Opinion Audit

Pacific Accounting Review
CEO’s Political Connections, Institutions and Audit Opinions

Fang Hu, Jenny Stewart, Weiqiang Tan,
Article information:
To cite this document:
Fang Hu, Jenny Stewart, Weiqiang Tan, "CEO’s Political Connections, Institutions and Audit Opinions", Pacific Accounting
Review, https://doi.org/10.1108/PAR-04-2016-0047
Permanent link to this document:
https://doi.org/10.1108/PAR-04-2016-0047
Downloaded on: 11 July 2017, At: 12:25 (PT)
References: this document contains references to 0 other documents.
To copy this document: permissions@emeraldinsight.com
Access to this document was granted through an Emerald subscription provided by emerald-srm:333301 []
For Authors
If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service
information about how to choose which publication to write for and submission guidelines are available for all. Please
Downloaded by Cornell University Library At 12:25 11 July 2017 (PT)
visit www.emeraldinsight.com/authors for more information.

About Emerald www.emeraldinsight.com
Emerald is a global publisher linking research and practice to the benefit of society. The company manages a portfolio of
more than 290 journals and over 2,350 books and book series volumes, as well as providing an extensive range of online
products and additional customer resources and services.
Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee on Publication
Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation.
*Related content and download information correct at time of download.

CEO’s Political Connections, Institutions and Audit Opinions
Abstract
Purpose – The purpose of this paper is to investigate whether audit opinions of listed firms in
China vary systematically with the political connections of the firm’s chief executive officer
(CEO). Prior literature only shows the importance of political influence to auditor choice and
audit quality.
Design/methodology/approach –A politically connected firm is defined as a firm in which

the CEO has a political background. We use a “difference-in-difference” model to control for
self-selection problems.
Findings – We find that the likelihood of receiving a favourable opinion in the subsequent
period is positively associated with a CEO’s political connections. This positive association is
stronger with CEOs connected to local government within the same region. We further find
that the CEO’s political connections have more influence on favourable audit opinions in
non-state owned enterprises (non-SOEs), in a less developed and lower investor protection
region. The influence is also less significant in the regions where there are more non-state
owned or foreign banks and where there are greater penalties for political corruption and
relationship-based contracting.
Originality/value – The study complements and extends the existing literature on the role of
political connections in the economy by providing evidence on the effect of a CEO’s political
connections on audit opinions. We extend the research on auditing in emerging markets by
explicitly accounting for unique institutional and market factors in China. We explore audit
quality by observing how audit opinions are directly shaped by political and institutional
factors.
Keywords Political connections, Audit Opinions, State-owned enterprises (SOE), China
Paper type Research paper

Acknowledgments (if applicable):
We appreciate Dr. Jing Liao, Dr. Jing Chi and two anonymous reviewers for their useful suggestions. This work
was supported by the Griffith University under New Researcher Grant and the National Natural Science
Foundation of China (NO.71210003). Dr. Tan acknowledges the financial support from Research Centre of Low-
carbon Economy for Guangzhou Region, Jinan University.
1. Introduction
In this paper we investigate whether audit opinions of listed firms in China vary
systematically with the political connections of the firm’s chief executive officer (CEO).
Although recent research analyses the importance of political influence to auditor choice and
audit quality (Chan et al., 2006; Chan et al., 2012; Gul, 2006; Wang et al., 2008; Guedhami
et al., 2009; and Guedhami et al., 2014), there is a lack of direct evidence on how political
connections affect audit opinions and whether the effect is confounded by other institutional
characteristics.
A number of studies have demonstrated that auditor choice and quality are shaped by
political influence in China’s state-owned enterprises (SOEs). For example, Chan et al.
(2006), Chan et al. (2012) and Wang et al. (2008) find that SOEs, especially local SOEs,
have a strong tendency to hire small local auditors or switch from a non-local auditor to a
local auditor in pursuit of desirable opinions. These studies assume that small local auditors
in China are subject to greater political pressure or have more reliance on local governments
so that they will issue more favorable opinions. Based on this assumption, they typically
examine the impact of political influence on auditor choice from the demand side, which
accordingly affects the quality of audit opinions. While these studies emphasize the
interaction of managers, auditors and government under political influence, they overlook
institutional differences. Moreover, it is unclear whether political influence varies with

features of the market and institutional factors. We fill this gap in the literature by empirically
examining and answering the following questions: (1) Do a CEO’s political connections
affect audit opinions? (2) Does this effect vary with institutional features such as the
existence of government ownership and other market characteristics?
We analyse all companies listed on the domestic stock exchanges in China during the
period 2005 to 2010. The descriptive data show that firms with politically connected CEOs
receive 11 per cent more favourable opinions than firms without political connected CEOs.
We find that the likelihood of receiving a favourable opinion in the subsequent period is
positively associated with a CEO’s political connections[1]. This positive association is
stronger with CEOs connected to local government within the same region. Our results are
robust when we use a “difference-in-difference” model to control for self-selection problems.
We interpret these results from two perspectives. On the demand side, we argue that
politically-connected CEOs will actively seek unqualified audit opinions. This is because
insiders with political connections wish to suppress information on actual economic
performance to ensure that their discretionary practices, largely stemming from political
cronyism and corruption, are kept hidden. On the supply side, auditors have incentives to
lower the threshold for issuing an unqualified opinion to clients with political connections.
This is because auditors have greater economic dependence on political connections in
China’s political and legal environment (Chan et al., 2006) and they may fear losing the
specific rents from politically connected clients if they issue a qualified opinion.
Alternatively, auditors may perceive that firms with political connections, especially those
connected to local government, have less litigation risk than other firms, thus raising the level
of acceptable errors.
Next, we analyse how the impact of a CEO’s political connections on audit opinions is
shaped by institutional factors. First, we find that the CEO’s political connections have more
influence on favourable audit opinions in non-state owned enterprises (non-SOEs) than in
state owned enterprises (SOEs). We then examine the CEO’s political connections interacted
with institutional characteristics based on regions in China. We find the impact of a CEO’s
political connections on audit opinions is less pronounced in regions where the market is
more developed and there is a higher level of investor protection[2]. The influence is also less
significant in the regions where there are more non-state owned or foreign banks and where
there are greater penalties for political corruption and relationship-based contracting. As
additional tests, the results show that the influence of political connections on audit opinions
is not related to auditor size and auditor locality and only persists in the firms that do not
switch their auditors.
This paper contributes to previous research in two ways. First, the study complements
and extends the existing literature on the role of political connections in the economy (Bliss
and Gul, 2012; Claessens et al., 2008; Fan et al., 2007; Guedhami et al., 2014; Li et al.,
2008). Prior studies mostly investigate political influence embedded in state ownership. For
instance, Wang et al. (2008) demonstrate that auditor choice in SOEs in China is subject to
political influence. We attempt to provide evidence on the effect of a CEO’s political
connections on this issue.
Second, we extend the research on auditing in emerging markets (Chan et al., 2006;
Chan et al., 2012; DeFond et al., 2000; Wang et al., 2008) by explicitly accounting for
unique institutional and market factors in China. In this single emerging market, we avoid the
possibility of contamination due to omitted variables resulting from country differences
(Faccio, 2006). Moreover, the Chinese market has a high level of government intervention,
with various layers of government and differences across regions. In contrast to prior studies
that focus on how auditor choice affects the pursuit of audit opinions (Chan et al., 2006; Chan
et al., 2012; DeFond et al., 2000; Wang et al., 2008), we explore audit quality by observing
how audit opinions are directly shaped by political and institutional factors.
The remainder of this paper is organised as follows. Section 2 provides the background
to the study and develops the hypotheses. Section 3 describes the research method including
the sample, the regression models and variables. Section 4 outlines our data and summarises
the descriptive statistics on the regression variables. Sections 5 and 6 report the results of the
multivariate analysis and the robustness tests respectively. Section 7 discusses the policy
implications of our results while Section 8 contains some concluding comments.

