Reducing unemployment: Waiting for high growth? Waiting for Godot?

Reads 28,246
Frederick Fourie, 12 March 2013
In trying to reduce unemployment in South Africa, the pursuit of higher economic growth is the single most agreed-upon policy strategy. The consensus on this ‘obvious solution’ may blind us to the fact that economic growth, though important, may only be half of the solution. Attempts to fine-tune and turbo-boost the formal-economy ‘engine of growth’ to absorb more labour are fundamentally constrained. Economic policy makers must look at other options for generating employment and self-employment for unemployed people.


The expression 'Waiting for Godot’ comes from a play of the same name by Samuel Beckett
in which two characters wait for a third named Godot, who never appears. It means to wait for
something that will never happen, with connotations of futility and despair.    [Wiki Answers]

The pursuit of higher economic growth (as measured by a growing real GDP) probably is the single most agreed-upon policy objective and strategy. Across all political and ideological divisions, in government, labour and business, it is seen as the prime solution to the unemployment problem. This essence of this solution is: grow the core economy, grow the formal economy – manufacturing, mining, etc.

The consensus on this ‘obvious solution’ is so strong and alluring that it may blind us to the fact that such growth, though important, may only be half of the solution.

Obviously the pursuit of a high growth rate is important, but…

For the last 20 years South Africans have seen policy documents stressing the need for a high GDP growth rate. These have included the RDP (1994), GEAR (1996), ASGISA (2006), the New Growth Path (NGP of 2010) and recently the National Development Plan (NDP). Various government documents such as the National Treasury’s annual Budget Reviews and govern­ment Medium Term Strategic Frameworks (MTSFs) have done the same.

The GDP growth rates that need to be attained and sustained to reduce unemployment to acceptable levels are relatively high. Those specified have ranged from 4.5% (MTSF 2004-2007) to ‘at least 6%’ (ASGISA for 2010-14), between 4% and 7% (NGP), 7% (the Minister of Finance and National Treasury), and most recently, with much precision: 5.4% (the NDP).

Our GDP growth performance has not been anywhere near those levels for many years. The latest Budget Review projects growth of 2.7% in 2013, 3.5% in 2014 and 3.8% in 2015, following several years in which growth was around 3% (National Treasury 2013: 6). There is a persistent, significant gap between the required or desired growth rate and the actual growth rate. And unemployment remains stuck at around 25% (narrow definition) and 35% (broad definition).

South Africa does not seem to have an alternative strategy to ‘going for growth’ and adopting a conventional menu of broadly agreed-upon strategies: removing the bottlenecks to output growth and employment in the formal sector (including the reform of education), promoting exports and investment (including investment in infrastructure), promoting small business and so forth. (The relaxation of labour market policies is the one strongly contested policy proposal, one favoured by the business sector but not by labour and government.)

We continue to hope and wait for the required higher growth. Is this a realistic position – or are we collectively ‘waiting for Godot’ (whilst continuing to churn out ambitious policy plans, conferences and workshops)?

Do we know how to increase growth? Not really

One problem is that economists and policy makers do not really know how to increase economic growth. Despite decades of study and research, increasing growth remains elusive because we still have an incomplete understanding of the human behaviour, determinants and causal relationships behind economic growth processes. The theory is there, all the mechanisms and ‘drivers of growth’ appear to make perfect sense. Yet all our efforts generate no major results.

This is very frustrating because ordinary people have been led to expect that growth can be increased by pulling a few policy levers or by ‘releasing entrepreneurial spirits’. Policy makers have to do their best, despite the incomplete understanding of growth processes (especially in a developing country such as South Africa). Complex global economic linkages obviously complicate this situation.

How ‘employment intensive’ is GDP growth in South Africa? Not very

The reason why an extraordinary high growth rate is required (and has to be sustained for many years) to make a meaningful impact on unemployment and poverty, is the relatively low rate of labour absorption rate in a modern formal economy – i.e. that economic growth does not produce many employment opportunities, that it is not ‘employment intensive’. These also apply to the South African economy.

One way of looking at this is to consider the ‘employment coefficient’ over time. This coefficient is defined as the ratio between the rate of growth in employment and the rate of growth of GDP. If employment grew at the same pace (e.g. 3%) as GDP, the coefficient would be 1; if employment grew more slowly (at a lower rate) than GDP, the value would be less than 1.

Hodge (2009) shows that for the period 1946-2007 the employment coefficient for the SA formal sector fluctuated around a value of approximately 0.5 (confirmed by Nattrass (2011) for 2001-2008). This means that, on average, economic growth leads to  employment growth in the formal sector of only half the real GDP growth rate. GDP growth, which measures mostly growth in the formal sector, tends to produce only limited growth in the total absorption of labour.

