How has graduate unemployment evolved since 2008? We situate trends in graduate unemployment in the contexts of improved graduation rates, the shifting composition of graduates, the broader labour market, and public expenditure on higher education.
Introduction
Graduate employability is a key metric for the return on public (and private) investments in higher education. University students in South Africa receive some of the highest state transfers per capita. The budget justification and sustainability of this expenditure is hotly debated. On the one hand, public finance decisions have trade-offs, and there is a tension between expanding access to university when the quality of basic education is poor and unequal, and when unemployed working-age adults lack a dedicated form of social protection. On the other hand, graduates who become active participants in the economy contribute to the tax base (in addition to other things), contributing to economic growth and the availability of public funds.
A concern, therefore, is the rising unemployment levels among South African graduates.
Between 2008 and 2023 graduate unemployment more than doubled – from 5.8% in 2008 to 11.8% in 2023. In this article, we use South Africa’s Quarterly Labour Force Surveys (QLFS) to profile graduate unemployment, finding that it appears to follow a trend of rising unemployment in the labour market more broadly, but that young, African women are most affected. We conclude by recommending an opportunity to comprehensively understand graduate employability by linking administrative data sources.
Background
Higher education is widely regarded as a key catalyst of social mobility. Recognising this, transforming access and success has been key on the policy agenda, and several state-funded programmes exist to this end. These include, for example, the National Student Financial Aid Scheme (NSFAS), which expands access for those who would otherwise find university education unaffordable, and Extended Curriculum Programmes, which provide remedial education at the university level. Additionally, the state provides subsidies to public institutions for teaching, learning and research activities.
Equity in public higher education participation has improved over time. The composition of graduates shifted significantly between 2008 and 2023, reflecting improved access to higher education for female and African students. In 2008, African graduates made up 37% of the South African graduate population, while white graduates constituted more than half. By 2023, African graduates comprised over 52% of the graduate population. Similarly, the proportion of female graduates has increased from 46% to nearly 54% over the same period.
Positively, graduation rates have also increased over time (albeit from a low base). While on-time graduation rates remain low, the share of students graduating has increased year-on-year, such that close to 80% of students in degree qualifications graduate overall. However, graduation rates are very low for distance learning students: even with the improvements seen over time, and allowing for extended time to completion, at most 45% graduate. Unfortunately, a limitation of household survey data is that there is no information identifying respondents who enrolled but did not complete their qualifications. The analysis in the next section thus cannot speak to the employment return on the investments made in such students.
Profiling graduate unemployment
QLFS data between 2008 and 2023 indicate a rapid rise in unemployment rates among graduates, defined as those with a degree from a university – public or private. Graduate unemployment (broadly defined) increased by over 100% in relative terms, from 5.8% in 2008 to 11.8% in 2023 (by quarter 4 of 2024 it had fallen slightly to 10.2%).
Rising graduate unemployment is concerning, but as Figure 1 shows it remains significantly lower than for individuals with lower levels of educational attainment. In fact, those with other tertiary qualifications (e.g. diplomas and certificates with matric) experienced a rise in their unemployment rate from 11.6% in 2008 to over 26% in 2023, a notably larger increase than that of degreed graduates. The high unemployment rates for all education groups up to completed secondary education imply more limited employment prospects for individuals without tertiary education.
Note: Estimates are adjusted for survey weights and rounded to one decimal place. “Other tertiary” refers to individuals with post-secondary certificate or diploma qualifications from university or college.
Importantly, rising trends in graduate unemployment do not necessarily reflect a decline in the value of degrees in the labour market, or the quality of graduates. Although graduate employment prospects have worsened over time, they have done so to a lesser extent than general employment levels. The rising graduate unemployment rate is largely due to a sluggish labour market, with graduates facing similar adverse shocks as the broader labour market. That degree holders nonetheless experience the lowest unemployment rates compared with other education levels, suggests that degrees remain sought after in the labour market.
Of concern, however, is persistent inequalities in employment opportunities: graduate unemployment is highest among young, female, and African graduates. Graduates in their twenties consistently face higher unemployment rates than any other age group, surpassing 30% in 2023. The unemployment rate among African graduates in 2023 was 18%, compared with just 3.7% among white graduates. In addition, the female graduate unemployment rate is 4.5 percentage points higher than that of males, reaching 14% in 2023. These inequalities are less pronounced among graduates than in the broader labour market, suggesting that higher education helps mitigate them to a certain extent.
