Imperfect Reasoning (was: reading recommendation)
Bryan Caplan wrote: At least on my reading, a lot of cognitive psychologists want to say more than "People occasionally reason imperfectly, and policy might improve on that." Rather, they are saying "We now know that human judgment is quite poor, and economic models that presume otherwise are kind of stupid." Of course, it depends on who you read, but I think this triumphalist message appears in Nisbett and Ross, Kahneman and Tversky, Thaler, Rabin, and others. These guys rarely stray into policy, but they clearly think their work is cosmically important. Bryan calls attention to the issue of how large or frequent are cognitive errors, suggesting that standard economic analysis is reasonable when they are rare or small. To me the central issue is instead human meta-rationality. If cognitive errors make workers sometimes miss-estimate the safety of a job, but workers realize that they might make such errors, then wiser-than-thou academics just need to *tell* workers that their particular job is more or less safe than they realize, and that should fix the problem. If people have time-inconsistent preferences, but realize this fact, then it can be enough to give them means to commit to future choices. If people can neglect possible ways a contract can go bad, but realize this fact, they can give arbitrators discretion to deal with this when settling contract disputes. In contrast, those who see large policy implications from imperfect reasoning tend to assume that people are not meta-rational. This may be true, but most of the evidence presented just show cognitive errors, and is silent on the issue of meta-rationality. Robin Hanson [EMAIL PROTECTED] http://hanson.gmu.edu Asst. Prof. Economics, George Mason University MSN 1D3, Carow Hall, Fairfax VA 22030- 703-993-2326 FAX: 703-993-2323
Re: reading recommendation
Bryan wrote: === This is almost orthogonal to my original point, but not quite. It wouldn't be interesting if the expected cost of bad judgment was $100/year, would it? So even taking a policy perspective, expected value of error matters. Agreed, but I think the evidence here is that cognitive anomalies can cause big losses. Sure the vast majority of the time confusing .01 with .1 doesn't make much difference in ones life, but if people consistently confuse the two when making decisions about (for example) occupational safety and health, that can have big effects in the aggregate the might be addressed by policy. -- Bill Dickens William T. Dickens The Brookings Institution 1775 Massachusetts Avenue, NW Washington, DC 20036 Phone: (202) 797-6113 FAX: (202) 797-6181 E-MAIL: [EMAIL PROTECTED] AOL IM: wtdickens
Re: Imperfect Reasoning (was: reading recommendation)
Robin Hanson wrote: To me the central issue is instead human meta-rationality. If cognitive errors make workers sometimes miss-estimate the safety of a job, but workers realize that they might make such errors, then wiser-than-thou academics just need to *tell* workers that their particular job is more or less safe than they realize, and that should fix the problem. If people have time-inconsistent preferences, but realize this fact, then it can be enough to give them means to commit to future choices. If people can neglect possible ways a contract can go bad, but realize this fact, they can give arbitrators discretion to deal with this when settling contract disputes. In contrast, those who see large policy implications from imperfect reasoning tend to assume that people are not meta-rational. This may be true, but most of the evidence presented just show cognitive errors, and is silent on the issue of meta-rationality. I think the reason they ignore it is that they think it is too far-fetched to be worthwhile studying. Have you got any empirical evidence to overcome that burden of proof? One thing that interests me: You might call my notion of "rational irrationality" a form of meta-rationality. But my model doesn't imply that you can make people compensate for their irrationality by pointing it out to them; the only thing that works is raising price of error. -- Prof. Bryan Caplan [EMAIL PROTECTED] http://www.gmu.edu/departments/economics/bcaplan "We may be dissatisfied with television for two quite different reasons: because our set does not work, or because we dislike the program we are receiving. Similarly, we may be dissatisfied with ourselves for two quite different reasons: because our body does not work (bodily illness), or because we dislike our conduct (mental illness)." --Thomas Szasz, *The Untamed Tongue*
Re: Imperfect Reasoning (was: reading recommendation)
Bryan Caplan wrote: ... If people have time-inconsistent preferences, but realize this fact, then it can be enough to give them means to commit to future choices. If people can neglect possible ways a contract can go bad, but realize this fact, they can give arbitrators discretion to deal with this when settling contract disputes. I think the reason they ignore it is that they think it is too far-fetched to be worthwhile studying. Have you got any empirical evidence to overcome that burden of proof? People talk a lot about their difficulty in committing to long term plans. They choose savings plans that they can't get out of. They take efforts to avoid being around tempting candy bars. People choose contracts that give arbitrators discretion. People talk a lot about various irrationalities that they might fall into and ways they try to compensate for that. People talk about realizing that each person tends to think highly of him/herself, and trying to compensate for that. How is this so far-fetched? Robin Hanson [EMAIL PROTECTED] http://hanson.gmu.edu Asst. Prof. Economics, George Mason University MSN 1D3, Carow Hall, Fairfax VA 22030- 703-993-2326 FAX: 703-993-2323
