The Impact of Firm‐Level Illiquidity on Crash Risk and the Role of Media Independence: International Evidence

• Date: 01 December 2018
• DOI: http://doi.org/10.1111/irfi.12162
• Published date: 01 December 2018

The Impact of Firm-Level

Illiquidity on Crash Risk and the

Role of Media Independence:

International Evidence*

ZHE AN

†

,WENLIAN GAO

‡

,DONGHUI LI

§

AND FEIFEI ZHU

¶

†

Monash Business School, Monash University, Melbourne, Australia

‡

Brennan School of Business, Dominican University, River Forest, IL

§

Management School, Jinan University, Guangzhou, China and

¶

College of Business and Economics, Radford University, Radford, VA

ABSTRACT

This study investigates the impact of firm-level illiquidity on stock price

crash risk by employing a sample of 21,986 firms across 36 countries span-

ning the years 1997 to 2007. In doing so, the role of media independence in

shaping this impact is also examined. The empirical results suggest that stock

illiquidity is significantly and positively associated with firm-level crash risk.

Furthermore, the positive association between illiquidity and crash risk has

been mitigated in countries with independent media, characterized by a

lower level of state ownership and a higher level of press freedom, in addi-

tion to the reduction effect of media independence per se on crash risk. Our

main conclusions remain valid after taking into account the endogeneity

issues and various robustness tests.

JEL Codes: G12; G15; G28

Accepted: 20 September 2017

I. INTRODUCTION

Stock price crashes are characterized as sudden and dramatic declines in stock

prices in the absence of significant news on the fundamentals. Survey studies

suggest that crashes are ranked as the first-order concern for investors (Olsen

1997; Koonce et al. 2005). Motivated by the model of agency conflicts in Jin

and Myers (2006), recent studies that identify the sources of stock price crash

risk primarily focus on information asymmetry between insiders and outsiders

that are captured by internal factors, such as various accounting

* We thank two anonymousreferees and conference participantsat 2015 FMA European conference,

as well as seminar participants at the Central University of Finance and Economics, and Sun Yat-Sen

University. All errors remain ours. E-mail: zhe.an@monash.edu (Z. An); wgao@dom.edu (W. Gao);

lidonghui@jnu.edu.cn (D. Li); fzhu2@radford.edu (F. Zhu).All correspondence to Donghui Li.

© 2017 International Review of Finance Ltd. 2017

International Review of Finance, 18:4, 2018: pp. 547–593

DOI: 10.1111/irfi.12162

characteristics.

1

However, the role of market microstructure-based information

asymmetry has been largely ignored. This is surprising as “market microstruc-

ture measures of information asymmetry are designed to capture adverse selec-

tion between a larger category of agents (informed traders) and the rest of the

market (uninformed traders) ...... Therefore, market microstructure measures of

information asymmetry are (imperfect) proxies for the financial markets’per-

ception of the information advantage held by firm insiders and the resulting

adverse selection costs”(Bharath et al. 2009, p. 3213). Thus, the information

advantage held by firm insiders prevents the full incorporation of information

in stock prices and should contribute to the occurrence of firm-specific crash

risk. This paper attempts to fill the gap by investigating the impact of market

microstructure-based information asymmetry on the crash risk of individ-

ual firms.

Drawn on the existing literature, we employ various illiquidity measures to

proxy for market-based information asymmetry between insiders and outsiders

(Amihud 2002; Bharath et al. 2009; Armstrong et al. 2010). Illiquidity can slow

the price discovery process and thus slow price adjustments (Bali et al. 2014).

As a consequence, the lack of full incorporation of information in stock prices

implies that the stock prices are distorted and biased. If the negative informa-

tion, as a subset of the full information set, is deliberately hidden from the mar-

ket by firm insiders, it is not easily detected in a market with limited investor

attention and illiquidity (Peng and Xiong 2006; Bali et al. 2014). When the

accumulation of negative information is beyond a certain threshold, the bubble

of overvalued stock prices will burst and a sudden stock price crash will happen.

