“Tyler Cowen”:http://www.marginalrevolution.com/marginalrevolution/2008/03/public-choice-o.html responds to the discussion on open publishing.

I don’t envision the free access system as the status quo but free. Papers would be ranked directly in terms of status and popularity rather than ranked through the journals they are published in. Ultimately there wouldn’t be journals and this would make a big difference as journals are the current carrier of selective incentives and status rewards. It would be easy to refuse to referee, since you wouldn’t fear being shut out of publication of that journal; I suspect refereeing might die. And if status were attached to the individual paper rather than the journal, who would bother to become an editor? It would be a very different world and in some ways more like (academic) blogging than its proponents may wish to think. In other words, the partial monopolization of for-fee journals makes it possible to produce status returns to motivate both editors and referees. Returning to the free setting, refereeing will survive insofar as writing detailed referee comments on other people’s work helps with your own research; it is interesting to ponder in which fields this might hold.

The interesting bit for me here is Tyler’s suggestion about the implicit incentives for reviewing; that people referee papers for fear of not being able to get published in the journal in question. My personal take on it (as is the take of a number of other people, if “this discussion”:https://crookedtimber.org/2008/02/12/how-much-should-we-referee/ is anything to go by), is a little different. I review not so much because I feel that if I don’t review a paper for journal _x_ that the editors of that journal will look unkindly on me in future, but because of a broad sense that I send papers out that others ought to review, and hence there’s a diffuse obligation on me to review other people’s papers in turn. In other words, I think that the motivating factor is general reciprocity rather than specific reciprocity. Not only that: when I have been on search committees where we are considering people who have been in the field for a few years, I usually check their resumes to see whether they have been reviewers for a few journals. This isn’t so much to figure out what the editors think of them (very often, editors are happy with whoever they can get as a reviewer), as because it seems to me to be the best publicly available proxy for whether the candidate is the kind of person who is likely to take on their share of the unofficial responsibilities that any school or department has.

This isn’t to say that Tyler may not be right when he suggests that an open publication world might not support the kinds of detailed and thoughtful review that we hope for, and sometimes get, in the current system. But I suspect (perhaps wrongly) that the mechanism that would undermine reviewing would primarily be a sociological one rather than an economic one. That is, it would have more to do with the disappearance of the social role of reviewer, and the set of perceived general responsibilities that go with it, than with the opportunities for specific quid-for-quo interactions between reviewer and editor that the current review system lends it to.

Usually, when I get invitations for talks or interviews with the press, the focus is my research. Last week, however, in an interesting twist, I got an email from the host of a Canadian radio show asking me to chat with her about my experiences with taking pictures of cheese labels.:) I was amused and was happy to talk. The interview is available here. I’m glad Spark contacted me, because I didn’t know about the show, but am now happy to have it in my RSS feed reader. Spark taught me about speedcabling, something I’ll have to try in my lab one of these days.

As a mini-update, right now I’m on my way to the University of Minnesota to speak in the seminar series of their Institute for Advanced Study about my research. It’s a campus-wide talk with people expected in the audience from all sorts of departments, which should be fun. It’ll also be nice to catch up with some prominent sociology bloggers.

A propos of nothing, I am blogging as I’m boarding the plane to Minneapolis. The flight attendant said I was working too hard (boarding with my laptop open), but who said I was working? I think it’s interesting that even in the age of YouTube, etc. laptops are primarily associated with work.

I can’t resist a racing metaphor to describe the problem that’s now facing the US Democrats, but one that is a more-or-less generic problem for democracy. In any system of government, there is a problem of succession, which has a large contingent element. In monarchies, for example, the absence of an adult male heir can produce crises of all kinds (in England, this problem recurred in different forms for all the Tudors from Henry VIII onward). Dictators rarely nominate a capable successor until the last possible moment, so their sudden death often brings about the collapse of the regime. To avoid this, it’s common to see a quasi-hereditary succession which rarely works well, indeed, at all, for more than one generation.

In democracy, close election results can cause big problems, since there is always a range of uncertainty in which normally unimportant procedural decisions or rule violations become critical. Obvious recent examples include the Bush-Gore race in 2000, the Mexican election of 2006, the recent election in Kenya and now the Democratic nomination race. Such close races inevitably produce a lot of bitterness and can lead to disaster. At the moment it seemed as if the threatened breakdown of democracy in Kenya has been averted, but it’s by no means certain that the power-sharing agreement there will hold, and lots of people have already died. At a less drastic level, but one with big consequences for the world, it seem quite possible that the closeness of the race between Obama and Clinton will produce a vicious contest that sinks the eventual winner.

It’s tempting, and sometimes correct, to argue that the sharp divisions that emerge at times like these were there all along. But often this is no more valid than the kind of analysis which ascribes civil strife to “ancient ethnic hatreds” when these are, in reality, little more than rationalisations of contemporary power politics. Certainly, in the case of the Democratic nomination, it’s clear that the vast majority of Democrats would be happy with either candidate and likely that the majority would prefer an immediate end, regardless of the choice, to a continued contest.

Rather than reflecting deeper underlying problems, to a large extent, these succession crises really are problems of institutional design. Some kinds of institutions manage succession problems better than others. Confining attention to democratic systems (broadly defined), I’d argue that there are substantial benefits to simple and definite procedures. If US national elections (including primaries) were based on popular vote (whether first-past-the-post or instant runoff) the likelihood of a result so close as to permit serious dispute would be very small. By contrast, when the result is reached from 50 state ballots, each operating under local and variable rules, the only surprise is that crises are as rare as they are.

As “Daniel”:https://crookedtimber.org/2008/03/12/us-election-horse-race/ notes, we don’t normally do horse race stuff here. And this is week old horse race stuff. But I thought there was some interesting stuff in the SurveyUSA 50 state polls on “Clinton vs McCain”:http://www.surveyusa.com/index.php/2008/03/06/electoral-math-as-of-030608-clinton-276-mccain-262/ and “Obama vs McCain”:http://www.surveyusa.com/index.php/2008/03/06/electoral-math-as-of-030608-obama-280-mccain-258/. The biggest thing was that they show up an interesting fallacy about probabilistic reasoning that, although pretty obvious when stated baldly, is also pretty hard to avoid in practice.

Those polls suggest that if we just look state by state at which candidate is likely to win, we see Obama and Clinton both narrowly ahead of McCain, with the differences between their performances well within any margin of error. That seems right, though by that measure I’d put Clinton a little ahead, and they put Obama ahead.

But the polls also suggest that if we look at two more important measures, Obama is (according to just this poll) a much stronger candidate. He has a higher expected electoral vote and, more importantly, a much higher win probability. “Darryl at Hominid Views”:http://hominidviews.com/?p=1370 produced one model that suggests this, though I suspect his numbers make both Obama and Clinton look more likely to win than they really are. So below I detail a model that I think is a little more realistic. (It’s still a very stylised model, and I’d be interested in knowing from people who do this kind of modelling well what changes might be made to make it better.)
[click to continue…]