Re: Imperfect Reasoning (was: reading recommendation)
Robin Hanson wrote: People talk a lot about their difficulty in committing to long term plans. They choose savings plans that they can't get out of. They take efforts to avoid being around tempting candy bars. These look more like conflicting preferences to me than "meta-rationality." But I'll agree that we see a fair amount of this. People talk a lot about various irrationalities that they might fall into and ways they try to compensate for that. People talk about realizing that each person tends to think highly of him/herself, and trying to compensate for that. People "talk a lot" about this?! Maybe in a few odd sub-cultures. I can't recall any family member every talking this way, for example. Maybe you're meta-rational, but I can't think of anyone else who resembles you in this way. :-) -- Prof. Bryan Caplan [EMAIL PROTECTED] http://www.gmu.edu/departments/economics/bcaplan "[W]hen we attempt to prove by direct argument, what is really self-evident, the reasoning will always be inconclusive; for it will either take for granted the thing to be proved, or something not more evident; and so, instead of giving strength to the conclusion, will rather tempt those to doubt of it, who never did so before." -- Thomas Reid, _Essays on the Active Powers of the Human Mind_
Re: reading recommendation
Let me second Bill's point. It's because decision heuristics are usually so useful that we can be lulled into following them when doing so is downright irrational! Alex -- Dr. Alexander Tabarrok Vice President and Director of Research The Independent Institute 100 Swan Way Oakland, CA, 94621-1428 Tel. 510-632-1366, FAX: 510-568-6040 Email: [EMAIL PROTECTED]
Re: reading recommendation
Regarding the exchange between W. Dickens and B. Caplan over decision heuristics: __ I encourage you to browse, if you haven't, the following: FAMA, EUGENE F., "Market Efficiency, Long-term Returns, and Behavioral Finance," The Journal of Financial Economics, 49 (1998). I keep a copy of the article at home and enjoy revisiting it from time to time . . .just as I enjoy watching a favorite comedy over and over again on television. In this article, which grew out of Fama's Hyde Park on-campus philosophical tussles with Richard Thaler and Company, we reach a very interesting conclusion. In short it goes something like this: [Yes, there are behavioral anomalies within investment markets. In fact, they often appear quite systematic and predictable. Furthermore, they are tenacious little "buggars" (that's my paraphrase). However, they'll cancel each other out in an efficient market.] Yes, that was the final answer after 40+ pages of statistical analysis and careful consideration: They'll cancel each other out. One morning, in the not-too-distant future, the folks in Stockholm will call Eugene in the wee hours to inform him that he'll be splitting a $1 million check with Kenneth French. I will smile over my morning coffee when I hear the news and applaud an incredible body of work and lifetime of academic contribution. They'll do so (in my opinion) in spite of that 1998 article. New York, NY
Re: reading recommendation
Sure, some important real world applications exist. But why is that interesting? I would think that the interesting question is: what's the *expected value* of the loss, averaging over situations of all importance levels? So would you argue that the interesting question about government policies is whether averaging over all of them net welfare effects are positive? Wouldn't you want to know something about particular policies with an eye towards identifying which ones are better or worse. Even if you thought that on average they were bad you probably couldn't convince most people that they shouldn't be considered on a case by case basis. Similarly, if you can identify even one situation in which judgement can be shown to fail and design an intervention to minimize the cost of that failure isn't that interesting? -- Bill Dickens William T. Dickens The Brookings Institution 1775 Massachusetts Avenue, NW Washington, DC 20036 Phone: (202) 797-6113 FAX: (202) 797-6181 E-MAIL: [EMAIL PROTECTED] AOL IM: wtdickens
reading recommendation
I just started a book that is so promising I'll recommend it after the first chapter: *Stereotype Accuracy* edited by Lee, Jussim, and McCauley. Basically they have a lot of chapters testing the accuracy of stereotypes, and mostly conclude that (in their words) "Stereotypes have been stereotyped!" They argue that in realistic situations, people using stereotypes are quite cognitively able. This provides concrete evidence that the pronounced cognitive biases emphasized in behavioral economics are to a large extent artifacts of unrealistic experimental conditions. More comments will be forthcoming once I finish the book. -- Prof. Bryan Caplan [EMAIL PROTECTED] http://www.gmu.edu/departments/economics/bcaplan "[W]hen we attempt to prove by direct argument, what is really self-evident, the reasoning will always be inconclusive; for it will either take for granted the thing to be proved, or something not more evident; and so, instead of giving strength to the conclusion, will rather tempt those to doubt of it, who never did so before." -- Thomas Reid, _Essays on the Active Powers of the Human Mind_
Re: Reading Recommendation with a Comment
