Fifteen years ago there was a raging debate between Stats SA, Adcorp and economists about the size of SA’s informal sector.
Adcorp economist Loane Sharpe, who was on the right of the ideological spectrum, had developed the Adcorp employment index (AEI) based on “alternative facts” about the country’s unemployment crisis. He was critical of Stats SA’s quarterly labour force survey (QLFS).
According to UCT economists Andrew Kerr and the late Martin Wittenberg, in December 2010 the AEI increased the number of employed people in SA by almost 50% to 19.2-million using a currency demand method to estimate the number of people employed in the informal sector. This increased the number by 6.2-million to about 8.5-million.
That had the unfortunate outcome of making total employment larger than the labour force. To avoid producing a negative unemployment rate, Adcorp added 2.5-million “illegal immigrants” — without explaining how it estimated the number — to arrive at an unemployment rate of 5%. The unemployment crisis had vanished.
At the time statistician-general Pali Lehohla said the AEI was “spewing bile” and that its methodology was “ad hoc rubbish”. He now recalls: “I scalped Sharpe’s head.” An Adcorp director called him for a meeting and the company dropped the index soon afterwards.
This week the debate erupted again when Capitec CEO Gerrie Fourie told Business Day the official unemployment rate would be 10%, not the reported 32.9%, if the vast informal sector was considered. If this is true, Fourie’s “methodology” has increased the number of people employed in the informal sector to 9.1-million, a 5.7-million or 171% increase on the 3.3-million reported by the QLFS. And the number of unemployed people has declined to 2.5-million from 8.2-million. The informality rate has increased to 54% from 19.9%.
Fourie stated — falsely — that Stats SA does not count self-employed people, and that the agency was therefore ignoring the country’s “large emerging market” and describing such people as “discouraged”.
There were 6-million small, medium and microenterprises (SMMEs) in SA, Fourie said, including 3-million formal businesses and 3-million emerging businesses. While the AEI’s quack methodology increased informal employment by 6.2-million, Fourie provided no evidence to back up his claim that there could be 5.7-million more people employed in the informal sector than in the official stats.
But the implication that there is more money in circulation in townships — there were informal entrepreneurs who had turnover of R1 ,000 a day, Fourie said — and that one can extrapolate this to develop precise labour market statistics with time series — has echoes of the AEI’s currency demand methodology, which was discredited by many economists. Let me be clear: using money in circulation to make estimates for the labour market is voodoo statistics.
Stats SA releases four major publications with labour market statistics. These are the QLFS (a household survey), the quarterly employment survey (a survey of formal enterprises and national, provincial and local government entities), Labour Market Dynamics (an annual publication) and the survey of employers and self-employed (SESE), which has 71 pages of data about the informal sector. Lehohla said these statistics should be seen as part of a system. “They are run independently but many of their findings support and correspond with each other. They say the same thing.”
This shows that the informal sector is a desperate, last-ditch survival strategy for poor people.
From the fourth quarter of 2008 to the first quarter of 2025, according to the QLFS, the number of people employed in the informal sector increased by 1-million to 3.3-million and the informality rate increased to 19.9% from 16% as the economy created only 1.2-million formal sector jobs. The Trade & Industry Policy Strategies’ (Tips) State of Small Business in SA publication, which uses Stats SA data for 2024, found that “only 6% of all working aged adults are employers or own account workers in SA, compared with 20% in other upper middle-income countries outside China”.
There were 700,000 formal small business that employed fewer than 50 people each. The informal sector had 1.7-million businesses. Median incomes in the informal sector ranged from R5,000 for employers to R2,000 for waged workers and R1 ,200 for own account workers. This shows that the informal sector is a desperate, last-ditch survival strategy for poor people. It is not clear where Fourie obtained his figure of 6-million SMMEs, which is more than double the Tips estimate using official data.
SA’s low informality rate is a development puzzle for many economists. “The data is not the issue, the low informality rate is a separate issue,” says Haroon Bhorat, a UCT labour economist.
According to the International Labour Organisation, the average for upper middle-income countries is 52.3%. The data also shows that informality declines as countries become richer, from 88.5% in low-income countries to 13.5% in high-income countries. Nobody disputes that apartheid legacies, where black enterprise was brutally suppressed, especially in urban areas, have contributed to the low informality rate.
There are two further explanations, informed by different ideological leanings. The mainstream view, outlined in a paper by Harvard University and UCT, blames regulations, laws, licensing, spatial inequalities, crime and infrastructure. While these explanations have merit, SA will not suddenly have an informal sector jobs boom if there is full deregulation.
There are only so many jobs that can be created in an economy that has excess supply (or inadequate demand) and has grown by an average of just 1.1% a year over the past 16 years. There could be saturation and further declines in already low earnings as more informal enterprises slug it out for a share of overcrowded markets where there is no demand.
In a classic paper for Tips, Kate Philip highlighted the centralised, monopoly structure of the core economy. The structure of the economy — big-farming, big retail, big manufacturing — limits opportunities on the margins that are available in other developing countries, where livelihoods in agriculture are often the first stepping stone into the informal sector.
In SA, markets in poor areas are thin and most people buy a limited basket of consumer goods that are mass produced in the core economy on a scale that makes it difficult to compete. It is hard to find a single item in a poor person’s shopping basket that is not mass produced in the core economy. Therefore, it is unrealistic to expect people to self-employ their way out of poverty. Under present conditions self-employment is a poverty trap.
“It is tragic when a businessman like Fourie dabbles in systems he does not understand... Just because companies like Capitec can make a fortune aggregating the incomes of poor people in townships does not mean there is no poverty at the level of individuals. There are many Stats SA publications that show the depth and scale of black poverty in townships and rural areas,” Lehohla said.
Fourie should withdraw his statement, which minimises the pain of 12.7-million unemployed people in SA.
• 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.