REVIEW OF ECONOMICS AND STATISTICS OVER THE PAST 100 YEARS: AUTHORSHIP

Transcription

1 REVIEW OF ECONOMICS AND STATISTICS OVER THE PAST 100 YEARS: AUTHORSHIP Asim I. Khwaja and Kunal Mangal To celebrate the 100th anniversary of the Review of Economics and Statistics, we are publishing a series of infographics chronicling the journal s evolution over this period. In the May issue we looked at trends in the volume and influence of articles published by the Review, as measured by word counts and citation counts. In this issue, we focus on changes in authorship patterns over time. Over the past 100 years the Review has gone from publishing a small number of repeat authors to reflecting the views of a wider and more diverse set of contributors. The graphs below help fill in the details of this transformation. Author and Article Counts Number of articles and unique authors by year of publication Data source: Metadata provided by JSTOR ( ) and the MIT Press ( ). Not all articles published in the Review had attributed authors. In the early years, the journal also published unattributed editorials and bulletins on economic statistics from around the world. Between the 1920s and the 2010s, the Review grew to publish seven times as many authors per year. About 70% of this growth came from an increase in the number of attributed articles published per year, and the remaining 30% came from increased coauthorship rates. The Review of Economics and Statistics, July 2018, 100(3): i v Ó 2018 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology doi: /rest_e_00743

2 ii THE REVIEW OF ECONOMICS AND STATISTICS Working with Others Coauthorship rates over time Data source: ReStat data provided by JSTOR ( ) and the MIT Press ( ). Top 5 authorship data obtained from Web of Science citation data. The set of Top 5 journals includes: American Economic Review, Quarterly Journal of Economics, Econometrica, Review of Economic Studies, and Journal of Political Economy. The average number of coauthors per article has grown at a rate of 0.2 authors per decade since 1960 both at the Review and at the Top 5 journals in economics and shows no sign of slowing down. Much of this trend is driven by a steep decline in single-authorship: Panel B shows that single-authored papers are now currently the exception, and nearly as common as papers with four or more authors.

3 REVIEW OF ECONOMICS AND STATISTICS OVER THE PAST 100 YEARS iii Newer Voices Repeats in the authorship pool within a five-year window Data source: Metadata provided by JSTOR ( ) and the MIT Press ( ). The main figure plots the fraction of articles within each five-year period from authors who published in the Review in that same five-year period. We see a similar pattern if we look at repeat authorship rates within a ten-year period instead. Over the past 100 years, 3,220 unique authors have contributed to the journal. Most of them over 60% have contributed only once over the whole lifetime of the journal (see inset). In the beginning, the Review relied almost exclusively on authors who made multiple submissions. Over 90% of articles in the first five years came from authors who published more than once. By contrast, between 2010 and 2015, this number was only 21%.

4 iv THE REVIEW OF ECONOMICS AND STATISTICS Increasing Diversity Fraction of authors who are female or have non-english names Data source: Metadata provided by JSTOR ( ) and the MIT Press ( ). We code genders from names using the gender coding method developed for, and employed in, Card, DellaVigna, Funk, and Iriberri (2018). Author names were identified as male or female with a combination of US and German Social Security data, as well as hand-coding. A non-english name is a first name that does not appear on the pages or The fraction of female or non-english names are calculated as a percentage of matched names only. Names remain unmatched mostly because in early years authors were attributed by first initial and last name only. For both series, the matching rate is greater than 90% overall, and better than 97.5% since Authors with repeat publications are counted for each of their publications in a given year. Before the 1970s, the female authorship rate remained low. The years 1937 to 1939 were an exception, partly because Dorothy Wescott, assistant editor of the journal, received attribution for articles that summarized quarterly trends in business statistics. These articles remained mostly unattributed in previous and future issues of the Review. In contrast, since the 1970s, the share of female authors has been growing at a rate of 4 percentage points per decade. Over the same period, the share of non-english names has been growing twice as fast. At the current rate of growth, women will represent half of published authors only by the end of the century.

5 REVIEW OF ECONOMICS AND STATISTICS OVER THE PAST 100 YEARS v Working in Diverse Groups: Rising Heterophily Coauthorship across groups Data source: Metadata provided by JSTOR ( ) and the MIT Press ( ). We code genders from names using the gender coding method developed for, and employed in, Card, DellaVigna, Funk, and Iriberri (2018). Author names were identified as male or female with a combination of US and German Social Security data, as well as hand-coding. A non-english name is a first name that does not appear on the pages or Sample restricted to articles with more than one author. Articles in which any coauthor wasn t identified using our databases of names have been omitted. Graph plots averages over ten-year periods between 1920 and 1950 and over five-year periods between 1950 and Heterophily the tendency for individuals to collect in diverse groups has been growing. Much of the growth in malefemale coauthorship has occurred since 1975, which is around the same time the share of female authorship started to take off as well. Heterophily across boundaries of English/non-English names is more common than coauthorship across gender lines. Reference Card, David, Stefano DellaVigna, Patricia Funk, and Nagore Iriberri, Are the Referees and Editors in Economics Gender Neutral? working paper (2018).