Measuring labour law violation: An application to South Africa

Mandela Initiative newsletter (Issue 3, July 2017)



Debates about the impact of labour regulations on labour market outcomes (e.g. wages and employment) are often contested and revolve around the legislation – whether it is too strict or too permissive. Very little attention is given to issues of enforcement and compliance, where the underlying assumption is that the regulations in question are adequately enforced. Yet in South Africa, and many other developing countries, this is not the case. Studies show that noncompliance with labour regulations is widespread. For one component of the Mandela Initiative’s labour markets theme, a research team led by Prof. Haroon Bhorat, DST-NRF South African Research Chair in Economic Growth, Poverty and Inequality: Exploring Interactions for South Africa, measured the extent of this noncompliance for both wage and non-wage labour regulations in South Africa by using an innovative technique.

Labour law violation was found more common in areas with higher unemployment, and smaller firms were more likely to be non-compliant

The study created a multidimensional violation index (MVI) that offers a macro perspective and also investigates some of the key determinants of violation in South Africa. This provides a comprehensive measure of violation, and serves as an indicator of vulnerability in the labour market. This new index also allows us to investigate some of the determinants of violation in South Africa and contribute to the broader discussion on labour market regulation and its impacts.

What is the MVI?

We measure violation using data from the 2014 Labour Market Dynamics Survey, adapting methodology from the multi-dimensional poverty literature to combine wage and non-wage measures into a single index. Different types of violation (e.g. wages, leave, Unemployment Insurance Fund) are grouped into three weighted dimensions: wages, leave, and 'other'. We then use these dimensions and the multi-dimensional poverty index methodology to create two indices: The first focuses on both wage and nonwage violation, while the second looks only at non-wage violation. We also use principle component analysis (PCA) to check our index weighting methods.

From these indices we can extract some summary statistics (see figure below) where:
H = proportion of violated employees;
A = average intensity of violation; and
MVI is the multi-dimensional violation index (or H multiplied by A).

Labour law violation diagram

What are the key findings?

The overall MVI measures suggest that non-wage violations in South Africa are substantial. While minimum wage violations are widespread, the index suggests that the intensity of non-wage violation is what contributes to the difference between the MVI outcomes. Examining some correlates of violation reveals that local labour market conditions are an important factor in the probability of violation, where unemployment levels matter.

For instance, we plotted the local unemployment rate against the violation level for both the wage and non-wage indices. In both cases we find a strongly positive relationship: labour law violation is more common in areas with higher unemployment. Similar to the wage curve literature, this may suggest that unemployment lowers worker bargaining power and results in higher levels of labour law violation by employers. Regression results suggest that a 1 percentage point increase in the local unemployment rate is associated with an increase of 0.22 points in the combined MVI. We also find strong relationships between violation and firm size, where smaller firms are more likely to be non-compliant. Variations of non-permanent contracts are also correlated with much higher levels of violation.



Labour law violation C.Kirschoffsmall

Image: Chris Kirchoff/

A further key finding is that non-wage violation indicators are positively correlated with minimum wage violation. What this means is that there is a higher probability of non-wage violation if the minimum wage is violated. This begins to present questions about the type of firms that are most likely to violate, and why they choose to comply with some laws and not others. Further research will also use the MVI to track levels of violation over time, and across countries.

Some policy implications

Quantifying violation in South Africa into an index format reveals that more attention should be paid to noncompliance and, in particular, aspects of non-wage regulation which are widespread, and where the 'intensity' of violation is high. Our findings take us into the heart of the debate around South Africa’s search for jobs, where the wage-employment trade-off seems in part to be working through the medium of violation of wage and non-wage conditions of employment. Analysis of the determinants of labour violation offers insights into where attention should be paid most keenly: Specifically small firms, areas with high unemployment, and workers with nonpermanent contracts.


There is a higher probability of non-wage violation, such as leave violations, if the minimum wage is violated



This article was written by Haroon Bhorat, Ravi Kanbur (Cornell University), Ben Stanwix and Amy Thornton, Development Policy Research Unit, University of Cape Town. The publication of this study is forthcoming.