Audit Committee Chair’s Geographic Distance and Earnings Quality

. This paper empirically examines the relationship between the geographical distance of audit committee chairs(ACCs) and corporate earnings quality by taking A-share listed companies in Shanghai and Shenzhen from 2007 to 2018 as samples. It is found that the geographical distance between ACC’s work place and the operation place of listed companies can signiﬁcantly reduce earnings quality, and the farther the geographical distance, the worse the earnings quality. In addition, the external audit conducted by the Big4 auditors can signiﬁcantly suppress the impact of geographical distance on earnings quality. Further analysis found that there is a partial mediation e ﬀ ect in the quality of internal control. The research conclusions of this paper provide a new idea for for companies to appoint ACCs. The research on accounting and ﬁnancial issues from the perspective of economic geography is reﬁned to the ﬁeld of audit committee, which enriches the research literature on factors inﬂuencing the functions of ACCs. It is of great signiﬁcance to improve the performance of the committee, the long-term development of the company and the protection of investors’ interests.


Introduction
As one of the guarantees for the efficient operation of the capital market, the higher the earnings quality, the more effective information it can provide to users of financial reports [1]. When preparing financial reports, enterprise managers may adjust earnings information based on subjective judgments and purposefully adjust financial report through earnings management within the scope permitted by current accounting standards [2] [3]. Adjusted financial reports would be confusing and misleading, therefore, managers sometimes may use such misleading information to guide users of financial reports to make biased decisions. Adjusted earnings information cannot reflect real operating conditions, it harms the interests of the public. Besides, companies may also evade effective financial supervision. Therefore, our study on earnings quality trying to maintain true and neutral financial information, can better protect users of financial reports, help them make decisions with less bias, and companies can receive more comprehensive supervision, and sustainable operation.
The Sarbanes-Oxley Act (2002) of the United States clearly requires companies to set up an audit committee under the board of directors, whose responsibilities are mainly focused unique due to different legal environments and investor protection mechanism [15]. From the perspective of internal corporate governance, the higher the level of corporate governance, the higher the earnings quality [15,16]. In addition, both non-executive directors and outside independent directors have the abilities to improve earnings quality [17]. From more detailed point of views, the proportion of independent directors and audit committee members with financial expertise have positive relationship with the earnings quality, this is because they have stronger supervision ability. [18,19]. Francis et al. (2015) [20] further verified that accounting academic independent directors have a more significant positive impact on earnings quality than non-academic independent directors.

Literature Review of Geographical Distance
Geographical distance is the spatial distance between two economic subjects, also known as geographical proximity. The variable of localization is mostly used in existing studies to measure spatial and geographical factors. First of all, the increase of geographical distance will deepen the information asymmetry between the research subject and the company. Studies have shown that information asymmetry decreases when shareholders, independent directors, ACCs and institutional investors are localized [21]. This is because the proximity will directly reduce the cost of information acquisition, and will make the attendance and field research become more convenient. therefore, it will be easier to directly access high-quality information.
Secondly, previous study has proved that soft information that is difficult to be directly transmitted is greatly affected by geographic distance [10,22]. also verified that geographic proximity could reduce the cost of information acquisition between economic entities, especially soft information acquisition. in summary, we believes that geographical distance is more appropriate to measure the relationship between spatial factors and earnings quality. Using ACC's geographical distance as the explanatory variable is still a relatively blank stage now. Therefore, our study about ACC's geographic distance and earnings quality is meaningful.

