Statistics show an increase in job creation over the past four years, but the real story is not so rosy Jeremy Baskin
Perhaps, like me, you thought the economy was shedding jobs. Well, think again and take a close look at the figures in the table below. According to Statistics SA (in statistical release P0317) about 1,1-million jobs were created between 1996 and last year. That’s right, more than one million new jobs! Now this is remarkable for it suggests that, despite our sluggish growth rate, the economy is creating jobs … and creating lots of them.
But, of course, the devil is in the detail. The real story is less rosy and sometimes as complicated as a Florida voting ballot. What are the basic facts about jobs in P0317? We can list five.
First, there are about 10,4-million jobs in the South African economy. This includes full-time jobs and part-time jobs. It counts the factory worker, the clerk, the manager, the taxi driver, even the street hawker and the “homeland” farmer. In short, all jobs. The key indicator in the jobs debate is the total number of jobs in the economy and the revised figures are much higher than previously thought. This is good news. But …
Fact number two: total jobs in the formal sector declined by just more than 1% between 1995 and 1999. This is a problem. We need an increase not just in total jobs but also in the number of relatively good jobs in modern sectors of the economy. Still, the new figures contain two hopeful signs. An overall drop of 1% is much less than was thought. And between 1998 and last year total formal-sector jobs actually increased by 174000. Gone is statistical support for the oft-heard claim that the last few years have seen 500 000 jobs shed. The latest figures may mean we have turned the corner. Or it could be just a blip on the statistical horizon. We will have a better idea when the figures for this year are released. While retrenchments are rampant in some sectors, the revised figures are detecting job creation in previously unmeasured sectors. These include jobs in micro-lending services and cellphone- related companies.
Fact three: the informal sector is where the job gains are happening. In 1996 informal employment and domestic service totalled 1,7-million jobs. By last year the figure had risen to 2,7-million, a 56% increase. Of course, the informal sector is a wide category. It includes self-employed professionals. It may also include some formal-sector jobs which have been outsourced. But mostly it is made up of people working in survivalist activities and earning very little.
Fourth, the official unemployment rate has been rising steadily. It stood at 15,5% in 1995, rising to 25% in 1998, before falling to just more than 23% last year. This is a high level of unemployment, even for a developing country. And you are more likely to be unemployed if you are African or female. But the official unemployment figures significantly under- estimate actual unemployment and Statistics SA also releases “expanded” figures. In reality the unemployment rate is undoubtedly higher than the official 23% figure but lower than the expanded rate of 36%. A fair guesstimate is that unemployment is slightly above 30%.
The unemployment rate is also affected by demographics and participation in the economy. The rise in the unemployment rate is partly a result of a growing population. But, importantly, a growing proportion of the population is in the jobs market. In the short term even real increases in employment will be accompanied by rising unemployment figures. Attracted by more jobs being available, or driven by poverty, more people will enter the jobs market. This means that for the next few years, if we see the unemployment percentage rise, we must analyse what is caused by job losses (a bad thing) and what is caused by increased labour force participation (basically a good thing).
Which brings us to the fifth and final point. How many jobs do we need to create? The working-age popu- lation is now growing by between 500000 and 600000 a year. Our labour-force participation rate is about 50%. So, in simple terms, we need to create between 250 000 and 300 000 jobs a year just to keep up with the growing labour force, and most of these should be formal-sector jobs.
Why should we believe the figures? Why should we have confidence that this data is better than what went before? After all, there are some disturbing features. Can agricultural employment really have risen so dramatically? And is it really possible that total 1997 employment now being reported is 1,1-million higher than reported in an earlier Statistics SA publication?
But overall the latest figures are probably more reliable than previously. Stats SA is trying to tackle long-standing weaknesses regarding labour statistics. They say they are getting better at collecting the data. They have integrated data from a range of sources. And studies have shown that the best big-picture data comes from reworking the October Household Survey, which is what these new figures do. The figures are not very good at giving sectoral detail. But they are the best guide to the big numbers around employment and unemployment.
The new figures undoubtedly tell a new story. And they have important policy implications. Previously the figures showed a very damaged relationship between economic growth and employment growth. The new figures suggest something different that growth of about 2,3% a year (between 1995 and last year) is associated with 1,3% yearly employment growth. This is very good news, suspiciously good. It implies that improved growth rates, even if not astronomically higher, could have significant job-creation effects. This supports the trade union argument that the government should focus on higher growth rates rather than on labour reforms. But much depends on whether the next set of figures (for 2000) is equally encouraging. And it depends on our economists giving us a better idea of the long-term relationship we can expect between growth and both formal and informal employment.
Another startling aspect is the changing relationship between informal employment (which is growing sharply) and formal employment (which is not). Informal and domestic service now makes up 26% of total employment. This is still substantially below most other developing countries. Compare it to Chile, a relatively successful economy, where infor- mal employment comprises 37% of the total. The growth of informal employment will continue for some time. At one level this is not a problem. It is a survivalist strategy in the absence of formal jobs. And there are now more opportunities for trading and self-employment.
The problem lies in the deep income divide between the sectors. Figures from the October Household Survey show the yearly income of those 10,3-million people with jobs. More than half of informal and domestic employees earn less than R500 a month and many more earn less than R1 000 a month. The situation is reversed in the formal sector. Worse, a large number of formal employees in the two lowest-earning categories are undoubtedly farm workers (which is really a special case).
Those in formal employment and earning more than R2 500 a month make up 27% of the whole workforce. This category covers a broad ambit and should be subdivided further. Not surprisingly it includes top executives. Less obviously, even a new police constable is in the top 27% of income earners.
While almost half (49%) of those in the informal sector were in unskilled jobs, only 18% in the formal sector were. This is an unstable situation and is part of the reason why we see so much subcontracting of unskilled tasks. Unskilled workers dropping from the formal to the informal sector face a sharp fall indeed. Conversely, informal entrepreneurs rarely make the transition into the formal sector. And first-time employees and the unskilled are basically confined to unemployment or informal jobs.
The point is that the labour market is deeply fragmented and the Chinese wall between sectors is unsustainable and unwise. This is bad news for the union movement since, although it might favour demands for some minimum standard enforcement, it mostly adds weight to calls for greater flexibility in the formal sector.
Jeremy Baskin is a public policy analyst