476 Corporate Governance: An International Review, 2009, 17(4): 476–491 The Determinants of Auditor Switching from the Perspective of Corporate Governance in China Z Jun Lin and Ming Liu* ABSTRACT Manuscript Type: Empirical Research Question/Issue: This paper reports on the association between the internal corporate governance mechanism of firms and their auditor switching types in the Chinese context Two types of auditor switching – namely switching to a larger auditor or switching to a smaller auditor – are identified and examined Research Findings/Insights: Controlling shareholders have the incentive to seek opaqueness gains The empirical results demonstrate that to realize opaqueness gains, firms with weaker corporate governance generally are more likely to switch to a smaller auditor rather than to a larger one Theoretical/Academic Implications: The empirical results demonstrate that firm-specific corporate governance devices will affect a firm’s auditor switching decision An effective corporate governance mechanism may inhibit the controlling shareholder from switching to a smaller auditor to exploit the minority shareholders The emerging economies are usually featured with concentrated ownership and insufficient legal protection of minority shareholders Compared with prior studies, this paper generates findings more applicable to the emerging economies Practitioner/Policy Implications: This study may facilitate market regulators and participants to maintain close monitoring of the structural arrangement of corporate governance of the listed firms, the independent auditing process and the credibility of financial reporting in an emerging market like China The findings also suggest that in order to bolster the confidence of the market participants, the Chinese government should promote the reform of the corporate governance system and enforce effective regulations, in particular on firms’ auditor switches Keywords: Corporate Governance, Supervisory Board, Government Ownership, State-Owned Enterprise (SOE), Auditor Switching, China INTRODUCTION T he purpose of this study is to investigate the association between a firm’s internal corporate governance mechanism and its auditor switching decisions in the Chinese context An independent auditing function can detect and disclose earnings management and other types of misconduct by business managers or controlling shareholders In general, if a firm has established a sound corporate governance mechanism, the firm’s management or its controlling shareholders will not have a free hand in making decisions on auditor selection, and vice versa Hence, there should be *Address for correspondence: Department of Accounting and Information Management, Faculty of Business Administration, University of Macau, Macau, China E-mail: Mliu@umac.mo an association between a firm’s corporate governance and its auditor selection decision This study empirically investigates the relationship between a firm’s internal corporate governance mechanism (proxied by ownership concentration, effectiveness of the supervisory board [SB] monitoring and shared board of directors [BoD] chairman and CEO) and their auditor switching decisions in the context of corporate governance practice in China This study was motivated by several factors First, as corporate governance has a positive impact on corporate financial reporting and auditing processes, a study of auditor switching with respect to the internal corporate governance mechanism may assist analysis of auditing quality and the auditor’s roles in ensuring the credibility of corporate financial disclosures Second, for the controlling owners, there is a tradeoff between hiring a high-quality auditor to lower the © 2009 Blackwell Publishing Ltd doi:10.1111/j.1467-8683.2009.00759.x THE DETERMINANTS OF AUDITOR SWITCHING FROM THE PERSPECTIVE OF CORPORATE GOVERNANCE IN CHINA costs of raising capital and hiring a low-quality auditor to maintain the gains from the opaqueness of corporate governance (called “opaqueness gains,” such as tunneling behaviors to transfer resources from a listed firm to its controlling shareholder) The ongoing bear market in China during the 2001–2004 period provides a good opportunity to disentangle these two incentives, thus allowing us to pinpoint the association between a firm’s internal corporate governance mechanism and its auditor switching decisions Third, the Chinese Institute of Certified Public Accountants (CICPA) began to rank auditors in China in the early 2000s in order to improve the transparency of the Chinese auditing market, thus allowing the possibility of identifying high-quality auditors in the Chinese market An investigation of the determinants of auditor switching from the perspective of corporate governance should contribute to an understanding of the necessity and utility of independent audits in China Our regression results show that firms with larger controlling shareholders (a higher degree of ownership concentration) or firms where the positions of board of directors’ chairman and CEO are held by the same person are more likely to switch to a smaller auditor rather than to a larger one However, the supervisory board monitoring strength is not a significant factor underlying auditor switching decisions The findings suggest that firms with a weak internal corporate governance mechanism generally tend to switch to smaller or more pliable auditors to sustain the opaqueness gains The finding that the supervisory board function (proxied by its size to represent monitoring effectiveness) does not have a significant influence on auditor switching decisions may imply that the supervisory board’s monitoring function is not consistently effective in practice The remainder of the paper is arranged as follows The second section reviews the relevant literature The third section develops the hypotheses to examine the association between the internal corporate governance mechanism and auditor switching decisions The fourth section introduces a regression model to test the hypotheses The fifth section presents and discusses the empirical results Sensitivity tests are presented in the sixth section Finally, the seventh section provides some conclusions LITERATURE REVIEW The Development of Auditing Profession in China Shortly after the founding of the People’s Republic of China in 1949, the auditing profession in China disappeared entirely due to the public (state) ownership of all production means An independent auditing function was virtually nonexistent in the planned economy before the 1980s when the state owned and ran enterprises directly But the mushrooming of Sino-foreign joint ventures, brought about by the government’s adoption of the “open-door” policy in the early 1980s, led to the emergence of independent auditing Due to the involvement of non-state equity interests in joint-ventures, it became necessary to have independent professionals, or certified public accountants (CPAs), to verify capital contributions and audit annual financial statements and income tax returns (Lin, Tang, & Xiao, 2003) Thus, the CICPA, a quasi-governmental organization in charge of national © 2009 Blackwell Publishing Ltd 477 administration of CPAs and auditing firms, was established in the early 1980s Following the business restructuring campaign, shareholding (stock) companies reappeared in the Chinese economy at the turn of 1990s, resulting in a further sharp increase in the demand for external audits The establishment of the Shanghai and Shenzhen stock exchanges and the promulgation of new accounting and auditing standards also played an important role in this process The China Securities Regulatory Commission (CSRC) required that the annual reports of all listed firms be audited by registered Chinese CPAs The Role of the Auditing Function In contemporary market economies, business incorporation leads to the separation of ownership and management Professional managers, rather than owners (shareholders), are directly involved in daily business operations Due to various self-interests and information asymmetries, business managers are able to pursue their own interests at the expense of those of the owners and other stakeholders (Jensen & Meckling, 1976) One of the binding mechanisms over management operations and information disclosure is the auditing function performed by independent professionals (Watts and Zimmerman, 1986) Nonetheless, the utility of the auditing function depends upon quality of the audit which is determined by the independence and expertise of the auditors (DeAngelo, 1981; Watkins, Hillison, & Morecroft, 2004) Audit quality is considered to be commensurate with the size of the auditors, i.e., larger auditors should have a higher degree of independence, possess more industrial expertise and resources, and bear higher reputation costs, so they can provide higherquality auditing services (DeAngelo, 1981; Lennox, 2005) Investors perceive the accounting numbers (e.g., earnings and book