Multifactor explanation of security returns in South Africa

Abstract

This paper evaluates the performance of the Fama and French three–factor model in South Africa for individual securities. We employed a multivariate time series methodology similar to Fama and French. The empirical results contradict the theoretical proposition of the Fama–French model and are inconsistent with the results documented by most studies in the developed and some emerging markets. The size and value premia are very weak when included in the regression model. Furthermore, the Fama and French three–factor model is unable to explain the return–generating process of securities trading on the Johannesburg Stock Exchange. This has important implication for corporate managers, investors as well as fund and portfolio managers in terms of estimating cost of equity, rate of return and portfolio allocation.