The national minimum wage debate: looking beyond a narrow focus on labour markets

Reads 29,951
Gilad Isaacs, Ben Fine, 17 March 2015
Most contributions to the debate on a national minimum wage adopt a narrow view of labour markets and accept that the structure of the economy will remain essentially as it is. We question both of these assumptions. Further, we argue that a national minimum wage, at a level to be determined through careful research, must be part of a well-designed package of longer-term policy reforms that look beyond the labour market and support employment growth through investment.


In their January Econ3x3 article on a national minimum wage, Seekings & Nattrass (2015) offer some useful and well-summarised perspectives regarding the relevant legislative frameworks – but unfortunately fail to grapple with the deeper issues at play or take seriously the arguments we have previously advanced.

In this article we interrogate Seekings & Nattrass’s position and argue that a national minimum wage, at a level to be determined through careful research, should be part of a well-designed package of policy reforms that go beyond labour market effects and support employment growth through investment.

We submit that Seekings & Nattrass’s approach involves a narrow consideration of labour markets (although they do differentiate between traded and non-traded sectors), revealing a fixation on low-wage growth models. Their proposal would do little to structurally reform current labour market dynamics and carries a high risk of largely entrenching the status quo. It also appears that they focus on short-term effects while ours is a longer-term perspective which situates the introduction of a national minimum wage within a restructuring of the economy.

Mechanical link? The relationship between wages and (un)employment

Seekings & Nattrass's article clarifies that they agree that there is no mechanical link between wage levels and employment levels according to which increased wages automatically lead to rising unemployment; it depends, they argue, on the level at which such a wage is set.

The implicit acceptance of such a rather simplistic causal relationship is a pervasive feature of many (or most) arguments against prescribed minima and wage increases in general. The proposition of such a mechanical link is theoretically unsound and not supported by the evidence. High wage levels do not necessarily have negative effects on employment. Arguments that involve a mechanical link are premised on limited, if not faulty, neoclassical economics that construe labour markets as being governed by the iron laws of supply and demand which, if left undistorted, would clear at a full-employment market equilibrium. No involuntary unemployment other than that caused by excessively high wages can exist, in this paradigm. This is at odds with reality: the wage level is neither the only nor the key factor that regulates employment levels.[1] The level of employment in an economy is determined primarily by the level of output which is predominately (but not exclusively) dependent on the level of investment. Wages do a play a role but it is often overstated and in any case does not operate independently from levels of output and investment.

If they could be both realised and sustained in the absence of other changes, inordinately high wages could have a negative effect on employment. To this extent, as Seekings & Nattrass suggest, the level at which a minimum wage is set is critical. However, we object to the focus of the debate being this aspect alone, which is what Seekings & Nattrass and other proponents of wage restraint do – and which amounts to a short-term, narrow approach that makes wage levels the predominant factor that determines levels of employment.

Considering the employment impact of wage increases in isolation has the dual implication of advancing labour markets as the sole (or the most important) locus of reform and allowing other aspects that comprise the basic structure of the economy to escape scrutiny. The impact of wages on employment surely should not be ignored but its analysis must be thoroughly embedded in a framework that is broader than labour markets alone. This has three dimensions (also see Isaacs & Fine 2014a; 2014b).

(a) Do wages determine global competitiveness?

First, to suggest that the level of wages is the decisive factor that determines global competitiveness and drives levels of investment and job creation flies in the face of evidence. Success relies on fostering increases in productivity through efficiency gains that reduce input costs and through innovations that increase the volume, quality or value of goods and services. Wages are only one input cost and reducing them is the least effective way in which to make labour more efficient.

Experts in this field (see Porter 2008; Delgado et al. 2012; Manyika et al. 2015) argue that the productivity increases needed to achieve foundational competitiveness[2] rely on a combination of elements such as: institutions; industrial clusters; state policy and technology; management techniques and good labour relations; social, physical and digital infrastructure; access to capital; workforce skills; and local and geographical advantages. There is no consistent evidence that wage restraint fosters significant increases in productivity or underpins foundational competitiveness.

Countries like Malaysia (referenced positively by Seekings & Nattrass), India, China and Singapore are not successful because of their low absolute wages levels, but because their wages are low relative to their foundational competitiveness; Venezuela’s wages are similar to Malaysia’s, but Venezuela’s lower level of foundational competitiveness makes it a less attractive location (Delgado et al. 2012).

