Introduction
If ever we are to have rational economic policy supporting a broad-based improvement in living standards in South Africa, we must have greater clarity in our positive and normative[1] thinking about the economy than we do. The intention of this brief is to identify obstacles to clear thinking about the national minimum wage. It does not, indeed it cannot, describe what this clear thinking is. The relevant work has not been done. Rather, the purpose here is to try and understand what holds us up in getting the purchase we need on the issues.
Analysis
- An official unemployment rate of 34.9%, as measured in the third quarter of 2021 by the Quarterly Labour Force Survey is generally recognized as a major economic failure. So it is. But it is not enough to say so and stop. Efficient markets clear. In the case of the labour market, this would mean that everyone who wants to work at the going wage can find it. This does not mean zero unemployment since workers quitting one job, for whatever reason, often need time to find another[2]. But the number of work seekers and the number of vacancies should be roughly in balance. This isn’t the case in contemporary South Africa. So the labour market is inefficient and the question is why – a question which is seriously under-investigated.
- Some take the view that to think in terms of a labour market is to make oneself an accomplice to the ‘commodification’ of labour, in itself dehumanizing, and to ‘neoliberalism’. While this is not the position in the NMWC Report as whole, it reflects in heavy emphasis on need and political determination with too little weight given to the economic consequences of policy decisions. One would be reluctant to disqualify the view as a normative position, though it has consequences which might startle. But leave that rabbit hole for another day. Here consider the addition of a second normative premise: that the private sector should bear the major responsibility for creating employment. Can both normative positions be held together coherently? No, because the private sector needs both factor (input) markets and product (output) markets to function. So which is it to be – private sector led employment growth or no labour market?
- There is a long and venerable tradition of concern with the subsistence wage. But we should not be blind to its limitations. One can trace it back to the first half of the twentieth century, and even to some official documents, most notably in the reports of the war-time Social and Economic Council and the Fagan Commission[3]. At its best, it was motivated by the moral concern that household incomes should not fall below subsistence level. Less admirable, but relevant was the realization, as one leading industrialist put it, that “if they don’t eat, we don’t sleep”. Self-regarding though that comment was, it at least recognized interdependence and thereby even a degree of social cohesion. But the weakness of the tradition was that it elided the distinction between individual income and household expenditure, inevitably so in an era when there was no information on the relation between the two. The information position has now changed and the tradition needs reconsideration and renovation.
- The terms ‘wage elasticity of employment’ and ‘output elasticity of employment’ may seem forbiddingly technical. But the idea behind them is not, and it is important. Usually, as the price of a good rises, the demand for it drops and as income rises, the demand for the good rises. The price elasticity of demand is simply a measure of the sensitivity of demand to price and the income elasticity of demand is a measure of the sensitivity of demand to income. The definition of price elasticity of demand is simply the ratio of the percentage change in quantity to the percentage change in price. Thus, if the price rises by 1% and demand drops by 2%, the price elasticity of demand is 2% divided by 1%, which is two, and we put a minus sign on it to indicate that the direction of change in demand is opposite to the direction of change in price. Similarly, if income rises by 1% and demand rises by 0.5%, the income elasticity of demand is 0.5% divided by 1% which is 0.5 and we put a plus sign on it to indicate that the direction of the change in demand is the same as the direction of the change in income.
Now, if both price and income increase at the same time, the effects offset each other: the increase in price reduces demand, while the increase in income increases it. But what happens if price increases and income decreases? There is then a doubly negative effect on demand: the price increase leads to a decrease in demand, and the income drop also leads to a decrease in demand. The price of labour is the wage, so that if wages increase while output drops, there is a doubly negative effect on employment. This is why policy makers have to be so careful under such circumstances. The size of the impacts will depend on the values of the elasticities, which is important to know. Estimating them is not easy. Characteristically there will be a lot of things changing at the same time so it is difficult to identify the specific effects one is looking for. But it is not impossible, and attempting it can yield important insights into the functioning of the labour market.
What we have in the report on this is two illogicalities and an absence of information. We read:
While the overall employment decreased, the study[4] indicates that the decline in employment was more pronounced amongst workers earning less than the national minimum wage (17%), compared to workers earning more than the national minimum wage (11%). The marginal disproportionate employment impact on low-wage workers cannot be attributed to the national minimum wage increase…[5]
But why are obliged to reach this conclusion? One may make exactly the opposite argument. After all, paying less than the minimum wage is now illegal, so that its existence puts pressure on employers who have been paying less than the minimum to move towards it, a pressure which does not exist for employers who have been paying more than the minimum wage. This would account for the quantitative finding cited. It is a consequence of the existence of the minimum wage and could be expected to stronger the higher the minimum wage becomes.
We also read:
A qualitative study conducted by the internal research unit of the Department of Employment and Labour in 2021[6] indicates no evidence of an effect on employment due to the national minimum wage in the domestic sector and construction sector. Any reported changes are attributed to the COVID-19 lockdown restrictions and the economic decline... The study also indicates that the impact of the national minimum wage on hours, employment and non-wage benefits cannot be distinguished from the impact of COVID-19 on businesses and other economic factors[7].
Qualitative research has the consequence that the sample in one or both of the studies is not a probability sample. Standard statistical techniques cannot be used, so we are left with an unrepresentative set of off-the-cuff opinions which cannot capture the full effects of the national minimum wage. Even worse: the passage cited is that the third sentence states that the impact of the national minimum wage cannot be identified, whereas the first sentence implies that it can. Which is it?
Conclusion
The aim of this brief is to amplify the argument made in our submission to the Department of Labour. In doing so, it has been necessary to confront some very poor argument. As advice to the Minister, the Report simply won’t do.
To read the Report of the National Minimum Wage Commission on the Review and Adjustment of the National Minimum Wage for 2022, click here.
To read the Helen Suzman Foundation’s Submission, click here.
Charles Simkins
Head of Research
charles@hsf.org.za
[1] Positive analysis is about what is the case. Normative analysis is about what the case ought to be.
[2] This form of unemployment is known as frictional unemployment.
[3] ‘Segregation has fallen on evil days’ said Smuts in an address to the SA Institute of Race Relations in 1942 and the documents cited dealt with the implications of an integrated future. They were, of course, repudiated by the National Party after 1948.
[4] The study referred to is by the Development Policy Research Unit at the University of Cape Town. The full reference is Haroon Bhorat, Adaiah Lilenstein and Ben Stanwix, The impact of the national minimum wage in South Africa: Early quantitative evidence, Development Policy Research Unit, DPRU Working Paper 202104, April 2021 available at http://www.dpru.uct.ac.za/sites/default/files/image_tool/images/36/Publications/Working_Papers/DPRU%20WP%20202104.pdf
[5] Government Gazette 45649, p 32
[6] No authors, no title, and above all no access. Accordingly, no basis for assessing the adequacy of the study.
[7] Is this the same study referred to in Annexure A of the Report? If it is, there is still no proper referencing and access, and there is a further problem, since Annexure A says the study was based on an earlier study by the Centre for Social Development at the University of Johannesburg, for which there is no title and no facilitation of access
