March 2017

Sun Mon Tue Wed Thu Fri Sat
      1 2 3 4
5 6 7 8 9 10 11
12 13 14 15 16 17 18
19 20 21 22 23 24 25
26 27 28 29 30 31  
Blog powered by Typepad

« How The Grinch Stole the Free Market | Main | Warren Samuels on George Stigler as a Political Economist »


Feed You can follow this conversation by subscribing to the comment feed for this post.

I haven't thought about your questions long enough to answer them, but I have one of my own.

What are your thoughts on Ikeda's "Dynamics of the Mixed Economy?"

I don't remember if he addressed the issue of mistakes or not - it's been a couple of years and it was a library book.

Mistakes can become embedded in the culture, especially if they are taught in the schools that everyone is supposed to attend.

Does mistake here refer to random error?

In any case, it's scientism. Just because there isn't a Theory of Mistakes doesn't mean you can't OBSERVE that people are making mistakes. Following Stigler, no paradigm would ever shift.

He's right, though, that Mistake Theory is insufficient to explain why it's always the same mistakes, assuming "mistake" means something like "error" and not "bias."

Ideas can lead a society astray, they can save a society from ruin. But we do not have the idea that a 'restart' button should be woven into the framework itself. We talk about what does not work, but we get totally caught in the discussion with no way to extricate ourselves from the thought process to actually do anything about it. In a sense we are overwhelmed by the noise of complexity, but the actual solutions that lie beneath the complexity are so obvious that they cannot even be seen, for instance the fact that money cannot represent everything necessary in society gets totally missed time and again. It was a huge mistake for government to take over the functions of culture itself. But now the damage has been done and while we don't want to go back to the societal rules we once had, we do not have functions built into the democratic process to carry out the kinds of cultural changes we actually want to achieve on our own.

"Can ideas lead societies astray, [or] save a society from ruin?"
Keynes thought they could: "The ideas of economists and political philosophers, both when they are right and when they are wrong, are more powerful than is commonly understood. Indeed the world is ruled by little else. Practical men, who believe themselves to be quite exempt from any intellectual influence, are usually the slaves of some defunct economist. Madmen in authority, who hear voices in the air, are distilling their frenzy from some academic scribbler of a few years back." (General Theory)

See Acemoglu, Daron "Why Not A Political Coase Theorem? Social Conflict, Commitment, and Politics," Journal of Comparative Economics, 2003.

His paper discusses these various approaches to political economy, and develops the argument that there are strong empirical and theoretical grounds for believing that inefficient policies and political institutions are prevalent

They are chosen because they serve the interests of politicians or social groups holding power, at the expense of the society at large.

At the centre are the commitment problems inherent in politics

Parties holding political power cannot make commitments to bind their future actions because there is no outside agency with the coercive capacity to enforce such arrangements.

Tullock wrote on entry barriers in politics with democracy as franchise-bidding for natural monopoly.

The workability of competition at the bidding stage depended on barriers to entry into politics. His barriers to entry approach emphasizes that suppression of political competition is especially important for dictators.

Free entry is central to a political Coase theorem as is enforceable contracts.

People can't help themselves. Everyone favors a limited government unless they themselves are the only rent-seekers. So everybody cheats.

It's the classic prisoner's dilemma.

Another good place to look to for thinking about this is the excellent Evans and Friedman article in the most recent CR, "'Search' v. 'Browse'".

When I was reading a draft of it, though, one of the things that bugged me about the article was exactly the point that there's little sense of exactly how to model these problems. I'm not even talking about formal modeling - just thinking about them in a structured way rather than recognizing the fact that we don't have such a model.

I think one of the major problems with identifying "mistakes", particularly in terms of peoples' judgements, is subjectivity. A "mistake" may not be a mistake given the goals at hand. Unemployment insurance is a good example. The public may or may not be aware of its impact on unemployment duration, but I think it's fair to say economists are universally aware of this - and yet there's strong support among even these informed economists for UI. Why? The reason, I think, was best captured by Martin Bailey who once wrote that "unemployment may increase as a result of UI, but it matters less". So is support for UI a "mistake", or not? Even defining these as "mistakes" is tough. I think two years of UI seems very long, and we can support families dealing with hardship without continuing to extend UI. But who am I to say that people who disagree with me on further extensions are making a "mistake". The UI-duration question is a very subjective thing.

Theorizing "mistakes" should be relatively easy. An agent has expectation E(X) when actually the expected value according to unknown probability distributions is X+b. Agents act on E(X), but reality acts on E(X+b). Voila - a mistake. The hard part is figuring out what's a mistake and what's not a mistake (reasonable people will disagree!), and what "b" is, so that the model is actually useful.

If Ron Paul gets elected a lot of people will be writing posts about how government is getting back on track, and then I'll be writing posts trying to figure out how the public makes irrational mistakes. When you think about exactly how you would construct and apply these models, you realize it's a lot harder than it first appears.

Which makes sense. If it were easier, then rational actors in an efficient intellectual market would have made the model by now!

I am not sure I see where we are so far from a good theory of error. Public choice goes a long way, starting with rational ignorance. Evolutionary psychology tells us about signaling and about our cave man ideas. Paul Rubin is good on such topics. Finally, complexity theory and other stuff show that errors are sure to happen. Should be a filter argument too.

Kuehn, you make a good point, but I feel you're side-stepping the core of Boettke's question.

There may be some rational (but subjective) justification for issuing unemployment insurance, but with today's technology we _do_not_ require a massive federal institution to do it. The whole process can be managed with software, which costs a little bit up front but greatly eliminates the perpetually growing state.

In short, you've managed to explain why a political position may be considered rational by some actors, but you haven't accounted for the fact that pretty much _everyone_ opposes government waste and gargantuan bureaucracy. Pretty much _everyone_ agrees that larger administrative expenses cut into our potential to deliver "the biggest bang for the buck."

In that sense, the growing state is always a mistake, unless you can provide some example of an "arguably necessary" state function that is incredibly labor intensive. That example is probably non-existent though.

The question of private unemployment insurance is interesting, but a little more complicated than that. Donald Parsons has a lot of good work on why the public and private systems we have differ and why UI is hard to imagine privately.

Which may be fine - one may not like unemployment insurance and prefer simple job loss insurance, which can and does emerge privately.

I don't quite understand your discussion of a growing state. UI is being automated, it is becoming less labor intensive. I've been involved in projects evaluating those systems on the program's efficiency, and they work well and the government knows that. That process has been going on for a long time. Nobody likes big labor intensive government for its own sake (as you seem to agree).

But you can't just say "nobody likes big government" and then observe "but we have big government" and then conclude "we've made a mistake". Our comparative advantage is the economic way of thinking, and we recognize that there are tradeoffs. A big government may not be a mistake at all - it may be an acceptable tradeoff for other benefits. And those things evolve over time. UI is less labor intensive than it was 10 years ago.

Isn't the logic of concentrated benefits/dispersed costs, when combined with voting institutions, a kind of mistake theory? The "misinformed populace" need not be misinformed about the policies, just misinformed about the properties of the voting institutions (falsely thinking majority rule protects the majority from exploitation by a well-placed minority).

Mike -
Certainly, but aren't there also a lot of concentrated benefits and dispersed costs that may reduce the size of government too?

How do the problems pointed out by social choice theory (in addition to public choice) affect the discussion, if they do at all?

“Stigler argues that we economists do not have an operational theory of mistakes…”

I doubt that econ needs a theory of mistakes any more than it needs a theory of ignorance. Some things are obvious and should be left that way. People have to state the obvious about ignorance only because some idiots insist that people are not ignorant. Is Stigler suggesting that people don’t make mistakes? Or is he just saying we don’t know why people make mistakes?

We make economic mistakes because we are ignorant and have short term perspectives and differing values. As McCloskey has well documented, economic success requires a specific set of values. She/he hasn’t said so yet, but values come from our worldview/religions. We have strayed from classical liberalism because we have lost those values. And we have lost those values because we have abandoned traditional Christianity.

I have a hard time explaining to my students that people can be rational and not wise. People can be rational and arrive at different conclusions because they have different values and time horizons and therefore different assumptions for their starting points. The majority of people must have bourgeois values, place some importance on the long run (part of the bourgeois values), and have correct ideas about how economies work in order to prosper.

We have to keep in mind that modern liberal philosophy happened by accident. At the time that the scholars of Salamanca wrote most people believed that God decided who would be wealthy and who would be poor, much like Tebow thinks God determines who will win football games. The Dutch didn’t create the first capitalist nation because they thought it would make them wealthy. They thought God determined those things. They protected property and implemented free markets because they had the bourgeois values, which they got from their theologians who got them from Salamanca.

Dutch success surprised everyone, but most Dutch writers attributed their success to God’s blessing. Only a few, like Peter Delacorte, actually analyzed rationally what had happened. Dutch success and Spanish failure (just as surprising) prompted scholars to look into the causes and economics began. Adam Smith held up the Dutch Republic as the best example of his philosophy of natural freedom.


First of all I also regularly assess the size and scope of federal agencies professionally, and my observations conclude the exact opposite of what yours conclude. My experience is that every act of automation has come with a growing number of federal employees. How can we reconcile my observation with yours? Which one of us is wrong?

This should be easy to sort out, since the BLS and other sources track the number of federal employees and contractors on the payroll. You're telling me the number is going down? Are you sure?

Second of all, you're contorting my point so that it looks more like yours. Under the assumption that we none of us want to waste money, fewer highly productive employees is better than more unproductive employees. It is obvious that if we attempt the former and end up with the latter, we have made a mistake according to our own assumptions.

How can you disagree with that?

From the public's perspective, big government is not a mistake in the sense that people don't want big government but get it anyway. It's a mistake in the sense that they don't want it in the abstract, but are *ignorant* of the fact that all the concrete programs they support add up to it.