Similarly, Turkey’s clothing and textile sector consistently outcompetes South Africa’s despite Turkish wages being 24% to 41% higher in 2011. Contrary to Seekings & Nattrass’ assertions, ‘excessive’ wages did not drive the SA clothing sector’s decline. This had far more to do with inefficient production techniques, rapid liberalisation and an unsupportive government policy, a reorientation of the sector towards exports and neglect of the local market, followed by an appreciation in the exchange rate that made exports uncompetitive. More recent (tentative) growth has been achieved by taking advantage of geographical ‘localness’ and improving operational efficiency to provide local retailers with fast and flexible supply (Morris & Barnes 2014). Consistently advocating low wages for this sector, as Seekings & Nattrass do, ignores all this, as well as the fact that wages in the clothing sector are rising globally, as shown in recent research by the Boston Consulting Group (2014).

In agriculture – a sector highlighted by Seekings & Nattrass – South Africa’s lack of competitiveness (and the ensuing poor performance) is predominantly the consequence not of high wages but of inappropriate and unsupportive policies. State support for farms has declined to very low levels – with the value of producer support as a share of the value of total gross farm receipts being about 3% in 2008–2010, well below the OECD average of 20%.  At the same time most middle-income developing economies increased their support over the past decade (Sender 2014; Black et al., Econ3x3, November 2014). Similarly, between 1993 and 2006 funds for agricultural research and development decreased steadily, together with a dramatic fall in the number of agricultural researchers employed in SA institutions. The failure to expand irrigation and fertilizer production, coupled with under-investment in farming capital goods, is part of the litany of policy failures that are ignored by Seekings & Nattrass but which account for this sector’s anaemic performance (Sender 2014).

On an aggregate level, a National Bureau of Economic Research (NBER) study of over 130 countries confirms that the ‘naïve interpretation of competitiveness as low costs, especially low wages, is clearly misguided if prosperity is the policy objective… [they] do not constitute a root cause of competitiveness that underpins economic performance’ (Delgado et al. 2012: 7).

(b) What restructuring is needed?

Secondly, like many pro-business commentators, Seeking & Nattrass tackle the question of minimum wages assuming that the structure of the economy remains unchanged. The economy they envisage – by positive reference to Mauritius’ low-wage growth model of the 1970s and advocating low wages in tradable sectors – is premised on wage restraint. This is neither necessary nor desirable.

We argue that, in order to start to eliminate low wages and (as the cause of low wages rather than their consequence) low investment, low productivity and low employment, we should raise the levels of investment in the productive sectors of the economy. In this regard South Africa lags hopelessly behind its peers. When played correctly, high investment, high productivity and high employment can be accompanied by higher wages. This would require a significant restructuring of the South African economy.

Therefore, the implementation of a minimum wage should be an integral part of a package of far-reaching industrial policy (and other) interventions as well as the reorientation of macroeconomic policy. It 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.

This restructuring can be achieved by building upon sectors in which productivity can be advanced at levels that pay decent wages. The mining capital goods sector is a good example because South Africa has both the necessary raw inputs and a large local market for the products. For uncompetitive sectors, improvements in productivity through improved efficiency and innovation (as described above) are needed together with a possible reorientation to move away from producing cheap goods towards higher value-added and niche products. Expanding domestic markets, boosted by higher wage income, could underpin economies of scale and productivity increases, while local industry, in an era of fast production cycles, could leverage its proximity to the continent’s major retailers to increase exports. In this vision for the economy, tradable sectors that can sustain higher wages would be promoted.

In non-tradable sectors – such as private security, domestic work, construction, retail and restaurants – higher wages may go hand in hand with increased labour productivity and a decreased demand for labour but overall these sectors would be buoyed by rising incomes and investment spending that spur growth and expand employment.  

It is critical to have a nuanced government policy that would provide the necessary incentives and regulation combined with direct investment in skills, research and development and world-class infrastructure, as well as a stable, corruption- and crime-free environment. This would include tackling the oligopolistic nature of certain sectors and the rent-seeking behaviour of certain large corporations (for example, domestic steel, polymers and other industrial inputs should be priced at developmental, not monopolistic levels). There is near unanimous agreement that debt-financed spending on infrastructure raises productivity, enhances competitiveness and can have strong multiplier effects (IMF 2014).

Abandoning restrictive macroeconomic policies is essential to support such a programme. This means more government and public investment, lower and regulated interest rates, preferential financing for chosen industries, limiting financial market speculation and curbing the ten to twenty percent of GDP that large corporations, legally and illegally, spirit out of the country annually.

Since raising wages is not the only way to increase effective demand and reduce poverty and inequality, other means of wealth redistribution must play a role; raising the tax intake will be necessary to facilitate this and the other interventions described.

