Accepted Manuscript

CEO media exposure, political connection and Chinese firms' stock price
synchronicity

Xiaoqing Li, Penghua Qiao, Lin Zhao

PII: S1059-0560(18)30381-2
DOI: 10.1016/j.iref.2018.08.006
Reference: REVECO 1664

To appear in: International Review of Economics and Finance

Received Date: 7 May 2018

Revised Date: 15 July 2018
Accepted Date: 13 August 2018

Please cite this article as: Li X., Qiao P. & Zhao L., CEO media exposure, political connection and
Chinese firms' stock price synchronicity, International Review of Economics and Finance (2018), doi:
10.1016/j.iref.2018.08.006.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

CEO media exposure, political connection and

Chinese firms’ stock price synchronicity

Xiaoqing Li
College of Economy and Management

PT
Hebei University of Technology
Tianjin, China
Tel: (86) 022-60435091

RI
Email: [email protected]

SC
Penghua Qiao
School of Business and Economics
Kunming University of Science and Technology

U
Kunming, China
Tel: (86) 13108709777
AN
Email: [email protected]
M

Lin Zhao
Department of Finance
D

Martha and Spencer Love School of Business

Elon University
TE

Elon, NC 27244
Tel: (336) 278-5963
E-mail: [email protected]
C EP
AC

______
The authors acknowledge the financial support by National Natural Science
Foundation of China (Grant No. 71702084), Humanity and Social Science Fondation
of the Ministry of Education of China (No. 16YJA630027 and 17YJC630112). Qiao is
the corresponding author.
 ACCEPTED MANUSCRIPT

CEO media exposure, political connection and

Chinese firms’ stock price synchronicity

ABSTRACT

PT
This paper investigates the impact of CEO media exposure and CEO political connection

RI
on stock price synchronicity using weekly data of Chinese firms from 2007-2016. We

SC
also examine the influence of three institutional factors – the firm’s ownership structure,

its financial and legal environments – on the interactions among them. Our results

U
suggest that CEO media exposure can enhance firm-specific information into stock prices,
AN
reducing stock price synchronicity, while the CEO political connection acts as a
M

counteracting force. The institutional factors appear to have significant influence on the

identified interaction among a CEO’s media exposure, his political connections, and the
D

company’s stock price synchronicity.

TE

Key words: Chinese stock market, CEO media coverage, political connection, stock price
EP

synchronicity
JEL: G30
C
AC

1
 ACCEPTED MANUSCRIPT

CEO media exposure, political connection and

Chinese firms’ stock price synchronicity

PT
1. Introduction

As an efficient information dissemination mechanism, the news media collect and

RI
release information about the economy, industries, and firms. By disclosing relevant

SC
information and imposing public pressure on firm behaviors, the news media is also a

guardian that monitors firms and can thus provide protection to investors (Jin & Myers,

U
AN
2006). The participation of the media in the capital market is therefore an important

factor that affects investors’ decisions and firms’ stock prices (Chan & Hammed, 2006).
M

Prior studies have documented the impact of the media coverage on stock returns, the
D

trading intensity of stocks, and corporate governance. For example, Fang and Peress
TE

(2009) find that returns of stocks with no media coverage are higher than those with high

media coverage. Aman (2013) finds that the crash frequency of stock price increases with
EP

the extent of media coverage, especially when there is intensive broadcasting of firms’

financial information disclosure by the media. Additionally, excess media coverage

C
AC

appears to be closely related to stronger momentum in a firm’s stock (Hiller, Jacobs and

Mϋller, 2014). According to Engelberg and Parson (2011), the coverage by local paper of

earnings announcements by S&P 500 Index firms is a causal factor of local trading

activities. By looking into the impact of newspaper strikes, Peress (2014) provides more

empirical evidence of the causal impact of the media on the trading volume of stocks,
2
 ACCEPTED MANUSCRIPT

intraday volatility and predictability of stock prices. It is also found in the literature that

the media coverage plays a role in the improvement of corporate governance. Farrell and

Whidbee (2002), for example, document that the board of director is more likely to

PT
remove the CEO when a firm’s poor performance is subject to more scrunity by the

RI
financial press, such as the Wall Street Journal. Joe, Louis, and Robinson (2009) find that

the media exposure of the board ineffectiveness incentivize firms to take correction

SC
actions which lead to improvement in the quality of board and enhance shareholder

U
wealth. AN
In this study, we contribute to the literature by exploring how the stock price

synchronicity of Chinese firms are affected by the media exposure of a firm’s CEO and
M

his political connections. Synchronicity in stock prices depends on the extent to which
D

firm-specific information is accessible to the public (Kim, Yu & Zhang, 2016). Typical
TE

media coverage on companies involves media exposure of the firm’s top executives, such

as the CEO. When widely covered by the media, a CEO is more likely to take
EP

precautionary actions to reduce the firm risk. By allowing external investors to obtain
C

more firm-specific information, CEO media exposure can lead to more informative stock
AC

prices and hence reduce stock price synchronicity, which will consequently promote the

overall efficiency of the capital market.

Given the active government involvement in the Chinese market, a CEO’s political

connection is another important factor considered in our analysis. The political

3
 ACCEPTED MANUSCRIPT

connections of CEOs at firms enable companies to acquire more resources and receive

extra support from the government. A close connection with key regulatory agencies is

expected to shield firms from full exposure to litigation risk and other costs related to

PT
incomplete or inaccurate disclosure of financial information (Chaney, Faccio & Parsley,

RI
2011). Therefore, politically connected companies are more likely to have an advantage

over firms without such connections in terms of government support and the level of

SC
scrutiny from regulatory departments (Lin et al., 2016). Cheng and Leung (2016) find

U
that political connected Chinese firms show better financial performance and are less
AN
likely to experience management turnovers. Political connections are also found to

positively affect a firm’s cash dividends (Fung, Huang and Shen, 2014). Political
M

connections, on the other hand, also catalyze some problems that may harm firms’
D

investors. For example, political connections of CEOs are found to negatively affect the
TE

post-IPO stock return performance of Chinese firms (Fang, Wong and Zhang, 2007). Due

to the lack of effective supervision, politically connected companies are discouraged from
EP

improving the quality of their financial information disclosure (Chaney, Faccio & Parsley,
C

2011). In this situation, it is hard for external investors to have full information to analyze
AC

the firm’s true performance. The problem with information disclosure will be reflected in

stock prices that are expected to be more synchronous when the firm-specific information

is not fully perceived by investors. Therefore, while providing extra resources to firms, an

executive’s political connections also block the effective dissemination of firm-specific

4
 ACCEPTED MANUSCRIPT

information. Ultimately, this will induce the stock price to move in a more synchronous

manner. Political connections, in this sense, act as a counteracting force against the

improvement of capital market efficiency.

PT
CEO’s media exposure and his political connections likely influence the

RI
dissemination of firm-specific information to a larger extent. This is typically the case in

a less developed financial and legal environment and with more government

SC
interventions. In these economies, the increased information transparency from media

U
exposure and resources received by politically connected companies are especially
AN
important factors that may influence the synchronicity of stock prices and the consequent

impact on capital market efficiency. The relatively more complicated market environment
M

in these economies also provides us valuable opportunity to fully examine the

D

interactions among CEO media coverage, his political connections, and the stock price
TE

synchronicity of target firms.

In this study, we focus on the emerging Chinese market, which is now the second
EP

largest economy in the world per GDP. The empirical investigation of this research can
C

provide important information to investors who are seeking to diversify their portfolios in
AC

international financial markets. This information is also beneficial to policy makers in

other emerging economies in the development of their equity markets. Specifically, we

explore the following research questions in the context of the Chinese equity market. Our

main test is to examine the relationship between a CEO’s media exposure and his firm’s

5
 ACCEPTED MANUSCRIPT

stock price synchronicity. We also examine the relationship between a firm’s political

connections and its stock price synchronicity. We expect that for Chinese firms, more

news coverage of CEOs can facilitate the dissemination of more firm-specific

PT
information to investors, and thus reduce the synchronicity in stock prices. In contrast

RI
with the CEO’s media exposure, his political connections may disturb stock prices. They

may also deter the effective incorporation of firm-specific information, and thus lead to

SC
more synchronous stock prices. Consequently, we expect to find a significantly positive

U
relationship between a firm’s political connections and its stock price synchronicity. The
AN
suppressed firm-specific information is not well received by investors of politically

connected firms. As the CEO’s media exposure affects the synchronicity of the stock
M

price in one direction, his political connections are expected to affect the synchronicity in
D

the opposite direction. Thus, we take a step further and also explore the counteracting
TE

influence of political connections on firms with a CEO who receives any media coverage.

Besides the enormous size of its economy, the Chinese market also has some distinctive
EP

institutional characteristics that are different from many other developed economies.
C

Besides, we also explore the previous research questions by considering the impact of
AC

institutional factors such as the company’s ownership structure, the financial environment

and legal environments of sample firms. We are curious to see if the significance and

strength of interactions among CEO media exposure, political connections, and stock

price synchronicity for Chinese firms are universally valid or subject to certain

6
 ACCEPTED MANUSCRIPT

institutional environment factors.

Our study contributes to the literature in the following aspects. First, we investigate

impacts of CEO media exposure, political connection, and their interactions on the stock

PT
price synchronicity. Our empirical findings about these relationships enrich the breadth

RI
and depth of the theory in this field. Second, we employ a comprehensive dataset that

includes nearly all listed Chinese firms on the Shanghai Stock Exchange and the

SC
Shenzhen Stock Exchange, the two leading stock exchanges in mainland China. Our

U
sample spans a ten-year period from 2007 to 2016, which prevents our test results from
AN
being affected by temporary market fluctuations. Third, we conduct a thorough empirical

investigation on the interactions among the CEO media coverage, his political
M

connections and stock price synchronicity with various fixed effects models. To identify
D

the influence of specific institutional factors in the Chinese market on these interactions,
TE

we divide the full sample into sub-samples according to a firm’s ownership structure, the

quality of its financial environment, and the legal environment of the firm. The
EP

sub-sample analysis allows us to examine the main research questions under different
C

institutional conditions. Fourth, we further conduct robustness tests for our main
AC

empirical findings. The results from these robustness checks are consistent with our

previous test results, proving that our methodology is appropriate and valid. Finally, our

empirical findings provide more insight into the emerging Chinese market, which is now

among the top economies in the world and attracts world-wide investors. Such

7
 ACCEPTED MANUSCRIPT

information is important for investors diversifying their portfolios and also for policy

makers in other emerging economies when regulating their own equity markets.

