Unemployment is worse than the official figures state

The latest figures might show that the unemployment rate has dropped to its lowest level since Statistics SA started measuring unemployment using its Quarterly Labour Force Survey (QLFS) in 2008, but the number of unemployed people in SA has actually reached a new record.

The number of people who work dropped to only 14.1 million – from a total working age population of more than 39 million – indicating that the majority of people in SA don’t work and don’t earn an income to sustain themselves.

Yet Stats SA calculates that the unemployment rate decreased from more than 30% in the first quarter of 2020 to 23.3% three months later.

This big drop in the unemployment rate seems totally wrong during a period when large parts of the economy were shut down completely and a lot of companies closed down.

The figure feels wrong for a reason …

A closer look at the figures reveals that Stats SA apparently classified people who could not work during the Covid-19 pandemic as not economically active due to “other” reasons.

The number of people deemed to have excluded themselves from the labour force increased by more than five million, decreasing the labour force and the number of unemployed by around the same number.

Effectively, Stats SA is simply ignoring more than five million unemployed workers in its calculation of the unemployment rate.

Read: Unemployment rate hit record high before virus

The first table in the QLFS report shows that the labour force decreased from nearly 23.5 million in the first quarter to 18.4 million in the second quarter, while the number of workers classified as non-economically active increased from 15.4 million to nearly 20.6 million.

Key labour market indicators (000’s)
	Q1 2020	Q2 2020
Population 15-64 years	38 874	39 021
Labour force	23 452	18 443
Employed	16 383	14 148
Unemployed	7 070	4 295
Not economically active – discouraged work-seekers	2 918	2 471
Not economically active – other	12 504	18 107
Rates (%)
Unemployment rate	30.1	23.3
Employed/population ratio (absorption)	42.1	36.3
Labour force participation rate	60.3	47.3

Source: Stats SA QLFS, June 2020

Siphamandla Mkhwanazi, senior economist at FNB, describes the decline in the official employment rate as artificial.

“The data showed an unprecedented decline in the number of employed persons, by 2.2 million to 14.1 million compared to the first quarter of year. However, and almost counterintuitively, unemployment also declined, by 2.8 million compared to the first quarter.

“Combined, these changes resulted in a significant 6.8 percentage point decrease in the official unemployment rate from 30.1% in the first quarter to 23.3% in the second.

Not a true reflection

“Needless to say, the fall in unemployment is not a true reflection of the current labour market conditions, but rather a technical issue in the definition of official unemployment,” says Mkhwanazi.

He says 5.2 million people were included in the economically inactive category because they were hindered from actively looking for employment by lockdown restrictions (and are thus excluded from the technical definition of unemployed).

“Importantly, the expanded unemployment rate – which is more encompassing and thus more useful under the circumstances – increased to 42% and is a true reflection of the severely weak labour market conditions,” says Mkhwanazi.

As always, statistics should be read carefully and considered even more carefully. This is particularly true of the employment statistics due to the somewhat rigid definitions used to calculate the end result.

Delving into the definitions

The very first figure begs for closer analysis: the working age population is defined as people older than 15 and younger than 65, creating a big problem in that the majority of children younger than the age of around 18 should not actually be included in the labour force as they are still at school. The numbers of students at tertiary institutions is also increasing.

Scholars and students are excluded from the labour force figures as non-economically active, together with those who retired earlier than 65 or people who do not work by choice, such as a stay-at-home parent.

While the calculations might be statistically correct, in the end – and in line with international guidelines – the 20.6 million people classified as non-economically active in SA can hide a lot. In the QLFS report this category hides those extra five million unemployed people without much of an attempt to explain it.

The Stats SA report says that in spite of the massive decline in employment, the number of discouraged work-seekers decreased by 447 000 and the number of people who were not economically active for reasons other than getting discouraged by not finding work increased by 5.6 million in the quarter, resulting in a net increase of 5.2 million in the not economically active population.

Misty explanation

The only comment on this was: “A decline in employment, accompanied by a larger increase in inactivity other than in unemployment, has been observed in most countries across the world.”

According to the Stats SA report: “It was observed that a large number of persons moved from employed and unemployed status to the ‘other not economically active’ category between the two quarters.

“The movement was proportionately more for the unemployed than for the employed, which resulted in a significant decrease of 6.8 percentage points in the unemployment rate to 23.3%. This is the lowest unemployment rate recorded since the start of the QLFS in 2008,” says Stats SA.

But the reason for the exodus of unemployed people from the active labour force seemed not to have attracted further attention.

Some clarity

Mkhwanazi nearly gets to a possible reason in his commentary on the unemployment figures when referring to Stats SA’s remarks on the impact of Covid-19 on income levels and his view on consumer spending over the next few months.

“The figures show that the impact of the pandemic has been disproportionately distributed, with lower income households more severely affected. Indeed, Stats SA estimates that approximately 90% of employed graduates continued to receive a full pay, compared to 75% of those with less than matric as their highest level of education.

“Looking ahead, we expect a continued weakening in labour market conditions, possibly extending into 2021,” says Mkhwanazi.

“So far, it appears that Ters [the Unemployment Insurance Fund’s Temporary Employer-Employee Relief Scheme] payments and the government’s support via extended social grants programmes have supported consumer spending.”

Is it possible that larger grants, albeit that they are temporary, motivated people to opt out of the labour force?

Stats SA’s comment that unemployed people were more likely to move to the non-economically active population seems to support this.

If correct, this will probably be temporary as well and unemployment will show a big ‘artificial’ increase in future surveys when special grants dry up.

Read: In South Africa, even skilled workers seek jobs on the street

The important figure to watch in the coming surveys remains the number of employed workers. This raw figure will show if SA will win back the more than two million jobs it lost in the last few months and if it can create jobs for some of the five million people who are actually out there looking for work.