Effect Of Tail Risk And Its Applications To Factor-pricing In The Chinese Stock Market

A stock’s tail risk describes the risk that the stock will incur a significant loss with a very small probability.Represented by the financial crisis that swept the world in 2008 and the COVID-19 epidemic that has ravaged all sides since 2020,cases of significant portfolio impairment due to the outbreak of small probability events have become common in global stock markets in recent years,bringing great risks and challenges to the stable development of global stock markets and investors’ wealth appreciation.Thus,the demand for investors in the market to manage the tail risk of stocks and avoid major losses has become increasingly urgent.At present,the central government attaches great importance to the prevention and control of financial risks and the maintenance of financial market stability.As a key node of the financial system,the potential risks in the Chinese stock market are bound to significantly impact the stable development of the entire economy and society due to its deep connectivity with various sectors of society,including enterprises,investors,and financial institutions.In this context,the tail risk of listed stocks has become the top priority of risk prevention work because of its wide impact and seriousness of consequences.Therefore,an in-depth study of the series issues on stock tail risk,especially the return-related ones,can not only help guide investors’ investment practice but also provide a micro basis for regulators to formulate reasonable risk management policies,which is of great significance to promote the stable and healthy development of the Chinese stock market.In academic research,the effect of risk on asset prices has been a significant theme in the realm of asset pricing.The early definition of risk emphasized the uncertainty of asset returns and was not concerned with whether returns were positive or negative.With the introduction of the safety-first principle,the disaster-avoidance motivation,the prospect theory,and others to reconceptualize asset risk,scholars have focused their understanding of risk on the loss component of asset returns,arguing that the uncertainty of loss is the exact source of asset risk and better reflects investors’ psychological perception of risk.On this basis,the tail risk of stocks,as a risk that portrays the uncertainty of extreme losses in stocks,has naturally become the focus of scholars’ attention.Currently,relevant domestic studies mainly focus on discussing the relationship between a stock’s tail risk and expected returns and testing the existence of the tail risk effect.However,the inconsistent conclusions of specific relationships,the incomplete analysis of the underlying mechanisms,and the lack of pricing-factor application make the process of how the tail risk of stocks affects their future returns still unclear to various market participants,and thus also needs to be explored more deeply and comprehensively.To this end,based on the listed stocks in the Chinese A-share market,this thesis takes the tail risk of stocks as the entry point of the research perspective,uses methods such as portfolio sorting analysis,Fama-Mac Beth regression,transition probability matrix,propensity score matching-difference in difference model,and numerical simulation,and constructs the theoretical models based on investors’ underreaction and risk preference,to thoroughly analyze the impact of a stock’s tail risk on its expected returns in the Chinese stock market.Around this core issue,specific researches are conducted in four aspects: the measurement and identification analysis of stock tail risk,the test on a stock’s tail risk effect,the mechanism analysis of the effect,and the pricing factor application for a stock’s tail risk,thus forming a systematic research framework on the series issues of stock tail risk in the Chinese market.Through the exploration and analysis of these contents,the thesis aims to strengthen the perception and understanding of market participants on the stock tail risk and ultimately serve the goal of preventing risks in the Chinese stock market and ensuring its stable operation.The main contents and related findings of this thesis are summarized as follows.First,the ability of tail risk proxies to identify the real tail risk of stocks is examined.After comparing the ability of three types of tail risk proxies in the Chinese stock market,the thesis finds that the relatively simple and intuitive loss-type proxy is the most effective in identifying real tail risk.In contrast,the investors in China may not be sensitive to the beta-type or idiosyncratic-type metrics that rely on complex computational capabilities.Second,it is also found that the dynamic tail risk indicator incorporating the time dimension information can also identify the real tail risk of stocks well,which lays the foundation for broadening the research horizon on a stock’s tail risk.Second,the thesis examines the specific performance of the stock tail risk effect in the Chinese market and examine the impact of various market characteristics on this effect.Firstly,the analysis from both static(level of tail risk)and dynamic(change in tail risk)perspectives confirms the existence of a significantly negative stock tail risk effect in the Chinese stock market,presenting a tail risk anomaly.The tail risk effect from the dynamic perspective is stronger and more stable,and the results of the interaction analysis of the two perspectives show that the static tail risk effect is a partial reflection of the dynamic effect,indicating that investors in the Chinese market mainly correspond to the dynamics of stock tail risk.This result largely expands on existing research findings of the tail risk effect and suggests that investors should percept the stock tail risk from a dynamic perspective.Secondly,a negative dynamic tail risk effect at the industry level is also significantly present in the Chinese stock market,providing a new channel for tail risk management at the industry level.After the industry adjustment of the dynamic tail risk of stocks,the negative tail risk effect at the stock level becomes weaker,suggesting that the stock tail risk effect is