system and the computational aspect of the price system. What you were telling me is that the price system, given certain institutions—private property, consent, exchange, contracts—can create a system where all these different bits of subjective dispersed knowledge can actually coordinate. And the price that pops up has managed to calculate all of that information the way we would in an economic model and spit out an objective number.
In equilibrium, that objective number that we can stick on things and all those subjective expectations about marginal costs, all of that lines up perfectly. That’s computational, and Hayek’s argument is, the price system is very low cost and efficient at computing in a way that we have no other human or technological institution that can compute.
The other half of it is, we actually don’t have any mechanism to coordinate either, and the computing and the coordinating elements are fundamentally different. The coordinating element comes from what you’re talking about, which is the learning, which needs to happen in an institutional environment and so on. Can we separate computation from the epistemic aspects? Is that the source of the confusion in Acemoglu, or is something else going on here?
BOETTKE: It’s a brilliant set of questions that you ask there, and I realize that I gave a very long-winded discussion of ’37 and ’45. I do want to address some of the things, but I want to read a quote from Hayek. It’s Hayek talking about the reception of Mises’ ideas by other economists. He says, “Mises’ arguments were not easily apprehended. Sometimes personal contact and discussion were required to understand them fully. Though written in a deceptively simple prose, they tacitly presuppose an understanding of economic processes, an understanding not shared by all his readers. We see this most clearly in his crucial argument on the impossibility of economic calculation under socialism.”
This is the key point. “When one reads Mises’ opponents one gains the impression that they do not really see why such calculation was necessary. They treat the problem of economic calculation as if it was merely a technique to make the managers of the socialist plant accountable for the resources entrusted to them and wholly unconnected with the problem of what they should produce and how. Any set of magic figures appear to them sufficient to control the honesty of those self-indispensable survivors of the capitalist system. They seem never to comprehend that it is not a question of playing with some set of figures, but one of establishing the only indicators that managers could have for deciding the role of their activities in the whole structure of mutually adjusted activities.”
It’s, again, this generative and discovery aspect of it. This is the key issue, is that to Hayek and to Mises, what calculation does is it allows us to sort from the array of technologically feasible production projects those that are economically viable. And you have to do that sorting. Otherwise you’re going to have systemic error, and you’ll end up not producing more with less, but less with more.
That’s the very definition of irrational. Again, you have to remember, they weren’t the ones who said the goal was rationality. The socialist goal was rationalization of production because the whole idea was that by rationalizing production, they would lead to a burst of productivity, which would deliver mankind from the kingdom of necessity to the kingdom of freedom. Unless they can get rationalization of production off the ground, the socialist project crashes to the ground. The search in all of this stuff from Marx to Neurath to all the way up to Lange and Lerner, but then also Kontorovich and all the rest of them, Leontief, is to get rationalized production.
If I can show you that you don’t have the generative process to be able to do that, then your project is going to crash and burn. That’s why it’s called the decisive objection, that they have to forgo the intellectual division of labor.
Subjectivism and the Market Process
BOETTKE: Let me go back to your other point, which I think is really important because this relates to Hayek’s issue about the tacit understanding of the process. What is the market made up of? Keynes had an idea of this; he called it the Keynesian beauty contest. He understood this idea that the market isn’t so much like what I think is the most beautiful. I’m trying to guess what I think other people think are the most beautiful and whatnot, and so the market is made up of judgments of others’ judgments.
That’s how the phrase Hayek uses there fit into the larger structure of the idea here. I’m trying to figure that out. In order to do that, I have to engage in some judgment, but I also have to have a tool of appraisement. I realize that whether or not I’m engaged in that activity the right way—and this all, of course, turns on the subjective costs that I face in my decision-making.
One of the key things, going back to your pluralism idea—and there’s a great exchange of letters between Jim Buchanan and Karen Vaughn that’s in the Buchanan Archives about, why didn’t the debate just get stopped once Arrow wrote his famous paper? Because if I can’t have a stable social welfare function, the planner doesn’t tangentially kiss the point of the social welfare function on the outer frontier of the production possibility frontier.
Buchanan’s whole point in that 1949 paper is there’s no fisc. Even in ’49 he’s arguing that—let alone in ’54 when he really goes after Arrow—there is no stable social welfare function. What collective action is all about is exchange and, as he says, we must always return to the swamp. He’s talking about mosquito abatement issue as an applied problem of collective action, but what he means by that is we have to be in this issue of the exchange process that makes up collective action. If there’s no stable social welfare function, there’s no single scaler of values. There is no residual claimant over the entire economy.
Part of the confusion, I think, in the debate, and maybe we’ll talk about this—because you were the one who turned me on to the quote from Jack Ma that I used to start the beginning of the feasibility paper. You’re the one who turned me on to that. He makes the argument that, “Look in my business, we use all this fancy computer knowledge and algorithms to help me make my decisions. I have a big business, and we’re doing all of that. And so, why can’t the world do that? And we can get rid of this messy process of the market, and we could just have this.”
I think that the issue is that there’s a residual claimant in his business. There’s a residual claimant for Amazon. There’s a residual claimant for Walmart. There’s not for the U.S. economy, let alone the global economy. Instead, we have this diversity of ends that are being pursued, and the only thing we can agree upon is the means by which we interact with one another. That allows for this greater scope of this. And that gets us back to the adaptation and agility of market mechanisms or whatever mechanism.
I would say the following thing, if we agree that you have to sort from the desirable to the feasible to the viable. Whatever social system we’re in, we have to go from desirable—“Oh, this is what I dream to be the dream”—to “That is actually a feasible dream,” to “That’s actually a viable dream.” What we know is that under markets where you have property, prices and profit and loss, we have built-in mechanisms that are going to prod us, are going to lure us, are going to guide us and are going to discipline us. We have constant feedback and constant movements they’re going through.
That functional significance of property, prices and profit and loss has to be mimicked by something else. If we get rid of private property, we’re going to have to have some other mechanism that does the same thing of prodding us, of guiding us, of luring us and disciplining us. Now the question is, what is that? Is that voting? Now we’re in comparative institutional analysis. Is it voting? Is it beauty? Is it some special—
RAJAGOPALAN: Status, caste, whatever.
BOETTKE: Yes. In history of mankind, we’ve had all kinds of different ways to do this—some of which are very ugly, actually, when we see them done, but we’ve done it. What they haven’t been able to do is produce advanced material production on a scale. They don’t rationalize production, we end up by producing less with more, and we waste resources and things like that, as opposed to the efficiency that’s generated by the market mechanism. I think that we should learn from that lesson and have a very important conversation about that.
Is There a Calculator?
RAJAGOPALAN: Do you think they went horribly wrong because they called it the impossibility of socialist calculation as opposed to something like the impossibility of being able to substitute the market process? Because now the entire focus is on calculation, which makes it seem like there is a calculator. They’ve, in some sense, personified the market process or what we get at the end of it as if there is a person who is sitting with a dinky little calculator making all those calculations. The main problem that Lange and Lerner had was that Leontief’s mainframe computer wasn’t large enough.
Do you think this is just, people didn’t read the entire thing because it was all old-school history of thought? What stuck was information economics and impossibility of calculation, and from there we have completely derailed because we have lost the epistemic part and divorced it from the computation part?
BOETTKE: Yes. I think there’s a couple of things that are going on. Again, your insight is very valuable, I think. One of Mises’ articles is not translated into English until 1935. Robbins and Hayek are actually more communicating the debate once it’s out of its German context. In the German context, though, from Neurath through Karl Polanyi and other people, and even if you go back before Mises to Weber, when Weber talks about socialism in one of his famous essays from 1919 or whatever. Or Mises in his book on war socialism, which is also 1919, “Nation, State and Economy,” where he talks about this issue.
It was other people who said there was a calculation problem that needed to be solved, but it could be solved by mechanisms other than the market. Again, Mises is adopting a challenge. Now, part of that challenge is in the old Marxists, was that you could use labor units. This addresses your subjectivism point, that you could use labor units to do it. Or Neurath was saying, “Look, we did it in the war economy. Why can’t we just do what we did in the war economy to do this?” The problem with that is that in war, you turn the economic problem into a technological problem because you have one end. And then how do you do that in a world where you have multiplicity of ends?
I want to read Milton Friedman. This is a very forgotten essay, but it’s brilliant. It’s Friedman’s review of Lerner from 1947. The reason why I’m picking it out is, you’ve got to put the context. Again, 1947 Samuelsonian economics, which is just about to start to emerge, but the economics that Lange and Lerner and all of them were doing was an effort to have an institutionally antiseptic theory. They wanted to purge institutions from economics, and so how do I bring them back in?
Again, this property, contracting, consent idea. He starts talking here; he’s quoting Lerner, and he’s talking about the incentive problem. Lerner basically says, “The incentive problem is not an economic problem. It’s a psychological-sociological problem. Don’t bring it up. When we want to talk about economics, we just talk about the efficiency properties of the mechanisms that we’re in.”
He says—this is Friedman, I’m quoting Friedman—“But this is only a part and probably the least difficult part”—that is, the incentive issue—“of the problem as an example of competitive entrepreneurs indicate. The manager’s intentions must not only be good, he must be able to translate his intentions into practice. The higher administrators who themselves need both incentives and tests of performance must have means of determining the extent to which the manager has been successful in his attempts to follow the rule.”
He needs that profit-loss accounting to tell him whether or not what they’ve been doing is a good way to do it or a bad way to do it. That’s what the calculation thing is. It’s a sorting mechanism, again, and his argument is, that’s the crucial thing. That’s Friedman, who many years later we would just say, “Oh, it’s about private property and incentives,” then give an example of collective farms versus private plots in China or in Russia, whatever, and say, “Look at the yield differences.”