(c) The importance of the social context

Third, a low-wage regime is untenable for other reasons. Growing civil unrest and increased tension in labour relations indicate that South African workers (and the extended families who rely on their earnings) will no longer accept the current model that has spectacularly failed to generate employment and decent livelihoods, with unemployment growing steadily from the 1970s. Related to the above is the moral argument that workers simply should not be paid starvation wages.

The risks of Seekings & Nattrass’s approach

Seekings & Nattrass (2015) argue that, if a national minimum is to be instituted in the near future, it should be set ‘at the level of the lowest sectoral determination’ and that, above that, a system of varying sector-determined minima should remain – with the potential for higher minima to be set ‘in other sectors if the expected negative employment effects (i.e. job destruction) in these sectors are assessed to be modest’. 

This runs a high risk of largely entrenching the status quo. Sectors already covered by higher sectoral minimum wages are unlikely to be affected. Sectors not currently covered by any prescribed minima will now be covered by the minimum wage, but only at the level of the lowest sectoral determination. This is an extremely modest aspiration. Having minimum wages that are differentiated by sector is an approach that experts such as the International Labour Organisation (ILO) (2014: 99) have cautioned against and one which would leave the current wage structure almost entirely intact.

Since the model advanced by Seekings & Nattrass places the sole focus on (mostly short-term) labour market effects, it obscures the structural changes that are needed elsewhere to sustainably effect much better outcomes for workers. In addition, if international competitiveness is reduced to relative wages, the implication is that, if wages elsewhere in the world were to fall, the South African minimum wage would have to be adjusted downwards – and this could lead to the proverbial ‘race to the bottom’ in which countries cut wages to outcompete one another.

The macroeconomic argument

Finally, Seekings & Nattrass doubt whether higher wages can significantly spur domestic demand and growth. Our view that it can boost demand is hardly a fringe view. An ILO paper (Lavoie & Stockhammer 2012, which summarises key empirical findings regarding wage-led growth) shows that shifts between wages and profits in favour of wages have a positive impact on demand and productivity, also (for demand) in South Africa.[3] Further, the United Nations' Trade and Development Report of 2014 (UNCTAD 2014) argues that lacklustre global trade can only recover via ‘robust domestic-demand-led output recovery at the national level’ premised upon, amongst other factors, incomes policies that support the growth of demand. Finally, as the World Bank, IMF and OECD acknowledge (Ostry & Berg 2011; Cingano 2014), growth is retarded by high levels of inequality – which is extreme in South Africa, but which would be (partially) addressed through increased wages.

We accept Seeking and Nattrass’ point that consumption-led growth – as seen in South African in the mid-2000s – does not automatically on its own lead to positive domestic growth benefits. This is precisely why guiding investment in the economy towards sectors (such as manufacturing) with strong forward and backward linkages in the economy is crucial. In addition, there is a salient difference between the consumption-led growth of the mid-2000s – which was skewed towards high-end, often imported goods acquired by the rich (and fuelled by rising debt levels) – and the possible positive consequences of increased demand from workers’ wages in the scenario that we outline.

A minimum wage at what level? The need for research

Setting a minimum wage should not occur in isolation and other policy measures to increase employment and support high wages must be implemented in tandem. This is one reason why we do not immediately propose a level at which the minimum wage should be set (although we believe we need to move towards a higher-wage growth path and that a higher minimum wage than that proposed by Seekings & Nattrass is a crucial step in that direction ).

Another reason is that the current evidence is insufficient. We accept that the range of approximately R4 500 to R6 000, frequently mentioned in labour union circles as a possible target range for a future national minimum wage (Cosatu et al. 2014: 5;  Coleman 2013: 34), must be underpinned by further careful investigation. Research is needed into the potential employment impact of different levels of minimum wages in various economic circumstances, of which the current context is but one. Thus far this has neither been offered by ourselves and Coleman, nor by Seekings & Nattrass and other advocates of low minimum wages.

Such research must include at least an analysis of the current wage structure and the challenges facing its transformation, lessons from other developing countries, policy design, and modelling the impact of a national minimum wage on poverty, inequality and employment (given a variety of economic and policy scenarios).


The central implication of Seekings & Nattrass’s argument is that there is limited scope for higher wages in South Africa. This is because they erroneously reduce international competitiveness to relative wages. They argue that we (the authors) suffer from ‘unemployment denialism’ – but their model, which implies that addressing the South African unemployment situation requires wage restraint, is misconceived. Cheap labour is not an engine of growth and low wages are not efficient, nor do they naturally generate investment and dynamic advantages.

