Sign up

The impact of audit firm size and locality on audit fees in an emerging economy: evidence from China.
Abstract:
This paper examines if audit firm size and location explain audit fees for audit firms in China. Using a sample of firms in the A share market, we find that local auditors, in general, do not charge significantly different audit fees as compared to non local auditors. However, after partitioning the local audit firms into large and small local auditors, we find State-owned Enterprises (SOEs) are more likely to choose small local auditors instead of large and non local audit firms. This result is consistent with prior literature that report SOEs have lower demand for reputable (presumably large or non local) auditors to signal their quality operation due to the preferential treatment SOEs receive from governments or state banks (Wang et al. 2009). As for audit fees, we find large local audit firms charge significantly higher fees to their clients than non local auditors. Small local auditors, on the other hand, offer fee discount to local SOEs, but not to central SOEs. The results have important implications for audit firms as well as for the Chinese standard setters.

Article Type:
Report
Subject:
Emerging markets (Forecasts and trends)
Economic conditions (Forecasts and trends)
Accounting firms (Management)
Authors:
Wang, Kun
O., Sewon
Chu, Baoping
Pub Date:
04/01/2012
Publication:
Name: Journal of Academy of Business and Economics Publisher: International Academy of Business and Economics Audience: Academic Format: Magazine/Journal Subject: Business; Business, general; Economics; Government Copyright: COPYRIGHT 2012 International Academy of Business and Economics ISSN: 1542-8710
Issue:
Date: April, 2012 Source Volume: 12 Source Issue: 4
Topic:
Event Code: 010 Forecasts, trends, outlooks; 200 Management dynamics Computer Subject: Market trend/market analysis; Company business management
Product:
Product Code: 8930000 Accounting & Auditing Services NAICS Code: 54121 Accounting, Tax Preparation, Bookkeeping, and Payroll Services SIC Code: 8721 Accounting, auditing, & bookkeeping
Geographic:
Geographic Scope: Texas; China Geographic Code: 1U7TX Texas; 9CHIN China
Accession Number:
312015187
Full Text:
1. INTRODUCTION

Auditor choice is an important strategic decision for companies that can help to solve agency problem and signal firm value (Fama 1980; Titman and Trueman 1986; Beatty 1989). It also directly affects the magnitude of audit fee. The purpose of this study is to investigate the joint effect of audit firm size and locality on audit fees in China, which is potentially one of the world's largest audit markets. In particular, we are interested in the following questions: Do companies pay different audit fee to auditors that are located in the same region (i.e., local auditors) than they do for non local auditors? Do companies pay different audit fees to large local auditors than they do for small local auditors?

Inspired by the China's fast growing economy and audit firms' international focus, China's audit market is playing an increasingly important role in the global accounting world. A productive stream of research have explored the unique features of the Chinese auditing profession in terms of audit fees (Li, Song, & Wong 2005; Chen, Su, & Wu 2007), audit opinion (Chen, Su, & Zhao 2000), auditor choice (Wang, Wong, & Xia 2008), and auditor industry specialization (Wang, O, & Iqbal 2009, Wang, O, & Iqbal 2010). However, much of these studies have employed aggregated firm-level data in the analyses, but overlooked the fact that local offices often enjoy significant autonomy over contracting and other audit decisions. Empirical studies focused on mature capital market, have demonstrated that market leadership calculations based on market share data at the national level do not accurately portray city-specific audit markets for about 70 percent of audits in their sample (e.g., Francis, Stokes, & Anderson 1999). In a similar vein, Penno and Walther (1996) note that local concentration measures could be more appropriate than the previously documented national measures for accounting services. Our study attempts to take a local perspective and examines whether audit fees in China are impacted by the location of audit firms.

In a sample of 1,028 publicly listed companies, we find that local auditors, in general, do not charge significantly different audit fees as compared to non local auditors. However, after partitioning the local audit firms into large and small local auditors, we find State-owned Enterprises (SOEs) are more likely to choose small local auditors instead of large and non local audit firms. This result is consistent with prior literature that report SOEs have lower demand for reputable (presumably large or non local) auditors to signal their quality operation due to the preferential treatment SOEs receive from governments or state banks (Wang et al. 2009). As for audit fees, we find large local audit firms charge significantly higher fees to their clients than non local auditors. Small local auditors, on the other hand, offer fee discount to local SOEs, but not to central SOEs. A possible explanation is that large local auditors are more capable of offering specialized local service than small auditors, and as a result, earn fee premiums from the differentiated service.

The issues addressed in this study are important at a time when the Chinese audit market is undergoing major changes due to mergers and acquisitions, altering the underlying market structure for audit services. Our findings that audit fees are affected by the location and the size of audit firms point to a dimension that may need to be considered by government officials when carrying out further plans for restructuring. Results from our study are also of potential relevance to accounting firms and listed companies, as the Chinese government is undertaking a wide range of initiatives to address accounting and auditing issues specific to the public sector.

The remainder of the paper is structured as follows. The next section offers background information regarding the audit market in China and reviews prior literature. Section three explains sample selection and data collection. Empirical results and analysis are provided in Section four. The final section presents conclusions and limitations of the study.

2. BACKGROUND AND HYPOTHESES

2.1 CPA Profession and Auditing Standards in China

Chinese Certified Public Accountants (CPAs) did not exist as a profession until 1980 when the government issued its first regulation on practicing accountants to meet the country's urgent need for direct foreign investments. The resurgence of the stock markets since the early 1990s further expanded the market for independent auditing in China and led to the promulgation of the first set of Independent Auditing standards (AS) in 1995, followed by the second and third set of standard enacted in 1997 and 1999 respectively. Overall, these AS were patterned after the generally accepted international auditing standards with some modifications.

