South Africa’s official unemployment rate sits at a dismal 32.9%, with the “expanded” rate topping 41.9% – a figure routinely cited as the highest in the world. It’s the kind of statistic that has become political shorthand for failure, repeated in headlines, debates and investment risk briefings.
Capitec CEO Gerrie Fourie set the cat among the pigeons last week, claiming that South Africa’s “real” unemployment rate may be closer to 10% – a claim quickly attacked by Stats SA and left-wing columnists.
It is crucial in this debate to know how both sides are arriving at the statistics behind their claims.
This is particularly true because it is possible that both Stats SA and Fourie are right from their own perspectives – at least in the sense that Stats SA is not inaccurately reflecting what its survey data shows, and Fourie is correctly extrapolating from the data he sees on his bank customer readouts.
Stats SA claims it is using definitions that comply with the International Labour Organisation (ILO). If you didn’t work at all in the previous week when you were surveyed – even for just one hour – but are actively looking and available to work, you’re “unemployed”.
The working age population of South Africa is 41.7-million people, with 16.7-million not economically active. Then there are people who are available to work but aren’t looking, called the “discouraged” group – an extremely large group.
The way this panned out in this year’s first Quarterly Labour Force Survey, for example, is this: the survey tallied the employed at 16.8-million, including 3.3-million informal workers. And it counted the unemployed at 8.2-million, which gives you your 32.9% unemployment rate. Throw some of the “discouraged” category back into the group, and you get Stat SA’s “expanded” definition, which totalled South Africa’s unemployment at an extraordinary 41% in this survey.
Fourie argues that Capitec’s data suggests that just under 3-million customers earn an income that does not originate from a formal salary, and more than 1-million run a sole proprietorship using their Capitec accounts.
That implies there are 4-million small and emerging businesses, formal and informal, roughly 1-million more at his bank alone than Stats SA is tracking, even if you include all its “informal and self-employed” in the bank’s numbers.
Capitec’s customers are drawn largely from township dwellers, so you would imagine they have a higher proportion of customers in this category. But still, it’s only one bank: Capitec banks about 30% of South Africa’s total banking clients. So assume all the other banks together have an equal number of customers in this group, and you could get to about 8-million customers in this category collectively. That’s at a guess.
So do the math: the base is 25-million economically active South Africans, 16.8-million of whom are employed (including the 3.3-million in the “informal” group). Add the extra 4.7-million (8-million people minus the 3.3-million Stats SA says are “informally employed”), whom the banking industry might consider employed, and you get to a roughly 14% unemployment rate. That’s a bit off Fourie’s 10% estimate, but it doesn’t seems impossible to me.
And bear in mind, Stats SA is doing a survey; there are additional questionnaires that go to employers, and the process is not casual, but generally, it’s about asking people questions – from some 30,000 households in total.
So why would people confronted with a person figuratively holding a clipboard say they are “unemployed” if they are not?
I guess that the answer has to do with what is generally considered to be “employed”. Because South Africa has a dual formal/informal economy, it’s not a huge leap to think many people would regard being “employed” as being technically employed with a contract and a UIF payment.
Or they may have other reasons to avoid being categorised as “employed”, because if they do, the South African Revenue Service will be onto them. Or they might consider that admitting this would disqualify them for a grant.
Out of sync
Just compare for a moment, South Africa’s unemployment rate with other developing countries, and more specifically the proportion of “employed” people who comprise the informal sector.
How is it possible that Rwanda has a 1.6% unemployment rate? Simple: 64% of people are “self-employed” but counted as “employed”. Magic! In Kenya, the informal group is even larger. So, is it really possible that South Africa’s informal sector is half the size, proportionately, of Rwanda’s? Or India’s? Smaller perhaps – but basically half?
Fourie’s numbers could be wrong in one important respect: the base could be too low. It certainly looks too low by international norms. In most developing countries about two-thirds of the population would normally be economically active. Stats SA is setting that figure at 25-million people; that’s barely 40%. If that’s the case, then the narrow definition of unemployment would jump to a massive 44%, even taking into account Fourie’s estimates.
But if that were the case, doesn’t it strengthen rather than weaken the argument that South Africa’s informal sector is much larger than the 3.3-million pegged by Stats SA? Doesn’t that tally with GG Alcock’s Kasinomics observations. And explain why it is that when Checkers advertises for drivers for its Sixty60 motorbike delivery service, only foreigners seem to apply?
There is another way of looking at this: let’s assume that South Africa’s level of informal employment is the same proportion as other developing countries, i.e. 50% rather than 32%. And, by the way, the reason we do this is because the implicit argument of Stats SA and all the left-wing commentators Rob Rose quotes is that a big chunk of South Africa’s population must be comparatively largely lazy, inept or incompetent – ironically exactly the same argument apartheid-era apparatchiks suggested. Let’s assume that isn’t the case, and re-run the numbers. So you would increase the proportion of the informal employed to 50% of the (expanded) estimate of the economically active population. That suggests an unemployment level of about 23% – which seems to me more or less right.
So, why would organisations like the ILO and the World Bank accept all these figures as correct?
The answer is that international organisations tend not to second-guess the home government. Rwanda wants to present itself to the world as having a 1.6% unemployment rate – which doesn’t come close to recognising the level of poverty in the country, but the international organisations tend to go along for the ride. There is no other conclusion because the numbers are obviously inconsistent.
The bigger question is why South Africa accepts a rather inflated unemployment figure? As Rob Rose suggests, doesn’t this make the government look bad?
Well, my feeling is that the whole philosophical orientation of our government is that it wants to intervene in the economy. That’s always been its fundamental raison d’être. And that philosophy is massively underpinned by the claim to the “world’s worst unemployment rate”. Intervention is the starting point. Under that rubric, everything else follows: high taxes, masses of red tape, stringent labour legislation, and onerous BEE rules. Everything.
So the short answer is that it does make South Africa look bad … internationally. But that’s a small price to pay for the ideological authorisation to jump into the economy with your boots on, not to mention buying votes with give-aways and bonsellas posing as social programmes.
And it’s underpinned by another piece of ideological rubbish: that South Africa’s core economy has a monopoly structure. This is a country with ten “high-street” banks, for heaven’s sake, twice as many as most European countries.
None of this is to deny high poverty rates, or South Africa’s poor economic condition, or pockets of great need. No country should be relying on its informal economy to boost its “employment” levels. South Africa’s labour market is undoubtedly in deep trouble.
But think about the warped policy that results from a distorted high technical unemployment rate. South Africa is now handing out social welfare to 28 million people, a remarkable 44% of the population. The amounts involved are mostly very small, yet if Fourie is right, then South Africa should be focused on giving more to fewer.
Make welfare meaningful, and give it to people who really need it.
And, if Fourie is right, then the government should be redoubling its efforts to encourage the informal economy to formalise. The problem is, the programmes to encourage formalisation are half-hearted and firmly resisted by the people in that informal economy because the tax rates and compliance terms are so onerous. The underlying suspicion of the government is, I would guess, something of a hand-me-down from apartheid days.
In 30 years, the democratic government hasn’t managed to convince informal businesses that it won’t unduly grab their money and make unreasonable demands on their fragile enterprises. That’s what happens if you spout a consistent anti-business message – something which is very visible in consistently low business-confidence levels.
This, of course, just compounds the problem and it is why, even by Stats SA’s figures, the unemployment level has been getting worse. And, after all is said and done, that is the rub.
ALSO READ:
Sign up to Currency’s weekly newsletters to receive your own bulletin of weekday news and weekend treats. Register here.
