Skip to main content Skip to navigation

Returns to higher education: variations by subject and degree class

The government is due to review the cap on "top-up" fees later this year. Robin Naylor, Jeremy Smith, and colleagues fear that a big rise will deter some people from going to university: their research shows that although on average a degree gets you a pay premium, there are wide variations in the returns to higher education by subject and degree class.

The burden of funding higher education in the UK has steadily shifted away from the taxpayer and towards students and their families. Since 1998, all full-time UK university students from within the European Union have paid uniform tuition fees. And since the autumn of 2006, universities have been able to charge "top-up" fees up to a regulated maximum, differentiated by university and by course.

When tuition fees were first introduced, much was made of estimates indicating that the average return to an undergraduate degree is high. On the basis of this evidence, proponents of the new arrangements argued that it was reasonable to expect students to make a more substantial contribution to the costs of higher education.

Some of the most influential evidence was based on analysis of the 1958 birth cohort, in which the typical graduate would have left university around 1980. But there been many changes for more recent cohorts in terms of both the supply and demand for highly skilled labour.

The pay premium for a first or upper second is considerably higher than for a lower degree class

One notable change is the significant increase in the proportion of young people going to university since the late 1980s. In a new study, we analyse data on one of these later generations for whom participation rates in higher education were higher: the 1970 birth cohort, in which the typical graduate would have left university in the early 1990s.

In policy discussions about tuition fees, substantial weight has been placed on estimates of the average return to a degree. But variations around the average are clearly an important dimension of the debate and they have been under-explored. Our research examines heterogeneities in returns by such factors as gender, family background, ability and subject studied.

But our main focus is the extent to which the returns to a degree vary with the class of degree awarded. The principal reason for this is that we might expect employers to rank graduates at the point of recruitment – and degree class is the characteristic by which employers are most likely to base a ranking of job applicants.

We use information on the gross hourly wages of 2,919 individuals – 1,497 men and 1,422 women – at age 30, drawn from a cohort of babies born in England, Wales, Scotland and Northern Ireland in one week in April 1970.

Our results confirm that the average wage return to a degree is substantial. We find that relative to otherwise similar individuals whose highest educational qualification level is two or more A-levels, a university graduate will earn a pay premium of 16% for men and 19% for women.

A social science degree gives the highest wage return, followed by science and then arts and humanities

But there is variation around these averages. The premium for a good degree (a first or an upper second) is 22% for men and 25% for women, while the premium for a lower degree class (a lower second, a third or a pass) is 12% for men and 15% for women. The difference in the wage returns between good and lower degree classes is remarkably similar by gender, at 10 percentage points.

The subject graduates studied also makes a considerable difference to the returns to a degree. For men, social science gives the highest wage return (32%), followed by science (22%) and arts and humanities (13%). For women, the ordering of subjects is the same: social science gives the highest wage return (28%), followed by science (22%) and arts and humanities (19%).

We find little evidence of significant differences in the returns to a degree by either family background or ability.

Our finding on the average return to a degree makes the investment decision of participating in higher education seem an attractive proposition. But our evidence of the marked variation around this average (according to both subject and degree class) renders the return on investment in higher education potentially much lower at the margin. This implies that any substantial increase in the regulated maximum for top-up fees risks deterring participation in higher education, at least for potential applicants from less financially privileged backgrounds.

Publication details

This article summarises Heterogeneities in the Returns to Degrees: Evidence from the British Cohort Study 1970, by Massimiliano Bratti, Robin Naylor and Jeremy Smith.

The authors

Massimiliano Bratti is at the University of Milan. Robin Naylor and Jeremy Smith are professors of economics at the University of Warwick.