In 1991, two thirds of the articles in Econlit, a comprehensive index of academic economic research, were written by people based in North America. By 2006, that share had dropped by one third to 45 percent.
These numbers are taken from a recent working paper by Cardoso, Guimaraes and Zimmermann. Their evidence suggests that the relative decline in North American share has come about because of a substantial increase in the productivity of European economists, and also because of an increase in Asian publication rates:
At one level, the dominance of North American and European authors in Econlit is not surprising. Das, Do, Shaines and Srinivasan argue that the US is not an outlier in terms of the production of economic research. They find, as do studies of other fields, that variables such as overall per capita GDP, governance, and English-language usage largely explain publications per capita. The Asian gain is readily explained by that region's rapid economic growth. The more interesting question is: why did North America decline and Europe rise so much between 1991 and 2006?
There has been a cultural revolution in European universities. I use the term advisedly. During the Chinese cultural revolution, the leadership empowered the masses to overthrow entrenched interests, creating radical change. In the same way, the British research assessment exercise empowered young research-active academics, generating a new, publication-intensive culture. The German university system has been reformed to introduce competitive systems of research funding, undermining the power of the old guard. The Bologna process aims to create a pan-European educational system, with more mobility for students and faculty - and hence more inter-university competition.
This next figure from the Cardoso et al paper shows the rise of Europe another way. The countries on the far right of the diagram are those with high levels of published economics articles per million inhabitants in 1991 through 1995 - Israel, the US, and Canada. The countries at the top of the diagram are those with the highest publication rates in 2002 to 2006 - a set of northern European countries including Norway, Britain, Sweden, and Denmark. Of the top 1991-1995 achievers, only Israel has maintained its publication per capita rate. The USA has slipped slightly, but Canada has slipped the most.
Note: ISO country codes: AUT – Austria; AUS – Australia; BEL – Belgium; CAN – Canada; DNK – Denmark; FRA – France; DEU – Germany; ISR – Israel; ITA – Italy; JPN – Japan; NLD – The Netherlands; NOR – Norway; ESP – Spain; SWE – Sweden; CHE – Switzerland; GBR – United Kingdom; USA - USA. Source: Authors computations based on Econlit and SSCI
So what's with Canada, eh? I can think of any number of explanations: an aging professoriate, European competition in fields Canadians have tended to specialize in, the strength of incentives to publish, or just random variation. I don't, however, have evidence in support of any of these hypotheses.
What is more interesting is how increased research activity in Europe and Asia is changing the profession. I've written before about my own personal feeling that increased competition for available spaces in journals means that it is getting harder to publish. Hold-ups in the publication process hinder the dissemination of research, as well as the career progression of academics, and this is a problem.
One response to a greater flow of research is to establish more publication outlets. The American Economic Association has introduced four new American Economic Journals. Rumour has it that when one very prominent professor's graduate students were no longer able to get published in the Association's journals, it was decided that the time had come to start some new journals. Likewise, there has been discussion in the Canadian Economics Association about starting a new Canadian journal, but as yet nothing has come of it.
More journals solve one problem and create another: what policy makers call "take-up." In the 20 year period studied by Das et al, there were 76,046 empirical economics papers written, and publication rates have increased since then. How do the ideas get read, taken up, and incorporated into the policy debate? Perhaps that's the reason why one of the most popular sessions at this year's American Economics Association meetings was on popular economics (webcast for AEA members only here). With so much being published, the really cut-throat competition is for audiences: getting someone to read, listen and pay attention.
Increased research output also changes what it means to be in the 'top 5'. When there are 100 journals, the top five represents the top five percent. When rates of research output increase sufficiently to support, say, 400 journals, the top five represents the top 1.25 percent. The simultaneous expansion of both research production and publication outlets has several consequences.
First, what it takes to publish in the "top five" or "top 20" changes. Whereas at one point one had to be in, say, at the 95th percentile or above, now one has to be, say, at the 98th percentile or above. What looks like an unchanged standard has, in fact, become tougher to meet, with real implications for anyone going up for tenure.
Second, while it might be possible to identify the 90th percentile of research output with some objectivity, or even the 95th, deciding whether or not an article is objectively good enough to be in the top 99.5 or 99.9 percent is another matter. Both the papers discussed here find that US-based authors remain dominant in the top five journals. When it is hard to tell whether or not a paper is objectively good enough to be in a particular top five journal, referees and editors may rely upon the signal provided by the individual's institutional affiliation. Yet as European and Asian economists come to dominate the mid-ranked journals, how long will it be before LSE is as good a signal of quality as Chicago and the US dominance of the top five journals dissipates?
On the other hand, why worry? Economic research is a public good. Everyone else free rides, why shouldn't economists? Surely the optimal strategy is to let others do the hard work, and take advantage of their efforts?