| This dissertation is an empirical investigation of the importance of liquidity and trading activity for asset pricing.; The first chapter analyzes the ability of trading activity to explain cross-sectional variation in expected stock returns. We depart from the previous literature in not taking for granted that turnover is solely a proxy for liquidity. Instead, we test the impact of trading activity on monthly stock returns, after controlling for the usual factors (firm size, book-to-market ratio and momentum) and for illiquidity costs (firm-level). We estimate illiquidity costs with intraday data, following a microstructure model for the price impact of a trade and transitory trading costs. We consider a large sample of NYSE and NASDAQ stocks from 1993 to 2002. The results for the entire sample period provide evidence that higher turnover rates are associated with lower future returns after controlling for these costs. We also find evidence that the effect of illiquidity costs is related to firm size. Yet, for large and glamour stocks, which are very liquid, the effect of trading activity is still statistically and economically significant. These findings call into question the presumption that trading activity is solely a proxy for liquidity.; In the second chapter, we re-investigate the importance of liquidity as an additional priced risk factor in a standard three-factor asset pricing model. Recent empirical findings reporting a, high premium due to liquidity risk motivate another look at its effects using an alternative measure of (aggregate) market liquidity. We follow previous studies in defining liquidity risk as the sensitivity of portfolio returns to market liquidity fluctuations. We contribute to the empirical literature by constructing a time-series of market liquidity innovations (liquidity factor) based on microstructure models of trading costs, and estimated with intraday data. We test a standard factor model specification including the liquidity factor, for 25 portfolios sorted by size and book-to-market. We show that liquidity risk is not able to significantly explain the cross-sectional variation in returns for this choice of portfolios. There is a weak improvement in the fit, but the liquidity risk premium is not statistically or economically significant. |
