In 2019, we published an analysis of graduate unemployment in South Africa using Stats SA’s Quarterly Labour Force Survey (QLFS). The analysis raised more questions than it answered. Sample sizes were small and did not lend themselves to meaningful segmentation. The headline finding that young, black, female graduates bore a disproportionate share of unemployment was plausible but fragile. We noted that better data exists (in HEMIS and SARS) and that someone should use it to explore this critical issue. That hasn’t happened yet. We’ve gone back to the QLFS (our only source of data), extended the analysis to 2025Q4, and asked the same questions again.
The picture is more troubling now than it was six years ago.
The graduate population has grown and changed shape
The supply of graduates in South Africa has more than doubled in under two decades. In 2008, there were 1.1 million working-age graduates. As of 2025, there were just under 2.5 million. This growth has not been demographically neutral. In 2008, the graduate population was roughly split: white males held the largest share at 27%, with black females at 20%. By 2025 those positions had switched. Black females now constitute 35% of all graduates, while white males have fallen to 13%.
Among young graduates specifically (under 35), the shift is even more pronounced: black women account for 39% of all young graduates in 2025, up from 24% in 2008. In absolute terms, this means roughly 350,000 young black female graduates are in the labour market today, a noticeable increase from roughly 86,000 fifteen years ago.
This is a significant transformation worth acknowledging. South Africa has materially expanded higher education opportunities for black women over the past decade and a half.
The harder question is what happens to them after graduation.
Graduate unemployment: worse than it should be
The aggregate unemployment rate for graduates sits at around 11% in 2025, which is almost exactly where it was in 2019 before the Covid shock pushed it to 15% in 2021. Relative to the broader labour market, graduates remain substantially better off: the unemployment rate for those with matric is approximately 35%. For those without matric it approaches 39%.
That differential is telling. A degree still represents a meaningful buffer against unemployment in South Africa.
But the aggregate figure conceals more than it reveals. Closer inspection of the data highlights a young graduate population that is inordinately affected by unemployment. Young graduates (under 35) face an unemployment rate of 18% in 2025, compared to 11% for the graduate population overall.
The gap between young and older graduates is nothing new, but it persists and appears to be widening.
In addition, within the young graduate population, the dispersion is stark: young white male graduates face an unemployment rate of roughly 3% compared to nearly 29% for young black female graduates.
This is not a rounding error. It is close to a ten-fold difference.
The racial and gender gap has not closed and in some respects has widened
In 2019, the unemployment rate for young black female graduates was 24%. In 2025, it was 29%. Over the same period, young black male graduate unemployment has edged down slightly, from 24% to 23%. The trajectories of these two groups, demographically similar in many respects, have quietly diverged over the recent years.
Small sample sizes give cause for caution. However, the 95% confidence intervals for 2008 and 2025 do not overlap. We are thus certain of the steady deterioration in employment prospects for young black female graduates.
What the data still cannot tell us
The QLFS remains the only available source for this analysis, and the samples have grown somewhat as the graduate population has expanded. But sub-group estimates still carry meaningful uncertainty. Confidence intervals are wide and year-to-year movements should be interpreted with caution.
Six years ago, we pointed out that HEMIS graduate records and SARS tax data contain critical information, and that linking them via a personal ID number would unlock a far richer analysis. That linkage still has not been done, at least not publicly or at scale. Meanwhile, universities are sitting on nearly R15 billion in unrecoverable student debt, SETAs are holding R26 billion in unspent levies, and the graduate unemployment rate for young black women has gone up, not down.
A thought worth sitting with
Young black women, while being the very demographic group that higher education has most successfully reached, is also the group most likely to be unemployed after graduating. They now comprise 39% of all young graduates and face unemployment rates nearly ten times higher than young white male graduates.
There are several ways to explain this gap.
- Networks and prior work experience matter for job-finding, and these are not evenly distributed
- Graduates from historically disadvantaged institutions may face structural disadvantages in the labour market that a degree alone cannot overcome
- Fields of study differ by demographic group, and some fields are more readily absorbed than others
All of these are plausible partial explanations, but none of them are fully testable with the data we have. What we can say, however, is that the story of South African higher education – growing, more inclusive and more female – is in friction with the story of the graduate labour market, which appears to provide an uneven return on investment for a degree.
A degree still pays. It just does not pay equally, and the gap between who holds degrees and who the labour market rewards for holding them does not appear to be narrowing.
Understanding why we see these outcomes require good analysis on sound data that exists but remains, frustratingly, out of reach.
