Comments on: Counterparties A slice of lime in the soda Sun, 26 Oct 2014 19:05:02 +0000 hourly 1 By: tomrus Thu, 19 May 2011 15:38:36 +0000 Re “Greg Mankiw is right about Treasury bonds,”

Here is what the great man said:

“No, I am not predicting the United States is about to become just like Japan. But it is not inconceivable.”

Tough to be wrong.

By: OnkelBob Thu, 19 May 2011 15:01:56 +0000 Re: Contaminating the wisdom of crowds. I wonder if there’s some of Maslow’s hierarchy of needs in play here too; specifically, the need to belong, to believe you are part of the group. You are happy to make a guess that is your own when not exposed to possible membership in the group. That is you’re happy with the groups you belong to, and this guess is an exercise in self-esteem. However, once you become aware of this group, you instinctively want to be part of it, and are willing to forgo the self-esteem with being correct in order to belong.

By: dWj Thu, 19 May 2011 13:42:29 +0000 The classic experiment in shared information and the breakdown of crowd wisdom presented volunteers with a rod and asked them to estimate its length. The real volunteer would answer after seeing about 7 people who were pretending to be volunteers give the same implausible guess, and frequently gave a guess that was at least shaded in the direction of the other “guesses”, sometimes closer to the implausible number than to the correct one.

This sort of thing can depend on the kind of information that is being shared. If you’re ever a jury foreman, resist the temptation to start with a poll of the jury; you can get two different biases by doing that. For one, if you go down the line having jurors state their opinion in public, you get the effect just described; if the first couple jurors happen to have the same prejudices, the whole jury will be infected by them. The other problem is that people who have stated conclusions become emotionally attached to them; if the jurors are split, it becomes harder for some of the jurors to convince the others that they’re wrong, and it may be that the jurors who ultimately split do so based on fatigue rather than thoughtful reconsideration.

You’re not going to get a perfect jury, but the better way to approach it is to let everybody discuss the evidence together — what some people might have missed, what might have particularly struck others — before any poll is taken. There are lesser emotional commitments involved, and, even if the first couple of jurors seem to be on the same side, there’s a much weaker signal to other jurors as to what the socially acceptable verdict is. Something like the wisdom of crowds has more room to take effect.

Financial markets obviously have the same emotional commitment problem; I think the social influence is qualitatively different due to the more anonymous nature of the markets, but to some extent you face, as Keynes described, a situation where people are trying to guess what other people are trying to guess, and with some degree of adverse selection thrown in the mix. Markets, as Hayek noted, can be good at aggregating information — or, perhaps more often, less bad than systems without markets — but they need good designs and/or organic informal institutional arrangements to make them do that particularly well.