values) audited by large auditors to have better information quality, and therefore attach greater market value to them (Lennox, 2005; Watkins et al., 2004) DeFond and Subramanyam (1998) argue that there are incentives for the controlling shareholders of firms to pursue their own interests by manipulating the accounting numbers or transferring resources through “tunneling” behaviors Thus, the controlling shareholders will weigh their own self-interests when making auditor selection decisions (Johnson, La Porta, Lopez-de-Silanes, & Shleifer, 2000; La Porta, Lopez-de-Silanes, Shleifer, & Vishny, 2002) On the one hand, selecting a large auditor will signal to the market that the financial statements are more reliable, thus the firms may benefit from lower capital-raising costs on the equity or debt market On the other hand, large auditors may be more stringent in detecting and reporting “tunneling behavior” and hence may deprive the controlling shareholders of their opaqueness gains (Johnson & Lys, 1990) In particular, when firms receive unfavorable audit reports, they might initiate an auditor switch, searching for a more pliable auditor with the goal of “opinion shopping” (DeFond & Subramanyam, 1998; Johnson & Lys, 1990; Watkins et al., 2004) Auditor switching may take different forms, including switching to a smaller auditor and switching to a larger auditor Nonetheless in the extant literature there is a general lack of research differentiating the two types of auditor switching Volume 17 Number July 2009 CORPORATE GOVERNANCE 478 Corporate Governance and the Auditing Function Corporate governance evolved with the separation of ownership and management underlying the modern corporation system (Ang, Cole, & Lin, 2000) There are various types of principal-agent relations, e.g., between owners and management, between creditors and owners/management, and between controlling shareholders and minority shareholders A primary objective of corporate governance is to monitor the behavior of the various interested parties and to reduce the agency costs underlying the various principalagent relations (Karpoff, Malatesta, & Walkling, 1996) Thus, corporate governance can be defined as “a set of mechanisms, both industrial and market-based, that induce the self-interested parties of a company to make decisions that maximize the value of the company to its owners” (Denis & McConnell, 2003) An audit provides external monitoring over a firm’s financial reporting by independent professionals (auditors) and therefore serves a fundamental role in reinforcing information credibility However, the true effectiveness of external auditing is subject to the actuality and the development of the corporate governance environment (Holm & Laursen, 2007) A sound corporate governance mechanism should ensure that firms appoint qualified auditors and that the auditors exercise independent and effective monitoring over the financial reporting process and attest to the financial statements’ conformity with the Generally Accepted Accounting Principles (GAAP) Thus, corporate governance plays a role in enhancing the effectiveness of the audit function Francis and Wilson (1988) examined the relationship between a firm’s agency costs and its demand for audit quality in the US market They find that, concentrated ownership, a proxy for close alignment of interest between a controlling shareholder and the listed firm, can substitute for good audit quality and therefore is associated with choice of a low-quality auditor However, in China, as ownership is highly concentrated and legal protection of minority shareholders is insufficient, ownership concentration may lead to severe entrenchment problems and therefore represent poor corporate governance The association between corporate governance and external auditing is an important issue worthy of serious study In particular, as the agency relationship, the business administrative system, and corporate governance practices in emerging markets such as China differ substantially from those in the developed economies (Tam, 2000), we are interested in finding out whether empirical results are similar in the Chinese market as in the United States This study will not only enrich the extant auditing literature, but will also further promote the development of corporate governance and independent auditing practices in the emerging economies DEVELOPMENT OF THE HYPOTHESES Auditing plays two major functions: internally, auditing can be used to monitor management behavior and reduce agency costs (Jensen & Meckling, 1976); and externally, auditing can be used by potential creditors and investors to evaluate the cost of capital of the company, as described by the results of a survey of members of the American Institute Volume 17 Number July 2009 of Certified Public Accountants (AICPA) (Carpenter & Strawser, 1971): “Almost universally, the reason expressed [for the change in auditors] was that the underwriters informed the client that a ‘national known auditor’ was necessary to sell their offerings at the highest possible price.” In China, on average two-thirds of the shares are held by block shareholders (Lin, Liu, & Zhang, 2007) The block shareholder has the power to appoint a management team, and through the management team has access to inside information However, because the controlling shareholders also have the power to select the auditor,1 the minority shareholders not expect to rely on auditing to effectively monitor management Hence, the internal function of auditing to reduce owner-manager agency costs does not exist in China and for this function there is insufficient demand for audit quality As there is no significant bond market in China, listed firms are primarily financed by equity rather than debt After the establishment of Shanghai Stock Exchange and the Shenzhen Stock Exchange, Chinese listed firms achieved cumulative financing of RMB 1.16 trillion (RMB 6.80 = U$1) between 1992 and 2004 In 2000 the total market capitalization hit RMB 1.61 trillion, however the bear market thereafter resulted in the slumping of the market value by RMB 44 trillion (CSRC, 2005) During the weak market period of 2001 to 2004 listed firms were not enthusiastic about offering new equity securities to the public, therefore the external use of audits was also substantially diminished In contrast, controlling shareholders are motivated to hire low-quality auditors to seek potential opaqueness gains at all times, especially if they are operating in a weak corporate governance environment where there are only loose monitoring and binding contracts (Felo, Krishnamurthy, & Solieri, 2003) To address the research question empirically, we set out to test the association between the firms’ internal corporate governance mechanism and their auditor switching decisions, i.e., whether firms with a weak internal corporate governance mechanism are more likely to switch to a lowerquality auditor If such an association does not exist, internal corporate governance mechanism may not impact the type of auditor switches Alternatively, if auditors of different size offer monitoring services with varied levels of quality, the firms’ internal corporate governance mechanism may impact the switch type, in respect of the varied monitoring functions on the firms’ opaqueness gains The association may be evidence of the cost-benefit based demand for audit monitoring: when the opaqueness gains outweigh the benefits of lowering capital raising costs, lower-quality auditors would be preferred by Chinese firms, and especially by firms with weak internal corporate governance mechanisms, as these firms have more opaqueness gains to protect (DeFond, Wong, & Li, 2000; Lin et al., 2007) By switching to a smaller auditor, the controlling shareholder (the agent) may comfortably exploit the wealth of the minority shareholders without being watched closely by the auditor In contrast, switching to a larger auditor leads to more rigid audit monitoring and hence tunneling behaviors will be confined We used three proxies to measure a firm’s internal corporate governance mechanism: ownership concentration (shareholding of the largest owner); the effectiveness of © 2009 Blackwell Publishing Ltd THE DETERMINANTS OF AUDITOR SWITCHING FROM THE PERSPECTIVE OF CORPORATE GOVERNANCE IN CHINA supervisory board (size of the supervisory board); and shared CEO-Chair (whether the board of directors’ chairman and CEO positions are held by the same person) A high ownership concentration is a distinct feature of listed firms in China A Chinese listed firm usually has a large controlling shareholder,2 who often is the government or the parent state-owned enterprise.3 Nonetheless, the ownership structure affects corporate governance and corporate value in many different ways Johnson et al (2000) argue that more narrowly held firms may face greater agency costs because the controlling shareholders will have a dominant influence on corporate affairs and they can easily bypass