Further Evidence of Firm Size and Earnings Anomalies on the Johannesburg Stock Exchange

Abstract

In spite of the large body of evidence in favour of capital market efficiency in the semi-strong form, several enigmatic findings persist. As much of the research is based on the theoretical foundation of the Capital Asset Pricing Model, the debate concerning these findings has revolved around whether the market is inefficient or the Capital Asset Pricing Model is mis-specified. This paper outlines a methodology for establishing alternative Arbitrage Pricing Theory based benchmarks for use in tests of market efficiency. The methodology is then applied to an analysis of the size and earnings anomalies evident on the Johannesburg Stock Exchange (de Villiers, Lowings, Pettit and Affleek-Graves, 1986; Page and Palmer, 1991). The persistent evidence of the anomalies suggests that any mis-specification within the Capital Asset pricing model must also exist within the multi-factor Arbitrage Pricing Theory lending weight to the inefficient market conclusion. This finding is contrary to that of Biger and Page (1993) who found that the apparent superior performance of unit trusts was no longer evident when using Arbitrage Pricing Theory based benchmarks.