The Conditional Relationship between Risk and Return: An Empirical Analysis of Listed Firms in Sri Lanka

Abstract

The study mainly focuses on the relationship between risk and return of the firms listed in the Colombo Stock Exchange of Sri Lanka during the period from June 2003 to May 2015, which includes the civil war period ended in May 2009 as well. Accordingly the main objective of this study is to examine whether the risk-return relationship that is prescribed in the Capital Asset Pricing Model (CAPM) still holds during varying market conditions. Thus the study sought to ascertain the conditional relationship between risk and return by using monthly cross sectional data which are obtained through secondary sources such as the Central bank of Sri Lanka and Colombo stock exchange of Sri Lanka. The study sample contains 91 companies under the individual analysis whereas 18 portfolios are formed to carry out the portfolio analysis. Initially the study employs the basic CAPM model to analyze the prescribed association between risk and return of stocks and secondly the full sample period is further segregated in to different market conditions as up market and down market, by considering the nature of the excess market return within a given period. Accordingly the study basically carries both unconditional and conditional relationship tests which have been suggested by Fama and MacBeth (1973) and Pettengill et al. (1995) respectively. The study reveals a flat risk-return relationship during its full sample period whereas some inter-temporal inconsistencies have also been identified under basic risk-return tradeoff. Thus it could not reject the null hypothesis and thus the prescribed unconditional relationship has become invalid. More importantly the data sample includes a large proportion of down market months which has negative market excess return and it could be due to the low stock market performance during the civil war period. However the empirical findings under conditional risk-return relationship test during times of positive and negative excess market returns reveal that the conditional relationship only holds during the down market period, whereas the up market risk premium proved to be positive as expected, but insignificant. However these findings do not necessarily reject the conditional relationship due to the empirical nature of the study. Thus the study concludes that there is no significant evidence to prove the existence of basic positive risk-return relationship, however an inverse relationship exists between risk-return during the periods of negative excess market returns. Hence the investors have to carefully analyze that how the prescribed risk-return relationship behaves during different market situations before making their stock investment decision.