| Abstract: |
The rising incidence of co-authorship has received much attention in recent
years especially in the sciences, but has received little attention so far in
economics, at least when using large data sets. This paper draws on a very
large new data set, covering around 175,000 articles in the top 250 or so
economics journals, over the period 1996 to 2014. On the basis of these data
several novel insightful charts, adjusted for various factors, are
constructed. The incidence of co-authorship by category type (two, three,
four-plus authors) is looked at in almost all charts and differences between
groups of journals by rank are also explored. Using this framework the
following are looked at: co-authorship over time, across countries, citations
per paper and per paper per author, length of papers, number of references,
frequency of alphabetical ordering of author names and the career profile in
terms of co-authorship for 133 top economists, using a specially constructed
data set based on detailed CV data. These charts were constructed to throw
light on the various explanations posited for the rise in the incidence of
co-authorship, and they do so to a surprising extent. In particular,
considerable doubt is thrown on the increased specialisation argument, at
least as applied to economics. They also throw major doubt on the hypothesis
that increased co-authorship can be traced to 'token' acknowledgement for work
done of a peripheral nature. It is also argued that several other explanations
are not well founded, with the exception perhaps of increased ease of
communication and greatly decreased cost of travel. The main explanation
though the paper posits could be linked to changing attitudes to the
discounting for number of authors by hiring and funding agencies. |