Corporate Governance Mechanisms and Auditor Choice : Evidence from China

This paper provides empirical evidence on the association between corporate governance mechanisms and the choice of high-quality and low-quality auditors (hereafter “auditor choice”) based on firm’s data from Chinese listed firms between 2007 and 2012. Consistent with most literature, this paper adopts Big 4 auditors as the proxy for high-audit quality. Our findings indicate that certain corporate governance factors, ownership of the largest shareholders, the aggregate of ownership of other large shareholders, percentage of independent directors in the board, dual-list status and marketization, have significant positive influence on auditor choice. More specifically, ownership of the largest shareholders exhibits a U-shape relationship for the auditor choice: in the lower ownership, a Chinese listed firm would prefer a low-quality auditor, while up to the threshold level of ownership, a Chinese listed firm would prefer a high-quality auditor. This paper also explains several possible reasons why Big 4 auditors are still uncommon in Chinese stock market.


Historical Background of Accounting (Auditing) Profession in China
External (or independent) audit (hereafter "audit") is regarded as one of the effective monitoring mechanisms in the global capital markets because in attesting the creditability of accounting information provided by management (the agent), an auditor (another agent) plays an external monitoring role on behalf of the shareholders (the principal) and an audit is an essential component of the corporate governance mechanisms (e.g., Ashbaugh & Warfield, 2003;Fan & Wong, 2005).
After the establishment of the People's Republic of China (hereafter "China") in 1949, the accounting (or auditing) profession in China became non-existent because all enterprises were owned by the state and run by the civil servants.After the open door policy in 1980s, auditing profession has been developed as demands emerged for verification of capital contributions and audits of annual financial statements and tax returns of the Sino-foreign enterprises by non-government employed professionals.The establishment of Shanghai Stock Exchange and Shenzhen Stock Exchange in 1990s facilitated the development of new accounting and auditing standards and accounting profession in China because the annual financial statements of any Chinese listed firm have to be audited by Chinese certified public accountants (Chinese CPAs).The governance of both public and private enterprises by independent auditors has gradually been employed by the government as an important mechanism in transformation of Chinese economy (Lin & Liu, 2009).
Most Chinese accounting firms were initially established or sponsored by government agencies or social institutions (Lin & Liu, 2009).Under such historical reason, the independence and audit quality services were challenged by various user groups, and especially, some scholars find that the audit opinion of financial statements of some local state-owned enterprises (SOEs) and those private firms with political connections and the auditors rarely issued modified audit opinion even though they had found some irregularities (e.g., Lin & Liu, 2009).Liu, Wang and Wu (2011) further find that two types of guanxi, "firm-level connections derived from state ownership and personal connections developed through management with external auditors", have a close association with auditors' independence in China.
Currently, the accounting profession in China is governed by the Chinese Institute of Certified Public Accountants, and only those accounting firms with "The Securities and Futures Related Business License" are eligible to conduct external audits of Chinese listed firms (Note 1).Institutional framework concerning the accounting profession has been well established in line with the international trend, and both civil and criminal responsibilities related to the frauds by the auditors have been established (see Appendix 1).Most audit frauds related to Chinese listed firms are investigated and punished by the China Securities Regulatory Committee (CSRC) (Note 2) (see two examples in Appendix 2).Therefore, the quality and independence of Chinese auditors have been substantially improved and such governance mechanism helps establish a more effective functioning of auditing services in China.Nonetheless, due to the historical, political and economic factors, auditing services in China are dominated by domestic accounting firms for the provision of external audit services to Chinese listed firms, and those accounting firms with international alliance (e.g., Big 4 and international second-tier firms) are still uncommon in Chinese stock market.

