Large educational disparities by socioeconomic status (SES) remain a pressing concern for policymakers aiming to ensure equal opportunities for all students to achieve their educational potential. While SES differences in educational attainment and the likelihood of obtaining a university degree have been widely studied, recent research has increasingly focused on gaps in the types of undergraduate programmes that students pursue. Evidence from the UK (Wyness et al. 2022), the US (Dillon and Smith 2017, Leighton and Speer 2023), and Ireland (Delaney and Devereux 2020) shows that lower-SES students are less likely to enrol in prestigious undergraduate institutions or programmes associated with high-achieving peers and strong labour market returns. However, despite the growing importance of graduate degrees – particularly in Europe – there has been relatively little research into how SES influences choices at this stage of education.
Students from lower SES backgrounds may face barriers to graduate education due to lower self-confidence in their academic abilities or financial constraints, including high debt levels and limited family support. These factors may also influence their selection of graduate programmes, potentially steering them towards fields with lower labour market returns. Consequently, SES-based differences in graduate education choices could have significant implications for income inequality and social mobility. In our recent study (Delaney and Devereux 2025), we analyse Irish administrative data to investigate SES-related disparities in graduate degree selection. Furthermore, we examine the impact of these choices on SES-related earnings gaps among university graduates.
Ireland presents an interesting case for study, as financial aid for graduate education is relatively limited, consisting primarily of partial or full tuition waivers and maintenance grants, with no loan system in place. Crucially, the available financial support is heavily targeted at students from low-income backgrounds. If SES-based disparities in graduate education persist within this context, they are likely to be even more pronounced in countries where financial aid is scarce or non-existent, potentially exacerbating challenges for students from disadvantaged backgrounds.
A key challenge in studying SES differences in graduate education is ensuring comparisons are made between individuals from different SES groups who are similar in terms of their undergraduate degree, academic interests, and achievements. Since our focus is on graduate study, we limit our sample to individuals who have completed an undergraduate degree by the age of 25. Utilising detailed administrative data, we control for undergraduate programme characteristics through fixed effects at the institution-field level (e.g. psychology at University College Dublin). We also incorporate undergraduate performance indicators based on degree classifications (first-class honours, upper second-class honours (2.1), and other categories).
At the secondary education level, Irish students take centrally administered Leaving Certificate examinations in seven to eight subjects, including compulsory (Irish, English, mathematics) and elective subjects. To further control for academic achievement and interests, we incorporate detailed information on subject choices, grades received, and overall Leaving Certificate performance (measured by ‘Leaving Certificate Points’).
To measure SES, we aggregate multiple SES proxies into a single index using the first principal component. For the analysis, we categorise this index into five quintiles.
We first examine whether students enrol in a graduate programme within three years of completing their undergraduate degree. Figure 1 illustrates that, after controlling for key variables, students from the lowest SES group are 1.4 percentage points less likely to pursue graduate study. Since 21% of students pursue graduate study within three years, this suggests that students from lower SES backgrounds are 7% less likely to enrol in graduate study. These findings contrast with those of Britton et al. (2020) and Britton and van der Erve (2020), who, using UK administrative data, found that students from lower SES backgrounds were slightly more likely to pursue graduate education.
Among students who do enrol in graduate programmes, we analyse their field of study. Specifically, we examine enrolment in STEM, economics/business (Stansbury 2023 shows that, at PhD level, socioeconomic diversity is lower in economics than in other disciplines) teaching, health, and other fields. We also assess the expected labour market earnings associated with each field based on historical graduate earnings data.
Figure 1 shows that students from the lowest SES quintile are 9 percentage points less likely to pursue an economics/business graduate degree, while they are 3 percentage points more likely to choose teaching and 4 percentage points more likely to enter the residual ‘other field’ category. The economics/business effect is particularly large: given that 30% of graduate students enter this field, the 9 percentage point gap implies that low-SES students are 28% less likely to study economics/business compared to their high-SES counterparts. Since lower-SES students are more likely to enter the lower-paying teaching field and less likely to choose high-paying economics/business programmes, they ultimately enrol in graduate programmes with expected earnings that are 3% lower than those of observationally similar high-SES students.
Figure 1 Effect of being in the lowest SES group relative to being in the highest SES group
When we analyse heterogeneous effects, we observe relatively small differences in SES gaps between males and females. However, we detect differences based on undergraduate academic performance. After controlling for a broad set of variables, there are minimal, if any, SES gaps in graduate choices or outcomes for students who achieved the highest degree classification (first class honours) in their undergraduate studies.
Humphries et al. (2022) report that the returns to graduate degrees differ greatly by programme. Our analysis reveals that SES-based differences in graduate programme choices substantially impact post-graduate earnings disparities. Notably, controlling for graduate programme attendance alone does not significantly alter SES-related earnings gaps. However, when we incorporate specific graduate programme choices into our models, the earnings gap narrows considerably. By age 33, persons from the lowest SES quintile who entered graduate programmes earn over 10% less than their highest SES counterparts. Crucially, controlling for the specific graduate programme chosen reduces this gap by 40%, underscoring the importance of field selection in shaping earnings outcomes.
Our findings suggest that disparities in overall graduate attendance (the ‘extensive margin’) are less consequential for SES earnings gaps than differences in graduate programme selection (the ‘intensive margin’). Further research is needed to explore why students from lower SES backgrounds disproportionately enrol in lower-paying graduate programmes compared to their more privileged peers.
It is important to emphasise that our findings persist even after controlling for undergraduate programme choice and a wide range of academic and demographic factors. This suggests that, even among highly educated individuals, new inequalities emerge at the graduate education stage. While addressing disparities earlier in the educational pipeline remains crucial, our results highlight the need to examine graduate education choices and their implications for long-term earnings and social mobility.
Source: cepr.org
Artificial intelligence differs from other technological advancements in finance, such as the initial adoption of…
Industrial raw materials such as nickel, cobalt, and rare earths are critical inputs in countless…
As European governments scale up investment, bond market stability is more critical than ever. This…
Economists have long warned of the negative consequences of excessive US public debt (e.g. Friedman…
Financial distress affects roughly one in five adults in OECD countries (OECD 2024). It constrains…
Until 2018, the US-China trade data gap was in line with the discrepancies found in…