Turnover: Liquidity or Uncertainty

Abstract

Turnover is usually used as a proxy for stock liquidity. The negative relation between turnover and expected stock returns is regarded as evidence of the existence of liquidity premium. Turnover is also often used to measure the firm-specific uncertainty. The positive relation between turnover and expected stock returns is considered as the compensation for uncertain risks. Obviously, there is a conflict in the interpretation of the turnover risk information. Taking A shares in the Chinese stock market as the research object, we find that stocks with low turnover generally have higher risk premium than those with high turnover through portfolio analyses and cross-sectional regression analyses. Moreover, after controlling variables such as size, trading volume, beta, liquidity and uncertainty, the result is still robust. However, after building more subdivided portfolios according to turnover, we find that turnover can positively predict stock returns for stocks with low turnover, but it can negatively predict stock returns for stocks with high turnover. That is, the cross-sectional expected return is the first increasing and next descending function of turnover. We take the directional reversal point of the return forecast ability as the dividing point of the turnover level, and find that the uncertainty proxied by the total volatility and the idiosyncratic volatility can significantly explain the positive relationship between turnover and expected returns for stocks with turnover below the cross-sectional dividing point, but the liquidity can significantly explain the negative relationship between turnover and expected returns for stocks with turnover above the cross-sectional dividing point. Thus, the lower degree of turnover contains more firm-specific uncertainty information, and the higher degree of turnover reflects more stock liquidity information. During our sample period, the trading mechanism in the Chinese stock market has undergone major changes. After the sample period is divided into four stages according to several major trading mechanism reforms in the Chinese stock market, the empirical results are still significant and therefore the reforms of the transaction mechanism have not fundamentally changed the turnover risk information. By defining the turnover risk information, this paper not only provides theoretical support for other related research by dint of turnover, but also has direct guiding significance for investment practice according to turnover.