partly derived from the industry level effect.Thirdly,small-cap stocks weaken the performance of the static tail risk effect,which reconciles the inconsistent findings in existing literature regarding this effect.In contrast,small-cap stocks do not significantly affect the dynamic tail risk effect.Fourthly,the release of short-selling constraints brought about by the margin trading system helps reduce the tail risk of the target stocks and mitigate their corresponding negative effect.However,the price limit and the switch in market conditions(sentiment,volatility,liquidity,macroeconomic conditions)has quite limited impacts on the tail risk effect of stocks.Third,the returns of tail risk anomaly in the Chinese stock market cannot be explained by the risk compensation channel,but stem from the stock mispricing due to investor’s behavioral biases.On the one hand,investor underreaction is an important cause of mispricing,but unlike the U.S.market,investors in the Chinese stock market mainly underreact to stocks with decreased tail risk and also lower their expectations of better future fundamentals in such stocks.Further,higher attention limits,greater information uncertainty,and stronger limits-to-arbitrage can all increase investor underreaction,leading to a stronger negative tail risk effect.However,institutional investors in the Chinese market do not mitigate the mispricing of stock prices,and they exhibit similar cognitive and behavioral biases as retail investors,indicating a different understanding of the role of institutional investors.On the other hand,the thesis reveals that investors’ risk preferences are also a possible cause of the negative effect of stock tail risk.After distinguishing between stock capital status and investors’ psychological expectations of stocks using the capital gain overhang(CGO)and prospect theory value(TK),this thesis confirms that investors prefer risk when they are in the loss domain or have higher psychological expectations of stocks,also resulting in a stronger negative tail risk effect.Fourth,stock tail risk is an important factor that drives the co-movement of stocks,and the tail-risk factor is thus constructed on this basis.The tail-risk factor is found to be significantly priced in the Chinese stock market,and the results of spanning tests show that the factor contains incremental information on return differences that cannot be explained by mainstream multi-factor models,and therefore,it is not a redundant factor.In terms of model pricing ability comparison,the hybrid four-factor model extended by the tail-risk factor performs very well.First,the hybrid four-factor model can provide an annualized maximal Sharpe ratio(2.05),and the return of corresponding tangent portfolio is mainly provided by the tail risk factor(with a weighting ratio of 0.57).Second,the hybrid four-factor model can explain 28 of the 37 anomalies significantly present in the Chinese stock market,which explains the largest number of anomalies.In detail,the hybrid four-factor model explains the price and underreactionrelated anomalies well.Overall,the contributions of this thesis can be summarized as follows:(1)The thesis proposes three criteria,i.e.,being able to identify the skewness and kurtosis of the return distribution and exhibiting the expected return effect,to assess the ability of proxies to identify the real tail risk of stocks in the Chinese stock market and subsequently finds the suitable tail risk proxy for stocks in the market,which makes up for the lack of comprehensive comparison of the tail risk proxies of stocks in the existing literature.(2)The understanding of stock tail risk is expanded from the existing static perspective to a dynamic perspective,which broadens the scope of mind to investigate the impact of a stock’s tail risk.The empirical results also confirm that investors in the Chinese market mainly correspond to the dynamics of stock tail risk,and the static tail risk effect is only a partial reflection of the dynamic effect.The results provide an important reference value for investors to accurately construct tail risk management or investment strategies,and for regulatory authorities to take oriented tail risk prevention measures.(3)Little literature discusses the tail risk effect of stocks from the perspective of the industry level.The thesis examines the existence of the industry tail risk effect and its impact on the stock tail risk effect.This finding also provides a practical basis for tail risk management and prevention at the industry level.(4)The mechanism analysis of the stock tail risk effect is deepened and supplemented.Through a simple model that incorporates investors’ underreaction and a two-period portfolio model that combines the expected utility theory and the cumulative prospect theory,the thesis aims to facilitate market participants’ understanding of the intrinsic mechanisms of the stock tail risk on affecting the future stock returns in the Chinese stock market.(5)By taking the tail risk factor as a behavioral factor,the thesis expands the existing multi-factor model with the factor,aiming to improve the pricing efficiency of the factor model on the stock markets.This exploration enriches the research literature on market anomalies and factor pricing in China and also provides an effective alternative benchmark pricing model for the market.Based on the above results and the characteristics of China’s stock market,this thesis puts forward the following policy recommendations: 1)Further improve the information disclosure system of listed enterprises and strengthen the punishment on financial fraud and insider trading,so as to provide the market with a transparent information environment and a fair platform for investment and financing;2)Continue to expand the scope of stocks covered by the margin trading system and appropriately lower the threshold of the system to improve the short-selling mechanism of the market and enhance its efficiency in pricing stocks;3)Increase efforts to cultivate and regulate the analyst market and enrich the channels for investors to understand the operating conditions of listed enterprises and the risks they face,thus promoting the information transmission efficiency of the market.