2. Background and hypotheses development
2.1 Political influence, audit profession and auditor quality in China
Chinese certified public accounting (CPA) firms were initially established and owned by
government bodies. Chan et al. (2006) suggest that this affiliation between audit firms and
local government led to a lack of independence as well as to regional protectionism. Between
1995 and 2000, the Chinese government launched a series of audit market reforms to improve
the independence of the auditing profession. These included (1) the adoption of international
auditing standards; (2) the separation of audit firms from government control; and (3) the
facilitation of mergers of small audit firms (Wang et al., 2008)[3].
Despite the reforms, political influence still affects the economic dependence of audit
firms in China (Chan et al., 2006). From the competition perspective, audit firms can gain
advantage by maintaining close relationships with government bureaucrats (Chan and Mo,
2002; Hofstede, 2001). For instance, auditors have more knowledge of government
procedures and regulations because of their close relationship with government (Yang, 2013).
Thus, they are better able to retain existing clients or find new clients. From the risk
perspective, auditors might encounter lower litigation risk given a connection to government
as the Chinese government strongly influences business activities (Fan et al., 2007; Chen et
al., 2011). This in turn impacts the quality and independence of audit firms (or individual
auditors). For example, many local auditors may have strong incentives to issue favourable
audit reports in order to maintain existing connections (Chan et al., 2006). Furthermore,
clients are unlikely to switch from auditors they perceive as lenient.
Based on the assumption that certain types of audit firms, such as small and local
auditors, are subject to more political influence, previous studies find that SOEs might select
or switch to these auditors in order to obtain desirable audit opinions. For example, using a
sample of Chinese listed firms for the period 1996–2002, Chan et al. (2006) find that local
government-owned companies switched from a non-local auditor to a local auditor following
receipt of a qualified opinion. Wang et al. (2008) find that SOEs have a stronger tendency to
hire small local auditors than non-state firms because governments may use political pressure
to coerce small local auditors to issue more favourable opinions to their SOEs. Chan et al.
(2012) provide further evidence that local government-controlled companies choose local
auditors to obtain more favourable audit opinions when they face the need for new equity
financing or the threat of exchange delisting.
2.2 Hypothesis development
2.2.1 Political connections and audit opinions. According to agency theory, external
auditing provides a monitoring role to minimise financial reporting bias stemming from
managerial incentives. Prior literature documents that political connections provide incentives
to distort financial reports and reduce accounting transparency. Bushman et al. (2004) report
that higher government share ownership is negatively associated with financial transparency
while Bushman and Piotroski (2006) find that earnings are less conservative in firms
operating in countries with higher state involvement in the economy. Chaney et al. (2007)
find a negative association between politically-connected firms and earnings quality.
According to Leuz and Oberholzer-Gee (2006: 416), “…high levels of transparency and
public attention might expose political favors of questionable legality. For instance, firms in
weakly regulated markets are often free to engage in undisclosed related-party transactions
benefiting controlling insiders and political backers.” Those politically connected insiders
could exploit their position to siphon corporate resources that they later conceal by distorting
the financial statements (Shleifer and Vishny, 1994; La Porta et al., 1998; Dyck and Zingales,
2004). Piotroski et al. (2008) find that politically connected firms are particularly eager to
suppress information about poor economic outcomes since these firms have incentives to
conceal from minority shareholders expropriation-related activities stemming from political
cronyism and corruption. To mitigate the agency costs arising from political connections, it
could be argued that external auditors are more likely to issue qualified or modified audit
opinions when the CEO is politically connected. However, this argument assumes a strong
regime of audit independence, where audit firms do not have incentives to develop their own
connections to government.
In contrast to agency theory, the resource dependence theory of the firm recognises the
value of a CEO’s political connections because these connections help firms obtain
favourable resources and thereby increase their negotiation power in contracting. As
discussed above, audit firms in China are economically dependent on political connections to
do business (Chan et al., 2006; Yang, 2013). As a result, auditors are expected to actively
lower the threshold of issuing unqualified opinions and be more lenient to politically
connected firms because (1) they can retain existing clients where politically connected
insiders might have more power to dismiss or appoint an auditor; (2) they can rely on their
political ties or relationships with governments to get more clients; and (3) they can reduce
auditing risk by maintaining close political ties as they can obtain insight into the regulatory
process and lobby for more favourable regulatory decisions (Yang, 2013). In particular,
auditors may perceive that firms with political connections, especially those connected to
local government, have less litigation risk than other firms and thus they may raise the level
of acceptable errors. Because China’s audit market is heavily influenced by political
relationships, resource dependence arguments are more relevant than agency theory
arguments in explaining auditor behaviour. We therefore develop the following hypothesis:
H1: Politically connected firms in China are more likely to obtain a favourable audit
opinion than firms without political connections.

2.2.2 CEO’s political connections, audit opinions and institutional factors. Institutional
factors, such as legal, political, financial, regulatory and cultural institutions, exert strong
pressures on economic agents’ incentives and behaviours. We therefore identify a number of
key institutional factors that may influence the relation between a CEO’s political
connections and audit opinions in China.
(i) Existence of government ownership
A unique feature of China’s market is the dominance of government ownership and high
state control of business. The CEO’s political connections might be mixed with the natural
political connections driven by government ownerships. Agency arguments suggest that
politically connected CEOs are appointed by the government owners in SOEs to align firm
goals with government objectives such as reducing unemployment. Hence, the impact of
politically connected CEOs in SOEs on external auditors may be diluted by government
ownerships[4]. Among non-SOEs (or privately owned firms) that lack ties with government
but operate in a relationship-based environment, having politically-connected managers helps
them to obtain favourable treatment and overcome market and institutional barriers (Li et al.,
2008). In accordance with resource dependency theory, CEO’s political connections can be
more advantageous for non-SOEs especially in emerging economies, which typically lack
property rights protection and the market-supporting institutions needed by private firms
(McMillan, 1995). In particular, non-SOEs lack political connections by nature of their
ownership. Therefore, when political connections are established by CEOs in non-SOEs,
audit firms may seek advantage by colluding with these clients as a way of maintaining close
relationships with government. These arguments lead to the following hypothesis:
H2: The effect of a CEO’s political connections on audit opinions is more pronounced in
non-SOEs than in SOEs.

(ii) the level of investor protection and market development
The level of market development and the resultant information environment is a key
institutional feature which shapes both financial reporting practices and monitoring
mechanisms (Piotroski and Wong, 2011). In general, stronger investor protection generates
higher quality financial reporting, i.e. less earnings management (e.g. Leuz et al., 2003),
more conservative reporting (e.g. Ball et al., 2000; Ball et al., 2003; Bushman and Piotroski,
2006) or higher earnings informativeness (e.g. DeFond et al., 2007). In particular, Francis et
al. (2005) find that a high quality (or more independent) auditor is used in the presence of
strong institutions. In an environment encouraging greater investor protection and fewer
relation-based contracts, the political connections or other relationships exert less pressure on
economic agents’ behaviour. For example, Faccio (2006) finds that the favourable treatment
enjoyed by firms with political connections is more pronounced in countries with
interventionist governments and weak protection of property rights. As such, we conjecture
that in a more developed market with stronger investor protection, an audit opinion is more
independent of political connections or more associated with information quality disclosed by
firms. Hence we test the following hypothesis:

H3: The effect of a CEO’s political connections on audit opinions is more (less)
pronounced in regions with lower (higher) level of market development.
(iii) Demand for information in contracting
Institutional factors also influence the demand for information and monitoring
mechanisms to ensure the quality of information in firms’ managerial and debt contracts. The
stronger the contractual demand for credible information, the higher is the quality of
disciplinary mechanisms, such as the independence of external auditors, to ensure the

reporting of high quality information to all contractual parties (Watts, 2006). Chinese firms
overwhelmingly rely on banks for financing and this increases the demand for information
quality. Chen et al. (2010) suggest that state-owned banks have weaker demand for
information quality as they adopt less accounting-based covenants than non-state owned
banks (including foreign banks). Hence, we propose that audit opinions are less influenced by
political connections in markets with stronger contractual demand for credible information.
This leads to the following hypothesis:
H4: The effect of a CEO’s political connections on audit opinions is more (less) pronounced
in regions with more (less) state-owned banks.
(iv) Penalties for accounting scandals and relationship-based contracting
How the market penalises misconduct is also a key feature to shape the incentives and
behaviours of market participants. When firms are caught in accounting scandals, the
reputation penalty is high. For example, studies show that a significant portion of a share
price decline is associated with scandals; there is a loss of potential new contracts or increase
in future contracting costs; the senior officers involved in scandals are dismissed and have
difficulty in finding comparable positions in the future (Karpoff et al., 2004; Karpoff et al.,
2008a, b). In China’s market, Hung et al. (2009) find that the share prices of the firms with
senior officers caught in political corruption charges and accounting scandals dropped more
significantly than those that were only involved in accounting scandals alone. After the
corruption charges, those firms find it hard to get bank loans and suffer from significant
board changes. These unfavourable economic consequences also have spillover effects on
auditors’ litigation risks. Hence, we conjecture that auditors are less likely to compromise
their opinions when a region has more severe penalties for managers’ and auditors’
misconduct based on relationships. We therefore test the following hypothesis:

H5: The effect of a CEO’s political connections on audit opinions is more (less)
pronounced in regions with less (more) penalties for accounting scandals and relationship-
based contracting.
3. Research design
3.1 Measures
3.1.1 Political connections. A politically connected firm is defined as a firm in which the
CEO has a political background. Given the dual presence of the Chinese Communist Party
(CCP) and government organs at each level of China’s political hierarchy (Whiting, 2001), a
CEO is identified as having a political background if (1) he or she has worked in government
entities (including the military); or (2) has held one of three main CCP positions – Party
Secretary, Central Committee Member, or Representative of the National People’s Congress.
In addition, given the importance of locality in China’s economy, we also classify the
political connections as local and non-local connections. A CEO with local political
connections is defined as a CEO who has a political background within the same region
(based on province in China).

3.1.2 Audit opinion. Auditors in China issue five types of audit opinion. These are
unqualified, unqualified with an explanatory paragraph, qualified, disclaimer, and adverse
opinions. Prior studies of the Chinese audit market define all opinions other than completely
unqualified opinions as “qualified” or “unfavourable” opinions (Chen et al., 2000; Chen et al.,
2001; DeFond et al., 2000; Chan et al., 2006)[5]. This is because unqualified opinions with
an explanatory paragraph are used by auditors in China as an alternative to qualified opinions
in order to retain clients (Chen et al., 2000). Therefore, this type of opinion is treated the
same as a qualified opinion by the China Securities Regulatory Commission (CSRC,

2001b)[6]. We therefore follow these other studies and include unqualified opinions with an
explanatory paragraph as unfavourable opinions.
3.2 Models
We first test the relationship between political connections and the likelihood of favourable
opinions. Since we focus our attention on the post-connection period to see if firms
successfully obtain a favourable opinion in the subsequent period after they establish political
connections, we follow Chan et al. (2006) in using a LOGIT model as follows:
(
)
= +
( _
) +
+ !
+ " #$_

+ % & _
+ ' ($
+ ) * ( +
+ , --

+ . / -ℎ
+ 1 2
+ 2&3_4

+ 5-ℎ

+ 6_#&77+ + * 3&-+_#&77+ + 8

Model (1)
where
Opinionit+1 = 1 if audit opinion in the subsequent period is favourable as explained earlier,
and 0 if it is unfavourable;
PCit = 1 if the firm’s CEO is politically connected as explained earlier, and 0 otherwise.
Local_PCit = 1 if the firm’s CEO is politically connected within the same region, and 0
otherwise.
Following prior studies (Chen et al., 2001; DeFond et al., 2000; Chan et al., 2006), we
include a number of control variables as follows:

