What is the relationship between the research intensivity of a university, and its impact on graduate earnings?
Here is a quantitative analysis, based on time series data from the Research Assessment Exercise (RAE) in the UK.
Analysis of higher education quality has become a central issue in light of UK government policies to introduce variable fees and to encourage more and more young people to attend university. In this context, an important question is whether institutional quality is reflected in labour market earnings. Such information could help to inform students, teachers and policy makers. However, empirical analysis of the link between institutional quality and labour market outcomes is rare outside the US. In this study we offer an empirical analysis of labour market returns to measures of institutional quality. We exploit the Graduate Cohort Studies for 1985, 1990, 1995 and 1999. We use data from the Higher Education Statistics Agency to obtain information about institutional quality. We use five different measures of quality: research assessment exercise (RAE) score; the faculty-student ratio; the retention rate; the total tariff score (which is based on A-levels or other eligible qualifications); mean faculty salary and expenditure per pupil. We explore how these variables can be combined to an aggregate proxy for quality. We attempt to control for all other variables that might influence both the quality of institution attended by the graduate and his/her wage. Like all studies in this literature, our analysis relies on an (untestable) assumption that relevant variables have not been omitted. Our key finding is that there is a positive return to attending a higher quality institution for most of the indicators, which is similar to what US studies have found. The earnings differential from attending a higher quality institution is about 6 per cent on average, when using an overall proxy for quality based on a combination of the measures. We also examine whether it makes a difference if an individual attends an institution in the second, third or fourth quartile of the quality distribution, as compared to an institution in the first (lowest) quartile. Results suggest that if a student attends an institution in the highest quartile of the RAE score, the retention rate or the total tariff, this leads to higher wages between 10 percent and 16 percent, compared to an individual who attends an institution in the lowest quartile. However, if an individual attends an institution in the second highest quartile of quality, the earning differential drops to 5-7 percent in comparison to the bottom ranked institutions. We also find that returns to institutional quality have increased over time, though within a modest range.
This is all very provocative. The cumulative effect of research intensivity on earnings across the lifespan is remarkable. We also get a good sense that research intensivity has an effect on the type of graduate produced by institutions. research-led teaching may not just be hyperbole – it may have a palpable and enduring impact on graduate quality and graduate earnings. There are caveats – we need to know how the extent to which the researchers generating these high RAE rankings teach – my ‘anecdata’ from being in the UK and from extensive interactions with UK colleagues is that research-only academic appointments (not research fellowships) are rare.
Iftikhar Hussain, CEP, London School of Economics and University of Oxford
Sandra Mcnally CEP, London School of Economics and IZA
Shqiponja Telhaj CEP, London School of Economics and University of Sussex