Liquidity premium and the Corwin-Schultz bid-ask spread estimate

Purpose - The purpose of this paper is to explore the price implication of a newly developed estimator of the bid-ask spread by Corwin and Schultz (2012). The paper focusses on whether the new measure as a liquidity proxy commands a significant premium. The research helps the understanding on the validity of the Corwin-Schultz estimate as a liquidity measure. Design/methodology/approach - The authors carry out their examination based on the portfolio approach, cross-sectional regressions, and time-series regressions. For comparison, the authors also adopt other three liquidity proxies and mainly rely on the Fama-French three-factor model as the benchmark. The sample includes NYSE/AMEX/ARCA/NASDAQ ordinary common stocks over 1926-2010. Findings - The paper finds that Corwin-Schultz spread lacks significant power to predict returns either in the pre- or post-1963 period. In contrast, other liquidity measures such as the price impact of Amihud (2002), trading discontinuity of Liu (2006), and turnover show stronger return predictability than the Corwin-Schultz spread estimate. Research limitations/implications - The evidence indicates the limited ability of the Corwin-Schultz spread estimate to describe liquidity. Practical implications - The comparison of the Corwin-Schultz spread with other liquidity measures helps practitioners and academic researchers to identify the appropriate proxy. Originality/value - This paper, for the first time, provides a thorough assessment of the Corwin-Schultz spread estimate as a liquidity proxy, which distinguish from Corwin and Schultz (2012) who focus on whether their spread estimate measures transaction costs. Our study not only helps practitioners and academic researchers to select an adequate liquidity measure and an asset pricing model to use, but it also sheds light on the current debate about whether transaction costs have the first order importance in asset pricing.