monitoring by other shareholders La Porta, Lopez-deSilanes, and Shleifer (1999) and La Porta et al (2002) show that in the emerging transitional economies, the controlling shareholders may expropriate the minority shareholders through aggressive “tunneling” behaviors They further argue that “the central agency problem in large corporations around the world is that of restricting expropriation of minority shareholders by controlling shareholders” (La Porta et al., 1999) This is particularly true for Chinese listed firms in which the controlling shareholders usually hold a very high percentage of the equity shares In China, the controlling shareholders have frequently intervened in the operations of the listed firms to benefit the parent companies, e.g., using the listed firms as guarantors for loan applications for the parent and related companies and therefore exposing the listed firms to extra financial and operating risks In fact, the controlling shareholders of many listed firms are keen to raise funds only on the stock market They frequently engage in benefit transfers through the misappropriation of funds and related-party transactions to expropriate the interests of the minority shareholders, which, if detected, may invite external intervention by minority shareholders and other stakeholders (Lin et al., 2007) The desire to maximize self-interest through “tunneling” behaviors leads the listed firms to avoid being monitored by a high-quality auditor The more concentrated the ownership structure (i.e., with a larger controlling shareholder), the weaker the internal corporate governance mechanism Therefore, firms with larger controlling shareholders are expected more likely to switch to pliable auditors to realize opaqueness gains through tunneling behaviors or other types of misconduct, as stated below: Hypothesis (H1): All other things being equal, a Chinese firm with a higher percentage of total shares held by its controlling shareholder will more likely switch to a smaller auditor Pursuant to the Chinese Company Law, all Chinese firms adopted a German-style dual-board governance system, thus each listed firm has both a board of directors and a supervisory board The supervisory board is composed of the shareholders’ representatives (including Chinese Communist Party officials) and an appropriate proportion of employee representatives, who are nominated by the firm’s employee union The Company Law specifically defines the supervisory board as a monitoring mechanism to carry out a series of responsibilities, including: (1) monitoring the performance of the directors and senior managers to ensure compliance with the laws, regulations, and the articles of incorporation; (2) reviewing the financial affairs of the firm; © 2009 Blackwell Publishing Ltd 479 (3) requesting the directors and senior managers to alter and/or rectify their personal activities if deemed in conflict with the firm’s objectives; (4) proposing specific shareholder meetings whenever deemed necessary; (5) fulfilling any other duties that are stipulated in the articles of incorporation of the firm; and (6) submitting a supervisory board report to the shareholders’ annual general meeting The Standard Code of Corporate Governance for Listed Companies in China issued by the CSRC and the State Economic and Trade Commission in 2002 further requires that supervisory board members should have some professional knowledge or work experience in the areas of law and accounting (CSRC, 2002) Based on the requirements of the Company Law, the supervisory board shall independently and effectively carry out supervision over the activities of the directors and the management as well as examine the financial affairs of the firm Such a German-style two-tiered board system with the co-existence of a board of directors and a supervisory board has become the backbone of corporate governance in most Chinese listed firms since the mid-1990s Using an event study, Dahya, Karbhari, Xiao, and Yang (2003) report that investors consider the supervisory board to be an important device of corporate governance in China Chen (2005) finds that there is a positive association between the size of the supervisory board and the level of corporate governance, suggesting that a larger supervisory board should be more effective in carrying out its legitimate monitoring responsibilities We use the number of supervisory board members as a proxy for the monitoring effectiveness of the supervisory board and have the second hypothesis: Hypothesis (H2): All other things being equal, a Chinese firm with fewer supervisory board members will more likely switch to a smaller auditor Within a sound corporate governance structure, the board of directors must ensure that the management acts in the best interests of the shareholders The board of directors is responsible for execution of the resolutions passed by the shareholders’ meetings and for appointing, removing, and remunerating senior managers Traditionally, the sharing of the position of board of directors’ chairman and CEO has been common in the United States However, in most European, British, and Canadian businesses, in an effort to ensure better corporate governance, these two positions are often split Combining the two positions does have its advantages, allowing the CEO multiple perspectives on the firm as a result of his/her multiple roles and empowering him/her to act with determination Nonetheless, this practice results in less transparency of the CEO’s activities, and as such his/her actions can go unmonitored, which paves the way for scandals and corruptions To the contrary, separation of the two positions allows the board of directors’ chairman, on behalf of the stockholders, to be more impartial in overseeing the work of the CEO and the overall performance of management (La Porta et al., 2002; Petra, 2006) Investors, researchers, and government officials have gradually accepted the view that the best practices of corporate governance require the separation of the roles of board of directors’ chairman and CEO Such a corporate governance device has received a boost since 2003 In practice, Volume 17 Number July 2009 CORPORATE GOVERNANCE 480 market regulators and professional bodies in many developed countries require that the two important positions be separated (Jiraporn, Young, & Davidson, 2005) In 2002 the CSRC also adopted this requirement in its Standard Code of Corporate Governance for Listed Companies in China Consistent with the association between internal corporate governance and the auditing function, we have the third hypothesis stated as the following: Hypothesis (H3): All other things being equal, a Chinese firm with the board of directors’ chairman and CEO positions held by the same person is more likely to switch to a smaller auditor RESEARCH METHODOLOGY Model Specification We intend to examine the determinants of audit switching from the perspective of the internal corporate governance mechanism of Chinese listed firms Our sample includes firms that switched auditors from 2001 to 2004 We classified all firms that switched auditors only once during the 4-year test period into two types – those switching to a larger auditor (upward switching, or US firms) and those switching to a smaller auditor (downward switching, or DS firms) according to the ranking order of auditors in China, which was compiled by the CICPA in terms of the CPA firms’ annual audit revenue (see Appendix) As elaborated earlier, the size of the auditing firm is regarded as an effective surrogate for the independence and monitoring strength of the auditors (Copley & Douthett, 2002; DeAngelo, 1981) Thus we construct a model to test whether the firm’s internal corporate governance mechanism (proxied by ownership concentration, supervisory board, and shared CEO-Chair) is associated with the different types of auditor switching (namely, upward switching or downward switching) Downward switching (DS) can be expressed as a function of the three corporate governance variables in which we are interested and the related control variables: DS = f (owner- ship concentration, supervisory board, shared CEO-Chair, control variables, error terms) As we classify all auditor switches into two types (upward switches or downward switches), there are only two values possible (0 or 1) for the dependent variable Hence, we run logit regression, which is used to predict the probability of an occurrence of an event by fitting the data to a logistic curve It makes use of several predictor variables that may be either numerical or categorical The logistic function is useful because it can take as an input any value from negative infinity to positive infinity, whereas the output is confined to values of either or The dependent variable represents the exposure to some set of risk factors, whereas f(dependent variable) = 1/(1 + exp (-dependent variable)) represents the probability of a particular outcome, given that set of risk factors The following logit model is