Theoretical Analysis and Research Assumptions
A lot of soft information, such as enterprise development prospects and technical secrets always contain core business information. However, geographical distance will directly affect the acquisition of soft information. Existing studies have shown that geographic proximity can reduce agency incentives, capital costs and especially soft information acquisition costs between economic agents [23].
Geographical distance plays an important role in supervisory functions [24]. The increase of geographical distance will reduce the efficiency of information acquisition, weaken the communication between various economic entities and enterprise management. Therefore, it will then weaken the effectiveness of supervision. Existing research has demonstrated that geographic distance undermine the oversight function of shareholders, investors, independent directors and auditors [21,25]. Liu (2014) [26] proved that the closer between auditor and client, the lower the degree of earnings management of the audited financial report. However, members of audit committee are also independent directors of the company, audit committee is another organizational form for independent directors to supervise company. Therefore, the influence of the geographical distance to information acquisition, financial statements and earnings quality should also apply to the audit committee.
One of the main duties of the audit committee is to review financial information and the disclosure. ACC has an important impact on supervising financial reports and both internal and external audits. There are findings about education level and localization of ACCs can significantly suppress real surplus management [9,27]. Cheng et al. (2022) [22] also proved that local ACCs were significantly related to higher earnings quality.
Therefore, as the distance increases, ACCs can obtain less soft information. The communication frequency and efficiency with the internal management and employees will decrease accordingly. Besides, director is the link between company and external audit agency. As the geographical distance of ACCs increases, the communication with the auditor also decreases. It weakens the bond between auditor and company. In addition, the diligence and supervision of ACCs decreases correspondingly because of distance. The various influences on the performance will have a negative effect on the reliability of financial information and reduce earnings quality.
To sum up, we argues that the farther the geographical distance between ACC's work place and company operating place, the lower earnings quality will be. Therefore, we propose Hypothesis 1.
Hypothesis 1. The greater the geographical distance of ACCs, the lower earnings quality As an important link in the process of financial information review, external auditors play an important role in supervising and reviewing the financial reports information. On the one hand, the reputation of the firm is an extremely important factor to support its operation and development. Larger firms have more professional and prestigious auditors and partners. However, the better the reputation, the higher the loss that the firm will lose clients and audit fees. The audit fee is far from being able to make up for the loss caused by the reputation damage. Larger auditing agencies tend to provide more reliable and higher quality audit reports out of reputational protection [28]. On the other hand, the larger the size of the auditing agency, the auditor training will be more systematic and comprehensive. Scholars have proved that the audit opinions issued by large-scale firms contain higher audit quality, and can restrain earnings management and improve earnings quality [29].
Due to the geographical distance, ACCs cannot play a sufficient role in supervising, which will reduce the quality of information disclosure. However, due to reputation protection and other aspects, large-scale audit firms will dutifully provide high-quality audit reports in order to reduce audit risks. Due diligence of large firms can alleviate low-quality financial information and improve the disclosure quality. Therefore, when the geographical distance of ACCs is large, ACCs will be absent in responsibilities. For reasons such as reputation protection and risk aversion, large auditing agencies will strengthen the supervision, increase audit procedures. The supervision to a certain extent has alleviated the negative impact of the large geographical distance of ACCs on the quality of financial reports. Therefore, we believes that the external auditing by the four major firms can moderate the impact of the geographical distance of ACCs on earnings quality to a certain extent. Therefore, we propose Hypothesis 2.
Hypothesis 2. Big4 Auditors can mitigate the negative relationship between geographical distance and earnings quality.

