Essays on the temporary employment services sector in South Africa: an analysis using administrative data

Abstract

In a country with high levels of inequality, the debate on the temporary employment services (TES) sector centres around whether the sector can provide productive employment opportunities or whether it perpetuates poor working conditions, reinforcing labour market segmentation. It became clear that even after revisions were made to the country’s overarching labour legislation at the advent of democracy, those employed within TES and other atypical workers had limited legislative protection. It was not until 2015 that amendments were made to that part of the Labour Relations Act No.66 of 1995 (LRA) that governs temporary employment, requiring client firms to treat TES workers no less favourably than their direct hires. The scope of existing data and literature on the TES sector in South Africa and in the Global South is limited and the inability to identify TES workers is a common problem across household surveys. Existing surveys and case studies on the TES sector in South Africa have been limited to small samples of firms that are not nationally representative, leading to divergent estimates of the size of the sector and the nature of TES employment, and limiting the assumptions that can be made about the sector as a whole. The administrative tax data released by the South African government in 2015 provided an opportunity to examine the TES sector in more detail as it accurately identifies TES firms in the formal sector as well as their workers. This study contributes to the body of work on the TES sector by addressing three unique aspects in three main chapters. The first builds a profile of the TES sector, describing what we currently know about the size and shape of the sector in South Africa. To do this, the chapter draws on a variety of sources, including historical accounts, legal cases, existing studies and surveys, and importantly a descriptive analysis of the newly released administrative tax data. The aim is to synthesise and critically analyse the information available on the sector and draw attention to the research gaps. The second chapter focusses on measuring the size of the gross wage penalty for TES workers (relative to non-TES workers), and examining how the administrative data can be used to unpack the drivers of this penalty, such as differences in worker characteristics between the TES and non-TES sectors; differences in the nature of the job itself; and differences in the conditions under which workers accept a job in the TES sector. Furthermore, this chapter considers whether wages are structured differently by comparing TES and non-TES workers’ contributions to certain benefits such as pensions and medical aid. A substantial TES gross wage penalty is found, which remains high, even after accounting for individual fixed effects and controlling for a set of job and 3 firm characteristics available in the tax data. Although the wage differential does not decline appreciably when benefits are taken into account, TES workers are much less likely to report benefit contributions than non-TES workers, and when they do, TES worker contributions are a fraction of what non-TES workers contribute. The third chapter estimates the short-term impact of the amendments to the LRA in 2015, which aimed to strengthen the rights of TES workers, on employment, earnings and job duration between 2015 and 2016. The findings suggest that for TES workers that transition into the nonTES sector, wages improved. The amendments had mixed effects on worker wellbeing in the short run.