used to test Hypotheses to DS = β0 + β1LSH + β2 SB + β3CEOCHR + β4 GOV + β5OPI + β6 LNASSET + β7 LEV + β8 MB + β9 LOSS + β10 NWISS + β11Yr 02 ∗ LSH + β12Yr 03 ∗ LSH + β13Yr 04 ∗ LSH + β14Yr 02 ∗ SB + β15Yr 03 ∗ SB + β16Yr 04 ∗ SB r + β17 Yr 02 ∗ CEOCHR + β18Yr 03 ∗ CEOCHR (1) + β19Yr 04 ∗ CEOCHR + ε Please see Table for a description of the variables in the model Three groups of variables, as discussed below, help explain why certain factors trigger a certain type of auditor switching The first group consists of corporate governance variables, namely ownership concentration (LSH), supervisory board (SB), and shared CEO-Chair positions (CEOCHR) Although the controlling shareholders may seek to influence auditor selection so as to facilitate their tunneling behaviors, they are subject to the constraints of the corporate governance structure in place The second group consists of company-specific variables that have been tested or are considered helpful to explain auditor switching in TABLE Description of Variables = = = = = = = = = LOSS NWISS Yr02 Yr03 Yr04 if the firm switches to a smaller auditor; otherwise largest owner’s shareholding as a percentage of total shares number of members of the SB if the CEO also holds the position of chairman of BoD; otherwise if the largest shareholder is a government agency; otherwise if the firm receives an unclean auditor’s opinion for the previous year; otherwise log of total assets at the end of the previous year long-term liabilities divided by total assets at the end of the previous year market to book ratio at the end of the previous year, calculated as the market value of stocks divided by the book value = if the firm experiences a loss for the previous year; otherwise = if there is a new equity issue in the two years immediately after auditor switching; otherwise = if the auditor switching occurs in Year 2002; otherwise = if the auditor switching occurs in Year 2003; otherwise = if the auditor switching occurs in Year 2004; otherwise Volume 17 Number DS LSH SB CEOCHR GOV OPI LNASSET LEV MB July 2009 © 2009 Blackwell Publishing Ltd THE DETERMINANTS OF AUDITOR SWITCHING FROM THE PERSPECTIVE OF CORPORATE GOVERNANCE IN CHINA prior studies, including government control (GOV), auditor’s opinion (OPI), size (LNASSET), financial leverage (LEV), market-to-book ratio (MB), profitability (LOSS), and new issues (NWISS) As there were new requirements in the Standard Code of Corporate Governance during the test period, we incorporate the third group of interactions variables to capture the yearly impact of the gradual adoption of these new requirements, namely Yr02, Yr03, and Yr04 With the interactions terms, we intend to test whether the impact of corporate governance mechanism on auditor switching varies with the progress of corporate governance practice In the regression, the dependent variable is defined by the types of auditor switching, thus it is coded if a firm switched to an auditor that was smaller than its predecessor For the independent variables, we expect b1 (for ownership concentration) and b3 (for shared CEO-Chair) to have positive signs as firms with a high ownership concentration and a shared board of directors’ chairman and CEO are more likely to switch to a smaller auditor But b2 (for supervisory board) is expected to be negative as a large or strong supervisory board may inhibit the firm from switching to a smaller auditor Government controlled and non-government controlled (privately owned or privately controlled) firms may have differing corporate governance structures and may also have different considerations when making auditor switching decisions In general, government agencies have a stronger influence over government-controlled firms and therefore can more easily access the firm’s financial information for their decision making (Chan, Lin, & Mo, 2006) Therefore, government-controlled firms may have less demand for high-quality independent audits and may have a greater propensity, compared with non-government controlled firms, to switch to smaller auditors Hence we add the variable GOV to capture the effect of government control over the firm’s auditor switching decisions This variable is coded as if the largest owner of the firm is a government agency and otherwise It is expected to be positively associated with a downward switch of auditors We also control for the effects the auditor’s opinion, firm size, financial leverage (risk), growth potential, profitability, and whether there is new issue of equity after the auditor switching Prior auditor switching studies focus mainly on the auditing markets in the Western countries Nonetheless, the basic theories and findings in prior research on auditor switching should be applicable to this study as well because the Chinese auditing profession in recent years has gradually adopted international accounting and auditing standards One very common reason cited for auditor switching is the qualifications of the auditor’s opinions Prior research has found that firms receiving unfavorable audit reports are more likely to switch auditors (DeFond & Subramanyam, 1998) We expect that the auditor’s opinion (receiving an unfavorable auditor opinion in the prior year = 1) to be positively related to a downward switch of auditors Large firms may be less likely to switch to a smaller auditor, as financial analysts and the financial press will more closely scrutinize their auditor switches Following Friedlan (1994), we use the log of the total assets to control for the size effect of the firms and we expect it to be negative in the regression model Reed, Trombley, and Dhaliwal (2000) find that firms select- © 2009 Blackwell Publishing Ltd 481 ing Big auditors tend to be highly leveraged, whereas Titman and Trueman (1986) predict otherwise As there are opposing arguments and findings regarding the association between a firm’s leverage and its auditor switching, we not predict the sign of the coefficient for financial leverage We also include the market-to-book ratio to control for the propensity of growing firms to switch to less conservative auditors (DeFond & Subramanyam 1998) Moreover, Sainty, Taylor, and Williams (2002) document that profitability may affect the selection of auditors In our model, we expect a net loss in the prior year (LOSS) to be positively associated with a downward switch of auditors Firms may change auditors (especially from a low-quality auditor to a high-quality auditor) to increase the marketability of new securities (Carpenter & Strawser, 1971) Pae and Yoo (2001) document a negative relationship between audit quality and the cost of raising capital, i.e., a firm can reduce its costs of raising capital by hiring a quality auditor We therefore include the variable of new issues (NWISS) in our regression model, which equals if the listed firm issues new equity to the public in the two years after its auditor switching and otherwise The new issue is used to proxy for the firm’s intentions to issue new equity at the time of the auditor switching Although the firms in our sample generally did not intend to issue new equity to the public during our test period, for accuracy and completeness we still include the variable Since 2002, new corporate governance requirements have been announced and implemented, therefore we add the year dummy variables for 2002 to 2004 to capture the potential impact of these new corporate governance devices on the firms’ auditor switching decisions during the test period Sampling Our sample covers A-share firms that switched auditors from the beginning of 2001 to the end of 2004.4 There are two main reasons to limit the sample firms that made auditor switches to this time period The first one is the availability of the ranking of Chinese auditors, which has been compiled by the CICPA since 2002 Thus it provides the possibility of identifying or classifying the different types of auditor switches (i.e., an upward switch or a downward switch) The second reason is that during this time period, firms had little intention to offer equity to the public, therefore the opaqueness gains from weak corporate governance significantly outweighed the benefits from lowering the costs of raising capital Hence, the bear market period from 2001 to 2004 is appropriate to test the association between the firms’ internal corporate governance mechanism and their auditor switching decisions Data were collected from the China Stock Market and Accounting Research (CSMAR) Database, the TEJ database (carrying financial information and stock market data compiled by the Taiwan Economic Journal), and authoritative national newspapers or magazines designated by the CSRC to publish financial reports of listed firms, such as China Securities Daily, Shenzhen Securities Times, and Shanghai Securities News The collected data on the sample firms were cross-checked and verified by different data sources to ensure their reliability A description of the data is provided in Table Volume 17 Number