October 18, 2002
More on the Economics Nobel Prize

A nice against-his-will summary by John Irons:

ArgMax Economics Weblog: Clarifying the Nobel Prize: I have, a few times, written about the research that won Nobel Prizes for economists. I am currently suppressing the urge to again summarize/translate the work of this year's winners, Daniel Kahneman and Vernon Smith, into a form that non-economists can understand. I'm being tested though. I've read more than one account of why they won the prize, and I'm not really happy. The problem with doing a summary is that either you (the reader) are (1) an economist and likely know their work and don't need a summary, or (2) a generally intelligent person who will read the summary and think, "gee, that's obvious, I knew people don't behave the way economists assume, they won a Nobel for that?" or (3) bored to death. The problem with not doing a summary is that the only summaries will be done by people who don't know what they're talking about. (Very few people who know what they're talking about write about economic Nobel prize winners - there is a very low signal to noise ratio when it comes to economic writings.)

(Warning: gratuitous name-dropping!) I was fortunate enough to meet both winners several years back: Kahneman at a two-week graduate-level behavioral economics workshop in Berkeley, and Smith at a two-week graduate-level experimental economics workshop in Arizona. So I do feel somewhat qualified to do some summarizing. However, the Nobel Prize committee has already done that reasonably well.

If I were to write something, these would be some points I would try to make...

1. Econ Nobels are more often than not awarded to "fields" - experimental and behavioral economics are very much "teams" in the sense that many, many people contributed to the foundations of the field. Amos Tversky was explicitly mentioned in the press release, but there are also many other people working away in these fields which have done very significant work. Smith and Kahneman were definitely leaders though.

2. Modern economists are aware of the work of Smith and Kahneman and dozens of their coauthors. We know their results, and that people are not always rational. You can't use the prize and their research to ignore the rest of the field or economists in general.

On Rationality...

1. A paper by John Miller and Scott Page has a nice quote about how there is only one way to be rational, but many ways to be non-rational. It's not enough to say that people are not always "rational" and thus ignore economic results. A more positive approach would be to specify *how* people are not rational, find the implications, and compare to a fully rational world. This isn't easy -- just ask theorists who deal in bounded rationality, or even those who deal with limited irrationality (myopia, procrastination, loss-aversion). The math is usually harder when you don't assume rationality.

2. Smith and Kahneman's work can be seen, in part, as trying to figure out the "how people are not rational." My only argument with some of their work is that the harder part -- figuring out what specific behavioral quirks mean for the behavior of markets/economies -- is often left undone (although people are working on this, especially with the "wind-tunnel" tests cited by the Nobel committee.)

3. Their research does not "disprove" all of more traditional economics. The assumption of rational economic actors was not / is not meant to be taken exactly literally - it is a simplification of reality, an approximation to the complexities of human behavior, and has allowed us to draw powerful conclusions in a variety of areas.

4. Rationality (usually) gets us close to the right answer, sometimes that's good enough.

And my favorite Vernon Smith "bigthink" paper, where he tries to explain what it all means:

Why is it that human subjects in the laboratory frequently violate the canons of rational choice when tested as isolated individuals, but in the social context of exchange institutions serve ujp decisions that are consistent (as though by magic) with predictive models based on individual rationality? Experimental economists have no good answers to this question.... It seems evident that an important part of the answer resides in the properties of exchange institutions and in how privatley informed, but globally poorly informed, decision making is mediated by institutions. Although institutional rules... have attracted increasing interest among theorists, that development has been slow... (full text here)

Patrick Sullivan points out an excellent post by Northwestern's Lynne Kiesling on Vernon Smith's (and others') experiments with simulated utility markets.

Posted by DeLong at October 18, 2002 02:11 PM | Trackback

Email this entry
Email a link to this entry to:

Your email address:

Message (optional):


The best explanation of Vernon Smith's work was given by Northwestern's Lynne Kiesling, on her blog, THE DAY BEFORE the prize was awarded.


Posted by: Patrick R. Sullivan on October 18, 2002 06:44 PM

Am I wrong in thinking that Dan Kahneman is originally trained a psychologist (and is currently in a psych department) and not an economist? I remember seeing him last year drop by the Yale Law School legal theory workshop pushing "behavorial law" along with Cass Sunstein.

If so, then it's all about winning a Nobel in a field you're not even in...what a badass...

Posted by: Eric on October 21, 2002 09:13 AM

Hmm, the Veron Smith bigthink link doesn't appear to work.

Posted by: Jason McCullough on October 21, 2002 11:16 AM

please, send me new informations about economic success.
Thanks a lot

Posted by: KHOA NGUYEN on April 1, 2003 11:16 PM
Post a comment

Email Address:



Remember info?