Authors: Ihsaan Bassier, London School of Economics and Political Science and Vimal Ranchhod, University of Cape Town
Minimum wage policies are typically aimed at reducing poverty. Yet there is little direct evidence of this effect, especially in developing countries. And none for South Africa.
In a recent paper, we consider the income, employment and poverty effects of the largest minimum wage increase South Africa has ever seen. In 2013 the agricultural minimum wage increased by about 50%, affecting nearly all workers in the sector.
South Africa’s agricultural sector is a low-wage industry contributing about 2% of GDP. Over the last four decades the sector has shed jobs, and relied increasingly on casual labour. Farmworkers have among the lowest average wages in the country, with poor living and working conditions. While farmworkers account for less than 5% of all employment, they comprise about half of employment in rural formal areas.
The agricultural minimum wage was first introduced in 2003, and has been adjusted annually in line with inflation. In November 2012 in the Western Cape province, worker protests quickly spread, with the demand for a wage of R150 a day. South Africa’s labour minister compromised by increasing the minimum wage from R69 to R105 a day in March 2013 (from $7.15 to $10.88 in 2013).
Such a large increase might have been expected to lead to job losses, with devastating consequences. This is what employers warned at the time – that the higher wages would need to come from existing profit margins.
But there weren’t in fact big job losses. And we found that household income and poverty rates strongly benefited in the short run. At the same time, our results also indicated low minimum wage compliance among the lowest paid workers – employers did not comply with the new minimum wage. This undermines the policy.
It may also have lessened its negative effects because employers who didn’t comply may have been the employers with the lowest profit margins. If they had complied, they might have had to fire a lot of workers.
Our study provides reason for optimism about the effects of minimum wage policies on poverty, while keeping in mind the complex relationship of these policies with compliance and institutional enforcement.
The 2013 hike
In our paper we document the implementation of the new minimum wage.
Notes: The average for quarter 4 of each year is shown. The last row shows the percentage change as calculated by comparing the years before the new minimum wage (2010-2012) to the years after (2013-2014). Agricultural employment is a percentage of all employment. The poverty rate refers to 1.5 times the reference poverty line of R1,042 (see Budlender, Leibbrandt and Woolard 2015). The wage bill is defined as the total hourly wages of farmworkers in millions of ZAR. Source: Own calculations using the Labour Market Dynamics surveys 2010-2014, Statistics South Africa.
Average hourly wages increased sharply from around R7.50 in 2012 to R9.70 in 2013 after the minimum wage implementation, which shows the policy had a clear impact on workers’ wages.
However, agricultural employment declined over the same period. Despite the employment loss, poverty dropped substantially, and the total wage payments to workers (the wage bill) rose.
Nevertheless, these trends may not reflect causal effects of the minimum wage. This is because there are a number of features of agriculture that complicate the story.
Firstly, an important feature of agriculture is the volatility of economic and agricultural conditions, including sensitivity to weather. Table 1 shows that much of the employment decline between 2012 and 2013 appears to be driven by an unusual spike in employment in 2012, perhaps reflecting such volatility.
Secondly, while average wages increased, they were still well below the new minimum wage. Research done in 2012 documented low compliance with minimum wages in South Africa, especially in the agricultural sector.
Thirdly, the average change in the poverty rate (the proportion of people who are poor) depends on who is classified as poor. For example, there was a decrease of only 4 percentage points in the poverty rate when classifying those with monthly income below R1,042 (US$110 in 2013) as poor (compared to a decrease of 10 percentage points when using a higher amount).
Overall, despite the large increase in average hourly wages and total wage bill presented in Table 1, these features of the agricultural sector make causal claims tricky.
Measuring the causal effect
To better understand the effects of the new minimum wage, we used data from the Quarterly Labour Force Surveys, which is the official source of labour market data.
We tracked a representative rotating panel of about a thousand workers who were paid below the new minimum wage just before the policy was implemented, and compared their income and employment to just after.
A worker who used to be paid a lot less than the new minimum wage would need to be paid a lot more for the employer to be compliant with the new minimum wage. We assessed the extent to which the changes in incomes and employment coincided with the gap between a worker’s previous wage and the new minimum wage.
On average across all workers, we estimated that the minimum wage increased hourly wages by 5.6%, increased the chance that a worker stayed employed, increased household income by about 6.3%, and decreased the poverty rate.
Notes. Outcomes are taken as a 1-quarter change between quarters 1 and 2. The outcomes are a percentage change in wage (column 1); percentage-point probability of employment retained (column 2); percentage change in household income per person (column 3); and percentage-point change in probability of falling below the poverty line (column 4). Source: Own calculations using confidential panel data from the Labour Market Dynamics survey 2010-2013, Statistics South Africa. * p < 0.1, ** p < 0.05, *** p < 0.01. Standard errors in parentheses. Source: Author Get the data. Created with Datawrapper.
However, these effects varied by how much workers were paid before the new minimum wage. Workers paid just below the new minimum had the biggest wage gains, while, counterintuitively, those who were initially the lowest paid workers had the smallest wage gains.
These lower-paid workers on average had the most positive employment effects though, such that their wage and employment effects offset each other. That is, lower paid workers were more likely to keep their jobs. Together then, workers across different levels of income had similarly large positive effects on household income and poverty.
Where this leaves the debate
Small wage gains for the lowest-paid workers means that minimum wage compliance was lowest where it mattered the most.
Why could this have happened?
One possibility is what economists call “endogenous compliance”, which would reduce unemployment effects. Workers, bosses, and enforcement authorities may “turn a blind eye” to minimum wage violations when they’re worried that strict enforcement would cause unemployment.
Consistent with this, we found much weaker compliance in small firms, where we would otherwise expect the most negative effects on employment.
Another explanation is that the lowest paid workers were the workers least able to force employers to comply, and these employers were more likely to keep them employed as a result.
Overall then, the effects of the minimum wage were strongly positive. The biggest limitation of our study is that we considered only the short run. And while we found similar effects two quarters out, we left estimates of medium and longer run adjustments to future work.
These positive short-run effects of such a large policy change provide some optimism that there is institutional space to substantially improve the lives of poorly paid workers, while being cognisant of those left behind in the process.
This article is republished from The Conversation under a Creative Commons license. Read the original article.
RELATED ARTICLE
Debunking some big and small myths about the national minimum wage