July 2009 Volume 17 Number July 2009 42 48.94 4.21 06 79 16 21.04 08 5.88 12 03 28 23 15 N 233 233 233 233 233 233 233 233 233 233 233 233 233 233 Variable DS LSH (%) SB (#) CEOCHR GOV OPI LNASSET LEV MB LOSS NWISS Yr02 Yr03 Yr04 Panel B: Descriptive Statistics of Variables 42 10 65 49 10 13 79 Industry Commerce Property Conglomerate Total Mean 2002 2001 Sector Panel A2: Sample Distribution by Sector and Year Firms switched auditors during 2001–2004 Less: Firms with missing data Financial, transportation, and utility firms Firms switched auditors for more than once during 2001–2004 Final sample Panel A1: Sample Selection 48.38 20.94 06 3.88 0 0 Median TABLE Description of Data 32 6 53 2003 43 0 19.03 00 1.27 0 0 0 MIN 89.51 12 1 24.60 51 230.39 1 1 MAX 17 10 36 2004 50 18.11 1.73 25 41 37 89 09 15.65 33 18 45 42 36 STD 140 29 22 42 233 Total 316 47 11 25 233 482 CORPORATE GOVERNANCE © 2009 Blackwell Publishing Ltd © 2009 Blackwell Publishing Ltd LSH 1.00*** -.05 07 20*** -.16*** -.04 04 -.04 21*** -.08 02 -.09† -.01 DS 1.00*** 11† 09† 13** -.18*** 04 13** 07 05 -.00 -.07 05 -.16*** 11** 1.00*** 01 -.07 05 -.02 -.03 05 -.03 -.06 06 01 -.00 SB 1.00*** -.15** -.16*** -.03 -.05 -.07 -.08 -.05 03 03 -.11** CEOCHR 1.00*** 25*** -.10† -.21*** -.14** 09† -.04 04 10† -.01 LNASSET 1.00*** -.01 -.08 06 02 03 -.00 -.02 05 LEV DS LSH SB CEOCHR LNASSET LEV MB LOSS OPI GOV NWISS Yr02 Yr03 Yr03 1.00*** 01 19*** -.01 -.04 -.05 -.08 -.04 MB 1.00*** 44*** 00 -.07 03 04 09† LOSS 1.00*** -.12** -.08 -.04 -.04 01 OPI 1.00*** -.13** -.01 -.05 02 GOV 1.00*** -.01 01 -.02 NWISS 1.00*** -.34*** -.27*** Yr02 1.00*** -.23*** Yr03 1.00*** Yr04 = if the firm switches to a smaller auditor; otherwise = the largest owner’s shareholding as a percentage of total shares = number of SB members = if the CEO also holds the position of the BoD chairman; otherwise = log of total assets at the end of the previous year = long-term liabilities divided by total assets at the end of the previous year = market-to-book ratio at the end of the previous year, calculated as the market value of stocks divided by the book value = if the firm experiences a net loss for the previous year; otherwise = if the firm receives an unclean (non-standard) auditor opinion for the previous year; otherwise = if the largest shareholder is a government agency; otherwise = if there is a new equity issue in the two years immediately after auditor switching; otherwise = if the switching occurs in Year 2002; otherwise = if the switching occurs in Year 2003; otherwise = if the switching occurs in Year 2004; otherwise The variables are defined as below: ***, **, and † denote significance at the 1%, 5%, and 10% levels, respectively DS LSH SB CEOCHR LNASSET LEV MB LOSS OPI GOV NWISS Yr02 Yr03 Yr04 Variable Panel C: Correlation Coefficient Matrix of Variables Table Continued THE DETERMINANTS OF AUDITOR SWITCHING FROM THE PERSPECTIVE OF CORPORATE GOVERNANCE IN CHINA Volume 17 Number 483 July 2009 CORPORATE GOVERNANCE 484 At the end of 2004, there were 1,387 A-share firms listed on the two stock exchanges in China, among which 316 firms (22.7 per cent) switched auditors during the four-year period from 2001 to 2004 This implies that generally a firm is not willing to switch auditors because of the potential high costs associated with auditor switching, such as the costs of searching for and renegotiating with a new auditor and the potentially unfavorable market responses to an auditor switch (Reed et al., 2000; Watkins et al., 2004) Panel A of Table presents the sample size for this study Financial, transportation, and utility firms are excluded because the nature of their operations is very different from that of other types of firms We also delete firms that switched auditors more than once during the four-year period The frequent switching of auditors may indicate some serious underlying reasons that are beyond the scope of this study Furthermore, firms that switched twice or more may have switched to a larger auditor at one time and to a smaller auditor the other time, making it difficult to categorize the type of switching The final sample consists of 233 firms EMPIRICAL RESULTS Panel B of Table presents the basic statistics on the tested variables Among the 233 sample firms, 134 firms switched to larger auditors and 99 firms switched to smaller auditors, based on the rankings prepared by the CICPA On average, the largest controlling owners held 48.94 per cent of the total shares of the sample firms, indicating a high ownership concentration in the Chinese listed firms About 79 per cent of our sample firms were directly owned by the government or governmental agencies, reflecting that most Chinese listed firms were originally carved out from state-owned enterprises and that various government agencies remain the largest owners of the listed firms In the sample, 16 per cent (37/233) of the firms received unfavorable auditor opinions before their auditor switches This high percentage may support the assertion that there is an association between receiving unfavorable auditor reports and switching auditors The average size of the supervisory board was about 4.21, with a minimum of member and a maximum of 12 members In per cent of the sample firms (15/233), the CEO also held the position of board of directors’ chairman Very few firms intended to issue equity during the bear market, as indicated in the table that only per cent (8/233) of the sample firms issued equity in the two-year period after their auditor switching Panel C of Table presents the correlation coefficient matrix for the variables used in the regression model A downward switch is significantly and positively correlated with ownership concentration, shared CEO-Chair, and market-to-book ratio; and it is significantly and negatively related to firm size The correlation coefficients among the independent variables are moderate, with no value exceeding 50 (the largest is 44) Table provides the empirical results from the regression which tests whether a firm with a weak internal corporate governance mechanism is inclined to switch to a smaller auditor Since there are only two values possible for the dependent variable (1 for a downward switch and for an Volume 17 Number July 2009 upward switch), we use a logit regression With a Chi-square of 40.03, p < 01, and a pseudo R-square of 21, the regression model is satisfactory in differentiating firms switching to smaller auditors from firms switching to larger auditors at an acceptable level of significance The probability of a downward switch can be expressed as f(DS) = 1/(1 + exp(-DS)), i.e., the larger the value of the DS, the higher the probability of a downward switch The coefficient for ownership concentration is positively significant at the per cent level (Coeff = 02, Wald = 4.78, p < 05), which supports H1 If under the current situation the probability of a downward switch f(DS) is exactly 50 (equal probability of an upward switch or a downward switch), then an increase of 10 per cent of the total shareholding will cause f(DS) to increase to 55 This suggests that firms with a higher degree of ownership concentration (i.e., higher percentage of equity shares held by the largest shareholder) are more likely to switch to a smaller auditor Although the test results seem to be the same as Francis and Wilson 1988 findings, our interpretation is completely different The relatively high internal ownership in upward switches should better align the interests between managers and shareholders, but may not be significant enough to expropriate the minority shareholders, and hence proxies for effective corporate governance (Francis & Wilson, 1988; Sabherwal & Smith, 2008) Effective corporate governance can substitute for a quality auditor, and therefore is associated with switching to a lower-quality auditor in the United States In contrast, in China, due to the awfully high level of concentration, the divergence between cash flow rights and control rights, and the ineffective legal protection, entrenchment problems become detrimental, hence ownership concentration proxies for poor corporate governance Poor corporate governance allows the controlling shareholders more space to hire a lower-quality auditor to realize their opaqueness gains, therefore leading to our results – a high ownership concentration, a proxy for poor corporate governance, is associated with switching to a lower-quality auditor The coefficient for the size of the supervisory board is insignificant (Coeff = 03, Wald = 05, p > 10), therefore H2 is not supported, indicating that the effectiveness of supervisory board monitoring may not be related to whether a firm switches to a larger or a smaller auditor Another possible interpretation is that in practice the monitoring role of the supervisory board is dubious, thus the supervisory board presently does not have a significant impact on a firm’s auditor switching decision In fact, as supervisory board members are mainly from inside the firm, whether the supervisory board can effectively play a monitoring role is controversial Some researchers contend that the supervisory board is mainly decorative in China (Dahya et al., 2003) and our findings seem to support this viewpoint Consistent with H3, the coefficient for the shared CEOChair is positive and significant at the per cent level (Coeff = 1.28, Wald = 3.85, p < 05) If the current status is the separation of the CEO and the board of directors’ chairman and the probability of a downward switch f(DS) is exactly 50, then a change to the combination of the two key roles will lead f(DS) to increase to 78 Thus, a firm is more likely to switch to a smaller auditor if its CEO also holds the position of board of directors’ chairman, as expected © 2009 Blackwell Publishing Ltd © 2009 Blackwell Publishing Ltd -.31 12 NWISS b 10 01 1.18 ? Yr02*LSH b 11 ? 