From a free-market economist's perspective, it's a mistake in that such an economist thinks many of the programs constituting big govt. produce unintended, counterproductive consequences. But the public is ignorant of these alleged consequences, so the programs are widely popular. From the economists' perspective, the public is mistaken, but that's because the public has not been exposed to the same economics teachings the economists have.

These types of mistake cannot be studied using non-Austrian economic theory, because both mistakes are caused by radical ignorance, which is impervious to incentives. In the first case, the public is inadvertently ignorant of the additive size of the programs it supports individually. In the second case, the public is inadvertently ignorant of the reasoning behind free-market economists' arguments against the programs the public supports.

(Don't tell me this is rational ignorance or rational irrationality on the part of the public: the public doesn't even realize that its members individual votes don't count, and without that realization, rational-ignorance and rational-irrationality theories can't get off the ground.)

Both mistakes have been extensively studied by political scientists. Young scholars interested in doing this type of political science--Austrian political science--should contact me through the Critical Review website.

How do we know what "the public wants"?

I've never said the number of federal employees is going down, I don't think. Anyway - no, I don't think that.

"Second of all, you're contorting my point so that it looks more like yours."

I wasn't entirely sure what you were getting at - there was certainly no deliberate contortion.

I guess I'm just not sure why you're so sure that we've ended up with more unproductive public employees. I'm not sure how to even talk about "productivity" in the public sector. The government seems to be doing it's job better than it did twenty years ago - particularly with respect to social welfare programs - but government employment as a share of the population has stayed fairly stable.

Anyway - I'm not sure you're understanding my point, which is quite different from what you're getting at. And I'm not even quite sure I'm sure what you're getting at (particularly if you think you're speaking to my comment).

I have some thoughts and numbers here:

If ever-expanding government is detrimental to general individual fulfillment, and if most people are through their political choices causing government to expand, then they are making a mistake. This is not a "theory," it's a fact. As a simpler analogy, suppose you point out that my arithmetic mistakes are causing my books not to balance. To call this a "mistake theory" is vacuous. Furthermore, it would be absurd to suggest that you can't explain why my books don't balance until you explain why I'm having trouble with arithmetic. (Which is not to deny that the latter explanation might itself be usefully illuminating.)

"These types of mistake cannot be studied using non-Austrian economic theory, because both mistakes are caused by radical ignorance, which is impervious to incentives."

I've never understood this claim, Jeff. First, I hope we all know that "non-Austrian" economic theory isn't just the sort of stereotypical rational agent stuff you get in the first chapter of an intermediate micro book. That should be clear. But aside from that, I don't see why you think this is so hard. We act on some assessments of the situation. We can jazz up that model to make our action more or less "rational" depending on what "irrationalities" we think are particularly important. If an agent is radically ignorant, then we just note that they're going to act on expectations that are inconsistent with reality. Give the agent expectations E(X), have them act on E(X) using whatever kind of rationality or irrationality you want them to have, and then observe what happens if these agents are acting in a world that actually has things happen with E(Y). The agents have no incentive to search out E(Y) because they are radically ignorant (they do have an incentive to search for things they are aware that they don't know - and they're going to conduct that search with a set of expectations that may or may not conform to reality too).

Modeling radical ignorance with "non-Austrian" economic theory seems easy. Assume an agent has knowledge (probabilistic or otherwise) that is wrong but, no knowledge that their knowledge is wrong. You could even model agents who - knowing that radical ignorance exists - discounts their expectations or deliberately stay "alert".

That all seems easy. The hard thing is to use that model for predictive purposes. If we could predict in what areas we are radically ignorant well enough to model them, we wouldn't be radically ignorant.

"If ever-expanding government is detrimental to general individual fulfillment"

This is not a self-evident "fact." This is a value judgment about the ends of political action. You need a political philosophy, with argument, to show why this is the case. Moreover, you need an argument to show why those who disagree with you are not just wrong, but unreasonable (or have made a mistake in logic).

I suppose I should say "have agents act with expectation E(X)=Y, but have reality be E(X)=Z". Voila - you have a mistake. One could differentiate between mistakes and some kind of radical ignorance by considering cases where no expectation of X even enters into an agents' behavior, but where an outcome X still affects what happens to them.

To me, "studying" something does not mean putting it into an equation. It means predicting it (as a practical matter, impossible) or retrodicting it, and in this case, that means knowing the culture and politics of the people you are studying. I.e., Weberian historical science.

The irrelevance of non-Austrian economics is that since radical ignorance is culturally dependent, it cannot be predicted by knowing or even imputing the agents' incentives or their interests. What they think is true of their individual or collective interests is unpredictably related to what actually is true of them.

Jeff -
I suppose I'm confused. Are you saying that non-Austrian economics disregards history and cultural context? This seems wrong to me. Certainly we can talk about and construct generic (ahistorical) models. That's a tough enough task itself, so it's no wonder people can spend a lot of time simply on doing that. But when these are actually applied by non-Austrian economic scientists, it's applied with a particular historical and cultural context in mind, right? Unless you're just describing how a particular mechanism works (i.e. - "what happens when people act on inaccurate information?") in a general way, the models are useless if they're not applied.

All that having been said - we're not really dealing with prediction here anyway because we're trying to understand a complex system. Understanding the system is a hard enough task itself. Forget predicting it.

That's like the difference between a meteorologist and a weather forecaster. The meteorologist is a scientist, like economists. The weather forecaster just does as well as he can letting me know if I should bring an umbrella or not. He does tolerably well at that task - well enough that people still read weather forecasts at least - but I wouldn't go as far as to say that therefore meteorology can give us predictions.

Hume, please re-read my statement. "IF something is counter-productive, then people doing it are making a mistake." One might of course disagree on whether the thing in question is counter-productive.

I'm quite familiar with your article, but I haven't read your book yet. Let me put it this way - non-Austrian economics does just fine explaining what happened. Non-Austrian economics can model radical ignorance simply by noting that people will act on expectations that are wrong and that they are unaware are wrong. Non-Austrian economics does just fine noting that bad things can happen as a result. Non-Austrian economics also does just fine noting that because people are aware that radical ignorance is a real thing, they learn to adapt to its existence. All of this non-Austrian economics can handle quite easily.

What non-Austrian economics and Austrian economics both have a much harder time doing is informing agents that radical ignorance is a problem before the bottom drops out. No economics is good as a predictive science because the economy is complex.

Lord Keynes at Social Democracy for the 21st Century has a great post today highlighting the huge swath of people who called the bubble from all sorts of theoretical background. Each looks equally ridiculous to the other. I think it's a mistake assign primacy in ability to predict a complex system.

"Hume, please re-read my statement. "IF something is counter-productive, then people doing it are making a mistake." One might of course disagree on whether the thing in question is counter-productive."

I was objecting more to the claim that "general individual fulfillment" is the standard for assessing political action.

I think my general concern is it seems that one needs (a) a moral/political philosophy as to the proper ends of political action, and/or (b) a moral/political theory for determining when it is accurate to say "the public" wants something. Part (b) brings in the problems brought to light by social choice theory (and all of the objections raised by deliberative democrats).

Ok, I guess my attempt at generic "good result" is itself open to challenge. But I don't think that was the issue in play. Assuming we can agree on what is the better result from a long-enough perspective to see how things work out, then I take it we would agree that people are mistaken if they act in way that is counter-productive of the better result. This is not a "mistake theory" that requires explication in terms of why people make such mistakes (although such explication might be desirable).

Friedman: "they don't want it in the abstract, but are *ignorant* of the fact that all the concrete programs they support add up to it."

I have noticed that about most neo-cons. The most obvious is their support for military intervention anywhere in the world. They can't seem to grasp how that contradicts limited government.

Not only is radical ignorance a problem, stubbornness reinforces ignorance. If you hold to certain values, socialist values, then you will not want to remove your radical ignorance about economics because you will see it as evil even if it is true.

Hayek touched on that in "Fatal Conceit" when he wrote that intelligence is overrated, especially by intelligent people. Hayek didn't say it, but intelligent people can be rational (Hayek called it pseudo-rationality) and unwise if they lack the bourgeois values.

Wisdom require intelligence plus bourgeois values. Foolishness lacks one or both.

It is less a problem of "mistakes" per se than it is of the general population making choices based on their evolved folk economics. We evolved to live in small groups that are closer to being organizations than spontaneous orders. Zero-sum thinking is also dominant. Folk economics tells you that, of course if the government spends more money that it will get us out of a recession (I just had this conversation recently, and had to explain why what is intuitive simply isn't true).

The general population makes mistakes because they accept their evolved folk economics. People also accept folk physics -- such as the intuition that if a ball shoots out the end of a spiral tube that it will continue to move in an arc (when it actually goes straight); or a bowling ball on a rope attached to the ceiling that you release from your nose will not swing all the way back and hit you in the face -- and have to learn actual physics to overcome it (and even then, many university physics students will still flinch and move at the ball's return). Folk biology argues against evolution. And folk sociology says society's an organization, not a spontaneous order.

So the real issue is that our evolved belief system does not match modern-day realities. We have to learn how the real world works -- and even then, those deep biases are hard to overcome. Just look at all the Keynesian economists -- they know economist, and yet accept his almost entirely folk economics-based economics. It makes "common sense," but in such cases as this, common sense and reality are two different things much of the time.

We should not be surprised that citizens who believe in folk economics elect politicians who believe in folk economics, and that both, then, support folk economics, and any economist who preaches what they already believe.

That at least is my theory of mistakes.

To borrow an overused meme:

"Radical ignorance. Daniel keeps using those words, but I do not think those words mean what he thinks they mean."

Claim: if you can write it down in an equation, it's not radical ignorance.


McKinney -
re: "The most obvious is their support for military intervention anywhere in the world. They can't seem to grasp how that contradicts limited government."

I don't want to necessarily defend neo-cons here, but I think it's important to note that we're not simply dealing with "pro-limited government" vs. "anti-limited government". The very term "limited government" is pretty vacuous. What limits? It's problematic to assume that because people don't share your understandings of the proper limits of government they do not recognize the importance of limiting government.

