Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Does the geographical proximity between the chairman and the CEO affect internal control quality?

Does the geographical proximity between the chairman and the CEO affect internal control quality? China Journal of aCC ounting StudieS , 2017 Vol . 5, no . 3, 344–360 https://doi.org/10.1080/21697213.2017.1375638 Does the geographical proximity between the chairman and the CEO affect internal control quality?* a b c Junli Yu , Xin Jin and Shangkun Liang a b antai College of e conomics & Management, Shanghai Jiao tong university, Shanghai, China; College of economics and Management, Zhejiang university of technology, hangzhou, China; China’s Management accounting research & d evelopment Center/School of accountancy, Central university of f inance and economics, Beijing, China ABSTRACT KEYWORDS culture; geographical This article redefines the concept of proximity governance and proximity; interlocking explains the determinants of internal control quality from the business network; internal perspective of an informal institutional arrangement. Based on an control internal network (chairman–chief executive officer geographical proximity), we examine the effects of proximity governance on internal control quality using the data of Chinese A-share listed companies from 2007 to 2013. We conclude that the geographical proximity has a negative impact on internal control quality and this effect is weaker in state-owned enterprises than in private enterprises. Further empirical evidence shows that the negative effect of the geographical proximity on internal control quality can be moderated if the company has an interlocking business network. This study enriches the research literature on internal governance in emerging markets and provides reference for management selection strategies and standards of internal control. 1. Introduction Internal control is defined in the ‘Examination of Financial Statements by Independent Public Accountants’, published by the American Institute of Accountants in 1936, and has been a hot topic in the field of accounting research. The literature (e.g. Altamuro & Beatty, 2010; Bargeron, Lehn, & Zutter, 2010; Brochet, 2010; Huang & Yang, 2010; Li, Lin, & Song, 2011) finds that internal control is helpful for improving the quality of accounting information, lessening enterprise risk, reducing insider trading, relieving information asymmetry, and protecting investor interests. Since internal control is becoming increasingly important to enterprises, China has begun to gradually improve its laws and regulations on internal con- trol. Five ministries (the Ministry of Finance, the China Securities Regulatory Commission (CSRC), the National Audit Office, the China Banking Regulatory Commission, and the China Insurance Regulatory Commission) jointly promulgated the ‘Basic Norms of Internal Control’ in 2008 and the ‘Supporting Guidelines of Internal Control’ in 2010, revealing the importance CONTACT Xin Jin jxgudong@163.com Paper accepted by Kangtao Ye. © 2017 a ccounting Society of China CHINA JOURNAL OF ACCOUNTING STUDIES 345 of establishing an efficient internal control standard system in China. To ensure the effec - tiveness of such a system, policymakers must understand the factors that affect internal control quality. These factors include the firm’s size and age, number of years listed, audit committee quality, efficiency of corporate governance, financial health, financial reporting and operation complexity, growth rate, restructuring, board and audit committee charac- teristics, chief financial officer’s professional qualifications and media supervision (e.g. Abbott, Parker, & Peters, 2007; Ashbaugh-Skaife, Collins, & Kinney, 2007; Doyle, Ge, & McVay, 2007; Hoitash, Hoitash, & Bedard, 2009; Krishnan, 2005; Li, Sun, & Ettredge, 2010; Naiker & Sharma, 2009; Zhang & Xu, 2015). Most studies have focused on developed and perfect markets and the conclusions may not fully explain the situation of emerging markets. Informal institutions, such as guanxi (relationship) in China, often have great influence on the efficiency of emerging or transi- tioning market economies. Wang (2005) used the framework of relationship governance to explain the implications and success of China’s gradual reform and pointed out that relational contracts have been the dominant form of governance for a long time. Studies show that relationship governance affects earnings management, which could have an impact on internal control quality (Hwang & Kim, 2009; Fracassi & Tate, 2012). The core of relationship governance is built on a network of social relations. Based on the different definitions of such a social network in the literature (Khwaja, Mian, & Qamar, 2008; Hwang & Kim, 2009), we conclude that the network involved in relationship governance covers both an internal network consisting of financial, familial, or social ties between owners, directors, and man- agers and an external business network of interlocking executives. The social ties between executives involve a mutual alma mater, military service, regional origin, academic discipline, or industry (Hwang & Kim, 2009). This paper uses the social tie of regional origin to examine the effect of relationship governance on internal control quality. Based on the availability and reliability of data, this paper uses only the social tie of regional origin between the chairman and the chief executive officer (CEO), called a geo - graphical proximity (Jacobs, 1982), as a measurement of internal relationship governance and the interlocking business network between the chairman and the CEO as a measurement of external relationship governance. This paper explores the mechanism and effect of internal relationship governance on internal control quality using manually collected relationship governance data of listed companies and internal control quality data from the DIB database from 2007 to 2013. It also tests the difference between the effect in state-owned enterprises (SOEs) and that in private enterprises, based on significant differences in their relationship governance in the Chinese context. The paper then examines the joint effect of two kinds of relationship governance on internal control quality using external relationship governance data. Compared with prior literature, this article makes the following contributions. First, few studies consider relationship governance an influential factor in internal control quality and this article enriches the research on internal control in emerging markets. Second, the dif- ferences in the association between internal relationship governance and internal control quality in firms with different property rights provide a practical reference for the construc - tion of an internal control standard system. Third, we extend the scope of relationship gov- ernance, which contains an internal network linked with ties between executives and an external network built from interlocking executives. The connection between these two types of relationship governance can effectively improve internal control quality, thus also 346 J. YU ET AL. providing new empirical evidence on the study of the relationship governance of China’s companies. The remainder of the paper is organised as follows. Section 2 develops the hypotheses based on the theoretical analysis. Section 3 presents the research design, including the sample selection, data, variables, and empirical methodology. Section 4 reports the empirical results and analyses the effects. Section 5 concludes and presents the implications. 2. Theoretical analysis and hypothesis development 2.1. Geographical proximity and internal control quality In modern enterprises, where ownership and management are separated, the job sepa- ration of the chairman and the CEO can promote resource allocation efficiency and firm performance. However, in enterprises under the influence of Chinese traditional culture, such separation is only the separation of two positions, and the relationship between the chairman and the CEO is still quite close, such that they can be relatives, classmates or countrymen; that is, the relationship is established by blood, school, or geography, respectively. For instance, the Shanxi Exchange Shop employs workers who share a geo- graphic relationship, such that the owners, managers, partners and apprentices are all from the same county, and it never employs the owner’s relatives and the members of his clan. This form of structure implies that geographical relationships are important in business (Cai, Zhou, & Wu, 2008). Such relationships facilitate cooperation, but supervision is harder to implement. We deduce therefore, that internal control quality may be low and these firms’ market evaluations may be discounted. Schmidt (2015) conducted a cost–benefit analysis on the relationship between the chairman and the CEO and found the relationship can reduce information asymmetry and strengthen the effects of direc - tors’ suggestions; however, the supervisory effect will be reduced. Fracassi and Tate (2012) find that close relationships between the CEO and directors lead to weak supervision and bad performance. In addition, CEOs tend to promote close relatives to become new directors. Although Chinese regulators have established a well-organised internal control system for listed firms, its implementation needs improvement. The implementation of internal control depends on the attitudes of managers. A close relationship between the chairman and the CEO may generate collusion and reduces information quality (Gul, & Leung, 2004; Wang & Liang, 2008). As a driving factor of earnings management, low-quality internal control is associated with low earnings quality (Doyle, Ge, & McVay, 2005). Thus, avoiding a close relationship between the chairman and the CEO will contribute to improving internal control. Furthermore, earnings management can be inhibited (Fang & Jin, 2011). According to the analyses cited above, the following hypothesis is proposed. H1: The geographical relationship between the chairman and the CEO is significantly neg- ative correlated with internal control quality. 2.2. Property rights, geographical proximity, and internal control quality The development of the market economy in China has led to a series of internal control problems. While the Chinese government has made great efforts to improve the internal CHINA JOURNAL OF ACCOUNTING STUDIES 347 control system of enterprises and strengthen risk management and control, its effect with different property rights varies (Liu, Luo, He, & Chen, 2012). Although the governance structure and regulation environment of SOEs vary, these enterprises are still controlled and supervised by the government and the absence of owner remains. Since the chairman and CEO of SOEs are usually appointed by the government, their goal is not to promote corporate values, but to seek political promotion (Zheng, Li, Xu, Lin, & Zhao, 2012). The promotion contradiction between the chairman and the CEO in SOEs thus removes the risk of collusion from the geographical relationship and places the internal control quality of SOEs under greater supervision and constraint. Hence, the negative ee ff ct of a geograph - ical relationship on internal control quality in SOEs is weaker. Since SOEs undertake various tasks for the government, such as economic development, employment, taxation and social stability (Lin, Cai, & Li, 1998; Lin & Li, 2004; Lin, Liu, & Zhang, 2004), the government will protect state-owned listed firms from bankruptcy. Even with lower internal control quality, SOEs are subject to fewer penalties than private enterprises. Therefore, private enterprises have more incentive to improve internal control quality and reduce business risks and legal risks. We suggest that the negative effect of geographical proximity on internal control quality will be greater for private enterprises. Accordingly, the following hypothesis is proposed. H2: Compared with state-owned listed firms, the negative effect of geographical proximity on internal control quality is stronger in private listed firms. 