Political Incentives to Suppress Negative Information: Evidence from Chinese Listed Firms

• DOI: http://doi.org/10.1111/1475-679X.12071
• Date: 01 May 2015
• Published date: 01 May 2015

DOI: 10.1111/1475-679X.12071

Journal of Accounting Research

Vol. 53 No. 2 May 2015

Printed in U.S.A.

Political Incentives to Suppress

Negative Information: Evidence

from Chinese Listed Firms

JOSEPH D. PIOTROSKI,

∗T. J. WONG,

†AND TIANYU ZHANG

†

Received 4 January 2012; accepted 24 November 2014

ABSTRACT

This paper tests the proposition that politicians and their affiliated firms (i.e.,

firms operating in their province) temporarily suppress negative information

in response to political incentives. We examine the stock price behavior of

Chinese listed firms around two visible political events—meetings of the Na-

tional Congress of the Chinese Communist Party and promotions of high-

level provincial politicians—that are expected to asymmetrically increase the

costs of releasing bad news. The costs create an incentive for local politicians

and their affiliated firms to temporarily restrict the flow of negative informa-

tion about the companies. The result will be fewer stock price crashes for the

affiliated firms during these event windows, followed by an increase in crashes

after the event. Consistent with these predictions, we find that the affiliated

firms experience a reduction (an increase) in negative stock return skewness

before (after) the event. These effects are strongest in the three-month pe-

riod directly preceding the event, among firms that are more politically con-

nected, and when the province is dominated by faction politics and cronyism.

Additional tests document a significant reduction in published newspaper ar-

ticles about affected firms in advance of these political events, suggestive of a

∗Stanford University; †The Chinese University of Hong Kong.

Accepted by Christian Leuz. The authors thank the Research Grants Council of the Hong

Kong Special Administrative Region, China for a Competitive Earmarked Research Grant No.

CUHK 452708. The authors also thank Katherine Schipper, Ross Watts, and two anonymous

referees for helpful comments and suggestions. An online appendix to this paper can be down-

loaded at http://research.chicagobooth.edu/arc/journal/onlineappendices.aspx.

405

Copyright C, University of Chicago on behalf of the Accounting Research Center,2014

406 J.D.PIOTROSKI,T.J.WONG,AND T.ZHANG

link between our observed stock price behavior and temporary shifts in the

listed firms’ information environment.

JEL codes: G39; M41

Keywords: state ownership; information environment; corporate gover-

nance; political costs; China

1. Introduction

In this paper, we examine the proposition that politicians and their af-

filiated firms (i.e., firms operating in the politician’s province) suppress

negative information in response to incentives created by political events

that temporarily increase the cost of releasing bad news. Prior research ar-

gues that politicians and governments like to avoid negative news about

their activities (e.g., Peltzman [1976], Watts and Zimmerman [1986]). To

the extent that certain events asymmetrically increase the cost of releasing

bad news, local politicians will have a heightened incentive to temporarily

restrict the flow of negative information about firms under their control

or influence. Similarly, self-interested firms reliant on the support of local

politicians will also respond to the incentives created by the political event.

If these temporary incentives are sufficiently strong, the result will be a re-

duced flow of negative information in advance of the events, followed by

an increase in the flow of negative information after the event as previously

suppressed information is subsequently released. For publicly traded com-

panies, the temporary suppression of negative information will generate a

distinct shift in stock price crash behavior around the political event if the

market is unable to fully unravel the suppression of bad news.

We test these arguments by examining the stock price behavior of Chi-

nese listed firms around two highly anticipated political events—meetings

of the National Congress of the Chinese Communist Party (CCP) and pro-

motions of high-level provincial politicians—over the period 1993 to 2011.

We employ a stock price crash methodology specifically designed to capture

the suppression and subsequent release of value-relevant negative informa-

tion (e.g., Chen, Hong, and Stein [2001], Jin and Myers [2006]). We focus

on Chinese listed firms because their information environment is expected

to be sensitive to political incentives for opacity. In China, government

and party officials exert significant influence on listed firms through di-

rect ownership arrangements and/or indirect channels (e.g., bureaucracy,

regulation, licensing requirements, and informal political and social net-

works); this influence creates a conduit by which listed firms are expected

to align their behavior with the preferences of local politicians. We focus on

meetings of the National Congress and provincial-level promotions as these

two events temporarily increase the political costs of revealing bad news.

From a career concerns perspective, local politicians incur a personal cost

when firms under their control report poor performance, as adverse eco-

nomic news limits their advancement within the party structure (e.g., Chen,

POLITICAL INCENTIVES TO SUPPRESS NEGATIVE INFORMATION 407

Li, and Zhou [2005], Li and Zhou [2005]). More generally, authoritarian

governments like China need to demonstrate strengths to stay in power

(e.g., Tullock [1987]). The revelation of bad news about Chinese firms and

their political supervisors reveals weaknesses, disrupts the government’s

narrative, and impacts social harmony. In both cases, the revelation of bad

news is very costly to both local politicians and their affiliated firms during

these event windows, generating a heightened incentive to suppress nega-

tive information. Finally, China’s institutional arrangements, such as weak

regulatory enforcement, limited private enforcement opportunities, con-

centrated ownership structures, and relationship-based contracting, may

be insufficient to support the necessary incentives for transparency during

these periods.

However, there exist countervailing forces capable of sheltering listed

firms from these transitory political pressures. First, the conversion of state

enterprises to joint stock companies and their subsequent listing on stock

exchanges expose state-controlled firms to western governance mecha-

nisms and market forces that demand transparency. Second, the Chinese

government also set up the State Asset Supervision and Administration

Commission to further decentralize control power from central and lo-

cal governments to the listed firms. Qian [1996] and Fan, Wong, and

Zhang [2013] argue that the Chinese government has established struc-

tural wedges between politicians and listed firms to credibly commit to

not intervene in these firms’ day-to-day operations. Third, suppression of

negative information is inherently difficult. Successful suppression requires

politicians to have control over an array of information dissemination and

production mechanisms that prevent self-interested market participants

and competing politicians from leaking negative information. Finally, the

market could interpret silence during political periods as bad news; this un-

raveling would eliminate the expected political benefits of suppression. To

the extent that these factors effectively mitigate the influence of local politi-

cians’ incentives for opacity, we should not observe Chinese listed firms

responding to transitory pressures to suppress bad news around political

events. More generally, we expect firms most (least) exposed to strong mar-

ket (political) forces, such as firms with weak political connections, non-

state-owned firms, or firms operating in market-oriented provinces or in

provinces without powerful local politicians, to be less sensitive to transi-

tory political incentives.

Consistent with the bad news suppression predictions, we find that af-

filiated listed firms are significantly less (more) likely to experience stock

price crashes in advance of (after) our two political events. Mirroring the

theoretical predictions of Chen, Hong, and Stein [2001], this intertempo-

ral pattern is consistent with the initial suppression, and subsequent release,

of adverse news in response to the heightened incentives of the local politi-

cians and affiliated firms. These results are robust to alternative empirical

specifications and various empirical measures of stock price crash behavior

and information flow (including stock price synchronicity). Our evidence