Mbeki and the jobs question: fact-checking employment from 1994 to 2008
- Mbeki understated the number of people employed in South Africa in 1994 by nearly a million, which meant he exaggerated the employment gains since then.
- Between 1994 and 2008, Statistics South Africa changed its survey methods three times, complicating direct comparisons. However, separate calculations by Africa Check and an expert put the increase in jobs during that period closer to 5 million, not the 6 million Mbeki claimed.
- Experts also warned that simply comparing raw employment numbers overlooks population growth. A more useful measure is the share of the working-age population in employment. This ratio grew by about four to six percentage points between 1994 and 2008, showing modest but real improvement.
In September 2025, former South African president Thabo Mbeki hosted his foundation's annual peace and security dialogue. During his address, one claim in particular sparked a public debate about employment.
"The number of people employed between 1994 and 2008 grew by six million," he said, from 8 to 14 million. Comparing gains made during that period to more recent times, Mbeki said the country had since gone in the "opposite direction".
Follow us on WhatsApp | LinkedIn for the latest headlines
When a clip of him making the claim was shared to X by the national broadcaster, hundreds of users questioned the accuracy of his figures. While some called them misleading, others praised Mbeki for his role.
We took a look at the data and asked experts to weigh in on the value of making direct comparisons like this.
According to findings from Statistics South Africa's October Household Survey (OHS), 8.9 million people were employed in 1994.
"The OHS collected labour market information for the whole country ..." Desiree Manamela, Stats SA's chief director of labour statistics, told Africa Check.
Though it has been criticised for having too few black respondents to be fully representative, it was considered a "benchmark ... in information gathering", surveying 30,000 households and for the first time including the nominally independent black "homelands" or Bantustans in country statistics.
Mbeki understated the number of people employed by almost a million.
Mbeki claimed that the number of people employed grew by 6 million. But there are issues with making direct comparisons.
In 1999, the OHS was replaced by the Labour Force Survey (LFS), which ran from 2000 to 2007.
Since 2008, the quarterly labour force survey (QLFS) has used updated sampling methodologies and definitions to bring it in line with various International Labour Organization standards. The results of the LFS have also been revised to make them comparable with QLFS data, Manamela said.
According to the QLFS data, 13.6 million people were employed in the third quarter of 2008. Subtract the 8.9 million employed in 1994, and it would appear that an additional 4.7 million people were employed at the end of 2008, over a million less than Mbeki claimed.
Dr Amy Thornton is a post-doctoral fellow at the University of Cape Town's Southern Africa Labour and Development Research Unit (SALDRU). Using a similar calculation method, she told Africa Check she came to a figure of about 5 million.
"There are certainly data quality concerns regarding the measurement of the surveys - but to some extent this is the case for almost every post-apartheid household survey," Thornton said. "Running large nationally representative household surveys is a difficult business, and pointing out data quality weaknesses does not always mean the data is 'wrong and unusable'."
According to Stats SA's Manamela, however, "data from 1994 cannot be compared with the current QLFS data". She told Africa Check that although there was continuity between the LFS and the QLFS, the OHS had a complex design and could not be "linked".
Other researchers have also noted the difficulties in comparing the OHS to later survey data.
We therefore rate Mbeki's claim as unproven.
Labour absorption rate
But comparability is not the only issue. Dr Neva Makgetla, senior economist at the Trade and Industrial Policy Strategies research institute, previously told Africa Check that looking only at raw numbers was problematic.
While the raw number of employed people has increased, so has the working-age population (ages 15 to 64), from 24.6 million in 1994 to 30.8 million in the third quarter of 2008. A more accurate way to look at the situation, Makgetla said, was as a ratio of those employed compared to the working-age population. In 1994, the ratio was 38.2%. In the third quarter of 2008, it was 44.3%.
Sign up for free AllAfrica Newsletters
Get the latest in African news delivered straight to your inbox
"I would say the employment-to-working age ratio - the share of the working age population that is employed; I call working age 18 to 59 years - did increase in this period by approximately five percentage points, from around 44 to 46% in the mid-90s, to 48 to 50% around 2008," Thornton said.
Employment did increase after 1994
While the data may be difficult to compare, experts agree that employment did increase post-1994. "Likely, a key factor ... was the end of apartheid and the extension of freedoms to all groups in the population," Thornton said. "Labour force participation increased, in particular for black African people and women."
Dr Miriam Altman, adjunct professor at the University of Cape Town's Nelson Mandela School of Public Governance, identified rising educational attainment and urbanisation as other contributing factors.
But that growth has stalled, particularly after the 2008 financial crisis. Thornton said that the economy had failed to recover, yet people, particularly young people, continued to join the labour market. When the economy cannot absorb them, the result is often unemployment.
