Dan Phiffer - Latest Comments in Dan Phiffer - Rationality vs. Economics

Re: Dan Phiffer - Rationality vs. Economics

aesylvatica — Thu, 24 Apr 2008 15:39:01 -0000

My apologies if this is going to be a repeat response, I don't have time to read all 46 comments at the moment, but thought I may have something to add....

I see the problem with his argument as two fold:

1. Traditional economics is grounded in the assumption that humans make decisions that are rationally based, but most don't. I don't mean this in some condescending way; I mean it very literally. Only 30% of the world's population (but 50% of the people in power) have developed to a stage where they can access the tools of rationality. Before that stage is reached, people just simply haven't developed the capacity to use rationality as a tool. Before people develop the capacity for rational though, they rely mostly on mythic thought processes: religion, nationalism, 'moral majority' (40% pop; 30% power)... Of course, the people who developed economics are capable of rational thought and it seems 'rational' to them that other people would behave 'rationally,' since that's 'rational.' However, that's just not the case. An 'economics' based on religious or other mythic-membership decision-making would be a more true reflection of the way in which the majority of humanity makes decisions.
(This is more of a critique of economics broadly than his argument specifically.)

2. Now more to the specifics of his argument (I needed to lay that groundwork first). One of the main challenges that I see with relying on 'experts' is that they are most often aligned with majority thought. It seems to me that if you've managed to stick with academia long enough to become a certified 'expert,' then you most likely either agreed with their point of view to begin with, or you've been converted long ago. (Given, there are always some rebels inside of any institution, but they are generally few and far between because it is such a painful place to be if you have a different belief.) The challenge with believing non-experts is that it is hard to distinguish between the nuts and the visionaries (and sometimes they are the same people!) There are pre-majority beliefs and post-majority beliefs in circulation, and it can be challenging for the average Joe to tell the difference. So, it is probably 'safer' for them to get their facts from an expert. However, it you're looking for cutting edge, paradigm-changing ideas, than experts are the last place you want to look. If you are undertaking your own education, you have to be more vigilant about what you take in as truth. It requires a lot more work, but can be a lot more rewarding.

Thanks.

Re: Dan Phiffer - Rationality vs. Economics

Abe — Mon, 17 Mar 2008 13:01:33 -0000

That's a really nice summation of the libertarian faith right there!

I'd agree that governments tend to be inefficient, but I disagree with libertarians on two points. One I don't think the inefficiency is tied to being a government per se, but to the fact that governments are almost always bureaucracies. Governments tend to be larger, more entrenched and more tied to the land then other large bureaucracies like corporations, but they are prone to the exact same failures. Two, I'm not sure that inefficiency is always a bad thing, but in fact it might just provide a necessary degree of stability and slowness to the world...

As for markets being efficient..... Ha, for one you can look at Bear Sterns... Established markets in calm waters, yeah they are pretty efficient, but a quick look at economic history shows that markets are prone to all sorts of bubbles, panics and crashes that are as far from efficient as you can get. In addition there is a huge issue of survivor bias when looking at markets, yes markets that survive need to have a degree of efficiency, otherwise why should they exist? But the world is littered with failed markets and potential markets, that are not efficient at all.

The of course there is the whole issue that markets no matter how efficient they are, need governments to provide the stability they need to exist. Without governments market transactions get trumped by guns at every maneuver. Not exactly efficient is it?

Re: Dan Phiffer - Rationality vs. Economics

Steve Roth — Mon, 17 Mar 2008 12:13:35 -0000

Alex J.: "I think you'd find that ongoing tax rates of 100% would have an effect on a country's long term growth."

Do I really have to explain what a profoundly meaningless argument that is?

Re: Dan Phiffer - Rationality vs. Economics

Alex J. — Mon, 17 Mar 2008 11:13:06 -0000

I think you'd find that ongoing tax rates of 100% would have an effect on a country's long term growth.

Re: Dan Phiffer - Rationality vs. Economics

Steve Roth — Fri, 14 Mar 2008 20:02:26 -0000

Getting back to the book, I have read it (not every page yet, but most) and keep coming back to Chapter 3, which underpins the whole rest of the book.

Basic argument:

o The main argument against economists and their recommendations is that the economists are biased. (Self-serving based on economic and other conditions, indoctrinated based on previous economists.)

o Through some regression analysis comparing economists to non-economists, we can determine that they're not biased as charged.

So, absent any other evidence, we should conclude that they're (mostly) right.

Maybe you've noticed the problem here" Caplan says it himself (p. 81):

"If A and B disagree, there are three logical possibilities. The first is that A is right and B is wrong. The second is that A is wrong and B is right. The third is that both A and B are wrong. But we can rule out the possibility that both A and B are right." (He's talking about economists and non-economists here.)

