Leverage Bias,Volatility Bias And Stock Returns

As one of the core research areas of modern finance,asset pricing theory aims to analyze assets value of future payments under the uncertain condition.Portfolio selection theory,capital asset pricing model and option pricing model,etc.have won the Nobel Prize in economic sciences in past years.The empirical research of asset pricing stress the importance of investigating the relationship between the equity returns and risks,and finding new effective asset pricing factors that help in explaining equity returns.Traditional asset pricing research consider the pricing factors mostly based on absolute measure of firm characteristics,such as market capitalization,book-to-market ratio,leverage,PE ratio,etc.In this paper,we try to extend the traditional theoretical framework in a way that captures prospect theory proposed by Kahneman and Tversky(1979)and re-examine the relationship between capital structure,volatility and equity returns based on reference-dependent perspective.Prospect theory has been regared as theoretical basis of behavioral finance research.The most important contribution of prospect theory is the introduction of the concept of reference-dependence in the decision-making process.People generally evaluate utility based on the value of their wealth relative to an individually specific reference point rather than on the absolute value of their wealth.Based on the reference dependence characteristic of prospect theory,leverage bias and volatility bias has been constructed as new explanatory factors for equity returns and we further investigate the relationship between leverage bias,volatility bias and equity returns based on the background of Chinese securities market in this paper.The target leverage ratio calculated based on partial adjustment model developed by Flannery and Rangan(2006)is taken as the reference point of financing decision.And leverage bias is constructed as the deviation of the observed leverage from the target leverage.Besides,expected volatility calculated based on rolling regression of GARCH and GJR model is taken as the reference point of volatility.And volatility bias is constructed as the deviation of the realized volatility from expected volatility.Both leverage bias and volatility base is proposed to be helpful in explaining equity returns.Firstly,we examine the relationship between leverage bias,volatility bias and equity returns based on cross-section regression and explain the results under the framework of prospect theory.Then using time-series regressions,we further investigate the pricing factors of leverage bias and volatility bias based on portorfolio return analysis,which is in the way of multi-factor analysis and taken FF-three factors(Fama and French,1993)model as the benchmark.And we find that:In leverage bias analysis,we get that:(1)The target leverage ratio(reference point)of firms is determined as a linear function of firms' characteristics.(2)The introduction of reference point in explaining equity returns is important.Leverage bias is significantly related with equity returns and it provides a better explanation for equity returns than observed leverage.(3)The firm's leverage bias(over/under leveraged firms)combined with market condition(up/down markets)together determine the firm's leverage position and the loss/gain domains it's located,and further leads to different leverage-returns relationship.In general,the leverage and expected equity returns exhibit a positive relationship in gain domain,however a negative one in loss domain.(4)The pricing factor of leverage bias(PMN)contains additional valuable information besides SMB,HML and RMRF and it helps in improving the pricing performance of FF-three factors model(Fama and French,1993).In volatility bias analysis,we get that:(1)Volatility bias is proved to be significantly and positively related with equity returns.(2)Based on the loss aversion characteristic of prospect theory,the effect of volatility bias on equity returns with positive volatility bias(when realized volatility is bigger than expected volatility)is bigger than the effect of volatility bias on equity returns with negative volatility bias(when realized volatility is smaller than expected volatility).(3)The pricing factor of volatility bias(RMW)also contains efficient pricing information and helps in improving the pricing performance of FF-three factors model(Fama and French,1993).The three-factor model which contains RMRF,HML,and PMN,and the four-factor model which contains RMRF,HML,PMN,and RMW turn out to be the best two pricing models in this paper.The contribution of this paper is mainly reflected in:(1)Traditional asset pricing research consider the pricing factors mostly based on absolute measure of firm characteristics.However,we extend the traditional framework in a way that captures prospect theory and constructed leverage bias and volatility bias as new influencing factors for equity returns based on the reference-dependent characteristic of prospect theory.This paper enrich the literature of assets pricing and behavior finance.(2)This paper re-examine the leverage-returns relationship based on prospect theory framework and provide a potential solution to reconcile the puzzling empirical results documented previously to some extent.(3)There has been little research about the relationship between volatility and cross-sectional equity returns in the past literature.This paper made up for the lack of previous research and focus on the effect of volatility on cross-sectional equity returns.