Tuesday, 16 February 2016

Attitudes after the killing fields

There has been quite a bit of recent literature looking at the effects of exposure to war and political violence on attitudes. Couttenier, Protu,  Rohner and Thoenig have a recent paper on immigrants in Switzerland that is beautiful in terms of identification. Now, another paper came across my desk that looks super-interesting - Lata Gangadharan and co-authors have a paper on the after-effects of the Cambodian genocide. This was one of the biggest exterminations in recent history, carried out with incredible ferocity -- probably, some 2 million Cambodians were killed, equivalent to 25% of the population.

The abstract is:

We use a field experiment to examine the long-term effects of exposure to the
Cambodian genocide (1975-1979), on individuals’ pro-social and anti-social behavior and risk
preferences. We show that individuals who were exposed to the genocide during childhood
and early adolescence are less trusting, less altruistic, and more risk averse than those who
were not exposed. Our results are corroborated by survey questions on personality traits. The
findings suggest that direct exposure to genocide during childhood and early adolescence has
a lasting impact on social capital and on attitudes towards risk and could make individuals less
extroverted and agreeable.

Wednesday, 10 February 2016

True "open source"

This should be the future of science: http://bigthink.com/neurobonkers/a-pirate-bay-for-science -- a way to download every paper ever written, without paywalls etc. I was glad to see that you can get "Persecution Perpetuated", say, here (thanks to whoever donated it). In general, it seems that all you need to do is to add "sci-hub.io" to the url, so that
"http://qje.oxfordjournals.org/content/127/3/1339.full" becomes

The ethics are quite clear... I just hope they don't end up the way Napster did. 

Out of the clutches of the evil empire...

I recently stepped down as editor of Explorations in Economic History. It was a fun period. Kris Mitchener and I introduced new article formats, and submission rules; turnaround times were down and submission numbers went up. Hopefully Ran Abramitzky, who took over from me, feels he inherited a house in good order.

But I want to talk a bit about the things I learned about dealing with a commercial publisher, Elsevier, which owns the title. I had heard complaints for years about how "evil" they were. I thought they did overcharge for some science titles, and Explorations is a bit too pricey too -- but not insanely so. So my prior was not particularly negative. Having a commercial motive, I hoped, would produce a focus on the bottom line and a passion for efficiency. After all, if you sell the (free) research of academics back to academic institutions, with only a few academics working for nearly free in between, you'd think they would run a smooth, efficient operation that would enable editors to focus on writing helpful decision letters so that authors know what they need to do to get in. Boy was I wrong.

Editing involves a lot of reading papers and reports, and thinking about them. It also involves a lot of looking up referees in online databases, sending them requests, loading up the reports, filing your decision letters, etc. All of this goes through online systems nowadays, and that is where the Elsevier system (and support) really falls apart. Editors work almost for free, and the tools they are given are as primitive as a stone knife. Websites are very slow. Half the functions you expect are not there, or not supported. When I started, I couldn't even add referees myself, and had to ask support staff (outsourced overseas - constantly on vacation, constantly changing, etc.) to do it. The online system is buggy, slow, and never more than one click away from firing off an offensive-sounding pre-spec'd template.

It is very clear that Elsevier doesn't care about small journals like Explorations or the academics that run them, and only treats them as cash cows without investing a penny in them. The publisher has no willingness to improve the journal, and seeing an issue through to production is harder than looking for hen's teeth.When I asked to use Editorial Express, the nearly-free (a few thousand $ a year) online system that the QJE, Econometrica, RESTUD, the EJ, etc. use, the answer was a charming "njet". I am now an editor at the Economic Journal, and one of the best things about it is Editorial Express - fast, clean, simple, and hugely user-friendly.

Should you boycott Elsevier journals? That would be at your own cost. There aren't that many highly-ranked echist journals. The best thing would be if a bunch of the good guys (led by the ClĂ­ometric Society? the EHA?) offered to take over the journal from Elsevier, stating clearly at the outset that if the answer is no, they would start a new, society-owned journal, focused at the same sweet spot between quantification and theory. Faced with the threat that EEH would go the way of the now almost entirely dead European Economic Review (after the European Economic Society left Elsevier twisting in the wind, starting JEEA instead), even Elsevier might think again. But then again... that laser-like focus on short-term returns will probably come out on top.