What matters most: teaching or research? Empirical evidence on the remuneration of British academics

Abstract

This paper examines the impact of productivity on pay within academia, drawing upon a detailed dataset of academics from five old established universities. Results outline the importance of publication; grant receipt and teaching quality in the determination of pay. A large financial penalty to time out of the profession is revealed, which, with productivity variables, explains away the gender salary gap. The relationship between teaching and research is investigated, and we find some evidence in support of the hypothesis that productive researchers are successful teachers. Results on reservation salary suggest that the best academics are willing to stay within the profession.

Acknowledgements

Since completing this paper, Melanie Ward has moved to the European Central Bank, Frankfurt. Both authors are affiliated with IZA, Bonn, and CEPR, London. The first version of the paper was written while both authors were at IZA, Bonn. We wish to thank one anonymous referee, Thomas Bauer, Hielke Buddelmeyer, Astrid Kunze, Arthur van Soest, Andrew Oswald, and the seminar participants at Leicester University, Essex University and ESPE 2000, Bonn, for valuable comments and helpful discussions.

Notes

Such a fixed salary system is not unique to UK academics. A similar payment system applies to UK public sector workers, and to academics and public sector workers within other countries like, for example, the USA, Germany, and the Netherlands.

For a part of the sample the 1994/1995 scale is relevant salary scale, while for another part the 1995/1996 scale is the relevant scale. Compared to the 1994/1995 scale, the salaries of the 1995/1996 scale were increased by 2.7%. This fact is taken into account in the analysis.

The limited availability of statistics for the UK academic population makes it difficult to comment to what extent the data is representative. Comparison with summary information on the parent population of academics in Scotland made available by the university personnel departments reveals that females are marginally over-represented in the dataset. Across rank and subject area, at least, the data are representative of the academic population.

It is recognized that there might be some causal effect the other way around – from salary to productivity. None of the papers on this topic mentions this potential problem, and also with the data one cannot correct for this endogeneity.

Results for this analysis are not presented here, but are available on request.