Sizeit is measured as the natural log of the client’s total assets. Chan et al. (2006) argue
that larger companies are more financially stable than smaller companies and hence Sizeit is
expected to be positively associated with the likelihood of receiving an unqualified opinion.
ROEit measures profitability by dividing net income by total equity. As prior studies indicate
that poorly performing companies are more likely to manage earnings opportunistically than
profitable companies (Chan et al., 2006; Chen et al., 2001; Schwartz and Menon, 1985), we
expect ROEit to be positively associated with the probability of receiving an unqualified
opinion.
Debt_Ratioit (debt-to-total assets ratio) measures the level of leverage. Prior studies
indicate that companies with higher leverage ratios are more likely to receive modified
reports (Chan et al., 2006; DeFond et al., 2000; Wilkins, 1997). We therefore expect
Debt_Ratioit to be negatively associated with the receipt of an unqualified audit opinion.
Current_Ratioit (the ratio of current assets to current liabilities) measures the financial
liquidity of companies. Prior studies indicate that companies with lower current ratios are
more likely to receive modified reports (Chan et al., 2006; DeFond et al., 2000; Wilkins,
1997). Current_Ratioit is therefore expected to be positively associated with such an opinion.
Receivableit (accounts receivable divided by total assets) and Inventoryit, (inventory divided
by total assets) are included to control for audit risk (Hay et al., 2006). We expect these two
variables to be negatively associated with the receipt of an unqualified opinion.
Lossit (a dummy variable for a firm making a loss) is included because companies that
have reported losses are more likely to manage earnings upwards to avoid delisting[7].
Similar to our arguments relating to ROEit, companies that manipulate earnings are more
likely to receive a qualified audit opinion (Chan et al., 2006; Chen et al., 2001).
Ownershipit is the ratio of the number of shares held by the largest shareholder over the
number of total shares of the firm. In the context of audit fee models, Hay et al. (2006: 171)
argue that “the existence of a dominant shareholder could indicate higher agency costs or
stronger control”, with either a positive or negative effect on audit fees. Similarly, the effect
of a dominant shareholder on the likelihood of receiving a qualified audit opinion may be
positive or negative. Hence we expect Ownershipit to be associated with the likelihood of
receiving an unqualified opinion but do not predict a direction.
Ageit is defined as the firm’s age at period t. Chen et al. (2001) find that companies
listed for some years will manipulate earnings to avoid delisting (see endnote [7]). We
therefore expect Ageit to be negatively associated with the receipt of an unqualified opinion.
Audit_Feeit is expected to be positively associated with the issue of a qualified opinion
as problems in completing the audit increase both the amount of audit work needed and audit
risk (Simunic, 1980; Hay et al., 2006) We measure Audit_Feeit as the natural log of audit fees
in period t.
B_shareit is a dummy variable equal to 1 if the client firm issues B shares and 0
otherwise. In China, B shares are eligible for foreign investors and are traded in foreign
currencies. For instance, B shares are traded in U.S. dollars on the Shanghai Stock Exchange,
and traded in Hong Kong dollars on the Shenzhen Stock Exchange. This variable controls for
the effect of foreign investors having a greater demand for high quality financial information.
In addition to compliance with generally accepted accounting principles in China, companies
with B shares are required to present financial statements in accordance with International
Accounting Standards. They are also required to hire one domestically licensed auditor as
well as an international audit firm to audit financial reports (Chan et al., 2006). Hence,
companies with foreign ownership are expected to be more likely to receive unqualified audit
opinions (Chan et al., 2006; Chen et al., 2001; DeFond et al., 2000). We therefore expect
B_shareit to be positively associated with the receipt of an unqualified opinion.
We also include year and industry dummy variables to control for any year effects and
industry effects.
We then test H2 to H5 using an extended model as follows, examining how the
likelihood of favourable opinions is affected by political connections interacted with the
various institutional factors.
(
)
= +
+
∗ * -& -
+
+ " !

+ % #$_
+ ' & _

+ ) ($

+ , * ( +
+ . --
+ 1 / -ℎ
+ 2

+ 2&3_4
+ 5-ℎ
+ 6_#&77+
+ * 3&-+_#&77+ + 8

Model (2)
To test H2, the institutional factor is:
NSOE = 1 if a firm is a non-SOE (its ultimate owner is not a government-related
organisation), and 0 otherwise;

MINDEX = the regional marketisation index reported by Fan et al. (2011), widely used in
prior studies to measure the level of investor protection and market development across
regions in China.
NSB = the number of non-state owned banks (including foreign banks) in a region (based on
provinces) of China.
Penalty = the number of firms caught in political corruption charges and accounting scandals
by the China Securities Regulatory Commission (CSRC) in a region (based on provinces) of
China.
The other variables are as defined in model (1).
4. Data and sample
4.1 Data
The financial data of all listed Chinese firms from 2005 to 2010 are obtained from the China
Securities Markets and Accounting Research (CSMAR) database. Information on the
backgrounds of board chairs, CEOs, and companies is collected by reading annual reports or
summaries of top management resumés in annual reports that are issued on the finance
website[8]. We choose the time period starting from 2005 because Chinese listed firms were
not required to disclose the directors’ and managers’ employment history and compensation
in the annual report prior to this year. Before 2005, the data is very limited as few firms
voluntarily disclosed this information. The CSMAR database serves as the primary source for
the audit data such as audit opinion, audit fee and auditor details. The institutional factors
such as the nature and number of banks in China and firms violating regulation caught and
penalised by CSRC are also extracted from the CSMAR database.
Firms lacking financial or audit information and financial institutions (1-digit SIC code:
I) are excluded[9]. The final sample consists of 1875 listed enterprises in China from 2005 to
2010. As data for some firms are not available in all years, the total number of firm years is
8617.
4.2 Sample description
Panel A of Table 1 reports details of the audit opinions partitioned by politically connected
(PC) firms and non-politically connected (NPC) firms during the period 2005 to 2010. It
shows that 2,273 firm year observations (26 per cent of the sample) are politically-connected
where the CEO has a political background. Consistent with prior studies (Chan et al., 2006),
most firm-year observations in China receive favourable opinions (86 per cent of firms in the
sample receive a favourable opinion). A greater percentage of PC firms (94 per cent) receive
a favourable opinion compared to NPC firms (83 per cent).
Panel B of Table 1 reports the number and percentage of changes in the type of audit
opinions (from year t-1 to year t) partitioned by changes in political connections of a firm
(from year t-1 to year t). It shows that more than half of the firm-year observations receive
the same audit opinion as that received in the previous year and retain connections or no
connections as in the previous year. Interestingly, the distribution in Panel B indicates that
firms currently establishing political connections are more likely to be given a favourable
opinion in the current year when they previously received a qualified opinion. Specifically, 7
per cent of the firm-year observations receiving a qualified opinion in the previous year but a
favourable audit opinion in the current year have political connections. That percentage is
higher than those without established political connections.
Table 1 Descriptive statistics of audit opinions partitioned by politically connected firms (PC)
and non-political connected firms (NPC).
Panel A: Audit opinions partitioned by PC vs. NPC
PC NPC
Audit Opinions number % number %
Unqualified 2126 94 5264 83
Unqualified with explanation 98 4 517 9
Qualified 35 2 286 5
Disclaimer & adverse 14 6 277 4
Observation No. 2273 100 6344 100
Note: unqualified opinions are identified as “favourable opinions”. The others (unqualified with explanation, qualified,
disclaimer & adverse) are identified as “unfavourable opinions”.
Panel B: Audit opinions from year to year partitioned by political connections from year to year
PC NPC
From NPC to PC No change From PC to NPC No change
Audit opinion number % number % number % number %
Change from unfavourable
7 0.07 37 0.03 7 0.03 101 0.02
to favourable
Change from favourable to
2 0.02 20 0.01 12 0.05 89 0.02
unfavourable
No change - favourable 88 0.86 1311 0.92 204 0.87 3714 0.91
No change - unfavourable 5 0.05 55 0.04 12 0.05 188 0.05
Obervation No. 102 100 1423 100 235 100 4092 100
Table 2 provides the descriptive statistics for firm characteristics and audit opinions
partitioned by PC firms and NPC firms[10]. The mean (median) total assets of RMB 5,013
million (RMB 1,917 million) in PC firms is significantly larger than the mean (median) assets
of RMB 3,935 (RMB 1,849) in NPC firms (t-statistic = -4.44, Wilcoxon z-statistic = 2.68).
The mean (median) leverage of 0.54 (0.52) in PC firms is significantly higher than the mean
(median) leverage of 0.52 (0.51) in NPC firms (t-statistic = -1.64, Wilcoxon z-statistic = 3.62).
The current ratio, accounts receivable and inventory in PC firms are significantly lower than
those in NPC firms. The results imply that PC firms are larger, have higher leverage but
lower financial liquidity than non-PC firms. Interestingly, the proportion of PC firms
choosing Big Four auditors is significantly more than the proportion of NPC firms. The
proportion of PC firms receiving favourable audit opinions is significantly more than the
proportion of NPC firms.
Table 2 Descriptive statistics on client firm characteristics partitioned by politically

connected firms (PC) and non-political connected firms (NPC)
PC NPC PC vs. NPC

N=2273 N=6344
Mean Median Mean Median Mean (t-test) Median (Wilcoxon)
Difference t-statistic Difference z-statistic
Total Assets (RMB Million) 5013 1917 3935 1849 1078 -4.44*** 68 2.68***
ROE 0.02 0.03 0.03 0.03 -0.01 1.39 0 -3.58***
Debt Ratio 0.54 0.52 0.52 0.51 0.02 -1.64* 0.01 3.62***
Current Ratio 1.49 1.12 1.79 1.26 -0.3 5.66*** -0.14 -9.10***
Receivable 0.09 0.06 0.11 0.08 -0.02 6.70*** -0.02 -7.74***
Inventory 0.16 0.12 0.18 0.14 -0.02 4.31*** -0.02 -7.57***

