A Test of Asset-Pricing Models at the Nairobi Securities Exchange

Abstract

The Capital Asset Pricing Model (CAPM) has for a long time been used to explain the variations in expected return on stocks. However, the discoveries of market anomalies such as the Size, Book-to-Market and the Momentum effects, have greatly undermined CAPM’s ability to explain the expected returns on stocks. These anomalies prompted Fama and French (1993) and Carhart (1997) to propound asset pricing models that captured the effects of these anomalies. This study sought to test whether the CAPM, Fama and French (1993) Three-factor model and the Carhart’s (1997) Four-factor model can explain the returns of stocks traded in the NSE, from a portfolio perspective. The stock returns used in this study were those for the forty eight companies that trade under the MIMS in the NSE, during the period January 2009 to December 2013. Six portfolios that were sorted for size and Book-to-Market were created and used to test the CAPM as well as the Fama and French (1993) Three-factor model. Also, an additional six portfolios that were sorted for size and past performance were constructed to test the Carhart’s (1997) Four-factor model. The data was then analyzed using time series regression analysis and the estimated parameters were tested for significance. This study finds that even though the CAPM has been highly regarded for many years, when tested at the NSE from a portfolios perspective, the evidence in support of it is weak. This study finds that other significant factors exists that were not captured by CAPM, implying therefore that beta is not an adequate measure of risk. Also, as for the Fama and French (1993) Three-factor model, this study finds that it doesn’t quite capture all the factors influencing the returns of stocks traded at the NSE. However, this study finds that the Carhart’s (1997) Four-factor model performs better relative to the CAPM and the Fama and French (1993) Three-factor model, as it was observed to have a better explanatory power of the variation of expected returns of most of the sets of portfolios that it was tested on. The findings of this study will be of great significance to the finance academia and policy makers as it will assist in boosting their understanding of an asset-pricing model that can explain better, the variations in returns of stocks traded at the security exchange.

Keywords: Asset pricing Models, NSE, Kenya