This is particularly the case because what most people want a government to do is limit coercion and limit involuntary cost imposition. People who oppose government intervention in those cases are thus perceived as the ones who are being coercive and anti-liberty.

If people were convinced libertarianism minimized coercion in society more people would be libertarian, period. It's a mistake to assume that the difference is over recognition of the value liberty. It's far more likely to be a difference in evaluation of the situation and relative priorities.

Steve -
Evans and Friedman provide this example of a piece of information that actors were radically ignorant of: "the housing boom is not being sustained by growing wealth of Americans and their desire to live in bigger houses" (Evans and Friedman, 2011, p. 78).

Let's call this outcome X. It has probability of being true 1.

Household decisions are a function of expectations for outcomes A, B, and C but not X. As Evans and Friedman point out, there's no incentive to search for information about the likelihood of X "if one did not already have reason to think that this proposition was indeed true" (p. 78). So perhaps we have households acting on expectations about A, B, and C which they know are important, expending resources searching for more information on A, B, and C based on expectations (that could be true or false) about the value of what they're looking for. But they are not expending resources searching for information on X and expectations about X do not enter their decision making process. As Bryan Caplan pointed out to Peter in their debate, we can always assign a non-zero probability to them finding out X and being surprised. If we want them to be alert or browsing, we can have them expend some resource to boost the non-zero probability that they'll find these things they don't know they're looking for. They will choose how much to commit to browsing based on an expectation of how successful the browsing might be, but of course there's no reason that that expectation is going to have to conform to reality (in that sense, it's another accident).

So decision making is a function of E(E(A)), E(E(B)), E(E(C)), any updates from search, and any random stumbling across E(X).

Add in whatever behavioral bells and whistles you want here. Hyperbolic discounting is likely important. Herd behavior. Whatever - I'm no expect in that. That's all certainly good stuff.

Churn it out to get human behavior. Show what that human behavior results in an environment where E(X)=1 even though the actors had no idea. You're going to get a result that explains human error. Error on information which you have no incentive to search for because you're ignorant is easy: make them not act on the information and don't give them a search function for the information.

How is this not the same radical ignorance that Evans and Friedman are talking about? Which of the faults that Evans and Friedman lists in mainstream micro am I committing here in dealing with modeling this?

1. I'm not assuming the agent has an incentive to search for information
2. I'm not assuming the agent knows the distribution of the information that he is searching
3. I'm not assuming the agents are omniscient
4. I'm not "ascribing irrationality" to explain why they did not act on E(X). They are rationally only acting on information they have or know is there to be had
5. I've incorproated the rational choice to browse for information, and I've recognized Evans and Friedman's point that these are accidents

I seem to have addressed all their points. Exactly what do you think I'm missing?

Certainly the ability to create an equation that someone argues represents some aspect of reality in no way indicates the person who created the equation does not have radical ignorance of what he is talking about. In fact, equations are so radically simplifying that they are probably a good signal that one does in fact have nothing but radical ignorance of what they are talking about. This become more and more true the more complex the process being discussed. One can create an equation for anything -- even to "prove" folk economics is true (as Keynes did).

The problem is that there is a set of math that describes certain simple processes very well, there is a set of math that gives probablistic descriptions and therefore hide the details, and there is theoretical math that describes nothing at all. It is possible to mistake the latter for the former two. In such cases, there is not just radical ignorance, but the person isn't even wrong. That, I think, is the main problem with Keynes and the Keynesians.

Daniel--First, just to clarify, Pete is posting about political errors, not economic errors. The latter are the subject of my paper with Tony Evans in Critical Review (free download: ). We view it as a contribution to Austrian economics. But Pete is asking for Austrian political science. This is what the main line of my work and that of my students is about.

In Austrian political science we apply a recognition of radical ignorance to politics and government to account for policy mistakes, just as in the Evans and Friedman article we apply radical ignorance to the economy to account for economic mistakes.

Now in saying that Austrian political science is not reducible to non-Austrian economics, I did not mean that it wasn't reducible to a formula. I simply meant that it explains mistakes that are not explainable or predictable by reference to political actors' objective interests or, therefore, their incentives. If you are saying that the heart of non-Austrian economics is not incentives but the ability to turn words into formulae, then more power to you--in that case, then "Austrian political science" is indeed reducible to non-Austrian economics.

"If people were convinced libertarianism minimized coercion in society more people would be libertarian, period."

I wish I were so sure of that. Granted, many folks think that free markets mean putting some kind of coercive power in the hands of robber barons to screw the rest of us. Nevertheless, it also seems to me that lots of people WANT coercion and, moreover, they want it used to produce outcomes that they prefer. What they need to be persuaded of is that the mechanisms of coercion will ultimately generate counter-productive outcomes by their own standards (or more colloquially, that those teeth inevitably turn around and bite them in their own butts).

Oh I would be one of the last people to say that economics is about turning words into formulae! I was just responding to Steve's request. I'd agree that non-Austrian economics (and Austrian economics, for that matter!) is mostly about response to incentives. I'm not sure it all is, but obviously that's very important. The leap I suppose I don't understand (and didn't understand in your article at the time) is why you think that prevents commentary on radical ignorance. You are quite right that there are unknown unknowns out there. Mainstream economists have noted this for at least a century and probably longer. If there is no incentive then there is no action under mainstream economics. That's exactly the sort of human behavior in conditions of radical uncertainty that we're looking for, right?

We can observe that its more complicated than that of course. I don't know the sorts of unknown unknowns that are going to be flung at me in my life. But I am aware that such a thing as an unknown unknown exists. That incentivizes me to do certain things, and we can talk about all that too.

As I said to you earlier when you were working on the paper - I guess my biggest issue is I'm not clear on how this isn't all thoroughly mainstream. Forget the equations that Steve is interested in seeing. How are these insights not mainstream insights? And if you want an equation to go along with the theory, I don't see why we couldn't write one. But I don't personally think that's necessary.

re: "Nevertheless, it also seems to me that lots of people WANT coercion and, moreover, they want it used to produce outcomes that they prefer."

Well of course - that's what government is, isn't it? But I'm not so sure this is wanting coercion per se. I think it's wanting to minimize coercion, and recognizing that zero coercion is impossible.

Daniel is suggesting that a recognition of radical ignorance is "thoroughly mainstream." In other words, he's suggesting that a mainstream economics that thoroughly incorporated radical ignorance would be no different from the status quo.

As a political scientist, I'm not qualified to respond except by speculating. But I'd sure like to hear what Austrian economists have to say about this.

My speculation: One difference would be that when it seems that massive mistakes are made in markets, economists would not turn to "irrational exuberance" (rather than radical ignorance) to explain it.

Another difference would be that economists would find themselves unable to predict what the effects of various public policies would be. My experience reading econ papers is that every single economist--unlike in any other discipline--ends his or her paper with unsolicited policy advice. An economist who recognized that people are unpredictable, due to radical ignorance, would not do that.

I believe this post from Sanford Ikeda appropriately gives the financial economics 'mainstream' on the matter of radical ignorance:

Oh look! The book is by one Jeffrey Friedman... huh, I wonder if he knows anything about radical ignorance....

Daniel, we have people desiring censorship and bans on Sunday sales, we have people wanting to put their neighbors in jail for growing, smoking, or selling pot, we have other people wanting to ban fatty foods or prohibit their neighbors from engaging in consensual sex that they don't approve of or marrying someone of another race, or who want the government to confiscate some people's earnings to pay the living expenses of others or force them to buy health insurance with politically dictated provisions or pay into a socialized retirement plan -- the list goes on and on. These things are not aimed at reducing coercion. They are coercive, period.

Sorry - I should have clarified what I meant by "this" in "this is thoroughly mainstream". I just mean that that way of talking about radical ignorance is a thoroughly mainstream way of talking about it, and it seems to answer all the criticisms you made in the article. No mainstream economist would raise an eyebrow over fully constructing the model I described and using it to talk about ignorance and the financial crisis. Again I have to refer back to Caplan v. Boettke. As Caplan noted, you simply can't say serendipity is incomprehensible from a mainstream view. There are immediately obvious ways of talking about serendipity in the mainstream. Likewise, there are immediately obvious, thoroughly mainstream ways of talking about radical ignorance.

Whether the mainstream talks about it often enough is a different question entirely. I'm not sure the answer to that question because I'm not entirely sure how often we should be talking about it, and I'm not even sure I could say how much they're talking about it now. They may be talking about it and calling it something different.


Do you think the work of Rustichini on Unawareness is value added to the contemporary literature? How about Weyl on Creativity? Or Ostroy on competition as a process? These are all very established economic theorists who are in fact trying to change the current practice of economic theory to be more amenable to the more radical insights from Austrian economics. What is your reaction to slightly out of sync work like Roman Frydman on Imperfect Knowledge Economics?

I really don't think you can persist in your claim that mainstream reasoning (as opposed to those trying to reform the mainstream) is quite comfortable with these ideas of uncertainty, ignorance, etc. Also, I'd be quite interested on your take on Franklin Fisher's Disequilibrium Foundations of Equilibrium Economics.

You will not be surprised by this, but the Caplan response from well over a decade ago was not persuasive then, and it remains not persuasive today. If you actually read theorists or work with theorists, taking these ideas into account with standard techniques has proven intractable. This is why the work on agent-based modeling is so very promising (have you read Axtell's paper on exchange in the EJ?) Scott Page's work on Diversity and Complexity must be trying to capture something that "mainstream" textbook models can not, otherwise why would his work be considered innovative within the social sciences?

Finally, I highly recommend Wagner's Mind, Society and Human Action for a very clear statement on the difference between close-ended models of choice and single exit theories, and open-ended models of choice and multiple exit theories. I'd also recommend Hardin's Indeterminacy, as a great discussion of the social sciences.