Motivation of the Study
There has been a lot of literature concerning the association of corporate governance mechanisms and audit choice in Chinese stock market mostly based on the firm's data prior to 2006 and those studies suggest that Big 4 auditors (Note 3) are the proxy for high-quality, because they are less politically influenced and have greater resources for the provision of auditing services (e.g., Boone, Khurana & Raman, 2010;Lin & Liu, 2009;Li & Luo, 2011).
Since 2006, there has been a substantial reform of institutional framework in China, including: (1) the adoption of International Financial Reporting Standards and International Statements on Auditing into respective standards in China with effect from 2007; (2) the implement of share reform under which ownership of the largest shareholder in a particular Chinese listed firm has been converted from non-tradable shares to tradable ones, and therefore, the largest shareholder can realize its shares for cash, rather than expropriating the listed firm; and (3) CSRC has allowed the foreign institutional investors (e.g., Morgan Stanley, Goldman Sachs) to participate directly in Chinese stock market under Qualified Foreign Institutional Investors (QFII), and therefore, those institutional investors would demand higher standard in corporate governance and enhance the effectiveness of audit monitoring in China.Further, there is a significant expansion of Chinese stock market between 2006 and 2012, in terms of the number of Chinese listed firms and the amount of raised capital, especially that from foreign investors, while the non-tradable ownership in Chinese listed firms have been reduced under the share reform (see Table 1).The change in ownership structure of Chinese listed firms may also lead the change of other corporate governance mechanisms.Accordingly, it is interesting to re-examine the effect of corporate governance mechanisms on the choice of high-quality and low-quality auditors (hereafter "auditor choice") in Chinese stock market.This study focuses on the firm's data from 2007 to 2012, reflecting the substantial reform of institutional framework and the change of Chinese accounting and auditing practices after 2007.The paper is organized as follows: Section 2 reviews relevant prior studies in the literature, Section 3 describes variable and regression model of this study, Section 4 presents the empirical results and interpretation and Section 5 concludes the results of the study.

Literature Review
The separation of ownership (the principal) and management (the agent) can result in opportunistic management behaviours and serve agency problems in firms (Fama & Jensen, 1983) and agency costs are likely to increase due to asymmetric information between managers and dispersed equity stakeholders (Jensen & Meckling, 1976).However, concentrated ownership of Chinese listed firms is common and therefore, the single largest (controlling) shareholders generally serve to monitor management and the demand for audit as a monitoring mechanism may be limited.Because of the protection of minority interest, there are great demands for audits to serve as a monitoring (governance) mechanism, depending on the levels of audit quality (Lin & Liu, 2009).

Big 4 Auditors as High Audit-Quality
Audit quality means "technical aspect -the ability to detect misstatements" and "independent aspect -willing to report the misstatements uncovered in the audit work" (Lee, Stokes, Taylor & Walter, 2003).
Most foreign scholars adopt Big 4 auditors as the proxy for high audit-quality.DeFond, Wong and Li (2000) find that big auditors are more likely to issue the qualified audit opinion in China.Francis (2004) also claims that audits of Big 4 auditors are of higher audit-quality than non-Big 4 auditors because Big 4 auditors can charge higher audit fee for higher audit quality through more audit effort and greater expertise of the auditor.Nevertheless, Boone, Khurana and Raman (2010) find little difference in actual audit quality but a more pronounced difference in perceived audit quality between Big 4 and second-tier firms from 2003 to 2006 in U.S.
However, Chinese auditing market presents an interesting issue for the study of auditor choice because Chinese accounting profession is not only regulated, but also administrated, by government agencies (e.g., Ministry of Finance, CSRC, etc.) (Lin & Liu, 2009).Some Chinese scholars adopt "Top 10" firms (including Big 4 auditors) as high-quality auditors in China (e.g.Lin & Liu, 2009;Li & Luo, 2011).Even though Big 4 auditors are rare in Chinese stock market, several scholars still adopted Big 4 and non-Big 4 auditors to proxy for high-quality and low-quality, respectively, mainly because Big 4 auditors should possess a higher degree of industrial expertise and are less politically influenced by local governments in China (e.g., Chen, Shrome & Su, 2001;Simunic & Wu, 2009;Chen, Su & Wu, 2009;Guedhami, Pittman & Saffar, 2009;Wang & Xin, 2011).
This study also adopts Big 4 auditors as the proxy for high-quality auditors.

