Graduate pay premium is two thirds lower for young women than previously thought

19th January 2026

Labour market returns for female graduates have been historically overestimated using tax data, as much of the documented graduate pay premium is the result of female graduates working longer hours, not higher hourly wages, a UCL led study finds.

The analysis, published in Economics of Education Review, challenges the conventional wisdom that early career earning potential is boosted by obtaining a university degree, particularly for women.

The researchers studied a sample of 2,878 millennials born in 1989 and 1990 who were participants in the UCL Centre for Longitudinal Studies' Next Steps cohort study.

The researchers were able to pinpoint their annual wages at age 25, but the Next Steps study provided additional data on hourly wages and hours worked - information that is not currently collected in administrative data.

When studying annual wages alone, female graduates earned on average 26.9% more than non-graduate women at age 26, which reduced to a 13% difference once background characteristics* were taken into account. These estimates were similar to others obtained using English administrative tax data.

However, when the number of hours worked was considered, the difference in earnings between graduate and non-graduate women dropped to 4.8%.

While the squeezing of the graduate premium can be partially explained by women without a degree being more likely to work part time, annual wages still overestimated the premium when the sample studied only full-time workers.

The researchers explain that this miscalculation occurs because graduate women work, on average, 2.3 hours more per week than non-graduate women. This small difference in weekly working hours is enough to reduce the magnitude of the graduate premium by two thirds at this early career stage.

Analysis of earnings has typically been based on HMRC tax records, which calculate earnings annually and do not record the number of hours worked. The researchers argue that using tax records alone distorts the size of the graduate premium, particularly for women.

Prior research from the UCL Social Research Institute has established that the graduate pay premium increases over time and that the lifetime university pay premium is bigger for men than for women.

Co-author Dr Nikki Shure (UCL Social Research Institute) said: "Our study found a crucial blind spot in the way the ‘graduate pay premium' is calculated. Without accounting for the number of hours worked, the economic returns to female graduates appear much larger than they actually are and this creates a distorted picture of the labour market.

"Hourly earnings are a great measure of economic productivity, and they should be recorded in official statistics, whether by employees or employers."
Co-author Dr Anna Adamecz (UCL Social Research Institute) said: "Our findings suggest that there is a limit on how much higher education can improve national productivity, particularly among graduates in the early stages of their careers.

“But we couldn't have discovered this without the additional information provided by the Next Steps cohort study. Next Steps is the only nationally representative cohort study of millennials in England. This is a key generation for understanding labour market dynamics as they came of age in the aftermath of the financial crisis and have experienced profound generational shifts as compared to Gen X and Gen Z. It is crucial that policymakers have accurate and reliable information when shaping fair and effective education policy."

The labour market returns to graduation: reconciling administrative and survey data estimates -
https://www.sciencedirect.com/science/article/pii/S0272775725000810?via%3Dihub