Variable Metric and Model Construction
(1) Measurement of surplus quality of listed companies (absDA) Referring to existing research, we choose the absolute value of manipulated accruals to measure earnings quality. The discretionary accrual profit is calculated according to the modified Jones model, and specific calculation processes are as follows.
Firstly, regress formula (1) by industry and year (excluding industries with less than 20 samples in the year), and then substitute the regression coefficient into formula (2) to calculate the non-manipulative accrual profit in year t (NDA t ). (1) Among them, T A t is the total accrued profit in year t, which equals to net profit in year t minus net cash flow from operating activities in year t; A t−1 is the total assets at the end of year t − 1; ∆S ale t and ∆Rec t are the changes in operating income and accounts receivable in year t, respectively ; PPE t is the original value of fixed assets at the end of year t; NDA t is the non-discretionary accrued profit for year t.
Calculate the manipulated accrual profit in year t DA t = T A t /A t − 1 − NDA t , and take its absolute value to get absDAt.
For the measurement of the geographical distance of ACC, referring to existing studies, the distance between the longitude and latitude of the administrative center of ACC's main workplace and the listed company operating address. The latitude and longitude distance is calculated by the Haversine formula (Sinnott, 1984), and the specific calculations are shown in models (3) and (4).
At the same time, due to the fact that the company may have multiple ACCs in a certain year, we use the average geographical distance of all ACCs in that year as the geographical distance, and takes its natural logarithm as the explanatory variable ( LnDis_J).
(3) Model construction The explanatory variable is the geographical distance of ACCs (LnDis_J), and the explained variable is the earnings quality of listed companies (absDA). If the coefficient in model (5) is positive, it proves that the greater the geographical distance, the greater the manipulative accrual earnings, and the worse the earnings quality. Hypothesis 1 is valid. in order to verify the hypothesis 2, we divide the samples into Big4 group (Big4 =1) and the non-Big4 group (Big4=0) to regress again. Controls is a control variable. According to relevant research (Lu et al., 2021; Xiang and Yang, 2016), we choose to control Size, Lev, Roe, Liquidity, Rec, Inv, Board, Dual, Growth, and ListAge. Year and Ind are year dummy variables and industry dummy variables, respectively. All regressions in our essay were performed with robust standard errors.

Correlation Analysis
The Pearson correlation coefficient and P value among the main variables are shown in table 2. From the results of the correlation analysis, it can be concluded that the geographical distance of ACC (LnDis_J) and company's earnings quality (absDA) are positively correlated at a significant level of 0.05, the correlation coefficient is 0.019, which indicating that the main working place of ACCs and company's operational location are positively correlated. The greater the geographical distance, the worse the earnings quality, and the conclusion can preliminary verify H1. The Big4 auditors is positively correlated with absDA at the 1% significant level with a coefficient of 0.037. The findings illustrate that using Big4 can weaken the influence of the location of ACCs on earnings quality, which verifies H2. The estimated results in the empirical model later show that the variance inflation factor (VIF) of Note: *, * *, * * * are significant at the 10%, 5%, and 1% levels respectively (two-tailed test), with a p-value in parentheses these variables does not exceed 5, so the empirical model of this paper does not have serious multicollinearity problems.

Regression Analysis
(1) The test of hypothesis 1 Table 3 shows the regression results of the geographical distance (LnDis_J) and earnings quality of ACCs. Columns (1) and (2) regressions used the absolute value of manipulative accrued profit (absDA) as the interpreted variable to represent earnings quality. Column (1) regression only controlled industry and year, and (2) adds all control variables. The results illustrated that there was a positive correlation between the LnDis_J and absDA at a significant level of 1%, with a coefficient of 0.0019. It is in line with H1's expectation that the greater the geographical distance of ACC, the worse the earnings quality. In terms of controlling variables, the asset-liability ratio (Lev), inventory to total assets ratio (Inv), growth rate of operating income (Growth) and listing age (ListAge) could reduce the earnings quality at a significant level of 1%; the current ratio (Liquidity) and the accounts receivable to total assets ratio (Rec) were positively correlated with absDA at a significant level of 5%. The company size, return on net assets (Roe) and the number of directors (Board) can significantly improve earnings quality, which had a significant positive impact on it.
In addition, we further divide absDA into positive and negative groups, and find out the different effects of geographical distance on positive accrued earnings (+DA) and negative earnings (-DA). As can be seen from table 3, there are significant positive effects on both geographical distance and each side of accrued earnings, indicating that with the increase of geographical distance, the control of positive and negative earnings will increase, and the coefficients are 0.0016 and 0.0017, respectively. The results verify H1.
(2) The test of hypothesis 2 In the regression of table 4, we add the restriction condition to the geography of ACC (LnDis_J) and earnings quality (absDA) regression model, under Big4=1 (i. e., employ Big4 auditors for external auditing) and Big4=0 (employ auditors from firms other than Big4). The results showed that when Big4=0, there was no significant correlation between LnDis_J 132 Note: *, * *, * * * are significant at the 10%, 5%, and 1% levels respectively (two-tailed test), with a p-value in parentheses and absDA. In addition to the operating income growth rate (Growth), other control variables have no significant correlation on earnings quality. However, at Big4=0, LnDis_J was significantly and positively correlated with absDA at 1% of the significance level (p <0.01). And the control variable results did not fundamentally change. The regression results showed that using Big4 auditing could weaken the impact of the geographical distance of ACC on earnings quality. On the contrary, if hiring auditors in other firms, the geographical distance still has a significant positive impact on earnings quality. Thus the results correspond with H2.  0.163 0.130 Note: *, * *, * * * are significant at the 10%, 5%, and 1% levels respectively (two-tailed test), with a p-value in parentheses