Of course, all economists understand that idea, but the issue is what’s going on in the way this industry is organized. You need to have that sorting mechanism so that you know that you’re not wasting resources because we live in a world of scarcity. If we didn’t live in a world of scarcity, we might not have to worry about that stuff, but we do live in a world of scarcity. So we have to always be worried about how do we produce more with less rather than trying to produce less with more.
Now, I have a paper that just came out, and it’s “Monitoring, Metering and Menger.” Because Menger himself actually makes a similar argument that hasn’t really been brought out that much about what’s going on in enterprises or whatnot. I’m reading stuff into it, but he has the monitoring issue—that’s the incentive issue—but he also has the metering issue. We have to be able to have that. I think when you read Alchian and Demsetz or any of the modern theory of the firm with both the concept of metering and monitoring in mind, you start to see the reason why what Hayek and Mises are doing, because it’s about, how do I not only monitor, but how do I meter? And that’s what Friedman’s talking about there.
Then once you get to metering, then that gets to judgment, that gets to appraisement, that gets to all the subjectivism stuff. This is why the Austrians are a little bit more attuned to those kinds of questions because they start already attuned to those kinds of questions.
RAJAGOPALAN: Speaking of Menger, I think the other debate which is relevant here is going back to the marginalists and the difference between Menger and Walras. I think some of the present-day conflation between the computational aspect and the epistemic aspect we can actually trace back to Menger and Walras. Because the MIT economist—and I’m just using that as a shorthand for all these computation arguments—they’re accustomed to Walrasian models. What we’re really saying there is that there are no false trades taking place. It’s a mathematically modeled world of N number of people with X goods and Y services.
All of this happens through an imaginary auctioneer who will be announcing prices. People are adjusting. We know that there are some specific conditions where price will equal marginal cost and output will be at the level that minimizes average cost and all the things that you said. We know that the output—after the auctioneer has announced everything and everyone has adjusted to everything, we’re going to get the output that maximizes production efficiency and exchange efficiency, and so on.
This is quite different from the Mengerian tradition, which is much more focused on subjectivism and process. To me, it seems like the problem is not that they don’t know the socialist calculation debate; that’s almost irrelevant at this point. It’s that they are in the Walras tradition and how that developed, and not the Mengerian tradition where the entire focus is on the process. The outcome that you get is almost beside the point because you know it’s going to change in the next instant. There’s going to be a new set of prices that everyone has to adapt to. There’s really no point studying that equilibrium point that has so beautifully come about because you know it’s going to change anyway.
BOETTKE: There’s a lot there. The first thing is that it was an important discovery of Pareto and the early neoclassical economists to point out that if a socialist economy was to be rational, it would have to mimic the optimality conditions of a general competitive equilibrium. Think about it like math is math. The relationship between marginals and averages is the same, whether or not you’re talking about the cost at a firm, or you’re talking about the test scores of your students, or whether or not you’re talking about in India or in China or whatever. The relationship doesn’t change.
That was the idea of having a universal theory was all tied into that. That drove a lot because, again, what’s the logic behind that? Well, here’s the logic: Economics has been debating for centuries over issues. People thought, man, we’re spinning our wheels. Why are we spinning our wheels? Well, we’re spinning our wheels because of ambiguity. The biggest ambiguities result from using the same words to mean different things or different words to mean the same thing.
You know what? We’re going to get rid of all that, and we’re going to get math. Math is math. Then we’re going to get rid of the ambiguities, then we’ll be able to have a science that’s worthy of the same idea. I kid around, as you know that what happened with Marshall when he was trying to formalize economics as a discipline—
RAJAGOPALAN: He said, “Burn the math,” right?
BOETTKE: Yes. Well, he called it economics. Everyone before that was political economy. Political economy, that means you end up with all the other disciplines that end up in Y, which is like history, philosophy, sociology. If you end in an X, you’re with the cool crowd. You’re with physics, you’re with mechanics.
One of my teachers was the great Kenneth Boulding. I asked Boulding—when I was a student I said, “Professor Boulding, why is it that everyone became a logical positivist?” He laughed and he said, “Oh, my dear Peter.” He goes, “Who would want to be an illogical negativist?” His point was a very good one about capturing words and where you belong on things.
I think using “economics” was important because it put us as a hard science, which was an aspiration. We’re not soft science like these other ones. We’re not squishy over there. We’re a hard science. What makes us a hard science eventually is going to be that we can follow the rules, just like in physics, and we can develop these hypotheses generated out of our formal models, and then we can test them.
I think one of the things that was a really big methodological issue—this is why Hayek goes off on the methodological idea—was that for most of the history of economics, what we cared about was having a logically sound theory. Which means that if your premises are true and your deductions are correct, then your conclusions are in fact true. This is how you would have a truth-preserving idea, but that’s really tough stuff. What happened with math is you don’t have truth-preserving. What you have is internal coherence of a validity. You have validity, not the soundness.
What we did was, we traded off the difficult stress for soundness and embraced validity as our goal. And this is why it took a while for it to take off and why it correlates with the rise of mathematics, because we’ve been trying to do math for a while. But it really takes off once we get statistical tools because the belief was, now we could have an array of logically valid models, but we’re going to sort among them by finding out which ones are empirically meaningful by having these very good statistical tests.
The problem, of course, is that even in physics, the Duhem-Quine thesis says that even these statistical tests are not going to unambiguously tell us which of the theories are right. Then we get stuck with a variety of these toy economy models, which seems so divorced from the world that we’re trying to study. That’s a long story, and it’s more complicated than that. That’s connected up to why it is that you could get this kind of model that you’re talking about.
I think the really important point that you made about Walras versus Menger is that, by definition, the approach of the general equilibrium and partial equilibrium approach, which builds on a system of simultaneous equations as solving of a system of simultaneous equations, whether or not it’s comparative statics or whether or not it’s a general equilibrium idea, it relies on not articulating the process.
First, system of simultaneous equations, you have pre-reconciliation of plans, and then you have a P and Q vector that clears the market. You know what that result is. As you said, you get simultaneously exchange efficiency, production efficiency and product mix efficiency. You get those things simultaneously achieved. In comparative statics, what you do is you have an equilibrium, you have an exogenous shock and you have a new equilibrium.
You don’t have any kind of process by which it goes there. When you’re telling the story in your class, you might talk about buying low and selling high. Or when you talk about Marshall’s fish market, you talk about this fish, and as the fish start to get old, the price goes down; you get rid of them faster. That’s your appreciative theory that you’re telling, but it’s not in the actual formula that’s there. The formula is either a system of clearing a vector or a change and then another vector. One vector that’s at equilibrium, a shock and then another vector like that, all of which preclude process analysis.
This has huge implications. One of the things about the Austrians is, their understanding of the market is one in which you begin as a monopolist. What happens is your monopoly power gets eroded by the entrance of other traders. You’re the first mover in a market. I’m the first person to come up with bottled water. Then for a while there I get monopoly rents for the bottled water. Then the next person says, “Huh. They’re making pretty good money.” They get lured by the profit opportunity. Then they come in. Then they erode that market share. The market becomes competitive.
It doesn’t ever start as competitive. It becomes more and more competitive. The whole thing about the Austrians—and not just the Austrians, but Buchanan and everything—is that the order is defined in the process of its own emergence. As that is happening, there’s all kinds of new generation of knowledge, there’s new evolution of product quality, characteristics, all kinds of various things that make for our life as consumers. It’s constantly shifting and changing, as you said.
This is where the system of simultaneous equations, which is a very determinate system, it doesn’t allow us to talk about these other kinds of things. There’s a very logical reason why they got rid of entrepreneurship. For years, entrepreneurship was one of the key things: land, labor, capital and entrepreneurship. All classical economists talked about that.
RAJAGOPALAN: Then we called it the residual claim, like the thing that’s left over that we can’t exactly put into our—
BOETTKE: We get rid of it. Schumpeter has to bring it back. Baumol and Kirzner have to bring it back. It’s not there. Baumol even refers to it as what economics was—it’s like “Hamlet” without the prince because it’s the entrepreneur that’s driving all of the things and the innovation machine. But with losing the entrepreneur, we also lose innovation except as an exogenous shock. That’s why all of the evolution of post-1950 economics is a rediscovery, in a lot of ways, of a lot of the earliest—
RAJAGOPALAN: What was lost.
BOETTKE: —what was lost in economics. I think that relates to why it is that we get sucked into these different puzzles.
Perfect Competition Without Markets
RAJAGOPALAN: I think it’s also got to do with the idea of perfect competition. If we look at stuff, say, late 19th century, early 20th century, it’s a competitive process. They will say dynamic market process or something like that. This idea of competition was always there, or contestability.
Perfect competition, if I’m not wrong, is Frank Knight’s innovation, right? I think he uses the phrase, but he uses the phrase to pedagogically show something and sneaks in all these assumptions that, if all these assumptions were true, then you would get what we have and what we call perfect competition, right? Now with the whole Marshall demand and supply curve, and how analytically easy and tractable and it’s got such great explanatory value in the classroom, you add these two things together, and then it becomes some kind of normative ideal that we want that.
That’s the thing the planners want. That’s the thing that the FTC wants. Now the entire question is, how do we get to that as if an individual can calculate up to that point, without all the things that are required to either have some kind of competitive process which mimics that or some kind of contestability process which will mimic that? We’ve erred on both sides, which is, if we don’t have exactly that, it’s not proper market competition. If we don’t have anything else, and we just have that, that’s enough to get us our simultaneous equations and calculate everything else that we need, right?
BOETTKE: I mean, I think it’s very important to keep in mind a point that you just made, which is that what Lange and Lerner did—I mentioned earlier about the formal similarity issue. What happens is they confused the formal similarity issue with a solution. If we could demonstrate that as a formal similarity, then why not just make the socialist regime or the collectivist economy—that’s what they normally called it—why don’t we just make it do exactly the same rules that would be in a competitive market economy? Except we don’t use the competitive market, we use the planning bureaucracy to do it.
