Which characteristics of economics departments predict productivity of publications?

In some recent work, Bosquet and Combes look at French data (only) and correlate the quality of economics departments with some of their underlying features. Why did they chose France?: “The most frequent way of becoming a full professor is via a national contest that allocates winners to departments in a largely random way.”

So what do we learn?  First, large departments are in per capita terms not so much more productive and not at all doing better in terms of quality.  Proximity to other economics departments also does not matter.

And then:

Heterogeneity among researchers in terms of publication performance has a large, negative explanatory power.

I suspect some of this is causal.  It is good for departments to get rid of their dead wood and good when departments insist that everyone produce.

There is also this:

The second department characteristic that has the highest explanatory power of individual publication performance is the diversity of the department in terms of research fields (within economics).

I wonder there how much the allocation of researchers is truly random.  I find the reverse causality story more plausible, namely that the strongest departments have the resources and heft to cover a larger number of fields, as it is less likely that having people scattered across many fields makes the department as a whole more productive.

In your spare time, you might also ponder this:

Finally, other department characteristics have interesting properties.

  • Contrary to common intuition, more students per academic do not reduce publication performance.

  • Women, older academics, stars in the department and co-authors in foreign institutions all have a positive externality impact on each academic’s individual outcome.

For the pointer I thank Mills Kelly.


I once might have believed their claim that "Women ...have a positive externality impact on each academic’s individual outcome" but now that I have read MR comment threads I know that this is absurd.

And you should at look at GMU's Econ dept for further proof of quality - around 5 women in a faculty of around 45 people (being charitable in terms of inflating the number, as some people only have a building as a profile picture). http://economics.gmu.edu/people/all_faculty

I don't think the Econ Dept at GMU is too bothered by any concerns of affirmative action, philosophically (one professor I knew personally was quite opposed to the very idea in the later 1980s, actually - and very vocal about it). At least this means that the people in charge of faculty hiring are consistent in their beliefs about not letting regulation impede their vision of a better world. Which, perhaps not too surprisingly, looks remarkably like the world of decades past in the Econ's dept mirror. Back when quotas where the rule - quotas based firmly on private principles, ones designed to keep things looking just the way they always had.

Given your laser-like focus on the issue of GMU economics department quality, I figured you'd be all over this post. But frankly, I'm a bit surprised that this is all you have. Someone is against affirmative action? Stop the presses!

Let's take your premise as true, that the department doesn't use affirmative action in hiring. What would we expect to see? We know that most other universities do use such policies, which means there is elevated demand for minority and female professors, and they are in short supply. Meanwhile, there is a surplus of white male professors because of the reduced demand for their services. An institution with color blind hiring practices would naturally tend to hire mostly white men in this environment. It does not follow, however, that mostly white men would be hired if everyone used a color blind policy.

Also, the color blind institution would make higher quality hires than would otherwise be possible, since they are taking advantage of a market inefficiency. It does not follow, however, that white men are inherently better economists than everyone else.

There is no denying that past discrimination was terrible. But the distortions caused by affirmative action, in my opinion, undermine the ultimate goal of equality.

Comments for this post are closed

Comments for this post are closed

Comments for this post are closed

A quick reading of the below indicates there's something wrong with the logic, unless "quality is quantity". It seems that it's just a trivial observation that if everybody in a group publishes, then of course the group as a whole will publish more than if there is a group that contains individuals that refuse to publish. But is publication frequency (quantity) equal to quality? - RL

TC: " So what do we learn? ... And then: ***Heterogeneity among researchers in terms of publication performance*** [ WHAT DOES THIS MEAN? DOES IT SIMPLY MEAN A DEPARTMENT IS DIVERSE RACIALLY OR GEOGRAPHICALLY OR OTHERWISE, OR, TAKEN LITERALLY, THAT THE DEPARTMENT SIMPLY HAS PEOPLE WHO DON'T PUBLISH? IF THE LATTER, AND YOU ARE MEASURING QUALITY BY QUANTITY OF PUBLICATIONS, THIS IS A TRIVIAL OBSERVATION--RL] has a large, negative explanatory power. I suspect some of this is causal. It is good for departments to get rid of their dead wood and good when departments insist that everyone produce."

Well in the gated abstract they don't define "heterogeneity" at all. But my fear that quality is measured by quantity is confirmed, note the top two indicia for quality are quantity:
From the abbreviated paper:
"We decompose individual productivity into three components:
The probability to publish.
The number of publications.
The average quality of these publications. (Quality measured as the product of journal quality weight and the relative number of pages of the article with respect to all articles published in the same journal-year) [EVEN THIS LAST METRIC IS MORE OR LESS A MEASURE OF QUANTITY--NUMBER OF PAGES--TRUE QUALITY ONLY ENTERS INDIRECTLY IN 'JOURNAL QUALITY WEIGHT'. ANOTHER EXAMPLE OF ACADEMICS DOING A STUDY BECAUSE THEY HAPPEN TO HAVE EASY DATA TO TRACK, AKIN TO SEARCHING FOR ONE'S LOST KEYS UNDER A LAMPPOST--RL]"

Comments for this post are closed

Considering the paper is about *productivity*, total amount of pages published strikes me as strongly correlated with overall productivity. So, I'm not sure what you're going on about.

