Latest Longitudinal Education Outcomes (LEO) shows that the gender pay gap is present just one year after graduation

19 June 2017

The Department for Education (DfE) have released the third in a series of higher education data releases using Longitudinal Education Outcomes (LEO) data. LEO data shows how much UK graduates of different courses at different universities were earning, one, three and five years after graduation. The data, which is still in a very raw format and comes with a number of caveats, has allowed analysts to start to identify a number of significant trends within the graduate labour market and see how these evolve over time.  

Emerging data trends 

Indeed, one of the most alarming trends to emerge from initial analysis of the data was the influence of gender. The data clearly indicates that the gender pay gap exists from the very beginning of the graduate career. The DfE found that in 12 subjects, male earnings are greater than female earnings at more than 75% of institutions. They also found that only one year after graduation, men’s median salaries outperform women’s by £2,000, rising to a gap of £3,400 after three years. Whilst university admissions have overall seen a higher proportion of women than men enrolling in recent years, the gender pay gap remains a problem for women entering the graduate labour market.   

In addition to the clear gender differential, the data allows highlights a number of other factors that influence graduates’ medium to long term earning potential.  

Unsurprisingly, the analysis shows higher earnings potential for STEM courses compared to arts and humanities courses. This shows that the labour market continues to have greater demand for graduate skills in STEM subjects. However, these courses remain the most expensive courses to teach, often requiring cross-subsidy from cheaper courses in the arts.  

Issues with the data   

It is important to remember that the LEO data is not contextualised in any way, meaning that certain factors are ignored when comparing different institutions, such as regional variations in salary. The DfE states in the supporting document that further analysis will be conducted which will look to contextualise this data set.  

The 23 subject areas are based on an amalgamation of JACS codes. These depend heavily on how institutions have coded their courses and can throw up some anomalies in analysis. The subject areas also amalgamate courses where the career trajectories and destinations may be hugely different – Architecture, Building and Planning is a good example where Architecture students would have a very different career trajectory to those choosing Town Planning, for example.  

An important element of the dataset revolves around graduates’ attainment prior to commencing their undergraduate degree. However, the assessment of students’ prior attainment merely covers their A-level scores and still doesn’t factor in any other qualifications (e.g. BTECs). It is also possible that other factors could be at play, such as gender, which could account for the large discrepancies in the data.

The LEO data has not been designed to identify the universities with the best teaching or ‘value added’ potential, nor can it be viewed as a performance or earning indicator. Instead it is purely historical data, and whilst it is very useful, future labour market outcomes are arguably far too complex for predictions to be made based upon previous outcomes. However, the stark gender and subject contrasts revealed in this report are simply too great to be ignored and need to be addressed within future policy making.    

View the full report