Does monetary policy uncertainty command a risk premium in the Chinese stock market?

• Published date: 01 September 2022
• Author: Lei Lin,Jing Tan,Wenzhen Liu
• Date: 01 September 2022
• DOI: http://doi.org/10.1111/irfi.12356

ORIGINAL ARTICLE

Does monetary policy uncertainty command a risk

premium in the Chinese stock market?

Lei Lin

1

| Jing Tan

2

| Wenzhen Liu

3

1

School of Finance, Southwestern University

of Finance and Economics, Chengdu, China

2

China Western Economic Research Center,

Southwestern University of Finance and

Economics, Chengdu, China

3

School of Economics and Management,

Anhui Normal University, Wuhu, China

Correspondence

Lei Lin, School of Finance, Southwestern

University of Finance and Economics,

Chengdu 610074, China.

Email: leilinfe@gmail.com

Abstract

We examine the pricing implications of monetary policy

uncertainty (MPU) in the cross-section of stock returns in

the Chinese stock market. Our results show that the long-

short portfolio that is long stocks with the lowest exposures

to innovations in MPU and short stocks with the highest

exposures to innovations in MPU earns about 6% annual-

ized alpha. The coskewness, idiosyncratic volatility, size, and

book-to-market effects cannot account for the low (high)

average returns earned by stocks with high (low) exposures

to innovations in MPU. Furthermore, the exposure to inno-

vations in MPU commands a significantly negative risk pre-

mium in the Fama-MacBeth regressions. These results

indicate that investors have preferences for the early reso-

lution of MPU.

KEYWORDS

consumption growth volatility, monetary policy uncertainty, risk

premium

JEL CLASSIFICATION

G11; G12

1|INTRODUCTION

Several important papers show that the policy uncertainty based on news affects financial markets and stock prices

(Baker et al., 2016; Brogaard & Detzel, 2015; Pastor & Veronesi, 2013). Moreover, both academic studies and inves-

tors agree that stock prices are significantly driven by monetary policy, and then respond differently in the cross-

section (Bernanke & Kuttner, 2005; Nielsen & Bender, 2010; Ozdagli & Velikov, 2020). However, how monetary

Received: 28 August 2020 Revised: 16 March 2021 Accepted: 19 June 2021

DOI: 10.1111/irfi.12356

© 2021 International Review of Finance Ltd.

International Review of Finance. 2022;22:433–452. wileyonlinelibrary.com/journal/irfi 433

policy uncertainty (MPU) based on news affects the investors' optimal investment decisions and then stock prices in

the cross-section, little empirical research has been done. We would expect to see investors care about more MPU

than other policy uncertainties because stock prices always respond directly to the changes in the monetary policy.

For example, Ai et al. (2020) show that stocks with high sensitivities to monetary policy announcements require sig-

nificant risk compensation. In this paper, we show that the cross-section of stock returns commands a significantly

negative risk premium for bearing the exposure to innovations in MPU.

To understand the underlying economic mechanism, the long-run risk model of Bansal and Yaron (2004) pro-

vides an insight. In the theoretical model, they show that investors care about the state variable that induces changes

in consumption growth, that is, consumption growth volatility (time-varying economic uncertainty). These investors

prefer to resolve the economic uncertainty in advance because they fear that a rise in economic uncertainty will

lower stock prices. Subsequent researches confirm their prediction that the consumption growth volatility is indeed

an important and separate risk factor. Bansal et al. (2014) show that the consumption growth volatility news requires

a risk premium different from the discount rate news. Ai and Kiku (2016) show that the aggregate volatility risks

affect the changes in the volatility of aggregate consumption and then stock prices. In sum, if the MPU is a state vari-

able that is positively correlated with consumption growth volatility, the investors prefer the MPU be resolved in

advance. Then, the negative risk premium of innovations in MPU is consistent with the long-run risk model.

We start our empirical analyses by investigating the above cross-sectional pricing implications of MPU. For this

purpose, we use the MPU index based on the news of Huang and Luk (2020) as a proxy to quantify the policy uncer-

tainty of China's central bank. Then, at the end of each month, we run the AR(p) processes for MPU to extract inno-

vations in MPU. After this, we calculate the time-series average correlations between the innovations in MPU and

several well-known factors. The small correlations preliminarily show that the innovations in MPU are a new risk fac-

tor that predicts the cross-sectional differences in stock returns. Hence, we can construct a tradable portfolio, which

is long stocks with the lowest exposures to innovations in MPU (βΔMPU) and short stocks with the highest βΔMPU,to

earn an excess return.

Specifically, our strategy for earning an excess return is as follows. First, at the end of each month, we estimate

βΔMPU using time-series regressions of excess returns on the innovations in MPU, market, size, value, momentum fac-

tors for all stocks in the Chinese stock market. Second, we show that the stocks with the highest βΔMPU have lower

average returns than the stocks with the lowest βΔMPU. In our single-sorted portfolio analysis, the value-weighted

high minus low (H-L) βΔMPU portfolio (long highest βΔMPU stocks and short lowest βΔMPU stocks) generates a negative

Fama and French (2015) 5-factor (FF5) alpha of 0.56% with a Newey and West (1987)t-statistic 2.97.

1

Third, in

the double-sorted portfolios, controlling for cross-sectional predictors of coskewness (COSKEW), idiosyncratic vola-

tility (IVOL), size (SIZE), and book-to-market (BM), respectively, the stocks with the highest βΔMPU still generate sig-

nificantly negative FF5 alphas than the stocks with the lowest-β

ΔMPU

. These results are also robust to the equal-

weighted portfolios. At last, using the Fama and MacBeth (1973) regressions, we estimate the risk premium of the

exposure to innovations in MPU at the firm level and find a significantly negative coefficient of 0.174 between

βΔMPU and 1-month-ahead excess returns after contemporaneously controlling for COSKEW, IVOL, SIZE, and BM. In

the robustness tests, we also show that the MPU is different from those state variables that are known to proxy for

economic uncertainty, consisting of market volatility, and economic performance index (EPI) (Ai & Kiku, 2016;

Brogaard & Detzel, 2015). After controlling for these competing variables, the negative risk premium on innovations

in MPU remains significant. In addition, our results continue to hold with different portfolio grouping.

Our finding contributes significantly to the policy uncertainty literature by strengthening the important role of

MPU in affecting the consumption growth volatility and then the cross-section of stock returns. First, we show that

the MPU is a new state variable of the long-run risk model that affects the consumption growth volatility. Second, to

our best knowledge, no research examines the cross-sectional effects of MPU in the Chinese stock market. Third,

we provide empirical evidence supporting that investors prefer to hold stocks with high βΔMPU for resolving the eco-

nomic uncertainty in advance. In sum, we confirm that the innovations in MPU are a newrisk factor reflecting a neg-

ative policy uncertainty premium in the Chinese stock market.

434 LIN ET AL.