Determinants of housing insecurity in a low income South African township

Abstract

The campaign to end poverty by 2030, as advocated by the World Bank, has multifaceted components. Food security and abject poverty have received more attention; however, little attention has been focused on housing insecurity. This is important, especially given the fact that it is widely acknowledged that housing satiates an essential human need for physical security, dignity and prosperity, among other things. Housing, for some time, has been observed as crucial to individual and family functioning. As indicated by organisations such as the United Nations Habitat, the importance of appropriate housing is explained by the fact that safe housing is seen as a basic need among other needs such as food, all of which are regarded as central components of the security of ordinary households, particularly in developing countries. In 2015, more than 1.5 billion people in the world were living in housing that was unsafe and inadequate, while in the meantime, millions experienced eviction from their homes every year, with at least 100 million people becoming homeless on any given day. Therefore, it is pertinent to understand what determines housing insecurity. The study aims to investigate factors that determines housing insecurity. Using data collected from Sharpeville and Bophelong, two low income townships in South Africa that are particularly vulnerable to housing insecurity, the study employs a multinomial logistic regression to investigate the effects of head of household and the general household characteristics on housing insecurity. The results of the regression analysis show that income, gender, number of people in the household and the amount paid to bond or rent are significant predictors of housing insecurity. Contentious issues like access to land and job opportunities are areas of further study in understanding the multifaceted phenomenon of housing insecurity.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 4 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.