You’re right about the role of perfect competition and all of that. I think it’s important to keep in mind that thinkers like Knight, like Mises and Hayek, like Ronald Coase, what they did was they were working with a foil concept of equilibrium to identify the world that isn’t the world that they’re in. Because the world out the window is too complex. They need to frame the world of change by isolating what a world without change would look like. In a world without change, in equilibrium, you have no firms. You have actually really no markets. You actually end up by having no law or anything like that. The institutions don’t matter.
What happens—for example, Hayek in the debate with Keynes says, “Look, you want to explain unemployment, you got to begin in a world of full employment before—identify what a world full of employment looks like.” Now tell me what gave rise to unemployment. Now I know a cause for unemployment. I can’t start with a world of unemployment and then say, “Oh, now fix it,” which is what the Keynesian idea did. That’s the logic why Hayek argues that Keynes is dealing with a world of abundance because there’s no tradeoffs in this. It’s just a technical solution.
Knight wanted to understand the phenomenon of profit. He comes out of studying theology. He’s very philosophical, and he’s very fascinated by the idea of profit. Can we justify profits? The way in which he’s trying to examine that risk, uncertainty and profit is, he wants to start with what would a world look like in which there was no profit at all. His argument is that rates of return are not due to risk because risk, that would be manageable, so that’s just another kind of return. What it is, is the bearing for uncertainty, right?
RAJAGOPALAN: Yes. We don’t know what we don’t know.
BOETTKE: Right, and that’s what he wants to get to. The joke is always, the Chicago guys read the first couple chapters but didn’t read the other chapters. That’s just facetious or whatever, but I think that one of the things that’s crucial about Coase’s zero-transaction-costs world—but he never wanted to study the zero-transaction-costs world. He wants to study—in fact, in his book summarizing his points of view, which is before his Nobel Prize, it’s called “The Firm, the Market, and the Law,” he goes through and explains it. I really recommend all your listeners to read that book because he also identifies why he got the idea from the socialist calculation debate, which is him, again, thinking about this issue of metering and monitoring, or what he would call transaction costs later on.
What you have is a fascinating evolution of modern economics, which is you have these early masters of neoclassical economics using a foil model, but at the same time, you have people using it as a benchmark—what you just said. On the other hand, you have people using it as an “as if,” so it’s like, “Oh, it’s as if.” And this issue of the Friedman in ’47—the difference between that Friedman and the evolution of Friedman or Stigler is that by the time you get to their more mature statements, what they’re trying to stress is not much the generative process of the market, discovering things, though they are not ignorant of that.
One of my favorite stories is Stigler in his “Memoirs of an Unregulated Economist”—people should look this up—there’s a story where Koopmans comes to him in an outrage because he says, “George, I heard that you argued that we should use the price system in the case of evacuating New York City in the case of a bombing.” Stigler writes back to him and says, “Well, Tjalling, you need not worry. No, I did not write that kind of memo, but on second thought—” and then he goes and explains why the market system would be more adaptable and adjustable and all these things like that to the moment on the time of the bombing, which is very brilliant. It’s just great. It’s Stigler to a tee because he’s snarky—
RAJAGOPALAN: You know, it’s happening today. When there is a disaster situation, the Uber price surge just goes through the roof. And you do see the market mechanism in some way allocating scarce resources for people who want to get the hell out of there. That’s what makes people mad.
BOETTKE: I get that. That’s a normative thing.
RAJAGOPALAN: Exactly. That’s a normative thing.
The Middle Ground
BOETTKE: Yes, you have that, but what happened in economics is the debate forgot the foil, and it just became a debate between the benchmark and the “as if.” This debate has huge implications for us because it affects the whole Econ 101 debate. Then you have, on the outside, the foil people, which also include someone like Buchanan who is sitting there saying in 2009, 2010, “Hey, there’s an old Chicago tradition which actually cared about institutions. What are you talking about?” He contrasts the old Chicago with the new Chicago, and then he has this whole thing about the emperor has no clothes, which people dismissed as an old man being cranky.
The reality is, those papers are brilliant that he is identifying because he’s talking about this Knightian tradition of where your economic system is embedded. I’ll give you an example of this. It’s in the economic organization in which Knight actually uses the argument that the theorems of economics are like the skeleton. And the institutions are like the muscles and the ligaments and everything like that. A skeleton, just a pure skeleton, can’t actually make your arm move. You need to have all the muscles to be able to do it. On the other hand, if you had just the muscles and no skeleton, your arm would—
I’m somewhat silly in my teaching, and there’s a scene in one of the early Harry Potters where Harry’s learning to play quidditch, and he falls off of his stick or whatever.
BOETTKE: Broom, and breaks his arm. There’s this weird wizard, and the wizard says, “I can fix your arm,” and then he takes all the bones out of Harry’s arm. To me, that’s what happens in economics because we don’t have theory. We don’t have theory; we just have data analysis, which is just this blobby flesh. On the other hand, if we just had the skeleton, we lose all the interesting details about what makes the system function.
This is the way Kirzner uses the analogy. I know I’m mixing up my analogies, but Kirzner uses the analogy of, it’s the friction between the soles of my feet and the ground that allows me to walk. If there was no friction, I’d slide and fall all over the place. And it’s the friction in the economy, the ligaments and the muscles, that make economic life come alive for us, as opposed to just the skeleton. But we need that skeleton.
Going back to Hayek ’37, the skeleton is a pure logic of choice; the muscles and ligaments are the institutional analysis; the outcome is, how do I take the two of them together to then explain the functioning of the body economic? Knight had that all in him. A lot of people just focus on the bones, and they miss this other part, and I think that’s because they confuse the foil. They just focus on the “as if.” They compare the “as if” to the “that.”
Think about our economics profession. You have Bob Lucas, who just passed away, brilliant economist. And then you have Joe Stiglitz, and both of them are what you would call mainstream economists. They taught at the best universities, they won the Nobel Prize, and they’re super cited and everything like that. Here they are, they’re contrasted with one another. Stiglitz is all about the benchmark, and it fails. The market constantly fails. He’s never met a market that didn’t fail. And Lucas never met a market that wasn’t perfect because he doesn’t actually have to have it be perfect. It’s an “as if,” right?
To go even further, if you go to Becker’s foundational book on “The Economic Approach of Human Behavior,” nobody says—he says, “These are assumptions: that you have maximizing behavior, stable preferences, and basically competitive equilibrium.” Price has to equal marginal costs. It’s not that he discovers price equal to marginal costs. It’s in a condition I impose on the world. He makes sense of the senseless by imposing that order on the world.
It’s brilliant, but Stiglitz does the same thing by imposing, “Okay, you say that, but I’m going to say that individuals suffer weakness of will, and they’re also going to suffer from monopoly structures.” Lo and behold, the price system doesn’t work, and so it fails on every occasion. This is the debate we have. Market never works; market always works.
I know that people don’t like this. Kirzner has made a living at doing this, but he always says, “Look, I’m in the middle ground. I’m in between these two things. The real world is constantly filled with change and whatnot, which means that there’s surpluses and unexploited opportunities all the time that need to be discovered and exploited, which means I’m in market failure at some level—
RAJAGOPALAN: Or disequilibrium at the very least.
BOETTKE: Disequilibrium, but I’m in the process of becoming.” On the other hand, the market doesn’t never have a problem. The market has a solution. As opposed to saying it has no solution, government has to be a deus ex machina, or the market never has a problem, therefore we just rely on the market. Instead, there’s this middle ground, which is the institutional and entrepreneurial innovations.
BOETTKE: If I could just say one thing about that, which I think is interesting for development economics, for all kinds of different things like that and the big debate—there’s a great paper by Deirdre McCloskey on the 126 market failures. Do you know this paper I’m talking about?
BOETTKE: McCloskey goes through all these different things and identifies all the market failures that people have identified in textbooks, which I find always amazing because when you hear the neoliberal Econ 101 criticism, they say, “Oh, market fundamentalism is taught.” I’m like, “I don’t know that any market fundamentalists are teaching at Harvard or whatever.” What they do is they teach, basically, supply and demand, and then the next lecture is monopoly, externalities, public goods, macroeconomic instability, inequality. They spend the rest of the semester on government as a corrective to the social ills that they’ve identified.
McCloskey has this whole list of these market failures that she’s identified. What I think is really weird about that is, okay, so what’s the response? The as-if response, which is what the majority of very skilled market economists do, they follow more of the Chicago approach. To me, I think there’s something a little bit wrong with that because it’s like a Jedi mind trick. There are no externalities. There is no monopoly problem.
Now, they do it sophisticatedly sometimes. For example, in the old structure-conduct-performance paradigm, the Chicago new learning was like, “Hey, what I’m going to do is, I’m not going to look at automobiles produced just in the United States by just United States auto producers. I’m going to look at the global market, and if I define the market broadly enough, no one’s a monopolist. If I define the market narrowly enough, everyone’s a monopolist. I’m going to define the market very broadly and then I have the broad market, therefore your monopoly criticisms aren’t true.”
Also identifying various costs—one of my favorite arguments that Alchian and others made was, “Oh, show me a normal monopoly diagram.” Then you say, “Why can’t the monopolist perfectly price discriminate?” Your answer is, “The monopolist can’t perfectly price discriminate because of transaction costs.” Then you go, “Transaction costs, that’s an additional cost to engage in an additional exchange.” Another word for that is called marginal cost. That marginal cost curve that you have that’s horizontal, we’re drawing it wrong. We have to let it go tick, tick, tick, tick, tick, and then all of a sudden, “Oh, price equals marginal costs given those contexts.” There’s something about that which is cheeky to us as economists, but it’s like a Jedi mind trick to the Stiglitzes of the world.