Loss 0.12 0 0.11 0 0.01 -1.39 0 1.42
Ownership (%) 0.33 0.31 0.32 0.3 0.01 -2.67*** 0.01 2.58***
Big Four 0.05 0 0.04 0 0.01 -1.77* 0 1.84*
Age (years) 12 12 11 11 1 -6.92*** 1 6.52
Bshare 0.08 0 0.05 0 0.03 -3.68*** 0 3.98***
Audit Fee (RMB Million) 0.59 0.46 0.6 0.48 -0.01 0.38 -0.02 -0.73
Opinion 0.94 1 0.83 1 0.11 -2.23* 0 0.23
Note: Total Assets is nominated in million RMB. Return on equity (ROE) is defined as the net income divided by the average net assets. Debt Ratio is
defined as the total debts divided by total assets. Current Ratio is defined as the ratio of total current assets to total current liabilities. Receivable is
defined as the account receivable divided by total assets. Inventory is defined as the inventory divided by total assets. Loss is defined as a dummy
variable equal to 1 if a firm’s net income is less than zero. Ownership is defined as the percentage of shareholdings held by the largest ultimate owner.
Big Four is defined as a dummy variable equal to 1 if a firm is audited by an international Big Four auditor. Age is defined as firm’s age. B share is
defined as a dummy variable equal to 1 if a firm has foreign-owned shares. Audit fees are nominated in million RMB. *, **, and *** indicate
significance at 10%, 5%, and 1% level, respectively.
5. Results
5.1 CEO’s political connections and subsequent favourable opinion
Table 3 presents the logistic regressions for the association between political connections and
the subsequent audit opinion based on model (1). The dependent variable is the dummy
variable for a favourable audit opinion in the subsequent period. The main independent
variable is the firm’s political connections. Various control variables for firm characteristics
are included. Column (1) shows that the likelihood of a favourable opinion in year t+1 is
significantly associated with political connections in year t (coefficient = 0.26, Wald Chi-
statistic = 3.33, p < 0.10). Column (2) shows that the likelihood of a favourable opinion in
year t+1 is significantly associated with political connections to local government in year t
(coefficient = 0.39, Wald Chi-statistic = 3.31, p < 0.10). Column (3) shows that the positive
association between subsequent favourable opinions and political connections to local
government is significantly stronger (coefficient = 0.83, Wald Chi-statistic = 5.36, p < 0.05).
Overall, our findings provide support for Hypothesis H1 that politically connected firms are
more likely to obtain a favourable audit opinion than firms without political connections[11].
Table 3 Logistic regression of subsequent audit opinions and political connections (use
Model (1)).
Dependent variable = Opinion it+1

Independent variables (1) (2) (3)
Intercept 0.27 -1.31 -1.71
(0.48) (0.36) (0.60)
PC it 0.26 0.31
(3.33*) (2.70*)
Local_PC it 0.39 0.83
(3.31*) (5.36**)
Size it 0.19 0.30 0.30
(5.24**) (9.06***) (9.19***)
ROE it 0.17 -0.05 -0.05
(0.06) (0.39) (0.34)
Debt_Ratio it -0.73 -2.76 -2.78
(22.99***) (33.24***) (33.58***)
Current_Ratio it 0.11 -0.07 -0.07
(2.45) (1.36) (1.36)
Receivable it -2.36 -1.65 -1.56
(15.56***) (5.15**) (4.61**)
Inventory it 0.99 1.65 1.69
(4.17**) (7.37***) (7.77***)
Loss it -1.39 -1.37 -1.36
(65.28***) (64.85***) (64.06***)
Ownership it 1.48 1.61 1.61
(12.71***) (11.03***) (10.99***)
Age it -0.10 -0.08 -0.08
(37.94***) (14.18***) (14.42***)
Audit_Fee it -0.17 -0.13 -0.13
(1.11) (0.48) (0.48)
B share it -0.21 -0.04 -0.06
(0.61) (0.01) (0.03)
Year dummy Yes Yes Yes
Indus dummy Yes Yes Yes
Psuedo R-square 0.08 0.06 0.06
Obs. No 8617 8617 8617
Note: The dependent variable is subsequent audit opinion (Opinion it+1), which is defined as 1 if audit
opinion is favourable and 0 otherwise. The independent variable is PC it, which is defined as 1 if a
CEO is politically connected (refer to as definition in Section 3) and 0 otherwise, or Local_PC it,
which is defined as 1 if a CEO is politically connected to local government (refer to as definition in
Section 3) and 0 otherwise. See Table 2 for other variable descriptions. The year dummy variables and
industry dummy variables are also included. The number in the parenthesis below is Wald Chi-square
value. *, **, and *** indicate significance at 10%, 5%, and 1% level, respectively.
5.2 CEO’s political connections and subsequent favourable opinion with existence of
government ownership
In Column (1) of Table 4, we extend model (1) to include an interaction term – political
connections with a dummy variable for non-SOEs. The results show that the likelihood of a
favourable opinion in year t+1 is significantly associated with local political connections in
year t (coefficient = 1.23, Wald Chi-statistic = 3.22, p < 0.10). The interaction of local
political connections and non-SOEs in year t is significantly positively associated with the
likelihood of a favourable opinion in year t+1 (coefficient = 0.33, Wald Chi-statistic = 3.65, p
< 0.10). The results suggest that auditors perceive local political connections to be more
important in non-SOEs since connections with local government can have a powerful effect
on their business. To further test these associations, in column (2) we partition the sample by
different layers of SOEs (local SOEs and central SOEs) and non-SOEs. In the local SOE
sample, column (3a) shows that the likelihood of a favourable opinion in year t+1 is
positively associated with political connections in year t but the result is not statistically
significant. Column (3b) shows that the likelihood of a favourable opinion in year t+1 is
significantly and positively associated with political connections to local governments in year
t (coefficient = 0.33, Wald Chi-statistic = 3.12, p < 0.10). In the central SOE sample, columns
(3c) and (3d) show that the likelihood of a favourable opinion in year t+1 is not significantly
associated with either political connections or local connections in year t. In the non-SOE
sample, columns (3e) and (3f) show that the likelihood of changing to a favourable opinion in
year t is significantly and positively associated with political connections (coefficient = 0.11,
Wald Chi-statistic = 3.27, p < 0.10) and political connections to local government in year t
(coefficient = 0.78, Wald Chi-statistic = 3.44, p < 0.10). Overall, the positive association
between political connections and audit opinions exists in local SOEs and non-SOEs but is
stronger in the sample of non-SOEs. These results arise because CEO’s political connections
are more crucial in influencing corporate activities and other related parties in non-SOEs
which lack natural political connections by way of ownership.
Table 4 Logistic regression of subsequent audit opinions and CEO’s political connections in
SOEs versus non-SOEs