6.42 2.25 ? -.01 2.21 Yr03*LSH b 12 01 2.24 ? ? 03 03 Yr02*SB b 14 + 1.28 3.85** CEOCHR b3 Yr04*LSH b 13 03 05 - + 02 4.78** SB b2 LSH b1 03 01 ? 09 24 Yr03*SB b 15 + GOV b4 ? 13 38 Yr04*SB b 16 + -.11 06 OPI b5 ? 26 53 Yr02*CEOCHR b 17 -.43 4.56** LNASSET b6 ? 09 20 Yr03*CEOCHR b 18 + 3.56 3.95** LEV b7 + 18 12 LOSS b9 ? 01 04 Yr04*CEOCHR b 19 + 13 3.70† MB b8 (1) DS LSH SB CEOCHR GOV OPI LNASSET LEV MB LOSS NWISS Yr02 Yr03 Yr04 = if the firm switches to a smaller auditor; otherwise = the largest owner’s shareholding as a percentage of total shares = number of SB members = if the CEO also holds the position of the BoD chairman; otherwise = if the largest shareholder is a government agency; otherwise = if the firm receives an unclean (non-standard) auditor opinion for the previous year; otherwise = log of total assets at the end of the previous year = long-term liabilities divided by total assets at the end of the previous year = market-to-book ratio at the end of the previous year, calculated as the market value of stocks divided by the book value = if the firm experiences a net loss for the previous year; otherwise = if there is a new equity issue in the two years immediately after auditor switching; otherwise = if the switching occurs in Year 2002; otherwise = if the switching occurs in Year 2003; otherwise = if the switching occurs in Year 2004; otherwise Notes: (1) The Wald-Wolfowitz test examines whether the mean (median) of each variable for Top 10 clients equals to that of non Top-10 clients ***, **, and † denote significance at the 1%, 5%, and 10% levels, respectively (2) N = 233, Chi-square = 40.03, and Pseudo R-square = 21 Prediction Coefficient Wald Prediction Coefficient Wald Intercept b0 DS = β0 + β1LSH + β2 SB + β3CEOCHR + β4GOV + β5OPI + β6 LNASSET + β7 LEV + β8 MB + β9 LOSS + β10 NWISS + β11Yr 02 ∗ LSH + β12Yr 03 ∗ LSH + β13Yr 04 ∗ LSH + β14Yr 02 ∗ SB + β15Yr 03 ∗ SB + β16Yr 04 ∗ SB + β17 Yr 02 ∗ CEOCHR + β18Yr 03 ∗ CEOCHR + β19Yr 04 ∗ CEOCHR + ε r TABLE Internal Corporate Governance Mechanism and Switching to a Smaller Auditor – Main Test THE DETERMINANTS OF AUDITOR SWITCHING FROM THE PERSPECTIVE OF CORPORATE GOVERNANCE IN CHINA Volume 17 Number 485 July 2009 CORPORATE GOVERNANCE 486 The coefficient of government control is positive but insignificant, so government-controlled firms, compared to non-government-controlled firms, are more likely to switch to smaller auditors, but the difference is only marginal In addition, the coefficient for the auditor opinion variable (OPI) is positively related to switching to smaller auditors, but not at a conventional level of significance Therefore, there is insufficient evidence to support the assertion that Chinese firms receiving unfavorable audit opinions are inclined to switch to smaller auditors for the purpose of “opinion shopping.” This is not surprising as other studies in the United States (Watkins et al., 2004) and China (Chan et al., 2006) report similar findings Consistent with our prediction, firm size is negatively related to switching to a smaller auditor at the 05 level of significance (Coeff = -.43, Wald = 4.56, p < 05) This finding confirms that larger firms are less likely to make downward switching decisions The reasons may be larger firms have more complicated operations and they care more about their reputation In addition, both financial leverage (Coeff = 3.56, Wald = 3.95, p < 05) and the marketto-book ratio (Coeff = 13, Wald = 3.70, p < 10) have positive and significant coefficients Therefore, firms with greater growth potentials and higher financial leverage ratios are inclined to switch to smaller auditors The profitability of firms (proxied by the loss incurred in the prior year) does not have a significant effect on auditor switching decisions, nor does a new issue (NWISS) We suggest that the rare cases of new equity issues (8 out of 233) may be the main reason for the insignificant result As indicated by Table 3, the coefficients for year dummy variables are not significant This suggests that although Chinese listed firms were required to adopt the independent non-executive director and audit committee practices in their corporate governance after 2002, auditor switches during the test period were not significantly affected Possible explanations are that most listed firms did not adopt the new practices until 2004 or the enforcement of the new practices was not satisfactory during the test period In summary, the empirical results support that there is an association between a firm’s internal corporate governance mechanism and its auditor switching decisions Firms with weaker internal corporate governance (proxied by ownership concentration and a shared CEO-Chair) are inclined to switch to smaller auditors so that they can shirk from more stringent audit monitoring and realize opaqueness gains This result implies that the opaqueness gains derived from self-interested activities such as tunneling behaviors might be a crucial consideration when Chinese firms switch auditors SENSITIVITY TESTS To examine the robustness of the empirical results, we performed varied sensitivity tests First, we adopt a more strict definition for a downward switch (DS), which is defined as switching from a Top 10 auditor to a non-Top 10 auditor The empirical results are similar (Table 4-1) In order to make sure that the largest shareholder controls the listed firm, we limit the sample to firms in which the largest shareholders Volume 17 Number July 2009 own a significant percentage of the total shares After deleting the observations in which the largest owners hold less than 10 per cent of the total shares, the empirical results remain qualitatively the same (Table 4-2) We also adopt alternative proxies to measure the control variables We use the log of revenues to proxy for firm size, the total debt to asset ratio to proxy for financial leverage, and the increase of assets for growth After making the changes for these control variables, the empirical results still reveal a significantly positive relation between a downward switch (DS) and ownership concentration (LSH) and between a DS and a shared CEO-Chair (CEOCHR) Again, the coefficient for the supervisory board is insignificant Therefore, H1 and H3 are robustly supported, and there is insufficient evidence for H2 CONCLUSIONS The purpose of this paper is to investigate the determinants of Chinese firms’ auditor switching decisions from the perspective of their internal corporate governance mechanism Three measures are used to proxy for the internal corporate governance mechanism, including the concentration of ownership (shareholding of the controlling owner), the effectiveness of the supervisory board’s monitoring (proxied by supervisory board size), and shared CEO and board of directors’ chairman We divided all auditor switches during the 2001–2004 period into two types: switching to a larger auditor and switching to a smaller auditor, and empirically examined the impact of the corporate governance variables on the firms’ auditor switching decisions Three hypotheses are used to test the association between the firms’ internal corporate governance mechanism and their auditor switching decisions H1 and H3 are supported, but there is insufficient evidence to support H2 Hence firms with larger controlling owners or firms in which the positions of board of directors’ chairman and CEO are held by the same person are more likely to switch to a smaller auditor However, it is inconclusive whether firms with smaller supervisory boards opt for a smaller auditor So generally, we conclude that firms with a relatively weak internal corporate governance mechanism are more likely to switch to smaller auditors in order to protect or realize the opaqueness gains associated with a weak corporate governance mechanism That the supervisory board does not have a significant impact on firms’ auditor switching may indicate supervisory board members’ inability to monitor effectively, possibly because they are mainly from inside the firms and/or not have sufficient corporate governance expertise This study contributes to the literature on auditing research First, although some prior studies have examined whether there is an association between firms’ auditor switching decisions and firm-specific characteristics, most of these studies not consider the different types of auditor switching, driven by different motivations In this study, two major types of auditor switching are classified, namely switching to a larger auditor and switching to a smaller one Second, compared with the Francis and Wilson, 1988, study of the US market, our results are more applicable to the © 2009 Blackwell Publishing Ltd © 2009 Blackwell Publishing Ltd -.18 03 NWISS b 10 01 1.11 ? Yr02*LSH b 11 ? 