Our main findings from the study are summarized as follows. First of all, we find a

PT
significantly negative relationship between a CEO’s media exposure and the price

RI
synchronicity of the firm’s stock, implying that news reports about CEOs disseminate

more firm-specific information to the capital market. Such information is incorporated

SC
into stock prices, as a result of investors’ trading activities. Second, our hypothesis

U
regarding the opposite role of the CEO’s political connections is also proven to be valid.
AN
In particular, the political connection variable is positively related to a firm’s stock price

synchronicity. This finding indicates that politically connected companies avoid

M

disclosing full financial information, which leads to a less informative stock price. Third,
D

the interaction term between a CEO’s media exposure and his political connections is
TE

positively related to stock price synchronicity, showing that the influence of the CEO’s

media exposure on the synchronicity of a firm’s stock price is counteracted by the

EP

existence of political connections. Fourth, the three institutional factors – a firm’s

C

ownership structure, its financial environment, and its legal environment – are proven to
AC

exert significant impact on the interactions among a CEO’s media coverage, his political

connections, and stock price synchronicity. In particular, the influence of CEO media

coverage is strengthened when the firm is not owned by the state, is operating in an

inferior financial environment, and in a weaker legal environment. On the other hand, we

8
 ACCEPTED MANUSCRIPT

find that state-owned firms, and firms that operate in inferior financial (and legal)

environments suffer more from the impact of political connections in terms of the

information content in stock prices. Additionally, the counteracting impact of political

PT
connections on the information-enhancing influence of CEO media coverage is

RI
significant only for non-state-owned enterprises, and when the financial and legal

environments are weaker.

SC
Findings from our empirical tests have the following implications. When considering

U
Chinese stocks for investment purposes, well-known firms with more media exposure are
AN
expected to be superior in terms of the alignment of stock price and a firm’s true

performance. Although the political connection of executives appears to be a detrimental

M

factor, its negative impact is largely alleviated for firms that operate in areas with a better
D

developed financial and legal environment.

TE

The remainder of the paper is organized as follows. We introduce the background and

develop hypotheses in Section 2. The data and methodology are described in Section 3.
EP

Empirical results are presented in Section 4. And Section 5 concludes the paper.
C

2. Background and Hypothesis Development

AC

2.1 CEO media exposure and stock price synchronicity

As the information hub of the capital market, the news media collect, analyze, and

report information on firms and the market (Kim, Yu, & Zhang, 2016). A higher level of

media exposure implies that more firm-specific information is known to the public

9
 ACCEPTED MANUSCRIPT

(Aman & Moriyasu, 2017), either voluntarily or involuntarily. These informational

activities function as an effective monitoring mechanism over firms’ behavior and affect

market participants’ investment decisions and firms’ stock prices (Qiao, Fung & Wang,

PT
2018). The media coverage on companies serves as an important information source for

RI
investors, especially small investors that are searching for relevant information in the

financial market (Aman, 2013).

SC
Media coverage also serves an important watchdog function, such as identifying

U
accounting fraud (Miller, 2006; Dyck, Morse & Zingles, 2010). More disclosure of
AN
firm-specific information from the media coverage can improve investor protection

through facilitating law/regulatory enforcements and imposing public pressure on related

M

firms. Better investor protection improves the financial market environment of the firm
D

and makes firms become more transparent to investors (Leuz, Nanda & Wysocki, 2003).
TE

In this case, investors can make more investing decisions based on firm-specific

information (Barber & Odean, 2008; Wu & Lin, 2017), which in turn decrease the stock
EP

price synchronicity. Prior studies also provide evidence that media coverage can affect
C

the pricing dynamics of firms’ stocks. For example, Hiller, Jacobs & Mϋller (2014) show
AC

that firms with media coverage exhibit strong stock price momentum. Additionally, Kim,

Yu, & Zhang (2016) and Morck, Yeung & Yu (2000) find that the extent to which stocks

move together depends on the relative amounts of firm-specific information capitalized

into stock prices. Hence, we propose the first hypothesis:

10
 ACCEPTED MANUSCRIPT

H1: CEO media exposure will reduce stock price synchronicity.

2.2 CEO political connection and stock price synchronicity

Political connections of executives play an important role in firms’ behaviors and

PT
economic activities (Lin et al., 2016). Such impact is more obvious in emerging markets

RI
where there are less developed legal systems, more government interventions, and

SC
weaker investor protection (Chen et al., 2017; Berkman, Cole & Fu, 2010). As a

double-edged sword, political connections can help corporations gain various benefits,

U
such as favorable bank lending (Claessens, Feijen & Laeven, 2008), lighter taxation and
AN
relaxed regulatory requirements (Faccio, 2006). On the other hand, it also has some

negative effects on firms. Prior research finds that political connections can significantly
M

decrease the quality of accounting information (Chaney, Faccio & Parsley, 2011).
D

Additionally, political connections distort the true performance of the firm by engaging in
TE

more earnings management activities (Chen, Pantzalis & Park, 2013).

As firms enjoy the resources and advantages brought by political connections, it is

EP

also more likely for them to hide information from the market, especially firm-specific
C

information that is quite valuable to investors (Liu et al., 2017). In particular, political
AC

connections can increase the stock price synchronicity in the following ways. First,

politically connected companies may acquire government’s extra support by rent-seeking

and underground trade. Although such transactions have to be presented in the annual

report in some ways, firm executives and related government officers will try to hide such

11
 ACCEPTED MANUSCRIPT

information from the public. Compared with weak political connections, firms with

strong political connections may be less eager to hire high quality auditors and become

less transparent (Liu et al., 2017). These activities decrease the quality of the firm’s

PT
disclosed information and therefore increase the stock price synchronicity. Second,

RI
politically connected companies are more acquainted with officials at government

agencies such as tax and securities regulation departments. A closer connection with the

SC
government can reduce the litigation risk and cost for the firm that arise from insufficient

U
public information, in view of the more lenient monitoring from regulators (Berkman et
AN
al., 2010). Thus, political connections discourage a firm from improving the quality of its

information disclosure, which in turn leads to increased stock price synchronicity.

M

In light of the above arguments, we hitherto propose that political connections

D

enable companies to hide firm-specific information from investors, leading to a higher

TE

level of stock price synchronicity. Additionally, the political connections act as a force

that counteracts the price synchronicity reduction effect of CEO media coverage. In other
EP

words, compared to firms without political connections, the firm-specific information in

C

stock prices for firms that have politically connected CEOs is not as great. Based on the
AC

foregoing discussion, our second hypothesis is as follows:

H2: CEO political connection will increase stock price synchronicity and counteracts

the reduction effect of CEO media exposure on synchronicity.

2.3 The impact of institutional environment on stock price synchronicity

12
 ACCEPTED MANUSCRIPT

According to institutional economics theory and contingency perspective (Aguilera,

et al.,2008; Williamson, 2000), the level of CEO media exposure and political

connection’s impact on stock price synchronicity is subject to different institutional

PT
environments. We will consider the influence of institutional environments in the

RI
following aspects: ownership structure, financial environment and legal environment.

First, as an important factor in the internal governance system of a company,

SC
ownership structure has significant influence on stock price synchronicity (Qian, Sun, &

U
Yu, 2018). Due to their special equity arrangement and the natural political connection
AN
with the government, Chinese state-owned enterprises (SOEs) are subject to less

supervision by regulators and thus discouraged from providing high quality financial
M

information. Consequently, we propose that compared to non-state-owned enterprises

D

(non-SOEs), the influence of CEO media exposure on stock price synchronicity is weaker,
TE

whereas the impact of political connection on the stock price synchronicity is stronger in

SOEs. Acknowledging the role of political connection as an informal mechanism that

EP

affects firms’ resources, non-SOEs in China seem to have an innate institutional

C

disadvantage because of their lack of political connections. For non-SOEs, obtaining

AC

shelter from government and regulators is more vital in enhancing their enterprise image

and legitimacy (Liu et al., 2017). Therefore, the proposed counteracting effect of political

connection on CEO media exposure’s influence over stock price synchronicity as in H2,

13
 ACCEPTED MANUSCRIPT

is greater for non-SOEs than SOEs. Our hypothesis regarding the firm’s ownership

structure is as follows:

H3a: CEO media exposure has a stronger negative effect on the stock price

PT
synchronicity in SOEs than non-SOEs; CEO political connection of SOEs has a

RI
stronger positive effect on the stock price synchronicity than non-SOEs.

Second, economic development and financial environment affect the information

SC
quality of disclosure by firms (Zhu et al., 2017; Bushman, Piotroski & Smith, 2004),

U
which in turn influences the stock price synchronicity. The financial environment impacts
AN
both the extent to which information is capitalized into stock prices, and the nature of

information. A favorable financial environment can enhance transparency of firms, while

M

a disadvantageous financial environment discourages firms from releasing more

D

firm-specific information because of weak local government and media scrutiny. For
TE

example, Morck, Yeung & Yu (2000) show that stock prices in economies with high GDP

move in a relatively unsynchronized manner, while those in low GDP economies tend to
EP

move up or down together. Therefore, we propose that the influence of CEO media
C

exposure and political connection on stock price synchronicity is more significant in

AC

worse financial environments than in better financial environments. Our hypothesis

regarding the financial environment is as follows:

H3b: CEO media exposure and political connection have a greater influence on the

stock price synchronicity in less developed financial environments than in more

14
 ACCEPTED MANUSCRIPT

developed financial environments. And the interaction of CEO political connection

and media exposure on the stock price synchronicity is also more significant.