(2)
Local SOEs Central SOEs Non-SOEs
Independent var.
(1) (a) (b) (c) (d) (e) (f)
Intercept 0.56 0.98 0.39 0.174 0.17 -0.64 -3.04
(0.23) (0.3) (0.15) (0.616) (2.00) (0.05) (0.9)
PC it 0.4 0.17 0.014 0.11
(2.36) (1.93) (0.786) (3.27*)
Local_PC it 1.23 0.33 0.003 0.78
(3.22*) (3.12*) (0.993) (3.44*)
PC it* NSOE it 0.18
(2.43)
Local_PC it* NSOE it 0.33
(3.65*)
NSOE it 0.63
(1.09)
Size it 0.51 0.03 0.12 0.124 -0.05 0.35 0.35
(2.15) (1.86) (2.71*) (0.58) (4.15**) (7.18***) (5.68**)
ROE it -0.03 -0.23 -0.16 -0.416 -0.62 -0.06 -0.01
(0.73) (0.22) (0.52) (0.15) (2.96*) (0.63) (0.003)
Debt_Ratio it -2.25 -2.23 -3.04 -1.864 -2.74 -2.47 -2.73
(12.21***) (8.91***) (14.30***) (1.63) (1.96) (15.87***) (13.25***)
Current_Ratio it -0.19 -0.18 -0.26 -0.576 -0.78 -0.13 -0.1
(1.17) (0.49) (2.30) (0.59) (1.05) (1.98) (0.84)
Receivable it -2.54 -2.45 -2.96 -2.154 -3.13 -0.81 -0.03
(1.64) (6.02***) (6.71***) (1.06) (2.69) (0.72) (0.001)
Inventory it 1.47 1.14 1.85 1.08 0.88 1.7 1.93
(2.44) (2.64) (4.32**) (1.86) (4.67**) (5.96***) (5.13**)
Loss it -1.26 -1.35 -1.29 -1.67 -1.84 -1.49 -1.57
(9.52***) (29.74***) (22.62***) (8.94***) (7.25***) (46.44***) (36.82***)
Ownership it 1.35 1.23 1.8 0.964 0.77 1.35 1.16
(2.78*) (4.72**) (7.83***) (2.20) (0.55) (3.59*) (2.08)
Age it -0.19 -0.21 -0.18 -0.58 -0.78 -0.1 -0.09
(12.39***) (9.58***) (4.06**) (11.81***) (10.16***) (14.69***) (9.35***)
Audit_Fee it -0.14 -0.15 -0.28 -0.53 -0.73 -0.25 -0.13
(0.44) (0.17) (0.30) (0.25) (2.67) (0.95) (0.19)
B share it -0.03 0.15 0.24 -0.03 -0.62 -0.3 -0.58
(0.39) (0.58) (0.74) (0.20) (2.61) (0.32) (1.14)
Year dummy Yes Yes Yes Yes Yes Yes Yes
Indus dummy Yes Yes Yes Yes Yes Yes Yes
Psuedo R-square 0.08 0.08 0.07 0.04 0.05 0.09 0.09
Obs. No 8617 3559 3559 1204 1204 3854 3854
Note: The dependent variable is subsequent audit opinion (Opinion it+1), which is defined as 1 if audit opinion is favourable and 0
otherwise. The independent variable is PC it, which is defined as 1 if a CEO is politically connected (refer to as definition in
Section 3) and 0 otherwise, or Local_PC it, which is defined as 1 if a CEO is politically connected to local government (refer to
as definition in Section 3) and 0 otherwise. NSOE is a dummy variable equal to 1 if a firm is not owned by government. See
Table 2 for other variable descriptions. The year dummy variables and industry dummy variables are also included. The number
in the parenthesis below is Wald Chi-square value. *, **, and *** indicate significance at 10%, 5%, and 1% level, respectively.
5.3 Political connections, subsequent favourable opinion and the level of market development
Fan et al. (2011) have developed a comprehensive index (MINDEX) to measure the strength
of market forces across regions (provinces) in every year in China. They score regional
market development from various perspectives including local legal enforcement, local
government intervention, development of non-state business and product market competition
in terms of regional trade barriers[12]. In column (1) of Table 5, we run model (2) including
the institutional factor MINDEX and its interaction with CEO’s political connections (PC).
The results show that the likelihood of a favourable opinion in year t+1 is significantly
associated with political connections in year t (coefficient = 0.08, Wald Chi-statistic = 3.04, p
< 0.10). The interaction of political connections and MINDEX in year t has a significantly
negative association with the likelihood of a favourable opinion in year t+1 (coefficient = -
0.12, Wald Chi-statistic = 10.07, p < 0.01). This result suggests that the CEO’s political
connections lead to more favourable audit opinions only in regions with a lower level of
economic development.
To further test this association, we divide the country into three areas based on the level
of economic development: eastern, middle, and western. The eastern area has the highest
level of economic development, whereas the western area has the lowest level[13]. The
middle region has more average levels of development. We re-estimate regression model (1)
in three samples partitioned on these areas. Column (4c) of table 5 shows that the likelihood
of a favourable opinion in year t+1 is positively and significantly associated with political
connections in year t in the subsample of Western China where the economy is less
developed (coefficient = 0.09, Wald Chi-statistic = 4.45, p < 0.05). In contrast, Column (4a)
shows that political connections in year t have a significantly negative impact on the
likelihood of a favourable opinion in year t+1 in Eastern China, suggesting that auditors may
regard CEO’s political connections as agency costs in regions with higher level of economy
development (coefficient = -0.04, Wald Chi-statistic = 2.98, p < 0.10). The association
between CEO’s political connections and audit opinions is not significant in the middle area
of China (see column (4b)). This may be because the middle area has a mix of relationship-
based and market-based forces. Taken together, the results suggest that the effect of CEO’s
political connections on audit opinions is stronger in less developed regions where markets
rely more on relationships and there is a lack of market competition and investor protection.
Hypothesis H3 is therefore supported.
5.4 Political connections, subsequent favourable opinion and the demand for information in
contracting
In column (2) of Table 5, we run model (2) including the institutional factor NSB, the number
of non-state owned banks in a region, and its interaction with CEO’s political connections
(PC). The results show that the likelihood of a favourable opinion in year t+1 is not
significantly associated with political connections in year t. The number of non-state owned
banks (NSB) in the region is negatively associated with the likelihood of a favourable opinion
(coefficient = -0.17, Wald Chi-statistic = 3.15, p < 0.10), suggesting that auditors are less
likely to issue favourable opinions in regions with more non-state owned banks that may
demand more credible information and accounting numbers for debt contracting. The
interaction of political connections and NSB in year t has a significant negative association
with the likelihood of a favourable opinion in year t+1(coefficient = -0.24, Wald Chi-statistic
= 3.07, p < 0.10). This result suggests that the influence of CEO’s political connections on
audit opinions is weakened by the information needs of accounting-based debt contracts. The
results support hypothesis H4 that audit opinions are less influenced by political connections
in regions with more non-state-owned banks where markets have more contractual demand
for credible information.

5.5 Political connections, subsequent favourable opinion and the penalty of political
corruption and accounting scandals
In column (3) of Table 5, we run model (2) including the institutional factor PENALTY, the
number of firms caught and penalised by the CSRC for political corruption and accounting
scandals in a region, and its interaction with CEO’s political connections (PC). The results
show that the likelihood of a favourable opinion in year t+1 is significantly positively
associated with political connections in year t (coefficient = 0.36, Wald Chi-statistic = 4.32, p
< 0.05). This result is consistent with previous results that the CEO’s political connections
could help firms obtain favourable opinions. The number of firms caught in political
corruption and accounting scandals in the region is negatively associated with the likelihood
of a favourable opinion (coefficient = -0.33, Wald Chi-statistic = 3.31, p < 0.10), suggesting
that auditors are less likely to issue favourable opinions in regions where more firms are
caught and penalised in scandals. The interaction of political connections and PENALTY in
year t has a significant negative association with the likelihood of a favourable opinion in
year t+1 (coefficient = -2.4, Wald Chi-statistic = 8.79, p < 0.01). This result suggests that the
influence of CEO’s political connections on audit opinions is weakened by the potential risk
for relationship-based economic behaviours, providing support for Hypothesis H5.
Table 5 Logistic regression of subsequent audit opinions, CEO’s political connections and
institutional factors.
(1) (2) (3) (4) Partitioned by AREAs
Independent variables MINDEX NSB PENALTY (a) East (b) Middle (c) West
Intercept 0.15 0.03 1.38 0.077 0.05 0.63
(0.70) (0.58) (0.98) (0.22) 0.89) (0.70)
PC it 0.08 -0.04 0.36 -0.04 0.003 0.09
(3.04*) (0.15) (4.32**) (2.98*) (0.99) (4.45**)
Institutional_Factor itr -0.05 -0.17 -0.33
(1.03) (3.15*) (3.31*)
PC it* Institutional Factor itr -0.12 -0.24 -2.40
(10.07***) (3.07*) (8.79***)
Size it -0.49 -0.61 -0.77 0.65 0.25 0.29
(8.24***) (3.36**) (6.96***) (7.32***) (0.28) (5.43**)
ROE it 0.03 -0.15 0.31 0.85 1.64 0.55
(0.63) (0.51) (1.91) (9.15***) (0.65) (0.74)
Debt_Ratio it -0.14 -0.28 -0.52 -8.30 -7.03 -7.09
(0.31) (0.19) (0.07) (0.75) (0.83) (0.38)
Current_Ratio it -0.18 -0.30 -0.56 -1.99 -1.13 -0.99
(0.94) (1.06) (1.32) (1.65) (1.40) (1.63)
Receivable it -0.008 -0.02 -0.04 -0.28 0.21 -0.45
(0.77) (0.89) (1.15) (0.81) (0.71) (1.07)
Inventory it 0.00 0.12 0.38 1.14 1.08 1.7
(1.23) (0.35) (0.61) (2.17) (2.01) (3.98**)
Loss it -0.17 0.05 -1.45 0.016 0.08 -0.07
(5.08**) (3.20*) (11.36***) (14.9***) (7.24***) (35.70***)
Ownership it 0.25 0.49 0.03 0.24 0.38 0.15
(0.92) (2.92*) (1.30) (1.58) (2.14) (2.93*)
Age it -1.13 -0.25 -0.41 -0.09 -0.07 -0.22
(0.56) (0.68) (0.64) (1.57) (1.67) (1.48)
Audit_Fee it 0.02 0.14 1.30 -0.27 -0.65 -0.37
(0.58) (0.70) (2.98*) (0.97) (0.68) (0.27)
B share it 0.14 0.26 0.42 0.03 -0.25 -0.42
(0.24) (0.12) (0.52) (0.73) (0.66) (1.09)
Year dummy Yes Yes Yes Yes Yes Yes
Industry dummy Yes Yes Yes Yes Yes Yes
Psuedo R-square 0.09 0.08 0.10 0.07 0.05 0.09
Obs. No 8617 8617 8617 3016 2240 3361
Note: The dependent variable is subsequent audit opinion (Opinion it+1), which is defined as 1 if audit opinion is
favourable and 0 otherwise. The independent variable is PC it, which is defined as 1 if a CEO is politically
connected (refer to as definition in Section 3) and 0 otherwise. See Table 2 for other variable descriptions. MINDEX
is a marketization index reported by Fan et al. (2011). NSB is defined as the number of non-state owned banks
(including foreign banks) in a region (based on provinces) of China. Penalty is defined as the number of firms caught
in both political corruption and accounting scandals by CSRC in a region (based on provinces) of China. The year
dummy variables and industry dummy variables are also included. The number in the parenthesis below is Wald
Chi-square value. *, **, and *** indicate significance at 10%, 5%, and 1% level, respectively.
6. Robustness tests and additional analysis
6.1 Change in political connections and change in opinions
To control for change in firm characteristics in the current year and self-selection problems,
we follow prior studies and use the “difference-in-difference” model (Chan et al., 2006)[14].
We run a logistic regression for the association between the change in political connections
and change in audit opinions. The dependent variable is a dummy variable equal to 1 if a firm
previously receiving a qualified opinion receives a favourable opinion in the current year, and
0 otherwise. The main independent variable is a dummy variable equal to 1 if a firm which
was not previously politically connected forms political connections in the current year, and 0
otherwise. Various control variables measuring changes in the firm’s characteristics are
included. The results (untabulated) show that the likelihood of changing to favourable
opinions in year t is significantly positively associated with political connections newly
formed in year t (p < 0.05). Overall, the findings provide further support for Hypothesis H1
since a firm forming new political connections is more likely to receive a favourable opinion
in the current year in contrast to a qualified opinion received in the prior year.
6.2 Politically connected CEOs and politically connected chairpersons
We retest the relationship between a CEO’s political connections and audit opinions based on
the sample including both board chair and CEO. Under Chinese corporate law, the chair is
the legal representative of the company. Therefore, this person is endowed with the highest
level of authority in the firm and bears the overall responsibility for firm operations. In most
cases, the chair is also the highest paid employee. For these reasons, some studies regard the
position of chair, rather than that of CEO, as the top management post in a firm (e.g., Firth et
al., 2006; Liao et al., 2009). Hence, we redefine our dependent variable in model (1) as a
dummy variable equal to 1 if either the chair or CEO has political connections. The
untabulated regression results are consistent with the results reported in Table 3.
6.3 Auditors’ features and audit opinions
It is well documented that audit opinions are influenced by audit firm attributes. For example,
DeAngelo (1981) has established a positive association between auditor size and audit quality.
Some Chinese studies find that local and small auditors are more likely to issue unqualified
opinions (Chan et al., 2006, Wang et al., 2008 and Chan et al., 2012). Studies also find that
companies use auditor switching in order to avoid receiving unfavourable audit reports
(Chow and Rice, 1982; DeAngelo, 1981; Haskins and Williams, 1990; Schwartz and Menon,
1985 and Smith, 1986). Following these prior studies, we further examine the influence of
political connections on audit opinions with features such as audit firm size, audit firm
locality and auditor switching.