South Africa does not need to reinforce a low-wage regime. Rather, a national minimum wage must be part of a well-designed restructuring policy package that stimulates investment and productivity and otherwise supports the growth of employment in the relevant sectors, thereby establishing a high-wage growth path for the economy.

Admittedly, we propose a completely different type of policy intervention than that of Seekings & Nattrass: ours is structural and long term, theirs is about the short-term impact of particular minimum wage levels. There is, however, a danger in focussing solely on short-term labour market effects: it blinds us to discussing the deeper issues and fails to challenge the entrenched interests that would see the status quo maintained. The point of the national minimum wage is to transform the wage structure that has been premised historically on cheap labour. The goal is not to set the minimum wage at a level that has a high risk of reinforcing this structure.

To eliminate extreme poverty, reduce inequality and improve lives, working people need to be guaranteed a minimum amount of money in their pockets through decent wages and wealth redistribution. In this a national minimum wage has an important part to play.


Boston Consulting Group. 2014. Apparel at a crossroads: the end of low-cost-country sourcing. Boston.

Cingano F. 2014. Trends in income inequality and Its impact on economic growth. OECD Social, Employment and Migration Working Papers. Paris.

Coleman N. 2014. Myths and so-called ‘facts’ about the national minimum wage. Daily Maverick, November 10.

Cosatu, Nactu & Fedusa 2014. Labour fact paper on the national minimum wage for the labour indaba 4 November 2014. Briefing document (not a position document).

Delgado M, Ketels C, Porter ME & Stern S. 2012. The determinants of national competitiveness. Working paper 18249. National Bureau of Economic Research.

Fine B. 1998. Labour market theory: a constructive reassessment. Taylor & Francis.

Herr H, Kazandziska M & Mahnkopf-Praprotnik S. 2009. The theoretical dbate about minimum wages. Working paper 6. Global Labour University.

ILO. 2014. Minimum wage systems. International labour conference, 103rd session. International Labour Organisation.

IMF. 2014. World Economic Outlook (WEO), October 2014: Legacies, clouds, uncertainties. International Monetary Fund.

Isaacs G & Fine B. 2014a. The minimum wage debate – logic, evidence and need. Daily Maverick, November 25.

---------  2014b. Minimum wage debate: the old cheap labour system will get us nowhere. GroundUp, December 17.

Kaufman BE. 2010. Institutional economics and the minimum wage: broadening the theoretical and policy debate. Industrial & Labor Relations Review 63 (3): 427–53.

Lavoie, M & Stockhammer E. 2012. Wage-led growth: concepts, theories and policies. ILO working paper 470936. International Labour Organization.

Leonard T. 2000. The very edea of applying economics: the modern minimum-wage controversy and its antecedents. History of Political Economy 32 (5): 117–44.

Lester, RA. 1947. Marginalism, minimum wages, and labor markets. American Economic Review 37(1): 135–48.

Manyika J, Woetzel J, Dobbs R, Remes J, Labaye E & Jordan A. 2015. Global growth: can productivity dave the day in an aging world? McKinsey Global Institute.

Morris M & Barnes J. 2014. The challenges to reversing the decline of the apparel sector in South Africa. TIPS Conference, Johannesburg.

Ostry, JD & Berg A. 2011. Inequality and unsustainable growth: two sides of the same coin?  IMF Staff Discussion Note 11/08.

Porter ME. 2008. On competition. Harvard Business Press.

Seekings J & Nattrass N. 2014. What is at issue in the minimum wage debate? Econ3x3, January 20.

Sender J. 2014. Backward capitalism in rural South Africa: prospects for accelerating accumulation in the Eastern Cape. Eastern Cape Socio-Economic Consultative Council (ECSECC), East London.

UNCTAD. 2014. Trade and development report, 2014. UN Conference on Trade and Development.






[1] Regarding increasing minimum wages, the case of the United States and the other international studies we cite proves this point without implying that South Africa and the USA have identical challenges and solutions. For another exposition, see Isaacs & Fine (2014a); for supporting academic material, see Lester (1947), Fine (1998), Leonard (2000), Herr et al. (2009) and Kaufman (2010).

[2] ‘Foundational competitiveness’ is defined by Delgado et al. (2012) as ‘the expected level of output per working-age individual given the overall quality of a country as a place to do business’.

[3] We advocate an investment-led growth strategy but this does not preclude the positive benefits of increased wages.

Tags: Minimum wage

Download article

PDF icon Isaacs & Fine 2015 Minimum wages 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.