Former Capitec CEO Gerrie Fourie caused a furore in June when he publicly stated that the country’s official unemployment statistics, as published by Statistics SA, were critically flawed and overstated the unemployment crisis.
The basis for his assertion seemed to be the mismatch between the declared unemployment figures — which are based on periodic samples undertaken by StatsSA — and the robust transactional activity Fourie observed on Capitec’s platforms.
Under this proposition the unemployment data is regarded as myopic as it focuses on narrow definitions of gainful employment and naturally leaves out those unable to be “boxed” into the traditional definition but who are nevertheless engaged in some form of economic activity — as shown by their transacting habits.
The idea that unemployment data — from which much national policymaking should stem — could be materially flawed naturally caused alarm across the political and policymaking landscape. The scale of the projected difference between the official rate and Fourie’s estimate — 33% versus 10% — would, if ever proven to be true, account for millions of citizens who are classified as unemployed being reclassified as something else.
For politicians eager to downplay the unemployment crisis the assertion was attractive, as it could be used to argue that things weren’t that bad in South Africa after all. Labour minister Nomakhosazana Meth duly published an article headlined “Capitec CEO isn’t wrong — unemployment data needs work”.
Amid the political noise, StatsSA had to defend the integrity of its data and went as far as engaging with Fourie to explain its processes. The problem underpinning the episode was that data collected for a specific purpose was being asked to explain why it wasn’t accounting for another crucial but different phenomenon.
The question of economic participation takes multiple forms, perhaps the most visible link between gainful employment and transacting capacity being found in the concept of remittances. Loosely defined as money sent to family members or friends within SA, remittances enable those outside the formal employment net to transact and hence fall within the ambit of the universe Capitec’s platforms track. Such participation reflects the habits of those with resources to share, rather than the robustness of the employment market.
Conflating these issues escalates the risk that those tasked with national economic and social planning responsibilities will assume that organic cures have found themselves rather than proactively working towards ensuring the maximisation of gainful employment opportunities for all those who wish to work.
The bigger risk with scepticism regarding StatsSA data is that any other proposition that claims to possess data that is different and even superior to official statistics could suddenly assume an amplified role in the social discourse.
This week FiberTime, a company dedicated to installing low-cost fibre countrywide, claimed that its granular, street-level data points to a far larger SA population than official figures suggest. The difference between its estimate of 95-million and the official population estimate of 63-million implies that South Africa has undercounted its population by half.
If this is the case it would represent an extraordinary set of coincidences where millions of people in a country with high reliance and dependency on public services have somehow mysteriously eluded the registration processes required to access these services. This is simply too fanciful to countenance and requires a clear rebuttal by StatsSA, outlining how such samples are captured.
Clarifying such complexities is the job of a properly capacitated Stats SA. In the absence of that, random statistics will keep emerging and risk making a mockery of the very idea of national planning.
• Sithole is an accountant, academic and activist.
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.