我国股票预期回报率中包含显著的非流动性补偿和流动性风险溢价,而且非流动性补偿和流动性风险溢价更为显著地出现在流动性较差或者说价格冲击弹性较高的股票上。由于流动性水平和流动性风险在股票横截面上存在强相关性,使得代理流动性水平的非流动性指标能够抓住流动性风险的溢价效应。经典资本资产定价模型的价格风险敏感度对股票回报率缺乏解释力,意味着股票的风险溢价更多地来自对流动性风险的补偿,而非来自对单纯的价格风险的补偿。
Based on a simple liquidity risk-adjusted asset pricing model, this paper empirically studies the liquidity risk premium and illiquidity compensation in Chinese stock markets. The evidence shows that there are significant illiquidity compensation and liquidity risk premium in stock expected returns. In addition, the lower liquidity or the higher price impact elasticity the stock may be, the more significant the tests on its illiquidity compensation and liquidity risk premium will be. Our new illiquidity proxy variable can grasp liquidity risk premium in stock returns because of high cross-sectional correlations between the level of liquidity and liquidity risk. Besides, the market β of classical CAPM shows less power in interpreting stock returns data. So the required risk premium in Chinese stock returns mainly compensate for systematical liquidity risks, not for pure price risk. This paper offers two contributions. First, a new illiquidity proxy measuring the level of illiquidity is established. The illiquidity proxy is a much more accurate measure than that of Amihud (2002). For, in order to denote the price impact by trading orders we use intra-daily price range instead of Amihud's use of the daily rate of return. This treatment could eliminate the case of the .change :of price not initiated by trading. Using this new proxy we show a significant illiquidity premium in our domestic stock markets consistent with other scholars' research. Second, in our asset pricing model we decompose market liquidity risk into the exogenous and endogenous liquidity risk in the investors' perspective. Besides showing the known exogenous risk premium, we find that the endogenous risk makes a significant impact on the cross-sectional stock price variation which has not been explored by other studies. Important implications of finding significant liquidity risk premium mean that Chinese stock markets are not as what some studies (for example Wu Wenfeng et al, 2003) have argued that there is no risk pr