Corporate Governance and Auditor Choice
There is a general perception that listed firms have to take a trade-off in their auditor choice decisions, (1) to hire high-quality auditors to signal effective audit monitoring and good corporate governance to lower their capital raising costs, and (2) to select low-quality auditors with less effective audit monitoring in order to recap private benefits derived from weak corporate governance and less-transparent disclosure (the opaqueness gains) (Lin & Liu, 2009).
Ownership structure.There are two controversial issues on the concentrated ownership on the auditor choice.On one hand, some scholars argue that with high ownership concentration, the firms' financial reporting is likely to be opaque due to the incentives for the controlling shareholders' rent-seeking and expropriation (Copley & Douthett, 2002), and because large shareholder would try to maximize their private benefits through tunneling or expropriation of other shareholders (LaPorta, Lopez-De-Silanes Shleifer & Vishny, 2002;Anderson, Kadous & Koonce, 2004).Lin and Liu (2009) further find Chinese listed firms with larger controlling shareholders are less likely to hire high-quality auditors from 2001 to 2004.On the other hand, a controlling shareholder may also introduce effective monitoring mechanisms that restrict his/her abilities to expropriate Chinese listed firms and therefore mitigate the agency conflict (Ang, Cole & Lin, 2000) and firm with such agency problems are more likely to hire Big 4 (previously Big 5) auditors (Fan & Wong, 2005).Further, after share reform, most controlling shareholders of Chinese listed firms can convert their ownership from non-tradable shares to tradable ones, so they can realize their shares of Chinese listed firms for cash.Accordingly those listed firms may have an incentive to hire high-quality auditors for the protection of their interests as well as that of other shareholders.
Other large shareholders may demand high-quality auditors.Leung and Cheng (2013) find that the higher the degree of ownership concentration among other large shareholders, the higher the firm value because the alignment of those large shareholders can challenge the acts of the largest (controlling) shareholders.Therefore, other largest shareholders are assumed to prefer high-quality auditors for the protection of their interests.
Internal management structure.Agency theory assumes that the directors and top executives, as agents, do not prefer an effective monitoring mechanism to them and an opaque financial reporting system.Xie, Davidson III and DaDalt (2003) suggest that board size and percentage of independent directors in the board can be used as proxies of strength of governance and board monitoring mechanism.As evidenced by Lin and Liu (2009), firms with smaller size of supervisory board and dual capacity of CEO and the board chairperson are less likely to hire high-quality auditors.External (or independent) directors are appointed to exercise the monitoring function on the operations of listed firms, and therefore, firms with higher portion of external directors are likely to hire high-quality auditors (Cheng & Leung, 2012).Nevertheless, the revised Company Law ( 2005) and Securities Law (2005) explicitly provide legal responsibilities for the listed firms together with their directors and managers which and who provide fraudulent financial information to their shareholders (Note 4).Therefore, it is uncertain whether internal management structure still has the significance influence in the auditor choice.
External governance mechanisms-institutional framework and marketization.In addition to ownership structure and internal management structure, external governance mechanisms may also have influence on the auditor choice.Wang and Xin (2011) Leung and Cheng (2013) further conclude that the firm value of Chinese listed firms registered in eastern coastal (well developed) region is higher than that of firms registered elsewhere.Therefore, external governance factors seem to have a significant influence on the auditor choice.

