Suchergebnisse

We study the effects of public evaluation of university research on the pay structures of academic departments. A simple equilibrium model of university pay determination shows how the pay-performance relationship can be explained by the incentives inherent in the research evaluation process. We then analyse the pay-performance relationship using data on the salary of all UK university full professors, matched to the performance of their departments from the 2014 UK government evaluation of research, the Research Excellence Framework (REF). A cross sectional empirical analysis shows that both average pay level and pay inequality in a department are positively related to performance. It also shows that the pay-performance relationship is driven by a feature of the research evaluation that allows academics to transfer the affiliation of published research across universities. To assess the effect of the REF on pay structure, we take advantage of the time dimension of our data and of inherent uncertainty in the evaluation of the performance of academic departments generated by the rules of the exercise. Our results indicate that higher achieving departments benefit from increased subsequent hiring and higher professorial salaries with the salary benefits of REF performance concentrated among the highest paid professors.

We study the problem of evaluating whether the selection from a set is close to the ordering of the set determined by an exogenously given measure. Our main result is that three axioms, two naturally capturing "dominance", and a stronger one imposing a form of symmetry in the comparison of selections, are sufficient to evaluate how close any selection from any set is to the given ordering of the set. This closeness is given by a very simple index, which is a linear function of the sum of the ranks of the selected elements. The paper ends by relating this index to the existing literature on distance between orderings, and also offers a practical application of the index.