In China, all CPA firms were initially established by, and affiliated with, a government-related organization (a government agency, a university, or a large state-owned enterprise or SOE). After the establishment of stock enterprise and opening of stock exchanges, the Chinese Institute of Certified Public Accountants (CICPA) required that all firms should gradually sever ties with their parent organizations in order to protect professional independence from undue interference. After cutting such ties, CPA firms are reorganized as private professional firms in the form of either a partnership or a limited liability company. Among the pool of entire Chinese CPA firms, only 105 were authorized to audit listed companies at the end of 1997. The practice of these firms is closely monitored by both the CICPA and Chinese Securities Regulatory Commission (CSRC), which have the power to revoke their licenses to audit listed companies if they violate standards.

China's publicly listed companies are the most sought-after clients of China's audit firms. Among the listed companies, the majority are former State-owned Enterprise (SOEs) with the government (both central and local) holding a certain percentage of shares (People's Daily 2006). In recent years, although the state remains the controlling shareholder, most listed companies have been granted considerable latitude in making business decisions, including the selection of auditors. Consequently, auditor selection and accounting choices within Chinese GAAP are now made at the discretion of the management. Given the small number of listed companies (1,309 at the end of 2006) relative to the number of CPA firms authorized to audit them, the Chinese auditing market is highly competitive.

In order to tackle competition from international accounting firms, CICPA is engaged in aggressive reform of the auditing profession by improving audit integrity and quality of local auditors (People's Daily 2006). One of its new policies states that, in the next five to ten years, China should cultivate ten accounting firms capable of operating internationally to support domestic companies that go global, and another 100 firms big enough to serve large domestic enterprises. To this end, China's local accounting firms have undergone more market-driven mergers and acquisitions, or partnerships with second-tier international auditors (e.g., Horwath and BDO). For example, seven Chinese local firms have partnered with Horwath since 2001, including Shanghai Shulun Pan CPAs, one of the largest local CPA firms in China. By capturing the market shares of its partners, Horwath became the largest audit firm immediately following the Big 4 operations in the Chinese market.

2.2 Research Hypotheses

There is a growing body of research that examines the Chinese auditing market in terms of audit fee (Chen, Su, & Wu 2007), audit opinion (DeFond et al. 2007), auditor choice (Wang, Wong, & Xia 2008), and auditor industry specialization (Wang, O, & Iqbal 2009; Wang, O, & Iqbal 2010). The main findings based on these research include: 1) Big 4 audit firms normally charge higher audit fees for their brand reputation, especially in the less competitive B share market; 2) industry specialized auditor could also earn fee premiums due to differentiated service quality; 3) the market share of the Top-10 audit firms, which presumably provide better quality and more independent audits, has been declining; and 4) SOEs controlled by local governments are more likely to hair small auditors within the same region (small local auditors) since they enjoy preferential access to capital and they can receive government bailout.

The findings from prior research are indeed valuable. However, much of these studies have employed aggregated firm-level data in their analyses, even though local offices often enjoy significant autonomy over contracting and other audit decisions. Empirical studies focused on mature capital market, have demonstrated that market leadership calculations based on market share data at the national level do not accurately portray city-specific audit markets for about 70 percent of audits in their sample (e.g., Francis et al. 1999). In a similar vein, Penno and Walther (1996) note that local concentration measures could be more appropriate than the previously documented national measures for accounting services. Our study attempts to take a local perspective and examine whether audit fees in China are impacted by the location of audit firms.

In their auditor choice research, Wang et al. (2009) provide two arguments that explain why Chinese listed companies prefer to choose local audit firms, namely the local knowledge argument and the collusion argument. Based on the demand argument, many local auditors were closely affiliated with their local governments until they separated themselves from the governments in 1998. After the separation, however, most accountants remained in their audit firms, carrying with them specialized knowledge of the companies that are in the same regions. This tendency is perhaps even stronger for local SOEs because local auditors have specialized knowledge of local government units that supervise or hold ownership stakes in the firm. In this case, we expect that a local knowledge advantage is likely to be important in auditor fee decisions. Prior research report that industry specialized auditors earn fee premiums due to differentiated service in the Chinese market (Wang et al. 2009, 2010). Therefore, our first hypothesis states:

H1: If the primary reason for companies to hire a local auditor is to seek specialized local knowledge, the benefits derived from differentiated service quality will enable local audit firms to charge higher audit fees.

On the other hand, the collusion argument suggests Chinese listed firms have incentives to hire acquiescent auditors to facilitate their meeting the CSRC earnings targets for IPO and seasoned equity offerings or to avoid delisting. An acquiescent auditor would allow its client to manipulate earnings by not issuing a modified opinion that may lead to a share price decline and trigger government sanctions such as delisting and loss of qualifications for seasoned equity offerings. In order for the companies to get these favorable treatments from the incumbent auditors, we expect them to pay higher fees. Therefore, our second hypothesis states:

H2: If the primary reason for companies to hire a local auditor is to seek favorable audit opinion, local auditors will charge higher audit fees.

