Relationship Networks and Earnings Informativeness: Evidence from Corruption Cases

• Published date: 01 September 2014
• Date: 01 September 2014
• Author: Feng Guan,Zengquan Li,Yong George Yang,Joseph P.H. Fan
• DOI: http://doi.org/10.1111/jbfa.12078

Journal of Business Finance & Accounting

Journal of Business Finance & Accounting, 41(7) & (8), 831–866, September/October 2014, 0306-686X

doi: 10.1111/jbfa.12078

Relationship Networks and Earnings

Informativeness: Evidence from

Corruption Cases

JOSEPH P.H. FAN,FENG GUAN,ZENGQUAN LIANDYONG GEORGE YANG∗

Abstract: The measurement difficulties arising from relationship-based business transactions

can result in accounting opacity. We test this hypothesis by exploiting a natural experiment.

Using a sample of firms that were networked with 45 high-level Chinese bureaucrats involved

in corruption scandals between 1996 and 2007, we examine the patterns in the earnings

informativeness of these firms before and after the exogenous break of the networks. We

predict that the costs and benefits of business-politics relationships, which are not measurable

by the current accounting systems, diminish the ability of accounting earnings to track a firm’s

economic performance. In turn, a break in a political relationship due to anti-corruption

enforcement reduces the measurement noise and improves the earnings informativeness. We

find that, relative to the matched control firms, there is indeed a significant increase in the

earnings informativeness of the networked firms following the public exposure of a scandal.

Robustness tests fail to show that the documented improvement in the earnings informativeness

is primarily due to systematic changes in the firms’ earnings management behavior or disclosure

policies.

Keywords: earnings informativeness, political networks, corruption, China

1. INTRODUCTION

Studies find that accounting numbers in emerging markets are less informative about

firms’ economic value than the numbers in developed economies (Ball et al., 2000,

∗The first author is from the Faculty of Business Administration and Institute of Economics and Finance,

The Chinese University of Hong Kong, Shatin, NT, Hong Kong. The second and third authors are from

the School of Accountancy, Shanghai University of Finance and Economics, GuoDing Rd. 777, YangPu

District, Shanghai, China 200433. The fourth author is from the School of Accountancy, The Chinese

University of Hong Kong. The authors appreciate valuable comments from Lawrence Brown, Ying Cao,

Mara Faccio, Xiaohong Liu, Chul Park, Joseph Piotroski, Gordon Richardson, the workshop participants at

the University of Hong Kong and 2009 AAA Annual Meeting (New York), and especially the Editor Martin

Walker and an anonymous referee. The authors acknowledge the financial support of the Research Grants

Council of the Hong Kong SAR government (CUHK442909). Zengquan Li acknowledges the funding

support of the Program for New Century Excellent Talents in University (No. NCET-12–0899), NSFC (No.

71372041), the MOE Project of the Key Research Institute of Humanities and Social Science in University

(No. 13JJD790019), the Program for Innovative Research Team of Shanghai University of Finance and

Economics, and the MOE Project of Humanities and Social Science (No.13YJA790057). Yong George Yang

acknowledges the funding support of the National Natural Science Foundation of China (No. 71272213).

(Paper received August 2012, revised version accepted April 2014).

Address for correspondence: Yong George Yang, School of Accountancy, The Chinese University of Hong

Kong, Shatin, NT,Hong Kong.

e-mail: yyong@cuhk.edu.hk

C

2014 John Wiley & Sons Ltd 831

832 FAN, GUAN, LI AND YANG

2003; Fan and Wong, 2002; Leuz et al., 2003; Bushman and Piotroski, 2006; Bushman

et al., 2006; Leuz and Oberholzer-Gee, 2006; DeFond et al., 2007). Poor corporate

governance and weak protection of investor rights are proposed as the main factors

leading to this accounting opacity. In environments with weak legal systems, corporate

insiders can distort accounting information to cover their expropriation of interests

from common investors.

However, the prevalence of relationship networks formed by familial, social and

political ties is an under-investigated factor that could also play an important role

in explaining accounting numbers’ lack of informativeness in emerging markets.

These networks provide the parties involved with trust and high-powered incentives

(Williamson, 1985) and help to enforce contracts (Klein and Leffer, 1981), thereby

constituting critical input of and adding significant value to the firm (Fisman, 2001;

Leuz and Oberholzer-Gee, 2006; Allen and Babus, 2009). Relationship networks pose

challenges to the current accounting systems. The ability of earnings to reflect changes

in firm value is particularly weak because the costs and benefits of these networks

cannot be properly measured.