12.80 3.20† ? 01 58 Yr03*LSH b 12 02 2.19 ? ? -.21 28 Yr02*SB b 14 + 1.33 4.00** CEOCHR b3 Yr04*LSH b 13 40 40 - + 04 4.23** SB b2 LSH b1 ? -.05 05 Yr03*SB b 15 + -.08 08 GOV b4 ? -.28 16 Yr04*SB b 16 + -.59 1.27 OPI b5 35 1.70 ? Yr02*CEOCHR b 17 -1.21 4.01** LNASSET b6 ? -.34 1.31 Yr03*CEOCHR b 18 + 3.05 51 LEV b7 + + -.22 01 LOSS b9 ? 15 60 Yr04*CEOCHR b 19 19 46 MB b8 (1) DS LSH SB CEOCHR GOV OPI LNASSET LEV MB LOSS NWISS Yr02 Yr03 Yr04 = if the firm switches to a smaller auditor; otherwise = the largest owner’s shareholding as a percentage of total shares = number of SB members = if the CEO also holds the position of the BoD chairman; otherwise = if the largest shareholder is a government agency; otherwise = if the firm receives an unclean (non-standard) auditor opinion for the previous year; otherwise = log of total assets at the end of the previous year = long-term liabilities divided by total assets at the end of the previous year = market-to-book ratio at the end of the previous year, calculated as the market value of stocks divided by the book value = if the firm experiences a net loss for the previous year; otherwise = if there is a new equity issue in the two years immediately after auditor switching; otherwise = if the switching occurs in Year 2002; otherwise = if the switching occurs in Year 2003; otherwise = if the switching occurs in Year 2004; otherwise Notes: (1) The Wald-Wolfowitz test examines whether the mean (median) of each variable for Top 10 clients equals to that of non Top-10 clients ***, **, and † denote significance at the 1%, 5%, and 10% levels, respectively (2) N = 62, Chi-square = 42.02, and Pseudo R-square = 67 Pred Coeff Wald Pred Coeff Wald Intercept b0 DS = β0 + β1LSH + β2 SB + β3CEOCHR + β4GOV + β5OPI + β6 LNASSET + β7 LEV + β8 MB + β9 LOSS + β10 NWISS + β11Yr 02 ∗ LSH + β12Yr 03 ∗ LSH + β13Yr 04 ∗ LSH + β14Yr 02 ∗ SB + β15Yr 03 ∗ SB + β16Yr 04 ∗ SB + β17 Yr 02 ∗ CEOCHR + β18Yr 03 ∗ CEOCHR + β19Yr 04 ∗ CEOCHR + ε r TABLE 4-1 Internal Corporate Governance Mechanism and Switching to a Smaller Auditor – Sensitivity: Top 10 to Non-Top 10 THE DETERMINANTS OF AUDITOR SWITCHING FROM THE PERSPECTIVE OF CORPORATE GOVERNANCE IN CHINA Volume 17 Number 487 July 2009 Volume 17 Number July 2009 -.37 17 NWISS b 10 ? 6.73 2.48 00 79 ? Yr02*LSH b 11 ? -.01 1.99 Yr03*LSH b 12 01 2.14 ? ? 02 02 Yr02*SB b 14 + 1.25 3.93** CEOCHR b3 Yr04*LSH b 13 04 05 - + 02 4.17** SB b2 LSH b1 05 01 ? 08 22 Yr03*SB b 15 + GOV b4 ? 12 24 Yr04*SB b 16 + -.13 08 OPI b5 ? 27 63 Yr02*CEOCHR b 17 -.44 4.66** LNASSET b6 ? 07 19 Yr03*CEOCHR b 18 + 3.45 3.70† LEV b7 + 16 10 LOSS b9 ? 02 15 Yr04*CEOCHR b 19 13 3.57† + MB b8 (1) DS LSH SB CEOCHR GOV OPI LNASSET LEV MB LOSS NWISS Yr02 Yr03 Yr04 = if the firm switches to a smaller auditor; otherwise = the largest owner’s shareholding as a percentage of total shares = number of SB members = if the CEO also holds the position of the BoD chairman; otherwise = if the largest shareholder is a government agency; otherwise = if the firm receives an unclean (non-standard) auditor opinion for the previous year; otherwise = log of total assets at the end of the previous year = long-term liabilities divided by total assets at the end of the previous year = market-to-book ratio at the end of the previous year, calculated as the market value of stocks divided by the book value = if the firm experiences a net loss for the previous year; otherwise = if there is a new equity issue in the two years immediately after auditor switching; otherwise = if the switching occurs in Year 2002; otherwise = if the switching occurs in Year 2003; otherwise = if the switching occurs in Year 2004; otherwise Notes: (1) The Wald-Wolfowitz test examines whether the mean (median) of each variable for Top 10 clients equals to that of non Top-10 clients ***, **, and † denote significance at the 1%, 5%, and 10% levels, respectively (2) N = 230, Chi-square = 38.88, and Pseudo R-square = 21 Pred Coeff Wald Pred Coeff Wald Intercept b0 DS = β0 + β1LSH + β2 SB + β3CEOCHR + β4GOV + β5OPI + β6 LNASSET + β7 LEV + β8 MB + β9 LOSS + β10 NWISS + β11Yr 02 ∗ LSH + β12Yr 03 ∗ LSH + β13Yr 04 ∗ LSH + β14Yr 02 ∗ SB + β15Yr 03 ∗ SB + β16Yr 04 ∗ SB + β17 Yr 02 ∗ CEOCHR + β18Yr 03 ∗ CEOCHR + β19Yr 04 ∗ CEOCHR + ε r TABLE 4-2 Internal Corporate Governance Mechanism and Switching to a Smaller Auditor – Sensitivity: Ten Per Cent Cutoff 488 CORPORATE GOVERNANCE © 2009 Blackwell Publishing Ltd THE DETERMINANTS OF AUDITOR SWITCHING FROM THE PERSPECTIVE OF CORPORATE GOVERNANCE IN CHINA emerging economies, where there is a high degree of ownership concentration and the main agency relationship is between the controlling and minority shareholders Third, while prior research (Jensen & Meckling, 1976) contends that auditing can serve a major role of coping with agency problems and reducing agency costs, this function is not hailed by Chinese firms (usually with highly concentrated ownership) when financing opportunities are rare This study will be helpful for understanding the recent development of auditing and corporate governance practices in emerging markets such as China There are also important policy implications from this study The quality of independent audits and corporate disclosure is identified as an important factor for the Chinese stock market, which is in transition towards a marketoriented economy Our findings on the determinants of auditor switching in the Chinese context will shed light on how to improve firms’ corporate governance and audit monitoring to enhance the credibility of corporate reporting and to promote smooth development of the capital market As CSRC has granted licenses to Qualified Foreign Institutional Investors (QFII) to participate directly in China’s domestic stock market, the findings suggest that international investors need to be aware of the structural arrangement of corporate governance of the listed firms and the effectiveness of audit monitoring in China To bolster the confidence of the market participants, the Chinese government should promote the reform of corporate governance of the listed firms and enhance the regulators’ surveillance over the behaviors of the controlling shareholders In particular, regulators should closely watch over firms’ auditor switches, and especially those downward switches, with an aim of preventing possible expropriation of minority shareholders’ interests thereafter As a result, the auditing profession and the stock market may be able to develop smoothly in China In line with the continuing progress of the economic reforms and business restructuring, the Chinese accounting and auditing practices have moved towards internationalization rapidly in recent years Therefore the Chinese experience involving auditor switching can be borrowed by other emerging economies in their development of the independent audit function as well In future research, as more Chinese firms report their auditing fee information, more rigorous simultaneous equation methods may be used to control both the demand and © 2009 Blackwell Publishing Ltd 489 the supply sides of the independent auditing function In addition, when sufficient data become available, other corporate governance variables, such as the characteristics of the independent directors and the different committees under board of directors may be incorporated into the regression model ACKNOWLEDGEMENTS We are grateful to the editor William Judge, the Guest Editor, two anonymous reviewers, and the participants of the Symposium on Corporate Governance in China and India for their helpful comments and suggestions We acknowledge the financial support from Faculty Research Grants of Hong Kong Baptist University and from the Research Committee of University of Macau NOTES Most listed firms in China did not establish audit committees responsible for engaging auditors until 2004 or 2005 Before the establishment of audit committees, the controlling shareholders, through the management, basically made the auditor selection decisions In China, the largest owner (mainly the parent state-owned enterprise) held, on average, around 50 per cent of the total equity of listed firms in 2000 Most Chinese listed firms were restructured from former stateowned enterprises (SOEs) There are normally three types of equity holding for a Chinese listed firm, namely the state-owned shares (representing the state’s interest in the