Third, the degree of investors’ protection and law enforcement effectiveness can

PT
affect the quality of financial information (Wang & Yu, 2015; Leuz, Nanda & Wysocki,

RI
2003). Agency problems in regions with poor legal environments are severe, and the

standard governance institutions to protect investors and maintain transparency are

SC
limited and weak (Allen, Qian & Qian, 2005). Political connections hinder information

U
transparency, so poor investor protection increases the cost and risk of collecting
AN
firm-specific information for investors. Thus, less information transparency leads to less

informative stock prices (Morck, Yeung & Yu, 2000), and more synchronous movement
M

in prices are thus expected. As an important informal supervision system in regions with
D

weak legal systems, media exposure can play the role of constraining a firm’s irregular
TE

activities (Qiao, Fung & Wei, 2018; Griffin, Hirschey & Kelly, 2011). Compared to

regions with a better developed legal system, the function of media is more important in
EP

regions with a weaker legal system, given media reports can accelerate the information
C

diffusion from managers to external investors (Zhu et al., 2017). Therefore, in these
AC

regions, the impact of CEO media exposure on stock price synchronicity is expected to

be greater. Meanwhile, firms with political connections in these regions tend to have

more power to control and disrupt the media. This will abate the influence of CEO media

15
 ACCEPTED MANUSCRIPT

exposure on stock price synchronicity to a larger extent than firms in regions with a better

legal system. Our hypothesis regarding the legal environment is as follows:

H3c: CEO media exposure and political connection have greater influence on stock

PT
price synchronicity in a weaker legal environment than a stronger one. Meanwhile,

RI
the interaction of CEO political connection and media exposure on the stock price

synchronicity is also stronger.

SC
3. Data and methodology

U
Our initial dataset includes all Chinese firms listed on the Shanghai and Shenzhen
AN
Stock Exchanges over the period of 2007-2016. Our analysis excludes financial and

public utility firms to avoid the impact of their unique firm behavior. Firms listed for less
M

than one year are also excluded. To minimize the impact of outliers, we follow common

practice in the literature and winsorize all variables at the 1st and 99th percentiles (i.e.,
D

Balatbat, Taylor & Walter, 2004; Giroud & Mueller, 2010; Nguyen, Locke & Reddy,
TE

2015; Qiao et al., 2017). Our final sample consists of 10,571 firm-year observations.

Weekly stock trading data and annual financial data are obtained from the Chinese
EP

Securities Market and Accounting Research (CSMAR) database to examine stock price
C

synchronicity. The CEO media exposure information is extracted from Baidu, which is
AC

the leading online search engine in China. Following the method in Nguyen (2015), we

use the CEO’s name and his firm’s name to quantify CEO media exposure. We wrote a

Python-based application to collect the number of news reports on CEOs from the Baidu

news website (http://www.news.baidu.com).

We follow Morck, Yeung & Yu (2000) and Piotroski & Ronlstone (2004) to compute

16
 ACCEPTED MANUSCRIPT

a firm’s stock price synchronicity in a two-step procedure. In the first step, we estimate

firm-specific weekly return for firm i in industry k in week t from the expanded market

model regression in Eq (1):

= + ∙ + ∙ + 1

PT
, , , , ,

where , , is the return of stock i in the industry k in week t; , is the average

RI
rate of market value-weighted market return in week t, which is calculated as the weekly

SC
tradable market value-weighted returns of all A-share stocks on the Shanghai and

Shenzhen Stock Exchanges; and , is the industry return in week t, which is calculated

U
as the tradable market value-weighted industry index, excluding firm i.
AN
In the second step, since R2 is highly skewed and bounded between unit and zero,
M

we measure the stock price synchronicity (Synch) as the natural logarithmic

transformation of the world market model in Eq (2).

D
TE

ℎ= 2
1−

Specifically, R2 is the fraction of variation in firm i’s returns explained by market

EP

and industry factors. This R2-based inefficiency measure has been used in various
C

empirical studies on corporate investment and emerging markets (e.g., Durnev, Morck &
AC

Yeung, 2004; Chen, Goldstein & Jiang, 2006; Cheng, Leung, & Yu, 2014). A higher value

of Synch indicates that less firm-specific information is incorporated into stock returns

(e.g., Morck, Yeung & Yu, 2000), and the stock price is more synchronized.

We propose the following model to estimate the relationship between CEO media

exposure and political connection on stock price synchronicity.

17
 ACCEPTED MANUSCRIPT

ℎ = + !"#$%&' + (" ") + * !"#$%&' ∗ (" ")

+, - ./ 0 + 3

PT
where CEOmediat and POLCONt denote the CEO media exposure and political

connection, respectively; and - ./ 0 represents control variables. Table 1

RI
summarizes all the variables in Eq (3).

SC
[Insert Table 1 about here]

U
CEOmediat measures the extent of CEO media exposure. Following prior studies,
AN
CEOmediat is computed using the natural log of one plus the number of CEO news

reports (i.e., Qiao, Fung & Wei, 2018; Fang & Peress, 2009). Similar to Fan, Rui & Zhao
M

(2008), political connection (POLCON) is measured by a dummy variable. Specifically,

POLCON equals 1 if the CEO of a particular firm is a member of the People’s Congress
D

(i.e., the national legislature of China) or the Chinese People’s Political Consultative
TE

Conference (i.e., the political advisory body), and 0 otherwise.

Following prior studies, we include a set of control variables that are expected to
EP

affect stock price synchronicity (i.e., Kim, Yu & Zhang, 2016; Chan & Hameed, 2006;
C

Boehmer & Kelley, 2009). In particular, the selected variables control for firm
AC

characteristics (Firm size, Lev, ROE, B/M, Inholding, First, Turnover), industry

characteristics (Ind_growth, HHI), and CEO attributes (Duality, Degree, Age, Gender).

First, we control for firm characteristics with the following variables: Firm size, Lev,

ROE, B/M, Inholding, First, and Turnover. Firm size is calculated as the natural

logarithm of the total assets at year end. Lev measures the financial leverage of the firm

18
 ACCEPTED MANUSCRIPT

and is defined as the ratio of total liabilities to total assets. The firm’s accounting

performance is measured by return on equity (ROE), which is the ratio of net income to

shareholders’ equity. B/M is book value to market value ratio, which reflects investor

perceptions of the firm’s current and future stock performance. Institutional investors

PT
have a great incentive to monitor managers and therefore influence stock price behavior

RI
(An & Zhang, 2013). We control for the institutional shareholdings (Inholding) with the

ratio of total institutional holdings to a firm’s total number of shares outstanding. The

SC
variable First represents the share proportion of the largest shareholder. We also control

for the de-trended share turnover (Turnover), calculated as the average monthly share

U
turnover in year t minus the average monthly share turnover in year t-1. Firms that
AN
receive high-quality audits are expected to have a lower synchronicity in stock prices, so

we use the dummy variable Big_4 to control for the auditing quality. It takes the value 1
M

if a firm uses a Big Four accounting firm and 0 otherwise.

D

Second, we control for industry characteristics with Ind_growth and HHI. The
TE

variable Ind_growth measures the industry growth rate, and the variable HHI is an

indicator of the industry concentration. Specifically, HHI is calculated as ∑34 , where

EP

is the proportion of sales by firm i in the industry, and n is the number of firms in the

same industry. HHI reflects the industrial environment of a company which may affect
C

the firm’s stock price changes.

AC

Third, CEOs attributes may also affect stock price synchronicity. We control for

CEO attributes variables including: Duality, Degree, Age, and Gender. The CEO duality

(Duality) is a dummy variable which takes the value 1 when the CEO is also the

chairman of the board and 0 otherwise. The variable of CEO degree (Degree) is equal to

19
 ACCEPTED MANUSCRIPT

2 if the CEO has an associate’s degree, 3 for a university degree, 4 for a master’s degree,

5 for a doctoral degree, and 1 otherwise. The CEO age (Age) is a time-variant variable

measured as the age of the CEO for each firm-year observation. CEO gender (Gender) is

a dummy variable and takes the value 1 when the CEO is male and 0 if female.

PT
Finally, industry (Industry) and year (Year) dummies are included in Eq (3) to

RI
account for the industry effect and time variations.

4. Empirical findings

SC
4.1 Descriptive statistics and univariate analysis

Table 2 reports descriptive statistics on the three key variables: stock return

U
synchronicity, CEO media exposure, and political connection. Panel A presents the
AN
sample distribution by year during 2007-2016. A few observations arise here. First, Synch
M

has a negative mean value in most of the sample years, except for 2008. We also notice

that its mean and standard deviation vary considerably over the years. Second, the mean
D

of CEO media increases from 1.830 in 2007 to 3.103 in 2012, and it stays around 3.100
TE

during 2013-2016. Third, among the 10,571 observations, about 1,617 (15.3%) firms

have a politically connected CEO. The number of sample firms with political connections
EP

has increased from 13.3% in 2007 to nearly 21.0% in 2016.

C

[Insert Table 2 about here.]

AC

Panel B presents the sample distribution by industry of firms with regard to stock

return synchronicity, CEO media exposure, and political connection. The industry

classification is based on specifications of the China Securities Regulatory Commission

(CSRC). The sample is divided into 21 industries, where nonmanufacturing industries

20
 ACCEPTED MANUSCRIPT

have a one-digit code and manufacturing industries have a two-digit code.1 As shown in

Panel B, Synch is relatively higher for firms in Conglomerate (-0.167), Paper making,

printing (-0.290), and Textiles, apparel (-0.297) sectors, and it is the lowest for companies

in the Communications industry (-0.800). The CEO media coverage has highest mean

PT
values for firms in Other manufacturing (3.606), Electronics (3.327), and

RI
Communications (3.290) industries. Additionally, firms in Agriculture (0.256), and

Textiles, apparel (0.254) industries have more political connections than others.

SC
Panel A of Table 3 summarizes the descriptive statistics of all variables in Eq (3).