I'm not familiar with that literature, and this is really out of what I spend most of my time on. My judgement, though, is that anything that brings insights from the Austrian school into the mainstream and insights from the mainstream into the Austrian school is a very positive thing.

What is unfortunate to me is that mainstream economics is dismissed as being unable to talk about something when there are quite obvious ways to talk about it. You say it's not persuasive. Why isn't it? I've met a lot of people that have agreed that Caplan was quite persuasive. And I haven't heard any responses from anyone yet about why the sort of model I sketched out couldn't accomplish everything that Evans and Friedman discuss.

I've explained why I was unpersuaded by your position. I've explained exactly how I would set about talking about these things. I think it's incumbent on you now to explain what's wrong with my response. Simply saying Caplan was unpersuasive is... well... not very persuasive for those of us who agree with him. Why?

Note I'm not critiquing what Austrians have to say (I may have some small critiques about how tractable their ideas are). I'm asking why you guys are being so critical of us when I've addressed every single point raised in the article. You are the one that seems closed to certain types of literature, not me.

Anyway - I work mostly in labor economics, some macro, and on Keynes. This stuff is out of my normal sphere except insofar as Jeff asked me to look at a draft, so it's not a literature that I'm deeply aware of.

But if it's a literature you're deeply aware of I'd love to understand what's so unpersuasive about the general sketch of a model I presented. I'm still not sure where you all think that is inadequate.

Ultimately you're a professor, an author, an editor. You can throw literature at me until the cows come home. We can do it that way.

But I've never said there's not a good non-mainstream literature out there on this stuff, so all I'm going to be doing is nodding my head. None of this clarifies the point at hand. Nobody has disputed the value of the Austrian literature. What's under dispute is the value of the non-Austrian literature, and I still have yet to understand exactly what's lacking in the relatively simple outline of that that I provided.

Jeff is right about big government not being a mistake for those who benefit.

Sam Peltzman pointed out that most of modern public spending is supported by the median voter. Most of this spending is income transfers their pockets.
• Governments at the start of the 20th century were a post office and a military.
• At the end of the 20th century, governments were a post office, a military and a welfare state.

The studies starting from Peltzman showed that government grew in line with the growth in the size and homogeneity of the middle class that was organised and politically articulate enough to implementing a version of Director’s law.

Director’s law is the bulk of public programmes are designed primarily to benefit the middle classes but are financed by taxes paid primarily by the upper and lower classes. Based on the size of its population and its aggregate wealth, the middle class will always be the dominant interest group in a modern democracy.

After the 1970s stagnations, the taxed, regulated and subsidised groups had an increasing incentive to converge on new lower cost modes of redistribution.

Reforms ensued led by parties both on the left and right, with some members of existing political and special interest groupings benefiting from joining new coalitions.

More efficient taxes, more efficient spending, more efficient regulation and a more efficient state sector reduced the burden on the taxed groups.

Most subsidised groups benefited as well from so called neoliberal reforms because their needs were met in ways that provoked less opposition.

Political systems converge on efficient modes of both regulation and income redistribution as their deadweight losses grow.

The deadweight losses of taxes, transfers and regulation limit inefficient policies.

Policies that are found to significantly cut the total wealth available for redistribution by governments are avoided relative to the germane counter-factual, which are other even costlier modes of redistribution.

An improvement in the efficiency of either taxes or spending would reduce political pressure for suppressing the growth of government and thereby increase total tax revenue and spending.

The post is about something very important. Because the overstretching and incompressibility of the public sector is one of the most important features of our societies, and one which worries me a lot.

First, a methodological point: I'm fed up with the "it doesn't exist because it is not operational" mentality. It has impoverished economics enough in the last century, so I prefer not to follow Stigler on the way the problem is posed. What is relevant in the world may or may not be operationalizable whether the theoretician likes it or not. Most operationalizations are just simplistic assumptions that downplay the role of our ignorance about reality (e.g., trying to prove causality with vector autoregressions is rubbish, and making macroeconomics with correlations but without causality is sterile).

Then, a political science view: neoclassical economics is what happens to economic reasoning when ignorance and complexity are downplayed. Austrian and neoclassical theorizing are compatible in that they play complementary parts with different focuses. The size of government can be explained by prisoner's dilemmas, transaction cost arguments for political lobbying, imperfect information by part of diffuse interests, asymmetries between public costs (taxes, inefficiencies, moral hazard) and private benefits (legal privileges and public outlays)...

When I ask myself "do I need Austrian economics to explain this?" I basically mean "do time, ignorance, complexity, heterogeneity, creativity play a fundamental role in the final outcome that cannot be explained by theorizing in a general equilibrium rational framework setting*?"

Well, I'm not particularly convinced that there is a residual explanandum that requires an Austrian toolbox. Monetary economics without coordination problem is a scam, but the size of government can be rather well explained by inefficiences explained by public choice theories. I would of course highlight that ignorance plays a key role in real politics and that micro-rationality (individuals respond to incentives in their local environment) cannot translate in macro-rationality (Walrasian auctioneering) in the real world. But even setting aside Austrian ideas, the size of government would know no bounds even if there were no problems with fairness or public goods, simply because power is an overproduced public evil and liberty is an underproduced public good in any social context. Those who fight for liberty don't gain much as a result. Those who fight for power can leave off the wealth of their subjects. Bastiat got it all 160 years ago.

* I may be conflating Austrian economics with transaction cost economics formulating the question in this way.

"live off", in the last paragraph, not "leave off"

"To get an satisfactory explanation we have to postulate an economic theory of mistakes -- and economics of error --- and Stigler says we haven't even developed a mistaken theory of mistakes as of yet within the economic way of thinking."

Could Stigler say this today in the world of behavioral economics?

Glen Whitman and I have worked out the implications in:

"Little Brother is Watching You: New Paternalism on the Slippery Slopes." While the argument is about new paternalism it is generalizable.

A more "rationalistic" approach is in "The Camel's Nose is in the Tent: Rules, Theories and Slippery Slopes."

The point of the first is that we can explain more government than "anyone" really wants either by reference to cognitive errors or even within a rational framework. The second article concentrates on a rational (but efficiently limited information) perspective.

I think Stigler's perspective is out-of-date and largely misses the point. It is a simple-minded application of a perfect knowledge perspective --as wrong-headed in the economics of the political process as it was in the economics of perfect "competition."

Wipe Stigler out of your mind!


You are smart, you are motivated, and you are an advanced student in economics. I gave you literature that is currently going on within economics (or in the case of Fisher went on a long time ago).

This is not about reading lists, go back and look what I asked --- what is your reaction to their work.

Lets go to your field --- don't you think there is a difference in the way that Diamond talks about market friction and Pissarides? And would you find the work of Alchian on market frictions (and his price searching model in general) to explain the positive, not negative, role of frictions?

These are questions. If you say the opposite of what I am trying rhetorically to get you to bite on, then our conversation stops. But if, as I think you would, take the bait, then our argument just begins. Because as Horwitz said in a comment, saying you are talking about something does not necessarily mean you are "dealing" with that something. I think uncertainty and ignorance is in that class of issues. The way we can officially talk about them, fails to really deal with them.

The reason why I find Caplan unpersuasive on that particular issue (though I find him persuasive at many levels) is because he is just relying on the absorptive capacity of "mainstream" to insulate from paradigmatic challenge. There is nothing that I could say that he couldn't restate in terms he is comfortable with.

So here is my question to you ... during your training at W&M, or GW, or American, did you ever get introduced to the "bucket" in your stats class? Is that standard to just throw all unexplained phenomena into a bucket so we capture ex ante, all possibilities that are realized ex post but we couldn't name them prior to their emergence?

I think that a careful reading of Hayek essays "Economics and Knowledge," "The Competitive Solution" and "competition as a discovery procedure" will reveal that he was NOT talking about efficient search, but the contextual nature of our knowledge. It is not bits and pieces dispersed that are costly to acquire that excites his theoretical imagination, it is knowledge that exists only because of the context within which it is utilized. Absent that context, the knowledge doesn't exist. How do you model contextual knowledge is a search framework?

Buchanan's "order defined in the process of its emergence" and V. Smith's "rational versus constructivist rationality" get at this idea of the contextual nature of our knowledge.

But the bottom line is that knowledge is different than information, and search is very different from discovery.

These ideas, ultimately, are too difficult to sort out via blogs --- as Cowen has pointed out, philosophical issues are not good blogging material, though clearly politics is. Thus, we run into a problem in the communication of subtle points.

Everyone, please read Stigler carefully --- he is contrasting the mistake theory (Mises-Friedman) with the bias theory (Olson, Tullock) with the rational theory (Peltzman and Becker).

He has public choice in there --- it is the bias theory. He just doesn't think it complete satisfies. Nor, for that matter, does the rational theory. It is just that the rational theory does better than the alternatives AT THE TIME HE WROTE.

Don't read my summary, read Stigler in the original. In fact, for the holidays give yourself a gift and read as much Stigler as you can get a hold of.

McCloskey's values probably comes down on the side bias. But I think it's hard to draw clear lines between mistakes, rationalism and bias. That may be the problem. Those three are highly correlated.

I think the problem here with Daniel is that he is also in a heterodox school and reads more people in that school, who happen to take uncertainty seriously, than in mainstream, orthodox economics. If most of the people you read do deal with uncertainty, you will come away thinking everyone does. Pete's point is that mainstream economics does not deal with this issue. If it did, you would not have so many people writing papers trying to convince people to take uncertainty seriously.

It's not an uncommon problem. Especially among specialists. We only have so much time to read so many things. And the more we specialize in something, the more focused our attention, and the less we read and thus are aware of other ideas -- even if those other ideas are the more mainstream ones.

This fact is also why Pete constantly tries to remind everyone to engage mainstream economists.

Off topic:
Professor Ebeling at one time pointed about Leontief being very upset that Hayek got the Nobel, and Hayek also replayed with a critique of the input-output bla bla. Can anyone tell me where I can find those texts?

I'm not an economist, so my answers plus a quarter, adjusted for inflation, will get you a cup of coffee.

