School of Economics
All jobs are equal, but some jobs are more equal than others: what a clustering algorithm reveals about the labour market segmentation in South Africa
Abstract
dc:description.abstractThe theory of labour market segmentation is often used to explain wage inequality in South Africa. It posits that labour markets consist of different segments that are differentiated from other segments based on some observable characteristics. Limited mobility between these segments prevents wages from equalising. In this paper, I make use of a clustering algorithm to identify these informal and formal segments in the labour market. I apply this methodology to a nationally representative panel dataset of employed South Africans. I find that employed South Africans fall into one of three segments. Two of these segments resemble what is commonly referred to as formal and informal employment. The third segment lies somewhere between these two segments and is referred to as semiformal employment. Findings suggest that informally and semiformally employed individuals have s relatively lower chance of entering formal employment. Those in formal employment are much more likely to remain formally employed. It also appears that much of the churning observed in the labour market is between semiformal and informal employment, rather than between informal and formal employment. The identification of this group of semiformally employed labour market participants helps contributes to a more nuanced understanding of labour market dynamics in South Africa and highlights the value that data driven methodologies can provide to applied labour economic analysis.
Degree
thesis:*- Grantor
- School of Economics
- Year dc:date.issued
- 2021
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Kensett, Jonathan Matthew
- Advisors dc:contributor.advisor
-
- Ranchhod, Vimal
- Leibbrandt, Murray
Subjects
dc:subject × 1Identifiers
dc:identifier.*- Handle dc:identifier.uri
- http://hdl.handle.net/11427/35760
- OAI identifier oai:identifier
- oai:open.uct.ac.za:11427/35760