Explainer: 2.2 million lose their jobs… what’s up with SA’s latest unemployment stats?

Statistics South Africa

Statistics South Africa has just released the latest employment statistics in the form of the Quarterly Labour Force Survey (QLFS), and they reveal the true impact of the Covid-19 pandemic on employment in the country. As expected, the numbers released on Tuesday make for grim reading, but there are things that we can do in the short term to cushion the blow. And, importantly, it is possible that the numbers will improve in the coming months, albeit slowly.

In a nutshell: about 42% of the population are currently unemployed, according to the expanded definition of unemployment (more on that later).

(If you’re curious as to how we got here in the first place, read and watch Part 1 of our unemployment series where we dig into what’s behind our unemployment crisis here.)

Here’s what you need to know from Tuesday’s announcement:

More than 2 million people were left without a job in the second quarter of 2020, bringing the total number employed people to 14.1 million. At the same time, the unemployment rate has actually decreased, and is now at 23.3% from 30.1% in the first quarter. Sounds confusing, right?

But there’s a perfectly reasonable explanation for this strange statistical phenomenon. (Spoiler alert: unemployment is not, in fact, improving.)

First, let’s understand the definitions (which seems technical but is key to understanding the facts).

Stats SA has two definitions for unemployment: the expanded definition, and the official one.

The official one includes people who are not employed, are actively looking for work, and those who are available for work. It also includes those who had a potential job somewhere in the future, so they were not actively looking for work.

The expanded definition includes people who are not employed, but also those who were available to work but did not look for work, either because they were discouraged or for other reasons.

During the pandemic and the lockdown, thousands of people joined the ranks of those who were not only unemployed, but who actually stopped looking for work. And because these people typically fall into the expanded definition of unemployed people, the number of “unemployed” people increased.

Based on the expanded definition, the real unemployment rate is more realistically at around 42%.

And this brings us to the real story behind the numbers. They show not only how many people lost their jobs during Covid-19, but how many actually stopped looking for work during Covid-19. This could be because they simply gave up hope, or because they were physically unable to go out and look for work because of lockdown restrictions.

You can get a pretty good idea of what this looks like over time by this graph by Stats SA:

Let’s dig into the details.

What’s the QLFS and how is it conducted?

The QLFS is a survey conducted by Stats SA every quarter of the year. It looks at employment trends among 15 – 64 year olds in the country. Traditionally, much of this research is done via face-to-face surveying. But because of the pandemic, this was not possible, and Stats SA had to conduct it’s research telephonically. This presented a number of problems, like people not having phone numbers, or phone numbers that had been changed.

All this was accounted for in the statistical analysis, but it does have a bearing on the results, as Josh Budlender, a PhD candidate in economics at the University of Massachusetts, explains.

“Stats SA normally does in-person surveys, and successfully interviews around 90% of the people they try to contact. Some people don’t have reliable phone numbers or refuse phone surveys, leading to a response rate in this QLFS of 57%.”

He said this can cause two problems. Firstly, the sample might not be truly representative of the whole population. Secondly, the sample might not be comparable to other QLFS samples, making it hard to compare against other samples.

“Stats SA has done what they can to adjust the sample to make it similar to an in-person survey, but this is a genuinely difficult problem to solve. While the QLFS results look reasonable, we should treat the estimates as somewhat approximate, especially when comparing to Q1,” he said.

What do the numbers really tell us?

Firstly, that this is a global phenomenon. As Stats SA points out, the decline in employment in the last quarter has been seen across the world.

The International Labour Organisation estimated in May that 1.6 billion informal sector workers, who make up about half of the world’s employed people, were at immediate risk of losing their jobs.

Secondly, that the situation varies across provinces and education levels.

Here are some key findings:

  • The Eastern Cape has been the hardest hit, with over 50% of the population now unemployed according to the expanded definition. The Western Cape was the least-hardest hit, and now has an expended definition unemployment rate of about 27%. The other provinces fall somewhere in between.
  • The biggest losses in jobs occurred in the formal sector. You can see the changes in unemployment per sector here: graph 3
  • Jobs were lost across all sectors, with community and social services workers the hardest hit.

  • While over 80% of the employed population received a salary during lockdown, for more than 20% of these people, it was a reduced one. Education played a big role in whether people took a pay cut, and nearly a quarter of those with an education lower than a matric were more likely to have a reduced salary.

Well, that sucks. Is there any hope at all?

As the economy opens up, more people are likely to start looking for work, which is a good thing, according to Budlender. But oddly, this will make the official unemployment figure look worse, because the number of unemployed people includes those who have recently looked for work. But the really important number, the number of employed people, could increase in the coming quarter.

So, the economy could recover, albeit slowly. The best thing government could do right now, according to Budlender, is to cushion struggling businesses and to try to prevent a second Covid-19 wave.

“We can expect the labour market to recover to some extent on its own as we move away from the initial short sharp shock of coronavirus and level 5 lockdown. However, this recovery may be slow. As an immediate priority to help recovery going forward, we need to try to prevent a serious COVID-19 second wave, prevent more businesses from failing, and try to help cushion the blow with grants and other social support. In the medium and long run we need a serious jobs program and better unemployment insurance infrastructure. These things are obviously all difficult to do, especially all at the same time,” he said.

(As explain.co.za showed in our second part of our unemployment series, there are other solutions available, looking at the examples of Greece and Spain.

Watch our video below and read more here.)