Is there really a problem of graduate 'over-education'?

Peter Dolton and Anna Vignoles challenge the notion that Britain has too many students.

The higher education sector in the United Kingdom has experienced virtually continuous growth in student numbers over the past 35 years. Between 1989 and 1995 alone there was a massive 61 per cent increase in university enrolments.

Any further expansion of higher education is currently on hold, pending the verdict of Sir Ron Dearing's National Committee of Inquiry. This seems an opportune time to ask whether the UK needs more graduates, as many people, including the Confederation of British Industry, argue.

Do most graduates get graduate level jobs? What is a graduate job? Can we measure the extent of any "over-education" problem? What has been the effect of expansion on both the private and the social rate of return to a degree? Specifically, do we have a situation of excess supply of graduates, caused in part by subsidising higher education to the extent that we do?

In response to similar concerns elsewhere in the world about a potential over supply of graduates, there is now a growing academic literature from the United States and Europe on the issue of over-education. US evidence suggests that "over-education" among the workforce as a whole may be as high as 40 per cent.

UK research into the issue is more limited and more recent but evidence confirms similar findings for this country. Our own work looks at people who graduated in 1980 and indicates that 38 per cent of these graduates did not get a "graduate" job when they left university. Six years into their careers 30 per cent still had not got a job that they considered to be of graduate level. Furthermore "over-educated" graduates earned less than those in graduate jobs, suggesting that "over-education" reduces both the private and social rate of return to a degree.

Analysis of how long graduates remained in jobs for which they were "over-educated" is equally gloomy. The majority of graduates who did not make it into a sought after "graduate" job appeared destined to remain under-utilised permanently. All this might indicate that indeed we have too many graduates.

Unfortunately, however, there are some fundamental difficulties in interpreting the evidence on this issue. Most of the literature on "over-education" has used self-surveys to determine whether graduates are indeed "over-educated". Graduates are simply asked whether or not they needed a degree to do or get their job.

Clearly this is not an objective measure of whether or not the person is "over-educated". The former prime minister might argue that he only needed O levels to do his job, Tony Blair, on the other hand, is likely to claim that a degree is essential. So some researchers have tried to construct more "objective" measures of "over-education", generally relying on job analysts to tell them what the average educational requirements of a particular job title are.

Even this approach has problems. Not all office managers do the same job or need the same qualifications. It is also possible that graduates "grow" jobs, bringing extra skills and knowledge to the job, and so transforming it into a graduate-level job. Recent National Institute for Educational Research findings suggest that employers who recruit graduates to non-graduate jobs do not generally upgrade the job content anyway. But none of the data is sufficiently sophisticated to anal-yse this problem clearly.

There is an even more fundamental issue. Do graduates bring skills and attributes to a job which enable them to do a non-graduate job more effectively? If graduate secretaries or graduate sales staff are better at their jobs because of their higher education, clearly "over-education" may be less of a problem than first thought. Until we can identify the skills and qualities that higher education develops in individuals and relate these attributes to activity at work, it is premature to claim that we have too many graduates.

There is a need to consider the effects of greater numbers of graduates on the private and social rate of return to a degree. It is important to stress that concerns about "over-education" are only relevant because, for the moment at least, we continue to subsidise students' higher education.

If there is indeed widespread graduate "over-education", it would not be optimal for the state to continue to subsidise growing numbers of graduates. This is not to say that greater numbers of people should not experience higher education, merely that they might pay for more or all of it themselves.

Perhaps most importantly, education generates externalities, benefits to the individual and society that are not taken into account by looking at job title and earnings. There is almost overwhelming evidence that education improves parenting skills, makes a person less likely to divorce and commit crimes and improves health, to name but a few benefits. When the cultural benefits of having a more educated population are weighed, worries about "over-educating" people seem less relevant.

The evidence on "over-education" is sufficient to suggest that this should be an issue of great importance for Dearing and others. However, we are not yet in a position to categorically say we have too many graduates. More research is needed to give a clearer answer to some of our questions.

Professor Peter Dolton and postgraduate Anna Vignoles are in the economics department at the University of Newcastle-upon-Tyne. They are presenting their findings at an Economic and Social Research Council/Society for Research into Higher Education seminar this week.