U
Several points are worth noting here. First, the stock price synchronicity shows
AN
substantial differences across firms. There are distinct variations among percentiles, with

the minimum and maximum values being -9.345 and 7.737, respectively. Second, we find
M

that most CEOs in the sample have a certain level of media exposure. The mean value for
D

CEO media is 2.937, indicating that there are around 18 news reports for an average CEO
TE

in the sample. And more than 75% of CEOs have media coverage greater than 1.792,

which is equivalent to around five relevant news reports during 2007-2016. Third, only a
EP

small portion of CEOs in sample firms have political connections. The dummy variable
C

POLCON has a mean value of only 0.153 and still has a value of 0 up until the 75th
AC

percentile. Fourth, control variables for firm characteristics exhibit substantial variations.

For example, the firm’s leverage variable (Lev) has highest and lowest level of 98% and

4.6%, respectively. Similarly, return on equity (ROE) ranges from -0.787 to 0.565 among

1
China’s economy is dominated by manufacturing. In our sample, the manufacturing sector comprises
almost 60.39% of the firms. We use a two-digit code to identify the manufacturing industries in this study
to control for the sector effects.
21
 ACCEPTED MANUSCRIPT

sample firms.

Panel B presents the correlations for CEO media, political connection, stock price

synchronicity and other variables. As expected, CEO media is negatively and

PT
significantly related to SYNCH. Political connection is positively and significantly

RI
related to SYNCH. These results are consistent with H1 and H2. Moreover, there is not

strong bivariate correlation between the CEO media, political connection, and other

SC
variables. The average variance factors (VIF) for all regression models are smaller than

U
the cut-off point of 10, indicating no significant multicollinearity issue in subsequent
AN
regression analyses.
M

[Insert Table 3 about here.]

D

To briefly examine the stock price synchronicity under different firm conditions and

institutional environments, we conduct a univariate analysis on this variable and present

TE

the results in Table 4. The main observations are summarized as follows. First, the stocks
EP

of firms with a CEO that has a higher (lower) level of media exposure show lower

(higher) synchronicity in their prices. The difference is statistically significant at the 1%

C

level, indicating that CEO media exposure is negatively correlated with stock price
AC

synchronicity. Second, firms without political connections have lower stock price

synchronicity (-0.477) than firms with a certain level of political connections (-0.313).

Again, the difference is statistically significant at conventional levels, implying that

political connections of CEOs are positively related to stock price synchronicity. Third,

the three institutional environment factors also exhibit significant influence on stock price
22
 ACCEPTED MANUSCRIPT

synchronicity. In particular, firms that are SOEs, firms that operate in a disadvantageous

financial environment, and firms that are subject to a weaker legal environment show a

lower level of stock price synchronicity.

PT
[Insert Table 4 about here.]

RI
4.2 Regression analysis

To minimize the effect of outliers on our regression results, we winsorize all

SC
variables at the 1st and 99th percentiles. In addition, we select the fixed effect model

U
according to the relevant test statistics. Table 5 presents regression results about the
AN
effects of CEO media exposure and political connection on stock price synchronicity.2

[Insert Table 5 about here.]

M

Four models are used to test our hypothesis. Model 1 only shows the effect of
D

control variables on stock price synchronicity. The estimation results of Model 1 indicate
TE

that firm size, lev, ROE have a negative and significant effect on Synch. The
EP

book-to-market ratio (B/M) has a positive effect on Synch. Model 2 examines the direct

effect of CEO media coverage on stock price synchronicity, and it reveals that the stock
C

price of firms with more CEO media coverage exhibits lower synchronicity (Model 2: β =
AC

-0.037, t-value = -4.24). In Model 3, we test the direct effect of political connection on

stock price synchronicity. The estimated coefficient of POLCON is positive and

2
For our panel analysis, we use the Hausman Chi Square test to select the proper model between the fixed-effects model
or the random-effects model, a standard procedure in the literature. The test results indicate that the fixed-effects model
seems to perform better than the random-effects model.

23
 ACCEPTED MANUSCRIPT

significant at the 1% level (Model 3: β = 0.210, t-value = 4.84), showing that firms that

have CEOs with political connections have higher stock price synchronicity. Finally, we

use Model 4 to examine the impact of the interaction between CEO media coverage and

PT
political connection on stock price synchronicity. With the interaction term added, the

RI
estimated coefficients of CEO media and POLCON are not affected. The estimated

coefficient for the interaction term CEO media ∗ POLCON is significantly positive

SC
(Model 4: β = 0.037, t-value = 1.94), implying that the synchronicity reduction effect of

U
CEO media coverage on stock prices is eroded by political connections of the firms. The
AN
results in Table 5 are generally consistent with our first two hypotheses (H1 and H2).

4.3 The impact of institutional environment

M

4.3.1 Ownership structure

D

Ownership structure of firms in China differs from most developed countries. In

TE

particular, state-controlled firms are still very common in China. A natural question is

whether the particular ownership setting in China has an influence on our previous results
EP

(Feng, Hu & Johansson, 2016). Investigating the role of a firm’s ownership structure in
C

the interactions among CEO media coverage, political connection, and stock price
AC

synchronicity provide more insight into our research questions. Thus, we divide sample

firms into SOEs and non-SOEs. We then conduct the same test as in the previous section,

using Model 2, Model 3, and Model 4. The test results are shown in Table 6.

[Insert Table 6 about here.]

24
 ACCEPTED MANUSCRIPT

Three findings are noteworthy here. First, the negative relationship between CEO

media coverage and stock price synchronicity exists for both SOEs and non-SOEs.

Specifically, the scale of impact of CEO media coverage is larger for non-SOEs as shown

PT
by the estimated coefficients (i.e., -0.045 vs. -0.031, p-value for testing their difference <

RI
0.01). Second, the positive relationship between political connection and stock price

synchronicity is still valid for both SOEs and non-SOEs, while the impact of political

SC
connection on price synchronicity is greater for SOEs than for non-SOEs (i.e., 0.259 vs.

U
0.145, p-value for testing their difference < 0.01). Third, the interaction variable CEO
AN
media ∗ POLCON is still positively related to stock price synchronicity, although the

relationship is only statistically significant for non-SOEs. Thus, the previously observed
M

significant coefficient on the interaction term with the full sample in Table 5 is mainly
D

driven by non-SOEs. In summary, these results are consistent with our hypothesis of
TE

H3a.

4.3.2 Financial environment and legal environment

EP

Given that different administrative regions/provinces in China typically exhibit

C

large variations in their economic and institutional development, we need to identify

AC

proxies to address the differences in the institutional development level across the 31

economically diverse regions (Kim, Yu & Zhang, 2016). Following Wang, Fan & Yu

(2017), we analyze the impact of provincial financial environment and legal environment

of these different economic regions.

25
 ACCEPTED MANUSCRIPT

Table 7 presents the impact of CEO media exposure and political connection on

stock price synchronicity under different provincial financial environments. According to

the provincial financial development index, 3 we divide Chinese provincial financial

PT
environments into relatively “superior” and relatively “inferior” groups. Estimation

RI
results from Model 2, Model 3, and Model 4 for each group are reported in Table 7.

[Insert Table 7 about here.]

SC
Some interesting results have emerged in Table 7. First, the stock price

U
synchronicity reduction effect of the CEO media coverage is significant in both financial
AN
environments, and such influence is greater for firms operating in a relatively inferior

environment than for those in a superior environment (i.e., estimated coefficients are
M

-0.042 vs. -0.035, p-value for testing their difference < 0.01). Second, the effect of
D

political connections on stock price synchronicity is also significant in both groups. The
TE

positive impact is larger for firms operating in an inferior financial environment (i.e.,

estimated coefficients are 0.263 vs. 0.153, p-value for testing their difference < 0.01).
EP

Third, the interaction variable CEO media ∗ POLCON is positively related to stock price
C

synchronicity in both groups. However, this positive relationship is only statistically

AC

significant for firms in an inferior financial environment. Again, these results are

consistent with our hypothesis of H3b.

3
We obtain the data of provincial financial development index and provincial legal development
index from the “Marketization index of China’s provinces: NERI report 2017”, by Wang Xiaolu,
Fan Gang and Yu Wenjing.
26
 ACCEPTED MANUSCRIPT

The influence of different provincial legal environments is also tested and the results

are presented in Table 8. According to the provincial legal development index, we divide

the legal environment into relatively “stronger” and “weaker” groups. Similarly, for both

PT
groups, we still find a significantly negative relationship between CEO media coverage

RI
and stock price synchronicity, and a significantly positive relationship between political

connection and stock price synchronicity. These impacts are greater for firms in provinces

SC
with a relatively weaker legal environment. The interaction term CEO media ∗ POLCON

U
is significantly positive only for provinces with weaker legal environments. These results
AN
are consistent with our hypothesis H3c.

[Insert Table 8 about here.]

M

In summary, by considering the possible influence of the three institutional factors

D

(i.e., ownership structure, financial and legal environment), we have confirmed that the
TE

negative impact of CEO media coverage and positive impact of political connection on

stock price synchronicity is valid under all these situations. Nonetheless, the strength of
EP

these impacts has significant variation across different institutional conditions. For
C

example, we find that the influence of CEO media coverage on price synchronicity is
AC

stronger for firms that are non-SOEs, and for firms that operate in provinces with a

relatively inferior financial environment and weaker legal environment. The influence of

political connection on stock price synchronicity appears to be larger for firms that are

SOEs, and for those that operate in inferior financial and weaker legal environments.

27
 ACCEPTED MANUSCRIPT

Another interesting finding is that the significant positive impact of the interaction term

CEO media ∗ POLCON found in the full sample seems to be driven by firms that are

non-SOEs, and firms that operate in weaker financial (and legal) environments.

PT
4.4 Robustness check

RI
We further conduct additional tests for robustness. The results are shown in Table 9.

