South Africa’s failure to create manufacturing jobs
To reduce unemployment South Africa needs to prioritise labour-intensive work instead of higher wages, write professors Jeremy Seekings and Nicoli Nattrass
W. Arthur Lewis is the only person of African descent to have won the Nobel prize for economics. Lewis, who might be considered the founding father of development economics in the 1940s and 1950s, combined writing and teaching with providing economic advice to policy-makers, including Ghana’s first president, Kwame Nkrumah.
Lewis thought extensively about the path along which farming societies in the Caribbean, Africa and elsewhere could develop. He focused on the structural changes needed in societies with what he called “unlimited supplies of labour” or “surplus labour”.
In these societies, he argued, labour can move out of the peasant farming sector — without affecting production there — and into low-productivity capitalist sectors, including industrial sectors such as clothing manufacturing, if wages in those sectors are low in line with productivity.
As capitalist economies grow, Lewis argued, they will reach a turning-point at which labour becomes scarce. They will then shift into higher-productivity sectors and wages will rise.
Lewis’s model was reflected in the economic history of East and South-east Asian countries. The economies of Korea, Hong Kong and China developed through job creation in labour-intensive manufacturing sectors before these economies reached the Lewisian turning-point and wages began to rise.
Lewis’ ideas are very relevant today across much of Southern Africa. In South Africa and most of its neighbours, “surplus labour” exists in the form of massive unemployment and under-employment (employed at less than full-time or paid too little for economic needs) rather than in subsistence agriculture.
Faced with this kind of economy, Lewis would recommend that the government expand the clothing manufacturing sector and other labour-intensive capitalist sectors.
But this is not how the South African economy has evolved over the past twenty-five years. Instead, South Africa’s industrial sectors have become more and more capital- and skill-intensive. The demand for less skilled labour in these sectors has collapsed.
Economists refer to the effect of economic growth on employment as the employment elasticity of growth.
An employment elasticity of 1 means that, when the economy or sector grows by, say, 2 percent, employment grows at the same rate, i.e. by 2 percent. If the employment elasticity is more than 1, then economic growth results in a faster rate of job creation. If the employment elasticity is less than 1, then employment rises more slowly than output.
Data suggests that in the South African manufacturing industry employment elasticity has not only been less than 1, it has been negative.
In an economy with surplus labour, the goal of job creation requires a high employment elasticity of growth. Only when the economy reaches the Lewisian turning-point should raising wages become more of a priority than creating jobs.
In developing economies, economic growth can generate lots of short-term jobs in sectors such as construction. Sustainable job creation on a large scale generally requires, however, job creation in labour-intensive manufacturing sectors such as clothing manufacturing. To create jobs on a large scale requires taking advantage of global markets by producing labour-intensive products for export.
This was what the East Asian tigers – including China – did. The south-east Asian countries are now doing the same thing: In Vietnam, Cambodia and Thailand, an additional one million jobs in clothing production were created between the mid-1990s and mid-2000s.
Burma is now following suit. In Bangladesh, also, as many as four million people work in clothing manufacturing. In these economies, the expansion of labour-intensive sectors has been crucial to poverty-reduction.
Data from the South African Reserve Bank, based on surveys of firms, suggest that manufacturing output expanded at 1.9 percent per year, taking inflation into account, between 1995 and 2017. Over the same period employment in manufacturing contracted by 1.3 percent per year. These data suggest that the employment elasticity of growth in manufacturing was -0.7.
These firm-level data probably exaggerate the contraction of employment. Data from labour force surveys suggest that employment in manufacturing barely fell at all, by only -0.02 percent per year, between 2000 and 2017. This gives an employment elasticity of growth in manufacturing of -0.01.
Even if we accept the labour force survey data, the conclusion is that rising manufacturing output has not resulted in any overall job creation in manufacturing.
In sectors such as clothing output has risen while employment has fallen dramatically, to barely more than 50,000. Bangladesh by comparison has about 4 million jobs in clothing production. Tiny Lesotho has almost as many clothing jobs as South Africa. Madagascar has many more.
Declining employment in labour-intensive sectors like clothing helps to explain why, in the South African economy as a whole, job creation has been so slow. Between 2000 and 2017, the South African economy grew by 2.8 percent per year. Total employment – including even informal and part-time jobs – grew by a paltry 1.6 percent per year. Growth was not entirely jobless, but the overall rate of job creation was not enough to absorb new entrants into the labour market, let alone reduce unemployment.
The employment elasticity of growth has been so low in South Africa because policy-makers have ignored Lewis’s insights. Faced with massive surplus labour, policy-makers should have been promoting labour-intensive sectors, such as clothing manufacturing.
Instead, policy-makers have been pursuing the goal of high-wage, high-productivity, “decent” work, without much or any regard for the costs in terms of both actual job destruction and obstacles to large-scale job creation.
South African government policies result in a paradoxical situation: Even when industrial output grows, jobs are not created. Instead, firms invest in more capital- and skill-intensive technologies. Accelerating economic growth does not raise the demand for labour in industrial sectors.
This is not what South Africans want. Every poll of public opinion since 1994 points to job creation being the top priority of South African citizens. South African citizens would like to see high-wage jobs being created but most believe that any job is better than nothing.
The most recent Afrobarometer survey in South Africa (in 2018) asked a representative countrywide sample to choose between two statements: (1) “It is better to have no job than to have a job with a low wage” and (2) “Any job is better than not having a job”.
Fewer than one in five people (17 percent) agreed or agreed strongly with the first statement. More (19 percent) agreed with the second statement. And a massive 60 percent agreed strongly with the second statement.
Policy-makers need to listen to South African citizens and focus on creating jobs in labour-intensive sectors.
The authors are professors at the University of Cape Town. Their new book, Inclusive Dualism: Labour-intensive development, decent work, and surplus labour in Southern Africa, will be published by Oxford University Press in 2019.
Views expressed are not necessarily those of GroundUp.
This is an interesting article on an important topic, but it raises the obvious question: why? It is not enough to observe, as the authors do that "South African government policies result in a paradoxical situation: Even when industrial output grows, jobs are not created. Instead, firms invest in more capital- and skill-intensive technologies. Accelerating economic growth does not raise the demand for labour in industrial sectors." What has led to this paradox? And what can be done to overcome it?
© 2019 GroundUp.
This article is licensed under a Creative Commons Attribution-NoDerivatives 4.0 International License.
You may republish this article, so long as you credit the authors and GroundUp, and do not change the text. Please include a link back to the original article.