Although difficult to accurately measure, political relationships affect a firm’s

growth potential and the economic value of its standard assets as well as the value

of the whole firm (Jenkins et al., 2009). For example, the sales and therefore

the value of a factory are very likely to depend on whether the factory is owned

by a politically connected entrepreneur who can secure future business from the

government. Nonetheless, the value of these relationships is typically not reflected in

the book value of intangibles or other assets in the accounting system, but is rather

manifested in the privileged granting of governmental contracts or in preferential

treatment by the government’s future policies. In addition, the costs incurred in

cultivating these relationships are normally immediately expensed, but the benefits

only materialize in the future, resulting in a mismatch between costs and benefits.

Moreover, the economic rent can be risky, because it depends on the stability of the

relationship and the political career of the connected bureaucrat. Using Indonesia as

a testing ground, Leuz and Oberholzer-Gee (2006) show that the profits associated

with a political relationship can be risky in regions with severe political strife. This

risk adds uncertainty to the rent obtained by the connected firm and, accordingly,

influences investors’ evaluation of the firm’s earnings persistence. For the above

reasons and according to the noise-in-signal model developed by Holthausen and

Verrecchia (1988) and Kothari (2001) that is empirically supported by Teoh and

Wong (1993) and Ou and Sepe (2002), we predict that firms connected to political

bureaucrats have a lower level of accounting informativeness than unconnected firms.

It is difficult to empirically disentangle the accounting effects of political networks,

due to, for example, the presence of severe endogeneity and the difficulty in estab-

lishing causality. We overcome these barriers by exploiting a natural experiment. We

collect information on 45 high-level Chinese government officials who were charged

with and punished for corruption and the managers of publicly traded firms that were

connected with these bureaucrats through previously existing relationships or outright

bribery. High-profile corruption cases are typically exposed during political strife or

for other reasons that have little to do with the business of the networked firms (Fan

et al., 2008). Enforcement against corruption effectively and unexpectedly breaks the

networks between the accused bureaucrats and the firms/managers. Our hypothesis

will be supported if we observe an improvement in the association between the

C

2014 John Wiley & Sons Ltd

RELATIONSHIP NETWORKS AND EARNINGS INFORMATIVENESS 833

accounting numbers and the firms’ values following enforcement and the subsequent

break of political networks.

Our treatment sample consists of two groups of firms. Bribing firms are firms

confirmed in the public press to have paid bribes to corrupt officials. These firms

represent the most conspicuous type of political connection and the sample is

relatively small. We increase the power of our statistical tests and generalize the

implications of our analysis by identifying the firms that are highly likely to have been

connected with the corrupt bureaucrats through a prior job-related relationship or

kinship, but were not directly linked with the corruption scandals in the public news.

We label these firms as related firms.

We examine the change in the earnings informativeness of the treatment firms

after their connections with the bureaucrats were broken by the exposure of the

bureaucrats’ corruption. We compare the treatment firms with matched firms that

are similar in various dimensions but were not connected to the exposed bureaucrats.

We measure the earnings informativeness using the annual-window earnings response

coefficient (ERC) (Lev and Zarowin, 1999; Ball et al., 2000; Fan and Wong, 2002;

Francis et al., 2005; Hanlon et al., 2008). Consistent with our hypothesis, we find that

the treatment firms’ ERCs significantly increase relative to their matched control firms

following the exposure of the corruption scandals. This result is robust to controlling

for factors that can affect the ERC, such as the firms’ general disclosure policies.

We argue that the measurement issue associated with relationship networks gener-

ates the noise affecting the earnings informativeness. However, it is possible that the

improvement of the earnings informativeness observed in our treatment firms is due

to a lower level of earnings management after the corruption exposure. For example,

when firms are connected with corrupt bureaucrats, they have greater incentives to

manage their earnings to hide their rent-seeking activities, the disclosure of which

may attract political and social scrutiny. We examine this possibility by comparing

the politically connected firms’ earnings management behavior before and after the

exposure of the corruption scandals. The empirical evidence indicates no significant

change in earnings management. Our main result is also robust to earnings measures

that are not likely to be susceptible to earnings management.

This study contributes to the networks and earnings informativeness literature.

Many studies show that networks facilitate an entrepreneur’s access to important

resources and markets (McMillan and Woodruff, 1999; Fisman, 2001; Johnson and

Mitton, 2003; Khwaja and Mian, 2005; Faccio, 2006; Khanna and Thomas, 2009;

Bunkanwanicha et al., 2009; also see Allen and Babus (2009) for a review of this body of

literature). Networks are particularly important in emerging markets, in which formal

institutions provide weak protection for business transactions. Even in developed

countries such as the United States, networks can affect corporate investment and

financing decisions (Cohen et al., 2007; Hochberg et al., 2007). However, the way

in which networks constrain the ability of accounting systems to measure economic

performance is only now attracting research attention.1We provide early evidence of

this constraint.

Our research complements a recent study by Chaney et al. (2011) that examines the

effect of political relationships on accounting properties in an international setting.

1 For example, Bae and Jeong (2007) find that the value-relevance of earnings is smaller for firms affiliated

with business groups in Korea.

C

2014 John Wiley & Sons Ltd