In column (1) of Table 6, we partition the sample by firms audited by Big Ten and non-
Big Ten auditors[15]. Consistent with prior studies indicating that small auditors dominate
China’s audit market (Chan et al., 2012; Chan and Wu, 2011), our results show that 77
percent of firms are audited by non-Big Ten firms in our sample. The results in columns (1a)
to (1d) show that the likelihood of a favourable opinion in year t+1 is significantly positively
associated with political connections (p < 0.10) and political connections to local government
(p < 0.05) in year t in both of the subsamples Overall, our findings show that politically
connected firms are more likely to receive favourable audit opinions in the subsequent period
regardless of whether they are audited by Big Ten or non-Big Ten auditors. This suggests that
the association between CEO’s political connections and favourable opinions is not affected
by audit firm size.
In column (2) of Table 6, we partition the sample by firms audited by local auditors and
non-local auditors. The results show that 57 percent of firms are audited by local auditors in
our sample. Columns (2a) and (2c) show that the likelihood of a favourable opinion in year
t+1 is significantly positively associated with political connections in year t in both
subsamples (p < 0.10). Further, columns (2b) and (2d) show that the likelihood of a
favourable opinion in year t+1 is significantly positively associated with political connections
to local government in year t in both subsamples (p < 0.05 for local audit clients and p < 0.10
for non-local audit clients). The results suggest that politically connected firms are more
likely to obtain a favourable audit opinion regardless of the locality of auditors.
In column (3) of Table 6, we partition the sample on the basis of whether firms switch or
retain their auditor. The table shows that 59 percent of firms in our sample retain their
incumbent auditors. The likelihood of a favourable opinion in year t+1 is positively but not
significantly associated with political connections or connections to local government in year
t in the subsample of firms that switch auditors (see columns (3a) and (3b). The likelihood of
a favourable opinion in year t+1 is significantly positively associated with political
connections and connections to local government in year t in the subsample of no auditor
switch (p < 0.10). Overall, our findings show that politically connected firms that retain their
incumbent auditors are more likely to receive favourable audit opinions in the subsequent
period. This suggests that politically connected firms may not select or switch to a new
auditor to obtain more favourable opinions, providing evidence that the favourable opinions
might be obtained in the absence of auditor switching.

Table 6: Logistic regression of subsequent audit opinions and political connections in a subsample of firms partitioned by auditors’ features
including Big Ten audit firms versus non- Big Ten audit firms, localauditors versus non-local auditors and auditor switching versus no auditor
switching.
(1) (2) (3)
Independent var. Big Ten Non-Big Ten Local Auditor Not Local Auditor Auditor Switch No Auditor Switch
(a) (b) (c) (d) (a) (b) (c) (d) (a) (b) (c) (d)
Intercept 7.46 8.09 0.39 -1.36 -1.76 -3.5 1.98 0.28 -2.96 -4.93 1.29 -0.04
(0.001) (0.002) (0.03) (0.3) (0.31) (0.87) (0.86) (0.01) (0.86) (1.73) (0.32) (0.00)
PC it 0.87 0.11 0.29 0.27 0.03 0.33
(2.92*) (3.48*) (3.27*) (2.92*) (0.02) (2.85*)
Local_PC it 0.27 0.74 1.16 0.25 0.69 0.6
(4.05**) (4.62**) (5.58**) (3.29*) (1.97) (2.87*)
Size it 0.21 0.2 0.33 0.35 0.34 0.44 0.13 0.23 0.23 0.35 0.26 0.26
(1.17) (0.61) (11.04***) (10.71***) (5.09**) (6.11***) (1.61) (3.62*) (2.45) (4.11**) (5.57**) (4.64**)
ROE it 0.06 0.92 -0.06 -0.04 -0.51 -0.12 0.91 -0.03 -0.01 -0.01 -0.02 -0.13
(0.19) (0.73) (1.12) (0.34) (0.15) (0.08) (1.14) (0.3) (0.07) (0.02) (0.81) (0.38)
Debt_Ratio it -3.3 -3.91 -1.84 -2.23 -0.57 -2.65 -0.84 -2.64 -1.42 -1.91 -2.67 -3.4
(6.35***) (4.05**) (16.10***) (19.68***) (4.59**) (10.34***) (15.46***) (16.38***) (3.32*) (4.65**) (24.67***) (28.47***)
Current_Ratio it 0.44 0.67 -0.06 -0.06 0.02 -0.07 0.19 -0.01 0.14 0.02 -0.11 -0.13
(1.68) (1.26) (1.16) (1.22) (0.08) (1.24) (3.15*) (0.01) (0.12) (0.02) (2.04) (2.26)
Receivable it -2.34 -5 -1.48 -1.01 -1.57 -1.37 -3.01 -2.09 -2.15 -2.39 -1.55 -1.21
(2.56) (6.50***) (3.81**) (1.57) (2.42) (1.31) (15.22***) (4.89**) (3.97**) (4.01**) (3.35*) (1.73)
Inventory it 2.72 5.35 1 1.22 2 4.65 -0.5 0.64 2.21 3.5 0.97 1.28
(2.66*) (3.54*) (3.39*) (3.82**) (4.92**) (12.93***) (0.72) (0.77) (5.61**) (7.11***) (2.31) (3.27*)
Loss it -0.64 -0.006 -1.49 -1.48 -1.57 -1.17 -1.25 -1.43 -1.51 -1.54 -1.25 -1.29
(2.12) (0.001) (79.10***) (66.41***) (25.69***) (12.39***) (34.22***) (46.44***) (35.34***) (28.83***) (40.78***) (32.15***)
Ownership it 1.59 1.45 1.36 1.64 1.59 2.05 1.46 1.49 0.82 1.11 2.02 2.02
(2.41) (1.31) (7.77***) (9.36***) (5.10**) (6.18***) (7.61***) (5.88**) (1.41) (1.98) (11.75***) (10.00***)
Age it -0.07 -0.12 -0.1 -0.07 -0.11 -0.08 -0.1 -0.08 -0.15 -0.13 -0.06 -0.04
(2.80*) (3.72**) (22.97***) (11.37***) (17.01***) (4.78**) (22.54***) (9.98***) (27.60***) (16.68***) (5.94***) (2.81*)
Audit_Fee it 0.2 0.11 -0.3 -0.26 -0.22 -0.34 -0.11 -0.09 0.13 0.01 -0.25 -0.16
(0.39) (0.08) (2.04) (1.34) (0.61) (1.05) (0.35) (0.15) (0.23) (0.001) (1.37) (0.5)
B share it -0.34 -0.2 0.46 0.2 -0.27 -0.05 -0.01 -0.05 0.5 0.25 -0.15 -0.25
(0.48) (0.12) (0.9) (0.16) (0.33) (0.01) (0.001***) (0.01) (0.23) (0.11) (0.16) (0.41)
Year dummy Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Indus dummy Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes
Psuedo R-square 0.08 0.1 0.06 0.06 0.08 0.07 0.09 0.07 0.08 0.08 0.06 0.07
Obs. No 1970 1970 6647 6647 4875 4875 3742 3742 3552 3552 5065 5065
Note: The dependent variable is subsequent audit opinion (Opinion it+1), which is defined as 1 if audit opinion is favourable and 0 otherwise. The independent variable is PC it, which is defined as 1 if a CEO
is politically connected (refer to as definition in Section 3) and 0 otherwise, or Local_PC it, which is defined as 1 if a CEO is politically connected to local government (refer to as definition in Section 3) and 0
otherwise. See Table 2 for other variable descriptions. The year dummy variables and industry dummy variables are also included. The number in the parenthesis below is Wald Chi-square value. *, **, and
*** indicate significance at 10%, 5%, and 1% level, respectively.
7. Policy implications
Our results have important implications for regulators, practitioners and investors. The
development of stock markets in China in the past two decades has created a demand for
greater auditor independence. In response to this demand, the Chinese government launched
a series of audit market reforms including transforming the organisation of audit firms and
adopting international auditing standards. While these reforms have led to considerable
advancement in accounting and auditing practices, our results suggest that the unique
political and institutional factors in China still have an important influence on audit
independence. Our findings indicate that relationship plays an important role in contracting in
China’s emerging market. Further, the quality of the auditing profession may be
compromised by the economic environment such as the level of market development. Hence,
there may be a need to increase the monitoring of audit independence, particularly in the
relatively less developed regions.
8. Conclusion
In response to calls for research on political influence and the audit market (Wang et al.,
2008), we examine the importance of corporate insiders’ political connections to audit
opinions. Based on a sample of listed firms in China, we find that CEOs’ political
connections are positively associated with subsequent favourable opinions. This political
influence appears to be stronger in non-SOEs which lack political ties with government.
Moreover, the association between CEO’s political connections and audit opinions is less
pronounced in regions with a more developed market and a better information environment
with stronger investor protection, stronger demand for credible information for contracting
and greater penalties for relationship-based contracts.