3. SAMPLE AND RESEARCH DESIGN

3.1 Sample Selection

To test the hypotheses, we collected data of all 1,309 Chinese companies that were listed in the A share market in 2006. We hand-collected company background, audit fee, and financial data from each company's annual report published by the Shanghai and Shenzhen Stock Exchanges. After removing firms with missing data (e.g., primarily on audit fee and auditor tenure), the final sample includes 1,028 firms. The industry breakdown for the final sample is reported in Table 1. The top five industries in our sample are Electronics and other Electrical Equipment (10.31%), Chemicals and Allied Products (10.12%), Industrial and Commercial Machinery (7.39%), Primary and Fabricated Metals (6.62%), and Real Estate (6.32). The remaining 32 industries consist 59.24% of the sample firms.

To identify local vs. non-local auditors, we used a similar approach employed by Wang et al. (2009). That is, a listed company is considered to have hired a local auditor if the audit firm is located in the same province (or region with provincial status, that is, autonomous administrative region or municipality under the central government) as the listed firm. Since in China, the signing auditors, not the affiliates, bear all legal and government sanction risks, we identify auditors based on the signing auditors, not the affiliates.

In addition, when auditors from two or more regions merge to form a new auditor, we treat the original registry regions of the merging auditors and the new registry region of the merged firm as the registry regions of the new audit firm. In other words, if a listed firm hires an auditor that results from a merger of auditors from different regions, this firm is considered as hiring a local auditor if the firm is located in any one of the registry regions of the new auditor. For example, if a Beijing audit firm merges with a Shanghai audit firm, the new merged firm will have a Beijing office and a Shanghai office and is considered a local auditor for any Beijing or Shanghai client firm. We classify a local auditor in this way because the new auditor can potentially be influenced by the local governments of both the new registry region of the merged firm and the registry regions of the original firms prior to the merger.

Next, we classify our sample audit firms as small vs. Top-10 because we also investigate whether large and small local auditors charge different audit fees. Following DeFond et al. (1999), we classify audit firms as small if they are not a Top-10 auditor based on total assets audited. We rank Top-10 auditors on a national rather than regional basis because the size of the top auditors of each province varies significantly. (1) Also, it is rather easy for the market to distinguish national Top-10 auditors from other auditors as the CSRC publishes an annual national Top-10 auditors list, which is relatively stable with at most two firms being removed from the list annually between 1993 and 2003. Table 2 lists the top-10 audit firms in China ranked by the CICPA based on revenue, audit quality, # of CPAs, the education background of CPAs, and other factors (CICPA 2009) (2).

3.2 Research Design

Using a standard audit fee model (Craswell and Francis 1999; Mayhew and Wilkins 2003; Chen et al. 2007; Cahan et al. 2008), we examine the determinants of audit pricing after controlling for the effects of client size, audit complexity, auditor-client risk sharing, and ownership structure that is unique to the Chinese Market. In Model (1), we assess whether the local audit firms are associated with significantly different audit fees. In Model (2), we further partition the local auditors based on their size and examine whether large and small local auditors implement similar pricing strategy.

Audfee=[b.sub.0] + [b.sub.1]Assets + [b.sub.2]Invrec + [b.sub.3]Sub + [b.sub.4]ROA + [b.sub.5]Opinion + [b.sub.6]Tenure + [b.sub.7]Leverage + [b.sub.8]Stateshr + [b.sub.9]Legalshr + + [b.sub.10]Local + [r.sub.1]Fixedindustry + e (1)

Audfee=[b.sub.0] + [b.sub.1]Assets + [b.sub.2]Invrec + [b.sub.3]Sub + [b.sub.4]ROA + [b.sub.5]Opinion + [b.sub.6]Tenure + [b.sub.7]Leverage + [b.sub.8]Stateshr + [b.sub.9]Legalshr + [b.sub.10]Biglocal + [b.sub.11]Smallocal + [r.sub.1]Fixedindustry + e (2)

where:

Audfee = natural log of total audit fee.

Assets = natural log of total assets.

Invrec = (accounts receivables + inventory)/total assets.

Sub = square root of number of consolidated subsidiaries.

ROA = return on assets.

Opinion = indicator variable (1 if modified opinion, 0 otherwise).

Tenure = natural log of auditors' tenure in years.

Leverage = total liabilities/total assets.

Stateshr = the % ownership of the state government.

Legalshr = the % ownership of the legal person.

Local = indicator variable (1 if a company is located in the same regions as its auditors, 0 otherwise).

Biglocal = indicator variable (1 if a company is audited by a top10 local auditor, 0 otherwise).

Samllocal = indicator variable (1 if a company is audited by a non top10 local auditor, 0 otherwise).

e = error term with a normal distribution.

With respect to the control variables, Assets is a proxy for client size, Invrec is a proxy for audit risk, and Sub is a proxy for audit complexity. We expect audit fee to have positive relationships with client size, audit risk, and audit complexity since higher values of these variables increase the workload and riskiness of the audit work. ROA is a proxy for firm profitability and we expect audit firms to require higher fees if the company has a lower return on assets. While the association between audit fee and Opinion is inconclusive for the developed audit market (Craswell, Francis, & Taylor 1995; Craswell and Francis 1999), we predict the association to be negative for the Chinese market because Chen et al. (2007) found Chinese listed companies receiving modified opinions tend to be smaller, poor financial performers, and unable to pay higher fees. We are unclear about the sign of TENURE since prior literature recognizes two opposing effects on audit fees from auditor tenure. On one hand, auditors with longer tenure tend to extract higher fees (i.e., future quasi-rents) from clients to recover losses incurred due to low-balling. On the other hand, longer tenure enhances auditors' understanding of the clients, enabling auditors to design efficient audit procedures and enjoy cost savings. Leverage is a proxy for audit risk and is calculated as the ratio of total liabilities to total assets. The audit fees are expected to be higher when companies have higher leverage ratio.