firm), the sociallegal-entity shares (mainly the interest of the parent state-owned enterprise or other social agencies), and the public shares held by institutional and individual investors, nevertheless, a major part of the total equity shares is usually controlled by the state (government agencies) and/or the parent state-owned enterprise The stock market in China is segregated A-share firms are for domestic investors whereas a small number of B-share firms are mainly for overseas investors According to the existing regulations, the financial statements of B-share firms must be audited by international auditing firms (usually the Big 5/4) There are few auditor switches for B-share firms because of the limited availability of alternative auditors In this paper, firms that issue both A- and B-shares are excluded Volume 17 Number July 2009 CORPORATE GOVERNANCE 490 APPENDIX Ranking of Auditors in China Ranking Auditor Auditor PwC Zhongtian KPMG Huazhen Deloitte Huayong EY Huaming Lixin Changjiang Yuehua Xinyongzhonghe Beijing Jingdu Jiangsu Gongzheng EY Dahua Zhongshen Zhongruihua Tianzhizixin Shanghai Zhonghua Li’anda Zhejiang Tianjian Tianjian Guangzhou Yangcheng Zhongtianhuazheng Shenzhen Pengcheng Shanghai Donghua Tianyi Hubei Daxin Shanghai Gongxin Jiangsu Suya Zhongxi Zhongxingcai Wuhan Zhonghuan Zhejiang Dongfang Beijing Zhongluhua Shanghai Shangkuai Tianjin Wuzhou Shandong Zhengyuan Shenzhen Nanfang Shenzhen Dahua Guangdong Zhengzhong Hu’nan Kaiyuan Beijing Yongtuo Yunnan Yatai Zhonglei Jiangsu Tianheng Guangdong Tianhua Beijing Tianhua 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 Ranking 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 Jiangsu Tianhua Huayan Gansu Wulian Zhejiang Wanbang Beijing Zhongzhou Huazheng Guangdong Hengxin Nanjing Yonghua Shandong Wanlu Chongqing Tianjian Xi’an Sigma Jiangsu Tianye Anhui Huapu Xiamen Tianjian Sichuan Huaxin Shandong Huide Beijing Zhongxingyu Beijing Zhongwei Shenzhen Tianjian Fujian Huaxing Beijing Zhongxing Shanghai Wanlong Shandong Tianhengxin Yatai Group Zhongtianyin Huajian Guangdong Kangyuan Shanghai Tongcheng Zhonghengxin Beijing Zhongzheng Shanghai Shangshen Liaoning Tianjian Beijing Xinghua Beijing Zhongtianheng Shandong Qianju Sichuan Hongri Zhongqin Wanxin Hebei Hua’an Beijing Zhongpingjian Sichuan Junhe Shanghai Jiahua Guangxi Xianghao (1) The rankings are based on average audit revenues of Year 2002–2004 as compiled by the CICPA (2) Auditors must be ranked among the top 100 based on revenues for each year of 2002–2004 (3) An auditor ranked higher is larger than an auditor ranked lower It is a downward switch if a listed firm switches from a higher ranked auditor to a lower ranked one, and vice versa Volume 17 Number July 2009 © 2009 Blackwell Publishing Ltd THE DETERMINANTS OF AUDITOR SWITCHING FROM THE PERSPECTIVE OF CORPORATE GOVERNANCE IN CHINA REFERENCES Ang, J S., Cole, R A., & Lin, J W 2000 Agency costs and ownership structure Journal of Finance, 55: 81–106 Carpenter, C G & Strawser, R H 1971 Displacement of auditors when clients go public Journal of Accountancy, 131(6): 55–58 Chan, K H., Lin, K Z., & Mo, P L 2006 A political-economic analysis of auditor reporting and auditor switches Review of Accounting Studies, 11(1): 21–48 Chen, J J 2005 China’s institutional environment and corporate governance In M Hirschey, J Kose, & A K Makhija (Eds.), Corporate governance: A global perspective, vol 11: 75–93 Boston: JAI Copley, P A & Douthett, E B., Jr 2002 The association between auditor choice, ownership retained, and earnings disclosure by firms making initial public offerings Contemporary Accounting Research, 19(1): 49–75 CSRC 2002 Standard code on corporate governance for publicly listed companies (in Chinese) Beijing, China CSRC 2005 An introduction to China’s securities and futures markets (in Chinese) Beijing, China Dahya, J., Karbhari, Y., Xiao, J Z., & Yang, M 2003 The usefulness of the supervisory board report in China Corporate Governance: An International Review, 11: 308–321 DeAngelo, L E 1981 Auditor size and audit quality Journal of Accounting and Economics, 3: 183–199 DeFond, M L., & Subramanyam, K R 1998 Auditor changes and discretionary accruals Journal of Accounting and Economics, 25: 35–67 DeFond, M L., Wong, T J., & Li, S 2000 The impact of improved auditor independence on audit market concentration in China Journal of Accounting and Economics, 28: 269–305 Denis, D K & McConnell, J J 2003 International corporate governance Journal of Financial and Quantitative Analysis, 38(1): 1–36 Felo, A J., Krishnamurthy, S., & Solieri, S 2003 Audit committee characteristics and the quality of financial reporting: An empirical analysis Working paper, Penn State University Francis, J R & Wilson, E R 1988 Auditor changes: A joint test of theories relating to agency costs and auditor differentiation Accounting Review, 63: 663–682 Friedlan, J M 1994 Accounting choices of issuers of initial public offerings Contemporary Accounting Research, 11(1): 1–31 Holm, C & Laursen, P B 2007 Risk and control developments in corporate governance: Changing the role of the external auditor? Corporate Governance: An International Review, 15: 322–333 Jensen, M C & Meckling, W H 1976 Theory of the firm: Managerial behavior, agency costs, and ownership structure Journal of Financial Economics, 3: 305–360 Jiraporn, P., Young, S K., & Davidson, W N., III 2005 CEO compensation, shareholder rights, and corporate governance: An empirical investigation Journal of Economics and Finance, 29: 242–258 Johnson, S., La Porta, R., Lopez-de-Silanes, F., & Shleifer, A 2000 Tunneling American Economic Review, 90(2): 22–27 © 2009 Blackwell Publishing Ltd 491 Johnson, W B & Lys, T 1990 The market for audit services: Evidence from voluntary auditor changes Journal of Accounting and Economics, 12: 281–308 Karpoff, J M., Malatesta, P H., & Walkling, R A 1996 Corporate governance and shareholder initiatives: Empirical evidence Journal of Financial Economics, 42: 365–395 La Porta, R., Lopez-de-Silanes, F., & Shleifer, A 1999 Corporate ownership around the world Journal of Finance, 54: 471–517 La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R W 2002 Investor protection and corporate valuation Journal of Finance, 57: 1147–1170 Lennox, C S 2005 Audit quality and executive officers’ affiliation with CPA firms Journal of Accounting and Economics, 39: 201– 231 Lin, Z J., Tang, Q L., & Xiao, J 2003 An experimental study of users’ responses to qualified audit reports in China Journal of International Accounting, Auditing, and Taxation, 12(1): 1–21 Lin, Z J., Liu, M., & Zhang, X 2007 The development of corporate governance in China The Company Lawyer, 28(7): 195–203 Pae, S & Yoo, S W 2001 Strategic interaction in auditing: An analysis of auditors’ legal liability, internal control system quality, and audit effort Accounting Review, 76: 333–356 Petra, S T 2006 Corporate governance reforms: Fact or fiction? Corporate Governance: An International Review, 6: 107–116 Reed, B J., Trombley, M A., & Dhaliwal, D S 2000 Demand for audit quality: The case of Laventhol and Horwath’s auditees Journal of Accounting, Auditing and Finance, 15(2): 183–198 Sabherwal, S & Smith, S D 2008 Concentrated shareholders as substitutes for outsider analysts Corporate Governance: An International Review, 16: 562–577 Sainty, B J., Taylor, G K., & Williams, D D 2002 Investor dissatisfaction toward auditors Journal of Accounting Auditing and Finance, 17(2): 111–136 Tam, O K 2000 Models of corporate governance for Chinese companies Corporate Governance: An International Review, 8: 52–64 Titman, S & Trueman, B 1986 Information quality and the valuation of new issues Journal of Accounting and Economics, 8: 159–172 Watkins, A L., Hillison, W., & Morecroft, S E 2004 Audit quality: A synthesis of theory and empirical evidence Journal of Accounting Literature, 23: 153–193 Watts, R L & Zimmerman, J L 1986 Positive Accounting Theory Englewood Cliffs, NJ: Prentice-Hall Z Jun Lin is a professor of accounting in the School of Business Administration, Hong Kong Baptist University His research interests are financial accounting, international accounting, auditing, and corporate governance Ming Liu is an assistant professor of accounting in the Faculty of Business Administration, University of Macau His research interests are auditing, capital market efficiency, and corporate governance Volume 17 Number July 2009 .. .THE DETERMINANTS OF AUDITOR SWITCHING FROM THE PERSPECTIVE OF CORPORATE GOVERNANCE IN CHINA costs of raising capital and hiring a low-quality auditor to maintain the gains from the opaqueness... r TABLE Internal Corporate Governance Mechanism and Switching to a Smaller Auditor – Main Test THE DETERMINANTS OF AUDITOR SWITCHING FROM THE PERSPECTIVE OF CORPORATE GOVERNANCE IN CHINA Volume... concentration (shareholding of the largest owner); the effectiveness of © 2009 Blackwell Publishing Ltd THE DETERMINANTS OF AUDITOR SWITCHING FROM THE PERSPECTIVE OF CORPORATE GOVERNANCE IN CHINA supervisory