Stiglitz writes in his book “Whither Socialism?,” “Chicago economics was well funded but not well founded,” because he doesn’t want to have the Jedi mind trick. To me, I think, what a better answer if we rely on Coase and Buchanan and Kirzner. Because what we’re going to have is first the possibility of different changes in the institutions, which will in fact be Pareto improvements. We can do them in changes of the law, which is, in the Coasian framework, what you do is you allow people to negotiate. If they can’t negotiate away the conflict, you adjudicate to clarify the rights so they can negotiate. The priority is to get them to negotiate away the conflicts. The legal changes are in fact to clarify the situation so they can negotiate away their conflicts.
In Buchanan’s idea, it’s you propose structural rule changes so that you can realize Pareto improvements. Now that has to be tested against the collective action decision-making, so your hypothesis as an economist is not taken as godlike or whatever. It’s tested in the collective process of democratic decision-making. Nevertheless, you can get Pareto improvements by changes in the rules. Elinor Ostrom: changes in the rules, but from the ground up. Buchanan, more from the top down. Her, more from the bottom up. Again institutional. There’s an institutional answer to these very hairy social problems.
On the other hand, you have Kirzner and Baumol and Schumpeter and other ones who sit there and say, “Today’s inefficiency is tomorrow’s profit opportunity for the person who can solve that problem.” They’ll jump in to try to arbitrage that way or come up with a new technological innovation to be able to achieve that. How robust is that answer to the market? I’m saying it to you, and you’re nodding your head because you buy this. It dumbfounds me that this is not the standard economics response.
Economics as the Golden Key
BOETTKE: During the pandemic, I read Matt Ridley’s book, “How Innovation Works,” and he has a line in there that I love. Of course, Deirdre’s been saying the same argument, but Ridley summarizes it up and says, “Innovation is the child of freedom and the parent of prosperity.” I look at that and I’m like, “Okay, that obviously seems true. Why is it that not everyone just says, ‘There we go’?” What we need is to have a permissionless economy that allows for constant entrepreneurial innovation to be able to achieve these greater outcomes than any of us ever could imagine for them to discover. To me, I think that’s like, “Why aren’t we teaching that in economics?”
Part of the difficulty—this is outside of this debate—but the broader difficulty of the critique of economics, “the economists’ hour is up” and all these other things like that, is that I think the critics have misidentified what it is that the sin in economics is. The sin in economics isn’t market fundamentalism because there are no market fundamentalists. The sin in economics is that we teach it as if it was social engineering. We forgot that economics at its best is a tool for the curious. It’s to excite the minds of the young people, not by giving them a catechism or telling them what to think, but to tell them how to actually frame questions when they look out the window, to allow them to have observational genius about making sense out of the senseless that’s all around them. Making a mystery out of the mundane.
This is what Adam Smith’s doing with the common woolen coat. Common woolen coat, what’s so mysterious about that? Then Smith says, “Look at that thing. Where the hell did it come from?” And then starts talking. 200 years later, Leonard Read says, “I, Pencil.” It’s the same kind of thing, right? We could say today, “Oh, it’s an iPhone.” How do we get that? Once we do, those kinds of things make a mystery of the mundane or allow us to make sense out of the senseless behavior that we see out in the world. Then, all of a sudden, we can see that economics is this tool for the curious to help them reason through it. It’s a golden key. We shouldn’t hoard it.
We should be really excited as teachers to share the golden key with people to recognize that we live in a world of scarcity. That scarcity implies tradeoffs. Tradeoffs mean that we must negotiate those tradeoffs. In order to negotiate those tradeoffs, we need aids to the human mind. The commercial society gives us those aids in the terms of property, prices and profit and loss.
If we’re not in a commercial society, give me another aid to the human mind. It turns out when we do the comparative institutions of those different institutions, they don’t do as well. As a result, we get stuck making tradeoffs that are not going to be corrected and redirected whenever we make mistakes. We get—mistakes are embedded into the system and privileges are embedded into the system.
Like you mentioned earlier, one of the ways in which we did answer these questions in the past was caste. What does caste do? Caste actually locks the world in. When we think about those kinds of ideas, it makes us really think about the dynamic nature of markets. And why don’t we teach markets that way—as exchange, the institutions within which exchange takes place and the entrepreneurial activity that drives economic progress?
Feasibility of Technosocialism
RAJAGOPALAN: When you were talking about the “as if” in Hayek’s 1945 paper, that’s an important part of the argument. Knowledge is dispersed, and it’s incomplete, and no one has this perfect information or perfect knowledge that we’re talking about.
BOETTKE: The tin example.
RAJAGOPALAN: Yes. People are able to behave as if they knew everything just by following the price signal because the price incidentally has managed to assimilate all the dispersed information and knowledge out there. It’s managed to accommodate everyone’s subjective choices and produce a lovely signal that can tell you something about the world and make you act. It can also incentivize you to act as if you have that knowledge.
Now, I want to come back to the socialist calculation debate because I think it can inform this whole AI question because what Oskar Lange and Abba Lerner and Fred Taylor and all these guys did, actually, they weren’t very socialist in the Marxist sense. They were very socialist in the marginalist sense. I would call them Walrasian socialists if that makes any kind of sense.
I don’t think anyone has used that weird phrase, but what I mean is they thought that, “Oh, there are these numbers out there. We have some kind of ranking that we can get in these markets outside. We can just borrow all of that, and then they can feed that into a computer, and that computer is going to tell us something.” What they forgot is that those numbers only tell us as if what was happening was actually happening and everyone knew what they were doing, but they don’t actually tell us that because you’ve removed that context. Can you walk me through what that debate was? Then we can talk about how that feeds into the computation question, which is in your technosocialism paper.
BOETTKE: Let me start by pointing to your first issue, is that the reason why Hayek was so animated in the discussion was because he thought that his liberal friends were making the same mistake that the German audience made a generation earlier. Now, the British, at the time that they were writing, the market socialists, their argument—this is like Dickinson, Durbin, Lerner—Lange counts, even though he is Polish, but he’s in London at the time that he’s doing this—they think that they are socialists in their economics because they’re liberal in their politics.
Going back to technosocialism, they now like to call this cybernetics socialism, and it never got implemented. It just was a plan that they were going to do. We’ll come back to that in a second. They thought that they needed to be collectivists in their economics precisely because they wanted to be liberal in their politics.
Part of the main thing of “The Road to Serfdom” was to show, “Time out, guys. You can’t be a collectivist and a democrat. Your goals of your collectivism will squash your democratic virtues, and that’s a tragedy.” Hayek’s dedication of that book to socialists of all parties is not a flippant comment. It’s a sincere comment. His whole point is the road to serfdom is a road to tragedy. You could have called it the tragic road of intellectual path. He even says in there, “What could be worse than people with high ideals end up by creating hell on earth by pursuing their high ideals?” That’s the first thing to keep in mind.
Now, the second thing is that they want to get rid of the monopoly problem, they want to address the business cycle problem and they want to address the social problem of inequality. Think about the Beveridge Report. The Beveridge Report has the five giants. I don’t know if you’ve ever seen the propaganda ad. It has government, which is like a tiny midget, and then it has the five giants. The government with the Beveridge Report is slashing the five giants. The five giants are poverty, ignorance, squalor, disease and idleness. They’re going to use these programs to slash this.
That’s 1942 that that program comes in. Between the 1930s and the 1940s, the Labour Party’s trying to push through this agenda to address—again, they’re in the midst of the Great Depression. They haven’t really gotten out of that. They’re just about to enter into a war. They’re very optimistic that they could actually do things the right way. Then that of course all comes into an issue. They knew they couldn’t go the full Marxist way. They knew that even going back—remember, the LSE is a Fabian school, and so they try to use marginalist principles, but they believed in socialist competition.
That’s like in the Edward Bellamy book “Looking Backward.” The argument was very simple. It’s like, “Look, if my firm, a socialist firm, only has the price equal to cost, not price equal to cost plus markup for profit, then I’m going to outcompete in the competitive marketplace. They’re going to go to the restaurant that sells the burger for three rather than for five. In the end, we’ll wake up one day and everyone will be in the future socialist world.” They had those ideas and they’ll address, and they thought distribution—again, think about all the issues that are involved—they think distribution is a question of politics, not a question of economics.
They’re like millions where the laws of production are different from the laws of distribution. Part of the whole thing about the marginalist revolution was to say, “No, the size of the pie is a function of how it is you slice up the pie,” so you can’t have production and distribution. The way Buchanan used to say it in a class was that basically, you can’t separate out the rules—you can’t just pick and choose whatever distributions you want. You have to actually pick rules of the game which engender a pattern of exchange, production and distribution. This lesson had to be learned. And these are all part of things Hayek is trying to get across in “The Road to Serfdom,” and then eventually in “The Constitution of Liberty,” as he addresses the move from hot socialism to cold socialism.
Now, the arguments at that time were using the greatest technology that they had available to them, which was a professionalized bureaucracy that could follow the rules that were given to them by the neoclassical model of market socialism. They didn’t have advanced computing yet like we know of. In fact, even if you look at NASA or whatever, the computers when we first were in this space program were individuals. That’s what they were called, the computing department. They were people with their slide rules doing all the math calculations.
The modern computing technology gets developed, and one of the things that it can do is it can produce given algorithms very, very quickly. That became a mode of optimism that, as long as you’re dealing with algorithms, can we just turn society into an algorithm and can we do that? If I treat data as given, it does seem to me that algorithms work. Now, Hayek is quite explicit that the data is never given, but they don’t read that part.
RAJAGOPALAN: That’s the “as if.”
BOETTKE: Yes. That’s not the part that they focus on. They focus on the idea that all I have to know is what the price of tin does, and then I can infer from that whether or not all the rest I can act appropriately on that. In itself, it’s a grand telecommunication system which at some sense is an algorithm.