Okay first, he doesn't mention an important case: what if A and B agree? What are the possibilities? They're both right or they're both wrong.

So of the five possibilities, ceteris parabus, the odds are:

Both right: 20%
One right: 40%
Both wrong: 40%

Allow me to address the latter case, which arises in the very first SAEE question, which asks (condensed) "here is a list of reasons for why is the economy not doing better than it is. Is each not a reason at all (0), a minor reason (1), or a major reason (2)?"

The first reason: "Taxes are too high."

Results:

Public 1.5
Economists .75
"Enlightened" public: .95

The Enlightened Public's answer lies between the economists' and the public's. According to Caplan, this means that "the truth lies somewhere in the middle."

Okay here's the problem: they're both wrong. (though the economists are significantly *less* wrong.) The consensus among economists who have studied the subject is that the level of taxation (and the size of government in general, for that matter), has no effect on long-term growth in developed countries. It's a myth.

http://trueconservative.typ...

Why do the economists get this wrong, even though the literature is clear, extensive, and pretty much unopposed (by real economists)? Must be bias. Here's a great example, in one mighty impressive economist, Greg Mankiw:

http://trueconservative.typ...

And that must mean--based on Caplan's analysis--that the whole society is biased to believe this thing.

It sort of casts a pall over the rest of the book.

for various criteria, popular

Re: Dan Phiffer - Rationality vs. Economics

Alex J. — Fri, 14 Mar 2008 18:07:18 -0000

I'll take your word that your beliefs held via faith, but you'd need to provide some evidence for me to believe that Caplan's are.

Re: Dan Phiffer - Rationality vs. Economics

AMW — Fri, 14 Mar 2008 13:19:38 -0000

As long as the uninformed are not systematically incorrect you will still get good results, because the ignorant-but-lucky will cancel out the ignorant-and-wrong, and the informed will win out.

The main thrust of Caplan's book is that when it comes to economic policies, the uninformed are very systematically incorrect, so you won't get good results.

Re: Dan Phiffer - Rationality vs. Economics

josh — Fri, 14 Mar 2008 12:38:45 -0000

Futurarchy. But this is purely utopian. The problem with your suggestions is that the systematic poor judgment of voters is natural, and essentially uncurable unless it becomes worth a persons while to become an expert. This would require changing the way we make collective decisions.

This best thing we can and should do is fight to leave as many decisions as possible to individuals.

Re: Dan Phiffer - Rationality vs. Economics

josh — Fri, 14 Mar 2008 12:28:02 -0000

Well, first you identify he problem (eg voters could do a much better job), then you think of solutions that are within the principles of justice. Caplan is only doing the first bit.

Re: Dan Phiffer - Rationality vs. Economics

josh — Fri, 14 Mar 2008 12:24:34 -0000

I think Hanson's futurarchy idea really does get to the heart of the issue. Would anyone not expect the prediction markets to come much closer to the "experts" opinions when it comes to maximizing the voters chosen utility function than does the result of a typical election. If this is the case, than essentially Caplan's is correct, experts>median voter (although still worse than prediction markets) when it comes to actually meeting some predetermined goal.

Re: Dan Phiffer - Rationality vs. Economics

josh — Fri, 14 Mar 2008 12:17:51 -0000

I agree that this is the obvious conclusion.

Re: Dan Phiffer - Rationality vs. Economics

RobD — Fri, 14 Mar 2008 11:26:11 -0000

I know that this may be slightly off topic, but as a blogger who tracks different "wisdom of the crowds" experiments, I couldn't help but notice what's going on here. Great stuff. As an amusing anecdote, I'm actually planning a wedding as my own little test case, where "the crowd" is making all of our wedding planning decisions for us. Would love any feedback:

http://thenerfherder.blogspot.com/2008/03/wedding-to-test-wisdom-of-crowds-theory.html

Re: Dan Phiffer - Rationality vs. Economics

Grant — Thu, 13 Mar 2008 18:56:37 -0000

For anyone who isn't convinced of Bryan Caplan's arguments, consider the efforts needed truly be able to predict the effects of one political candidate being elected versus another. Consider all the economics and history a voter needs to know. Consider how incredibly difficult (impossible?) it is for anyone to understand all the effects of even one of the bills passed by congress. Consider how hard it is to judge what a candidate will actually do once elected. Consider how much more complicated the effects of legislation become year after year, as technology and society grow in complexity.

Who would go to all this work? Following politics would be an 80-hour a week job.

More voices are good when those voices have incentives to provide good information (e.g., prediction markets). But if the uninformed and incorrect have the same incentives to share their information as the experts, you don't get good results.

Or put another way, who do you want designing airplanes and skyscrapers, professional engineers or voters? Do we vote on the aluminum alloys used in airframes? Why would we think "engineering" society would be any different?