Is an economically coherent theory of mistakes possible?

Yes. In normal non-economist language, a "mistake" is when you do something under the assumption that reality is a certain way, but it turns out that reality is another way, so the results are quite other that what you intended. Seems to me quite obvious that you can incorporate that into economics and, in fact, do. The "loss" side of the profit-and-loss calculation wouldn't happen if there were no mistakes. There also seems to be no reason to assume that mistakes can be limited in time horizon or scale, other than that they must be finite.

Can ideas lead societies astray, and can ideas also save a society from ruin?

Yes on both counts. No one who followed Lenin or Mao into the crushing poverty and brutal tyranny of Communism thought that was the destination. And in much of the world, there is a deep suspicion of business and trade, married to a belief that just getting the right powers into the government will result in prosperity. Not to mention that most religions come with certain economic prescriptions, which may not be healthy. And on the flip side, the popularity of ideas about "liberty" and "rights" in the West seem to have done us well. In fact, I think the affirmative answer to this question is the subject of a recent book by Deirdre McClosky, "Bourgeois Dignity."

Persistent and widespread social phenomena cannot be satisfactorily explained by mere reference to mistaken opinion.

If Islam is true, than most of the Western world is in the grip of a persistent, widespread, mistaken opinion. If Islam is not true, then a huge swath of human civilization spanning a couple continents is in the grip of a persistent, widespread, mistaken opinion.

I think this quote shows where Stigler gets off base:

If deception by intellectuals were the motive force of social change, we would expect to observe on numerous occasions on which a group of conservatives with large powers of persuasion had captured the public's fancy, and succeeded in initiating a regime of declining governmental activity.

False ideas don't sweep the public imagination because of the erudition, subtlety, and persuasiveness of their proponents. They become popular because they make popular promises. Marx's ideas didn't gain traction because Marx was such a great rhetorician. They gained traction because they promised a carefree life in a Garden of Eden. Often times, the fantasy is simply more appealing to our instincts and desires than reality, and so the champions of reality must fight a decidedly uphill battle to convince the public. The power of a false opinion's promises may greatly outweigh the arguments against it.

Troy, I'll let Daniel speak for himself but I would venture to say that Daniel is most definitely NOT in the heterodox school associated with Keynes.

What a frustrating discussion to read. OF COURSE rational ignorance/rational irrationality is compatible with the so-called "radical" ignorance of the public. People don't know that their votes don't matter? So what? The point is that they don't bear the costs of their beliefs. Understanding is hardly necessary. They don't experience higher costs, regardless of how they vote individually. They keep voting based on their (biased, ignorant) beliefs, and they don't know their votes don't matter. Those premises are not mutually exclusive. Costs and benefits do not have to be consciously perceived to be experienced, and they must only be experienced to provoke a reaction.

For example, take pessimistic bias. Most people believe that the economy is bad and getting worse, even during periods of sustained growth. They may even be pessimistic about their own fate, and their own purchasing power. This does not stop people from responding to real affluence with real increases in consumption. In fact, the ignorance of increases in their own consumption allows them to remain ignorant of increases in their purchasing power, and of improving economic conditions generally. The mistake in all this is to assume that purposiveness (i.e. procedural rationality) requires the conscious weighing of costs and benefits. It often doesn't. People don't consciously say to themselves that they're saving on cognitive costs. That would defeat the purpose!

But they *do* bear the costs, according to free-market economists! People vote for policies that hurt themselves!

The crucial role of "knowing" that one's vote doesn't count in Caplan's theory is that otherwise, a voter who bears the costs of the bad policies would have to think they are actually *beneficial* policies, such that by voting for them, he would be helping in a significant way to implement these beneficial policies.

YET Bryan Caplan and his students think these supposedly beneficial policies are really bad policies. (After all, they're free-mkt economists.) This leaves them with only two ways to explain why voters behave as if (and answer polls as if) they actually *agree with* these policies, even though free-mkt economists think the policies are counterproductive. (1) The voters must be ignorant of the reasons the policies might actually be counterproductive (this is what an Austrian political scientist would say). (2) The voters magically know that the policies are bad, even though they usually have had no exposure to economics of any kind. But they also magically "know" that their vote doesn't matter, despite being told all their lives that "every vote counts." So they figure that they may as well "indulge their taste for irrationality" by voting for the bad policies, because their votes won't make a difference anyway (this is what non-Austrian economists playing political scientist would say).

Daniel: If mainstream economists (Caplan having gotten his Ph.D. at Princeton) were fully comfortable with radical ignorance, why would they come up with such completely unsubstantiated (indeed, contradicted-by-evidence, e.g., the evidence of millions of people voting) notions as "rational irrationality" when they think about politics?

Brad W., you must not read much Daniel writes -- he's about the proudest, most true-believing Keynesian I have ever met.

Troy, I read a lot of what Daniel writes. Sometimes it's insightful, sometimes it's frustrating. The point is, I certainly wouldn't categorize him as a heterodox Keynesian -- that is, a post Keynesian in the same vein as Kalecki, Davidson, Wray, Robinson, Sraffa, et al. I realize it's probably oversimplifying his views but he seems much closer to a mainline New Keynesian to me. At any rate, he can speak for himself.

Keynesianism is heterodox relative to mainstream neoclassical economics, generally understood as being orthodox economics. Thus, as Keynesian, whether New or Post, he's in an orthodox school.

Jeffrey, thank you for the comments about Caplan's view on voters. His book got so much praise that I began to wonder if I were nuts for disagreeing with him.

Troy, your last sentence, did you mean a Keynesian is in a [heterodox] school?

Anyways, I disagree with you, but it's really not important to quibble over this.

There's also an assumption that the public even knows what the policies actually are, or that the policies they think they're voting for are the ones that get implemented.


You can find Hayek's reply to Leontief in Hayek's "New Studies in Philosophy, Politics, Economics and the History of Ideas" (1978), chapter 14, 'The New Confusion About Planning,' pp. 232-246. (It originally appeared in "The Morgan Guranty Survey" (January 1976)

Richard Ebeling

As Alchian was mentioned, he defined efficiency as "Whatever is, is efficient."
1. If it wasn't efficient it would have been something different.
2. Of course, if you try to change anything that is there, that is efficient too.

The best way Alchian related this was the question of optimal taxes.

Alchian would ask "If something is so optimal, why don't we see it then?"

The notion was essentially that there must be other costs that you left out of your model: either costs involved in the political system, in organizing support, or changes required for this other solution which might seem to be such a low-cost option.

The basic point was why are we weighing only some costs and not others?

Why are these costs (involved in minimizing that particular dead-weight losses that would be involved in setting a particular tax) less important than other types of costs (those involved in informing people of what the options are or of organizing them to go out and try to adopt the alternative option)?

Why is the cost of fixing mistakes not taken into account when labelling them a mistake?

Alchian’s analysis of institutions spent much time showing that they arose to lower various costs of decision-making and social interaction.

A mistake is a decision we regret, or is a mistake a decision we would undo if the costs of decision-making and social interaction were less that they actually were at the time.

On people voting for policies that hurt themselves, remember the theory of expressive voting.

People vote against their interests because their vote is not decisive, so they vote for what gives them a sense of identity and self-worth.

Many gain pleasure, excitement and self-definition for cheering for particular parties and worthy causes in the same way as they cheer and boo for sports teams.

Consider Obama had nearly 80% of the Jewish vote in 2008.

Now see Expressive voting and identity: evidence from a case study of a group of U.S. voters A review of Norman Podhoretz, Why are Jews Liberals? Doubleday, New York, 2009:
• Podhoretz describes behavior that substantiates the hypothesis that people vote expressively to confirm identity.
• Podhoretz is concerned that liberal Jews vote against their self-interest.

Most of the Jews who voted for Obama did care about Israel but, downplaying or dismissing or ignoring his anti-Israel associations, they voted for him anyway

The expressive behavior hypothesis explains why they do so.

With a single vote not decisive in determining actual policies, the sole source of benefit from discretion exercised in voting is expressive utility.

Liberal Jews, in behavior that is rational, choose the expressive utility from voting against the ‘Right”, which is identified with past prejudice against Jews and with contemporary privilege.

The identity of a person who opposes privilege and cares about social justice is confirmed through the act of voting for the “Left”

Liberal Jews expressively support liberal principles through the low-cost actions of voting and rhetoric, so as to place the individual with freedom of choice at the centre of society, in the hope of being safe. Is this a mistake?

Mistaken policies are not mistakes under the theory of expressive voting because people are voting for feel-good policies they want.

The rules of the game allow the expressive voter’s policies to win, so the solution is better institutions, but there is expressive voting at the constitutional level too. The U.S. Constitution has immense expressive value – a romantic fantasy almost - for many.

We are then left with Churchill’s defence of democracy; the worst possible system of government except for all the others.

Is the best available system a mistake? If democracy is not a mistake, then maybe a Political Coase theorem is reborn?

Naturally, those who intend to remake society in their interests, adhere to the faction that has been successful in remaking society in in their interests. The left is simply the gang that has been winning, and so keeps on winning, while the right is whoever is getting run over by the bandwagon.

It is always safe, and apt to be profitable, to be lefter than thou, whereas right wing deviation is apt to be punished. This has been the case for a couple of hundred years.

@Professor Ebeling:

Thank you very much.

(1) To form an expectation, E, about whether I will buy a blue cashmere scarf at Bloomingdales today, X, assumes that I know the probability of X and of W (buying a grey felt hat) and of Y, etc. So to form E(X) I must have a proper probability distribution that exhaustively lists all mutually exclusive states of the world. If there are states of the world of which I am radically ignorant then the list won't be exhaustive, I won't have a proper probability distribution, and so I can't form E(X).