When Hayek responds—which is actually before “The Use of Knowledge in Society” paper, in “The Competitive ‘Solution’”—he argues that, “Look, what everyone’s forgetting is that you have to discover anew each day what is the least-cost technology to employ. If you don’t discover it, the guy next to you is going to discover it and bring it in and compete you away.” It’s very contestable, as you were mentioning before, and all these things. It’s this generative process that’s getting completely lost.
Now, when I show up to graduate school, I’m excited about this debate. And Don Lavoie, my professor, was working on these issues, so I wanted to go study with Lavoie. I decided, through a various different process of search, I’d get into George Mason, and I decide I’m going to come here to do it. I end up by working for Lavoie. Do you know what the very first thing he gave me to read? It wasn’t Mises, it wasn’t a Hayek, it wasn’t Michael Polanyi, which eventually he does. It was Hubert Dreyfus’ “What Computers Can’t Do.” I think it’s important.
I have a new paper out on SSRN. It’s going to come out in the Independent Review. It’s an appreciation of Lavoie, and I hope people read it. The paper on SSRN is much larger than the one that will actually be published, but Lavoie was a professional computer programmer and actually a quite talented computer programmer. He eventually, at the back end of his career, went back to computer simulations and programming, but he gave us the Hubert Dreyfus book on “What Computers Can’t Do.” Then we also had to read the debate between Searle and other people, John Searle, the philosopher, on what hard AI could do and all these things like that.
I’m reading this stuff, and I’m like, “Why am I doing this when I’m supposed to be learning microeconomics and macroeconomics and everything?” He had a reason for that, which is, he wanted to understand what the nature of human knowledge is in the market. Because, again, go back to our earlier idea: If markets are about judgments of judgments and appraisement, these are very human characteristics that computers don’t have. It’s that the world isn’t in an algorithm. We have to actually decide between algorithms, or which algorithms we need to modify through combinatorial thinking. It goes back again to social learning. I think that’s the key thing.
Kasparov vs. Ronaldo
BOETTKE: Acemoglu, in the more recent Twitter thing, he has some papers on social learning. When you reframe it in terms of social learning, he’s like, “Oh, is that what Hayek’s arguing? If that’s what Hayek’s arguing, that’s pretty cool, but that’s not how I read Hayek.” That’s what he’s basically saying. “But I could see how you could read Hayek that way, and if that’s cool, then that’s cool. That’s a different kind of argument. Let’s talk about that.” I think he’s probably the “it factor” economist in the world today. It’s pretty cool that he’s actually thinking about these kind of issues.
I like to think in terms of this social epistemology distinction between what they call wicked learning environments and kind learning environments. In our technosocialism paper, we talk a little bit about this. In kind learning environments, the parameters are fixed, so it’s more analogous to chess. The game of chess is fixed. There’s very strict rules. A bishop can only move so many ways. A rook can only move so many ways. They’re very defined rules, and there’s only so many spots on the board. Therefore, what a computer can do is, it can process the algorithms and get through those pretty quickly.
But if you try to make a computer play a game in which the parameters are relatively free—going back to Lavoie, Lavoie was able to get a computer to play Bach, but he couldn’t get a computer to play Miles Davis. There’s a difference there—or John Coltrane. There’s a kind of a difference there. I think we can get computers to play Kasparov, but we can’t get him to be Ronaldo. The reason is that Ronaldo is constantly dealing with a spin of a ball and the way it comes to him in ways that require him constantly to be agile, adapt, readapt and constantly adjust. The question you have to ask is, is an entrepreneur more like Kasparov in seeking out a chess move, or is it more like Ronaldo trying to adjust and adapt?
You and I both love tennis. Roger Federer—one of the things that was amazing about him was his ability to come up with amazing shots in the most improbable things, especially in his prime. He just did amazing things. He reintroduced the squash shot into major tennis, that he would pull off. He was the one that did the between the legs before everyone else was doing. He mastered it.
I was just watching the match the other day with Alcaraz versus Tsitsipas, and for the first nine-tenths of that match, Alcaraz was on a whole other level. If you look at the shot-making that he was doing, it was off the charts. It was all him doing various different variations that required constant adaptation and adjustment to whatever this other guy threw him. The No. 5 player in the world playing the No. 1 player in the world, he’s about to take him out at one of the grand slams in less than three games that the guy had at the time. He was just blowing him out. You’re watching, and you’re in awe because of this adaptation and adjustment.
If I go to MIT’s Creative Lab and I watch their robots play soccer, it doesn’t look too good. If I watch them play, they can have a dog that jumps from an already known angle. Think about it. They have this dog that’s going to race. It’s going to race your dogs or whatever. They have it already known that it’s going to land on a surface that’s tilted this way or tilted that way. They program that in because they program in what they already know what the future is going to hold. Of course, the robot can adapt and adjust, but that’s the fixed algorithm.
Imagine if what was happening was like what Marshall puts out in the “Principles” book, which is that it’s not just a pendulum. Remember what he says about the market system. He goes, “You have a pendulum, and it swings.” He says, “But now imagine that that pendulum is in a turbulent sea of water. That’s what commercial life is like.” We don’t know where the pendulum is going to go because the turbulent seas are pulling it all the way.
The individual—that’s the judge on the spot, the man on the spot, Hayek’s man on the spot—is going to have to make those judgments and appraisements constantly to be able to actually move the market in a certain way. That’s not going to be fixed by better and better computers. That requires human intelligence. This goes back to the Searle point. In the Chinese language box, there’s a difference between syntactic knowledge and semantic knowledge, the meaning of terms and understanding everything like that. We as humans, we rely on all those kind of various things in our appraisements and our judgments, and our computers don’t have those kind of things.
Now, these tools can be extremely good tools. I’m trying to go in a course of self-improvement on myself right now. I have a tool on this thing, and I track everything that I do, and it follows me. It can do all the things, and I love it and everything like that. And I use it all the time, and I can try to organize things like that. But it’s still my judgment when I sit down at a restaurant to decide whether I’m going to have a piece of bread or stay gluten-free or whatever. I have to actually act on that judgment in those appraisements. I think that that human element of us is what makes our economic life so fascinating.
RAJAGOPALAN: It’s funny you say about the judgment thing. People typically think that I’m a snob because I won’t eat Hershey’s chocolate or something like that. What I constantly tell them is, it’s not worth being fat for. I’m not getting fat eating that chocolate. I’m going to get fat eating a far superior chocolate. I’m willing to do it, but this is not worth it and that is. And for some people, Hershey’s chocolate might do the trick in some other cases.
BOETTKE: It’s funny that you say that because now that I’m tracking all these things—one of the things that I say to my wife, Rosemary, all the time is like we’ll pick a food, and I’ll discover how many calories it’s worth. I’m like, “You know, that’s not worth that.” I make exactly your argument. I’m like, “That’s not worth it. That extra banana, that’s a lot. That’s 100 calories. Maybe I could have something else for that.”
To go back, there’s a really good paper. I hope I don’t butcher his name. He’s at the University of Pennsylvania. His name is Jesús Fernández-Villaverde. He has a great paper that’s on a public domain, it’s not like a journal article at the moment, on why AI doesn’t solve the knowledge problem. But he’s working on developing a paper, a more full-blown paper of these arguments.
RAJAGOPALAN: He has a paper on AI that he did for one of my edited volumes in the International Review of Law and Economics, but that had to do with the complexity of the world. This was like Richard Epstein’s “Simple Rules for a Complex World,” and how AI can’t solve the computation problem.
BOETTKE: He’s got his finger on the pulse on this. He’s a very successful economist. He can talk to these guys on their terms and stuff. I’m very hopeful that he’ll be able to communicate a lot of ideas to economists that—we should be paying attention to how he’s developing these arguments and learn from him on that. I’m very hopeful. Of course, I’m not dogmatic about it. Who knows? In the future we might have computers that become fully human or something, and we might have a singularity moment.
Language and the Calculation Problem
RAJAGOPALAN: I recently saw the Apple Vision Pro launch. One of the things that I learned was that, actually, it uses some subliminal messages to figure out how you’re going to move next or almost predict how you’re going to twitch and move and what you want to see. It’s figured out some part of this cognitive function in a predictive way, right? We also have Neuralinks. We have all this brain-mapping stuff.
What would it take to not just solve the computation problem, which is just a question of computing power and how much we throw at it, but this aspect of the tacit knowledge and the knowledge that is not institutionally antiseptic, and it’s subjective, and it comes from the individual and their judgment—but what would it take to be able to predict that, such that as if we could solve the calculation problem? What would that look like?
BOETTKE: Smith long ago ties a lot to language and our ability to engage in what communication is. And I think communication is a lot more than, say, information theory, which is signal-to-noise ratio, and there’s a sender and the receiver, and all that stuff like that. There’s a lot more subliminal things to this.
There’s a book—it’s a weird book, it’s by a guy named Dan Rose. He was an anthropologist at the University of Pennsylvania. He was a student of Erving Goffman. And it’s a Russell Sage volume; it’s called “Living the Ethnographic Life.” I met him when I was a fellow at Boston University at Peter Berger’s Institute for the Study of Economic Culture years and years ago. He was a fascinating guy.
I’ll tell you just a very quick story about him. He wrote a book which made his career. It’s called “Black American Street Life.” As a Goffman student, he went and he lived in south central Philly and tried to blend in and do his notes, and it became a best seller and everything like that. Then after that, he’s at Penn. What he decided to do was write a book.
He wrote that book in six years. The next 20 years of his career was in the process of writing a book about living the lifestyle on the Main Line. He was belonging to cricket clubs and things like that. When he first did the presentation, I was sitting there, and I was like, “Can you tell me how you got those research grants? Because I want to sign up for that deep ethnographic divide.” Anyway, he’s a fascinating guy. I love talking to him and I learned so much about him.