2.3. Joint effect of internal–external relationship governance Close geographical proximity between directors and managers usually leads to inefficiency. If the chairman or CEO is also responsible for other enterprises, he or she can construct an external relationship network. External relationship governance can help the chairman or CEO acquire more experience in improving internal control quality. Especially when man- agers have little experience evaluating project risks and benefits, they usually look for related information from social networks (Ellison & Fudenberg, 1995). Liao and Chen (2009) studied the economic effect of social networks on research and development strategies and found that firms whose directors have close connections have similar strategies on R&D. Kang and Tan (2008) studied the association between directors’ external relationships and accounting strategies and found the stock option plans of firms with closely connected directors to be similar. To diminish resource constraints and information asymmetry, managers can build social networks to gain more information (Dyer, & Nobeoka, 2000; Ma, Dong, & Ge, 2010). We suggest that a rich external network can improve the efficiency and quality of information flow, so that internal control quality will be increased. In firms where the chairman and CEO do not have a close geographical relationship, there is little room for improvement. Thus, the positive effect of building external networks on internal control systems may not be evident. On the contrary, however, this positive effect may be significant in firms with close chairman–CEO geographical proximity. Accordingly, the following hypothesis is proposed. H3: External relationship governance reduces the negative effect on internal control quality of a close geographical relationship between the chairman and the CEO. 348 J. YU ET AL. 3. Research design 3.1. Sample and data The sample consists of Chinese A-share listed companies. The internal control index is from 2007 and the sample period ranges from 2007 to 2013. For the initial sample, we exclude: (1) observations missing crucial personal information; (2) observations where the chairman and CEO are the same person or from the same family; (3) financial and insurance firms; (4) Special Treatment (ST) or Particular Transfer (PT) firms ; and (5) observations with missing variables. Our final sample includes 5,457 firm–year observations. To avoid the effects of extreme values, all continuous variables are winsorised at the first percentile and 99th per - centile to eliminate the influence of extreme values. Using executive information (name list, tenure, etc.) from the China Stock Market & Accounting Research (CSMAR) listed company database, we manually collect personal infor- mation (such as birth location, living location) through prospectuses, annual reports, com- pany websites, the SINA Finance website, and other public channels. To characterise the strength of the geographical relationship, we quantify the distance (proximity) between the chairman and the CEO using the latitude and longitude of their birth (or currently domiciled) locations. Due to the homogeneity of internal control disclosure of listed companies in China, there are small differences between listed companies’ internal control effectiveness and audit opinions and other indicators, such that it cannot reflect the actual internal control situation of different companies. We therefore use internal control data that refer to the DIB listed company’s internal control index (included in the CSMAR database), which was also adopted by Mao and Meng (2013), Zhang and Wu (2014), Zhou, Hu, Lin, & Liu (2013). The external governance environment is measured by the marketisation index (Fan, Wang, & Zhu, 2011; Wang, Yu, & Fan, 2016). Other financial data are from the CSMAR database. 3.2. Model settings and variable definitions The model, based on the work of Doyle et al. (2007) and Gong, Ke, & Yu (2012), is designed as follows. 3.2.1. Model construction Hypothesis 1 predicts the negative association between a geographical relationship and internal control quality. In our study, we employ the following regression to test H1: IC (ICW )=  +  Province (Geodist )+  Controls + it it 0 1 it it i t Hypothesis 2 predicts the effects of different geographical proximity between the chairman and the CEO from companies with different property rights on internal control quality. We employ the following regression to test H2: IC (ICW )=  +  Province (Geodist )+  Soe it it 0 1 it it 2 it +  Province (Geodist )∗ Soe +  Controls + 3 it it it i t Special treatment (St ) firm reports net losses two consecutive years, Particular t ransfer (Pt ) firm suffers losses with three consecutive years of regulation. St or Pt firm means financial difficulty. t he internal control index is aimed at measuring the efficiency and effectiveness of the implementation of internal control norms. it covers all companies listed in China since 2007. CHINA JOURNAL OF ACCOUNTING STUDIES 349 Hypothesis 3 predicts the adjustment effect of external relations governance on geographical relationship and internal control quality. We employ the following regression to test H3: IC (ICW )=  +  Province (Geodist )+  Fnet it it 0 1 it it 2 it +  Province (Geodist )∗ Fnet +  Controls + 3 it it it i t 3.2.2. Variable definitions (1) Dependent variables. Internal control indicators, measuring internal governance and risk management, include the internal control index (IC, starting in 2000) and the number of internal control deficiencies (ICW, starting in 2007 ), with the internal con- trol data derived from the DIB’s Internal Control Index for Chinese Listed Companies (Hu, 2012). The index uses an internal control system framework as the basis for the system. Firms’ internal control strategy, management, reporting, compliance, and asset security are used for the basic design of the index of internal control, with major defects of internal control as revised targets to supplement and amend the internal control index. The index or internal control defects proxy for internal control quality. (2) Independent variables. Geographical relationship indicators are used to measure the geographical relationship in two dimensions to capture the internal manage- ment culture and the heterogeneity of the concept, including provincial relations (Province) and geographical distance (Geodist). Google Earth is used to obtain infor- mation regarding the latitude and longitude of the birthplace or current domicile of the CEOs and chairmen (Yu, Jin, & Lei, 2015). (3) Moderating variable. We use an executive external network index aimed at measuring the ability of executives in a commercial network to access technical innovation resources. We refer to Mintz and Schwartz (1985), Stokman, Ziegler, & Scott (1985), Khwaja et al. (2008), and Jin, Lei, & Yu (2016) and use the number of chairmen and CEOs who directly serve in other listed companies (Fnet). (4) Control variables. In this article, the following key factors are controlled for in our model: SOE represents the nature of enterprise property rights, Dual indicates the situation in which the chairman and CEO are the same person, Turnover indicates whether there was a change in the chairman or CEO that year, Law represents the legal environment in the company’s area quantified by the legal environment in the marketisation index, Size refers to the company’s size, First refers to equity con- centration, Share refers to the executive shareholding ratio, Big4 refers to whether the company that year hires a Big Four accounting firm as an auditor, Opinion refers to whether the company’s annual report obtained a standard unqualified audit opinion, ROA refers to the company’s return on assets, Leverage refers to the com- pany’s asset–liability ratio, BM refers to the book-to-market value, and C_edu, D_edu, C_age, D_age, C_gender and D_gender represent the educational background, age, and gender of the CEO and directors, respectively. In the above regression model, we also control for year and industry fixed effects. g oogle earth is virtual globe software developed by g oogle; it allows one to accurately determine latitudes and longitudes. f or example, if the Ceo ’s birthplace is a, and the chairman’s is B, geographical distance is calculated as the following equation: ∆longitudes = longitudes a − longitudes B∗ 1000∗111.413∗cos* latitudes B∗π/180 − 0.094∗cos3∗latitudes B∗π/180 ∆latitudes = latitudes a − latitudes B*1000*(111.133 − 0.059*cos(2* latitudes B*π/180))g eographical 2 2 1/2 distance = ((∆longitudes) + (∆latitudes) ) . 350 J. YU ET AL. Table 1. Variable definitions. Name Definition IC internal control index ICW t he number of internal control deficiencies Province indicator variable that is equal to 1 if the birthplaces of the board chairman and the Ceo are in the same province, and 0 otherwise Geodist negative distance between the birthplaces of the board chairman and the Ceo (mileage) Fnet t he number of interlocking companies SOE indicator variable that is equal to 1 if the nature of company’s property rights is state-owned, and 0 otherwise Law indicator variable that is equal to 1 if marketisation index of the province is higher than median value, and 0 otherwise Size natural log of total assets First largest shareholders’ holdings divided by the total number of shares Big4 indicator variable that is equal to 1 if the company hires a Big f our accounting firm as an auditor that year, and 0 otherwise Opinion indicator variable that is equal to 1 if the company’s annual report obtained a standard unqualified audit opinion, and 0 otherwise Share executive shareholding ratio ROA net income divided by total assets Leverage t otal liabilities divided by total assets BM Book-to-Market value Dual indicator variable that is equal to 1 if the year is board chairman and Ceo is the turnover year, and 0 otherwise Turnover indicator variable that is equal to 1 if the year is board chairman or Ceo turnover year, and 0 otherwise C_edu Categorical variable that is equal to 3 if the Ceo get Master or Phd , 2 if Bachelor, and 1 otherwise D_edu Categorical variable that is equal to 3 if the board chairman get Master or Phd , 2 if Bachelor, and 1 otherwise C_age Ceo ’s age D_age Board chairman’s age C_gender indicator variable that is equal to 1 if the Ceo ’s gender is male, and 0 otherwise D_gender indicator variable that is equal to 1 if the Board chairman’s gender is male, and 0 otherwise Year Year control variable Industry industry control variable 4. Empirical results 4.1. Univariate analysis Table 2 presents the main descriptive statistics of the variables. The dependent variables include the internal control index (IC) and the number of internal control weaknesses (ICW). The mean of IC is 687.50 and its standard deviation is 99.39. The mean of ICW is 0.77 and its standard deviation is 1.890, indicating that great variations exist in internal control quality. Internal geographical relationship indicators include Province and Geodist, with mean values of 0.60 and –5.03, respectively, showing that fewer than half of firms have the same or similar geographical relationship, which is similar to the result of Yu et al. (2015). The mean of the moderator external index (Fnet) is 0.13, with a maximum value of two, indicating the chair- man and/or CEO served in at most two listed companies. The variable SOE has a mean value of 0.55, indicating that the numbers of SOEs and private enterprises in the sample are roughly the same. Table 2 also presents the descriptive statistics of the other control variables, which are consistent with those of previous studies. The internal controls have a negative correlation with internal geographic relationship, preliminarily supporting H1. Both indicators of internal geographic relationship (Province and Geodist) are significantly and highly relevant, with a coefficient of more than 0.50, verifying that the two types of indicators are complementary and alternatives of each other. The other correlation coefficients between the control CHINA JOURNAL OF ACCOUNTING STUDIES 351 Table 2. Variable descriptive statistics. Variable N Mean S.D. Min p25 p50 p75 Max IC 4,167 687.50 99.39 303.50 662.20 688.10 724.50 938.20 ICW 4,167 0.77 1.89 0.00 0.00 0.00 0.00 13.00 Province 4,167 0.61 0.49 0.00 0.00 1.00 1.00 1.00 Geodist 4,167 −5.03 9.83 −125.20 −8.33 −0.83 0.00 0.00 Fnet 4,167 0.13 0.39 0.00 0.00 0.00 0.00 2.00 SOE 4,167 0.55 0.50 0.00 0.00 1.00 1.00 1.00 Law 4,167 0.66 0.47 0.00 0.00 1.00 1.00 1.00 Size 4,167 21.89 1.54 19.04 20.87 21.60 22.61 27.52 First 4,167 0.37 0.15 0.09 0.25 0.36 0.49 0.77 Big4 4,167 0.48 0.50 0.00 0.00 0.00 1.00 1.00 Opinion 4,167 0.96 0.20 0.00 1.00 1.00 1.00 1.00 Share 4,167 4.87 7.40 0.00 0.02 1.23 6.90 34.32 ROA 4,167 0.05 0.06 −0.20 0.02 0.05 0.08 0.26 Leverage 4,167 0.48 0.23 0.05 0.31 0.47 0.63 1.23 BM 4,167 0.28 0.13 0.00 0.21 0.35 0.57 0.85 Dual 4,167 0.23 0.42 0.00 0.00 0.00 1.00 1.00 Turnover 4,167 0.45 0.50 0.00 0.00 0.00 1.00 1.00 C_edu 4,167 2.26 0.81 1.00 2.00 2.00 3.00 3.00 D_edu 4,167 2.26 0.82 1.00 2.00 3.00 3.00 3.00 C_age 4,167 47.89 6.39 33.00 44.00 47.00 52.00 64.00 D_age 4,167 51.51 6.46 37.00 47.00 51.24 56.00 67.00 C_gender 4,167 0.94 0.24 0.00 1.00 1.00 1.00 1.00 D_gender 4,167 0.97 0.17 0.00 1.00 1.00 1.00 1.00 variables are less than 0.50, indicating no serious collinearity problem and significant positive correlation with internal control. 