Bryan Caplan  

Twilight of My Liberal Idol

PRINT
Politics and Cults... Thoughts on Medicare...

I've been singing Krugman's praises for years. Anyone who could write The Accidental Theorist has a lot of credit in my bank. But his latest column on the failure of betting markets is so beneath him, it hurts to read:

[P]rediction markets — which you see, again and again, touted as having some mystical power to aggregate information, know no more than the conventional wisdom.
I can only shake my head and wonder how any statistically literate thinker can say such things. There is ample evidence that prediction markets have better average performance than almost anything else. To dismiss prediction markets on the basis of a few high-profile failed predictions is just demagoguery.

HT: Robin.


Comments and Sharing





COMMENTS (13 to date)
Student writes:

In the link you provided, Hanson only cites a single paper that studied the ability of prediction markets to forecast election results. That paper was published in 1992 and studied only the 1988 Iowa Presidential Stock Market. Not exactly cutting edge.

I don't think one would have to be a demagouge to question whether that paper is representative of the effectiveness of modern presidential prediction markets. After all, the growth of the internet has no doubt changed the composition of individuals with access to these sorts of markets.

Do you have any other papers that would support your and Hanson's view? It would be very helpful for those us unfamiliar with the literature.

Acad Ronin writes:

Here are a couple of surveys if you have access to NBER.

Justin Wolfers & Eric Zitzewitz. 2004. Prediction Markets.Journal of Economic Perspectives, 18(2), Spring 2004. Previously: NBER Working Paper #10504

Abstract: We analyze the extent to which simple markets can be used to aggregate disperse information into efficient forecasts of unknown future events. Drawing together data from a range of prediction contexts, we show that market-generated forecasts are typically fairly accurate, and they outperform most moderately sophisticated benchmarks. Carefully designed contracts can yield insight into the market's expectations about not only probabilities, means and medians, and also uncertainty about these parameters. Moreover, conditional markets can effectively reveal the market's beliefs about regression coefficients, although we still have the usual problem of disentangling correlation from causation. We discuss a number of market design issues and highlight domains in which prediction markets are most likely to be useful.

Justin Wolfers & Eric Zitzewitz. Prediction Markets in Theory and Practice. Forthcoming, The New Palgrave Dictionary of Economics, 2nd ed. Larry Blume and Steve Durlauf (eds.)
Previously: NBER Working Paper #12083

Abstract: Prediction Markets, sometimes referred to as "information markets," "idea futures" or "event futures", are markets where participants trade contracts whose payoffs are tied to a future event, thereby yielding prices that can be interpreted as market-aggregated forecasts. This article summarizes the recent literature on prediction markets, highlighting both theoretical contributions that emphasize the possibility that these markets efficiently aggregate disperse information, and the lessons from empirical applications which show that market-generated forecasts typically outperform most moderately sophisticated benchmarks. Along the way, we highlight areas ripe for future research.

PrestoPundit writes:

How many years has Krugman's dishonesty been looking you in the face? This looks to me like a willful refusal to see the obvious -- and I'd credit it to a "rally around the guild" mentality of most economists.

Chuck writes:

http://www.overcomingbias.com/2008/01/once-more-with.html#comment-96493546

This, it seems to me, is the real issue at play here:

"There is reason for disappointment, or could be, depending on what one's hopes were. The real disappointment is not that the winner got less than 50% (or whatever it was that happened) but that the markets did very little to reduce the level of uncertainty. As an analogy, suppose that it rains on average about 1 day out of every 10. Suppose, then, that a prediction market in weather predicts, every day, a 1/10 chance of rain. The market is "not wrong" in the sense that, indeed, the events that it assigns a 1/10 probability happen 1/10 of the time. However, it is disappointing to someone who was hoping that the weather prediction market could reduce his level of uncertainty. The political markets seem disappointing in this sense. So far, anyway."

I think a better criticism of Krugman's post would be that systems, like a market, have "settling times", an amount of time requried to digest new information and clear the market, and he just found an example of a market that was required to respond to multiple events without sufficient settling times.

Kubi writes:

Prediction markets...know no more than the conventional wisdom.

Even if this is true, wouldn't prediction markets still be useful by telling us quantitatively exactly what is the conventional wisdom on a topic? After all, currently, the CW is only expressed via vauge pronouncements issued by pundits, whose own biases and filters may make their CW different from the actual CW.

ed writes:

Kubi is exactly right.

Krugman actually has a point, I think...prediction markets ARE often touted as having some kind of mystical power, rather than simply being relatively efficient information aggregators.

Dr. T writes:

I agree with PrestoPundit. Paul Krugman spent his intellectual capital long ago. His recent writings are not worth reading, unless his avoidance of facts and distortions of sound economic principles into unrecognizable shapes appeal to you.

JR writes:

Unfortunately, I am too young to remember when Krugman was an economist.

Matt writes:

Insurance agents use prediction markets, no?

Snark writes:

And hast thou slain the demagogue?
O frabjous day! Callooh! Callay!

Pat writes:

instead of overcoming bias, krugman has submitted to it

bill writes:

"There is ample evidence that prediction markets have better average performance than almost anything else"
Krugman does not state that this is not true, he does however say that they know no more than the conventional wisdom. Are you sure this is incorrect? And this was just a blog post btw, not his column

Terry writes:

There is always the old efficient market test of criticisms that a given market is inefficient -- show us how you would take advantage of the obvious inefficiencies of the market. Show us how you would consistently earn abnormal profits by buying when the market is obviously low and selling when it is obviously high. OK, Mr. Krugman, once a week or so tell us which prices are too high and which are too low, form a portfolio based on that wisdon, and keep track of your profits over the coming year.

It is interesting to note that while Mr. Krugman says the prices he displays are obviously wrong, he doesn't tell us NOW, i.e., ex ante which prices are obviously wrong and how they are wrong.

Comments for this entry have been closed
Return to top