Some time ago this was debated heatedly when the question was asked whether the SA economy was characterised by jobless growth. A coefficient value below zero would indicate jobless growth. This was the situation in the 1990s when aggregate labour absorption declined in several sectors (notably mining, agriculture and manufacturing). But it lasted only for a short ‘abnormal’ period, whereafter the rate of growth of aggregate labour absorption returned to normal levels.

This indicates that jobless growth has not been a long-term characteristic of the South African economy. But low jobs growth has been, as evidenced by the long-term level of the employment coefficient being approximately 0.5.

This pessimistic conclusion captures an important implication of the low employment coefficient: growth within the formal sector alone is unlikely to absorb sufficient numbers of people to reduce unemploy­ment rates significantly (unless real growth rates increase drastically, of course). The formal sector cannot absorb all the people who entered the labour force since the 1990s, many of them (notably women) from being economically inactive, others from the subsistence segment and the informal economy; large flows of job seekers from other African countries contributed to the growing labour force. This situation is bound to continue.

The value of the employment coefficient is critical

The importance of good current estimates of the employment coefficient (alternatively the ‘output elasticity of employment’) is that too high a value can lead to over-optimism regarding the prospects for reducing unemployment through GDP growth. For example, the National Development Plan uses a value of 0.56 for the overall employment coefficient. Yet, if it actually were at the somewhat lower value of 0.5, the employment rate would drop to only 12.7% by 2030 rather than the 6% projected by the NDP. Or, GDP would have to grow by 6.2% (rather than 5.4%) for 20 years to attain the unemployment goal of 6% by 2030.[1]

Unfortunately, declining employment intensity is normal for South Africa

Whatever the precise value of the employment coefficient, the findings from researchers show that, no matter how high the GDP growth rate, the growth rate of employment within the formal sector persistently tends to be significantly lower.

Any employment coefficient below one implies that the absorption of labour will continually decline relative to output.  In this sense the employment performance of the South African economy has been declining steadily for many decades. A continually declining employment intensity (or declining ‘labour intensity’) actually is normal.

The magnitude of the steady decline in labour intensity (aggregate employment relative to GDP) in the long run underlies much of the observed increase in unemployment. Pollin et al (2006:10) highlight the decline in labour intensity in the period 1967-2001: from 8.2 formal economy workers per R1m real output in 1967 there was a sustained drop to 4.9 in 2001 – a total drop of 40% in 44 years. In 1994-2001 (the bad patch years) the drop was 28%.

To a large extent, this long-term decline reflects trends in technology and cost management techniques. However, the possibility of the trend being aggravated (in the recent past or in the future) by the negative employment impacts of real wage increases cannot be dismissed – consider the reported employer responses to the recent increase in the agricultural minimum wage.

Whatever the causes may be, the sustained downward trend in labour absorption in the core economy is a fundamental reality which all but overshadows all strategies to reduce unemployment significantly via accelerated GDP growth combined with increasing employment intensity.

Can employment intensity be increased? Are we looking in the right places?

It will take persistent hard work and good fortune to slow the downward trend in employment intensity, let alone reverse it. While it is exemplary for government to see the private sector as a partner in improving growth and employment, one should not forget that private enterprises do not have job creation as an objective. Since labour is a cost to business, employment growth in the private sector is, at most, a side-effect of growing sales and output.

Increasing the employment intensity of GDP growth has been a theme of government policies for a long time. The New Growth Path (NGP) is a recent initiative to propose strategies to increase the employment intensity in infrastructure, manufacturing, mining, agriculture and services. The National Development Plan aligns itself with the NGP and proposes measures to accelerate growth and make it more labour absorbing – inter alia by focusing on export-linked domestic services which have significant employment effects (NPC 2012:39).

This being so, Black (2012) points out that the cards of industrial policy have been stacked against increasing labour absorption for many decades – and that it still is, many years post-1994. Large incentives still go towards increasing capital expenditure. While such industrial policies are in place (compare IPAP2, the industrial policy closely aligned to the NGP), we should not be surprised that employment intensity continues to decline, at least in industry.

Current NGP-derived government plans to invest massively in infrastructure are also not likely to increase employment intensity directly – the construction of infrastructure tends to be very capital-intensive.

If the (slowly) expanding core formal economy is unlikely, or slow, to absorb large numbers of unemployed people, what is to be done?

Other options: Develop a vibrant informal micro-enterprise sector

One option, not sufficiently explored at all, is for the unemployed to find meaningful and gainful employment and self-employment in the informal economy – the world of informal micro-enterprises (which is to be distinguished from survivalist activities).

It is true that the informal economy in SA is much smaller than in other developing countries. This could mean that it has reached a plateau and cannot make much of a contribution to employment growth.

However, it could also mean that the informal economy has significant untapped potential – if only it is developed and boosted in appropriate ways. This could mean that the informal economy might become more ‘self-absorbing’ as far as unemployed people are concerned, rather than being a place where people wait to be absorbed into the formal economy – where they are waiting for Godot, once again.