Compositional changes
Despite improvements in the demographic composition of graduates, the increase in African and female graduates has not sufficiently translated into a proportional increase in employment for these graduates. We use a propensity score reweighting method to examine whether graduate composition changes have contributed to graduate employment changes. The weighting adjustment serves to create comparable groups between 2008 and 2023, such that the reweighted 2023 graduate employment rate reflects the employment rate for graduates with a similar demographic composition as those in 2008.
Results show that the change in the composition of graduates does play a role in explaining the drop in graduate employment. This result suggests that the decline in graduate employment cannot simply be attributed to declining graduate quality. Given growing earnings returns to tertiary education, growth in graduate unemployment therefore signals a potential mismatch between the skills or qualifications of unemployed graduates and the labour market’s needs. That unemployment rates are lower among graduates who studied in certain fields (e.g. education and health) could be suggestive that there is a mismatch between supply and demand or that graduates from certain fields are inappropriately skilled.
Conclusion
The large investment into creating a university graduate warrants concern around graduate employability. Ensuring graduates find employment is key to maximising both private and social returns on investment in university education. Although graduate unemployment is rising, it is largely due to a stagnant labour market that affects all participants. Moreover, tertiary education, particularly for degreed individuals, remains valuable in terms of employment prospects and it appears that the employment return on investment remains considerable.
Nonetheless important challenges remain and there are gaps in information needed to create well-aligned policies. First, addressing inequalities by age, race and gender in the graduate labour market is critical, especially because this limits the extent to which transformative social mobility can take place. It is unclear whether inequalities are driven by labour-market discrimination, or rather differential sorting of demographic groups into qualifications, specialisations, or institutions with lower employment prospects.
Second, and relatedly, we do not yet know enough from survey data alone, about the factors that contribute to unemployment. For example, do graduates lack the necessary qualifications, are those with the required qualifications inappropriately skilled, or do too many have qualifications in specialisations that are not in demand. We also lack information on the return on state investment in students who do not complete their degrees.
Lastly, it is unclear how institutional factors are associated with employment. If targeted interventions at the institutional level are required, we cannot pretend that outcomes of university students are not linked to the institutions they attend.
To answer these questions, a labour-market intelligence system that links administrative data sources is crucial (e.g. school exit data, data on students and graduates of public universities, and tax record data). This would allow for longitudinal tracking of individual trajectories from secondary education through to the labour force. This major data advancement would offer low-cost, high-frequency, detailed insights to enhance policy monitoring and inform the design of sustainable and collaborative interventions in the higher education and labour market (and other) sectors. For example, such a system could inform enrolment plans and/or incentives to study for the qualifications and specialisations required in the labour market. This could go some way towards alleviating the potential mismatch between unemployed graduates and the labour market’s needs, maximising the employment return on educational investments.
References
Branson, N., & Lam, D. (2022). The Economics of Education in South Africa. In A. Oqubay, F. Tregenna, & I. Valodia (Eds.), The Oxford Handbook of the South African Economy (707–734). Oxford University Press.
Branson, N. & Whitelaw, E. (2025). Post-School Education and Training Policies - From 1994 to Now. Economic Research Southern Africa (ERSA) Policy Paper 34.
Branson, N., Kerr, A., & Muller, S. (n.d.). Towards equity in post-secondary education outcomes: What can we learn from a large-scale change to state financial aid in South Africa? (Report No. TBC). Council of Higher Education. Cape Town
DHET. (2022). Funding and expenditure trends in Post-School Education and Training
DHET. (2024a). Funding and expenditure trends in Post-School Education and Training.
DHET. (2024b). Statistics on Post-School Education and Training in South Africa: 2022.
DHET. (2024c). 2000 to 2022 First Time Entering Undergraduate Cohort Studies for Public Higher Education Institutions.
Kerr, A., Piraino, P., & Ranchhod, V. (2017). Estimating the size and impact of affirmative action in undergraduate admissions at the University of Cape Town. South African Journal of Economics, 85 (4), 515–532.
Wildschut, A., Rogan, M., & Mncwango, B. (2020). Transformation, stratification and higher education: Exploring the absorption into employment of public financial aid beneficiaries across the South African higher education system. Higher Education, 79 (6), 961–979.