First, we use the fixed-effect model to examine the standard errors, which are clustered

SC
by effects of both firms and years. The results are reported in column (1). As expected,

U
the clustered standard errors become larger. The impact of CEO media coverage on stock
AN
price synchronicity remains negative at the 1% level and the effect of political connection

remains positive at the 1% level. Second, we use Driscoll & Kraay’s (1998) covariance
M

matrix estimator in the regression, and the results are reported in column (2). The effect
D

of CEO media coverage still remains significantly negative and that of political
TE

connection also remains significantly positive. The interaction variable CEO

media ∗ POLCON is positively related to stock price synchronicity at the 10% level.
EP

Third, for standard errors and confidence sets, we use 200 bootstrap repetitions and report
C

the results in column (3), which again shows similar findings. Finally, we attempt to
AC

mitigate the potential endogeneity problem of CEO media exposure and political

connection by including lagged media variables. That is, we used CEO mediat-1 and CEO

mediat-2 as proxies for the CEO media variable, and POLCON t-1 and POLCONt-2 as

proxies for the POLCON variable. The results estimated by the two-stage least squares

28
 ACCEPTED MANUSCRIPT

for panel-data models are shown in column (4). The finding is that lagged CEO media

coverage reduces stock price synchronicity at time t and lagged political connection

increases stock price synchronicity at time t, effects which are unlikely to be caused by

PT
endogeneity. Our method is valid as the data pass the Sargan-Hansen tests at the

RI
conventional significance levels in examining the over-identifying restrictions. All of the

above tests include industry and year dummies. Therefore, the negative effects of CEO

SC
media and the positive effects of POLCON on Synch are robust across different models,

U
which proves the validity of our main results.
AN
5. Conclusions

In this study, we examine with weekly data the impact of firm CEO’s media exposure
M

and political connections on the firm stock’s synchronicity. Our sample consists of
D

10,571 firm-year observations for Chinese companies listed on the Shanghai Stock
TE

Exchange and the Shenzhen Stock Exchange during the period of 2007-2016. We apply

fixed effects models to test whether the CEO’s media coverage and political connections
EP

influence the incorporation of firm-specific information into stock prices, which is

C

reflected in the level of synchronicity of the stock price. We further test the impact on
AC

stock price synchronicity of the interaction term between CEO media coverage and

synchronicity. Given the specific market environment in the Chinese economy, we

consider the above interactions under different institutional situations. In particular, we

investigate the potential effect of the firm’s ownership structure (i.e., state-owned vs.

29
 ACCEPTED MANUSCRIPT

non-state-owned), the quality of financial environment (superior vs. inferior), and the

legal environment (stronger vs. weaker). We also conduct a thorough robustness check

for additional validity of our methodology and main findings.

PT
In summary, the empirical test results are generally consistent with our hypotheses.

RI
First, CEO media coverage has a negative influence over stock price synchronicity,

implying that news reports facilitate the release of firm-specific information. Second,

SC
when the firm executives have a certain level of political connections, the synchronicity

U
of the stock price tends to increase. This suggests that firms are reluctant to disclose
AN
valuable firm-specific information when they have access to more resources from their

connections with government and regulatory agencies. Third, the interaction term of CEO
M

media coverage and political connection has a positive impact on stock price
D

synchronicity for the full sample data, which proves that if the firm is politically
TE

connected, the positive influence of CEO media coverage on the informational content of

stock price will be abated or even eliminated. Finally, we reexamine the above patterns
EP

under different institutional environments. Our sub-sample analysis reveals some

C

interesting findings. The institutional factors (i.e., firm ownership, financial and legal
AC

environments) indeed play an important role in the statistical significance and scale of the

above interactions. In particular, CEO media coverage has a greater impact on the

reduction of stock price synchronicity for firms that are not owned by the state, that

operate in inferior financial environments and weaker legal environments. The impact of

30
 ACCEPTED MANUSCRIPT

political connections also varies under different institutional environments. It appears that

the stock price synchronicity is exposed to a stronger influence of political connection

when firms are state-owned and operate in relatively poor financial and legal

PT
environments. Interestingly, the counteracting effect of political connections against the

RI
impact of CEO media coverage becomes insignificant for state-owned firms and firms

that operate in healthier financial and legal environments.

SC
Findings from this study suggest that, in addition to traditional fundamental factors of

U
a company, investors should also look at factors such as media exposure and political
AN
connections of a firm when making investment decisions. Additionally, institutional

factors also play an important role in the evaluation procedure. Therefore, our study
M

provides meaningful information for investors that seek to diversify their portfolios and
D

to those who are interested in China’s emerging market.

TE
C EP
AC

31
 ACCEPTED MANUSCRIPT

Reference

Aman, H. (2013). An analysis of the impact of media coverage on stock price crashes and

jumps: Evidence from Japan. Pacific-Basin Finance Journal, 24, 22-28.

PT
Aman, H. & Moriyasu, H. (2017). Volatility and public information flows: Evidence from

RI
disclosure and media coverage in the Japanese stock market. International Review of

SC
Economics & Finance, 51(9), 660-676.

U
Aguilera, R., Filatotchev, I., Gospel, H., and Jackson, G. (2008). An Organizational
AN
Approach to Comparative Corporate Governance: Costs, Contingencies, and

Complementarities. Organization Science, 19(3), 475-492.

M

Allen, F., Qian, J., & Qian, M., (2005). Law, finance, and economic growth in China.
D

Journal of Financial Economics, 77(1), 57–116.

TE

An H., & Zhang T. (2013). Stock price synchronicity, crash risk, and institutional
EP

investors, Journal of Corporate Finance, 21(6), 1-15.

C

Aman., H. (2013). An analysis of the impact of media coverage on stock price Crashes
AC

and Jumps: Evidence from Japan. Pacific-Basin Finance Journal, 24(9), 22–38.

Balatbat, M. C. A., Taylor, S. L., &Walter, T. S. (2004). Corporate governance, insider

ownership and operating performance of Australian initial public offerings.

Accounting and Finance, 44(3), 299–328.

32
 ACCEPTED MANUSCRIPT

Barber, B. M., & Odean, T. (2008). All that Glitters: The Effect of Attention and News on

the Buying Behavior of Individual and Institutional Investors. Review of Financial

Studies, 21(2), 785~818.

PT
Berkman, H., Cole R.A., & Fu, L. (2010). Political Connections and

RI
Minority-Shareholder Protection: Evidence from Securities-Market Regulation in

SC
China. Journal of Financial and Quantitative Analysis, 45(6), 139101417.

Boehmer, E., & Kelley, E. K. (2009). Institutional investors and the informational

U
efficiency of prices. The Review of Financial Studies, 22(9), 3563-3594.
AN
Bushman R., Piotroski J. & Smith A. (2004). What Determines Corporate Transparency.
M

Journal of Accounting Research, 42(2), 207-252.

D

Chan, K., & Hameed, A. (2006). Stock price synchronicity and analyst coverage in
TE

emerging markets. Journal of Financial Economics, 80(1), 115-147.

EP

Chaney P. K., Faccio M., & Parsley D. (2011). The Quality of Accounting Information in

Politically Connected Firms. Journal of Accounting and Economics, 51(1-2), 58-76.

C
AC

Chen, Q., Goldstein, I., & Jiang, W. (2006). Price informativeness and investment

sensitivity to stock price. The Review of Financial Studies, 20(3), 619-650.

Chen C. W., Pantzalis C., & Park J. C. (2013). Press Coverage and Stock Price Deviation

from Fundamental Value. Journal of Financial Research, 36(2),175-214.

33
 ACCEPTED MANUSCRIPT

Cheng, L.T.W., & Leung, T.Y. (2016). Governance protection, political connection and

management turnover in China. International Review of Economics and Finance, 45,

160-176.

PT
Cheng, L.T.W., Leung, T.Y., & Yu, W.(2014). Information arrival, changes in R-square

RI
and pricing asymmetry of corporate news. International Review of Economics &

SC
Finance, 33(9),67-81.

Claessens, S., Feijen, E., & Laeven, L. (2008). Political connections and preferential

U
access to finance: The role of campaign contributions. Journal of Financial
AN
Economics, 88(3), 554–580.
M

Driscoll, J. C., & Kraay, A. C. (1998). Consistent Covariance Matrix Estimation with
D

Spatially Dependent Panel Data. Review of Economics and Statistics, 80(4), 549-560.
TE

Durnev, A., Morck, R., & Yeung, B. (2004). Value‐enhancing capital budgeting and
EP

firm‐specific stock return variation. The Journal of Finance, 59(1), 65-105.

Dyck, A., Morse, A., & Zingles, L. (2010). Who blows the whistle on corporate fraud?
C
AC

Journal of Finance, 65(6), 2213–2254.

Engelberg, J.E., & Parsons, C.A. (2011). The causal impact of media in financial markets.

The Journal of Finance, 66(1), 67-97.

34
 ACCEPTED MANUSCRIPT

Faccio M. (2006). Politically Connected Firms. The American economic review, 96(1),

369-386.

Fan, J. P. H., Rui, O.M., & Zhao, M. (2008). Public governance and corporate finance:

PT
Evidence from corruption cases. Journal of Comparative Economics, 36(3), 343–364.

RI
Fan, J.P.H., Wong, T.J., & Zhang, T. (2007). Politically connected CEOs, corporate

SC
governance, and post-IPO performance of China’s newly partially privatized firms.

Journal of Financial Economics, 84(2), 330-357.

U
AN
Fang, L., & Peress, J. (2009). Media coverage and the cross-section of stock returns.

Journal of Finance, 64(5), 2023–2052.

M

Farrell, K.A., & Whidbee, D.A. (2002). Monitoring by the financial press and forced
D

CEO turnover. Journal of Banking & Finance, 26(12), 2249-2276.

TE

Feng, X., Hu, N., & Johansson, A. C. (2016). Ownership, analyst coverage, and stock
EP

synchronicity in China. International Review of Financial Analysis, 45, 79-96.

C

Giroud, X., & Mueller, H. M. (2010). Does corporate governance matter in competitive
AC

industries? Journal of Financial Economics, 95(3), 312–331.

Griffin J. M., Hirschey N. H., & Kelly P. J. (2011). How Important Is the Financial Media

in Global Markets? The review of financial studies. 24(12), 3941-3992.

35
 ACCEPTED MANUSCRIPT

Hillert, A., Jacobs, H., & Mϋller, S. (2014). Media Makes Momentum. The Review of

Financial Studies, 27(12), 3467–3501.