(2) True, Austrians don't have a theory of why people make mistakes. We do have theories of whether or not and how people correct their mistakes, depending on the rules of the game. Austrians assume mistakes will simply happen, the interesting thing then is to try to figure out why and under what conditions they will correct their mistakes. Neoclassicals like Stigler assume that, if you take into account people's choices to remain rationally ignorant, then people don't really make mistakes at all. That's why from a Stiglerian point of view people making genuine mistakes is either a mystery or more likely a mirage.


"But they *do* bear the costs, according to free-market economists! People vote for policies that hurt themselves!"

They bear costs, and they enjoy benefits. But which costs they bear and which benefits they enjoy aren't in any way connected to how one individual votes. So in that sense, individuals do not bear the costs of their votes.

"The crucial role of "knowing" that one's vote doesn't count in Caplan's theory is that otherwise, a voter who bears the costs of the bad policies would have to think they are actually *beneficial* policies, such that by voting for them, he would be helping in a significant way to implement these beneficial policies."


"YET Bryan Caplan and his students think these supposedly beneficial policies are really bad policies. (After all, they're free-mkt economists.) "

For the sake of argument, yes.

"This leaves them with only two ways to explain why voters behave as if (and answer polls as if) they actually *agree with* these policies, even though free-mkt economists think the policies are counterproductive. (1) The voters must be ignorant of the reasons the policies might actually be counterproductive (this is what an Austrian political scientist would say). (2) The voters magically know that the policies are bad, even though they usually have had no exposure to economics of any kind. But they also magically "know" that their vote doesn't matter, despite being told all their lives that "every vote counts." So they figure that they may as well "indulge their taste for irrationality" by voting for the bad policies, because their votes won't make a difference anyway (this is what non-Austrian economists playing political scientist would say)."

It's not either-or, it is in some sense both. Some large number of voters, maybe a majority, maybe not, clearly fall under option 1. But plenty of voters have opened up a newspaper or flipped on cable news long enough to get some vague sense that there is some sort of economic argument against the minimum wage, for free trade, etc. I would venture that a majority are aware at least that some people disagree with them. Do they investigate further? Do they look for facts? Do they try to figure out which beliefs are correct and which are incorrect? They do not. They are being rational in the procedural (economic) sense but irrational in the instrumental sense. They save on cognitive costs, while enjoying the benefit of clinging to their cherished beliefs (and the signaling value of expressing these beliefs can often be quite high). They reduce costs while retaining benefits. That's rational. And for the individual in a democracy, policy is exactly what it would have been if they a) had voted differently or b) not voted at all. So what that individual experiences in terms of the costs of "bad" policy doesn't change, even when she believes that her vote *does* make a difference.

For example: I had tonsillitis this week. I went to the doctor's office. It could have been viral or bacterial. In most cases, tonsillitis is viral, but if it's bacterial antibiotics are a very effective treatment. He could have tested for strep, but a rapid strep test gives false negatives 20% of the time, and there are plenty of other kinds of bacterial infections. Checking for all of those would be costly, mostly in terms of time. The PA saved on those costs. He prescribed antibiotics. I took them. My symptoms improved rapidly. I think it was probably bacterial, but I'm aware of the post hoc fallacy. My immune system had been fighting the infection for days, so it's possible that the infection was viral and the timing of my recovery just coincided with the use of antibiotics. Knowing the truth is difficult, so guess what? I get to choose what I believe. I choose to believe that it was a bacterial infection. Most people wouldn't even consider all of this, they would just believe that antibiotics fixed a sore throat and push for more the next time their tonsils swell up like golf balls. Most people are radically ignorant in that they don't know the difference between bacterial and viral infections. That belief (and that ignorance) has social costs, namely antibiotic resistant infections. People who share that belief share in those costs. But the costs of antibiotic overprescription do not impact the typical patient's eagerness to receive antibiotics. The costs of uninformed, incorrect beliefs are swamped by the sizable, immediate benefits to all actors involved. The PA saves time, which is no small savings. The patient gets a treatment that is perhaps 35% likely to work. Everyone is being rational. Some people are ignorant, some are not. But they are indulging their irrational beliefs at the same time. If anything, economic policy's effects are more complicated.

McKinney--Critical Review published a whole issue of critical commentary on Caplan's book, including a point-by-point critique by me. Ordering from Routledge is difficult; contact me at edcritrev@gmail if you're interested.

Josh S., you have this point nailed down tight. Contact me if you want to go into political science!

Steve--It's a nice theory, but as I said, there's no evidence for it and plenty against it.

The most overduplicated finding in political science is that most people do *not* have even the minimal political knowledge you attribute to them in saying that "plenty of voters have opened up a newspaper or flipped on cable news long enough to get some vague sense that there is some sort of economic argument against the minimum wage." See Philip E. Converse, "The Nature of Belief Systems in Mass Publics," reprinted in Critical Review vol. 18 nos. 1-3--again, contact me.

That's not to say that voters are unaware that people disagree with them, in the general sense that there are people on "the other side." But they attribute that to the other party's self-interested motives, stupidity, ignorance, or...irrationality. So if you're a liberal, why waste your time investigating conservative views, when everyone knows conservatives are stupid tools of the rich?

Therefore, I disagree with Sandy Ikeda a little bit. I think Austrians do have a theory, of sorts, of mistake: the world is too complicated to be understood at a glance. Even after a lifetime of study, one will, at best, grasp only a small part of the world, and contained in the other parts may be unknown unknowns that could overturn one's lifetime of learning. Given this theory, it's harder to dismiss people who disagree with you as knaves. But most people do not have this theory, so they do dismiss people who disagree with them as knaves. They think they can account for what their opponents think, so they feel no need to actually read what their opponents think. This is not irrational; it is evidence of most people's ignorance of the theory behind radical ignorance.

A final note on rational irrationality; this applies to "expressive voting" as well. Caplan, like Brennan and Lomasky, provides neither evidence nor logic for why people would feel better insisting that the minimum wage helps people (do libertarians feel bad when they say it hurts people?), why people would feel better insisting on a pessimistic economic outlook (surely the opposite is true), etc.

Caplan's book can be quite deceiving if read carelessly, because it has a big chapter filled with opinion survey data. But these data show only that voters disagree with free-market economists about several things, and that is all the data show. These data can either be interpreted as due to people's radical ignorance of (or disagreement with!) free-market economics, or as due to "rational irrationality," but the book doesn't contain an iota of evidence for the latter interpretation, and in my article I provide abundant evidence against it.

The third from last paragraph in Jeff's comment above is worth reading and re-reading. There's a lot going on in there that seems just right to me.

"That's not to say that voters are unaware that people disagree with them, in the general sense that there are people on "the other side." But they attribute that to the other party's self-interested motives, stupidity, ignorance, or...irrationality. So if you're a liberal, why waste your time investigating conservative views, when everyone knows conservatives are stupid tools of the rich?"

This is exactly what I'm arguing -- and I wouldn't argue any further. But the point is, in instrumental terms this is irrational. But it's procedurally rational. Thus the term, "rational irrationality".

Argh. I keep saying instrumental when I mean to say "substantive". My bad.

The overarching argument is that people buy more of things as the price falls. "Irrationality" in terms of indulging one's cherished beliefs -- and indulging them includes ignoring challenges, confirmation bias, etc. -- has a low price. Nowhere does Caplan argue that this is a necessarily conscious process. I'm not sure it's always a conscious process that leads people to buy more strawberries at lower prices. They certainly don't sit down and equate MU of strawberries/Price of strawberries with MU of everything else/Price of everything else. You'd have to have an extremely uncharitable interpretation of consumer behavior theory to think that's what economists are arguing.

There are a whole lot informational problems dealing with government besides:

a) The speed of feedback loops. You touch a hot stove, you burn your hand, you don't do that again. You develop the Ford Edsel, the public hates it and doesn't buy it, you stop selling it. The effects of public policy are not nearly so rapid and not so obvious (if they were, Adam Smith would have titled his book, "Common Knowledge Everyone Obviously Knows," Frederic Bastiat wouldn't be interesting, and Henry Hazlitt would have had no reason to write anything). It's pretty easy to figure out that the Edsel is losing you money. It's not so easy to figure out that progressive policy is the reason crime is high, the schools are failing, and businesses are fleeing. I mean, no one in Detroit wants that city to be a garbage dump. Seniors can see the Social Security check in the mail as a direct result of Social Security policy. They can't see the causal relationship to the decades of lost capital investment, the inflation, the debt, the unemployment, and the graft quite so easily, and it doesn't so directly affect them.

b) The frequency of choice. I choose what to eat three times a day. I choose which President I want once every four years. For every one time I have voted on what I prefer for President, I have voted on what I want to eat on the order of four thousand times. A market is, in a sense, a radical democracy where anyone can run for any office with sufficient backing, everyone is constantly voting on whether the office-holders should continue to hold office, and representatives are constantly being chosen all the time. Expecting information to travel through a political system, where you get one vote every few years rather than many votes per day, at anything near that rate or with anything resembling a market's efficiency is silly.

c) The scarcity of options. In 2012, I will be able to choose between a Republican and a Democrat. For lunch, I got to choose between Wendy's, McDonald's, Applebee's, a variety of local pubs, Subway, making my own sandwich, heating up some soup, cooking something fancy, in addition to the many, many menu options associated with each class of choice. I don't even know how to count the different choices I had for lunch! Whatever public preference for government actually is, it can't evolve as quickly through the political system as the public lunch-preference does in the market.

It's remarkable how the pathological "science" of the math economists have forced Friedman to come up with the weird word "radical ignorance" for what ordinary people call just ignorance.

Ditto how Frydman is forced to use the strange term "genuine uncertainty" because of the pathologies of the math economists, where the rest of the world gets to use the everyday word uncertainty.

If economists are to talk sense, the math economists have forced them to put a modifier on every word the math economists have stolen -- and wrecked -- from ordinary language.

There is a deep point to be made here about how "technical" terms get their ultimate hinge to turn on ....

Josh, you can choose between R and D, but you can't necessarily get your choice even there. If you choose Wendy's and I choose McD's, we can both have our choice.