We also control two ownership variables that are unique to institutional environment of the Chinese stock market. Stateshr is the percent ownership from the state agencies. Prior studies argue state ownership representatives lack a direct personal stake in the company's profits, and they are more likely to hire small local auditors because they don't have high demand of audit quality (Wang, Wong, and Xia 2008). As a result, we expect audit fees will be lower if the company has more state ownership. On the other hand, Legalshr is the percent ownership from legal person(s) that are more motivated to monitor firms because they are geared more toward profit-making than fulfilling political and social goals. This inference is supported by empirical evidence that legal person ownership is positively associated with corporate performance and voluntary disclosure on the Internet (Xiao, Yang, and Chow 2004). As legal person(s) have more resources and expertise to monitor the firm management, they are more likely to hire high quality auditors and pay higher audit fees.

3.3 Descriptive statistics

The first two columns of Panel A of Table 3 presents descriptive statistics for the sample firms that have complete information (1,028 firms). In the A share audit market, the mean and median total assets (Assets) are RMB 4,298 million and RMB 1,674 million, respectively. The mean and median total audit fee (Audfee) charged by the audit firms are RMB 571 thousand and RMB 410 thousand, respectively. In addition, the data show that accounts receivables and inventory (Invrec) are about 27% of total assets and that an average sample firm has 8.73 consolidated subsidiaries (Suo). The findings on ROA show the average return on assets for the sample companies are 2.4%. Untabulated data also show that about 9.3% of the firms experienced financial loss. During the sample period, 7% of the companies received modified opinions (Opinion), and the average tenure period for engaged auditors (Tenure) is 6.3 years. The total liabilities of the sample companies are about 52% of the total assets as indicated by Leverage. The state and legal person account for 11% and 19% of the total ownership of the companies, respectively. Finally, around 79% of the sample firms choose local auditors, and among them, 58% choose top10 local auditors and 21% choose small local auditors.

The next two columns in Panel A present the summary statistics for the sample based on the partition of local and non local auditors. It seems that local and non local audit firms, in general, do not charge significantly different audit fees. The clients of local auditors, however, have significantly larger assets, more operational subsidiaries, longer auditor tenure periods, less modified audit opinions, and higher state ownership. Finally, we report the mean and median values of the interested variables separated by large and small local audit firms in the last two columns. This time we find large local auditors are associated with companies that are substantially larger, own more subsidiaries, pay higher audit fees, but have relatively lower state ownership. Taken together, the descriptive results are consistent with the findings in Wang et al. (2008) and suggest companies with higher state ownership (e.g., SOEs) have preference for small local auditors due to potential benefits such as less modified opinions. Another explanation is that SOEs have lower demand for reputable (presumably large or non local) auditors to signal of their quality operation due to the preferential treatment SOEs receive from governments or state banks.

4. EMPIRICAL RESULTS AND ANNLYSES

4.1 The Main Analyses

In Table 4, we present the multivariate results for the impact of audit firms' location on the magnitude of audit fee. The F-statistics are significant at p < 0.00, implying that the independent variables explain a significant portion of the variance in audit fee. The adjusted [R.sup.2] for the model is 0.55, which is similar to that reported for U.S. and Australia, but much higher than the 0.25 adjusted [R.sup.2] reported for China's B share market in Wang et al. (2009). To examine potential multicollinearity in the regression model, we regress all the explanatory variables on Audfee. These results indicate that the variance inflation factor (VIF) is below 1.68 and tolerance levels are above 0.82 for all the explanatory variables. This result suggests that multicollinearity between the explanatory variables is not likely to pose a serious problem in our interpretation of the regression results. We also remove outliers from the sample firms if they have extreme variable values (i.e., rstudent >=3).

Among the control variables, the coefficient of Assets is positive and significant, which is consistent with the findings on the positive firm size--audit fee relation documented in earlier studies (DeFond et al. 2000; Ferguson, Francis, & Stokes 2003; Mayhew and Wilkins 2003). The coefficient for Sub is also positive and significant, suggesting that audit firms charge higher fees for clients with a large number of subsidiaries. The coefficient for ROA is negative and significant, indicating that audit firms lower their prices if the clients are more profitable in the audited period. The percentage of inventory and receivables (Invrec) is negative and inconsistent with U.S. results. This is not surprising, as in China, inventory is less associated with audit risk because listed firms prefer to use methods other than inventory manipulation to manage earnings. We fail to find any conclusive evidence on auditor opinion (Opinion), auditor tenure (Tenure), and Leverage. For the two ownership variables, state ownership (Stateshr) is slightly positive, suggesting the influence of state agencies on auditor choice and audit fees is not significant. The legal person ownership (Legalshr), however, is substantially associated with increased audit fees. This result implies Chinese listed companies are motivated to select higher quality auditors in order to retain the investment from legal person shareholders.

As for the interested variable (LOCAL) in equation (1), we find the coefficient is positive but statistically insignificant (t = 0.86, p = 0.39) after controlling for other factors. It indicates that local auditors, in general, do not charge significantly different audit fees as compared to non local auditors.

In Equation (2), we separate local audit firms into top10 and small local auditors to examine whether the higher fees associated with local auditors are related to the firm reputation or firm size. This time we find the parameter for top10 local (0.11) auditors is significantly positive (t = 2.95, p = 0.003), suggesting that top10 local audit firms charge higher fees to their clients than non local auditors. On average, this result translates to top10 local auditors having a premium of 11.63 percent over the comparison group. (3) In contrast to the top10 local auditors, the parameter for small local auditors (-0.01) is negative but insignificant (t = -0.17, p = 0.86). This result indicates that small local auditors cannot charge as high as non local auditors.

