Connectedness among stocks and tail risk: Evidence from China

• Published date: 01 December 2021
• Author: Zhijun Hu,Ping‐Wen Sun
• Date: 01 December 2021
• DOI: http://doi.org/10.1111/irfi.12320

ORIGINAL ARTICLE

Connectedness among stocks and tail risk:

Evidence from China

Zhijun Hu

1

| Ping-Wen Sun

2

1

School of Finance, Jiangxi University of Finance and Economics, Nanchang, China

2

Newhuadu Business School, Minjiang University, Fuzhou, China

Correspondence

Ping-Wen Sun, Newhuadu Business School,

Minjiang University, Fuzhou, China.

Email: sunpingwen@gmail.com

Funding information

National Natural Science Foundation of China,

Grant/Award Numbers: 71463018,

71801117; National Office for Philosophy and

Social Sciences, Grant/Award Numbers:

14CJY064, 18BJY241; Science and

Technology Project of Jiangxi Provincial

Department of Education, Grant/Award

Number: GJJ190263

Abstract

Applying the market tail risk measure proposed by Kelly and

Jiang in the China's A-shares market, we find that the

monthly market tail risk significantly and negatively predicts

the monthly industrial output growth rate up to 1 year. In

addition, from July 2007 to June 2019, we find that stocks

with a higher tail risk outperform stocks with a lower tail risk

by 0.62% (0.30% after risk adjustment) per month. Using

the institutional holding weight within an industry and cor-

relations in return on assets as proxies for the connected-

ness among stocks associated with firm fundamentals, and

treating the sentiment risk and correlations in the three fac-

tor risk-adjusted residuals as proxies for the connectedness

among stocks associated with investor sentiment, we fur-

ther show that the connectedness among stocks signifi-

cantly affects individual stocks' tail risk and tail risk

premium. Moreover, our findings show that the connected-

ness components of tail risk associated both with firm fun-

damentals and with investor sentiment can significantly and

positively predict stock returns. Our finding suggests that

the connectedness among stocks provides an important

channel through which firm fundamentals and investor sen-

timent influence the tail risk premium in the China's A-

shares market.

Received: 3 December 2019 Revised: 28 April 2020 Accepted: 10 June 2020

DOI: 10.1111/irfi.12320

© 2020 International Review of Finance Ltd. 2020

International Review of Finance. 2021;21:1179–1202. wileyonlinelibrary.com/journal/irfi 1179

KEYWORDS

China, connectedness, firm fundamentals, investor sentiment,

tail risk

JEL CLASSIFICATION

G11; G12

1|INTRODUCTION

Recently, Kelly and Jiang (2014) propose a market tail risk measure by identifying a common fluctuation component

from firm-level price crashes in each month to overcome the difficulty of directly constructing a market tail risk esti-

mate due to infrequent extreme events. Kelly and Jiang (2014) find that, in the U.S. market, stocks with returns more

sensitive to the lagged market tail risk measure have higher returns in the future. In our study, to see whether the

market tail risk is also a priced factor in the China's A-shares market, we first examine whether the market tail risk

can predict real economic activities in China. Second, in contrast to the U.S. market which is dominated by institu-

tional investors in Kelly and Jiang (2014), we explore whether the tail risk premium also exists in the China's A-shares

market which is dominated by individual investors. Third, we attempt to identify a potential channel through which a

firm's characteristics may influence its tail risk premium. As Kelly and Jiang (2014) point out that the tail risk of public

firms' sales growth is significantly correlated with the market tail risk, their finding suggests that a stock's connected-

ness with other stocks due to firm fundamentals affects a firm's tail risk premium. Furthermore, because Barberis,

Shleifer, and Wurgler (2005) contend that, in addition to firm fundamentals, investor sentiment influences the com-

ovement among stocks as well, we also examine whether the connectedness among stocks resulting from investor

sentiment contributes to the tail risk premium.

Our first research question is to investigate whether the market tail risk can predict real economic activities in

China. Kelly and Jiang (2014) find that the market tail risk highly correlates with the sales growth tail risk in the

U.S. market and attribute this correlation to a tail uncertainty shocks channel. As Bansal, Kiku, Shaliastovich, and

Yaron (2014) contend that an increase in macroeconomic volatility (uncertainty) leads to an increase in discount

rates, there may also exist a negative relationship between the market tail risk and real economic activities. Specifi-

cally, we investigate whether the market tail risk negatively predicts industrial output growth in China. This investiga-

tion can help us understand whether economic fundamentals can support the market tail risk to be a priced factor.

Our second research question is to examine whether the tail risk premium also exists in the China's A-shares

market. Many studies document that investors require a premium to hold a stock whose price drops significantly

when the market declines. For example, in the U.S. market, Ang, Chen, and Xing (2006) document an annualized 6%

downside risk premium. Bali, Cakici, and Whitelaw (2014) document a significant and positive relation between the

hybrid tail covariance risk and a firm's expected stock return. Kelly and Jiang (2014) find that stocks with high load-

ings on past market tail risk earn an annualized three-factor risk-adjusted return of 5.4% more than stocks with low

market tail risk loadings. Chapman, Gallmeyer, and Martin (2018) verify the tail risk premium documented by Kelly

and Jiang (2014) and find that the tail risk premium results from the discount rate component of returns rather than

the cash flow component of returns. Chabi-Yo, Ruenzi, and Weigert (2018) find that investors receive compensation

for holding crash-sensitive stocks. Although those studies document a significant and positive tail risk premium in

the U.S. market which is dominated by institutional investors, it is not clear whether the tail risk premium also exists

in the China's A-shares market which is dominated by individual investors. While many studies assume that individ-

ual investors are more irrational than institutional investors, DeVault, Sias, and Starks (2019) find that the sentiment

metric proposed by Baker and Wurgler (2006) capture institutions' demand shocks rather than individuals' demand

shocks.

1

Therefore, our investigation that whether the tail risk premium also exists in the China's A-shares market

can help us understand whether the investor composition may influence the existence of the tail risk premium.

1180 HU AND SUN