It has been suggested that illiquidity plays a significant role in causing crashes

in many international financial markets (CGFS 1999), evidenced by the delayed

and costly capital movements even in the presence of buying opportunities

post a crash. As cited in both Pastor and Stambaugh (2003) and Huang and

Wang (2009), a vivid example of Long-Term Capital Management in 1998

showed the influence of illiquidity on stock price crashes. This highly levered

hedge fund suffered from the liquidity deterioration due to the Russian debt cri-

sis. It had to go through liquidation to meet margin calls, which further low-

ered its portfolio value with high liquidity sensitivity. As argued by Grossman

and Miller (1988), the limited liquidity availability associated with large trading

pressure could lead to significant price shifts. Huang and Wang (2009) modeled

that, when market presence is costly, the need for liquidity could arise endoge-

nously due to infrequent trading and cause large selling pressure and price

drops.

The second question to be investigated by this study is to identify what fac-

tors contribute to the differences in the positive association between illiquidity

1 See Jin and Myers (2006) for firm opaqueness; Hutton et al. (2009) for earnings management;

Kim et al. (2011a) for corporate tax avoidance; Kim et al. (2011b) for CFO option sensitivity;

Callen and Fang (2013) for institutional ownership; DeFond et al. (2014) for mandatory Inter-

national Financial Reporting Standards (IFRS) adoption; Andreou et al. (2016) for ownership

structure and CEO incentives; Kim and Zhang (2016) for conservatism in financial reporting.

© 2017 International Review of Finance Ltd. 2017548

International Review of Finance

and crash risks across countries. It has been well documented that institutional

infrastructures vary significantly across countries and their importance has been

highlighted in asset pricing and firm valuation. Therefore, we conjecture that

the effect of stock illiquidity on firm-specific crash risk could be shaped by insti-

tutional features. Specifically, we propose that media independence can play a

significant role in explaining the variations in the association between illiquid-

ity and firm-level crash risk across countries. Independent media without gov-

ernment involvement is probably one of the most effective institutions to

ensure timely release of unbiased and accurate information to the public and

uncover corporate wrongdoings.

2

Bushman and Smith (2001) and Bushman

et al. (2004) argued that firm information flow is impeded without a well-

developed communication infrastructure. Some legendary episodes such as the

exposure of Enron’s accounting irregularities and WorldCom accounting fraud

buttress the view that the press can make a difference. Therefore, well-

developed media channels should be able to help reduce information costs, the

market friction identified in Merton (1987)’s pricing model, and promote mar-

ket efficiency. According to Huang and Wang (2009), with reduced information

costs, investors’market presence will be less costly and the endogenously arisen

liquidity needs will be lowered. As a result, the occurrence of stock price

crashes, which arises from order imbalances in the form of excess supply, will

be less likely.

Accordingly, our empirical tests proceed in the following two steps. We first

investigate the impact of illiquidity on individual stock’s crash risk on our wide

sample of international stocks. We adopt yearly quoted relative spread, effective

spread, Amihud’s illiquidity ratio, and zero-return proportion, together with

their principle components as our primary illiquidity measures. Results show

that all these four illiquidity measures as well as their principle components are

significantly and positively associated with future crash risk. This finding is

robust to the use of various crash-risk measures, that is, negative conditional

return skewness, down-to-up volatility, and the likelihood of crashes. We

attempt several ways to mitigate the concerns over simultaneous determination

and reverse causality between stock illiquidity and crash risk, including employ-

ing an exogenous event—the passage of IFRS in different countries and two-

stage least squares (2SLS) regressions. Our results continue to hold after addres-

sing the endogeneity issue. We also perform a few robustness tests, taking

account the Asian Financial Crisis (AFC), American Depositary Receipts (ADRs),

different time periods, and the exclusion of dominant countries. The main con-

clusion remains valid.

We next turn to examine how the positive association between stock illi-

quidity and firm-specific crash risk is affected by media independence across

2 For example, Chen et al. (2013) found that, by discovering and disseminating information,

mass media facilitates public monitoring and enhance audit quality. Dyck et al. (2010) docu-

mented that, due to reputational incentives, journalists play a key role in bringing the fraud

to light.

© 2017 International Review of Finance Ltd. 2017 549

Firm-Level Illiquidity and Crash Risk