In summary, the overall findings are consistent with the local knowledge argument in Wang et al. (2009) rather than the collusion argument because top10 auditors are presumably more capable of providing specialized local service than small auditors. In addition, it has been indicated in the literature that large auditors normally provide more independent audits (DeAngelo 1981). If the collusion argument dominates, we would observe small local auditors charge higher fees since it is unlikely to collude with large auditors. Therefore, our Hypothesis 1 is supported and Hypothesis 2 is rejected as Chinese listed companies have incentives to hire top local auditors in order to seek local specialized service. The differentiated service quality associated with specialized knowledge enables top10 local auditors to earn fee premiums from their clients.

4.2 The Interaction of Small Local Auditors with Central and Local SOEs

The descriptive statistics in Table 3 suggest Chinese listed companies with dominant state ownership (i.e., SOEs) are more likely to choose small local auditors. To further examine the impact of state ownership on the audit fees charged by small local auditors, we perform the analysis with a focus on small local auditors and two interaction variables, local auditor x Central SOEs and small local auditor x Local SOEs. Based on Wang et al. (2009), Central SOEs are owned by the central government (e.g., the Ministry of Finance and Central Industrial Enterprise Administration Committee), while local SOEs are owned by local governments (e.g., The Bureau of State Assets Management and the Finance Bureau). If two or more types of owners control a listed firm, we classify the firm's ownership type based on the identity of the owner that has the largest ownership control in the firm.

As shown in the third column of Table 4, the coefficient on small local auditor (Smallocal) is insignificantly negative, while the coefficient on small local auditor x Central SOEs is not significant and that on small local auditor x Local SOEs is significantly negative. The findings indicate that, in general, small local auditors do not offer a fee discount. However, compared with non-state firms, small local auditors significantly reduce their audit fees to local SOEs but not to central SOEs.

4.3 Further Analyses of Local Audit Firms

Among the listed companies in our sample, some hire local auditors that are headquartered in the same region as the company, and others hire local auditors that are branch offices of a non local audit firm. To explore whether audit fees are related to the headquarter or a local branch office of auditors, we re-examine the audit fee model using only the 778 companies audited by local audit firms and added an indicator variable for headquarter offices. The results, reported in the last column of Table 4, suggest that headquarter offices of local auditors charge significantly higher fees than branch offices. A possible reason is that headquarters are presumably able to offer more specialized local service to their clients given the large scale of human and technology resources available.

4.4. Sensitivity analysis

We perform sensitivity analyses to test the robustness of our results. First, the relation between client size (log of client assets) and audit fees may not follow a non-linear relation. We conduct a model by adding the square root of client assets and client assets in the audit fee model (using audit fees instead of log of audit fees as the dependent variable); the main results remain the same. The coefficients on the square root of client assets and on client assets are significantly positive. The first column of Table 5 reported results using the square root of client assets as a proxy for client size.

In addition, the fixed industry effect variables reveal there are variations of audit fees among different industries. Specifically, the top two industries in our sample, Electronics and Petroleum's/Chemicals that comprise 20 percent of the sample firms, could have affected the regression results. To further control the industry influence, we re-estimate the fee model with these two industries removed from the sample. The findings reported in the second column of Table 5 show that, the signs and magnitudes on all of the experimental variables basically remain unchanged with the smaller sample size. It suggests that main results are not driven by particular industries in the sample.

Audfee = natural log of total audit fee.

Assets = natural log (square root) of total assets.

Invrec = (accounts receivables + inventory)/total assets.

Sub = square root of number of consolidated subsidiaries.

ROA = return on assets.

Opinion = indicator variable (1 if modified opinion, 0 otherwise).

Tenure = natural log of auditors' tenure in years.

Leverage = total liabilities/total assets.

Stateshr = the % ownership of government.

Legalshr = the % ownership of the legal person(s).

Biglocal = indicator variable (1 if a company is audited by a top10 local auditor, 0 otherwise).

Smallocal = indicator variable (1 if a company is audited by a non top10 local auditor, 0 otherwise).

Finally, we tested whether our results are sensitive to the inclusion of Big 4 audit firms. After removing Big 4 clients from the sample, we run the model and find our results did not change with the magnitude of top local auditors being reduced very slightly. The new findings are reported in the third column of Table 5.

5. CONCLUSIONS AND LIMITATIONS

This study investigates the audit market in China's transitional economy using data from annual reports prepared by publicly traded companies. We examine variables that explain audit fees for Chinese local audit firms with a focus on audit firm size and location. The results support our expectations that the development of industry specialization for Chinese local audit firms have unique features than that for the Big 4 firms in the western regions. In particular, we find that large local auditors, not small ones, charge significantly higher audit fees than non local auditors. This result supports the demand argument that large auditors are more capable of providing specialized local knowledge, and as a result, earn fee premiums from differentiated service quality. The findings of our study will enhance understanding of the audit markets in China, and help Chinese standard setters in their efforts to nurture a robust and efficient audit market.

This study is not without limitations. First, since hand collecting data for the Chinese audit market was a long and tedious process, we limited our sample to the most recent year which was year 2006 at the time of data collection. This excluded new mergers and acquisitions among top local audit firms and new branches set up by the Big 4 firms after 2006. Our findings, therefore, do not reflect new changes in the market position and ranking of the accounting firms and their impacts on auditor industry specialization and audit pricing. Second, since both audit firms and public companies in the Chinese environment have unique features, such as the stronger government and regional/geographical influences in the selection of audit firms, our findings may not be generalized to other audit markets. Replications of audit fee models in other national settings warrant potential research extensions of this paper.