Looking it up from the paper, heterogeneity in this case refers to the average of individual publication output within department d at date t.

How do you propose measuring paper quality?

I think this study was trivial. I would define paper quality by the number of cites in other journals. What this paper found was the trivial clustering effect summarized as: "there are places where everybody publishes a lot, and publish long articles, and there are places where they don't publish all that much, or just a few people in the place publish"--what does that prove? Nothing about quality.

Comments for this post are closed

Comments for this post are closed

Comments for this post are closed

It's the drunk looking for his keys under the streetlight syndrome. They can't measure quality and quantity sounds a bit like quality so it will have to do.

Great minds think alike, I said that streetlight before I read your comment.

Comments for this post are closed

Comments for this post are closed

True story: I searched for "economics" with the top million sites excluded and economics.gmu.edu is the top result.

If it's any consolation, Cal and MIT are both lower on the search list

Comments for this post are closed

I find this part surprising, actually:
"Heterogeneity among researchers in terms of publication performance has a large, negative explanatory power. "

I can see how this could sometimes be true; some faculty may not be good at anything. But if we believe in specialization and division of labor--and we economists do, right?--then shouldn't departments that have a few faculty focus on teaching courses (especially big intro courses) have even more productive research faculty? That is, shouldn't having some faculty specialize in teaching allow other faculty to specialize in research, allowing more research (and more teaching) to get done overall?

I have a hard time envisioning a world in which it is always the case that one gets more research done by expecting everyone to do lots of research, while also expecting everyone to teach.

Comments for this post are closed

Although full professors are assigned at random, assistant professors (or lecturers) and associates (maitre des conferences) are not. Moreover, those below full prof status have "tenure" (cannot be fired) from the very first time they are hired under French law. This would lead to substantial endogeneity which might explain why heterogeneous departments do worse. Not only do they have deadwood but they have a track record of picking people who don't publish. In addition, as the paper notes, many professors are listed as being in multiple departments at the same time. They can't distinguish between those stars who are spread thin and the low ranking profs who teach at multiple schools just to make ends meet.

Finally, there is a problem with using quantity of publications (even quality adjusted by journal) because these are the precise criteria many European universities use for hiring and bonuses even if full professorship is based on the concours. Absent an independent measure of ability, we might suspect an endogeneity problem. Top departments that can hire on the basis of faculty agreement in the U.S., UK, and elsewhere may siphon off the top guys with fewer pubs but a stronger record. The French may often get those who look good on paper but are not as well regarded in the profession (standard lemons problem). Two European professors (Dutch and German) I recently met expressed similar reservations about the way incentives are being structured in many parts of the world with the state often assigning bonuses based on a general formula rather than using judgment as is typical in the U.S.

Comments for this post are closed

I found that my publication rate went up when I retired. My cleverest stuff, though, happened earlier, my most commercialisable stuff later. From this I draw no particular conclusion.

Comments for this post are closed

It's a pity you don't have a donate button! I'd definitely
donate to this superb blog! I suppose for now i'll settle
for bookmarking and adding your RSS feed to my Google account.
I look forward to fresh updates and will talk about this blog with my Facebook group.

Talk soon!

Comments for this post are closed

I used to bbe suggested this blog via mmy cousin. I'm no longer positive whether
this put up is written by means of him as no one
else recognise such particular approximately my problem. You're wonderful!
Thank you!

Comments for this post are closed

So this is my first time commented on Marginal Rev, but I wanted to suggest that the impact of diversity may not be explained (or at least fully explained) by the reverse causality Tyler suggests. The papers findings actually largely mimic current findings in management literature. Take for example Woolley et al (2010).

The main findings of this literature is that collective intelligence is driven by 1) the social intelligence of group members, 2) the diversity of the group and 3) (yes Alan) the number of women (though this is sometimes thought of acting through 1 & 2).

Also interesting, groups of "diverse problem solvers" outperform groups of "high-ability problem solvers"
(Krause through Salminen 2012)

Woolley et al Woohttp://web.mit.edu/press/2010/collective-intel.html
Salminen http://arxiv.org/ftp/arxiv/papers/1204/1204.3401.pdf

Comments for this post are closed

Hellо there! Would you mind if I share your blog with my twitter grοup?
There's a lot of folks that I think would really еnjoу уour content.
Please let me know. Thanks

Comments for this post are closed

Comments for this post are closed