IMF Blog IMF Blog

South Africa’s Unemployment Puzzle

Among the havoc wrought by the global financial crisis, unemployment ranks at the top. This discussion often focuses on the situation in advanced countries. Unemployment in the United States, for example, continues to hover around 9 percent.

Take that and double it. Then you can begin—yes, just begin—to get a sense of the magnitude of the problem in South Africa. Unemployment in South Africa now stands at some 24 percent. Youth unemployment is phenomenally higher still at some 50 percent.

Unemployment in South Africa was already very high before the crisis due to a number of structural factors—such as mismatches between the kinds of jobs available and workers’ skills, or large distances between population centers and where businesses are located.

But, the enormous job losses during 2008-09 made the already dire situation much worse. South Africa did not have a financial crisis and its recession was, relatively speaking, mild. Still, the country lost proportionately as many formal sector jobs (1 million) as those countries at the center of the global financial crisis.

Like the government, we see reducing unemployment as the foremost economic challenge facing the country. Trying to understand why recent labor markets outcomes have been so bad and, more important still, what it will take to reverse the increase in unemployment is a thus an ongoing focus of our work.  

Pieces to the puzzle

One interesting finding is that, despite high unemployment, South Africa’s labor markets are relatively dynamic. During the growth upswing of the mid-2000s, the country was very good at creating jobs; but when the economy hits a rough patch, job losses also tend to be very high. For each 1 percentage point increase in growth, employment growth tends to increase by more than 1 percentage point. Unfortunately, the same is true when growth declines.

This goes some way towards explaining the relatively large magnitude of job shedding during the recent recession. But it also implies that wages do not respond much to changes in the demand for labor. When there is an adverse shock to demand, it is the level of employment rather than wages that adjust. Of course, in most countries, wages are generally ‘sticky’ downwards and do not decline in nominal terms. But seldom does one also see large economy-wide increases in real wages—wages rising faster than inflation—during a recession. Yet, that is exactly what happened in South Africa in 2009.

Last December I had the opportunity to meet with some Members of South Africa’s Parliament. We exchanged views on the impact of the global financial crisis on the country. When I laid out the ideas above, one of the Members asked if I thought that the country’s employment protection laws were the cause. As I noted then, this is not my view. Indeed, I firmly believe that the country’s labor legislation provides important and necessary—and hard won—protection for workers. Rather, what would be worthwhile is a close look at the wage bargaining framework to ensure that most of the adjustment in the labor market does not continue to fall mainly on the number of jobs. Wage moderation during downturns would not seem that unreasonable a trade-off.

Likely solutions

Beyond a more flexible wage bargaining framework, here’s my take on what is required to make significant inroads into high unemployment in South Africa.

Higher growth. The government’s recent budget document noted that annual growth of the order of 6‑7 percent will be needed to meet the target of creating 5 million jobs by 2020. The current fairly supportive monetary and fiscal policies will help in the near-term. But it is unclear where the impetus for the higher growth will come from once macroeconomic policies turn less supportive, with a view to rebuilding the policy buffers run down over the last couple of years. Ideally, growth should be private investment and export led. Looking for ways to promote private investment thus needs to be firmly on the reform agenda in the coming months.

Making growth more labor intensive. Another approach that should help is targeted interventions to address unemployment in particularly problematic areas—such as youth unemployment. This is, in part, because current wage setting mechanisms do not allow differences in wages that would fully reflect productivity differences between young and old workers. A wage subsidy scheme along the lines recently announced by the government should serve to make it cheaper for firms to employ young workers. Provided the subsidy is carefully designed, it should avoid the displacement of existing workers and also minimize substitution away from older workers.

More competitive product markets. One of the other problematic features of the South Africa economy is the relatively high cost structure of many of its markets for goods and services. By contributing to higher input costs, this inhibits the external competitiveness of manufacturing and other tradable sectors. Enhancing domestic competition should lower costs for companies but importantly also for consumers.

The big blemish on South Africa’s otherwise strong economic performance since the mid-1990s is stubbornly high unemployment. Of course this is an important exception, especially as it has exacerbated income inequality. Doubling growth—from the current 3½ percent—is the first order of business. And this in turn requires changing the incentives facing firms and employees.

Recent