Our study has several limitations. For example, our data period is restricted to the period
2005 to 2010 and may not be generalisable to other periods. The overall strength of our
models is quite low and some of our significance levels are marginal (10 per cent). The
models may be strengthened by using alternative measures for some of our control variables
or including additional control variables. Alternatively, expanding our definition of political
connections may strengthen our results. For example, our measure of CEO’s political
connections does not include other types of political influence such as corporate political
connections by donations or contributions to governments. Broadening the time period of our

study and strengthening our models are opportunities for further research. Further, testing
whether these various political connections facilitate opinion shopping or reduce audit risks
in different levels of economic and institutional environments are additional opportunities for
future research.
Notes
1. A ‘favourable opinion’ means a non-modified opinion. A firm favours a non-modified opinion

issued by the auditor because it indicates that its financial statements give a true and fair view
and have been prepared in accordance with accounting standards.
2. In China, the market is less developed and investor protection is poorer in Western and inland
areas whilst the market is more competitive and developed in Eastern China.
3. The market share of the Big Ten audit firms, which are assumed to provide better quality and
more independent audits, has been declining (DeFond et al., 1999).
4. In accordance with prior studies (Chan et al., 2012; Wang et al., 2008), SOEs have a stronger
tendency to select small local auditors since those auditors are more likely to issue favourable
opinions under political influence of government owners.
5. Qualified opinions are typically issued for events such as uncertainty about asset values,
limitations of scope, and financial distress/going concern issues. The events and transactions for
disclaimer or adverse opinions are similar to those of qualified opinions but with more pervasive
and material effects.
6. As indicated in the official guidelines, events and transactions for unqualified opinions with
explanatory paragraphs are those which are important to financial statement users but do not
have a direct influence on the financial statements. The events include uncertainty about asset
values, related party transactions, and financial distress/going concern issues.
7. In accordance with Chinese regulations, listed companies are required to achieve a target
profitability level in order to raise additional capital through a rights issue. Further, if they report
losses for three consecutive years, they will be de-listed.
8. The link to the website is http://finance.sina.com.cn/stock/.
9. The SASAC’s mandate does not cover financial institutions. The supervision of financial
institutions is the responsibility of the Central Banking Supervisory Committee (CBRC). In
financial institutions, ownership and regulatory functions are more clearly separated.
10. Our regression diagnostics indicate that some of our control variables are not normally
distributed. Eliminating some outliers improved the strength of our models in terms of the
overall r squares but had no material impact on our reported results. In order to maximise our
sample size, we have therefore chosen to retain all observations.
11. We have also examined the contemporaneous relationship between political connection and
audit opinion but results are not significant. The most likely reason for this is that establishing
political connections has a more pronounced effect on subsequent audit opinions.
12. This index assesses the relative progress in marketisation of Chinese districts using a
comparative method, considering 23 basic indicators in five fields. The index has been
sponsored by the National Economic Research Institute (NERI) and the China Reform
Foundation. The index captures: (1) the development of market intermediaries based on the ratio
of the number of lawyers and registered accountants to local population; (2) the protection of the
legal rights of firms based on a nation-wide survey on the frequency of local economic crimes
(scaled by local GDP) and managers’ ratings on local investor protection; (3) intellectual
property rights (IPRs) enforcement based on the total number of patents applied for and
approved (adjusted by the number of engineers in the region); and (4) consumer rights protection
based on the frequency of consumer complaints received by the Consumer Association of China
(adjusted by local GDP).
13. The eastern region includes the eleven provinces of Beijing, Fujian, Guangdong, Hainan, Hebei,
Jiangsu, Liaoning, Shandong, Shanghai, Tianjin, and Zhejiang. The middle region is composed
of eight provinces: Anhui, Hei Longjiang, Henan, Hubei, Hunan, Jiangxi, Jilin, and Shanxi. The
western region consists of twelve provinces, namely, Chongqing, Gansu, Guangxi, Guizhou,
Inner Mongolia, Ningxia, Qinghai, Shaanxi, Sichuan, Xinjiang, Xizang, and Yunnan.
14. In the “difference-in-difference” model, all variables are transformed as changes between
current and prior period, namely, Xit-Xit-1. This method is used to control for changes in firm
characteristics (Chan et al., 2006).
15. As Big Four auditors have merged with domestic auditors in China, Big Ten auditors based on
total revenues or market share are more often used in research as a proxy for reputation and
quality of auditors in China.
References
Bliss, Mark A. and Gul, Ferdinand A. (2012), “Political connection and cost of debt: Some
Malaysian evidence”, Journal of Banking & Finance, Vol. 36 (5), pp. 1520 –1527.
Bushman, R. and Piotroski, J. (2006), “Financial Reporting Incentives for Conservative
Accounting: The Influence of Legal and Political Institutions”, Journal of Accounting
and Economics, Vol. 42, pp. 107-148.
Bushman, R., Piotroski, J. and Smith, A. (2004), “What Determines Corporate
Transparency?”, Journal of Accounting Research, Vol. 42, pp. 207-252.

Chan, K.H. and Wu, D. (2011), “Aggregate Quasi Rents and Auditor Independence:
Evidence From Audit Firm Mergers in China”, Contemporary Accounting Research,
Vol. 28, No. 1, pp. 175-213.
Chan, K.H., Lin, K.Z. and Mo, P.L. (2006), “A Political-economic Analysis of Auditor
Reporting and Auditor Switches”, Review of Accounting Studies, Vol. 11, pp. 21-48.
Chan, K.H., Lin, K.Z. and Wang, R. (2012), “Government Ownership, Accounting-based
Regulations, and the Pursuit of Favourable Audit Opinions: Evidence from China”,
Auditing: A Journal of Practice & Theory, Vol. 31, No. 2, pp. 47-64.
Chaney, P.K., Faccio, M. and Parsley, D. (2011), “The Quality of Accounting Information in
Politically Connected Firms”, Journal of Accounting and Economics, Vol. 51, No. 1-2,
pp. 58-76.
Charumilind, C., Kali, R. and Wiwattanakantang, Y. (2006), “Connected Lending: Thailand
Before the Financial Crisis”, Journal of Business, Vol. 79, No. 1, pp. 181-218.
Chen, J.P., S. Chen and Su, X. (2001), “Profitability Regulation, Earnings Management and
Modified Audit Opinions: Evidence from China”, Auditing: A Journal of Practice &
Theory, Vol. 20, pp. 9–30.
Chow, C.W. and Rice, S.J. (1982), “Qualified Audit Opinions and Auditor Switching”, The
Accounting Review, Vol. 57, pp. 326–335.
Claessens, S., Feijen, E. and Laeven, L. (2008), “Political Connections and Preferential
Access to Finance: The Role of Campaign Contributions”, Journal of Financial
Economics, Vol. 88, No. 3, pp. 554-80.
Copley, P.A. and Douthett, E.B. (2002), “The Association Between Auditor Choice,
Ownership Retained, and Earnings Disclosure by Firms Making Initial Public
Offerings”, Contemporary Accounting Research, Vol. 19, No. 1, pp. 49–75.
Datar, S.M., Feltham, G.A. and Hughes, J.S. (1991), “The Role of Audits and Audit Quality
in Valuing New Issues”, Journal of Accounting and Economics, Vol. 14, No. 1, pp. 3-
49.
DeAngelo, L. (1981), “Auditor Size and Audit Quality”, Journal of Accounting and
Economics, Vol. 3, pp. 183–199.
DeFond, M.L., Wong, T.J. and Li, S. (2000), “The Impact of Improved Auditor
Independence on Auditor Market Concentration in China”, Journal of Accounting and
Economics, Vol. 28, pp. 269-305.
Dinç, S. (2005), “Politicians and Banks: Political Influences on Government-owned Banks in
Emerging Markets”, Journal of Financial Economics, Vol. 77, pp. 453-479.
Dyck, A. and Zingales, L. (2004), “Private Benefits of Control: An International
Comparison”, Journal of Finance, Vol. 59, pp. 537-600.
Faccio, M. (2006), “Politically Connected Firms”, American Economic Review, Vol. 96, pp.
369-386.
Faccio, M., Masulis, R. and McConnell, J.J. (2006), “Political Connections and Corporate
Bailouts”, Journal of Finance, Vol. 61, pp. 2597-2635.
Fan, G., Wang, X. L. and Zhu, H. P. (2011), “Marketization indices of China: The relative
progress of regional marketization indices in 2011”, China: Economic Science Press.

Fan, J.P.H. and Wong, T.J. (2005), “Do External Auditors Perform a Corporate Governance
Role in Emerging Markets? Evidence from East Asia”, Journal of Accounting Research,
Vol. 43, No. 1, pp. 35-72.
Fan, J.P.H., Wong, T.J. and Zhang, T. (2007), “Politically Connected CEOs, Corporate
Governance and Post-IPO Performance of China’s Newly Partially Privatized Firms”,
Journal of Financial Economics, Vol. 84, pp. 265-590.
Firth, M., Fung, P. and Rui, O.M. (2006), “Firm Performance, Governance Structure, and
Top Management Turnover in a Transitional Economy”, Journal of Management

Studies, Vol. 43, pp. 1289-1330.
Francis, J.R., Khurana, I.K. and Pereira, R. (2005), “Disclosure Incentives and Effects on
Cost of Capital around the World”, The Accounting Review, Vol. 80, No. 4, pp. 1125-
1162.
Guedhami, O., Pittman, J.A. and Saffar, W. (2009), “Auditor Choice in Privatized Firms:
Empirical Evidence on the Role of State and Foreign Owners”, Journal of Accounting
and Economics, Vol. 48, pp. 151-171.
Guedhami, O., Pittman, J.A. and Saffar, W. (2014), “Auditor Choice in Politically Connected
Firms”, Journal of Accounting Research, Vol. 52, No. 1, pp. 107-162.
Gul, F.A. (2006), “Auditors’ Response to Political Connections and Cronyism in Malaysia”,
Journal of Accounting Research, Vol. 44, pp. 931-963.
Haskins, M. and Williams, D.A. (1990), “Contingent Model of Intra-Big Eight Auditor
Changes”, Auditing: A Journal of Practice & Theory, Vol. 9, No. 3, pp. 55-74.
Hay, D.C., Knechel, W.R. and Wong, N. (2006), “Audit Fees: A Meta-analysis of the Effect
of Supply and Demand Attributes”, Contemporary Accounting Research, Vol. 23
(Spring), pp. 141-191.

