Wanting to work – the story of South Africa’s labour market since 1994
That South Africa has an unemployment problem is well understood. Less well understood is the exact nature and full extent of the problem. This analysis seeks to shine some light into the darkest recesses of the labour market to determine just how serious the problem is, how and why it has arisen, who is most affected by it, what continues to underpin it and whether it can be resolved.
Contrary to more dire assessments, democratic South Africa’s first two decades were not an era of jobless growth. StatsSA shows that the number of employed South Africans increased from 7 971 000 in 1994, to 15 545 000 in 2016, or by 95%. This amounts to roughly 340 000 net new jobs per annum, or the creation of roughly 1 000 net new jobs per day. The number of employed black people more than doubled, from 4 980 000 to 11 506 000, or by 131% – making black South Africans the greatest beneficiaries of the expanding labour market.
There is more good news. In 1994, there were 380 South Africans not working for every 100 who were employed. That ratio has declined to 260 people not working for every 100 people employed – a trend that is consistent with an increase in the labour market participation rate. The rate measures the proportion of people of working age either working or looking for work. In 1994, it was 47.7%. It increased to a peak of 61.3% in 2000, and subsequently stabilised before slipping a bit in the aftermath of the 2009 global financial crisis to reach a level of 57.9% in 2016.
However, notwithstanding the progress made, and in fact largely because of the increase in the participation rate, the number of unemployed people has also increased – and at an even faster rate than the employed. On the official definition of unemployment, the number of unemployed people has increased from 1 988 000 in 1994 to 5 634 000 in 2016, or by 183.4%. On the expanded definition, the number increased from 3 672 000 to 8 880 000, or by 141.8%. The narrowly defined unemployment rate increased from 20% in 1994 to 26.6% in 2016, while the broad rate increased from 31.5% to 36.4%.
This, in turn, explains why the labour market absorption rate has remained low since 1994. The absorption rate measures what proportion of people of working age work. In 2016, it was 42.5%, while emerging market norms are a comfortable thirty percentage points higher.
Despite black people being the greatest beneficiaries of the expanding labour market, the burden of unemployment falls largely on their shoulders – a burden felt most acutely by young black people. In 2016, the narrowly defined black unemployment rate was 30.1% and the broadly defined rate 40.9%, while the white population recorded rates of 6.6% and 8.6% respectively. Among young black people (aged 15 − 24) the narrow unemployment rate was 58.7% in 2016, while the broad rate was an incredible 71.6%.
Being unemployed has also become a long-term affair. Since 2008, the number of people who have been unemployed for between one and three years increased by 18.9%, and those unemployed for between three and five years increased by 26%. However, those unemployed for more than five years increased by 52.9% – roughly twice the rate of the other two groups.
A sectoral analysis shows that the primary and secondary economic sectors – those most likely to absorb relatively unskilled workers – have performed particularly poorly. The number of mine workers has fallen from 692 900 in 1990 to 457 309 in 2016, or by 34%. The number of manufacturing workers decreased from 1 537 511 to 1 136 465, or by 26.1% over the same period. More highly skilled and services sector jobs have increased quickly. The number of government workers increased from 1 310 484 in 1990 to 2 079 987 in 2016, or by 58.7%. Data for financial and hi-tech services sector jobs is not strictly comparable over time, but the available data suggests a possible doubling in the number of such jobs to roughly two million. Changing sectoral employment trends, in turn, reflect the changing structure of GDP. In 1991, for example, the primary and secondary economies generated over a third of GDP. That figure has subsequently fallen to less than a quarter. Manufacturing saw its share of GDP fall by almost half.
The trouble is that levels of labour market absorption closely correlate to skills levels. The absorption rate for people with a qualification lower than matric sits at an average of around 34%. For people with matric as their highest level of education, the rate increases to just below 50%. But for people with a tertiary qualification (degree, diploma or certificate), the rate is just above 75% where, for the first time, it is comparable with international norms. Yet, only 28% of South Africans aged 20 and above have matric as their highest level of education and only 14% have a tertiary qualification.
A fair assessment is that, despite creating an average of over 300 000 net new jobs per annum since 1994, the rate of job creation has been too low to meet increasing demand for work, while the changing structure of GDP and the poorly performing education system have conspired to engineer uniquely low levels of labour market absorption for young black people. Reaching the equality milestone that would arise from reducing the black unemployment rate to the white rate (which is internationally competitive), and doing so over the next decade, will require doubling the number of black people with a job or creating roughly one million net new jobs per annum. We estimate that GDP growth rates of around 6% together with significant labour market deregulation could take South Africa halfway there by the late 2020s. Maintain that growth rate for 20 years and much of South Africa’s unemployment problem could be eradicated.
Frans Cronje is a scenario planner and CEO of the IRR – a think tank that promotes political and economic freedom.