This study empirically tested the standard CAPM and DR-CAPM to examine the cross-sectional variation in stock returns of financial institutions listed at PSE over the period 2002-2016. The analysis is also performed for three sub-periods: pre-financial crisis period (2002-2005), during- financial crisis period (2006-2009), and post-financial crisis period (2010-2016). The empirical analysis is carried out following the two-pass regression analysis of Fama & Macbeth (1973). The results of the study suggest that the standard CAPM is not suitable for calculating the risk and required rate of returns for Pakistani financial institutions. To examine the validity of DR- CAPM, we estimate the downside beta developed by Bawa and Linderberg (1977), Fishburn (1977), Harlow and Row (1989), and Estrada (2002). Out of these models, the DR-CAPM of Fishburn (1977) appeared to be more appropriate for calculating required rate of returns. The DR-CAPM of Harlow and Row (1989) also provides evidence of a positive and statistically significant risk-return relationship in most of the examined sub-periods. However, the results for Estrada (2002) are inconclusive at best, suggesting that the downside beta of Estrada (2002) is an inefficient measure of risk. The results help investors in identifying an appropriate and suitable measure of risk for financial institutions of Pakistan, which, in turn, enables the investors to design an efficient and well-diversified portfolio. The results are also of great significance to managers of financial institutions as they help them in making capital budgeting decisions and calculating the cost of equities.
Read full abstract