Nevertheless, through their political selections, by ceding more power to government, people have in fact chosen to have fewer and less frequent choices and less responsive feedback, i.e., the situation you describe. In so doing, they have been mistaken. It is a challenge, and a worthy one, to help them learn to choose differently.

Greg--I agree. And "inadvertent ignorance" might be a better term, because the problem with mainstream econ, as I see it, is that, armed with "incentives matter" as the only tool in their arsenal, mainstream economists have no conceptual space for the inadvertent--e.g. mistake. But I defer to Sandy, who I think originated "radical ignorance."

Steve--the key point you seem to want to agree with is that until one has learned, say, free-market economics, one is inadvertently ignorant of its potential value in overturning one's existing beliefs about, say, the minimum wage. That value is an unknown unknown, and, *if* free-market economists are right about the minimum wage, it is a mistake not to learn what they have to say about the subject. Austrian political science, step 1.

Step 2: One can't know whether what anyone has to say about a given subject will be worthwhile until one finds out what they have to say, and even then, one could spend one's whole life researching the matter, whatever it is. So there's nothing irrational about radical ignorance. It is an inescapable aspect of the human condition. Step 3 is therefore charitability to those with whom one disagrees--a form of Verstehen.

This is directly at odds with Caplan's approach, which judges the public's failure to agree with Bryan Caplan's opinion as somehow irrational, not merely ignorant.

You refer to Herbert Simon's key 1985 American Political Science Review article, "Human Nature in Politics," which distinguishes between two definitions of "rationality": Simon's own bounded or "procedural" rationality, i.e., doing the best one can despite one's ignorance; and what mainstream economists mean by rationality, "substantive" rationality--i.e., not making a mistake (have you read this, Daniel?)! The two definitions are *opposed.* Simon is taking a slap at mainstream economics for assuming that agents know, hence do, what is in their objective interest to do.

So yes, it is true that Bryan is saying that voters are substantively irrational, i.e., wrong--because they disagree with Bryan and his colleagues. In that, he shows his true mainstream colors. But he says explicitly in the introduction to the book that this is *not* because voters are ignorant of what Bryan and his colleagues say. He says that voters are *not* ignorant--as political scientists have discovered. It's "worse" than that, he says: they are irrational.

Therefore, I have to disagree when you maintain that Bryan's story is that procedural/bounded rationality, i.e., ignorance, leads to substantive irrationality, i.e., error. Bryan never acknowledges ignorance on the part of voters after the introduction, and he goes out of his way to deny it. Thus, he claims that the truth about economics "falls into people's laps" every day, so it takes an active effort to avoid the truth for them to reach their mistaken conclusions.

This is *precisely* what partisans in political debate think of their opponents: that what the other side is saying is so self-evidently false that they cannot actually believe it. The truth is obvious, but they don't want to believe the truth, because it is somehow more satisfying to believe what is false.

"Emotion and ideology" do not connote ignorance, but Caplan constantly blames disagreement with him on emotion and ideology without providing a shred of evidence that this is true. (One of the first things political scientists discovered about most voters is that they have no clue about ideology. See the Converse article cited above.)

The one good thing about the book, I said in my reply to it, is that Caplan acknowledges that Adam Smith discovered *counterintuitive* truths. But if they are counterintuitive, then they do *not* "fall into our laps" unless we actively avoid them. Rather, we will believe intuitively plausible falsehoods until we are taught economics. Until then, our errors in economic reasoning will be due to our intuitions plus our ignorance of counterintuitive truths.

What was Bryan's response? He withdrew the acknowledgement that Adam Smith discovered anything counterintuitive! Thus, people who disagree with free-market economics intuit that they are wrong, but insist on indulging their incorrect views because it makes them feel good, and because they know that their votes won't actually help enact these views into law.

Josh, whoever you are, right on and send me an email. You are doing Austrian political science steps 4, 5, 6....

Again, the issue is that we can't expect the average person, whose economics beliefs are rarely anything more than folk economics, to make educated decisions, since they aren't educated in economics. Economics is counterintuitive, because our intuitions evolved in a tribal setting. Yet, it is actually pretty easy to teach people the basics of economics. Intuition tells people that prices are high because businesses are raising prices due to greed; economics tells us that prices are high because of competition among consumers, and that they are low because of competition among sellers. Once the latter is explains, you can see the light come on. It makes sense, but that's not how we evolved to understand the world to work. That matters. This is the source of most people's errors in regards to economics and who and what they support politically. If you want to know where the source of error is, that's it.


At the very least you've created a caricature of Myth of the Rational Voter. The standard is *not* disagreement with Bryan Caplan, or even disagreement with free-market economics. It's disagreement with the consensus of economists as he summarizes it from the SAEE. So that's one tweak on what you're saying.

The heart of the problem is Buchanan's "every man his own economist" *and* the ignorance that Converse observes. That's because ignorance is not neutral, it leads to reliance on certain other heuristics (yes, heuristics that evolved and were effective in hunter-gatherer or tribal societies). Those heuristics are, in market settings, "irrational". Further, they're not merely incompatible with market exchange and an understanding of complementarity of wants, etc. -- those heuristics lead to hostility against markets.

Once upon a time I wondered how or why this could matter, but I understand now it matters in terms of the most effective strategy for economic education (or Austrian Political Economy or what have you). If people are merely ignorant in some neutral fashion, give them a copy of The Use of Knowledge in Society or I, Pencil, or whatever and watch their minds change. I guess one option is to stick them in a room with Jeffrey Friedman for a few days. But guess what? That doesn't change the rational irrationality story at all. By locking someone in a room for a weekend with Hayek or Coase or Jeffrey Friedman or Bryan Caplan, you have, ahem, *raised the cost of irrationality* when it comes to their views on markets. The rest of the argument is little more (a bit more, but little more) than semantics. There is in some sense a demand for irrationality, and the only argument you're really making is *how* to raise the costs, not whether to.

I'll say this, your own approach is probably more market-oriented that some of Bryan's suggestions, like giving the CEA veto power. Interesting. BTW, if you're saying that Econ 101 doesn't do the job, you're basically right (I can imagine the *right* teacher with the *right* textbook and the *right* examples making a difference). I have no idea what step it is, but it seems to me the people to go after are not college freshmen, but scholars throughout the humanities and social sciences. Make them aware that the advantages of markets are not in terms of cold efficiency, but the enhancement of community and strengthening of peaceful bonds -- the kind of stuff that Virgil Storr writes about so eloquently. But don't throw out Supply and Demand with the hyper-rationalist bathwater.

When Dan Ariely or Robert Frank say people are irrational, they are not necessarily insulting them. They are saying that people's own perceptions and cognition often fail in predictable ways. Certainly such a thing can be said about the typical layman's views on economic issues.

"Make them aware that the advantages of markets are not in terms of cold efficiency, but the enhancement of community and strengthening of peaceful bonds"

The problem is that most experts in social theory completely disagree with this. Why such a casm between economists and social theorists? Are they rationally irrational? Simply ignorant?

Perhaps Mises had the better take on rationality. As I read him some years ago, he was saying that people are rational in the sense that they make choices in the pursuit of their own aims, given the resources at their disposal, given their knowledge, and GIVEN their ability to process that knowledge. (Checking my copy of Human Action, p. 20 seems to support this reading.) The opposite of rationality might be automatic reflexes. What is more often referred to as "rationality" might better be called "cognitive skill" -- skill which may not be entirely general in nature but might vary with circumstance. I'm not convinced, for example, that learning formal logic is the key to success in life.

One of the main reasons people make mistakes in business is long lag times between policy and effect. In business, lag times are much shorter, so business people expect something similar in macro economics. But the lags between fiscal or monetary policy and the effects can be four years to four decades. The lack of correspondence in time fools people and they look for events closer to the effect as the cause.

Of course, mainstream econ has an even harder time with lags because they have no time at all in their theories or equations. Everything happens instantaneously.

The main problem with the Stigler's argument quoted by Dr. Boettke is that it presumes that somehow the people as a collective really choose the social arrangement as a group of friends chooses where to go out at night.

But one doesn't even need to have an in-depth knowledge of history to understand how detached from reality this understanding is.

The main problem with the Stigler's argument quoted by Dr. Boettke is that it presumes that somehow the people as a collective really choose the social arrangement as a group of friends chooses where to go out at night."

Exactly. It simply assumes, without argument, the adequacy/legitimacy of the status quo social choice function/institutional arrangments as the accurate representation of the group's choice.

Wow - this discussion has grown!

Peter, from the 16th -
Yes, I think both Pissarides and Alchian are valuable on frictions. I don't think I've ever viewed Pissarides as saying that frictions are "negative" though, the way you've framed it. Search is essential to making good matches - that's part of the point of the theory. It explains unemployment, which is I guess what you mean by "negative", but I think it's a little misleading to valorize things like that. I suppose frictionless, perfect matches would be nice but I don't see the sense in comparing he real world to that fiction and then calling the real world "negative".

Regarding this: "The reason why I find Caplan unpersuasive on that particular issue (though I find him persuasive at many levels) is because he is just relying on the absorptive capacity of "mainstream" to insulate from paradigmatic challenge. There is nothing that I could say that he couldn't restate in terms he is comfortable with."

I think that's the whole point. That's precisely why we don't feel as if you're offering a paradigmatic challenge - because nothing you seem to offer strikes us as challenging our paradigm. It's just restating a problem we feel entirely comfortable with in new words. Parsimony is a feature, not a bug. Caplan doesn't go into nearly as much detail as Kirzner, it's true. But if I'm grasping around for a way to understand surprise and error in the economy, the parsimonious neoclassical answer seems to give me a lot more mileage than the extended Austrian hermeneutics. Until that ceases to be the case, there isn't much of a challenge to the paradigm to speak of.

Sorry, I'm not familiar with the bucket in statistics.