One of the things that was fascinating is, he pointed out the following thing. He goes, “When we do most anthropology, a kid goes to college and decides that they want to study, say, the Balinese cockfight. Then what they do is they start learning the language. They learn the language when they’re 19 years old, and then they study it when they’re 19, 20, 21, 22. They go to graduate school, they think they know the language, they do field work and things like that. Then they write their dissertation.”
He says, “How long have they known the language?” They’ve known the language and lived in that community for a total maybe of like six or seven years. He goes, “Would we rely on the reports of a seven-year-old to tell us about this community that they’re living in?” No, we wouldn’t. And that’s because so much of the human experience is tied up into the subtleties of language, and that’s the key issue, I think.
If we go back and we look at what Smith was arguing about in terms of language, or what McCloskey is arguing about in terms of language, and the way we communicate and the unspoken spoken, what Hayek calls they’re tacit presuppositions, these kinds of things. I think that we would have to somehow be able to render not useful but explicit these tacit understandings. If we could do that, then the judgments could be algorithms rather than these judgment calls that are made.
The other thing is that I do understand that when I use the phrase “combinatorial,” that it does suggest computation. But there is something weird about combinatorials that are unique to people.
RAJAGOPALAN: Context, right?
BOETTKE: Yes, context. Here’s an interesting story; it’s outside of economics. I heard it on NPR. I have no idea how robust this is, but it was Yo-Yo Ma. One of the things Yo-Yo Ma has decided to do since he’s very successful is, he gathers every year musicians from all over the world. He’ll find like the best bagpipe player ,and then he’ll find the best person like that, and they meet in his compound and they try to come up with music with all these various different instruments or whatever, who no one would normally put together.
When I was a graduate student, I read this fantastic article about what was found on Jacob Viner’s desk when he died. Jacob Viner died, and they had to clear out his desk. Jacob Viner had this habit that he had developed all over the years in which every Friday he went to the library. He didn’t do any other thing at the library, but he took notes on the journals. Now, the journals weren’t as big as they are now. But what he did was he’d start with Journal A, and then for the next quarter he’d be reading every Friday until he got to Journal Z. Then the quarter would be up, so then he’d be back at A again. And he took notes. He took notes on all the articles he found of interest, but he didn’t know that the articles existed until he picked out the journal.
Instead, what we do now when we have a search engine is we know, “Oh, Hayek wrote that paper in ’37. Where can I find it on JSTOR? Shruti Rajagopalan has this paper. How can I find that on JSTOR?” And then we’ve got it. It’s amazing. I do this, and then an article appears in front of me. But what I don’t do is, I don’t get all the other articles that were in the International Review of Law and Economics or in the JLS that are in the same issue as you. I get your article. I don’t get the serendipity of like, “Oh, I’m looking at Shruti’s article, but wow, look at that other article. That might be really cool. Let me look at that.”
I think part of the discovery of science is these motions of these a-ha serendipity moments where I sit there and I think, “That guy did that and that girl did that, and I’m going to take that and that and combine them, and I’m going to do this,” and that’s where you come up with a new idea. There’s no necessary algorithm that would ever link those two together. That’s this human judgment, this human appraisement of how I’m going to try it.
A large part of them fail. Going back to invoking Boulding again, one of the things Boulding used to say to us in class all the time is, economic science would make more progress if we publish people’s wastebaskets, because you try acts of creativity and they fail.
I think that the 20th century had a shift in the way we conceived of the scientist and economics and the discipline of economics. We used to do economics from the inside out, and then what happens is, we ended up by doing economics from the outside in. When you do economics from the inside out, you put priority on the clever and creativity of the actors in the economy that you’re studying. When you do economics from the outside in, you put the creative and cleverness of the theorist who’s trying to control the economy.
RAJAGOPALAN: Because everyone’s a representative agent, right? That’s what we do.
Mainline vs. Mainstream Economics
BOETTKE: Exactly. For example, think about the difference between Hayek’s knowledge problem and Lucas’ knowledge critique in the critique of Keynes. What Lucas argues in his critique of Keynes is that no agent in the economy doesn’t know what the economic theorist knows. Hayek’s argument is slightly different. Hayek’s argument is that no economic theorist can know what the agent in the economy knows because the agent in the economy has that knowledge of time and place which the theorist doesn’t have. They only have this general knowledge, what he calls the difference between knowledge in a system versus knowledge between systems. There’s this subtle distinction that matters. When we practice economics from the inside out—let me give you another example.
RAJAGOPALAN: I do think Lucas’ models are also doing the foil thing which we often miss.
BOETTKE: I agree with that.
RAJAGOPALAN: He’s got the Hayek critique embedded in it.
BOETTKE: I think Lucas is a genius; he transformed economics, which is an amazing thing to do. It was a huge mistake for the Austrian economists of that generation to not try to communicate with him and view him as a partner in getting a more micro foundations of macroeconomics.
RAJAGOPALAN: Can I ask you a weird question about Lucas?
RAJAGOPALAN: Why is he not in the mainline economics that you draw out between Adam Smith to Vernon Smith? You have a whole bunch of other people in there, and when I read it I didn’t think about that. All the people who are in it belong there; that makes perfect sense. What is it about Lucas that he doesn’t quite belong there?
BOETTKE: Friedman and Stigler aren’t in there either.
RAJAGOPALAN: I know, but Lucas is different from Friedman because Lucas is not a positivist like Friedman. That’s one point of difference that Lucas might belong in this list.
BOETTKE: I could be persuaded, but this is how I would put it, is that what I mean by mainline economics is those individuals that tried to square two basic facts of the world. The one fact is that individuals strive to do the best that they can given their situation. We call that the rational choice postulate. I would make a qualification and say rational choices if the choosers are human. They’re fallible but capable human choosers, not infallible choosers. They’re not maximizing in the sense that we know the full array.
The second one is Paris gets fed, which is the idea that there’s an invisible hand mechanism. That’s the invisible hand. We have the rational choice postulate and the invisible hand proposition. The question is, how do you square the two together? One way is to deny that they go together, and that tends to be the critics of the market: Marx, Veblen, Keynes. They go at it in various modern behavioral economics. They go at it in various different ways.
I actually would highlight Keynes’ essay “The End of Laissez-Faire,” because in that, he explicitly says that there is no invisible hand, basically. He argues there are individuals who suffer from weakness of will, so there’s no rational actor. The price system doesn’t coordinate economic activity. There is no force from above that guarantees that they could coordinate, and there’s no force from below. To Keynes, you’re stuck; that’s the whole point of an unemployment equilibrium. You’re stuck in a disequilibrium, and the only solution out of it is deus ex machina who can come in and fix from outside the system, which is why you have to be outside the system to come in. That’s one side.
The other side goes back to your as-if people, which is what they do is, they collapse the one onto the other. They collapse one onto the other. What I want to say is that all of the people—this goes back to the early part of our conversation—all of the individuals that are in what I call the mainline derive the invisible hand proposition from the rational postulate, but via institutional analysis. The institutional analysis becomes a critical part of whether or not you can see the invisible hand of the coordination of Paris getting fed. That means that you’re going to have that property, contract and consent.
It’s that focus on those individuals. All those individuals from Adam Smith and David Hume up through the French economists, like Bastiat and Say, into the late classical economists like Mill, into the early neoclassical economists to the Austrians. Then, in the post-1950 era, to the people like Buchanan, Coase, [Douglass] North, [Vernon] Smith and Elinor Ostrom. There are of course other people, but the reason why those names get names is because they’re Nobel Prize winners.
A part of that is sociological. I’ll explain this very, very quickly. It’s my own insecurities and weaknesses. When I got out of graduate school, I got the job of a lifetime. I got to work at New York University as an assistant professor. It was a dream come true for me. It’s where Ludwig von Mises taught. It’s where my undergraduate professor got his Ph.D. It’s where my graduate adviser got his Ph.D. I was going to go work with Israel Kirzner and Mario Rizzo and Ludwig Lachmann.
Both there and also at various functions that I would go to because I was at NYU—and I was an IHS baby, Institute for Humane Studies baby; they had been supporting me forever; I became their poster child along with David Schmidtz. David Schmidtz at the time was at Yale. They used to bring us out all the time to all their seminars and talk to all the students. I’d meet very bright students like yourself and others that are wanting to maybe go and get a degree.
Anyone that was in economics, they would sit there, and they would say to me, “You’re not really mainstream.” I’m really like, “I don’t know what that means.” I studied with Ken Boulding; he’s the second John Bates Clark medal winner. I studied with Jim Buchanan; he’s a Nobel Prize winner. I studied with Gordon Tullock; he actually is a distinguished fellow of the American Economic Association. “Who is it that you studied with?” “I studied with academic royalty in economics. What are we talking about here? Why are you making these comments?”
One of the things was, Boulding used to make the distinction between mainline economics and other economics. He has a great essay. It’s called, “After Samuelson, Who Needs Smith?” In it, he develops this notion of what he calls the extended present. His argument is that the evolutionary potential of Adam Smith’s science is still there. Adam Smith’s still part of our conversation, and that’s part of the mainline. I got this idea from him. I used to say, “No, maybe I’m not mainstream, but I’m mainline.”
The thought experiment that came to my head was, “If I was a martian and I came to the earth in 1780 and I said I was an economist, what would people think that economists believed? If I came to the earth in the middle of the 19th century, what would they think that economists believed?” I know the answer to that because all you have to do is read Charles Dickens and his discussion of the philosophical radicals in “Hard Times,” which is Mr. Gradgrind’s school, which is the utilitarians. If I go to 1890s and I listen to Thorstein Veblen criiticize economists, and he’s like, “Oh, the lightning calculator of pleasure and pains,” that’s what an economist believes.
