Within and across department variability in individual productivity : the case of economics

University departments (or research institutes) are the governance units in any scientific field where the demand for and the supply of researchers interact. As a first step towards a formal model of this process, this paper investigates the characteristics of productivity distributions of a population of 2,530 individuals with at least one publication who were working in 81 world top Economics departments in 2007. Individual productivity is measured in two ways: as the number of publications until 2007, and as a quality index that weights differently the articles published in four journal equivalent classes. The academic age of individuals, measured as the number of years since obtaining the PhD until 2007, is used to measure productivity per year. Independently of the two productivity measures, and both before and after age normalization, the main findings of the paper are the following five. Firstly, individuals within each department have very different productivities. Secondly, there is not a single pattern of productivity inequality and skewness at the department level. On the contrary, productivity distributions are very different across departments. Thirdly, the effect on overall productivity inequality of differences in productivity distributions across departments is greater than the analogous effect in other contexts. Fourthly, to a large extent, this effect on overall productivity inequality is accounted for by scale factors well captured by departments’ mean productivities. Fifthly, this high degree of departmental heterogeneity is found to be compatible with greater homogeneity across the members of a partition of the sample into seven countries and a residual category.