Jin, L., & Myers, S. C., (2006). R2 round the world: new theory and new tests. Journal of

PT
Financial and Economics, 79(2), 257–292.

RI
Joe, J.R., Louis, H., & Robinson, D. (2009). Managers’ and investors’ response to media

SC
exposure of board ineffectiveness. The Journal of Financial and Quantitative

Analysis, 44(3), 579-605.

U
AN
Kim, J. B., Yu, Z., & Zhang, H. (2016). Can media exposure improve stock price

efficiency in China and why? China Journal of Accounting Research, 9(2), 83-114.
M

Leuz, C., Nanda, D., & Wysocki, P. D. (2003). Earnings management and investor
D

protection: an international comparison. Journal of Finance and Economics, 69(3),

TE

505–527.
EP

Lin, C., Ho, P., Shen, C., & Wang, Y. (2016). Political connection, government policy,

and investor trading: Evidence from an emerging market. International Review of

C
AC

Economics and Finance, 42(3), 153-166.

Liu, Y., Li, X., Zeng, H., & An, Y. (2017). Political connections, auditor choice and

corporate accounting transparency: evidence from private sector firms in China.

Accounting & Finance, 57(4), 1071-1099.

36
 ACCEPTED MANUSCRIPT

Miller, G. S. (2006). The press as a watchdog for accounting fraud. Journal of Accounting

Research, 44(5), 1001-1033.

Morck R., Yeung B., & Yu W. (2000). The information content of stock markets: why do

PT
emerging markets have synchronous stock price movements? Journal of Finance and

RI
Economics, 58(1), 215-260.

SC
Nguyen, B. D. (2015). Is more news good news? Media coverage of CEOs, firm value,

and rent extraction. Quarterly Journal of Finance 5 (4), 1–38.

U
AN
Nguyen T., Locke S., Reddy K. (2015). Does boardroom gender diversity matter?

Evidence from a transitional economy. International Review of Economics and

M

Finance, 37(5), 184-202.

D

Piotroski, J. D., Roulstone, B. T. (2004). The influence of analysts, institutional investors,

TE

and insiders on the incorporation of market, industry, and firm-specific information

into stock prices. The Accounting Review, 79(4), 1119–1151.

EP

Peress, J. (2014). The media and diffusion of information in financial markets: Evidence
C

from newspaper strikes. The Journal of Finance, 69(5), 2007-2043.

AC

Qiao, P. H., Fung, A., Miao, J. C., & Fung, H. G. (2017). Powerful Chief Executive

Officers and Firm Performance: Integrating Agency and Stewardship Theory.

China & World Economy, 25,100-119.

37
 ACCEPTED MANUSCRIPT

Qiao, P.H., Fung, H.G., & Wei, W. (2018). Chinese firms’ crash risk and CEO media

exposure. Applied Economics Letters,25(14):1034-1037.

Qian, M.F., Sun, P., & Yu, B. (2018). Top Managerial Power and Stock Price Efficiency:

PT
Evidence from China. Pacific-Basin Finance Journal, 47(2), 20-38.

RI
Su, Z.-q., Fung, H.-G., Huang, D.-s., & Shen, C.-H. (2014). Cash dividends,

SC
expropriation, and political connections: Evidence from China. International Review

of Economics and Finance, 29, 260-272.

U
AN
Wang, X., Fan, G. & Yu, W. (2017). Marketization index of China’s provinces: NERI

report. Social Sciences Academic Press (CHINA)

M

Wang, J. W., & Yu, W. W. (2015). The Information Content of Stock Prices, Legal
D

Environments, and Accounting Standards: International Evidence. European

TE

Accounting Review, 24(3), 471-493.

EP

Williamson O. E. (2000). The new institutional economics: taking stock，looking ahead.

Journal of Economic Literature, 38(3), 595-613．

C
AC

38
 ACCEPTED MANUSCRIPT

Table 1. Description of variables

Variable Definition
Synch Stock price synchronicity
CEO media A CEO’s media coverage, measured by the natural log of one plus the
number of CEO news.

PT
A CEO’s political connection, measured by a dummy variable that equals
POLCON 1 when a CEO is the member of the People’s Congress or the Chinese
People’s Political Consultative Conference, and 0 else.

RI
Firm size Size of a firm, measured by the natural logarithm of total assets
Lev Financial leverage of a firm

SC
ROE Return on equity of a firm
B/M Ratio of book to market value of a firm
Inholding Institutional shareholdings, measured by the ratio of total institutional
holdings to firm’s total shares outstanding

U
Tunrover Difference of average monthly share turnover of year t and t-1
AN
First Share proportion of the largest shareholder
Big_4 A dummy variable that equals 1 if a firm uses a Big 4 international
auditor and 0 otherwise.
M

Ind_growth Industry growth rate

HHI Indicator of the industry concentration
Duality A dummy variable that takes 1 if the CEO is also the chairman of the
D

board and 0 otherwise.

Degree A categorical variable of a CEO’s education level
TE

Age Time-variant variable of the age of a CEO

Gender A dummy variable that equals 1 if the CEO is male and 0 if female
C EP
AC

39
 ACCEPTED MANUSCRIPT

Table 2. Descriptive statistics of Synch, CEO media, and POLCON

Panel A – Sample distribution by year

Synch CEO media POLCON
Year Obs.
Mean Std. Min Max Mean Std. Min Max Mean Std. Min Max
2007 444 -0.537 0.970 -6.001 1.360 1.830 1.255 0.000 6.807 0.133 0.340 0.000 1.000

PT
2008 585 0.522 0.671 -3.432 2.043 1.911 1.308 0.000 6.460 0.128 0.335 0.000 1.000
2009 715 -0.247 0.851 -6.021 6.205 2.210 1.450 0.000 6.967 0.130 0.337 0.000 1.000
2010 947 -0.525 0.939 -5.411 7.737 3.048 1.532 0.000 7.763 0.141 0.349 0.000 1.000

RI
2011 1,150 -0.395 0.675 -4.014 1.538 3.092 1.494 0.000 8.001 0.142 0.349 0.000 1.000
2012 1,304 -0.334 0.787 -6.954 1.553 3.103 1.517 0.000 9.776 0.142 0.349 0.000 1.000

SC
2013 1,323 -1.099 0.967 -8.429 1.034 3.102 1.581 0.000 8.314 0.130 0.336 0.000 1.000
2014 1,374 -1.230 1.002 -6.761 3.250 3.214 1.654 0.000 10.477 0.143 0.351 0.000 1.000
2015 1,364 -0.017 0.932 -6.957 1.908 3.070 1.663 0.000 10.609 0.185 0.388 0.000 1.000

U
2016 1,365 -0.124 0.995 -9.354 2.196 3.182 1.763 0.000 10.633 0.210 0.408 0.000 1.000
Total 10,571 -0.457 1.016 -9.354 7.737 2.937 1.625 0.000 10.633 0.153 0.360 0.000 1.000
AN
Note: This panel presents the sample distribution by year during 2007–2016.
M
D
TE
C EP
AC

40
 ACCEPTED MANUSCRIPT

Panel B – Sample distribution by industry

Industry Firms Synch CEO media POLCON
Industry name
code Obs. % Mean Std. Mean Std. Mean Std.
Agriculture A 195 1.84 -0.494 0.992 2.667 1.616 0.256 0.438
Mining B 190 1.80 -0.669 0.975 2.374 1.551 0.079 0.270
Food, beverage C0 424 4.01 -0.612 1.018 3.055 1.722 0.163 0.370

PT
Textiles, apparel C1 335 3.17 -0.297 1.026 2.941 1.503 0.254 0.436
Timber, furniture C2 68 0.64 -0.380 0.923 2.998 1.540 0.147 0.357

RI
Paper making, C3 185 1.75 -0.290 0.947 2.899 1.499 0.146 0.354
Petroleum, C4 1014 9.59 -0.349 0.971 2.796 1.466 0.178 0.382
Electronics C5 669 6.33 -0.501 1.074 3.327 1.586 0.169 0.375

SC
Metal, nonmetal C6 787 7.44 -0.356 0.978 2.954 1.481 0.113 0.317
Machinery, C7 2057 19.46 -0.414 0.991 3.115 1.634 0.191 0.393
Medicine, C8 748 7.08 -0.583 1.069 2.863 1.663 0.134 0.341

U
Other C9 97 0.92 -0.472 1.018 3.606 2.028 0.144 0.353
AN
Power, gas, and D 387 3.66 -0.520 1.019 2.332 1.197 0.114 0.318
Architecture E 210 1.99 -0.453 0.956 2.769 1.449 0.157 0.365
Transportation F 326 3.08 -0.351 0.924 2.672 1.455 0.101 0.302
M

IT G 1024 9.69 -0.589 1.101 3.271 1.790 0.132 0.338

Retail H 580 5.49 -0.351 0.931 2.508 1.726 0.112 0.316
Real estate J 520 4.92 -0.612 1.064 2.705 1.472 0.108 0.310
D

Social services K 409 3.87 -0.430 0.921 2.926 1.647 0.134 0.342
Communications L 151 1.43 -0.800 1.187 3.290 2.247 0.159 0.367
TE

Conglomerate M 195 1.84 -0.167 0.984 2.632 1.431 0.144 0.352

Note: This panel presents the sample distribution by industry during 2007–2016.
C EP
AC

41
 ACCEPTED MANUSCRIPT

Table 3. Descriptive statistics and correlation coefficients of all variables

Panel A: Descriptive statistics
Variable Mean Std. Min p25 p50 p75 Max
Synch -0.457 1.016 -9.354 -0.994 -0.341 0.233 7.737
CEO media 2.937 1.625 0 1.792 2.708 4.331 10.633
POLCON 0.153 0.360 0 0 0 0 1

PT
Firm size 21.922 1.235 16.757 21.054 21.77 22.63 27.41
Lev 0.431 0.213 0.046 0.261 0.427 0.594 0.98
ROE 0.068 0.112 -0.787 0.031 0.069 0.113 0.565