What I started realizing is that if I’m a martian and I come, everyone’s going to say an economist believes that Paris gets fed and that individuals are pursuing their self-interest. What is it that made these things? The only time it ever stops being that is after Keynes. Once we get to Keynes, all of a sudden all hell breaks loose. In fact, if you were a martian in Keynes’ time, you still would’ve thought that’s what an economist believed because Keynes refers to himself as part of the rogues’ gallery in economics. That he’s part of the new economics, which is actually the old economics, which was the criticism of all this other economics.
To me, I was like, “Okay, those who sit in the seat of Adam Smith.” This is Buchanan’s line: “What should economists do? It’s those who sit in the seat of Adam Smith.” I was carrying out that idea. Again, the idea is that individuals are striving to do the best that they can given their circumstances. The by-product of all these individuals striving within a certain institutional context is going to give us this invisible hand of the coordination of the market. That united the economics of that.
That’s what I wanted to try to see, and then contrast it with people that either deny that or collapse it because when they collapse it, they forget the institutions that make it possible. I don’t think that we can have the coordination of economic activity without private property, the rule of law and things like that. That’s why I argue that Hayek’s move in that second and third arc of his career is the founding of new institutional economics.
Robert Lucas and the Mainline Tradition
RAJAGOPALAN: I think Lucas would fit that description. Lucas thinks about how Paris gets fed. Lucas thinks about rational choice very, very seriously. In fact, Lucas has the third element, a very Buchananist element, which is when there is some kind of, let’s say a given structure, an exogenous structure of someone announcing interest rates or someone announcing monetary policy, how does everyone around them adapt, whether it’s the right one or the wrong one? There is this process that has to take place for people to adapt to government success or government failure, in this case.
I feel like Lucas checks all those boxes. Lucas also comes from the seat of Adam Smith in the second arc of his career, which is as a growth economist and a development economist, because he’s very much directly studying those questions. When I first read “Living Economics,” I didn’t think about, “Oh, Lucas belongs here and he’s missing.” Lucas recently passed away. I was thinking about Lucas in a broader context than what we’ve learned. I thought, again, Lucas is very similar to Adam Smith, is unwilling to treat people within the economy like pawns on a chess board who can just be moved around.
BOETTKE: Definitely, that is part of his whole idea. In fact, I remember when I was a graduate student, there was an Economist cover. They were trying to explain the invariance proposition because that was big in time. Remember that new classical macroeconomics, when it first became popular, it was summarized in the invariance proposition. Later, it just was a methodology of having these microfoundations. Especially when you have the new Keynesians like Larry Summers and Greg Mankiw and Janet Yellen and others who are trying to show that you could have unemployment models with rational agents, which is a key part of all this.
It’s intriguing, by the way. I just started reading Ned Phelps’ new book on the journey through economic theory. I’m sure this will make me think about this again. I do agree with you about Lucas and the making of an economic miracle in that paper, and his work in that, just like Romer’s work. Romer focuses on charter cities for a reason, which is that that’s basically constitutional economics. As Acemoglu and Robinson want us to think about institutions and constitutions as well. Again, that work is all pushing in that way. Shleifer, in the work that he did in the ’90s on law and finance, very much in those—but I don’t mention those guys. My mistake, probably.
RAJAGOPALAN: No. It’s interesting because there are points when the mainline and mainstream converge, and there are points when they diverge. The Nobel Prizes are these weird moments when suddenly the mainstream will acknowledge—
BOETTKE: That drove a lot of the naming, is that in the modern idea—and those guys haven’t won yet. They’ll win, and then I’ll take credit for him or whatever.
RAJAGOPALAN: I think the next edition should have Lucas.
BOETTKE: I’m going to think about that, actually. It’s a very good idea. Especially Romer, I think, too.
RAJAGOPALAN: Romer too.
An Economy of Verbs
BOETTKE: Even in Romer’s Nobel address, he emphasizes this issue about the move toward treating agents with novelty and creativity. I’ve been thinking about that a lot because in our community, in the Coasian kind of world, we think about someone like Steven Cheung. He has this wonderful piece called “The Fable of the Bees,” and he explains this idea. A lot of people, what they focus on is the end product, which is that an efficient contract was formed. I want to try to say to everyone, “Okay, I get it, that an efficient contract, that’s amazing, but why don’t we recognize the following thing?
Step one, there must have been a massive externality that they didn’t recognize to begin with. Step two, creative and clever people came up with a solution. It took them a while to get that solution to work out. But what we should focus on is not so much that they came up with a solution that was at one point optimal, but that they were coming up with a solution and what conditions allowed them to negotiate around to do that.” If you do the Coasian interpretation of Cheung, you end up with the focus on prior to the contract, or the activity that led to the contract. To me, that’s where all the action is in economics.
I really like this recent article by Brian Arthur called “Economics in Nouns and Verbs.” His argument is that for us to make progress in the next generation of economic theory, we have to move from an economics of nouns, an economics of states of affairs, to an economics of verbs, which is about activity. To me, that’s what we need to actually embrace. He acknowledges that we already have a body of economic theory that works with verbs, that’s called Austrian economics. To me, I think that’s what we need to work on.
If we’re going to be doing that economics of activity, we have to remember that that activity doesn’t take place inside of a vacuum. It always takes place inside of what? Inside an institutional context. In that institutional context, structures are incentives, and we shouldn’t deny that, but it more importantly controls and dictates the flow and quality of the information and knowledge that we can discover and work with.
Let me just say one last thing about this knowledge issue. One way to think about the great mainline economists—Buchanan, Hayek, Coase, Ostrom, Vernon Smith—is that they look at the world through, in many ways, a neoclassical prism. Actors are choosing on the margin against constraints, and so it’s not like they’re saying there’s no constraints or there’s no margins or whatever. They’re choosing on the margin against the world of given constraints. They’re facing these tradeoffs. What they do is, they slightly always twist the prism. They twist the prism differently than what, say, you would get in a standard graduate-level textbook variant. By twisting that prism, it opens up these new lights.
Elinor makes us think about collective action problems in a way that when Mancur Olson or Garrett Hardin, I’m in a remorseless trap and I can’t get out of it. But she twists it, and all of a sudden I can get out of it by various different creative and clever actors inside of my model doing things. Elinor is a fascinating example. At the very end of her book, she has a section—it was 214 and following in “Governing the Commons.” It’s called “The Challenge Toward Scholarship” that her position takes.
The main idea is that she says, “We don’t want to have an economics about governing over; we need to have an economics that’s suited for governing with.” Then she goes to a whole thing about self-governing democratic societies, and she ends up at the end not talking about common-pool resources, but instead self-governing democratic societies and the nature of those, and Tocqueville and Madison and Hobbes and Smith and all these people like that. That’s how she ends her book.
To me, what that tells us is that what her book was about was test-betting democratic self-governance within a really hairy problem, which is collective action faced with a common-pool resource or basically fugitive resources like the fisheries or whatever, and how is it that individuals come to solve their problems themselves from inside of the model themselves. To me, we need more of that kind of stuff and more embracing of that. I think that’s different than all the people that do the economics. I’m going to think about this Lucas thing a lot. Thank you.
What Is Endogenous to the Model?
RAJAGOPALAN: One way I would frame that question is, what do each of these theorists consider endogenous to the model versus exogenous to the model? You do have people who sit on the seat of Adam Smith, someone like Kirzner for instance, who still thinks institutions are given and fixed, not because he actually thinks institutions don’t change or something like that, but for what he’s studying, which is the process within the market, institutions are given and fixed. We’re going to take that as given and fixed, and then we’re going to do what economists do.
Now, if you were to introduce a slightly different kind of Austrian economist, if you bring in Lachmann or Mario Rizzo, they made their career out of endogenizing the institutions that many Austrians before them took as given and fixed. They were very Hayekian in that sense. Now, if you look at what, for instance, Buchanan has done, he’s endogenized a whole number of institutions including lawmaking. But what is given to us is the Constitution, because we did something in the Constitution-making. We got this moment, we got the Constitution, then the relatively absolute absolutes. That’s relatively given and fixed.
Now, Ostrom goes one step further. We don’t have this relatively given or exogenous and fixed Constitution. In fact, part of solving this specific problem includes solving that bigger constitutional problem. They’re actually the same thing to Elinor Ostrom. She also endogenizes that element, which Buchanan has left on the table as exogenous. I honestly think of it as really where they are placing the model.
For economists, until very recently, preferences are given and fixed. No human being’s preferences are actually given and fixed. It’s only very recently—Buchanan is part of that. He starts endogenizing preference formation in some sense. Then behavioral economists are like, “No, no, no. Preferences are given and fixed, and that’s what they should look like as a normative benchmark.” That’s how I think about reframing the mainline and mainstream economics. If I go with this, I think Lucas—then even Schelling belongs.
BOETTKE: Sure. Schelling definitely belongs.
RAJAGOPALAN: Schelling belongs. Do you know the weird thing about Romer is, I feel like he’s the one who’s really answering the original Adam Smith question of, “What makes certain places poor versus rich?” Wherever I look—and I’m thinking about this from a development economics point of view, everything that people are trying to engineer into people having better lives, people not living in poverty and squalor and all the things you mentioned. Romer has, all the places that look good have figured out the endogenous growth theory. Something is working there. All the ideas are cross-pollinating, or as Matt Ridley would say, ideas are having sex. Where all the good places look like that, and all the places in decline look slightly different.
BOETTKE: Just to start with the issue with Romer, I think that one of the key issues is that his notions of increasing returns is Smithian, not Ricardian. Most people don’t make that distinction. They just go with Ricardian increasing returns, and that’s the way they think about these things. They don’t get the point of the generalized increasing returns, which, of course, Buchanan emphasized as well with his work with Yoon. That’s a key issue. How do we work with that idea?
I think if we go back to my earlier discussion about the move from logical soundness to logical validity, that one of the issues that happened was, with logical validity, we ended up by having the notion that abstractions entailed unrealism. We allowed for unrealism of assumptions.