REFERENCES:

Beatty, R. P., "Auditor reputation and the pricing of initial public offering", The Accounting Review, 64, 1989, 693-709.

Chen, C. J. P., X. Su, & X. Wu, "Market competitiveness and Big 5 pricing: Evidence from China's binary market", The International Journal of Auditing, 42, 2007, 1-24.

Chen, C. J. P., X. Su, & R. Zhao, "An Emerging Market's Reaction to Initial Modified Audit Opinions: Evidence from the Shanghai Stock Exchange", Contemporary Accounting Research, 17 (3), 2000, 429-455.

Cahan, S., J. M. Godfrey, J. Hamilton, & D. Jeter, "Auditor specialization, auditor dominance, and audit fees: The role of investment opportunities". The Accounting Review, 83 (6), 2008, 1393-1423.

CICPA. 2009. Available at: http://www.cicpa.org.cn/top100/top2009.htm

Craswell, A., J. R. Francis, & S. Taylor, " Auditor brand name reputation and industry specializations" Journal of Accounting and Economics, 20, 1995, 297-322.

Craswell, A., and J. R. Francis, "Pricing initial audit engagements: A test of competing theories", The Accounting Review, 74 (2), 1999, 201-216.

DeAngelo, L.," Auditor Size and Audit Quality", Journal of Accounting and Economics, 3, 1981, 183-199.

Fama E. F.," Agency Problems and the Theory of the Firm", The Journal of Political Economy, 88 (2), 1980, 288-307.

Francis, R., D. J. Stokes, & D. Anderson., "City Markets as a Unit of Analysis in Audit Research and the Re-examination of Big 6 Market Shares", Abacus, 35 (2), 1999, 185-206.

Li, C., F. M. Song, & S.M. Wong, "Audit firm size effects in China's emerging audit market", Hong Kong Institute of Economics and Business Strategy Working Paper No. 1114, 2005.

Mayhew, B., and M. Wilkins, "Audit firm industry specialization as a differentiation strategy: Evidence from fees charged to firms going public", Auditing: A Journal of Practice & Theory, 22 (2), 2003, 33-52.

Penno, M., and B. Walther, "The Concentration of Local Markets: A Study of Accounting, Advertising and Law", Accounting Horizons 10, 1996, 88-99.

People's Daily, Accounting firms reposition for international business. People's Daily online, May 30, 2006.

Titman, S. and B. Trueman, " Information quality and the valuation of new issues", Journal of Accounting & Economics, 8 (2), 1986, 159-172.

Wang, K., S. O, & Z. Iqbal, "Audit pricing and auditor industry specialization: Evidence from China", Journal of International Accounting, Auditing, & Taxation, 18 (1), 2009, 60-72.

Wang, K., S. O, & Z. Iqbal, "Non Big 4 Audit Market in An Emerging Economy: Does Industry Specialization Pay?", 2010, Working paper.

Wang, Q., T. J. Wong, and L. Xi, "State ownership, the institutional environment, and audit choice: Evidence from China", Journal of Accounting & Economics, 46, 2008, 112-134.

Xiao, Z., H. Yang, and C. Chow, "The determinants and characteristics of voluntary internet-based disclosures by listed Chinese companies", Journal of Accounting and Public Policy, 23, 2004, 191-225.

Kun Wang, Texas Southern University, Houston, Texas, USA

Sewon O, Texas Southern University, Houston, Texas, USA

Baoping Chu, Qinghai University, Xining, China

Dr. Kun Wang is an associate professor of accounting at Texas Southern University, Houston, Texas, USA.

Dr. Sewon O is an associate professor of accounting at Texas Southern University, Houston, Texas, USA

Dr. Baoping Chu is an associate professor at Qinghai University, Xining, China.

(1) For example, the mean size of the top auditor based on client assets in each of the bottom 10 provinces (with at least one local auditor) was US$5.5 billion in 2003, which is significantly smaller than the mean size of the Top-10 auditors in the country in the same year, US$43 billion.

(2) We use the 2009 data instead of the 2006 data because the ranking of the top 10 audit firms in the past five years is relatively consistent.

(3) Because model (1) is linear in logarithms, the antilog of Top10s coefficient minus 1 is the percentage effect on audit fees of choosing a top10 auditor.
Table 1
Industry Representation of the Sample Companies

Industry Description 2                 2 Digit   Number of   % of the
                                         SIC     Companies    Sample

Electronics and Other Electrical         36        106        10.31
  Equipment
Chemicals and Allied Products            28        104        10.12
Industrial and Commercial Machinery      35         76         7.39
Primary and Fabricated Metal             33         68         6.62
  Products
Real Estate                              65         65         6.32
Pharmaceuticals                          28         63         6.13
Retails                                  53         63         6.13
Utilities                                49         51         4.96
Transportation Equipment                 37         47         4.57
Transportation                           40         43         4.18
Food and Beverages                       20         43         4.18
Business Services                        73         33         3.21
Stone, Clay, Glass, and Concrete         32         31         3.02
  Products
Whole Sales                              50         27         2.63
Constructions                            15         24         2.33
Agriculture                              01         22         2.14
AH Other Industries                                162        15.76

Total Sample Firms                               1,028       100.00

Table 2
Comprehensive Evaluation of the Top 100 Accounting Firm for 2009
(Excerpt)

