A national minimum wage: moving the debate forward?

Reads 12,464
The public debate on a national minimum wage sometimes appears to occur in different universes. Two recent contributions to Econ3x3 may help to take the debate forward. This article analyses and contrasts these views and finds that, though they emphasise (and underplay) different aspects, the differences may not be insurmountable – especially once one recognises that the proposals apply to different time frames. [A shorter version of this article appeared as an op-ed article in Business Day on 25 June 2015. See references.]


The debate on a national minimum wage appears to occur in different universes. The main institutional protagonists – organised labour and the business sector – seem to talk past each other. In op-ed and other media articles, spokespersons, lobbyists and even academic analysts frequently resort to flag waving or labelling. The issue drowns in the adversarial relationship between labour- and business-oriented approaches. Amidst this furore, government must formulate policy on the matter.

Contrasting two recent contributions to Econ 3x3 may help to take the debate forward.

Seekings & Nattrass: the risk of negative employment effects

The first contribution was that of Jeremy Seekings and Nicoli Natrass (2015) of UCT. The real debate, they argue, is not between a free-market position (which favours no regulation) and Cosatu’s stance (which argues for a single, relatively high national wage floor). Rather, it is between Cosatu’s position and one that takes seriously the negative employment effects that too high minimum wages may have in particular sectors or areas if they are set too high for the conditions in those labour markets.

When framed in this way, there are two key points of dispute: the level of the national minimum wage and whether there should be a single minimum wage or a differentiated system.

Seekings & Nattrass are in favour of a national minimum wage. However, they caution against a uniform minimum wage that is set too high because the higher wage costs would result in job losses in vulnerable sectors: there would be higher wages, but for a shrinking group of workers, while unemployment would grow. This is likely to increase rather than reduce poverty and inequality. The vulnerable sectors are, notably, those which are exposed to competition from imports (tradable sectors, such as clothing and agriculture) and/or where mechanisation is an attractive option (as in agriculture, for example). In less vulnerable sectors higher minimum wages are unlikely to result in major job losses (domestic and security work are examples).

Thus they propose a differentiated system: a national minimum wage floor at a relatively low level to prevent job losses in the vulnerable sectors – combined with higher minimum wages in other sectors where employment effects will be modest. This is preferred to ‘setting a national minimum at a level that has a high probability of destroying jobs in sectors such as agriculture, destroying entire sectors such as clothing and reducing the possibility that labour-intensive sectors could grow in future’.

Isaacs & Fine: minimum wages as part of a package to boost investment and output

The second contribution to Econ3x3, in broad support of Cosatu’s position, comes from Gilad Isaacs of Wits University and Ben Fine of the School of Oriental and African Studies in London (Isaacs & Fine 2015). They argue that wages is not the key factor that determines employment levels. These are determined ‘by the level of output, which is predominantly determined by the level of investment.’ Also, wages is not the main factor affecting international competitiveness. Efficiency gains through, for instance, management techniques, good labour relations, technology, infrastructure and supportive government policies are more important.

Given the goal of fighting poverty and inequality by improving the earnings of workers, a relatively high minimum wage should be instituted. Yet this should not be done in isolation, but rather as an integral part of a policy package of structural reforms that are intended to boost investment, output growth and thus the demand for labour. While there is a possibility of negative employment effects from a new minimum wage, the positive effects of an expanding, restructured economy – aided by the extra spending power of higher-paid workers – would override any negative employment effects of a relatively high minimum wage. This could also be true for tradable sectors, including agriculture (whose competitiveness, they argue, has suffered not due to minimum wages but due to sustained unsupportive government policies; the latter also applies to the clothing sector, where inefficient production techniques have contributed to its lack of competitiveness). ‘Such restructuring of the economy would require the careful planning and coordination of the introduction, roll-out and phasing-in of the various elements of the policy package. These are neither short-term nor quick-fix remedies.’

In their view, Seekings & Nattrass’s position is focused too narrowly on labour market reactions, while the proposal of a relatively low national minimum wage (albeit coupled with higher minima in some sectors) carries the risk of entrenching a status quo of predominantly low wages. It is a short-term approach that leaves the structure of the economy untouched.

Coleman: the Brazilian experience; the issue of phasing in

Neil Coleman, the Cosatu economist, has presentedcomprehensive arguments for a national minimum wage in a paper published by the ILO (Coleman 2013). He highlights the experience in Brazil, where minimum wages, coupled with social grants, have substantially reduced inequality and poverty. He notes that the implementation of minimum wages in Brazil did not lead to negative employment effects; unemployment actually decreased in the relevant period. (The period was characterised, though, by an economic strategy to boost domestic demand and employment, linked to government-led industrial strategies that created more employment for unskilled workers.)

Coleman argues that, to address the needs of workers, a national minimum wage in South Africa must be substantially above current sectoral minimum wages. While quoting the International Labour Organisation’s preference for a uniform national minimum wage, he indicates that a two-tiered model may be preferable – but primarily as a phasing-in mechanism: start with a ‘two tier arrangement, which initially covers domestic and farm workers separately, and phase in one national minimum wage which would include domestic and farm workers, over an agreed period.’ Sectoral minimums in specific industries can then be negotiated above the national minimum wage. As in Isaacs and Fine’s view, the introduction of the national minimum wage must be combined with appropriate macroeconomic and sectoral policies to have the desired employment impact.

