Abstract
When measuring poverty in South Africa, much of the theoretical and empirical work has focused on money-metric measures of poverty. The
conventional approach has been the use of a poverty line sufficient to meet primary human needs, often derived from consumption, expenditure or
income levels. This narrow perspective has tended to divert the attention of development economists towards the pursuit of income growth, which has
been associated with development and a necessary condition for poverty reduction. However, with a changing emphasis on development, the model,
analytical methods, and related frameworks have been subjected to critical scrutiny by populists and neoliberals alike. The purpose of this study is
to contributes to the existing literature by using an alternative approach ─ a non-monetary approach to the measurement of poverty in South Africa.
To the best of our knowledge, this is the first study in South Africa to compute an asset poverty index using principal component analysis in a panel
setting and use appropriate panel data models to investigate the key determinants of asset poverty. We used data drawn from the existing five waves
of the National Income Dynamic Study. Using the random effect probit model, we found that some factors, such as levels of education of the head of
household (primary, secondary, matric and tertiary) and land ownership, have a reducing influence on asset poverty. Factors The results also revealed
that household size and unemployed people are more prone in South Africa.