Research Question
Previous literature in developed and emerging markets (including China) supports that audit is one of corporate governance mechanisms (CGMs), but most scholars find that Big 4 auditors are more independent from audit clients and governments (e.g., DeFond, Wong & Li, 2000;Wang, Wong & Xia, 2008) and are more likely to issue modified (qualified) audit opinion once they detect the irregularities on the financial statements of their audit clients, and therefore they are often regarded as "high-quality" auditors.In China, previous literature evidences that certain CGMs show respective positive and negative associations with the auditor choice.Therefore, the following research question (RQ) is recited as: RQ: What is the key corporate governance factors related to the auditor choice in Chinese listed firms?

Data Source and Sample Selection
This  Cheng, 2013).There are 11955 Chinese main-board A-share firm-year observations for these 6 years available from the CSMAR.

Regression Model
Section 2 describes that many scholars use Big 4 auditors as the proxy for high-audit quality.The logistic regression of this study is run to examine the association between the corporate governance mechanisms and auditor choice in Chinese listed firms as follows: Where ɛ is the random error term of the model; i is the i*th firm and t is the year.
As mentioned in Section 2, there are two controversial issues on the concentrated ownership of the largest shareholders on the auditor choice.Leung and Cheng (2013) find that there is a non-linear association between ownership of the largest shareholders and firm valuation, and therefore, in the additional test, TOP1 2 is inserted into (1) to replace TOP1 to examine whether there is a non-linear association between ownership of the largest shareholders and firm valuation.
Table 2 contains summary descriptions of the variables used in the empirical analysis.More specifically, this study includes some variables which may be missing in some related studies: (1) the proportion of unpaid directors, UNPAID_DIR, is included because those unpaid directors are mostly appointed by the large shareholders and they may act towards the wills of those who appoint them; and (2) the potential development indicators of listed firms, GROW_TA and GROW_SALES, are included., where TA is the total assets of listed firms at year end

SALES SALES SALES
, where SALES is the total sales for the year., where NP and OCF are the profit (loss) and operating cash flows for the year respectively whereas TA is the total assets at year end (representing "accounting accruals")

TQ
Tobin-Q value as a ratio of the market value of equity of a firm to the book value of its assts FIXED_EFFECTS Dummy variables controlling the fixed effects of calendar years and industries (see

Multiple Regression Analysis
Table 8 presents regression results on the association between corporate governance mechanisms and auditor choice.TOP1 is positive to BIG4 at the 1% significant level, indicating that with an increase in ownership of the largest shareholder, a Chinese listed firm is more likely to hire high-quality auditors; such association is consistent with Ang, Cole and Lin (2000) and Fan and Wong (2005), but inconsistent with Copley and Douthett (2002) and Lin and Liu (2009).Additional sensitive tests are performed to support this conclusion (see below).TOP2_5 is positive to BIG4 at the 1% significant level, indicating that other large shareholders would also prefer high-quality auditor for the protection of their interests in the listed firm as expected.EXT_DIR is positive to BIG4 at the 1% significant level consistent with Cheng and Leung (2012), indicating that the higher portion of external directors in the board, the higher likelihood the listed firm to hire BIG4 as the external directors can relieve their fiduciary duties with the effective audit monitoring.DUAL_LIST is positive to BIG4 at the 1% significant level, and such association is expected; otherwise, the listed firm has to hire another auditor which is qualified for statutory audit (B-shares listed firm) or which is qualified for the purpose of Hong Kong Stock Exchange (H-shares listed firm).MI is positive to BIG4 at the 1% significant level, indicating that Chinese listed firms registered in the eastern coastal region (well developed) are more likely to hire high-quality auditor.
More surprisingly, BIG4 is not significant to lnBOS and DUAL_CAP, which is inconsistent to Lin and Liu (2009) who found that the size of supervisory board and DUAL_CAP showed respective positive and negative associations with BIG10.Our interpretation is that since both the largest (TOP1) and other large shareholders (TOP2_5) also demand for high-quality auditors, there is no other significant resistance.Additional sensitive tests are performed to support this interpretation (see below).
lnTA and TQ are positive to BIG4 at the 1% significant level, indicating that the listed firms with larger size and higher perceived market value are more likely to hire high-quality auditor.LEV and TAC are negative to BIG4 at 1% and 5% significant levels respectively, indicating that firms with higher gearing and accounting accruals are unlikely to hire high-quality auditors.
Additional test is rerun as TOP1 is replaced with TOP1 2 .TOP1 2 is positive to BIG4 at the 1% significant level, indicating that such association is a U-shape.Combing the effects of TOP1 and TOP1 2 indicates that when ownership of the largest shareholder is low, a Chinese listed firm is more likely to hire a non-Big 4 auditor because the largest shareholder can extract private benefits without the effective audit monitoring from the high-quality auditor.Once ownership of the largest shareholder reaches certain level, a Chinese listed firm would hire high-quality auditor to reduce its capital raising costs.Besides, the directions and significances of the associations between BIG4 and other variables remain unchanged.
A sensitivity test is performed.The ultimate controlling shareholders may have an influence on the auditor choice.For example, Leung and Cheng (2013)  Therefore, following Leung and Cheng (2013), two dummy variables, C_SCLF and L_SCLF, are added into Eq.
(1).As shown in Table 6, the directions and significances of those associations remain unchanged.

