David Kingman reports on the findings from the most recent attempt to estimate the value of the typical graduate premium
The long-standing government policy of expanding access to higher education is premised on the so-called “graduate premium” – the idea that workers who hold university degrees will earn more than non-graduates. But how much is the typical degree really worth? And is that enough to justify the enormous levels of debt that English students now have to incur?
A full-scale review of the debate surrounding the graduate premium is the subject of a forthcoming IF research paper. The latest empirical contribution to this debate was launched recently by the Institute for Fiscal Studies (IFS), who used “big data” for the first time to try and answer some of these questions – with results which may make uncomfortable reading for plenty of students, their parents and university leaders alike.
Is a degree worth £8,000?
The IFS research project, How English domiciled graduate earnings vary with gender, institution attended, subject and socio-economic background, attempted to measure the graduate premium using tax return data from 2012/13 for 260,000 anonymized English graduates, most of whom had been in the labour market for around ten years by that point. This large sample was also cross-referenced with matching data from the Student Loans Company to calculate a proxy-measure for their parental income, in order to work out what impact a graduates’ social background has on their future earning potential.
The results, which were broken down by social class, degree subject and institution, revealed a number of interesting insights about the labour market performance of recent English graduates. First and foremost, this analysis confirmed that there is a substantial graduate premium: the typical male degree-holder was earning £9,000 per year more than the typical non-graduate by their early thirties, while the difference for females was £8,000 (the median graduate, regardless of gender, was earning £30,000 per year compared to £22,000 for the median non-graduate). Graduates were also about twice as likely to be employed as non-graduates, so in a general sense holding a degree was found to be highly advantageous.
However, the IFS analysis also emphasised just how variable earnings are in relation to the other factors they were looking at. In particular, the differences between graduates who had studied different courses were vast: the median male graduate who studied medicine was earning £55,000 ten years into their career, in comparison to just £26,200 for the median male graduate who studied a subject in the social sciences. On average, the two least remunerative subject areas were “mass communication”, where the median male graduate earned £19,300 ten years later, and courses in the creative arts, for which the equivalent figure was £17,900.
In other words, graduates in these subjects were actually earning less than the median non-graduate – so they had experienced a negative economic return from studying their courses, which would look even greater if the cost of paying tuition fees and their lost earnings from delaying the start of their careers were factored in.
A different class?
The research findings also dramatically illustrate the impact of a graduate’s social background on their future earnings, which turns out to be remarkably persistent despite the strong role which obtaining a university education is supposed to play in improving social mobility.
Strikingly, it showed that graduates who came originally from wealthier backgrounds earned significantly more than their poorer counterparts ten years after graduation, even if they had studied the same course at the same institution. The data from the Student Loans Company enabled the IFS researchers to identify the graduates within the sample who came from households in the top fifth of the earnings distribution, and they were found to earn an average of £8,000 for males (or £5,300 for females) more than graduates who originally came from lower down in the earnings distribution after ten years of work.
The true scale of the advantages enjoyed by graduates from wealthy families may well be even greater than this, because the sample was restricted to those students who took out student loans to fund their studies, who were only 85% of the total – most of the remainder are likely to be the wealthiest ones whose families could afford to support them instead.
A large proportion (although not all) of the differences in graduate earnings by social class were explained by the greater propensity of students from wealthier backgrounds to attend the most selective institutions, which are associated with the highest graduate premium. At one extreme, over 10% of male graduates from LSE, Oxford and Cambridge were found to be earning over £100,000 ten years after they graduated, whereas at the other, there were 23 institutions where the average male graduate had received no graduate premium at all.
Of course, the raw financial returns are far from the only benefit that someone gets from attending university. However, these findings from the IFS pose some tricky questions for higher-education policy-makers. In particular, should prospective students be told about the average earnings profile of former students at particular universities or for particular subjects? And should taxpayers – who ultimately fund higher education through the subsidies within the student loan system – be expected to fund courses and institutions if it is has been demonstrated that people who graduate from them aren’t benefiting in the labour market?
Unfortunately, it seems that “Big Data” can sometimes pose more questions than it answers.