RI
B/M 0.505 0.245 0.013 0.311 0.476 0.673 1.516
Turnover 5.851 4.057 0.003 2.875 4.862 7.842 37.955
Inholding 0.376 0.233 0 0.178 0.375 0.561 0.979

SC
First 0.353 0.150 0.043 0.233 0.333 0.453 0.900
Big_4 0.048 0.214 0 0 0 0 1
Ind_growth 0.161 0.135 -0.298 0.074 0.14 0.224 1.879

U
HHI 0.062 0.054 0.016 0.033 0.046 0.068 0.374
Duality 0.257 0.437 0 0 0 1 1
AN
Degree 2.548 1.372 1 1 3 4 5
Age 48.426 6.327 27 44 48 52 79
Gender 0.945 0.227 0 1 1 1 1
M
D
TE
C EP
AC

42
 ACCEPTED MANUSCRIPT

Panel B: Correlation coefficients

CEO Firm
Synch POLCON Lev ROE B/M Turnover Inholding First Big_4 Ind_growth HHI Duality

PT
media size
Synch 1
CEO

RI
-0.13*** 1
media
POLCON 0.06*** 0.15*** 1

SC
Firm size -0.03*** 0.11*** 0.01 1
Lev -0.01 -0.13*** -0.05*** 0.44*** 1

U
ROE -0.11*** 0.11*** 0.00 0.10*** -0.11*** 1
B/M 0.18*** -0.09*** -0.01 0.57*** 0.44*** -0.12*** 1

AN
Turnover 0.02** -0.01 0.00 -0.34*** -0.15*** -0.09*** -0.33*** 1
Inholding -0.11*** 0.00 -0.02* 0.40*** 0.17*** 0.13*** 0.11*** -0.51*** 1

M
First -0.06*** -0.02 -0.03*** 0.22*** 0.03*** 0.08*** 0.12*** -0.12*** 0.29*** 1
Big_4 -0.04*** 0.06*** -0.01 0.36*** 0.09*** 0.09*** 0.15*** -0.16*** 0.17*** 0.16*** 1

D
Ind_growth 0.01 -0.02* -0.01 -0.08*** 0.03*** 0.08*** -0.06*** -0.02* -0.03*** 0.01 -0.02** 1
HHI -0.04*** -0.02* -0.03*** 0.02** 0.00 -0.02** -0.03*** 0.02** -0.01 0.04*** 0.04*** 0.01 1

TE
Duality -0.03*** 0.39*** 0.19*** -0.15*** -0.15*** 0.00 -0.17*** 0.10*** -0.15*** -0.05*** -0.06*** -0.01 0.01 1
Degree -0.05*** 0.21*** 0.17*** -0.02* -0.14*** 0.02** -0.10*** 0.01 -0.01 -0.02* 0.04*** -0.05*** 0.03*** 0.12***
EP
Age 0.04*** 0.03*** 0.09*** 0.15*** 0.04*** -0.02* 0.06*** -0.06*** 0.11*** 0.02* 0.04*** -0.07*** -0.01 0.16***
Gender 0.03*** -0.04*** -0.05*** 0.04*** 0.04*** -0.01 0.04*** -0.03*** 0.04*** -0.01 0.01 0.00 0.00 0.03***
Note: *, ** and *** represents significant at the 10%, 5%, and 1% level, respectively.
C
AC

43
 ACCEPTED MANUSCRIPT

Table 4: Stock price synchronicity under different firm conditions

Type Level SYNCH T-Test for SYNCH Kruskal-Wallis

High -0.654
CEO media exposure 7.542*** 0.000
Low -0.333
With -0.313

PT
POLCON 6.077*** 0.000
Without -0.477
SOEs -0.513
Ownership 7.380*** 0.000

RI
Non-SOEs -0.367
Provincial financial Better -0.294
13.649*** 0.000
environment Worse -0.374

SC
Provincial legal Better -0.295
11.367*** 0.000
environment Worse -0.369
environment of *** represents significant at the 10%, 5%, and 1% level, respectively. High CEO
Note: *, ** and

U
media exposure is above its overall 75% level, and Low CEO media exposure is below its overall
25% level. Provincial financial environment is a dummy variable equals to 1 if the provincial
AN
financial development index is higher than the overall median index (i.e., it has better provincial
financial environment), and 0 otherwise. Provincial legal environment is a dummy variable equals
to 1 if the provincial legal development index is higher than the overall median index (i.e., it has
M

better provincial legal environment), and 0 otherwise.

Source: We obtain the data of provincial financial development index and provincial legal
D

development index from the “Marketization index of China’s provinces: NERI report 2017”, by
Wang Xiaolu, Fan Gang and Yu Wenjing.
TE
C EP
AC

44
 ACCEPTED MANUSCRIPT

Table 5. CEO media exposure and stock price synchronicity

Variable Model 1 Model 2 Model 3 Model 4

Firm size -0.230*** -0.211*** -0.238*** -0.219***
(-8.36) (-7.59) (-8.66) (-7.86)
Lev -0.389*** -0.412*** -0.362*** -0.388***

PT
(-4.16) (-4.40) (-3.87) (-4.15)
ROE -0.226** -0.222** -0.219** -0.213**
(-2.50) (-2.45) (-2.42) (-2.36)
B/M 2.112*** 2.087*** 2.109*** 2.083***

RI
(24.86) (24.53) (24.86) (24.51)
Turnover -0.028*** -0.028*** -0.028*** -0.027***
(-8.54) (-8.42) (-8.45) (-8.33)

SC
Inholding -0.373*** -0.374*** -0.375*** -0.376***
(-5.93) (-5.96) (-5.98) (-6.00)
First -0.067 -0.063 -0.050 -0.043

U
(-0.39) (-0.37) (-0.29) (-0.25)
Big_4 -0.167 -0.172 -0.175 -0.180
AN
(-1.37) (-1.41) (-1.43) (-1.47)
Ind_growth 0.307*** 0.305*** 0.317*** 0.316***
(3.33) (3.32) (3.45) (3.44)
HHI -1.812*** -1.763*** -1.869*** -1.824***
M

(-2.86) (-2.79) (-2.96) (-2.89)

Duality 0.043 0.084** 0.022 0.063
(1.09) (2.07) (0.56) (1.54)
D

Degree 0.025* 0.025** 0.024* 0.024*

(1.90) (1.97) (1.82) (1.86)
TE

Age 0.003 0.003 0.002 0.002

(1.04) (0.94) (0.78) (0.58)
Gender 0.030 0.037 0.032 0.041
EP

(0.34) (0.42) (0.36) (0.47)

Constant 4.023*** 3.705*** 4.185*** 3.863***
(6.82) (6.23) (7.09) (6.50)
Industry Yes Yes Yes Yes
C

Year Yes Yes Yes Yes

CEO media -0.037*** -0.039***
AC

(-4.24) (-4.42)
POLCON 0.210*** 0.206***
(4.84) (4.72)
CEO media*POLCON 0.037*
(1.94)
N 10571 10571 10571 10571
Adjust R2 0.328 0.330 0.330 0.332
F 193.537 186.570 186.904 173.773
Note: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively.
45
 ACCEPTED MANUSCRIPT

Table 6. Impact of ownership structure

SOEs Non-SOEs
Variable
Model 2 Model 3 Model 4 Model 2 Model 3 Model 4
Firm size -0.061 -0.091** -0.078* -0.321*** -0.350*** -0.324***
(-1.39) (-2.10) (-1.79) (-8.04) (-8.86) (-8.12)
Lev -0.604*** -0.550*** -0.564*** -0.267** -0.224* -0.256*

PT
(-4.38) (-3.99) (-4.09) (-2.03) (-1.70) (-1.94)
ROE -0.171 -0.159 -0.159 -0.347** -0.354** -0.341**
(-1.57) (-1.46) (-1.46) (-2.32) (-2.37) (-2.28)
B/M 1.599*** 1.622*** 1.603*** 2.574*** 2.597*** 2.561***

RI
(14.00) (14.27) (14.07) (20.10) (20.29) (19.99)
Turnover -0.033*** -0.032*** -0.032*** -0.024*** -0.025*** -0.024***
(-5.87) (-5.70) (-5.71) (-5.78) (-5.91) (-5.74)

SC
Inholding -0.444*** -0.452*** -0.456*** -0.268*** -0.267*** -0.268***
(-4.72) (-4.82) (-4.86) (-3.13) (-3.13) (-3.14)
First 0.012 0.025 0.018 0.049 0.048 0.063

U
(0.05) (0.10) (0.07) (0.20) (0.20) (0.26)
Big_4 -0.120 -0.137 -0.142 -0.283 -0.274 -0.282
AN
(-0.78) (-0.90) (-0.93) (-1.43) (-1.38) (-1.42)
Ind_growth 0.466*** 0.472*** 0.472*** 0.189 0.201 0.198
(3.69) (3.74) (3.75) (1.45) (1.54) (1.52)
HHI -1.113 -1.253 -1.164 -2.404*** -2.489*** -2.453***
M

(-1.10) (-1.24) (-1.15) (-2.89) (-2.99) (-2.95)

Duality 0.056 -0.001 0.025 0.105* 0.035 0.093
(0.93) (-0.01) (0.42) (1.85) (0.64) (1.63)
D

Degree 0.012 0.010 0.011 0.036* 0.035* 0.035*

(0.70) (0.59) (0.63) (1.85) (1.79) (1.84)
TE

Age -0.001 -0.001 -0.001 0.006 0.006 0.005

(-0.21) (-0.35) (-0.30) (1.61) (1.58) (1.32)
Gender -0.067 -0.046 -0.039 0.065 0.050 0.055
EP

(-0.54) (-0.37) (-0.31) (0.52) (0.40) (0.43)

Constant 1.026 1.565 1.333 5.463*** 5.975*** 5.563***
(1.07) (1.64) (1.39) (6.57) (7.23) (6.69)
Industry Yes Yes Yes Yes Yes Yes
C

Year Yes Yes Yes Yes Yes Yes

CEO media -0.031** -0.031** -0.045*** -0.047***
AC

(-2.27) (-2.26) (-3.83) (-4.01)