Healy, P.M. and Palepu, K.G. (2001), “Information Asymmetry, Corporate Disclosure, and
the Capital Markets: A Review of the Empirical Disclosure Literature”, Journal of
Accounting and Economics, Vol. 31, pp. 405 - 440.
Hofstede, G. (2001), Culture’s Consequences: Comparing Values, Behaviors, Institutions,
and Organizations across Nations, 2nd edition, Thousand Oaks: Sage Publications Inc,
The U. S.
Johnson, S. and Mitton, T. (2003), “Cronyism and Capital Controls: Evidence from
Malaysia”, Journal of Financial Economics, Vol. 67, No. 2, pp. 351-82.

Khwaja, A.I. and Mian, A. (2005), “Do Lenders Favor Politically Connected Firms? Rent
Provision in an Emerging Financial Market”, Quarterly Journal of Economics, Vol. 120,
pp. 1371-1411.
Kornai, J. (1980), “Hard and Soft Budget Constraints”, Acta Oeconomia, Vol. 25, pp. 3-4.
Krishnan, J. (1994), “Auditor Switching and Conservatism”, The Accounting Review, Vol. 69,
pp. 200–215.
Krishnan, J. and Stephens, R. (1995), “Evidence on Opinion Shopping from Audit Opinion
Conservatism”, Journal of Accounting and Public Policy, Vol. 14, pp. 179–201.
Krishnan, J. and Stephens, R. (1996), “The Simultaneous Relation between Auditor
Switching and Auditor Opinion: an Empirical Analysis”, Accounting and Business
Research, Vol. 26, pp. 224-236.
La Porta, R., Lopez-de-Silanes, F. and Shleifer, A. (1999), Corporate Ownership around the
World”, Journal of Finance, Vol. 54, No. 2, pp. 471-517.
La Porta, R., Lopez-de-Silanes, F. and Shleifer, A. (2006), “What Works in Securities
Laws?”, Journal of Finance, Vol. 61, pp. 1-32.
Lennox C. (2000), “Do Companies Successfully Engage in Opinion Shopping: Evidence
from the UK?”, Journal of Accounting and Economics, Vol. 29, pp. 321-37.
Lennox, C. (2003), Opinion Shopping and the Role of Audit Committees when Audit Firms
are Dismissed: the US Experience, The Institute of Chartered Accountants of Scotland,
Scotland.
Leuz, C. and Oberholzer-Gee, F. (2006), “Political Relationships, Global Financing, and
Corporate Transparency”, Journal of Financial Economics, Vol. 81, pp. 411-439.
Li, C., F.M. Song and Wong, M.L. (2004), “Audit Firm Size Effects in China's Emerging
Audit Markets”, working paper, Hong Kong University.
Li, H., Meng, L., Wang, Q. and Zhou, L. (2008), “Political Connections, Financing and Firm
Performance: Evidence from Chinese Private Firms”, Journal of Development
Economics, Vol. 87, pp. 283-299.
Li, S. and Wu, X. (2003), “Supplementary Audit Patterns and Auditor Independence:
Evidence from China’s B Share Market”, China Accounting and Finance Review, Vol.
5, No. 3, pp. 72–102.
Liao, G., Chen, X., Jing, X. and Sun, J. (2009), “Policy Burdens, Firm Performance, and
Management Turnover”, China Economic Review, Vol. 20, pp. 15-28.
Mobarak, A.M. and Purbasari, D.P. (2006), “Corrupt Protection for Sale to Firms: Evidence
from Indonesia”, working paper, University of Colorado at Boulder.
Piotroski, J.D. and Wong, T.J. (2010), “Institutions and Information Environment of Chinese
Listed Firms”, working paper, NBER, the U.S.
Piotroski, J.D., Wong, T.J. and Zhang, T. (2015), “Political Incentives to Suppress Negative
Financial Information: Evidence from State-controlled Chinese Firms”, Journal of
Accounting Research, Vol. 53, pp. 405–459.
Sapienza, P. (2004), “The Effects of Government Ownership on Bank Lending”, Journal of
Financial Economics, Vol. 72, No. 2, pp. 357-384.
Schwartz, K.B. and Menon, K. (1985), “Auditor Switches by Failing Firms”, The Accounting
Review, Vol. 60, pp. 248– 261
Securities and Exchange Commission (1988), Disclosure Amendments to Regulation S-K,
Form 8-K and Schedule 14A Regarding Cchanges in Accountants and Potential
Opinion Shopping Situations, SEC Financial Reporting Release No.31, Washington,
D.C..
Simunic, D.A. (1980), “The Pricing of Audit Services: Theory and Evidence”, Journal of
Accounting Research, Vol. 18 (Autumn), pp. 161-190.
Smith, D.B. (1986), “Auditor ‘Subject to’ Opinions, Disclaimers, and Auditor Changes”,
Auditing: A Journal of Practice & Theory, Vol. 6, pp. 95–108.
Teoh, S. (1992), “Auditor Independence, Dismissal Threats, and the Market Reaction to
Auditor Switches”, Journal of Accounting Research, Vol. 30, pp. 1-23.
Wang, Q., Wong, T.J. and Xia, L. (2008), “State Ownership, Institutional Environment, and
Auditor Choice: Evidence from China”, Journal of Accounting and Economics, Vol. 46,
No. 1, pp. 112-134.
Watts, R. (2006), “What Has the Invisible Hand Achieved?”, working paper, Massachusetts
Institute of Technology, the U. S.
Whiting, S. (2001), Power and Wealth in Rural China: The Political Economy of Institutional
Change, Cambridge University Press, Cambridge.
Wilkins, M.S. (1997), “Technical Default, Auditors” Decisions and Future Financial
Distress”, Accounting Horizons, Vol. 11, No. 4, pp. 40–48.
Willenborg, M. (1999), “Empirical Analysis of the Economic Demand for Auditing in the
Initial Public Offerings Market”, Journal of Accounting Research, Vol. 37, No. 1, pp.
225–238.
Yang, Z. (2013), “Do Political Connections Add Value to Audit Firms? Evidence from IPO
Audits in China”, Contemporary Accounting Research, Vol. 30, No. 3, pp. 891-921.
Zheng, Jun, Lin, Z., Peng, L. and Zhang, T. (2010), “Can Political Connection Realize
Opinions Shopping: a Perspective of Investors Protection”, Journal of Finance and
Economics (in Chinese), Vol. 36, No. 11, pp. 104-114.


Opinion Audit

Uploaded by

Copyright:

Available Formats

Opinion Audit

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Opinion Audit

Uploaded by

Copyright:

Available Formats

Pacific Accounting Review

CEO’s Political Connections, Institutions and Audit Opinions

visit www.emeraldinsight.com/authors for more information.

*Related content and download information correct at time of download.

Design/methodology/approach –A politically connected firm is defined as a firm in which

connections on audit opinions. We extend the research on auditing in emerging markets by

Keywords Political connections, Audit Opinions, State-owned enterprises (SOE), China

Paper type Research paper

institutional differences. Moreover, it is unclear whether political influence varies with

existence of government ownership and other market characteristics?

positively associated with a CEO’s political connections[1]. This positive association is

robust when we use a “difference-in-difference” model to control for self-selection problems.

insiders with political connections wish to suppress information on actual economic

This is because auditors have greater economic dependence on political connections in

switch their auditors.

political influence. We attempt to provide evidence on the effect of a CEO’s political

connections on this issue.

possibility of contamination due to omitted variables resulting from country differences

implications of our results while Section 8 contains some concluding comments.

2. Background and hypotheses development

2.1 Political influence, audit profession and auditor quality in China

local government led to a lack of independence as well as to regional protectionism. Between

facilitation of mergers of small audit firms (Wang et al., 2008)[3].

clients are unlikely to switch from auditors they perceive as lenient.

government-owned companies switched from a non-local auditor to a local auditor following

financing or the threat of exchange delisting.

2.2 Hypothesis development

conceal from minority shareholders expropriation-related activities stemming from political

favourable resources and thereby increase their negotiation power in contracting. As

of acceptable errors. Because China’s audit market is heavily influenced by political

arguments in explaining auditor behaviour. We therefore develop the following hypothesis:

opinion than firms without political connections.

pressures on economic agents’ incentives and behaviours. We therefore identify a number of

connections and audit opinions in China.

(i) Existence of government ownership

political connections driven by government ownerships. Agency arguments suggest that

politically connected CEOs in SOEs on external auditors may be diluted by government

but operate in a relationship-based environment, having politically-connected managers helps

(McMillan, 1995). In particular, non-SOEs lack political connections by nature of their

ownership. Therefore, when political connections are established by CEOs in non-SOEs,

relationships with government. These arguments lead to the following hypothesis:

non-SOEs than in SOEs.

(ii) the level of investor protection and market development

strong institutions. In an environment encouraging greater investor protection and fewer

enjoyed by firms with political connections is more pronounced in countries with

interventionist governments and weak protection of property rights. As such, we conjecture

independent of political connections or more associated with information quality disclosed by

firms. Hence we test the following hypothesis:

pronounced in regions with lower (higher) level of market development.

(iii) Demand for information in contracting

disciplinary mechanisms, such as the independence of external auditors, to ensure the

This leads to the following hypothesis:

in regions with more (less) state-owned banks.

(iv) Penalties for accounting scandals and relationship-based contracting

misconduct based on relationships. We therefore test the following hypothesis:

Secretary, Central Committee Member, or Representative of the National People’s Congress.

(based on province in China).

unqualified, unqualified with an explanatory paragraph, qualified, disclaimer, and adverse

an explanatory paragraph are used by auditors in China as an alternative to qualified opinions

Inventory 0.16 0.12 0.18 0.14 -0.02 4.31* -0.02 -7.57*