4.2. Multivariate tests Table 3 reports multivariate results based on panel data fixed effects regression analysis. The controls are lagged by one period to address contemporaneous correlation problems and heteroscedasticity. As shown in columns (1) and (2), the coefficients of Province and Geodist are significantly negative at the 1% and 5% levels (–8.839, t-value = –3.89; –0.355, t-value = –2.41), respectively, which means internal control quality will be lower if the geographical relationship is closer. This result also shows that the geographical proximity between the executives (such as fellow villagers) will reduce internal control quality and is not conducive to corporate governance. If the chairman’s and the CEO’s homes are located in the adjacent cities of two provinces, the value of Geodist may be less than the distance between two towns from the same city. Since there may be significant difference between two provinces, we exclude the sample for which Province equals one to better describe the effects of birth- place distance, as presented in columns (3) and (6). The results show H1 is still supported. The results for the control variables show that the top management turnover variables are significantly negative, indicating that the internal control quality of listed companies will be lower in a year in which the chairman or CEO changes than in a year when they remain the same. The coefficients of firm size, audit opin- ions, executive shareholding, and the market-to-book ratio are significantly positive. The coefficients for Big4 accounting firms and the leverage ratio are significantly negative, show - ing that internal control quality will be low when a Big4 accounting firm is hired to audit the firm as well as when liability is high. Similarly, if we replace the IC indicator with the ICW 352 J. YU ET AL. Table 3. g eographical relationship and internal control quality. IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −8.839*** 0.094*** (−3.89) (3.69) Geodist −0.355** −0.687** 0.002** 0.011** (−2.41) (−2.31) (2.25) (2.17) Dual 5.131 2.409 2.163 −0.083 −0.015 −0.053 (0.87) (0.44) (0.26) (−0.48) (−0.09) (−0.23) Turnover −7.881*** −7.416*** −16.009*** 0.070 0.059 0.188* (−2.86) (−2.72) (−4.22) (0.88) (0.75) (1.80) SOE 8.446* 8.251* 6.857 −0.060 −0.058 −0.002 (1.90) (1.85) (1.20) (−0.46) (−0.45) (−0.01) Law −4.639 −4.645 −16.377** 0.247 0.246 0.290 (−0.89) (−0.89) (−2.31) (1.63) (1.63) (1.48) Size 31.945*** 31.643*** 21.484*** −0.208 −0.206 −0.199 (6.93) (6.87) (3.24) (−1.55) (−1.54) (−1.09) First 38.358 38.338 −38.962 0.205 0.186 0.658 (1.32) (1.31) (−0.91) (0.24) (0.22) (0.55) Big4 −7.419** −7.759** −13.052*** −0.071* −0.072* −0.321** (−2.02) (−2.11) (−2.77) (−1.67) (−1.67) (−2.47) Opinion 27.906*** 28.044*** 47.684*** −0.274 −0.284 −0.373 (2.81) (2.82) (3.49) (−0.95) (−0.98) (−0.99) Share 1.350*** 1.336*** 1.253*** −0.003 −0.003 0.003 (5.30) (5.25) (3.81) (−0.47) (−0.47) (0.35) ROA 467.351*** 467.174*** 408.093*** −0.043 −0.042 0.656 (16.38) (16.37) (11.23) (−0.05) (−0.05) (0.65) Leverage −55.758*** −55.012*** −35.442* 0.283 0.262 0.544 (−3.78) (−3.73) (−1.73) (0.66) (0.61) (0.96) BM 211.360*** 212.497*** 214.783*** 1.605 1.601 1.506 (5.11) (5.14) (4.72) (1.34) (1.33) (1.20) C_edu −6.648** −6.635** 0.325 −0.125 −0.128 −0.003 (−2.09) (−2.09) (0.07) (−1.35) (−1.39) (−0.02) D_edu 0.804 0.859 −8.997 −0.023 −0.027 0.076 (0.29) (0.31) (−1.57) (−0.28) (−0.33) (0.48) C_age −0.002 0.009 0.191 0.010 0.009 −0.013 (−0.01) (0.02) (0.35) (0.87) (0.79) (−0.84) D_age 0.588 0.517 0.703 −0.014 −0.013 0.002 (1.54) (1.37) (1.11) (−1.26) (−1.16) (0.14) C_gender −5.835 −5.887 4.034 −0.006 0.006 0.260 (−0.51) (−0.52) (0.20) (−0.02) (0.02) (0.48) D_gender −5.604 −4.894 −2.525 0.542 0.533 0.388 (−0.46) (−0.41) (−0.13) (1.55) (1.52) (0.70) Constant −43.831 −42.768 172.972 5.981** 6.025** 5.068 (−0.45) (−0.44) (1.19) (2.11) (2.12) (1.26) Firm&Year&Industry Yes Yes Yes Yes Yes Yes observations 4,167 4,167 2,530 4,167 4,167 2,530 a dj. R 0.188 0.188 0.162 0.042 0.042 0.051 notes: a djusted t -values (double) are in parentheses. *, **, and *** represent significance at the 10%, 5%, and 1% levels, respectively. indicator, the results remain signic fi antly consistent. The regression coec ffi ient is in the oppo - site direction. The results are shown in columns (4) and (5) of Table 3, which also support H1. Table 4 reports the result for H2 (how a geographical relationship affects internal control quality for firms with different property rights). As shown in columns (1) and (2) of Table 3 , the coefficients of Province or Geodist are significantly negative at the 1% and 5% levels (–8.705, t-value = –3.16; –0.390, t-value = –2.54), respectively, indicating a significant neg- ative correlation with internal control quality, while the interaction terms (Province*SOE CHINA JOURNAL OF ACCOUNTING STUDIES 353 Table 4. g eographical relationship, property rights, and internal control quality. IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −8.705*** 0.091*** (−3.16) (2.76) Geodist −0.390** −2.585*** 0.003** 0.020** (−2.54) (−2.90) (2.44) (2.25) Province*SOE 0.398** −0.002** (2.19) (−2.16) Geodist*SOE 0.002* 1.476** −0.001* −0.008** (1.83) (2.21) (−1.74) (−2.23) SOE 8.082* 8.367* 6.047 −0.064 −0.029 0.000 (1.80) (1.85) (1.05) (−0.49) (−0.22) (0.00) Constant −57.087 −54.815 152.319 6.264** 6.278** 4.958 (−0.58) (−0.56) (1.04) (2.19) (2.20) (1.23) Controls Yes Yes Yes Yes Yes Yes Firm&Year&Industry Yes Yes Yes Yes Yes Yes observations 4,167 4,167 2,530 4,167 4,167 2,530 a dj. R 0.192 0.191 0.164 0.042 0.043 0.052 notes: a djusted t -values (double) are in parentheses. *, **, and *** represent significance at the 10%, 5%, and 1% levels, respectively. and Geodist*SOE) are significant positive, indicating that the negative correlation between geographical relationship and internal control in SOEs is weaker than in private enterprises. The finding means that the impact of internal relationship governance on internal control quality differs under different property rights. Hypothesis H2 is empirically supported. This may be due to the state-owned properties of SOEs and the personal political demands of executives. The internal relations between the chairman and the CEO are bound by organ- isational departments. The actual effects of internal control development and practice are not governed by internal relations. The impact mainly depends on national policy guidance and institutional constraints. To reduce risks and promote sustainable development, the executives of private firms have more incentive to improve internal control quality due to fewer political arrangements. Therefore, internal control quality depends mainly on the influ - ence of internal management on corporate governance, especially that of the chairman and the CEO. The same conclusion is also obtained using the ICW indicator (see columns 4–6 of Table 4). Table 5 validates the results of H3.As shown in columns (1) and (2) of Table 4, the coef- ficients of Province or Geodist are significantly negative at the 1% and 5% levels (–6.280, t-value = –3.30; –0.297, t-value = –2.16), respectively, and the coefficients of Province*Fnet or Geodist*Fnet are significantly positive at the 5% and 10% levels (19.066, t-value = 2.16; 0.847, t-value = 1.83), respectively, which indicates a significant negative correlation with internal control quality, but a positive correlation with the external network of top man- agement. The findings indicate that a close relationship between the chairman and the CEO can inhibit internal control quality due to internal and external supervisory pressure; the chairman and CEO thus construct external networks to enhance internal control quality and strengthen the weak corporate governance due to geographical proximity. Therefore, strengthening external relations governance can effectively enhance internal control quality. The same conclusion is also obtained using the ICW indicator (see columns 4–6 of Table 5). 354 J. YU ET AL. Table 5. g eographical relationship, external network, and internal control quality. IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −6.280*** 0.107*** (−3.30) (2.86) Geodist −0.297** −0.740*** 0.000** 0.011** (−2.16) (−3.31) (2.30) (2.37) Province*Fnet 19.066** −0.092* (2.16) (−1.76) Geodist*Fnet 0.847* 1.232** −0.020** −0.009** (1.83) (2.33) (−2.17) (−2.19) Fnet 6.249* 6.463* 19.262* −0.023 −0.174* −0.162 (1.91) (1.86) (1.65) (−1.14) (−1.83) (−1.50) Constant −53.275 −46.125 170.984 5.944** 5.970** 5.049 (−0.54) (−0.47) (1.17) (2.09) (2.10) (1.25) Controls Yes Yes Yes Yes Yes Yes Firm&Year&Industry Yes Yes Yes Yes Yes Yes observations 4,167 4,167 2,530 4,167 4,167 2,530 a dj. R 0.190 0.188 0.163 0.042 0.043 0.051 notes: a djusted t -values (double) are in parentheses. *, **, and *** represent significance at the 10%, 5%, and 1% levels, respectively. 4.3. Robustness test 4.3.1. Sub-sample test of the previous and following years for chairman or CEO turnover Since there is no change in geographical relationship when there is no chairman or CEO turnover, the changes in internal control may be caused by other factors. When there is a change, especially a geographical change, it is better to observe the difference in inter - nal control due to geographical relationship. Taking into account the huge difference in firm characteristics in the turnover year and the effect on internal control, we choose to observe the changes in the previous and following years for chairman or CEO turnover. Table 6 shows that the results are more pronounced and are consistent with previous findings. 4.3.2. Sub-sample with job separation of the chairman and the CEO The strongest geographical relationship that can create a corporate governance problem and affect internal control quality is CEO duality. Thus, the robustness test keeps only the sub-sample with separation of the chairman and CEO, Table 7 shows that the results are consistent with previous findings. 4.3.3. TSLS test The effect of corporate governance on internal control often leads to a very important endog - enous problem that must be addressed. We refer to Fracassi and Tate (2012) and Lu and Hu (2014) to alleviate problems of endogeneity. We construct an instrumental variable (IV ) with an external departure factor (Leave) and a social trust factor (Trust), since the departure of a chairman or CEO will affect changes in geographical relationship but will not have a direct untabulated tests showed that future t obin’s Q and cumulative excess returns decrease (increase) in companies whose geographical proximity becomes closer (more distant) after a change in executives. CHINA JOURNAL OF ACCOUNTING STUDIES 355 Table 6. t he sub-sample of the previous and following years for chairman or Ceo turnover. Panel A: Geographical relationship and internal control quality IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −5.576** 0.118** (−2.08) (2.12) Geodist −0.663** −7.037** 0.002* 0.049** (−2.10) (−2.55) (1.71) (2.58) Constant −285.589*** −296.702*** −280.925*** 1.222 1.254 1.915 (−6.19) (−6.53) (−4.56) (0.94) (0.98) (1.20) Controls Yes Yes Yes Yes Yes Yes Firm&Industry Yes Yes Yes Yes Yes Yes observations 1,292 1,292 838 1,292 1,292 838 a dj. R 0.521 0.523 0.498 0.059 0.059 0.077 Panel B: Geographical relationship, property rights, and internal control quality IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −6.090* 0.109* (−1.78) (1.75) Geodist −0.545* −6.531** 0.004* 0.054** (−1.70) (−2.40) (1.79) (2.25) Province*SOE 0.425** −0.010** (2.38) (−2.12) Geodist*SOE 0.002** 0.868** −0.002*** −0.001** (2.19) (2.43) (−2.80) (−2.49) SOE 0.514 1.428 −5.856 −0.126 −0.101 −0.174 (1.10) (1.28) (−0.91) (−0.86) (−0.69) (−1.02) Constant −287.204*** −299.445*** −290.909*** 1.128 0.964 1.774 (−6.17) (−6.52) (−4.67) (0.85) (0.74) (1.08) Controls Yes Yes Yes Yes Yes Yes Firm&Industry Yes Yes Yes Yes Yes Yes observations 1,292 1,292 838 1,292 1,292 838 a dj. R 0.530 0.530 0.513 0.060 0.065 0.077 Panel C: Geographical relationship, external network, and internal control quality IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −3.151** 0.077* (−2.57) (1.81) Geodist −0.781** −5.215* 0.002 0.040* (−2.39) (−1.75) (1.44) (1.72) Province*Fnet 11.490** −0.095** (2.13) (−2.35) Geodist*Fnet 0.896* 5.959** −0.013** −0.026** (1.88) (2.28) (2.38) (−1.26) Fnet 7.103 −2.952 3.670 −0.171 −0.229 −0.000 (1.12) (−0.42) (0.41) (−0.96) (−1.16) (−0.00) Constant −292.229*** −299.789*** −281.206*** 1.303 1.217 1.905 (−6.29) (−6.59) (−4.56) (0.99) (0.95) (1.19) Controls Yes Yes Yes Yes Yes Yes Firm&Industry Yes Yes Yes Yes Yes Yes observations 1,292 1,292 838 1,292 1,292 838 a dj. R 0.522 0.524 0.499 0.060 0.060 0.077 notes: a djusted t -values (double) are in parentheses. *, **, and *** represent significance at the 10%, 5%, and 1% levels, respectively. 356 J. YU ET AL. Table 7. t ests in the sub-sample with chairman–Ceo separation. Panel A: Geographical relationship and internal control quality IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −9.656* 0.042* (−1.86) (1.78) Geodist −0.446* −0.776** 0.002* 0.776** (−1.72) (−2.30) (1.68) (2.30) Constant −161.738 −161.158 58.169 7.920** 7.911** 58.169 (−1.43) (−1.42) (0.32) (2.42) (2.41) (0.32) Controls Yes Yes Yes Yes Yes Yes Firm&Year&Industry Yes Yes Yes Yes Yes Yes observations 3,175 3,175 1,581 3,175 3,175 1,581 a dj. R 0.188 0.188 0.162 0.042 0.042 0.051 Panel B: Geographical relationship, property rights, and internal control quality IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −9.501* 0.035** (−1.83) (2.23) Geodist −0.494* −3.131* 0.001 0.061* (−1.76) (−1.81) (1.49) (1.69) Province*SOE 1.263** −0.006** (2.36) (−2.36) Geodist*SOE 0.006** 1.648* −0.001** −0.014** (2.32) (1.72) (−2.08) (−2.39) SOE 10.818** 11.649** 7.500 −0.148 −0.094 0.034 (2.09) (2.22) (1.02) (−0.98) (−0.62) (0.16) Constant −184.790 −178.137 25.634 8.319** 8.193** 6.577 (−1.63) (−1.57) (0.14) (2.52) (2.49) (1.28) Controls Yes Yes Yes Yes Yes Yes Firm&Year&Industry Yes Yes Yes Yes Yes Yes observations 3,175 3,175 1,581 3,175 3,175 1,581 a dj. R 0.216 0.214 0.183 0.044 0.046 0.047 Panel C: Geographical relationship, external network, and internal control quality IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −8.229* 0.057* (−1.63) (1.67) Geodist −0.435 −0.779* 0.004 0.043* (−1.54) (−1.71) (1.54) (1.69) Province*Fnet 10.077** −0.107* (2.02) (−1.82) Geodist*Fnet 0.167** 1.167** −0.031* −0.002** (2.26) (2.28) (−1.67) (−2.01) Fnet 3.770 −0.120 −13.967 0.099 −0.207 −0.124 (0.60) (−1.01) (−0.94) (0.55) (−0.88) (−0.90) Constant −168.534 −163.395 55.445 7.748** 7.801** 6.713 (−1.48) (−1.44) (0.30) (2.35) (2.37) (1.31) Controls Yes Yes Yes Yes Yes Yes Firm&Year&Industry Yes Yes Yes Yes Yes Yes observations 3,175 3,175 1,581 3,175 3,175 1,581 a dj. R 0.210 0.209 0.180 0.045 0.046 0.047 notes: a djusted t -values (double) are in parentheses. *, **, and *** represent significance at the 10%, 5%, and 1% levels, respectively. CHINA JOURNAL OF ACCOUNTING STUDIES 357 Table 8. t wo-stage least squares test between geographical relationship and internal control quality. IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −21.802*** 3.602*** (−3.43) (2.80) Geodist −16.768** −19.295*** 0.271* 1.167* (−2.07) (−2.80) (1.82) (1.90) Constant −306.425*** 92.222 −260.022*** 6.547*** −0.098 4.583*** (−3.63) (1.13) (−2.90) (3.74) (−0.07) (2.86) Controls Yes Yes Yes Yes Yes Yes Firm&Year&Industry Yes Yes Yes Yes Yes Yes observations 4,167 4,167 2,530 4,167 4,167 2,530 a dj. R 0.24 0.14 0.13 0.23 0.14 0.13 f value 65.98 98.08 32.21 67.64 29.94 28.33 Weak identification test (f-statistic) 48.82 18.89 7.59 48.82 18.89 7.59 J-Statistic 2.24 1.73 0.19 2.28 1.69 0.01 (p-value) (0.12) (0.18) (0.66) (0.13) (0.19) (0.89) notes: a djusted t -values (double) are in parentheses. *, **, and *** represent significance at the 10%, 5%, and 1% levels, respectively. impact on internal control. Similarly, social trust is also exogenous, which will ae ff ct whether owners have a conspiracy motive to hire a geographically close CEO, but does not directly impact internal control. Table 8 reports the correlation tests, endogeneity tests, and second-stage regression results for the IVs. In the correlation test, the F-statistic is greater than 10, which means that the selected IV satisfies the dependency condition. The over-identification constraint p -value is greater than 0.10 in the IV endogeneity test, which means that the two IVs cannot be rejected due to the endogenous nature of the original hypothesis. The dependence and exogenous conditions provide strong evidence of the validity of the use of IVs in our study. After introduction of the IVs, the results of the main variables are still consistent. 5. Conclusion The rapid development of the market economy and the acceleration of the process of eco- nomic integration have made information technology increasingly prominent in the enter- prise, with a gradual increase in enterprise risk. As an important means of management, internal control plays a strong role in controlling and avoiding business risk and financial risk and its implementation is a key step in ensuring operating efficiency and sustainable development for enterprises. Therefore, internal control quality improvement has become key to corporate management success and the focus of research. From the perspective of informal institutional arrangements, our study empirically tests the effect of chairman–CEO relationship governance on internal control quality using data of Chinese companies listed from 2007 to 2013. This differs from prior literature, which con- centrates on financial activities and management. We find that strengthening the internal relationship governance or establishment of a geographical relationship between the chair- man and the CEO can harm internal control quality, and this negative effect should be weaker in SOEs than in private firms. However, the establishment and scale of an external network could moderate the negative effect of geographical proximity on internal control quality. The close geographical proximity of executives often cause serious damage to internal 358 J. YU ET AL. control quality, especially for private enterprises that lack resources and receive less support from local government, and executives can improve internal control quality and efficiency through external relationship governance. We conclude that collusion caused by a close geographical relationship between the chairman and the CEO will lead to a reduction of internal control quality and have a negative influence on firm value and market performance. However, an interlocking business network could regulate the negative impact of such collusion. Therefore, business owners should consider the advantages and disadvantages of relationship governance when deciding on the selection and appointment of senior managers and personnel arrangements should be conducive to enhancing the efficiency of corporate governance and internal control quality. Acknowledgements This work was supported by the Social Science Foundation of Zhejiang [No. 14NDJC166YB], the National Natural Science Foundation of China [No. 71402169], the China Postdoctoral Science Foundation [No. 2016M601606, 2017T100301, 71702102], the Humanity and Social Science Youth Foundation of Ministry of Education of China [No. 15YJC790137], the Social Science Foundation of Beijing [No. 15JGC176], and the Young Elite Teacher Project of the Central University of Finance and Economics [No. QYP1608]. Disclosure statement No potential conflict of interest was reported by the authors. References Abbott, L.J., Parker, S., & Peters, G.F. (2007). Corporate governance, audit quality, and the Sarbanes-Oxley Act: Evidence from internal audit outsourcing. The Accounting Review, 82, 803–835. Altamuro, J., & Beatty, A. (2010). How does internal control regulation affect financial reporting? Journal of Accounting and Economics, 49, 58–74. Ashbaugh-Skaife, H., Collins, D.W., & Kinney, W. (2007). The discovery and reporting of internal control deficiencies prior to SOX-mandated audits. Journal of Accounting and Economics, 44, 166–192. Bargeron, L.L., Lehn, K.M., & Zutter, C.J. (2010). Sarbanes–Oxley and corporate risk-taking. Journal of Accounting and Economics, 49, 34–52. Brochet, F. (2010). Information content of insider trades before and after the Sarbanes-Oxley Act. The Accounting Review, 85, 419–447. Cai, H.H., Zhou, L.A., & Wu, Y.Y. (2008). The clan rule, the faith of merchants and the management of merchant organizations: A comparative study on merchants of Anhui province and Shanxi province in the period of the Ming and Qing dynasty. Management World, 8, 87–99 (In Chinese). Doyle, J., Ge, W., & McVay, S. (2005). Determinants of weaknesses in internal control over financial reporting and the implications for earnings quality. Unpublished manuscript. Retrieved from https://papers. ssrn.com/sol3/papers.cfm?abstract_id=677622 Doyle, J., Ge, W., & McVay, S. (2007). Determinants of weaknesses in internal control over financial reporting. Journal of Accounting and Economics, 44, 193–223. Dyer, J.H., & Nobeoka, K. (2000). Creating and managing a high performance knowledge-sharing network: The Toyota case. Strategic Management Journal, 21, 345–367. Ellison, G., & Fudenberg, D. (1995). Word-of-mouth communication and social learning. Quarterly Journal of Economics, 110(1), 93–125. Fan, G., Wang, X.L., & Zhu, H.P. (2011). NERI index of marketization of China’s provinces 2011 report. Beijing: Economic Science Press (In Chinese). CHINA JOURNAL OF ACCOUNTING STUDIES 359 Fang, H.X. & Jin, Y.N. (2011). Can high quality internal control reduce earnings management? An empirical research based on voluntary internal control audits reporting. Accounting Research, 8, 53–60 (In Chinese). Fracassi, C. & Tate, U. (2012). External networking and internal firm governance. Journal of Finance, 67(1), 153–194. Gong, G., Ke, B., & Yu, Y. (2012). Home country investor protection, ownership structure and cross- listed firms’ compliance with SOX-mandated internal control deficiency disclosures. Contemporary Accounting Research, 30, 1490–1523. Gul, F.A., & Leung, S. (2004). Board leadership, outside directors’ expertise and voluntary corporate disclosures. Journal of Accounting and Public Policy, 23, 351–379. Hoitash, U., Hoitash, R., & Bedard, J.C. (2009). Corporate governance and internal control over financial reporting: A comparison of regulatory regimes. The Accounting Review, 84, 839–867. Huang, S.C., & Yang, X.S. (2010). Internal control report, quality of financial reports and information asymmetry: Empirical evidence from listed companies in Shanghai Stock Exchange. Journal of Finance and Economics, 36, 81–91 (In Chinese). Hwang, B.H., & Kim, S. (2009). It pays to have friends. Journal of Financial Economics, 93(1), 138–158. Jacobs, J.B. (1982). The concept of guanxi and local politics in a rural Chinese cultural setting. In S. Greenblatt, R. Wilson, & A. Wilson (Eds.), Social interaction in Chinese society. New York: Praeger. Jin, X., Lei, G.Y., & Yu, J.L. (2016). Government governance, executive networks and enterprise R&D expenditure. China Journal of Accounting Research, 9(1), 59–81. Kang, E., & Tan, B.R. (2008). Accounting choices and director interlocks: A social network approach to the voluntary expensing of stock option grants. Journal of Business Finance & Accounting, 35, 1079–1102. Khwaja, A.I., Mian, A., & Qamar, A. (2008). The value of business networks. Unpublished manuscript Harvard University. Retrieved from https://www.hks.harvard.edu/fs/akhwaja/papers/KhwajaMianQamar_ NetworksImpact_July25_final.pdf Krishnan, J. (2005). Audit committee quality and internal control: An empirical analysis. The Accounting Review, 80, 649–675. Li, C., Sun, L.L., & Ettredge, M. (2010). Financial executive qualifications, financial executive turnover and adverse SOX 404 opinions. Journal of Accounting and Economics, 50(1), 93–110. Li, W.F., Lin, B., & Song, L. (2011). The role played by the internal control in companies’ investment: Is it a promotion of efficiency or a repression thereof? Management World, 2, 81–99 (In Chinese). Liao, C., & Chen, J. (2009). Director networks and their similarity in R&D expenditure decision. International Research Journal of Finance and Economics, 23, 75–88. Lin, J.Y., Cai, F., & Li, Z. (1998). Competition, policy burdens and state-owned enterprise reform. American Economic Review, 88, 422–427. Lin, Y.F., & Li, Z.Y. (2004). Policy burden, moral hazard and soft budget constraint. Economic Research Journal, 2, 17–27 (In Chinese). Lin, Y.F., Liu, M.X., & Zhang, Q. (2004). Policy burden and enterprise’s soft budgetary binding: A case study from China. Management World, 8, 81–89 (In Chinese). Liu, Q.L., Luo, L., He, W.F., & Chen, H.W. (2012). State ownership, the institutional environment, and internal control quality: Evidence from Chinese listed firms. Accounting Research, 3, 52–61 (In Chinese). Lu, Y., & Hu, J.Y. (2014). The impact of hometown connectedness between the CEO and board of directors on corporate risk. Management World, 3, 131–138 (In Chinese). Ma, H.J., Dong, B.B., & Ge, B.S. (2010). The empirical research on the relationships among network capabilities, information acquisition and performance of high-tech firm. Studies in Science of Science, 28(1), 127–132 (In Chinese). Mao, X.S., & Meng, J. (2013). Internal control and litigation risk. Management World, 11, 155–165 (In Chinese). Mintz, B., & Schwartz, M. (1985). The power structure of American business. Chicago: University of Chicago Press. Naiker, V., & Sharma, D.S. (2009). Former audit partners on the audit committee and internal control deficiencies. The Accounting Review, 84, 559–587. 360 J. YU ET AL. Schmidt, B. (2015). Costs and benefits of friendly boards during mergers and acquisitions. Journal of Financial Economics, 117, 424–447. Stokman, F.N., Ziegler, R., & Scott, J. (1985). Networks of corporate power. Cambridge: Polity Press. Wang, B., & Liang, X.X. (2008). Corporate governance, financial condition and quality of disclosure: Evidence from Shenzhen Stock Exchange. Accounting Research, 2, 31–38 (In Chinese). Wang, X.L., Yu, J.W., & Fan, G. (2016). Eight-year process report on China’s marketization. Retrieved from http://www.rmlt.com.cn/2016/0414/423199.shtml (In Chinese). Wang, Y.Q. (2005). Reputation, commitment and organizational form: The comparative institutional analysis. Shanghai: Shanghai People’s Press (In Chinese). Yu, J.L., Jin, X., & Lei, G.Y. (2015). Geographical relationship of management and corporate investment efficiency. Contemporary Finance & Economics, 10, 116–128 (In Chinese). Zhang, H.L., & Wu, Y.H. (2014). Internal control, cash holdings and economic consequences. Accounting Research, 3, 71–78 (in Chinese). Zhang, P., & Xu, W. (2015). Does media supervision improve the effectiveness of internal control? Empirical evidence from Chinese listed enterprises. Journal of Accounting and Economics, 5, 88–105 (In Chinese). Zheng, Z.G., Li, D.X., Xu, R., Lin, R.T., & Zhao, X.J. (2012). The political promotion of the SOE executives: A case study on a certain corporation of a certain province. Management World, 10, 146–156 (In Chinese). Zhou, S.H., Hu, W.M., Lin, B., & Liu, C.L. (2013). Research on internal control of Chinese listed companies in 2012. Accounting Research, 7, 3–12 (In Chinese). http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png China Journal of Accounting Studies Taylor & Francis

Does the geographical proximity between the chairman and the CEO affect internal control quality?

Download PDF
17 pages

Loading next page...
 
/lp/taylor-francis/does-the-geographical-proximity-between-the-chairman-and-the-ceo-wCZ66cQUYN

References (50)

Publisher
Taylor & Francis
Copyright
© 2017 Accounting Society of China
ISSN
2169-7221
eISSN
2169-7213
DOI
10.1080/21697213.2017.1375638
Publisher site
See Article on Publisher Site

Abstract

China Journal of aCC ounting StudieS , 2017 Vol . 5, no . 3, 344–360 https://doi.org/10.1080/21697213.2017.1375638 Does the geographical proximity between the chairman and the CEO affect internal control quality?* a b c Junli Yu , Xin Jin and Shangkun Liang a b antai College of e conomics & Management, Shanghai Jiao tong university, Shanghai, China; College of economics and Management, Zhejiang university of technology, hangzhou, China; China’s Management accounting research & d evelopment Center/School of accountancy, Central university of f inance and economics, Beijing, China ABSTRACT KEYWORDS culture; geographical This article redefines the concept of proximity governance and proximity; interlocking explains the determinants of internal control quality from the business network; internal perspective of an informal institutional arrangement. Based on an control internal network (chairman–chief executive officer geographical proximity), we examine the effects of proximity governance on internal control quality using the data of Chinese A-share listed companies from 2007 to 2013. We conclude that the geographical proximity has a negative impact on internal control quality and this effect is weaker in state-owned enterprises than in private enterprises. Further empirical evidence shows that the negative effect of the geographical proximity on internal control quality can be moderated if the company has an interlocking business network. This study enriches the research literature on internal governance in emerging markets and provides reference for management selection strategies and standards of internal control. 1. Introduction Internal control is defined in the ‘Examination of Financial Statements by Independent Public Accountants’, published by the American Institute of Accountants in 1936, and has been a hot topic in the field of accounting research. The literature (e.g. Altamuro & Beatty, 2010; Bargeron, Lehn, & Zutter, 2010; Brochet, 2010; Huang & Yang, 2010; Li, Lin, & Song, 2011) finds that internal control is helpful for improving the quality of accounting information, lessening enterprise risk, reducing insider trading, relieving information asymmetry, and protecting investor interests. Since internal control is becoming increasingly important to enterprises, China has begun to gradually improve its laws and regulations on internal con- trol. Five ministries (the Ministry of Finance, the China Securities Regulatory Commission (CSRC), the National Audit Office, the China Banking Regulatory Commission, and the China Insurance Regulatory Commission) jointly promulgated the ‘Basic Norms of Internal Control’ in 2008 and the ‘Supporting Guidelines of Internal Control’ in 2010, revealing the importance CONTACT Xin Jin jxgudong@163.com Paper accepted by Kangtao Ye. © 2017 a ccounting Society of China CHINA JOURNAL OF ACCOUNTING STUDIES 345 of establishing an efficient internal control standard system in China. To ensure the effec - tiveness of such a system, policymakers must understand the factors that affect internal control quality. These factors include the firm’s size and age, number of years listed, audit committee quality, efficiency of corporate governance, financial health, financial reporting and operation complexity, growth rate, restructuring, board and audit committee charac- teristics, chief financial officer’s professional qualifications and media supervision (e.g. Abbott, Parker, & Peters, 2007; Ashbaugh-Skaife, Collins, & Kinney, 2007; Doyle, Ge, & McVay, 2007; Hoitash, Hoitash, & Bedard, 2009; Krishnan, 2005; Li, Sun, & Ettredge, 2010; Naiker & Sharma, 2009; Zhang & Xu, 2015). Most studies have focused on developed and perfect markets and the conclusions may not fully explain the situation of emerging markets. Informal institutions, such as guanxi (relationship) in China, often have great influence on the efficiency of emerging or transi- tioning market economies. Wang (2005) used the framework of relationship governance to explain the implications and success of China’s gradual reform and pointed out that relational contracts have been the dominant form of governance for a long time. Studies show that relationship governance affects earnings management, which could have an impact on internal control quality (Hwang & Kim, 2009; Fracassi & Tate, 2012). The core of relationship governance is built on a network of social relations. Based on the different definitions of such a social network in the literature (Khwaja, Mian, & Qamar, 2008; Hwang & Kim, 2009), we conclude that the network involved in relationship governance covers both an internal network consisting of financial, familial, or social ties between owners, directors, and man- agers and an external business network of interlocking executives. The social ties between executives involve a mutual alma mater, military service, regional origin, academic discipline, or industry (Hwang & Kim, 2009). This paper uses the social tie of regional origin to examine the effect of relationship governance on internal control quality. Based on the availability and reliability of data, this paper uses only the social tie of regional origin between the chairman and the chief executive officer (CEO), called a geo - graphical proximity (Jacobs, 1982), as a measurement of internal relationship governance and the interlocking business network between the chairman and the CEO as a measurement of external relationship governance. This paper explores the mechanism and effect of internal relationship governance on internal control quality using manually collected relationship governance data of listed companies and internal control quality data from the DIB database from 2007 to 2013. It also tests the difference between the effect in state-owned enterprises (SOEs) and that in private enterprises, based on significant differences in their relationship governance in the Chinese context. The paper then examines the joint effect of two kinds of relationship governance on internal control quality using external relationship governance data. Compared with prior literature, this article makes the following contributions. First, few studies consider relationship governance an influential factor in internal control quality and this article enriches the research on internal control in emerging markets. Second, the dif- ferences in the association between internal relationship governance and internal control quality in firms with different property rights provide a practical reference for the construc - tion of an internal control standard system. Third, we extend the scope of relationship gov- ernance, which contains an internal network linked with ties between executives and an external network built from interlocking executives. The connection between these two types of relationship governance can effectively improve internal control quality, thus also 346 J. YU ET AL. providing new empirical evidence on the study of the relationship governance of China’s companies. The remainder of the paper is organised as follows. Section 2 develops the hypotheses based on the theoretical analysis. Section 3 presents the research design, including the sample selection, data, variables, and empirical methodology. Section 4 reports the empirical results and analyses the effects. Section 5 concludes and presents the implications. 2. Theoretical analysis and hypothesis development 2.1. Geographical proximity and internal control quality In modern enterprises, where ownership and management are separated, the job sepa- ration of the chairman and the CEO can promote resource allocation efficiency and firm performance. However, in enterprises under the influence of Chinese traditional culture, such separation is only the separation of two positions, and the relationship between the chairman and the CEO is still quite close, such that they can be relatives, classmates or countrymen; that is, the relationship is established by blood, school, or geography, respectively. For instance, the Shanxi Exchange Shop employs workers who share a geo- graphic relationship, such that the owners, managers, partners and apprentices are all from the same county, and it never employs the owner’s relatives and the members of his clan. This form of structure implies that geographical relationships are important in business (Cai, Zhou, & Wu, 2008). Such relationships facilitate cooperation, but supervision is harder to implement. We deduce therefore, that internal control quality may be low and these firms’ market evaluations may be discounted. Schmidt (2015) conducted a cost–benefit analysis on the relationship between the chairman and the CEO and found the relationship can reduce information asymmetry and strengthen the effects of direc - tors’ suggestions; however, the supervisory effect will be reduced. Fracassi and Tate (2012) find that close relationships between the CEO and directors lead to weak supervision and bad performance. In addition, CEOs tend to promote close relatives to become new directors. Although Chinese regulators have established a well-organised internal control system for listed firms, its implementation needs improvement. The implementation of internal control depends on the attitudes of managers. A close relationship between the chairman and the CEO may generate collusion and reduces information quality (Gul, & Leung, 2004; Wang & Liang, 2008). As a driving factor of earnings management, low-quality internal control is associated with low earnings quality (Doyle, Ge, & McVay, 2005). Thus, avoiding a close relationship between the chairman and the CEO will contribute to improving internal control. Furthermore, earnings management can be inhibited (Fang & Jin, 2011). According to the analyses cited above, the following hypothesis is proposed. H1: The geographical relationship between the chairman and the CEO is significantly neg- ative correlated with internal control quality. 2.2. Property rights, geographical proximity, and internal control quality The development of the market economy in China has led to a series of internal control problems. While the Chinese government has made great efforts to improve the internal CHINA JOURNAL OF ACCOUNTING STUDIES 347 control system of enterprises and strengthen risk management and control, its effect with different property rights varies (Liu, Luo, He, & Chen, 2012). Although the governance structure and regulation environment of SOEs vary, these enterprises are still controlled and supervised by the government and the absence of owner remains. Since the chairman and CEO of SOEs are usually appointed by the government, their goal is not to promote corporate values, but to seek political promotion (Zheng, Li, Xu, Lin, & Zhao, 2012). The promotion contradiction between the chairman and the CEO in SOEs thus removes the risk of collusion from the geographical relationship and places the internal control quality of SOEs under greater supervision and constraint. Hence, the negative ee ff ct of a geograph - ical relationship on internal control quality in SOEs is weaker. Since SOEs undertake various tasks for the government, such as economic development, employment, taxation and social stability (Lin, Cai, & Li, 1998; Lin & Li, 2004; Lin, Liu, & Zhang, 2004), the government will protect state-owned listed firms from bankruptcy. Even with lower internal control quality, SOEs are subject to fewer penalties than private enterprises. Therefore, private enterprises have more incentive to improve internal control quality and reduce business risks and legal risks. We suggest that the negative effect of geographical proximity on internal control quality will be greater for private enterprises. Accordingly, the following hypothesis is proposed. H2: Compared with state-owned listed firms, the negative effect of geographical proximity on internal control quality is stronger in private listed firms. 2.3. Joint effect of internal–external relationship governance Close geographical proximity between directors and managers usually leads to inefficiency. If the chairman or CEO is also responsible for other enterprises, he or she can construct an external relationship network. External relationship governance can help the chairman or CEO acquire more experience in improving internal control quality. Especially when man- agers have little experience evaluating project risks and benefits, they usually look for related information from social networks (Ellison & Fudenberg, 1995). Liao and Chen (2009) studied the economic effect of social networks on research and development strategies and found that firms whose directors have close connections have similar strategies on R&D. Kang and Tan (2008) studied the association between directors’ external relationships and accounting strategies and found the stock option plans of firms with closely connected directors to be similar. To diminish resource constraints and information asymmetry, managers can build social networks to gain more information (Dyer, & Nobeoka, 2000; Ma, Dong, & Ge, 2010). We suggest that a rich external network can improve the efficiency and quality of information flow, so that internal control quality will be increased. In firms where the chairman and CEO do not have a close geographical relationship, there is little room for improvement. Thus, the positive effect of building external networks on internal control systems may not be evident. On the contrary, however, this positive effect may be significant in firms with close chairman–CEO geographical proximity. Accordingly, the following hypothesis is proposed. H3: External relationship governance reduces the negative effect on internal control quality of a close geographical relationship between the chairman and the CEO. 348 J. YU ET AL. 3. Research design 3.1. Sample and data The sample consists of Chinese A-share listed companies. The internal control index is from 2007 and the sample period ranges from 2007 to 2013. For the initial sample, we exclude: (1) observations missing crucial personal information; (2) observations where the chairman and CEO are the same person or from the same family; (3) financial and insurance firms; (4) Special Treatment (ST) or Particular Transfer (PT) firms ; and (5) observations with missing variables. Our final sample includes 5,457 firm–year observations. To avoid the effects of extreme values, all continuous variables are winsorised at the first percentile and 99th per - centile to eliminate the influence of extreme values. Using executive information (name list, tenure, etc.) from the China Stock Market & Accounting Research (CSMAR) listed company database, we manually collect personal infor- mation (such as birth location, living location) through prospectuses, annual reports, com- pany websites, the SINA Finance website, and other public channels. To characterise the strength of the geographical relationship, we quantify the distance (proximity) between the chairman and the CEO using the latitude and longitude of their birth (or currently domiciled) locations. Due to the homogeneity of internal control disclosure of listed companies in China, there are small differences between listed companies’ internal control effectiveness and audit opinions and other indicators, such that it cannot reflect the actual internal control situation of different companies. We therefore use internal control data that refer to the DIB listed company’s internal control index (included in the CSMAR database), which was also adopted by Mao and Meng (2013), Zhang and Wu (2014), Zhou, Hu, Lin, & Liu (2013). The external governance environment is measured by the marketisation index (Fan, Wang, & Zhu, 2011; Wang, Yu, & Fan, 2016). Other financial data are from the CSMAR database. 3.2. Model settings and variable definitions The model, based on the work of Doyle et al. (2007) and Gong, Ke, & Yu (2012), is designed as follows. 3.2.1. Model construction Hypothesis 1 predicts the negative association between a geographical relationship and internal control quality. In our study, we employ the following regression to test H1: IC (ICW )=  +  Province (Geodist )+  Controls + it it 0 1 it it i t Hypothesis 2 predicts the effects of different geographical proximity between the chairman and the CEO from companies with different property rights on internal control quality. We employ the following regression to test H2: IC (ICW )=  +  Province (Geodist )+  Soe it it 0 1 it it 2 it +  Province (Geodist )∗ Soe +  Controls + 3 it it it i t Special treatment (St ) firm reports net losses two consecutive years, Particular t ransfer (Pt ) firm suffers losses with three consecutive years of regulation. St or Pt firm means financial difficulty. t he internal control index is aimed at measuring the efficiency and effectiveness of the implementation of internal control norms. it covers all companies listed in China since 2007. CHINA JOURNAL OF ACCOUNTING STUDIES 349 Hypothesis 3 predicts the adjustment effect of external relations governance on geographical relationship and internal control quality. We employ the following regression to test H3: IC (ICW )=  +  Province (Geodist )+  Fnet it it 0 1 it it 2 it +  Province (Geodist )∗ Fnet +  Controls + 3 it it it i t 3.2.2. Variable definitions (1) Dependent variables. Internal control indicators, measuring internal governance and risk management, include the internal control index (IC, starting in 2000) and the number of internal control deficiencies (ICW, starting in 2007 ), with the internal con- trol data derived from the DIB’s Internal Control Index for Chinese Listed Companies (Hu, 2012). The index uses an internal control system framework as the basis for the system. Firms’ internal control strategy, management, reporting, compliance, and asset security are used for the basic design of the index of internal control, with major defects of internal control as revised targets to supplement and amend the internal control index. The index or internal control defects proxy for internal control quality. (2) Independent variables. Geographical relationship indicators are used to measure the geographical relationship in two dimensions to capture the internal manage- ment culture and the heterogeneity of the concept, including provincial relations (Province) and geographical distance (Geodist). Google Earth is used to obtain infor- mation regarding the latitude and longitude of the birthplace or current domicile of the CEOs and chairmen (Yu, Jin, & Lei, 2015). (3) Moderating variable. We use an executive external network index aimed at measuring the ability of executives in a commercial network to access technical innovation resources. We refer to Mintz and Schwartz (1985), Stokman, Ziegler, & Scott (1985), Khwaja et al. (2008), and Jin, Lei, & Yu (2016) and use the number of chairmen and CEOs who directly serve in other listed companies (Fnet). (4) Control variables. In this article, the following key factors are controlled for in our model: SOE represents the nature of enterprise property rights, Dual indicates the situation in which the chairman and CEO are the same person, Turnover indicates whether there was a change in the chairman or CEO that year, Law represents the legal environment in the company’s area quantified by the legal environment in the marketisation index, Size refers to the company’s size, First refers to equity con- centration, Share refers to the executive shareholding ratio, Big4 refers to whether the company that year hires a Big Four accounting firm as an auditor, Opinion refers to whether the company’s annual report obtained a standard unqualified audit opinion, ROA refers to the company’s return on assets, Leverage refers to the com- pany’s asset–liability ratio, BM refers to the book-to-market value, and C_edu, D_edu, C_age, D_age, C_gender and D_gender represent the educational background, age, and gender of the CEO and directors, respectively. In the above regression model, we also control for year and industry fixed effects. g oogle earth is virtual globe software developed by g oogle; it allows one to accurately determine latitudes and longitudes. f or example, if the Ceo ’s birthplace is a, and the chairman’s is B, geographical distance is calculated as the following equation: ∆longitudes = longitudes a − longitudes B∗ 1000∗111.413∗cos* latitudes B∗π/180 − 0.094∗cos3∗latitudes B∗π/180 ∆latitudes = latitudes a − latitudes B*1000*(111.133 − 0.059*cos(2* latitudes B*π/180))g eographical 2 2 1/2 distance = ((∆longitudes) + (∆latitudes) ) . 350 J. YU ET AL. Table 1. Variable definitions. Name Definition IC internal control index ICW t he number of internal control deficiencies Province indicator variable that is equal to 1 if the birthplaces of the board chairman and the Ceo are in the same province, and 0 otherwise Geodist negative distance between the birthplaces of the board chairman and the Ceo (mileage) Fnet t he number of interlocking companies SOE indicator variable that is equal to 1 if the nature of company’s property rights is state-owned, and 0 otherwise Law indicator variable that is equal to 1 if marketisation index of the province is higher than median value, and 0 otherwise Size natural log of total assets First largest shareholders’ holdings divided by the total number of shares Big4 indicator variable that is equal to 1 if the company hires a Big f our accounting firm as an auditor that year, and 0 otherwise Opinion indicator variable that is equal to 1 if the company’s annual report obtained a standard unqualified audit opinion, and 0 otherwise Share executive shareholding ratio ROA net income divided by total assets Leverage t otal liabilities divided by total assets BM Book-to-Market value Dual indicator variable that is equal to 1 if the year is board chairman and Ceo is the turnover year, and 0 otherwise Turnover indicator variable that is equal to 1 if the year is board chairman or Ceo turnover year, and 0 otherwise C_edu Categorical variable that is equal to 3 if the Ceo get Master or Phd , 2 if Bachelor, and 1 otherwise D_edu Categorical variable that is equal to 3 if the board chairman get Master or Phd , 2 if Bachelor, and 1 otherwise C_age Ceo ’s age D_age Board chairman’s age C_gender indicator variable that is equal to 1 if the Ceo ’s gender is male, and 0 otherwise D_gender indicator variable that is equal to 1 if the Board chairman’s gender is male, and 0 otherwise Year Year control variable Industry industry control variable 4. Empirical results 4.1. Univariate analysis Table 2 presents the main descriptive statistics of the variables. The dependent variables include the internal control index (IC) and the number of internal control weaknesses (ICW). The mean of IC is 687.50 and its standard deviation is 99.39. The mean of ICW is 0.77 and its standard deviation is 1.890, indicating that great variations exist in internal control quality. Internal geographical relationship indicators include Province and Geodist, with mean values of 0.60 and –5.03, respectively, showing that fewer than half of firms have the same or similar geographical relationship, which is similar to the result of Yu et al. (2015). The mean of the moderator external index (Fnet) is 0.13, with a maximum value of two, indicating the chair- man and/or CEO served in at most two listed companies. The variable SOE has a mean value of 0.55, indicating that the numbers of SOEs and private enterprises in the sample are roughly the same. Table 2 also presents the descriptive statistics of the other control variables, which are consistent with those of previous studies. The internal controls have a negative correlation with internal geographic relationship, preliminarily supporting H1. Both indicators of internal geographic relationship (Province and Geodist) are significantly and highly relevant, with a coefficient of more than 0.50, verifying that the two types of indicators are complementary and alternatives of each other. The other correlation coefficients between the control CHINA JOURNAL OF ACCOUNTING STUDIES 351 Table 2. Variable descriptive statistics. Variable N Mean S.D. Min p25 p50 p75 Max IC 4,167 687.50 99.39 303.50 662.20 688.10 724.50 938.20 ICW 4,167 0.77 1.89 0.00 0.00 0.00 0.00 13.00 Province 4,167 0.61 0.49 0.00 0.00 1.00 1.00 1.00 Geodist 4,167 −5.03 9.83 −125.20 −8.33 −0.83 0.00 0.00 Fnet 4,167 0.13 0.39 0.00 0.00 0.00 0.00 2.00 SOE 4,167 0.55 0.50 0.00 0.00 1.00 1.00 1.00 Law 4,167 0.66 0.47 0.00 0.00 1.00 1.00 1.00 Size 4,167 21.89 1.54 19.04 20.87 21.60 22.61 27.52 First 4,167 0.37 0.15 0.09 0.25 0.36 0.49 0.77 Big4 4,167 0.48 0.50 0.00 0.00 0.00 1.00 1.00 Opinion 4,167 0.96 0.20 0.00 1.00 1.00 1.00 1.00 Share 4,167 4.87 7.40 0.00 0.02 1.23 6.90 34.32 ROA 4,167 0.05 0.06 −0.20 0.02 0.05 0.08 0.26 Leverage 4,167 0.48 0.23 0.05 0.31 0.47 0.63 1.23 BM 4,167 0.28 0.13 0.00 0.21 0.35 0.57 0.85 Dual 4,167 0.23 0.42 0.00 0.00 0.00 1.00 1.00 Turnover 4,167 0.45 0.50 0.00 0.00 0.00 1.00 1.00 C_edu 4,167 2.26 0.81 1.00 2.00 2.00 3.00 3.00 D_edu 4,167 2.26 0.82 1.00 2.00 3.00 3.00 3.00 C_age 4,167 47.89 6.39 33.00 44.00 47.00 52.00 64.00 D_age 4,167 51.51 6.46 37.00 47.00 51.24 56.00 67.00 C_gender 4,167 0.94 0.24 0.00 1.00 1.00 1.00 1.00 D_gender 4,167 0.97 0.17 0.00 1.00 1.00 1.00 1.00 variables are less than 0.50, indicating no serious collinearity problem and significant positive correlation with internal control. 4.2. Multivariate tests Table 3 reports multivariate results based on panel data fixed effects regression analysis. The controls are lagged by one period to address contemporaneous correlation problems and heteroscedasticity. As shown in columns (1) and (2), the coefficients of Province and Geodist are significantly negative at the 1% and 5% levels (–8.839, t-value = –3.89; –0.355, t-value = –2.41), respectively, which means internal control quality will be lower if the geographical relationship is closer. This result also shows that the geographical proximity between the executives (such as fellow villagers) will reduce internal control quality and is not conducive to corporate governance. If the chairman’s and the CEO’s homes are located in the adjacent cities of two provinces, the value of Geodist may be less than the distance between two towns from the same city. Since there may be significant difference between two provinces, we exclude the sample for which Province equals one to better describe the effects of birth- place distance, as presented in columns (3) and (6). The results show H1 is still supported. The results for the control variables show that the top management turnover variables are significantly negative, indicating that the internal control quality of listed companies will be lower in a year in which the chairman or CEO changes than in a year when they remain the same. The coefficients of firm size, audit opin- ions, executive shareholding, and the market-to-book ratio are significantly positive. The coefficients for Big4 accounting firms and the leverage ratio are significantly negative, show - ing that internal control quality will be low when a Big4 accounting firm is hired to audit the firm as well as when liability is high. Similarly, if we replace the IC indicator with the ICW 352 J. YU ET AL. Table 3. g eographical relationship and internal control quality. IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −8.839*** 0.094*** (−3.89) (3.69) Geodist −0.355** −0.687** 0.002** 0.011** (−2.41) (−2.31) (2.25) (2.17) Dual 5.131 2.409 2.163 −0.083 −0.015 −0.053 (0.87) (0.44) (0.26) (−0.48) (−0.09) (−0.23) Turnover −7.881*** −7.416*** −16.009*** 0.070 0.059 0.188* (−2.86) (−2.72) (−4.22) (0.88) (0.75) (1.80) SOE 8.446* 8.251* 6.857 −0.060 −0.058 −0.002 (1.90) (1.85) (1.20) (−0.46) (−0.45) (−0.01) Law −4.639 −4.645 −16.377** 0.247 0.246 0.290 (−0.89) (−0.89) (−2.31) (1.63) (1.63) (1.48) Size 31.945*** 31.643*** 21.484*** −0.208 −0.206 −0.199 (6.93) (6.87) (3.24) (−1.55) (−1.54) (−1.09) First 38.358 38.338 −38.962 0.205 0.186 0.658 (1.32) (1.31) (−0.91) (0.24) (0.22) (0.55) Big4 −7.419** −7.759** −13.052*** −0.071* −0.072* −0.321** (−2.02) (−2.11) (−2.77) (−1.67) (−1.67) (−2.47) Opinion 27.906*** 28.044*** 47.684*** −0.274 −0.284 −0.373 (2.81) (2.82) (3.49) (−0.95) (−0.98) (−0.99) Share 1.350*** 1.336*** 1.253*** −0.003 −0.003 0.003 (5.30) (5.25) (3.81) (−0.47) (−0.47) (0.35) ROA 467.351*** 467.174*** 408.093*** −0.043 −0.042 0.656 (16.38) (16.37) (11.23) (−0.05) (−0.05) (0.65) Leverage −55.758*** −55.012*** −35.442* 0.283 0.262 0.544 (−3.78) (−3.73) (−1.73) (0.66) (0.61) (0.96) BM 211.360*** 212.497*** 214.783*** 1.605 1.601 1.506 (5.11) (5.14) (4.72) (1.34) (1.33) (1.20) C_edu −6.648** −6.635** 0.325 −0.125 −0.128 −0.003 (−2.09) (−2.09) (0.07) (−1.35) (−1.39) (−0.02) D_edu 0.804 0.859 −8.997 −0.023 −0.027 0.076 (0.29) (0.31) (−1.57) (−0.28) (−0.33) (0.48) C_age −0.002 0.009 0.191 0.010 0.009 −0.013 (−0.01) (0.02) (0.35) (0.87) (0.79) (−0.84) D_age 0.588 0.517 0.703 −0.014 −0.013 0.002 (1.54) (1.37) (1.11) (−1.26) (−1.16) (0.14) C_gender −5.835 −5.887 4.034 −0.006 0.006 0.260 (−0.51) (−0.52) (0.20) (−0.02) (0.02) (0.48) D_gender −5.604 −4.894 −2.525 0.542 0.533 0.388 (−0.46) (−0.41) (−0.13) (1.55) (1.52) (0.70) Constant −43.831 −42.768 172.972 5.981** 6.025** 5.068 (−0.45) (−0.44) (1.19) (2.11) (2.12) (1.26) Firm&Year&Industry Yes Yes Yes Yes Yes Yes observations 4,167 4,167 2,530 4,167 4,167 2,530 a dj. R 0.188 0.188 0.162 0.042 0.042 0.051 notes: a djusted t -values (double) are in parentheses. *, **, and *** represent significance at the 10%, 5%, and 1% levels, respectively. indicator, the results remain signic fi antly consistent. The regression coec ffi ient is in the oppo - site direction. The results are shown in columns (4) and (5) of Table 3, which also support H1. Table 4 reports the result for H2 (how a geographical relationship affects internal control quality for firms with different property rights). As shown in columns (1) and (2) of Table 3 , the coefficients of Province or Geodist are significantly negative at the 1% and 5% levels (–8.705, t-value = –3.16; –0.390, t-value = –2.54), respectively, indicating a significant neg- ative correlation with internal control quality, while the interaction terms (Province*SOE CHINA JOURNAL OF ACCOUNTING STUDIES 353 Table 4. g eographical relationship, property rights, and internal control quality. IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −8.705*** 0.091*** (−3.16) (2.76) Geodist −0.390** −2.585*** 0.003** 0.020** (−2.54) (−2.90) (2.44) (2.25) Province*SOE 0.398** −0.002** (2.19) (−2.16) Geodist*SOE 0.002* 1.476** −0.001* −0.008** (1.83) (2.21) (−1.74) (−2.23) SOE 8.082* 8.367* 6.047 −0.064 −0.029 0.000 (1.80) (1.85) (1.05) (−0.49) (−0.22) (0.00) Constant −57.087 −54.815 152.319 6.264** 6.278** 4.958 (−0.58) (−0.56) (1.04) (2.19) (2.20) (1.23) Controls Yes Yes Yes Yes Yes Yes Firm&Year&Industry Yes Yes Yes Yes Yes Yes observations 4,167 4,167 2,530 4,167 4,167 2,530 a dj. R 0.192 0.191 0.164 0.042 0.043 0.052 notes: a djusted t -values (double) are in parentheses. *, **, and *** represent significance at the 10%, 5%, and 1% levels, respectively. and Geodist*SOE) are significant positive, indicating that the negative correlation between geographical relationship and internal control in SOEs is weaker than in private enterprises. The finding means that the impact of internal relationship governance on internal control quality differs under different property rights. Hypothesis H2 is empirically supported. This may be due to the state-owned properties of SOEs and the personal political demands of executives. The internal relations between the chairman and the CEO are bound by organ- isational departments. The actual effects of internal control development and practice are not governed by internal relations. The impact mainly depends on national policy guidance and institutional constraints. To reduce risks and promote sustainable development, the executives of private firms have more incentive to improve internal control quality due to fewer political arrangements. Therefore, internal control quality depends mainly on the influ - ence of internal management on corporate governance, especially that of the chairman and the CEO. The same conclusion is also obtained using the ICW indicator (see columns 4–6 of Table 4). Table 5 validates the results of H3.As shown in columns (1) and (2) of Table 4, the coef- ficients of Province or Geodist are significantly negative at the 1% and 5% levels (–6.280, t-value = –3.30; –0.297, t-value = –2.16), respectively, and the coefficients of Province*Fnet or Geodist*Fnet are significantly positive at the 5% and 10% levels (19.066, t-value = 2.16; 0.847, t-value = 1.83), respectively, which indicates a significant negative correlation with internal control quality, but a positive correlation with the external network of top man- agement. The findings indicate that a close relationship between the chairman and the CEO can inhibit internal control quality due to internal and external supervisory pressure; the chairman and CEO thus construct external networks to enhance internal control quality and strengthen the weak corporate governance due to geographical proximity. Therefore, strengthening external relations governance can effectively enhance internal control quality. The same conclusion is also obtained using the ICW indicator (see columns 4–6 of Table 5). 354 J. YU ET AL. Table 5. g eographical relationship, external network, and internal control quality. IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −6.280*** 0.107*** (−3.30) (2.86) Geodist −0.297** −0.740*** 0.000** 0.011** (−2.16) (−3.31) (2.30) (2.37) Province*Fnet 19.066** −0.092* (2.16) (−1.76) Geodist*Fnet 0.847* 1.232** −0.020** −0.009** (1.83) (2.33) (−2.17) (−2.19) Fnet 6.249* 6.463* 19.262* −0.023 −0.174* −0.162 (1.91) (1.86) (1.65) (−1.14) (−1.83) (−1.50) Constant −53.275 −46.125 170.984 5.944** 5.970** 5.049 (−0.54) (−0.47) (1.17) (2.09) (2.10) (1.25) Controls Yes Yes Yes Yes Yes Yes Firm&Year&Industry Yes Yes Yes Yes Yes Yes observations 4,167 4,167 2,530 4,167 4,167 2,530 a dj. R 0.190 0.188 0.163 0.042 0.043 0.051 notes: a djusted t -values (double) are in parentheses. *, **, and *** represent significance at the 10%, 5%, and 1% levels, respectively. 4.3. Robustness test 4.3.1. Sub-sample test of the previous and following years for chairman or CEO turnover Since there is no change in geographical relationship when there is no chairman or CEO turnover, the changes in internal control may be caused by other factors. When there is a change, especially a geographical change, it is better to observe the difference in inter - nal control due to geographical relationship. Taking into account the huge difference in firm characteristics in the turnover year and the effect on internal control, we choose to observe the changes in the previous and following years for chairman or CEO turnover. Table 6 shows that the results are more pronounced and are consistent with previous findings. 4.3.2. Sub-sample with job separation of the chairman and the CEO The strongest geographical relationship that can create a corporate governance problem and affect internal control quality is CEO duality. Thus, the robustness test keeps only the sub-sample with separation of the chairman and CEO, Table 7 shows that the results are consistent with previous findings. 4.3.3. TSLS test The effect of corporate governance on internal control often leads to a very important endog - enous problem that must be addressed. We refer to Fracassi and Tate (2012) and Lu and Hu (2014) to alleviate problems of endogeneity. We construct an instrumental variable (IV ) with an external departure factor (Leave) and a social trust factor (Trust), since the departure of a chairman or CEO will affect changes in geographical relationship but will not have a direct untabulated tests showed that future t obin’s Q and cumulative excess returns decrease (increase) in companies whose geographical proximity becomes closer (more distant) after a change in executives. CHINA JOURNAL OF ACCOUNTING STUDIES 355 Table 6. t he sub-sample of the previous and following years for chairman or Ceo turnover. Panel A: Geographical relationship and internal control quality IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −5.576** 0.118** (−2.08) (2.12) Geodist −0.663** −7.037** 0.002* 0.049** (−2.10) (−2.55) (1.71) (2.58) Constant −285.589*** −296.702*** −280.925*** 1.222 1.254 1.915 (−6.19) (−6.53) (−4.56) (0.94) (0.98) (1.20) Controls Yes Yes Yes Yes Yes Yes Firm&Industry Yes Yes Yes Yes Yes Yes observations 1,292 1,292 838 1,292 1,292 838 a dj. R 0.521 0.523 0.498 0.059 0.059 0.077 Panel B: Geographical relationship, property rights, and internal control quality IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −6.090* 0.109* (−1.78) (1.75) Geodist −0.545* −6.531** 0.004* 0.054** (−1.70) (−2.40) (1.79) (2.25) Province*SOE 0.425** −0.010** (2.38) (−2.12) Geodist*SOE 0.002** 0.868** −0.002*** −0.001** (2.19) (2.43) (−2.80) (−2.49) SOE 0.514 1.428 −5.856 −0.126 −0.101 −0.174 (1.10) (1.28) (−0.91) (−0.86) (−0.69) (−1.02) Constant −287.204*** −299.445*** −290.909*** 1.128 0.964 1.774 (−6.17) (−6.52) (−4.67) (0.85) (0.74) (1.08) Controls Yes Yes Yes Yes Yes Yes Firm&Industry Yes Yes Yes Yes Yes Yes observations 1,292 1,292 838 1,292 1,292 838 a dj. R 0.530 0.530 0.513 0.060 0.065 0.077 Panel C: Geographical relationship, external network, and internal control quality IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −3.151** 0.077* (−2.57) (1.81) Geodist −0.781** −5.215* 0.002 0.040* (−2.39) (−1.75) (1.44) (1.72) Province*Fnet 11.490** −0.095** (2.13) (−2.35) Geodist*Fnet 0.896* 5.959** −0.013** −0.026** (1.88) (2.28) (2.38) (−1.26) Fnet 7.103 −2.952 3.670 −0.171 −0.229 −0.000 (1.12) (−0.42) (0.41) (−0.96) (−1.16) (−0.00) Constant −292.229*** −299.789*** −281.206*** 1.303 1.217 1.905 (−6.29) (−6.59) (−4.56) (0.99) (0.95) (1.19) Controls Yes Yes Yes Yes Yes Yes Firm&Industry Yes Yes Yes Yes Yes Yes observations 1,292 1,292 838 1,292 1,292 838 a dj. R 0.522 0.524 0.499 0.060 0.060 0.077 notes: a djusted t -values (double) are in parentheses. *, **, and *** represent significance at the 10%, 5%, and 1% levels, respectively. 356 J. YU ET AL. Table 7. t ests in the sub-sample with chairman–Ceo separation. Panel A: Geographical relationship and internal control quality IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −9.656* 0.042* (−1.86) (1.78) Geodist −0.446* −0.776** 0.002* 0.776** (−1.72) (−2.30) (1.68) (2.30) Constant −161.738 −161.158 58.169 7.920** 7.911** 58.169 (−1.43) (−1.42) (0.32) (2.42) (2.41) (0.32) Controls Yes Yes Yes Yes Yes Yes Firm&Year&Industry Yes Yes Yes Yes Yes Yes observations 3,175 3,175 1,581 3,175 3,175 1,581 a dj. R 0.188 0.188 0.162 0.042 0.042 0.051 Panel B: Geographical relationship, property rights, and internal control quality IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −9.501* 0.035** (−1.83) (2.23) Geodist −0.494* −3.131* 0.001 0.061* (−1.76) (−1.81) (1.49) (1.69) Province*SOE 1.263** −0.006** (2.36) (−2.36) Geodist*SOE 0.006** 1.648* −0.001** −0.014** (2.32) (1.72) (−2.08) (−2.39) SOE 10.818** 11.649** 7.500 −0.148 −0.094 0.034 (2.09) (2.22) (1.02) (−0.98) (−0.62) (0.16) Constant −184.790 −178.137 25.634 8.319** 8.193** 6.577 (−1.63) (−1.57) (0.14) (2.52) (2.49) (1.28) Controls Yes Yes Yes Yes Yes Yes Firm&Year&Industry Yes Yes Yes Yes Yes Yes observations 3,175 3,175 1,581 3,175 3,175 1,581 a dj. R 0.216 0.214 0.183 0.044 0.046 0.047 Panel C: Geographical relationship, external network, and internal control quality IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −8.229* 0.057* (−1.63) (1.67) Geodist −0.435 −0.779* 0.004 0.043* (−1.54) (−1.71) (1.54) (1.69) Province*Fnet 10.077** −0.107* (2.02) (−1.82) Geodist*Fnet 0.167** 1.167** −0.031* −0.002** (2.26) (2.28) (−1.67) (−2.01) Fnet 3.770 −0.120 −13.967 0.099 −0.207 −0.124 (0.60) (−1.01) (−0.94) (0.55) (−0.88) (−0.90) Constant −168.534 −163.395 55.445 7.748** 7.801** 6.713 (−1.48) (−1.44) (0.30) (2.35) (2.37) (1.31) Controls Yes Yes Yes Yes Yes Yes Firm&Year&Industry Yes Yes Yes Yes Yes Yes observations 3,175 3,175 1,581 3,175 3,175 1,581 a dj. R 0.210 0.209 0.180 0.045 0.046 0.047 notes: a djusted t -values (double) are in parentheses. *, **, and *** represent significance at the 10%, 5%, and 1% levels, respectively. CHINA JOURNAL OF ACCOUNTING STUDIES 357 Table 8. t wo-stage least squares test between geographical relationship and internal control quality. IC ICW Dep. Var. (1) (2) (3) (4) (5) (6) Province −21.802*** 3.602*** (−3.43) (2.80) Geodist −16.768** −19.295*** 0.271* 1.167* (−2.07) (−2.80) (1.82) (1.90) Constant −306.425*** 92.222 −260.022*** 6.547*** −0.098 4.583*** (−3.63) (1.13) (−2.90) (3.74) (−0.07) (2.86) Controls Yes Yes Yes Yes Yes Yes Firm&Year&Industry Yes Yes Yes Yes Yes Yes observations 4,167 4,167 2,530 4,167 4,167 2,530 a dj. R 0.24 0.14 0.13 0.23 0.14 0.13 f value 65.98 98.08 32.21 67.64 29.94 28.33 Weak identification test (f-statistic) 48.82 18.89 7.59 48.82 18.89 7.59 J-Statistic 2.24 1.73 0.19 2.28 1.69 0.01 (p-value) (0.12) (0.18) (0.66) (0.13) (0.19) (0.89) notes: a djusted t -values (double) are in parentheses. *, **, and *** represent significance at the 10%, 5%, and 1% levels, respectively. impact on internal control. Similarly, social trust is also exogenous, which will ae ff ct whether owners have a conspiracy motive to hire a geographically close CEO, but does not directly impact internal control. Table 8 reports the correlation tests, endogeneity tests, and second-stage regression results for the IVs. In the correlation test, the F-statistic is greater than 10, which means that the selected IV satisfies the dependency condition. The over-identification constraint p -value is greater than 0.10 in the IV endogeneity test, which means that the two IVs cannot be rejected due to the endogenous nature of the original hypothesis. The dependence and exogenous conditions provide strong evidence of the validity of the use of IVs in our study. After introduction of the IVs, the results of the main variables are still consistent. 5. Conclusion The rapid development of the market economy and the acceleration of the process of eco- nomic integration have made information technology increasingly prominent in the enter- prise, with a gradual increase in enterprise risk. As an important means of management, internal control plays a strong role in controlling and avoiding business risk and financial risk and its implementation is a key step in ensuring operating efficiency and sustainable development for enterprises. Therefore, internal control quality improvement has become key to corporate management success and the focus of research. From the perspective of informal institutional arrangements, our study empirically tests the effect of chairman–CEO relationship governance on internal control quality using data of Chinese companies listed from 2007 to 2013. This differs from prior literature, which con- centrates on financial activities and management. We find that strengthening the internal relationship governance or establishment of a geographical relationship between the chair- man and the CEO can harm internal control quality, and this negative effect should be weaker in SOEs than in private firms. However, the establishment and scale of an external network could moderate the negative effect of geographical proximity on internal control quality. The close geographical proximity of executives often cause serious damage to internal 358 J. YU ET AL. control quality, especially for private enterprises that lack resources and receive less support from local government, and executives can improve internal control quality and efficiency through external relationship governance. We conclude that collusion caused by a close geographical relationship between the chairman and the CEO will lead to a reduction of internal control quality and have a negative influence on firm value and market performance. However, an interlocking business network could regulate the negative impact of such collusion. Therefore, business owners should consider the advantages and disadvantages of relationship governance when deciding on the selection and appointment of senior managers and personnel arrangements should be conducive to enhancing the efficiency of corporate governance and internal control quality. Acknowledgements This work was supported by the Social Science Foundation of Zhejiang [No. 14NDJC166YB], the National Natural Science Foundation of China [No. 71402169], the China Postdoctoral Science Foundation [No. 2016M601606, 2017T100301, 71702102], the Humanity and Social Science Youth Foundation of Ministry of Education of China [No. 15YJC790137], the Social Science Foundation of Beijing [No. 15JGC176], and the Young Elite Teacher Project of the Central University of Finance and Economics [No. QYP1608]. Disclosure statement No potential conflict of interest was reported by the authors. References Abbott, L.J., Parker, S., & Peters, G.F. (2007). Corporate governance, audit quality, and the Sarbanes-Oxley Act: Evidence from internal audit outsourcing. The Accounting Review, 82, 803–835. Altamuro, J., & Beatty, A. (2010). How does internal control regulation affect financial reporting? Journal of Accounting and Economics, 49, 58–74. Ashbaugh-Skaife, H., Collins, D.W., & Kinney, W. (2007). The discovery and reporting of internal control deficiencies prior to SOX-mandated audits. Journal of Accounting and Economics, 44, 166–192. Bargeron, L.L., Lehn, K.M., & Zutter, C.J. (2010). Sarbanes–Oxley and corporate risk-taking. Journal of Accounting and Economics, 49, 34–52. Brochet, F. (2010). Information content of insider trades before and after the Sarbanes-Oxley Act. The Accounting Review, 85, 419–447. Cai, H.H., Zhou, L.A., & Wu, Y.Y. (2008). The clan rule, the faith of merchants and the management of merchant organizations: A comparative study on merchants of Anhui province and Shanxi province in the period of the Ming and Qing dynasty. Management World, 8, 87–99 (In Chinese). Doyle, J., Ge, W., & McVay, S. (2005). Determinants of weaknesses in internal control over financial reporting and the implications for earnings quality. Unpublished manuscript. Retrieved from https://papers. ssrn.com/sol3/papers.cfm?abstract_id=677622 Doyle, J., Ge, W., & McVay, S. (2007). Determinants of weaknesses in internal control over financial reporting. Journal of Accounting and Economics, 44, 193–223. Dyer, J.H., & Nobeoka, K. (2000). Creating and managing a high performance knowledge-sharing network: The Toyota case. Strategic Management Journal, 21, 345–367. Ellison, G., & Fudenberg, D. (1995). Word-of-mouth communication and social learning. Quarterly Journal of Economics, 110(1), 93–125. Fan, G., Wang, X.L., & Zhu, H.P. (2011). NERI index of marketization of China’s provinces 2011 report. Beijing: Economic Science Press (In Chinese). CHINA JOURNAL OF ACCOUNTING STUDIES 359 Fang, H.X. & Jin, Y.N. (2011). Can high quality internal control reduce earnings management? An empirical research based on voluntary internal control audits reporting. Accounting Research, 8, 53–60 (In Chinese). Fracassi, C. & Tate, U. (2012). External networking and internal firm governance. Journal of Finance, 67(1), 153–194. Gong, G., Ke, B., & Yu, Y. (2012). Home country investor protection, ownership structure and cross- listed firms’ compliance with SOX-mandated internal control deficiency disclosures. Contemporary Accounting Research, 30, 1490–1523. Gul, F.A., & Leung, S. (2004). Board leadership, outside directors’ expertise and voluntary corporate disclosures. Journal of Accounting and Public Policy, 23, 351–379. Hoitash, U., Hoitash, R., & Bedard, J.C. (2009). Corporate governance and internal control over financial reporting: A comparison of regulatory regimes. The Accounting Review, 84, 839–867. Huang, S.C., & Yang, X.S. (2010). Internal control report, quality of financial reports and information asymmetry: Empirical evidence from listed companies in Shanghai Stock Exchange. Journal of Finance and Economics, 36, 81–91 (In Chinese). Hwang, B.H., & Kim, S. (2009). It pays to have friends. Journal of Financial Economics, 93(1), 138–158. Jacobs, J.B. (1982). The concept of guanxi and local politics in a rural Chinese cultural setting. In S. Greenblatt, R. Wilson, & A. Wilson (Eds.), Social interaction in Chinese society. New York: Praeger. Jin, X., Lei, G.Y., & Yu, J.L. (2016). Government governance, executive networks and enterprise R&D expenditure. China Journal of Accounting Research, 9(1), 59–81. Kang, E., & Tan, B.R. (2008). Accounting choices and director interlocks: A social network approach to the voluntary expensing of stock option grants. Journal of Business Finance & Accounting, 35, 1079–1102. Khwaja, A.I., Mian, A., & Qamar, A. (2008). The value of business networks. Unpublished manuscript Harvard University. Retrieved from https://www.hks.harvard.edu/fs/akhwaja/papers/KhwajaMianQamar_ NetworksImpact_July25_final.pdf Krishnan, J. (2005). Audit committee quality and internal control: An empirical analysis. The Accounting Review, 80, 649–675. Li, C., Sun, L.L., & Ettredge, M. (2010). Financial executive qualifications, financial executive turnover and adverse SOX 404 opinions. Journal of Accounting and Economics, 50(1), 93–110. Li, W.F., Lin, B., & Song, L. (2011). The role played by the internal control in companies’ investment: Is it a promotion of efficiency or a repression thereof? Management World, 2, 81–99 (In Chinese). Liao, C., & Chen, J. (2009). Director networks and their similarity in R&D expenditure decision. International Research Journal of Finance and Economics, 23, 75–88. Lin, J.Y., Cai, F., & Li, Z. (1998). Competition, policy burdens and state-owned enterprise reform. American Economic Review, 88, 422–427. Lin, Y.F., & Li, Z.Y. (2004). Policy burden, moral hazard and soft budget constraint. Economic Research Journal, 2, 17–27 (In Chinese). Lin, Y.F., Liu, M.X., & Zhang, Q. (2004). Policy burden and enterprise’s soft budgetary binding: A case study from China. Management World, 8, 81–89 (In Chinese). Liu, Q.L., Luo, L., He, W.F., & Chen, H.W. (2012). State ownership, the institutional environment, and internal control quality: Evidence from Chinese listed firms. Accounting Research, 3, 52–61 (In Chinese). Lu, Y., & Hu, J.Y. (2014). The impact of hometown connectedness between the CEO and board of directors on corporate risk. Management World, 3, 131–138 (In Chinese). Ma, H.J., Dong, B.B., & Ge, B.S. (2010). The empirical research on the relationships among network capabilities, information acquisition and performance of high-tech firm. Studies in Science of Science, 28(1), 127–132 (In Chinese). Mao, X.S., & Meng, J. (2013). Internal control and litigation risk. Management World, 11, 155–165 (In Chinese). Mintz, B., & Schwartz, M. (1985). The power structure of American business. Chicago: University of Chicago Press. Naiker, V., & Sharma, D.S. (2009). Former audit partners on the audit committee and internal control deficiencies. The Accounting Review, 84, 559–587. 360 J. YU ET AL. Schmidt, B. (2015). Costs and benefits of friendly boards during mergers and acquisitions. Journal of Financial Economics, 117, 424–447. Stokman, F.N., Ziegler, R., & Scott, J. (1985). Networks of corporate power. Cambridge: Polity Press. Wang, B., & Liang, X.X. (2008). Corporate governance, financial condition and quality of disclosure: Evidence from Shenzhen Stock Exchange. Accounting Research, 2, 31–38 (In Chinese). Wang, X.L., Yu, J.W., & Fan, G. (2016). Eight-year process report on China’s marketization. Retrieved from http://www.rmlt.com.cn/2016/0414/423199.shtml (In Chinese). Wang, Y.Q. (2005). Reputation, commitment and organizational form: The comparative institutional analysis. Shanghai: Shanghai People’s Press (In Chinese). Yu, J.L., Jin, X., & Lei, G.Y. (2015). Geographical relationship of management and corporate investment efficiency. Contemporary Finance & Economics, 10, 116–128 (In Chinese). Zhang, H.L., & Wu, Y.H. (2014). Internal control, cash holdings and economic consequences. Accounting Research, 3, 71–78 (in Chinese). Zhang, P., & Xu, W. (2015). Does media supervision improve the effectiveness of internal control? Empirical evidence from Chinese listed enterprises. Journal of Accounting and Economics, 5, 88–105 (In Chinese). Zheng, Z.G., Li, D.X., Xu, R., Lin, R.T., & Zhao, X.J. (2012). The political promotion of the SOE executives: A case study on a certain corporation of a certain province. Management World, 10, 146–156 (In Chinese). Zhou, S.H., Hu, W.M., Lin, B., & Liu, C.L. (2013). Research on internal control of Chinese listed companies in 2012. Accounting Research, 7, 3–12 (In Chinese).

Journal

China Journal of Accounting Studies – Taylor & Francis

Published: Jul 3, 2017

Keywords: culture; geographical proximity; interlocking business network; internal control

There are no references for this article.