                        Compre     Total      Revenue       # of
                        hensive   Revenue    from Audit    CPAs *
  Accounting Firms       Rank      2008     Service (M)

PWC                        1      275518         260984      587
Ernst & Young              2      270000         226251      750
Deloitte & Touche          3      249882         170449      668
KPMG                       4      243517         154017      550
RSM China                  5       65217          55433     1013
Horwath (Shulun Pan)       6       66639          51817      679
Wanlong Asia               7       39839          30011      556
Pan-China                  8       31466          24401      339
Daxin                      9       31373          27550      360
Shinewing                 10       26153          22287      590

                        Compre     # of       # of        # of
                        hensive    Local    Employee   Partners *
  Accounting Firms       Rank     Offices

PWC                     1227.99        9       4583            2
Ernst & Young           1216.29        7       4094            2
Deloitte & Touche       1122.00        6       4371            2
KPMG                    1087.59        3       4890            2
RSM China               358.61        18       1825           38
Horwath (Shulun Pan)    344.24         9       1315           37
Wanlong Asia            216.67        18       1183           34
Pan-China               177.96         2        930           27
Daxin                   173.29         9        946           20
Shinewing               166.97         6        803           25

                                CPA Age Analysis

  Accounting Firms      [less than    40 [less    > 60
                         or equal     than or
                          to] 40     equal to]
                                         60

PWC                           569           18       0
Ernst & Young                 733           16       1
Deloitte & Touche             629           39       0
KPMG                          538           12       0
RSM China                     643          310      60
Horwath (Shulun Pan)          481          156      42
Wanlong Asia                  388          152      16
Pan-China                     298           38       3
Daxin                         286           70       4
Shinewing                     489           97       4

                                 Education of CPAs

  Accounting Firms        Some     B.S.    M.S.    Ph.D.
                        College

PWC                           5     447     133       2
Ernst & Young                14     550     182       4
Deloitte & Touche            24     457     181       6
KPMG                         10     376     164       0
RSM China                   303     621      84       5
Horwath (Shulun Pan)        280     360      39       0
Wanlong Asia                190     317      42       7
Pan-China                    26     265      46       2
Daxin                       109     211      36       4
Shinewing                   115     430      45       0

* Notes

1. Total revenue: as shown in the accounting firms'
financial statements from 2008.

2. # of CPAs: as of the end of 2008.

3. All the Big 4 firms in the table are joint ventures.
They don't have individual partners.

4. The information already reflects the mergers and
acquisitions of the accounting firms as the end of 2008.

Table 3
Descriptive Statistics and Correlation Coefficients for the Audit Fee
Model

Panel A: Descriptive Statistics

                                            Companies with Local
                   Full Sample (1,028)         Auditors (820)

Variables *        Mean        Median        Mean        Median

Audfee (000)      571.55       410.00       573.90       420.00
Assets (000)    4,298,033    1,673,678    4542324 *    1,721,778
Invrec            0.270        0.240         0.27         0.23
Sub               8.730        5.000         9.37 *       6.00
Tenure            6.300        6.000         6.64 **      7.00
ROA               0.024        0.026         0.03         0.03
Opinion           0.070        0.000         0.06 **      0.00
Leverage          0.522        0.535         0.51         0.53
Stateshr          0.110        0.000         0.11 **      0.00
Legalshr          0.185        0.090         0.18         0.09
Local              0.79         1.00
Largelocal         0.21         0.00
Smallocal          0.58         1.00

                   Companies with Non         Companies with Large
                  Local Auditors (208)        Local Auditors (218)

Variables *        Mean        Median        Mean        Median

Audfee (000)      562.24       400.00      838.34 *       500
Assets (000)    3,333,788    1,623,999    10,007,201 * 2,183,356
Invrec            0.240        0.210         0.27         0.24
Sub               6.240        5.000        11.7 *        7
Tenure            4.980        5.000         6.17         6
ROA               0.020        0.024         0.031 **     0.031
Opinion           0.090        0.000         0.055        0
Leverage          0.550        0.570         0.51         0.52
Stateshr          0.090        0.000         0.1          0
Legalshr          0.180        0.080         0.17         0.06
Local
Largelocal
Smallocal

                  Companies with Small
                  Local Auditors (602)

Variables *        Mean        Median

Audfee (000)      477.71        400
Assets (000)    2,554,270    1,515,357
Invrec             0.27         0.25
Sub                8.51         5
Tenure             6.8          7
ROA               0.024         0.025
Opinion            0.06         0
Leverage           0.52         0.53
Stateshr           0.12         0
Legalshr           0.19         0.1
Local
Largelocal
Smallocal

Panel B: Pearson Correlation Coefficients

Variables *       Audfee       Assets       Invrec        Sub

Audfee (000)        1          0.4 *         -0.1        0.23 *
Assets (000)                     1           -0.07 **    0.11 *
Invrec                                        1          0.08 *
Sub                                                        1
Tenure
ROA
Opinion
Leverage
Stateshare
Legalshare
Local
Largelocal
Smallocal

Variables *       Tenure        ROA        Opinion

Audfee (000)      -0.04        0.03        -0.04
Assets (000)      -0.005       0.054 **    -0.03
Invrec            -0.02       -0.001       -0.08 *
Sub                0.08 *      0.004       -0.09 *
Tenure              1          0.09 *      -0.06 **
ROA                              1         -0.35 *
Opinion                                       1
Leverage
Stateshare
Legalshare
Local
Largelocal
Smallocal

