Research and Policy Brief: Unemployment's Statistical Illusion - 17th November 2010

Unemployment remains a central obstacle to South Africa’s attaining the ANC’s 1994 promise of a ‘better life for all’. The current rate of unemployment is officially recorded at approximately 25%. This is significantly higher than that of almost any other comparable developing economy. Among young black African South Africans the figure now climbs to over 50%. Generating an environment to boost employment therefore remains a major policy priority for the Government as evidenced by the recently announced “new growth plan” and its target to create five million net new jobs over the next decade. Yet boosting levels of employment may perversely require a rise in unemployment before employment levels are able to show any substantial future growth.

Statistics on employment in South Africa are widely published. One way of presenting the data that is not widely used is the ‘labour market participation rate’. The ‘labour market participation rate’ essentially assesses demand for jobs in the country. It measures the proportion of people of working age (15-64) who are either employed or unemployed. In other words it measures the number of economically active people as a proportion of the working age population. It does not take into account people who are classified as ‘discouraged workers’ - those who want to work but have given up looking for work. It also does not take into account people classified as ‘not economically active’ such as housewives or full time students.


The labour market participation rate in the first quarter of 2010 was 55%. This is 10 to 15 percentage points lower than that of comparable developing economies. The rate has also dropped by 10% or six percentage points over the past decade from 61% to 55%.


This suggests that a diminishing proportion of people of working age actually wish to work. This is at first glance borne out by official data showing that while the working age population grew by 4.1 million over the past decade a total of 3.4 million people were classified as ‘not economically active’ over the same period. Only 114 000 were classified as ‘discouraged’ workers.   


This data is puzzling because one would have expected the economically active population (those wishing or needing to work) to have grown by a figure in the same ball-park as the increase in the working age population. Instead, the number of not economically active people grew by that staggering figure of 3.4 million and the number of economically active people by a paltry 538 000. In other words, the bulk of the increase in the working age population over the past decade has been officially classified as being ‘not economically active’.


Persons classified as ‘not economically active’ include, in addition to students and housewives, pregnant women, people who can’t afford to pay for transport to go and look for work, and ‘other’. It seems most unlikely that the numbers of these people could have grown so fast relative to the working-age population. Why are so many people apparently dropping out of, or not even entering, the labour market?


Are the proceeds of crime now so widely spread that fewer people need to work? Has the rollout of social grants – they are now paid to nearly 30% of the population as against less than 8% ten years ago – caused a decrease in the demand for jobs? Are far more people working in the informal sector than are captured in the official job statistics? Has subsistence agriculture grown to such an extent that more people can live off the land? Has there been so great a widening in the gap between the actual unskilled pay on offer and the minimum that would-be workers would rather stay at home? Is some of the Stats SA data simply wrong? There is a great need for a proper study of the unemployed based on extensive fieldwork.


It is worth noting that if the labour market participation rate had remained at its earlier level of 61%, then the number of unemployed people would today be much higher at 6.3 million than the current figure of 4.3 million. This would push up the unemployment rate from the official figure of 25% to 33%. In other words by classifying so many of the working age population as ‘not economically active’, and thereby bringing down the labour market participation rate, South Africa has by official classifications and definitions of statistics understated the seriousness of its unemployment problem.


This places the country in a peculiar predicament. If South Africa does stage any significant growth recovery and thereby creates the environment for more people to look for jobs then the labour market participation rate must be expected to show a rapid increase. That must initially cause unemployment levels to increase as large numbers of people previously classified as ‘not economically active’ become active and can therefore be counted as unemployed – at least until they find work. In other words one early indicator that economic circumstances in the country are changing for the better should be an initial, temporary, spike in unemployment numbers. Of course other growth, spending, and debt indicators would at the same time be showing an improvement.    


This will doubtless cause a political predicament for the Government if its critics were to read this initial increase in unemployment numbers as evidence that whatever policy was responsible for increased economic growth was causing higher levels of unemployment – a statistical illusion. The lesson is always to read South Africa’s unemployment figures in conjunction with changes in the labour market participation rate.    


That said, we have long puzzled over why our very high levels of unemployment have not led to greater social instability than we already see. Clearly, competition for jobs has fuelled the resentment of foreign Africans that has from time to time caused outbreaks of violence against them. Idleness and hopelessness may be among the ingredients of so-called ‘service delivery’ protests. Indeed we expect that maintaining very high levels of unemployment will increasingly serve introduce volatility into the political system in the country and that the ANC may ultimately be made to pay a political price by the very large numbers of unemployed people in the country – especially young black unemployed people.


We can take some encouragement therefore from the fact the Government now regularly and openly acknowledges the problem. We can also take some encouragement from the fact that there remains a pragmatic core within the ANC that promotes the importance of growth over greater state intervention, or even nationalisation, to solve the unemployment problem. But that is about as far as we can concede any credit to the Government for their handling of the unemployment problem. Getting serious about reducing unemployment in the country will require major policy adjustments on labour markets and education, as well as well as in the Government’s often hostile attitude to the private sector. Labour market regulation remains inappropriate for a country at our level of development and with our skills set. Trade unions exercise far too much policy control over what happens in our schools with the result that the minister of education in South Africa is in effect a shadow minister. The Government can therefore not do the things it needs to do to improve the quality of learning and teaching – let alone instil discipline in our schools. While the Government’s recent ‘new growth plan’ acknowledges the importance of private sector led growth and entrepreneurship, too many other Government policies, especially those relating to empowerment and employment equity, operate as disincentives to private sector investment. Our unemployment problem therefore has its current origins in official policy on business, labour, and education and can only be solved if policy shortcomings in those three areas are addressed.        






Total aged 15-64

27.2 million

31.3 million

4.1 million

Total not economically active

10.7 million

14.2 million

3.4 million

Total economically active

16.5 million

17.1 million

538 000

Labour market participation rate



-10% (or 6 percentage points)

Total employed

12.5 million

12.8 million

309 000

Total unemployed

4.0 million

4.3 million

300 000


-          John Kane-Berman and Frans Cronje


This article was first published in the Cape Times as part of a National Dialogue initiative launched in March by the Ministry of Economic Development in association with the Cape Times and the South African New Economics Network.  Copies of earlier contributions to the dialogue can be found via the SANE website,