It cannot be said that much, if any, policy effort has gone into making the informal micro-enterprise economy a vibrant employment-creating and employment-intensive component of the South African economy – compared to the effort that has gone into trying (often in vain and at much expense) to boost the growth of the formal economy. How much economic policy (as opposed to social and welfare policies) has focused on turning economic marginalisation around? How much effort, in terms of economic policy, has gone into determining and tackling the constraints facing informal enterprise, self-employment and employment (in contrast to those that face the formal SMME sector)?

The NGP and NDP are not exemplary in this regard – yet there is actually so much to consider and analyse from an economic policy point of view. Philip (2012) demonstrates that informal or micro-enterprises face many structural obstacles that emanate from large mainstream competitors, while Charman (2012) highlights how government actions (especially at local level) tend to actively discourage and even undermine informal businesses. Whatever employment growth has occurred in the informal economy, has occurred without economic policy support.

The key point is that this line of inquiry would approach the informal, micro-enterprise economy not as a problem segment, but rather as a promising segment in which self-employment and employment can be grown. It would also see the informal economy as a training ground for business skills development, which will enable those who gain experience as workers and/or as entrepreneurs to enter the formal economy successfully at a later stage. (And surely one’s conception of informal enterprise should not be limited, patronisingly, to spaza shops and street traders!)

Growing productive activity and increasing skills and productivity in the informal economy as such could transform this segment into becoming a source of economic growth. Policy interventions with a relatively low cost may have a large impact on employment, also producing economic growth.

Conclusion: Adopt a two-pronged, formal-plus-informal policy approach

If the formal economy continues to grow at around 3% per annum and create employment at historical rates, employment is likely to grow at approximately 1.5% per annum. This is only half of what is required to reduce unemployment significantly, which means that unemployment will remain at around 25% (narrow definition).

Pursuing higher employment through the growth of the formal sector will always be important (as will the formal economy’s providing most of the tax revenue which finances pro-poor social expenditure.) However, a strategy that continually attempts to prime, fine-tune and turbo-boost this ‘engine of job growth’ in order to absorb more labour is fundamentally constrained. At some point it becomes less and less productive, even futile.

Furthermore, as long as economic policy makers rely exclusively on a growing formal sector to ‘suck in’ unemployed people, they ignore an important avenue to improve the employment prospects, earnings, livelihoods and human dignity of millions of poor people.

A carefully-designed and balanced two-pronged economic strategy that targets both the informal and formal economies (and, notably, linkages between them) is likely to have much more success at reducing unemployment. Economic policy makers should not allow formal sector issues (and interests) to claim the entire economic policy thinking space and to absorb all policy resources.


Altman, M. (2008). Revisiting South African employment trends in the 1990s. South African Journal of Economics, 76(S2): S126-47.

Black, A. (2012).  Industrial policy and unemployment: Can South Africa do better in labour-demanding manufacturing? Econ3x3, November.

Charman, A. (2012). Is informality being disallowed by government? Econ3x3, November.

Economic Development Department (2010). New Growth Path. Pretoria.

Fourie F.C.v.N. (2011). The South African unemployment debate: three worlds, three discourses? Working Paper 63, SALDRU, University of Cape Town.

Hodge, D. (2009). Growth, employment and unemployment in South Africa. South African Journal of Economics. 77(4):488-504.

National Treasury. (2013). Budget Review. Pretoria.

Nattrass, N. (2011). The new growth path: game changing vision or cop-out? South African Journal of Science 107(3/4), commentary section.

NPC (2012). National Development Plan 2030. National Planning Commission, Pretoria.

Philip, K. (2012): How structural inequality limits employment and self-employment in poor areas (or: Why South Africa’s informal sector is so small), Econ3x3, November.

Pollin, R. et al (2006). An employment-targeted economic program for South Africa. United Nations Development Programme, New York.

[1] The estimation of the employment coefficient is complex, inter alia because different data series on employment provide different estimates; measurement at sectoral level also provides a variety of estimates (see Fourie 2011:50-52 for an overview). Altman (2008: S143), presenting revisions to employment data between 1995 and 2006 and revised formal-plus-informal employment data, estimated the simple output elasticity of employment in the interval 0.45-0.66, depending on the period and data source. In its employment scenarios the National Development Plan (NPC 2012:121) appears to use relatively optimistic values of around 0.6 for the formal sector and 0.5 for the informal sector. (Especially the latter appears to be quite high; see Altman (2008: S143) for informal sector estimates that are around 0.25.)


Download article

PDF icon Fourie March 2013 Unemployment - Waiting for growth or Godot FINAL.pdf
Post a commentary

This comment facility is intended for considered commentaries to stimulate substantive debate. Comments may be screened by an editor before they appear online. To comment one must be registered and logged in.

This comment facility is intended for considered commentaries to stimulate substantive debate. Comments may be screened by an editor before they appear online. Please view "Submitting a commentary" for more information.