(I don't mean to be arguing for any sort of aristocratic elitism here, I'm just trying to point out the differences in the decision making of market processes vs. democratic processes)

Re: Dan Phiffer - Rationality vs. Economics

AMW — Thu, 13 Mar 2008 17:12:26 -0000

Oh, and note also that experts accept both the theory of gravity and the theory of evolution. Non-experts are more likely to reject the theory of evolution. Make of that what you will.

Re: Dan Phiffer - Rationality vs. Economics

AMW — Thu, 13 Mar 2008 17:09:29 -0000

I tried to post this over at econlog, but I'm having trouble with their server. Also, it's directed to you, so I might as well post it here where you are more likely to see it.

Dan Wrote: "Also, how would you guys respond to commenter Abe's claim that there's an inconsistency between trusting non-experts in the marketplace but not in the voting booth?"

My response:

Dan, I posted a short reply to Abe on your blog. It all comes down to incentives. You may be surprised to hear this (and I'm not being sarcastic), but economists are generally pessimistic about how markets will perform when there are externalities (like, say pollution). In those cases markets are not predicted to be perfectly efficient. Same goes with markets with asymmetric information (like used cars, health care and plubming).

So economists are willing to grant that people will behave badly in markets under certain situations. (Economists are generally of the mind that people will behave badly whenever it is in their interest.) But the question isn't just "do markets fail?" The question is "do they fail worse than government addressing the same issue?" And in most cases the answer is no.

There is an enormous existing body of literature called Public Choice theory that explores government failure. Public Choice generally assumes that people are rational, but their desire for good economic policy is thwarted by lobbyists (in a nutshell), so we end up with bad policy. Caplan goes further and argues that voters are themselves irrational when it comes to abstract economic ideas, so we don't even need the lobbyists to screw up economic policy.

So back on topic, the asymmetry between behavior in the voting booth and in the market is that in the latter there are strong incentives not to be irrational. If you, as an individual, make systemmatically stupid decisions in the market, you will first go bankrupt, then go hungry. Since so much is on the line, people tend to wise up quickly. But you, as an individual, have virtually zero impact on what happens in government through your vote. So you have nothing to punish you individually for making stupid decisions at the voting booth. Hence, irrationality at the voting booth tends to persist, while in the market it tends to disappear (or at least shrink).

Caplan likes to draw the analogy with the theory of gravity and the theory of evolution. People who consistently reject the theory of gravity tend to be heavily punished by their own actions. People who consistently reject the theory of evolution suffer pretty much no ill effects as a result. Lo and behold, very few people reject the theory of gravity, and loads of them (in the U.S. at least) reject the theory of evolution.

I hope this comment is helpful (if a bit long).

Re: Dan Phiffer - Rationality vs. Economics

AMW — Thu, 13 Mar 2008 16:45:05 -0000

Markets achieving efficient outcomes and governments achieving inefficient outcomes are not matters of faith for libertarians. They are stylized facts well butressed by both theory and empirical evidence. People are people; they are generally self-interested and will exert the most effort to make themselves and those they are closest to better off. Markets are generally good at channeling this self-interest into socially beneficial behavior, (activist) governments are generally good at channeling it into behavior that is socially neutral at best and detrimental at worst. To be sure, free rider problems and externalities throw a wrench in the system for markets. But even in these instances minimal government involvement is all that is called for (to tax or subsidize).

Re: Dan Phiffer - Rationality vs. Economics

Silas — Thu, 13 Mar 2008 14:58:27 -0000

Apparently, Dan thinks this is still a novel response,

http://econlog.econlib.org/...

despite this alleged inconsistency having been resolved several times above on this very page. To repeat: voters bear none of the cost of irrational voting, while they bear the full cost of irrational market choices.

Who's wasting whose time here? I don't like sounding rude, but even more, I don't like people who claim to want to hear ideas and then ignore them.

Re: Dan Phiffer - Rationality vs. Economics

Ajay — Thu, 13 Mar 2008 13:12:02 -0000

Exchange of knowledge is the defining feature of the internet, yes, but social capital will always be important in that process. For example, you can currently go read on the internet that linux is the best thing since sliced bread and blows away all other operating systems and you can also read the converse. The only way for a layman to know which fact is probably true is by knowing who to trust. You're right that a new decentralized editorial process will spring up to provide this reputation currency on the internet, an editorial filter that wasn't as necessary before when there was a lot less information being communicated (Google is a first hack at a more automated version of this editorial filter). The nice thing is that since this filter is much larger and widespread than the previous filters of peer review or getting published, there will be many more people filtering at varying levels of expertise. You don't have to slog away for decades as an editor at The Economist or get tenure to get to your status (link blogs are an early version of this, where people post their best links and you get to use them as your filter). As for your notion of intuitive judging, I don't think that's what will happen. I think you're just saying that to try and justify not having any formal editorial filter at all, but I think there will be one, it will just be much bigger, decentralized, and quicker for people to start contributing to than society's current filters, sort of like google search is today. :)

Re: Dan Phiffer - Rationality vs. Economics

Abe — Thu, 13 Mar 2008 11:35:02 -0000

Dan man, don't waste your time.