Bill's comment was interesting. The main thing I got from Harris is that you *can* affect how much fun your kid has while he/she is around you, and how much they like you in the future. As she says, you don't spend time with your spouse in order to "mold" her, but to have fun together; why would kids be any different? -- Prof. Bryan Caplan [EMAIL PROTECTED] http://www.gmu.edu/departments/economics/bcaplan "We may be dissatisfied with television for two quite different reasons: because our set does not work, or because we dislike the program we are receiving. Similarly, we may be dissatisfied with ourselves for two quite different reasons: because our body does not work (bodily illness), or because we dislike our conduct (mental illness)." --Thomas Szasz, *The Untamed Tongue*
Re: Reading Recommendation with a Comment
William Dickens wrote: However, Harris goes on to say that there are appreciable "neighborhood effects." It doesn't matter if your dad stays around, but if all kids in a peer group lack dads, this makes them all (environmentally) worse off. I don't think JRH would agree with this. She thinks there are neighborhood effects, but they don't come from parent to child. They are transmitted child to child. I suspect she would say that keeping fathers with families wouldn't do anything to improve children's neighborhood culture. I don't have the book with me, but in elaborating her model, she repeatedly concedes that in many cases we have: all parents -- all kids Rather than: each set of parents -- their own kids That's why immigrant kids speak English mostly with the accent of their friends' American-born *parents*. If kids were a closed system, they would be developing much more distinct kid-only dialects on their own. And she does favorably cite this study showing that it is bad for kids to grow up in fatherless neighborhoods, even though it isn't bad for them to grow up in fatherless homes. If there's a textual dispute I can confirm this at home tonight. -- Prof. Bryan Caplan [EMAIL PROTECTED] http://www.gmu.edu/departments/economics/bcaplan "We may be dissatisfied with television for two quite different reasons: because our set does not work, or because we dislike the program we are receiving. Similarly, we may be dissatisfied with ourselves for two quite different reasons: because our body does not work (bodily illness), or because we dislike our conduct (mental illness)." --Thomas Szasz, *The Untamed Tongue*
Re: Reading Recommendation with a Comment
All three of the other admitted readers of the Harris book - Bill, Robin, and Alex - are parents. Has this in any way affected your "parental investment" or anything like that? My own take: The behavioral genetic literature shows that while the *marginal* benefits of parenting are less than you might think, the *average* benefits may well be higher. *Assuming* that you like your genes (and your spouse's), we know that your kids will resemble you to a high degree no matter what you do. That's a big fixed benefit. But the MB of shared environmental influence is smaller. It all depends on parameters, but I think the net message of this literature is that it's better to have kids than you thought, but not so useful to spend your time preaching at them. -- Prof. Bryan Caplan [EMAIL PROTECTED] http://www.gmu.edu/departments/economics/bcaplan "We may be dissatisfied with television for two quite different reasons: because our set does not work, or because we dislike the program we are receiving. Similarly, we may be dissatisfied with ourselves for two quite different reasons: because our body does not work (bodily illness), or because we dislike our conduct (mental illness)." --Thomas Szasz, *The Untamed Tongue*
Reading Recommendation with a Comment
*The Nurture Assumption* by Judith Rich Harris is very good. The main thesis is that genetics can explain almost all of parent-child similarity, so parents have little environmental effect on their kids. The people who actually do additional "molding" of personality, etc., are kids' peers. One point of interest to economists: Harris harshly critiques studies showing that one-parent families are worse for kids than two-parent families, chastising Quayle for his remarks on Murphy Brown. On Harris' account, one-parent families are disproportionally comprised of disfunctional people, who pass on disfunctional genes to their kids. One supporting piece of evidence: Kids whose parents died in accidents look a lot better than kids whose parents left - presumably because there is a much stronger genetic component to leaving your kids than accidentally dying. However, Harris goes on to say that there are appreciable "neighborhood effects." It doesn't matter if your dad stays around, but if all kids in a peer group lack dads, this makes them all (environmentally) worse off. Now Harris doesn't draw any implication from this, but it should suggest to any economists that Quayle was actually righter than he knew. On Harris' account, fathers don't help their own kids much by staying, but they do help all of the neighborhood kids a little bit through this peer channel. Thus, it's easy to see how families start breaking up - most of the costs are externalized on neighboring kids anyway. Family breakup hurts not so much your own kids, but the proverbial "fabric of society." There is a lot of additional food for thought, economic and otherwise, in this fine book. -- Prof. Bryan Caplan [EMAIL PROTECTED] http://www.gmu.edu/departments/economics/bcaplan "We may be dissatisfied with television for two quite different reasons: because our set does not work, or because we dislike the program we are receiving. Similarly, we may be dissatisfied with ourselves for two quite different reasons: because our body does not work (bodily illness), or because we dislike our conduct (mental illness)." --Thomas Szasz, *The Untamed Tongue*