This is one of Friedman’s mistakes because—even go back and look at the debate between Friedman and Samuelson in the AER [American Economic Review], where Samuelson defends the realism of assumptions, but Friedman unleashed economists by saying it doesn’t matter. I think that’s a huge issue because I think that in the Austrians, they deal with abstractions, but they’re realistic. They’re not unrealistic abstractions. They’re realistic abstractions which hold their feet to the fire of the world in some sense. I think that’s important.
The other thing I just wanted to mention to you about the endogenizing of institutions is, Buchanan has this great passage in the beginning of “The Demand and Supply of Public Goods” in which he argues that economists can no longer be content treating the institutional framework as fixed and given, but they must derive the institutional framework itself. Then he adds, from the ordinary behavioral assumptions of the economists themselves. I think the way he chose to do that was his constitutional economics.
That’s why he gets to all of the things. Now, the reason why he does the veil of ignorance and those kind of issues, that goes back to an issue with Knight, which has to do with draining us of having motivated reasoners, because motivated reasoners will undermine the voluntary process of it all. Instead, he wants to have various different efforts. The difference between him and Ostrom in that regard is that that exercise, the veil of ignorance, allows him to have rational discourse without any squabbling. Ostrom, if you listen to her—there’s a great movie on her called “Actual World, Possible Future.” She does an interview. She was a wonderful human being, just absolutely a wonderful person.
RAJAGOPALAN: She came and did her Nobel Laureate tour starting with Mason when I was a graduate student.
BOETTKE: Yes. You remember.
RAJAGOPALAN: She drank Scotch with all of us, and she was so fun and so cool, but also just so involved in what every single person was working on.
BOETTKE: Yes, when you were talking to her, she made you feel like you’re a million bucks.
RAJAGOPALAN: She made us feel human.
BOETTKE: Yes, she’s just amazing. What she does, she laughs and she says, “You know what we found when we studied people making decisions? We found that they quarrel, they fight all the time.” She chuckles about it, and she says, “They have to figure out a way.” In my book “Struggle for a Better World,” I take that idea of that conflict. Wagner, of course, has worked with this conflict issue as well. Dick Wagner is a Buchanan student, your teacher or whatever. He’s brilliant. I use this idea of what I call sharp objects. Now, I will admit it comes from watching the HBO show on “Sharp Objects.” I’m not that brilliant, but I thought, “Hey, that’s pretty cool.”
These notions that we interact in the world in sharp objects—what happens is that we have to figure out a way that our institutions dull the edges of the objects because in our conflicts with one another, if they’re not dulled, what we’re going to do is we’re going to have mortal wounds. But instead, if they’re dulled, we’re going to end up by getting scraped and bruised, but we’ll still be able to cooperate with one another. I thought this is an analogy for the way in which we interact in a pluralistic society. That comes from Elinor because we’re wrestling over these scarcities and who has what, when and where, and we’re arguing over these things.
RAJAGOPALAN: Also who’s included and excluded from the decision-making.
BOETTKE: That has to be a constant conversation and negotiated and adjudicated all the time. Power relations are all involved. Think about all the things that she allows us to open up into that space of institutional analysis, which is different from the more rarefied discussion of institutions that you see in other economies. Another way to think about it is, they all fill in gaps of one another. You start with some and they get there, but there’s a gap. Then the next person picks up their framework and then fills in that gap.
Hayek gives us a program, then they build on that. I think, again, to go back about Hayekian ideas, what you’re talking about is the evolution of these various arcs. He has the economics of coordination problem, the abuse of reason project, the restatement of liberalism and then philosophical anthropology. What happens in the philosophical anthropology is pushing the co-evolution and spontaneous order story further and further into all of us. The origins of our mores are, in fact, in these spontaneous orders. He was pushing that agenda as far as he could run it.
I think there’s so much richness in the discussion of political economy and social philosophy that these thinkers have that we can learn from and continually push out. Again, it’s not about the thinker. I think that’s a mistake. It’s about the concepts that these thinkers work with.
RAJAGOPALAN: It’s the context that they choose to bring into the model or leave out of the model.
Economics as a Tool for the Curious
BOETTKE: We have to make these things alive for us today. The young people that are listening to the show here, whether or not they’re overseas or here in the States or whatever, they have to learn how to make these ideas speak to them today and speak to their students. That’s a key thing. If I could stress again, I think that the key thing about mainline economics, when it’s taught right, is that it’s a tool for the curious. Economics is a tool of social understanding for those who are curious about how the world works. It is not a tool for social control by those who want to engineer the system.
Adam Smith used just a simple phrase of the “man of systems.” He also says that it’s nowhere as dangerous as in the man who fancies himself to presume that he could, in fact, control the lives of others. Menger referred to it as Prussian police science. When we teach economics as a tool of social control, we’re teaching economics as Prussian police science. When we teach economics as a tool of social understanding, then we’re teaching economics as a branch of moral philosophy.
RAJAGOPALAN: There’s also the in-between. We think we are teaching economics the right way, but it can become an instrument of social control if we get it wrong. Which is what the whole AI and computation question tells me. That if you’re really thinking that people’s tacit knowledge doesn’t matter, and it’s just a matter of given data or information that now we have the supercomputer that can resolve, not only have we gotten the model of how the real world works or the process works wrong, we can also do a lot of damage.
BOETTKE: Maybe this might give a good example. George Soros made billions of dollars as an international currency trader.
RAJAGOPALAN: Yes, and bet against the Bank of England, famously.
BOETTKE: One of his policy prescriptions in his retirement has been that he wants to have a world central bank because he wants to get rid of the spread in international currencies that he arbitraged and made so much money on. I think that it’s really counterintuitive to a guy like him that his very behavior was a thing that stabilized markets. It was that he and others were in fact arbitraging that actually gave the market its order. And if we didn’t do that, how the hell is the central bank going to know how to actually do things now?
RAJAGOPALAN: An alternate view of the world which can achieve the same thing is lots of decentralized banks, which is the Hayekian view. You have lots of denationalization. You have lots of different currencies, and now you have all the crypto guys who come in and who say, “Oh, we can do the same thing.” Not controlled by an individual.
BOETTKE: That’s the thing that so many people who don’t get the mainline idea miss. They miss the benefits of the competitive process of contestation. They miss it in terms of experts, that experts need to be contested. Hayek is a criticism. He quotes you when he says he’s going to use reason to whittle down the claims of reason. He’s not denying reason. He’s using reason to whittle down the claims of reason so that we can have reasonable reason. It’s the same thing with experts. No one here is anti-expert. What they want is their anti-monopoly of experts.
We don’t want to concentrate the power in the hands of a few, because if we concentrate the power in the hands of a few, even a sincere error by the hands of a few can threaten the entire system. It’s a system we can’t afford. We need to have constant contestation and built-in resiliency and tests and whatnot. That’s this institutional design question.
Uniformity vs. Decentralization
RAJAGOPALAN: I think a lot of people don’t like the decentralized process stuff that we talk about because there is some attraction for—we want everything universally to be uniform. When there are lots of decentralized actors and we don’t know exactly what they know and we don’t know exactly what they think and we don’t know exactly what context they inhabit, that makes people very nervous because if we don’t know it all, then how can we intervene and create certain uniform outcomes for them?
In our pursuit for uniformity of outcome or what we try to call this kind of equality that everyone is looking for, I think the appreciation for diversity, decentralization and pluralism gets taken away. That’s the institutional or intellectual tradeoff that we’re constantly making at the political level.
BOETTKE: Scott Page has these great models where he tests the idea of diversity and the power of diversity because it gives us creativity, novelty and whatnot that we otherwise wouldn’t have if it wasn’t there. It’s the same reason why you want to have different voices. You want to test your ideas out in different environments and everything like that. It is a very fascinating issue to think about what generates a creative society.
Hayek has, in “The Constitution of Liberty,” one of my favorite chapters called “The Creative Powers of a Free Civilization.” If you read that chapter, he doesn’t tell you that he knows what the creative society would ever give us. In fact, it’s the opposite. It’s that what it does is, we maximize the chances of accidents to give us things that we didn’t know we wanted,] but now we know we want. He actually quotes in there Lewis, who was one of his students at the LSE, the first Black man to win the Nobel Prize in economics and a professor at LSE. And Sir Arthur Lewis, in that discussion that Hayek has there, it’s all about innovation and how innovation comes from minorities because minorities see things differently than others see it.
A fish doesn’t recognize the water that it swims in. If I’m a native, if I’m a majority, I don’t recognize what the opportunities might be. But I come in, and I’m a minority, and I’m blocked off. Necessity is the mother of invention. I’m like, “You know what? That’s pretty cool.” I come in there, and I do that.
I think that permissionless innovation, that system that is constantly churning, it’s upsetting to people because they want stability. But at the same time, it’s the source of why it is that we have all the cornucopia of beautiful things that we have. And most of human history, in fact, didn’t have that. It was like this and then, “Whoa,” like this. The question is, what just happened here that unleashed this power of a free civilization?
RAJAGOPALAN: I think Hayek sums that up beautifully in the title because it’s “Individualism and Economic Order.” You get the order, but you also get the individualism. In fact, you get the order because of the individualism.
Thank you so much for doing this, Pete. This was such a pleasure. It feels like I’m back in your classroom. It was a lot of fun.
BOETTKE: It’s easy to have this conversation with you, and I learned a lot. I’m going to now rethink some of my characterizations and take into account some other people.
RAJAGOPALAN: I think it’s a great opportunity for the next edition, and we know that in the second edition, if there’s another chapter on Lucas, Romer and Schelling, then the first one can’t compete with it.
RAJAGOPALAN: You can not only create a new product, but you can also make more money out of this. This was such a pleasure. Thank you.
BOETTKE: Well, thank you very much.