Is unemployment really just 10%? Not likely …

South Africa’s unemployment rate – officially 32.9% – has always been hotly contested.

Back in 2005, then-president Thabo Mbeki complained that it “seems quite unlikely that the Stats SA figure is correct”, because if there were millions of people roaming our streets looking for work, “nobody could possibly have failed to notice”.

For years, ANC politicians, trying to defend their political record, have argued that the official gauge of unemployment is far too high – and this week, they found an unlikely ally in Gerrie Fourie, the CEO of Capitec Bank.

In a surprising interview, Fourie told Business Day that the real figure is closer to 10%, since “Stats SA doesn’t count self-employed people”, saying “that is an area we must correct”. Presumably, based on his insight into Capitec’s 24-million customers, Fourie argued, rather like Mbeki did, that if a third of the nation truly was jobless, “we would have had unrest”.

Soon enough, politicians – who would love to believe they have been far more successful in fixing the economy than facts suggest – leapt on Fourie’s opening. Parks Tau, minister of trade, industry and competition, said in parliament that he agreed with Fourie that there “is a significant undercount, particularly if you take into account the informal sector”.

So, could Fourie and Tau be right? Not likely, say statistical and labour experts. While the jobless number might be slightly off on the margin, there’s pretty much no way Stats SA would have mistakenly tripled the unemployment number.

“Fourie’s 10% number is not helpful – there is no question we have a much more serious unemployment problem than this,” says Andrew Donaldson, the former deputy director-general of National Treasury and now a senior research associate at the Southern Africa labour and development research unit at the University of Cape Town (UCT).

Donaldson tells Currency there is no serious debate about whether Stats SA includes informal work in its Quarterly Labour Force Survey (QLFS) – self-employment, or any kind of work, unpaid or not, for at least one hour in the previous week, qualifies as “employment” in the national surveys.

That survey, of 30,000 households, was clear that this included anyone involved in “selling things, making things for sale, construction, repairing things, guarding cars, brewing beer, collecting wood or water for sale”.

Even providing “help without being paid” in their household would qualify.

The last survey, released last month, said that a fifth of all the 16.7-million people employed – 3.34-million people – worked in the “informal sector”.

Still, Donaldson says there is a legitimate question about how rigorous the survey process is: it is possible, for instance, that people who answer the survey might think that irregular informal work counts as “unemployed”, and so they will give the wrong answer. But this is unlikely to skew the answers as radically as Fourie suggests.

Andrew Kerr, a professor at UCT’s school of economics, is less charitable. He describes Fourie’s comments as “an embarrassment”, as it is clear to everyone that informal work is included in the employment statistics.

Like Donaldson, Kerr concedes that there are legitimate concerns about Stats SA’s QLFS, including poor quality data on earnings – but it is simply wrong to claim that people who are self-employed aren’t included in the numbers. “It may be that unemployment is a bit lower than the QLFS says, but there is no way the unemployment rate is 10%,” he says.

As it is, there are 16.7-million employed people. So, even if 6-million customers are depositing R15,000 or more every month into Capitec’s accounts, they could easily fit into the category of “employed” people in the QLFS.

Far-reaching effects

Were Fourie right, and Stats SA had got it wrong by such an immense factor, it would have colossal implications for South Africa’s economy and the integrity of all the financial information it publishes.

Foreign investors, after all, consider the country’s GDP growth rate when deciding whether to invest, and if the country’s jobless numbers are wrong by many multiples, why believe the growth number? Equally, the South African Reserve Bank makes interest rate decisions based on inflation numbers calculated by Stats SA, and if these are wrong, so much else would stand to fall.

As one senior government official told Currency this week: “You can’t easily dismiss what the chief executive of a major bank says – he knows what’s happening on the ground – but it seems implausible that those numbers would be so vastly incorrect.”

Critically, the official statistics body isn’t backing down. “We stand by the methods that we use to collect and publish our labour market statistics. They are solid,” said Risenga Maluleke, South Africa’s statistician-general, in the wake of this debate.

Maluleke says Stats SA follows the International Labour Organisation’s best practice on calculating the informal sector, and factors in all the available data it can lay its hands on, including surveys of employers and the self-employed, and reports on the informal sector.

“To suggest our unemployment is sitting as low as 10% would be misleading the nation,” he says.

He points out that if Fourie were correct on his 10% estimate, there would be 2.5-million people unemployed in South Africa’s 25-million-strong workforce, rather than triple that number.

One argument Fourie’s advocates make is that it would be artificially low to assume that only a third of South African businesses are “informal”. But Donaldson says international studies show that for various historical and structural reasons, “our economy has far less informal or household-based economic activity than most developing countries in Latin America, or Southeast Asia”.

Better data needed

Like many South African debates, there is some nuance to this, which means Fourie isn’t entirely wrong to raise this issue.

For one thing, Donaldson says Fourie is right to argue that informal and small businesses should be treated with far more importance in local economic initiatives and policy debates.

“There is faster growth in informal housing, building, trade, personal services, repairs and maintenance and other activities, than in the dominant formal economy. And sustainable job creation costs far less in townships and informal contexts than in the formal sector,” he says.

And Maluleke does admit that if South Africa had a proper register of informal sector enterprises, like that which exists for formal businesses, collecting data would be much easier. And no doubt, more correct.

Nonetheless, the point remains that if the figures are off, it is by a small percentage, rather than by whopping multiples.

Jonny Steinberg, who teaches at Yale University’s Council on African Studies, was eviscerating of Fourie this week.

He said that for the Capitec boss to simply assume the country’s official statistical office excludes informal activity “without bothering to check” suggests an intellectual laziness born from overconfidence. “Fourie degraded public discourse” by claiming that his own, untested views, based on his own intuition, were somehow correct, and everyone else was wrong.

And this error matters, Steinberg argued, precisely because we are in a world where powerful people are just making up their own facts, and ignoring what doesn’t suit them.

“Two influential men, Fourie and Tau, decided between them to rewrite the tale of South Africa’s unemployment crisis. They did so on the foundations of clear factual inaccuracy,” he said. The country, Steinberg said, needs a better public discourse – and this “debate” only retarded that effort.

_{Top image: Capitec’s Gerrie Fourie and statistician-general Risenga Maluleke.}

ALSO READ: