Qian Wang, T.J. Wong, Lijun Xia Presented by Carl Chen State Ownership, the institutional environment, and auditor choice: Evidence from China Qian Wang, T.J. Wong, Lijun Xia Presented by Carl Chen Are your classroom colors different than what you see in this template? That’s OK! Click on Design -> Variants (the down arrow) -> Pick the color scheme that works for you! Feel free to change any “You will…” and “I will…” statements to ensure they align with your classroom procedures and rules!

Abstract Chinese state-owned enterprises controlled by province, city and county governments(local SOEs) are more likely to hire small auditors within the same region(small local auditors). In regions with less developed institutions, SOEs controlled by central government(central SOEs) also have such a tendency. This tendency attenuated as the institutions develop. Possible answers: lack of demand for large or non-local auditors, small local auditors’, superior local knowledge, and SOEs’ collusion incentives.

Facts: In 2013, 54% of Chinese listed firms hired auditors that are small(non-Top 10) and from the same locality(small local auditors), where as only 25% of Chinese listed firms hired Top-10 audit firms.

Role of political and economic institutions on firm behavior Institution features such as: Extend of state ownership The Level of market and legal institutions’ development And the degree of government power over auditors

Empirical results: Using a sample of Chinese listed firms from 1993 to 2003, compared with non- state-owned firms (non-state firms), state-owned enterprises ultimately controlled by local (province, city, and county) governments (local SOEs) are more likely to hire small local auditors. in regions with more local government intervention and a less developed credit market, both local SOEs and SOEs under the control of the central government (central SOEs) have a stronger tendency than non-state firms to hire small local auditors. However, the tendency of local and central SOEs to hire small local auditors is significantly attenuated when the state is less involved in controlling the economy and as the market and legal institutions develop.

The potential arguments Demand argument Local knowledge argument Collusion argument

Demand Argument Local and central SOEs’ preferential access to capital (IPO, Access to loans) Government bailout: In contrast, without the protection afforded by the government, non-state firms are under more pressure to hire reputable (large or non-local) auditors to mitigate the agency problem and provide an early warning of any possible financial distress.

Local Knowledge Argument Many auditors in China were closely affiliated with their local governments until they separated themselves from the government in 1998. Local auditors have specialized knowledge of local government units that supervise or hold ownership stakes in the firms. This local knowledge advantage is likely to be greater in regions with weaker market and legal institutions and stronger government involvement in business.

Collusion argument Chinese listed firms have incentives to hire acquiescent auditors to facilitate their meeting the China Securities Regulatory Commission’s (CSRC) earnings targets for IPO and seasoned equity offerings or to avoid delisting. Government owners of local SOEs face the least collusion costs because of their political power over local auditors. Government owners may also have incentives to hire acquiescent auditors to pursue private gains

Hypotheses based on arguments: Hypothesis 1. Local and central SOEs are more likely than non-state firms to hire small local auditors. Hypothesis 2. The difference in the propensity to choose small local auditors between non-state firms and local or central SOEs is larger in regions with more local government involvement in business and with less developed market and legal institutions.

To measure the quality of institutions: Credit market index Government decentralize index Legal environment index Local Knowledge and Collusion argument

Data:

Sample selection:

Result: Hypothesis 1 test Regressing a dummy variable indicating whether the auditor is a small local auditor on the following three sets of variables: (1) two ownership-type variables, Local SOEs and Central SOEs; (2) three index variables, Credit Market Index, Government Decentralization Index, and Legal Environment Index; and (3) firm characteristic and financial control variables.

Characteristic and financial control Growth and Equity Issuance to capture the effect of a firm’s capital needs on the demand for auditing. Log of Client Assets and Return on Assets are used to capture the effects of size and performance, and Total Debt/Total Assets and Current Assets/Current Liabilities are included to control for risk. To control for firm complexity, we include Receivables/Total Assets and Inventory/Total Assets.

Analysis: Panel A Local and central SOEs’ median credit market index and government centralization index are significantly lower than those of the non-state firms. This indicates that local and central SOEs are more likely than non-state firms to be operating in regions with more government intervention and less developed credit markets. Compared with that of non-state firms, local SOEs’ mean and median legal environment index are significantly lower, while central SOEs’ mean and median legal environment index are significantly higher. This suggests that local SOEs are more likely than nonstate firms to be in regions with less developed legal institutions, whereas central SOEs tend to operate in more developed legal environments.

Analysis: Panel B The coefficient on Local SOEs is positive and statistically significant at the 1% level, while that on Central SOEs is positive but statistically insignificant. This suggests that local SOEs have the strongest propensity to hire small local auditors, while central SOEs are not different from non-state firms in their likelihood of hiring small local auditors. The change in odds for Local SOEs indicates that local SOEs’ odds (the ratio of the probability of hiring small local auditors over that of hiring other auditors) are 84.5% greater than that of non-state firms. The Chi-square test shows that the coefficient on local SOEs is significantly greater than that on central SOEs, with a p-value below 0.001, which provides support to the conjecture that local governments have more direct political power than the central government over small local auditors.

Analysis: Panel B (continued) Chinese listed firms have a weaker tendency to hire small local auditors in regions with a better institutional environment, the coefficient on each index is significantly negative, and at the 1% level in models (2)–(4). The coefficient on the government decentralization index is negative but insignificant while the coefficients on the other two indexes remain significantly negative in model (5). This suggests that although there is multicollinearity among the three indexes, the credit market index and the legal environment index are able to capture unique aspects of a province’s institutions that are distinct from the other indexes.

Result: Hypothesis 2 test Add the interaction terms between government ownership (Local SOEs and Central SOEs) and the three institutional indexes to the regression model.

Result: Hypothesis 2 test The coefficient on local SOEs is significantly positive, while the coefficient on the interaction between Local SOEs and each of the three indexes in models (1)–(3) is significantly negative. When market and legal institutions are more developed or when the government intervenes less, local and central SOEs are more likely to behave the same as non-state firms in choosing auditors. We expect the coefficients on these interaction terms to be negative. This result is consistent with our prediction that when institutions develop, the positive effect of local government ownership on the choice of small local auditors decreases. When the three indexes are included in the same regression in model (4), the coefficients on the three interaction terms with Local SOEs remain negative but only the one with Credit Market Index is statistically significant, indicating that the three index variables are likely to be capturing similar underlying institutional factors.

Result: Hypothesis 2 test (central SOEs) The coefficient on Central SOEs is significantly positive and the coefficient on the interaction term is significantly negative when the Credit Market Index and Government Decentralization Index are used in models (1) and (2) of Panel C. The coefficients on the interaction terms remain marginally significantly negative for Credit Market Index (Z-stat ¼ 1.50) and significantly negative for Government Decentralization Index (Z-stat ¼ 1.81) in model (4). These results indicate that although central SOEs as a whole do not exhibit a significantly stronger tendency than non-state firms to hire small local auditors, the former are more likely to hire small local auditors in regions with less developed institutions.

Result: Hypothesis 2 test (central SOEs) The coefficient on Central SOEs Legal Environment Index is significantly positive in models (3) and (4). We find that due to collusion incentives, central SOEs located in Beijing are more likely to hire small local auditors than central SOEs located in other regions. This indicates that the collusion incentives probably outweigh the legal institutions effect of the Beijing location.

Change in ownership control and auditor choice

Change in ownership control and auditor choice The coefficient on Local SOEs to Non-state Firms in models (1)–(4) is consistently negative and statistically significant. This indicates that compared with the matched sample that has no ownership change, local SOEs that change control ownership to private owners are more likely to switch from small local auditors to non-small local auditors. In summary, the evidence in Table 5 provides further support for Hypotheses 1 and 2.

Additional analyses Audit opinion: the significantly negative coefficient on Small Local Auditors can be interpreted as small local auditors having less ability to find errors or being less willing to report errors when they are found. (demand and collusion) Audit fees: the significantly negative coefficient on Small Local Auditors can be interpreted as small local auditors having less ability to find errors or being less willing to report errors when they are found.

Conclusion We find that local SOEs are more likely than non-state firms to hire small local auditors. In addition to local SOEs, we find that central SOEs are also more likely than non-state firms to hire small local auditors in regions where institutions are less developed, but this tendency of local and central SOEs to hire small local auditors is significantly attenuated in regions with more developed institutions. Additional analyses show that local and central SOEs’ tendency to hire small local auditors is consistent with three arguments. The first argument is that local and central SOEs have less demand for Top-10 or non-local auditors due to preferential access to capital or government bailouts in times of financial distress. Local and central SOEs’ likelihood to hire small local auditors can also be explained by collusion incentives or local auditors’ superior local knowledge.