Coleman debunks arguments – opportunistically emphasised by business, in his view – that wage increases automatically lead to job losses. Still, he recognises such effects by allowing for the differential treatment of domestic workers and agricultural employees in a phasing-in period, as well as emphasising the importance of a package of supportive policies to get the desired employment impact. He also lists ‘consideration of the employment effects of a national minimum wage’ as a key challenge in a minimum wage strategy. Similarly, Isaacs & Fine concede that the demand for labour may decline in response to wage hikes – but argue that it is often overstated and that it can be overridden by appropriate expansionary and sectoral policies (as part of a broader restructuring that focuses on boosting investment and productivity).

Is there common ground? The importance of the time frame

What is one to make of these opposing views? Is there enough common ground to take the debate forward – given the apparent differences in ideology, sentiments, analytical arguments and the interpretation of empirical evidence?

All these authors seem to recognise the negative employment effects of higher labour costs. However, they give differing weight to their practical importance – especially relative to policies that could increase employment by boosting demand and investment (as well as a positive impact due to higher spending by higher-paid workers). In other words, they differ on whether labour cost or investment and output demand are the most important aspects in protecting, or boosting, employment.

While Seekings & Nattrass are criticised for not mentioning broader structural changes and longer-term employment growth, their warning about likely job losses in certain sectors cannot be ignored. Isaacs &Fine (and Coleman) may be open to criticism for their reluctance to acknowledge the risk of negative employment responses (and also the crucial role of demand stimulation in Brazil’s good minimum-wage experience). However, their positive and ambitious approach to achieve a high-investment growth path with higher wages and employment growth cannot be easily dismissed.

It has also become evident that the protagonists differ mainly with respect to the time frames they have in mind. The restructuring of the economy, which Isaacs & Fine propose, is a long-term project. If that is so, it is not inappropriate for Seekings and Nattrass to propose a system that manages the short-term risk of labour-demand responses to a national minimum wage. If a ‘high’ minimum wage were to be introduced on its own or before appropriate investment- and sector-boosting policies have taken effect, or in a stagnant economy, there could be significant job losses.

Thus Isaacs & Fine could be right in the medium to long run (if the restructuring can be done successfully and sustainably), while Seekings & Nattrass could be right in the short to medium run. Something like Coleman’s differentiated phasing in could also manage some short-run risks, but would still leave the medium to long run as a major balancing act.


The differences in the debate do not seem insurmountable. The question is whether government, organised labour and business can be open-minded enough to find the consensus necessary to improve the lot of workers while at the same time boosting employment – thus decreasing inequality and poverty.

For business, it would require a rethink of, for example, a singular focus on labour costs versus the challenge, and appeal, of building an economy with higher wages, lower poverty and lower inequality. Can business recognise that minimum wages are about more than merely labour costs and that lower inequality and poverty can be beneficial to economic growth and stability? Likewise, can labour balance the needs of workers against the reality of employment decisions and the at-risk sustainability of many farms and firms (in certain sectors and of particular sizes)? Finally, organised labour may need to rethink its stance on the purity of a single national minimum wage versus the complexity of necessary differentiation (at least in the short to medium run) – and on the risk of instituting a (‘high’) national minimum wage irrespective of economic conditions in individual sectors (or in the country as a whole).


Fourie FCvN & Green P. 2015. Flexibility needed to find consensus for workers’ sake. Business Day, 25 June 2015, p. 11. http://www.bdlive.co.za/opinion/2015/06/25/flexibility-needed-to-find-consensus-for-workers-sake

Coleman N. 2013. Towards new collective bargaining, wage and social protection strategies in South Africa – Learning from the Brazilian experience. Working Paper 17. ILO and Global Labour University (GLU).

Isaacs G & Fine B. 2015. The national minimum wage debate: looking beyond a narrow focus on labour markets. Econ3x3, March. http://www.econ3x3.org/article/national-minimum-wage-debate-looking-beyond-narrow-focus-labour-markets

Seekings J & Nattrass N. 2015. What is at issue in the minimum wage debate? Econ3x3, January. http://www.econ3x3.org/article/what-issue-minimum-wage-debate

Tags: Minimum wage

Download article

PDF icon Fourie & Green 2015 Minimum wages - moving the debate forward 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.

In a debate on an issue such as minimum wages, I think there is always a risk of simplistically linking issues such as labour, minimum wages and social grants as plausibly correlated.

In the first place, social grants should (in my view) be a safety net for individuals that are, or were, structurally excluded from gainful employment either in the formal or informal sectors - and not, as viewed by some, as an alternate income source, or worse still as the price tag of sustaining democracy.

In the second place, the implication of the word 'labour' tends to relegate people to the level of a resource, like energy, raw materials or land, where price is the only consideration - rather than as a productive asset that can, and should, contribute to the overall success of an enterprise.

Thirdly, the fact that treating labour as a resource allows, even encourages, participants to find the 'point of surrender' where both labour and capital leave the table less than satisfied, probably with escalated adversarial positions and already strategising about leverage for the next round of dividing up a shrinking cake. This situation may also contribute, or give credence to, the gulf in income disparity between labour and management, based on perceptions of competency and a need for increased control.

Finally, the very fact that the labour/capital discourse is about a minimum wage tends to relegate the matter to one of solidarity with either protagonist and the division of the spoils rather than the potential that exists for growth and opportunity.

Is South Africa really still plagued by such entrenched meta-level vacillation between communist and capitalist ideology?

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.