Supplementary Tests
In order to address the U-shape pattern of the association between TOP1 and BIG4, the authors conduct two supplementary tests.
As mentioned in Lin and Liu (2009), even though most Chinese listed firms have a very concentrated ownership structure in the hand of their largest shareholders, some do not (e.g. the largest shareholder of Minseng Bank holds less than 10%).First, the authors rerun Eq. ( 1) by limiting ownership of the largest shareholders (1) not lower than 20%, (2) not lower 30% and (3) more than 50%, of the total equity shares from the sample.Table 9 presents the empirical results of this test.In the 50% cut off, TOP1 is positive to BIG4, without significance, implying that the incentive for the largest shareholder to engage the Big 4 auditors becomes weaker, because it can exercise an effective control on the listed firm when its ownership is over 50%.Other than this, all associations are substantially the same as those in Table 8.Second, the authors set the threshold of ownership at 20% as the largest shareholders can get significant influence in Chinese listed firms.The authors rerun Eq. ( 1) by limiting ownership of the largest shareholders (1) lower than 10%, (2) lower than 15% and (3) lower than 20%, of the total equity shares from the sample.Table 10 presents the empirical results of this supplementary test.In all those ranges, TOP1 is negative to BIG4 at 1% and 5% significant level, implying that the incentive for the largest shareholder to engage the low-quality auditor becomes clear, because there is high deviation between cash flow right and control right of ownership of the largest shareholders and the largest shareholders do not prefer the higher audit quality for the possibility of tunneling.The associations between other corporate governance factors and BIG4 are similar to those shown in Table 8.

Summary of Regression Results
This study finds that there is a U-shaped pattern between ownership of the largest shareholders and the engagement of high-quality auditors, and in general, other large shareholders, the higher independent board, dual-listed firms and those registered in the eastern costal (more developed) regions also prefer the engagement of high-quality auditors.Other than those, no CGMs seem to have significant influence the auditor choice.

Conclusion
Even though several scholars believe Big 4 auditors as high-quality auditors for Chinese stock market, auditing services in China are still dominated by domestic accounting firms, and Big 4 auditors only account for about 6% from 2007 to 2012 in Chinese stock market, maybe because, on one hand, Chinese listed firms in which their largest shareholders who hold less than the threshold value (say 20%) (i.e.greater deviation between cash flow right and control right) would not engage Big 4 auditors as they do not prefer the better audit quality and therefore, they can maximize their private benefits through tunneling or expropriation of other shareholders (internal factor), and on the other hand, Big 4 auditors would not select the listed clients which are registered in non-eastern coastal regions, higher accounting accruals, smaller firm size and higher gearing (external factors).Besides, Wang, Wong and Xie (2008) find that Chinese listed firms have a preference to hire auditors within the same regions.Further, Liu and Subramaniam (2013) illustrate large auditors tend to charge the central SOEs lower audit fees than local SOEs taking the audit risk assessment and audit effort into account.Therefore, in order to enhance the audit quality in Chinese accounting profession, the policy makers can encourage the foreign auditors (including Big 4 auditors) to collaborate and merge with the domestic accounting firms, and provide incentive to foreign auditors to set up branches in non-eastern coastal region.