As best I can tell Caplan is just trying to work out a libertarian paradox, that they think masses are doing the right thing when it comes to markets and the wrong thing when it comes to government. It's only a relevant issue if you buy into two of the key tenets of the libertarian faith, that "markets" (or "the market") are good and that government is bad. If you don't believe both those things, ignore the fucker. Otherwise you're gonna be stuck in essentially a religious argument...

Re: Dan Phiffer - Rationality vs. Economics

Dan Phiffer — Thu, 13 Mar 2008 11:14:43 -0000

I think my argument (which isn't fully baked) might be that on the Internet, the exchange of knowledge can be a powerful force, independent of social capital (which helps identify trustworthy sources). The scarcity lies in attention, as you point out, rather than published works. In this environment the editorial process (collective or individual) becomes relatively more valuable. Maybe I'm talking about a different kind of expertise (skill?) that works to identify experts intuitively rather than rationally. This idea "feels" right vs. this idea is logically sound.

Re: Dan Phiffer - Rationality vs. Economics

Ajay — Thu, 13 Mar 2008 10:40:02 -0000

I think you're trivially right that the internet will lead to a lot more people being able to build reputations online and take part. You're wrong however, if you take that to an extreme and say that EVERYONE or even a large majority of people will contribute expertise to a significant degree. Caplan is right in that only a minority of people ever wants to take the time to become expert and that reputation will always be important, but the internet will allow many more people to take part in these markets by making them more efficient and allowing the work to be split up many more ways. For example, rather than going to grad school and getting a PhD in economics, people will be able to do independent research on their own, blog about it, and build up a reputation through their writing. Also, others will be able to easily share small snippets of data that they have momentarily become experts in, say on the best digital camera below $500 for prosumers, because they just researched it for themselves or someone paid for them to. People can already do this but they mostly do not get paid for it yet. They will.

Re: Dan Phiffer - Rationality vs. Economics

Dan Phiffer — Thu, 13 Mar 2008 10:34:06 -0000

Aw, Silas, you're being that guy! Don't be that guy...

You may be correct that I'm not fully grasping Caplan's argument (and I've already committed to reading his book to clarify any misconceptions), but I don't intend this to only be a referendum on The Myth of the Rational Voter (you'll notice I said "what response would *you* advocate," not which Caplan would). Surely there are other valid reactions in addition to the one you outlined above. Does media literacy have no place in this discussion?

Re: Dan Phiffer - Rationality vs. Economics

SilasX — Thu, 13 Mar 2008 00:03:29 -0000

Dan, you're still failing to understand the distinction between "uninformed" (ignorant) and "irrational", which I have explained twice now. Either state why you disagree with Caplan's argument that irrationality exists and is the key problem, or stop talking about how you're going to cure the problem of ignorance.

Several people have, at this point, already given you excellent, concise explanations of Caplan's work that apparently were not represented in his talk. With all due respect, I do not think you have made a similar effort in learning from them and formulating a response that articulates where and why you disagree. Please do so.

Re: Dan Phiffer - Rationality vs. Economics

another bob — Wed, 12 Mar 2008 23:12:45 -0000

Here's an irrational voter (David Mamet, no less) explaining why his choices are less irrational now.

http://www.villagevoice.com...

Bryan's argument (I've read the book but only once) is neatly summarized in his book - only about 5 bullets - would fit on one powerpoint slide.

Plebes have systematically inaccurate economic beliefs (when compared to economists) possibly hardwired as a result of our past as hunter-gatherers.

Plebes have little incentive to change their own beliefs. It is costly (in time and emotional certainty) to become an economist. Plebes who do educate themselves garner no benefit (only one more smart vote among zillions of idiots).

Bryan might say that we don't expect everyone to be a lawyer. We delegate that to courts. We can't expect everyone to be an economist, let's have a supreme court of economics to invalidate truly idiotic policies.

the only thing wrong with democracy is the demos.

how 'dat?

Re: Dan Phiffer - Rationality vs. Economics

Dan Phiffer — Wed, 12 Mar 2008 20:25:21 -0000

But what does this elitism imply policy-wise? That is, what response you would advocate in light of these deficiencies? My response isn't to tout the qualities of the better-informed but rather to think about ways to inform people about their candidates. Some of this has to do with media literacy, and some of it is a cultural argument -- you might say I'm in favor of Anti-anti-intellectualism. (Google informs me there is already Anti-anti-anti-intellectualism.)