Essay I Liquidity Premium and Catching-up Effect: Evidence from the Japanese Stock Market...[ Read more ]
Essay I Liquidity Premium and Catching-up Effect: Evidence from the Japanese Stock Market
We extend the asset-pricing model with focus on liquidity of Amihud and Mendelson (1986) and Brennan and Subrahmanyam (1995) to incorporate previous market information. Consistent with their results, we find that without knowing the previous market information, expected returns are negatively related to the turnover ratio. If we further assume that liquid and illiquid stocks react to the previous market information asymmetrically, then the return-turnover relationship becomes more (less) significant for an up (a down) market in the previous period. Using the monthly data from the Japanese stock market where there are no formal market makers, we find that the empirical results support the prediction of the model. The pattern of the asymmetric return-turnover relationship is insensitive to the estimation methods and is robust to the Fama-French three factor model. We also find that about half of the asymmetric liquidity premium is due to the correlation between the portfolio return and the lagged market return, i.e., the "lead-lag" effect.
Essay II Asymmetric Liquidity Premium and Lead-Lag Effect Across Firm Size and Exchange*
In this article we are concerned with the return-turnover relationship in the U.S. stock market across firm size and exchange. We find that expected returns are negatively related to turnover ratios for both NYSE/AMEX firms and NASDAQ firms. Nonetheless, this relationship is more profound for small stocks than large stocks. In addition, this relationship is stronger for NASDAQ firms than NYSE/AMEX firms. This evidence indicates that the market structure in the NYSE/AMEX provides more liquidity than that in the NASDAQ. However, both market structures provide similar liquidity for the transactions of large stocks. We also document that the return-turnover relationship is more profound conditioned on an "up market" than a "down market" in the pervious month. This pattern of asymmetric liquidity premium is robust even after controlling for the "lead-lag" effect. This asymmetric pattern also exhibits in the size and book-to-market effects. Finally, we find that the liquidity premium does not exhibit any seasonal pattern for both the NYSE/AMEX firms and the NASDAQ firms.
Essay III A re-examination of the relationship between stock returns and predetermined firm attributes
This paper re-examines the relationship between stock returns and a set of' predetermined firm attributes, namely past return, turnover, firm size and book-to-market ratio. The results from NYSE/AMEX stocks and the NASDAQ stocks indicate that past return and turnover are the two major determinants for cross-sectional expected stock returns even after controlling for the Connor and Korajczyk (1988) risk factors. The evidence suggests that the profit of the momentum strategy (i.e., to buy past winners) could be enhanced when the turnover ratios for winner and loser stocks are taken into account. Furthermore, while the momentum and size effects are more profound when conditioned on a previous "down market", the turnover and book-to-market effects are more significant when conditioned on a previous "up market." This implies that glamour stocks (as characterized by large firm size, low book-to-market ratio, high turnover, and past winner) outperform the value stocks (as characterized by small firm size, high book-to-market ratio, low turnover, and past loser) more when the previous market movement was down than when it was up. These patterns are consistent with the notion that glamour stocks overreact to the common and firm-specific information and/or value stocks react with a delay to the common and firm-specific information. Finally, we find that the momentum effect is concentrated in non-January months, while the turnover, size and book-to-market effects are more apparent in January months.