Variables *      Leverage     Stateshr     Legalshr

Audfee (000)      0.06 **       0.03       -0.06 **
Assets (000)      0.03          0.04       -0.08 **
Invrec            0.16 *        0           0.1 *
Sub               0.16 *       -0.006       0.02
Tenure           -0.04         -0.023       0.025
ROA              -0.38 *        0.003      -0.05
Opinion           0.28 *       -0.02        0.07 *
Leverage            1           0.0004      0.09 *
Stateshare                       1         -0.02
Legalshare                                   1
Local
Largelocal
Smallocal

Variables *       Local      Largelocal   Smallocal

Audfee (000)       0.29 *       0.2 *       -0.16 *
Assets (000)       0.21 *      -0.01 *      -0.1 *
Invrec             0.08         0.02         0.05
Sub                0.11 *       0.07 *      -0.04
Tenure             0.18 *       0.05        -0.007
ROA                0.04         0.05        -0.014
Opinion           -0.04        -0.03        -0.03
Leverage          -0.05 ***    -0.03        -0.03
Stateshare         0.02         0.017       -0.003
Legalshare        -0.004        0.02        -0.003
Local               1           0.06 *       0.26 *
Largelocal                       1          -0.62 *
Smallocal                                     1

*, **,*** indicate significance at the 0.01, 0.05, and 0.10 level,
respectively.

Table 4
Regression Results

                     Expected      Full      Large and Small
                       signs      Sample      Local Auditors

F-statistics                       37.94          37.72
Sample size                        1,014          1,014
Adjusted [R.sup.2]                 0.55            0.55

Independent
  variables *
Intercept                ?         0.93            1.03
                                   4.71 *          5.23 *
Assets                   +         0.35            0.34
                                  24.23 *         23.5 *
Invrec                   +        -0.01           -0.01
                                  -0.09           -0.14
Sub                      +         0.08            0.08
                                   8.32 *          8.26 *
Tenure                   ?         0.01            0.02
                                   0.75            1.12
ROA                      -        -0.51           -0.53
                                  -2.65 *         -2.65 *
Opinion                  +         0.00           -0.001
                                   0              -0.05
Leverage                 +        -0.05           -0.04
                                  -0.66           -0.49
Stateshr                 -         0.02            0.02
                                   0.87            0.93
Legalshr                 +         0.22            0.24
                                   3.6 *           3.6 *
Local                    +         0.03
                                   0.86
Biglocal                 ?                         0.11
                                                   2.95 *
Smallocal                ?                        -0.01
                                                   0.17
Smallocal x              +
  Central SOEs
Smallocal x              -
  Local SOEs
Origlocal

                     Central or    Companies with
                     Local SOEs    Local Auditors

F-statistics            36.12           15.53
Sample size            1014.00           779
Adjusted [R.sup.2]      0.55            0.39

Independent
  variables *
Intercept               1.01            1.64
                        5.11 *          6.01 *
Assets                  0.35            0.29
                       23.86 *         14.46 *
Invrec                 -0.02            0.02
                       -0.2             0.16
Sub                     0.08            0.07
                        8.4 *           6.18 *
Tenure                  0.04            0.04
                        1.32            1.03
ROA                    -0.51           -0.57
                       -2.64 *         -2.39 **
Opinion                -0.04           -0.03
                       -0.53           -0.42
Leverage               -0.04           -0.03
                       -0.53           -0.34
Stateshr                0.02            0.02
                        1.02            0.84
Legalshr                0.17            0.18
                        2.58 **         2.31 **
Local                   0.08
                        2.34 **
Biglocal

Smallocal              -0.03
                       -0.76
Smallocal x            -0.01
  Central SOEs         -0.13
Smallocal x            -0.07
  Local SOEs           -1.77 ***
Origlocal                               0.096
                                        2.21 **

*, **, *** indicate significance at the 0.01, 0.05, and 0.10 levels,
respectively.

Table 5
Robust Tests

                      Expected
                       signs        (1)         (2)        (3)

F-statistics                       29.77       33.27      27.07
Sample size                        1,014        806        958
Adjusted [R.sup.2]                 0.48        0.56        0.47
Independent
  variables *
Intercept                ?          5.3        0.83        1.73
                                  78.09 *     3.71 *      8.85 *
Assets                   +        0.0002       0.35        0.28
                                  19.34 *     21.47 *    19.43 *
Invrec                   +         0.02        -0.05       0.02
                                   0.17        -0.57       0.27
Sub                      +         0.12        0.07        0.08
                                  12.21 *     6.69 *      8.91 *
Tenure                   ?         0.03        0.03        0.02
                                 1.81 ***    1.66 ***      1.4
ROA                      -         0.19        -0.55      -0.49
                                   -0.96     -2.45 **    -2.75 *
Opinion                  +         -0.05       0.01       -0.02
                                   -0.81       0.14       -0.33
Leverage                 +         0.12        0.001       0.03
                                 1.63 ***      0.02        0.39
Stateshr                 -         0.01        0.02        0.02
                                   0.44        1.03        1.01
Legalshr                 +         0.11        0.17        0.21
                                 1.72 ***     2.41 **     3.73 *
Biglocal                 ?          0.1         0.1        0.07
                                  2.42 **     2.19 **    1.98 **
Smallocal                ?         -0.02       -0.02       0.05
                                   -0.68       -0.47       1.53

*, **, *** indicate significance at the 0.01, 0.05, and 0.10 levels,
respectively.
Gale Copyright:
Copyright 2012 Gale, Cengage Learning. All rights reserved.