POLCON 0.259*** 0.260*** 0.145** 0.124*
(4.58) (4.57) (2.24) (1.87)
CEO media*POLCON 0.008 0.051**
(0.27) (2.00)
N 4405 4405 4405 6166 6166 6166
Adjust R2 0.372 0.374 0.375 0.318 0.317 0.319
F 94.391 95.439 88.368 101.170 100.582 93.889

46
 ACCEPTED MANUSCRIPT

Table 7: Impact of financial environment

Superior Financial Environment Inferior Financial Environment
Variable
Model 2 Model 3 Model 4 Model 2 Model 3 Model 4
Firm size -0.238*** -0.256*** -0.242*** -0.174*** -0.214*** -0.187***
(-5.99) (-6.50) (-6.09) (-4.45) (-5.55) (-4.77)
Lev -0.299** -0.265** -0.278** -0.555*** -0.496*** -0.537***

PT
(-2.29) (-2.02) (-2.13) (-4.17) (-3.74) (-4.04)
ROE -0.489*** -0.489*** -0.481*** -0.050 -0.046 -0.041
(-3.29) (-3.28) (-3.24) (-0.45) (-0.42) (-0.37)
B/M 2.131*** 2.150*** 2.126*** 1.978*** 2.013*** 1.980***

RI
(18.61) (18.80) (18.58) (15.49) (15.83) (15.55)
Turnover -0.028*** -0.029*** -0.028*** -0.026*** -0.025*** -0.025***
(-6.54) (-6.67) (-6.55) (-5.19) (-4.99) (-4.94)

SC
Inholding -0.409*** -0.407*** -0.410*** -0.296*** -0.304*** -0.300***
(-5.00) (-4.97) (-5.01) (-3.01) (-3.09) (-3.06)
First -0.060 -0.038 -0.046 -0.036 -0.023 -0.003

U
(-0.24) (-0.16) (-0.19) (-0.15) (-0.10) (-0.01)
Big_4 -0.084 -0.092 -0.098 -0.366* -0.342 -0.342
AN
(-0.56) (-0.61) (-0.65) (-1.71) (-1.60) (-1.61)
Ind_growth 0.425*** 0.431*** 0.430*** 0.111 0.134 0.133
(3.52) (3.57) (3.56) (0.78) (0.95) (0.94)
HHI -1.728** -1.796** -1.792** -1.928* -2.090** -1.942*
M

(-2.13) (-2.21) (-2.21) (-1.88) (-2.05) (-1.90)

Duality 0.101* 0.044 0.083 0.066 -0.001 0.044
(1.87) (0.84) (1.53) (1.07) (-0.01) (0.72)
D

Degree 0.035** 0.035** 0.034** 0.009 0.004 0.006

(2.03) (2.02) (2.01) (0.48) (0.20) (0.31)
TE

Age 0.002 0.002 0.001 0.003 0.002 0.002

(0.49) (0.42) (0.31) (0.76) (0.59) (0.37)
Gender 0.039 0.036 0.045 0.060 0.050 0.057
EP

(0.33) (0.30) (0.38) (0.45) (0.38) (0.44)

Constant 4.155*** 4.455*** 4.238*** 3.101*** 3.846*** 3.381***
(4.85) (5.22) (4.95) (3.77) (4.73) (4.11)
C

Industry Yes Yes Yes Yes Yes Yes

Year Yes Yes Yes Yes Yes Yes
AC

CEO media -0.035*** -0.036*** -0.042*** -0.045***

(-3.00) (-3.09) (-3.13) (-3.33)
POLCON 0.160*** 0.153** 0.263*** 0.263***
(2.61) (2.47) (4.34) (4.36)
CEO media*POLCON 0.029 0.052*
(1.13) (1.83)
N 6519 6519 6519 4052 4052 4052
Adjust R2 0.328 0.327 0.329 0.342 0.344 0.346
F 113.314 113.177 105.035 75.863 76.435 71.300

47
 ACCEPTED MANUSCRIPT

Table 8. Impact of legal environment

Stronger Legal Environment Weaker Legal Environment
Variable
Model 2 Model 3 Model 4 Model 2 Model 3 Model 4
Firm size -0.269*** -0.289*** -0.274*** -0.152*** -0.192*** -0.166***
(-7.04) (-7.62) (-7.16) (-3.67) (-4.68) (-3.98)
Lev -0.272** -0.233* -0.249** -0.657*** -0.601*** -0.639***

PT
(-2.17) (-1.85) (-1.99) (-4.62) (-4.25) (-4.50)
ROE -0.401*** -0.403*** -0.395*** -0.084 -0.078 -0.074
(-2.83) (-2.84) (-2.79) (-0.74) (-0.68) (-0.65)
B/M 2.217*** 2.233*** 2.213*** 1.900*** 1.942*** 1.901***

RI
(20.02) (20.20) (19.99) (14.01) (14.39) (14.06)
Turnover -0.030*** -0.031*** -0.030*** -0.022*** -0.021*** -0.021***
(-7.23) (-7.33) (-7.22) (-3.99) (-3.86) (-3.82)

SC
Inholding -0.434*** -0.432*** -0.435*** -0.217** -0.218** -0.215**
(-5.51) (-5.49) (-5.53) (-2.05) (-2.07) (-2.04)
First -0.017 0.004 -0.003 -0.130 -0.119 -0.093

U
(-0.07) (0.02) (-0.01) (-0.51) (-0.46) (-0.36)
Big_4 -0.115 -0.125 -0.128 -0.241 -0.200 -0.235
AN
(-0.79) (-0.86) (-0.88) (-0.99) (-0.82) (-0.96)
Ind_growth 0.448*** 0.457*** 0.455*** 0.045 0.065 0.069
(3.82) (3.90) (3.88) (0.30) (0.43) (0.46)
HHI -1.532** -1.693** -1.639** -2.295** -2.211** -2.207**
M

(-1.97) (-2.18) (-2.11) (-2.04) (-1.96) (-1.96)

Duality 0.075 0.016 0.056 0.131** 0.070 0.108
(1.44) (0.32) (1.06) (1.98) (1.08) (1.64)
D

Degree 0.038** 0.037** 0.038** 0.002 -0.004 -0.003

(2.27) (2.24) (2.25) (0.10) (-0.20) (-0.17)
TE

Age 0.003 0.002 0.002 0.002 0.001 -0.000

(0.70) (0.62) (0.53) (0.37) (0.21) (-0.01)
Gender 0.079 0.082 0.087 0.010 -0.018 -0.003
EP

(0.69) (0.72) (0.76) (0.07) (-0.13) (-0.03)

Constant 4.718*** 5.048*** 4.810*** 2.842*** 3.583*** 3.140***
(5.74) (6.18) (5.86) (3.24) (4.12) (3.57)
C

Industry Yes Yes Yes Yes Yes Yes

Year Yes Yes Yes Yes Yes Yes
AC

CEO media -0.033*** -0.034*** -0.045*** -0.049***

(-2.92) (-2.99) (-3.15) (-3.37)
POLCON 0.183*** 0.176*** 0.252*** 0.260***
(3.10) (2.93) (3.89) (4.01)
CEO media*POLCON 0.025 0.054*
(1.01) (1.71)
N 6985 6985 6985 3586 3586 3586
Adjust R2 0.340 0.340 0.341 0.325 0.326 0.329
F 126.918 126.985 117.742 61.615 61.938 57.903

48
 ACCEPTED MANUSCRIPT

Table 9. Robustness check

Cluster Driscoll-Kraay Bootstrap 2SLS
(1) (2) (3) (4)
Firm size -0.219*** -0.219*** -0.219*** -0.029*
(-6.74) (-3.44) (-7.43) (-1.93)
Lev -0.388*** -0.388*** -0.388*** -0.462***
(-3.73) (-3.39) (-3.51) (-7.01)

PT
ROE -0.213* -0.213*** -0.213* -0.589***
(-1.93) (-3.37) (-1.83) (-5.59)
B/M 2.083*** 2.083*** 2.083*** 0.767***
(20.63) (10.09) (20.67) (11.20)

RI
Turnover -0.027*** -0.027** -0.027*** 0.028***
(-7.01) (-2.01) (-6.99) (7.66)
Inholding -0.376*** -0.376*** -0.376*** -0.054

SC
(-5.26) (-3.24) (-5.31) (-0.83)
First -0.043 -0.043 -0.043 -0.267***
(-0.21) (-0.16) (-0.23) (-3.18)

U
Big_4 -0.180 -0.180* -0.180 -0.149***
(-1.25) (-1.70) (-1.33) (-2.67)
AN
Ind_growth 0.316*** 0.316*** 0.316*** 0.661***
(3.69) (2.83) (3.74) (6.59)
HHI -1.824** -1.824 -1.824** -2.655***
(-2.56) (-1.59) (-2.38) (-3.95)
M

Duality 0.063 0.063*** 0.063 -0.041

(1.35) (5.34) (1.43) (-1.21)
Degree 0.024 0.024*** 0.024 -0.006
D

(1.49) (3.25) (1.37) (-0.99)

Age 0.002 0.002 0.002 0.010***
TE

(0.50) (0.61) (0.47) (5.34)

Gender 0.041 0.041 0.041 0.078
(0.48) (0.65) (0.49) (1.58)
EP

Constant 3.863*** 3.863*** 3.863*** -0.367

(5.56) (2.88) (6.19) (-1.11)
Industry Yes Yes Yes Yes
Year Yes Yes Yes Yes
C

CEO media -0.039*** -0.039*** -0.039*** -0.031**

(-4.09) (-6.57) (-4.32) (-2.05)
AC

POLCON 0.206*** 0.206*** 0.206*** 0.273**

(3.87) (3.67) (3.35) (2.18)
CEO media*POLCON 0.037 0.037* 0.037 0.020
(1.62) (1.95) (1.56) (0.59)
N 10571 10571 10571 7761
Adjust R2 0.332 0.332 0.332 -
F 191.244 1830.036 - -
Ward Chi2 - - 10294.07 575.64
Note: *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively.
49