I'm also not quite sure what you're getting at with your question about Hayek on knowledge. Taking your 2002 RAE discussion of this issue, it seems to me that mainstream models do make the information/knowledge distinction that you consider a conflation. There's plenty of information in models, some of which is accessed or known by agents and interpreted subjectively, some of which is not. Mainstream economists may not know that you like to make this distinction, and so they may more casually talk about it all as "information", but the distinction seems to be made when it needs to be made. Hayek also points out that knowledge is revealed in the market process itself. Perhaps I'm misunderstanding your question, but I'm not sure what you think is lacking in the recognition of this point. Mainstream economists are often more interested in equilibria than you are, but if you think about non-static depictions of those equilibria, there seems to be a recognition of what I would agree with you on - that the market process generates relevant knowledge. I think about economic science as a science of the determinants of economic phenomena. Why do we see phenomena we see. What are their determinants? So equilibria are obviously relevant to talk about. I think you see economics more as a science of the process of the market itself. That's a different interest, but I don't think it follows that there's an under-appreciation of the market process from people who have different objectives.

I agree strongly on Vernon Smith. When I was at GMU for your summer workshop I got introduced to a lot of Hayek (before I was introduced to Keynes, actually!), and thought he had a hard time getting to his point. But Vernon Smith made the point very clearly and has been an important resource to me ever since.

Jeff -
re: "Daniel: If mainstream economists (Caplan having gotten his Ph.D. at Princeton) were fully comfortable with radical ignorance, why would they come up with such completely unsubstantiated (indeed, contradicted-by-evidence, e.g., the evidence of millions of people voting) notions as "rational irrationality" when they think about politics?"

Does the mainstream fall in with Caplan on this point? I guess I'm confused by the question. I think Caplan has done an excellent job highlighting why a lot of the critiques of neoclassicism by Austrians are way off base. I don't agree with a lot of other things he says,and I'm not sure I'd take him to be representative. The mainstream is not a monolith - the unity is more in the methodology than any single idea.

I also doubt Bryan would reject the idea of ignorance. He may disagree with you on how relevant it is (I may disagree with you on that too, in fact).

I guess I just don't think we should take an either/or approach to a question this multifaceted.

Sandy Ikeda -
re: "To form an expectation, E, about whether I will buy a blue cashmere scarf at Bloomingdales today, X, assumes that I know the probability of X and of W (buying a grey felt hat) and of Y, etc. So to form E(X) I must have a proper probability distribution that exhaustively lists all mutually exclusive states of the world. If there are states of the world of which I am radically ignorant then the list won't be exhaustive, I won't have a proper probability distribution, and so I can't form E(X)."

I don't think this is quite right, which explains Peter and Jeff's confusion about how radical ignorance fits into mainstream modeling.

If you are radically ignorant of a state of the world, your list won't exhaust the actual states of the world. But you will still calculate an E(X).

The crucial thing to remember in thinking about economic decision making is that these decisions are subjective. The neoclassical agent is NOT an omniscient statistician. He is an actor that acts on his subjective knowledge. E(X) may not include all states of the world, but it will reflect his subjective judgement of the states of the world. And that judgement will motivate an action. And that action will have consequences. And that judgement may or may not conform to reality. When it doesn't conform to reality, that's ignorance, error, and a mistake.

Neoclassical models aren't meant to study all the states of reality. They are meant to study human behavior. And humans act on subjective judgements that may not conform to actual states.

Jeff -
Yes, I have read Herbert Simon and bounded rationality, and I was first introduced to "bounded rationality" in Prof. Gerlach's introductory microeconomics class nine and a half years ago, and it's come up in every micro class since then. (I didn't actually read Simon until a later sociology course).

I really think a lot of people who were around when some of this stuff was newer genuinely don't appreciate that this is standard fare. The tools of classic incentives-based micro are still tools that no one has bested. But I've never been told agents are omniscient. I've never been told that rationality isn't qualified. It's true - we don't sit around and talk about the implications of bounded rationality all day. There are a lot of technical tools to master, and that's what time is spent on. But I don't think these illusions are out there in the way that a lot of heterodox commenters think they are.

Also - I've been rereading Evans and Friedman more carefully since this conversation, and I think what they call "surprise" a lot of mainstream economists call "shock". People have certain subjective expectations and act accordingly. There is some shock that does not conform to their expectations. They react. Shocks, of course, are all over the place in mainstream economics - both micro and macro.

It's like the information/knowledge distinction Peter makes that I noted earlier. I think non-Austrians think about both things, they just may not semantically make the same distinctions.

Daniel--Strictly speaking, the burden of proof is on me to show that mainstream neoclassical economists *in practice* treat agents as effectively omniscient. And strictly speaking, pointing to a mainstream neoclassical economist (Caplan) who in practice treats political agents as effectively omniscient (at the intuitive level, about economics) does not discharge this burden of proof. That, and my Critical Review article with Tony Evans about how information economics effectively treats agents as omniscient, is the best that I, as a political scientist, can do.

But I think you're getting on shakier ground when you suggest that Herbert Simon--a president of the AEA--didn't know what he was talking about when he made the same point.

Still, I wish we would hear from more Austrian economists about exactly how a recognition of radical ignorance would change the routine practices of economists.

Steve--I don't know why you think you are criticizing my argument. It was I who praised Caplan for originally allowing that economics is counterintuitive, even though this destroys his claim that people have to push this intuitive knowledge away to "indulge their taste for irrationality." Now you are hammering me on the grounds that economics is counterintuitive? Where did I say or imply that ignorance is random? People are not blank slates. Perhaps you should read my critique of Caplan in Critical Review. It goes into this subject in depth.

Nor did I provide a caricature Bryan's attitude toward those who "disagree with him" or "disagree with him and his colleagues." It just so happens that he castigates the public as irrational ("emotional and ideological," not merely "mistaken") on four issues where he agrees with the majority view of his peers. It would have been good for him to consider a case where he disagrees with his peers. In such cases is he, like the public, "irrational"?


I don't know if I am criticizing your argument or not! It just strikes me that you're lashing out at a strawman. Does Caplan say people are blank slates? A major thread of his research (and mine, obviously) rests on the notion that people are not blank slates. Where is the contradiction within Caplan? Of course I've read your entire issue on MRV, and to be honest I think we're getting much closer to the heart of your criticism here than I would from picking over your essay from that issue. Caplan is saying that people push away reason on economic issues the same way that Ariely would say people push away reason when buying a car. "Pushing away" is actually probably the wrong way to say it, and uncharitable to the nature of Caplan's argument. Something other than reason is indulged in. That something other is known as irrationality in everyday language. But as you seem to agree, those indulgences follow a rational pattern. You could say that Caplan is too blunt for his own good, but it could just as easily be said that you are too subtle for your own good.

It's interesting that you bring up the cases where Bryan disagrees with his peers -- I don't have MRV handy (it's in my office) -- but I believe you're right that he doesn't discuss that in MRV, and my recollection is that he doesn't discuss it in the EJ article on systematic bias, either. As I think about it, I'm not sure that has been published in any of his work on the SAEE, because it's beside the point. But, last I checked, it's actually a required exercise in his graduate classes: He gives examples of where he disagrees with the economic consensus from the SAEE (the importantce of education and job training to macro performance is an example) and challenges students to do the same. It would be hard to find an economist who agrees with the entire economic consensus, and yet there is a consensus. The four types of bias can either be viewed as arbitrarily chosen or as four clear patterns seen in the general public's view. As it happens, Caplan's findings for the general public's views in the SAEE can be replicated using the GSS (see our 2010 Intelligence article). I don't see how he went out of his way to cherry-pick the issues he agrees with. Like most economists (most of whom obviously do not share Bryan's ideological priors) he is in agreement with most of the economic consensus. Why? Because they're not merely mistaken, although they are mistaken, they are emotional and ideological. And they are emotional and ideological because they are not blank slates. The emotion and ideology lead to mistakes (sounds like Behavioral Economics, eh?). You say you're not sure why I think I'm disagreeing with you; I say I'm not sure why you think you disagree with Caplan or MRV. As far as I can tell, your only disagreement comes down to MRV's lack of charity. You object to the word "irrational" in this context because of its connotation. Okay. Do you have similar complaints about Behavioral Econ? The point is that, to take a somewhat "Austrian" example, spontaneous orders (or whatever people want to call them) are counter-intuitive. And to change people's thinking on order, planning, design, and coordination, you have some biases to overcome. You have stupidity to overcome. You have emotion to overcome. You have ideology to overcome. Those things get in the way, and you generally have to raise the costs of those things (emotional and ideological priors) to make a hole big enough for economically literate insights to penetrate.

"Claim: if you can write it down in an equation, it's not radical ignorance.


I'm afraid I have not been following this discussion so much. But I did want to comment on Steve's remark about equations. I think it is mistaken. I just gave a paper at the SEAs in which I use the "Brazilian beta" (as I called it) to model discretionary monetary policy. I lifted the trick directly from
Tsuji, M., daCosta, N.C.A., Doria, F.A., 1998. The incompleteness of theories of games. Journal of Philosophical Logic 27, 553–564.

This "beta" is zero or one depending on the solution to an (Turing) undecideable question. I have the authorities picking a high or low inflation depending on beta. Thus, you have a situation in which the high or low rate is not a matter of probability, is not computable, but is "determinate" in some vague metaphysical sense. That's not everything you might include in the category "radical ignorance," but some. And I think similar techniques borrowed from computabilty theory can get you other aspects of "radical ignorance." It is *classical analysis* (such as Bourbaki math) that might be incapable of representing radical ignorance in a satisfying way, not "math" or, as in the quote here, "an equation."

"Whatever public preference for government actually is, it can't evolve as quickly through the political system as the public lunch-preference does in the market." - Josh S

Well...unless you allowed taxpayers to directly allocate their taxes...

Friedman, if you haven't already seen my recent comment over at Koppl's oldish blog would you differentiate your argument from Buddha's argument that we are all just blind men touching different parts of an elephant?

The comments to this entry are closed.