Title

Systematic Risks and Asset Pricing

Abstract

This study examines the asset pricing mechanism in Pakistani equity market for the period 6/1998 to 6/2008 under general equilibrium framework and proposes an efficient multifactor model for asset pricing. CAPM Tests indicate that historical beta is unable to predict expected returns. However, market risk premium is priced and has significant positive relationship with returns. Size, book to market, illiquidity and P/E ratio factors are priced by market. However, momentum factor is not priced by market. Small stocks earn higher return in comparison to big stocks. Small P/E stocks earn higher return in comparison to high P/E stocks. High B/M stocks earn higher return in comparison to low B/M stocks. In Pakistan, high turnover stocks earn more than low turnover stocks. So it can be concluded that value stocks in general outperform growth stocks. Fama and French’s Three Factor model is valid model and size and value premium have significant positive relationship with equity returns. Its explanatory model is 15% higher than conventional CAPM. Carhart’s Four Factor model does not provide additional insight as momentum factor is not priced. The proposed liquidity based Four Factor model reveals that illiquidity factor can explain the equity returns and high liquidity stocks earn more than low liquidity stocks. It means Pakistani investors prefer liquidity. Proposed Five Factor model extends Carhart’s Four Factor model by considering illiquidity premium and it confirms the presence of significant negative relationship between ILLIQP and equity returns. The explanatory power of the model touches 80% in this case. The proposed Six Factor model further extends Five Factor model by adding P/E premium and reports that P/E premium has significant positive relationship with equity returns but iii does not add value to model. The comparison of explanatory powers of models reveals that Five Factor model best explains the returns in Pakistani equity market.

This study also provides evidence about existence of long term causal and dynamic relationship between macroeconomic variables and equity returns by using multivariate cointegration analysis. Unidirectional causality is also observed flowing from X Rate, TBill, Money Supply, and CPI to equity market returns. However, no Granger Causality is observed in industrial production and equity market returns. ARDL Approach also confirms that industrial production, oil prices, inflation and foreign portfolio investment are not found statistically significant while interest rates, exchange rates and money supply have significant long run effect on equity prices. The error correction model based upon ARDL approach also reports the same behavior in short term. It is worth mentioning that foreign portfolio investment is not significant in long term but it is statistically significant in short term. Study further indicates that adjustment process is quite fast and 39% of the disequilibrium in equity prices from its equilibrium path is corrected in subsequent time period.

Download full paper