The great labour market statistics debate of 2025 has shown that ignorance about the measurement of employment and unemployment is on steroids. Capitec CEO Gerrie Fourie started the debate when he said Stats SA did not count self-employment and that the official unemployment rate was closer to 10% than the reported 32.9%. A debate in parliament’s standing committee on finance was like a comedy show as ANC MPs showed they had the statistical literacy of children in a school playground.
Lusizo Makhubela said the expanded definition of unemployment included housewives who chose not to work. Seaparo Sekoati said the definition included people who were engaged in the informal sector, which was not accounted for. “There is no crisis in this country. Yes, we have challenges, but it is not a crisis,” he said. In a rambling essay, employment & labour minister Nomakhosazana Meth made so many false statements it is difficult to decide where to start. We have a problem if the minister of labour does not understand internationally accepted definitions of employment and unemployment.
Since 1994 there have been three main surveys that collected labour market statistics: the annual October Household Survey (OHS) from 1995-1999, the biannual Labour Force Survey (LFS) from 2000-2007 and the Quarterly Labour Force Survey (QLFS) since 2008. Stats SA says it cannot account for labour market statistics from the OHS. In 2004, stung by trade union criticism about “jobless growth” the government started questioning the statistics. The economy was recovering from the disastrous 1996 Gear macroeconomic policy, but that there was a limited increase in employment.
Former president Thabo Mbeki raised concerns with then IMF MD Rodrigo de Rato when he visited SA in September 2004, and the head of the IMF’s statistics department identified two experts to review the situation. Stats SA then commissioned World Bank adviser Alfonso Rodriguez Arias and IMF staff member Richard Phillips to look at the LFS. In 2005 the two economists produced a report, and Stats SA decided to re-engineer the LFS. A 2008 Stats SA report outlined the recommendations, which included the frequency of the survey and changes to the questionnaire.
The 66-page QLFS questionnaire now asks: “Did you work for a wage, salary, commission or any payment in kind (including domestic work), even if it was only for one hour?” It lists examples as “a regular job, contract, casual or piece work for pay, work in exchange for food or housing, paid domestic work”. The next question asks: “Did you run or do any kind of business, big or small, for yourself or with one or more partners, even if it was only for one hour?” Listed examples include guarding cars, selling things, making things for sale, collecting wood or water for sale, and taxi or other transport businesses.
Maria Payet, the International Labour Organisation’s statistics specialist, who works at the Pretoria office and is responsible for 20 countries in the region, says: “The QLFS is the best survey in the region and within the Southern African Development Community.” Trade & Policy Strategies senior economist Neva Makgetla says: “The breadth of the definition of employment in the QLFS emerges from its findings on the number of hours worked. For the first quarter of 2025 4% of people recorded as informally self-employed worked no more than a single hour every week; 7.5% no more than an hour a day; and 22% worked 20 hours a week or less.” It is difficult to understand how there could be a broader definition of employment.
In an interview on Kaya FM Fourie backed up his “alternative facts” by saying FinScope had estimated that there were 10-million to 13-million people employed in the informal sector. Author GG Alcock had put the figure at between 9-million and 10-million people. Fourie’s statement that Stats SA does not measure self-employment is false. Kingstone Mutsonziwa, Finscope’s executive for data and analytics, says Fourie has “got lost in the numbers” and used the organisation’s estimate of all people employed in SMMEs. “If we eliminate data overlaps and double counting, our estimate is that there are about 3-million people working in the informal sector.” Alcock says: “That is not my number.”
To the ANC MPs, the expanded definition of unemployment does not include housewives. They are classified as not economically active and not part of the labour force. The definition also does not include people who are in the informal sector. Meth said: “Millions officially classified as ‘unemployed’ are actively engaged in vital, though precarious economic activities such as street vending, waste recycling, home-based production and numerous micro-services.” But this statement is a fabrication.
According to the QLFS, there were 489,000 street vendors and 94,000 waste collectors during the first quarter of 2025. Stats SA says that this data must be used with caution since the survey is not meant to drill down to this level of detail. One would expect the minister of employment & labour to know that home-based production and subsistence agriculture are not part of employment, though the QLFS counts them as part of people involved in nonmarket activities using ILO reporting guidelines. The QLFS found that there were 2.5-million subsistence farmers during the first quarter of 2025.
Meth continued: “Standard definitions, which prioritise formal employment structures like fixed hours, registered businesses and regular wages, fail to capture this fluid, irregular and self-directed work, misrepresenting significant economic participation as idleness.” This is fake news — the scale of the minister’s ignorance about official definitions that are based on ILO guidelines is mind-boggling. The department has been hollowed out. In the past, its labour market policy directorate was a respected think-tank that had economists and statisticians, some of whom were trained at the US Bureau of Labor Statistics. It had great minds such as the late Guy Mhone, the renowned development economist.
The department should be leading debates about the unemployment crisis and the effect of the platform economy, climate change and artificial intelligence on jobs. The department, not the National Treasury, should be doing the review of active labour market policies. Instead, it is nowhere to be seen on any burning issue, except to draft a labour migration policy that panders to xenophobic sentiments.
Our public sector is broken but Stats SA produces an excellent labour force survey despite having a 22% vacancy rate because of the Treasury’s irrational austerity policies. The QLFS and other related Stats SA publications should form the basis of national debates about how to solve our jobs crisis and achieve full employment. We should not wish away our unemployment crisis.
• Gqubule is an adviser on economic development and transformation.
Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.
Please read our Comment Policy before commenting.