How to draw conclusion on Vavi’s R4,500 minimum wage

Vavi has disappointed me with his call for a R4,500 minimum wage.

There is no question that higher pay is better, all else being equal. I expect the overwhelming majority of eyes that have ever glanced at this blog belong to beings that earn more than R4,500 a month.

There is no question that lower income inequality would be more desirable. There is no question that we would be a happier society with fewer people living off very little income.

There is also no question that a higher minimum wage would, all else being equal, lead to an increase in unemployment and therefore quite likely an increase in income inequality and a decrease in the productive capacity of our country, a likely increased tax burden on tax payers and greater social ills.

Let’s look at this a little more clearly. This took me all of a couple of minutes to draw, so I can’t believe Vavi hasn’t seen the same drawing somewhere. (If he disagrees with it, fine, but then he should address the problem head-on and not just ignore the problem.)

Increasing the minimum wage had a doubly negative impact on employment
Increasing the minimum wage had a doubly negative impact on employment

The starting position

  • The demand for labour is downwards sloping. More labour is demand if it costs less. Theory says, quite sensibly, that companies will add more labour as long as the marginal cost of labour (the cost of hiring another employee) is less than the marginal product of that labour.
  • The supply of labour is interesting. Mostly it’s upwards sloping. The higher wages are offered, the more labour will be supplied. (This is true up to a point, a point far away from where we are here.)  The increase in supply comes from two causes.  Firstly, people who are happily not working or under-employed because they value leisure time, because it’s expensive to have someone to look after dependants, because the cost of getting to work is too great, some of those will go to work if the wage offered is higher. Some of this then doesn’t reflect a decrease in unemployment since it reflects an increase in the labour participation rate. However, for some it will definitely reflect the ability to get a job because the search for employment can extend farther from the home. Then, those who are currently working may opt to work 6 days rather than 5 or more or longer shifts. So overall the supply of labour is upwards sloping.
  • The supply of labour curve does something interesting at a certain low point. It becomes zero for all values of P (the price of labour of the wage) below a specific point. This is because below a certain wage, the costs of working (including in large part transport) are higher than the wage offered for virtually all labour. Thus, no labour will be supplied. If I were to draw the diagram again I might have made the curve more differentiable, with a gradient starting nearly flat and moving to vertical at full employment to reflect this relationship more accurately. But this one explains the point well enough.
  • Where we are now is at Pc and Qc, the current price (wage) and supply of labour. I’m also naively assuming that there is no minimum wage now either (I think another graph is actually needed to show that deadweight loss already.)
  • The current problem is twofold.  Pc is viewed as too low by many, and Qc is far below Qf (the full employment level of supply of labour, where the supply of labour should really be vertical…) which is a catastrophic problem.

The first impact

So that’s the starting point, with all the problems clearly identified. Now comes Vavi’s R4,500 minimum wage. We’re presenting this as an increase in the minimum wage from Pc to Pv

  • Instantly, this causes the quantity of labour that wants to be supplied to increase to “?”. This is the mystery magical hoped for number that doesn’t exist, because the demand for labour decrease to Qv0.
  • At higher wages, the point where the marginal cost of labour is equal to the marginal product of labour comes at a smaller amount of labour. Thus we have a reduction in employment and an increase in the rate of unemployment (possibly also a decrease in the labour participation rate, which is only slightly less bad than an increase in unemployment since we are still losing productive capability in the economy and families are still not getting the income.)

The double whammy

That’s the first negative impact. The second negative impact is that all those newly unemployed people are now consuming less.

  • The reduction in income is (Pv-Pc)*Qv0 – (Qc-Qv0)*Pc-0.5*(Pv0-Pc)*(Qc-Qv0) although that last compound term is a clear approximation. The triangle is in fact the deadweight loss to society from the minimum wage since nobody is benefiting from that component any more.
  • For a range of assumptions (although not all to be fair) the result will be an overall decrease in income, which will give rise to a decrease in consumption.
  • The decrease in consumption will decrease the demand for labour overall, shifting the labour demand curve from 1 to 0. This further decreases the quantity of labour demanded at Pv to Qv1.
  • The change from Qc to Qv1 reflects an increase in unemployment and a decrease in the labour participation rate.
  • It also means lower GDP per capita, lower tax revenues, more income inequality and more social problems

So well done Vavi. Genius.

What we should be trying to do

  • Shift the supply of labour curve downwards by reducing the costs of working, specifically the cost of getting to work (an awful Apartheid legacy)
  • Shift the demand for labour upwards (training, skills development, youth wage subsidies, maybe payroll tax credits)
  • Maybe, just maybe, decreasing the total cost of employment to employers through very careful, very limited increases in labour market flexibility. But very carefully and very limited.

The problems with our country’s labour force are structural. We need problems that fix the underlying structural problems, not populist and germ-infected band-aids that actually stop the sores from healing.

Published by David Kirk

The opinions expressed on this site are those of the author and other commenters and are not necessarily those of his employer or any other organisation. David Kirk runs Milliman’s actuarial consulting practice in Africa. He is an actuary and is the creator of New Business Margin on Revenue. He specialises in risk and capital management, regulatory change and insurance strategy . He also has extensive experience in embedded value reporting, insurance-related IFRS and share option valuation.

Join the conversation


  1. David – how would one go about doing this: “…decreasing the total cost of employment to employers through very careful, very limited increases in labour market flexibility.”?

    1. The entire quote probably reflects my nervousness here:

      Maybe, just maybe, decreasing the total cost of employment to employers through very careful, very limited increases in labour market flexibility. But very carefully and very limited.

      I probably need to do a little more digging and research before I have a useful answer, but the constant pressure against labour brokers seems to me to be counter-productive.

Leave a comment

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.