Table 3 .
Table3presents the details of the sample of this study.Our sample covers from 2007 to 2012 based on the firm's data available from the CSMAR.There are 11955 firm-year observations for these 6 years, of which 210 observations from the financial sector (Note 6) and 758 observations with missing variables are removed.Our final sample contains 10987 firm-year observations.Details of the sample

Table 4
presents the descriptive statistics.The mean of BIG4 is 6.0%, indicating that the engagement of Big 4 auditors in Chinese listed firms is rare.The means of TOP1 and TOP2_5 are 36.6%and 16.8% respectively, indicating that ownership of the largest shareholders in Chinese listed firms remains the same before and after the share reform and the alignment of other large shareholders is unlikely to restrict the acts of the largest shareholders in Chinese listed firms.The mean of DUAL_CAP is 21.0%, indicating that most Chinese listed firms have followed the professional recommendation to assign different persons to these two roles.The mean of EXT_DIR is 36,6%, fulfilling the requirements of independent directors in the management board of Chinese Listing Rules.The mean of UNPAID_DIR is 23.6%, indicating that about one-fourth of Chinese listed firms still have the directors assigned by the large shareholders.The mean of MI is 58.0%, indicating that more than half of Chinese listed firms are registered in the eastern coastal (well-developed) region.

Table 4 .
Descriptive statistics of the variables State-controlled listed firms (C_SCLF and L_SCLF) still amount to 51% in Chinese listed firms, but there is a decreasing trend, i.e. privatization of state owned enterprises.More surprisingly, DUAL_CAP and DUAL_LIST are increasing and decreasing respectively, indicating that there may a perceived detriment of the development of corporate governance in Chinese listed firms.

Table 6
presents the distribution of ownership of the largest shareholders in Chinese listed firms.It indicates that more than 86% of the largest shareholders hold ownership over 20% in Chinese listed firms and they are at least able to exercise significant influence on them.More surprisingly, more than 20% of the largest shareholders hold ownership over 50% in those firms, i.e. they can even have a dominant influence on particular firms.

Table 6 .
Distribution of ownership of the largest shareholders in Chinese listed firmsNote.* Correlation is significant at the 5% level (two-tailed) and ** Correlation is significant at the 1% level (two-tailed).
Lin and Mo (2006)porate governance mechanisms of central state-controlled listed firms (SCLFs) differ with those of local SCLFs from 2007 to 2009;Li and Luo (2011)also find that political connected private firms lack sufficient incentives to choose high-quality auditors because of auditor independence from 2004 to 2009; and Chan,Lin and Mo (2006)find that local (non-Big 4) auditors, who have greater economic dependence on local clients and are subject to more political influence from local governments than non-local auditors, are not willing to issue qualified audit opinion from 1996 to 2002.

Table 8 .
Regression results on the association of corporate governance mechanisms and auditor choice Note.P test values are in parentheses.*, ** and *** indicate significance at the 10%, 5% and 1% levels, respectively.

Table 9 .
Supplementary test on the regression results on the association of corporate governance mechanisms and auditor choice (20%, 30% and 50% cut off) Note.P test values are in parentheses.*, ** and *** indicate significance at the 10%, 5% and 1% levels, respectively.

Table 10 .
Supplementary regression on the regression results on the association of corporate governance mechanisms and auditor choice (Top1 < 20%) Note.P test values are in parentheses.*, ** and *** indicate significance at the 10%, 5% and 1% levels, respectively.