Robustness Test
Where: ACC it calculated by operating profit minus net cash flow of operating activities; CFO i,t and CFO i,t+1 is net cash flow from operating activities for i company t − 1, t and t + 1, respectively. DCFO i,t−1 is a dummy variable, which equals 1 when CFO i,t − CFO i,t+1 < 0 , Otherwise is 0. ε i,t is regression residuals, which is the operational accrued profit as well. The greater the absolute value of the residual difference, The greater the probability for earnings management, The lower the earnings quality will be. To eliminate the scale effect, it is divided by the total assets ending in phase t − 1.
The regression result after replacing the explained variables is shown in table 5. Column (1) is the regression result of hypothesis 1, and in the case where the dependent variable  (2) Control  Control  Ind  Control  Control  Control  N  17705  1166  16539  Adjusted R2 0.071 0.082 0.073 Note: *, * *, and * * * are significant at the 10%, 5%, and 1% levels, respectively, with a p-value in parentheses is NLIN, the coefficient of the LnDis_J is 0.0010, it is significant at the level of 1%. The result means that the greater the geographical distance, the worse the earnings quality. The conclusion is consistent with H1. Our conclusion is robust. Columns (2) and (3) are the regression results of H2 under the nonlinear accrual model, that is, to distinguish whether a company adopts Big4 auditors, and the results are still representative. The results of column (2) show that there is no significant correlation when Big4=1, while the results of column (3) show that when Big4=0 , LnDis_J has a significant positive impact on earnings quality. It can be concluded that in the case of applying Big4 auditors, the impact of the geographical distance of ACC on the earnings quality will be weakened, and the regression results are consistent with previous conclusions, our conclusions are stable.
(2) Explanatory variable replacement  [10], the natural logarithm of the geographical distance between ACC's workplace and the company's registered place (LnDis_Z) is used in our test as the replacement variable into the original model (5) for re-inspection.
The results after the replacement variables are shown in table 6. It can be seen that in the whole sample regression column (1), when the dependent variable is absDA, the coefficient of LnDis_Z is 0.0020, and it is significant at the 1% significance level. The regression results of remaining control variables had no significant difference from previous test. H1 can be proved again. Column (2) and (3) are the regression to distinguish whether the listed companies adopt Big4 auditors. The results showed that LnDis_Z and absDA failed the significance test in the Big4=1 regression, with no significant correlation. However, the regression of Big4=1 showed a significant positive correlation with earnings quality at the 1% level, and the conclusion consistent with H2. The conclusions remains robust after replacing the measure of location.

Influence Mechanism Test
The previous test has confirmed that as the geographical distance of ACC increases, earnings quality of listed companies is worse. Internal control, as an important part of the corporate governance mechanism and the core path to monitor earnings quality within the governance system, is a significant external influence in the process of preparing financial report. The existing essay has found that the quality of internal control can affect the earnings quality degree [30]. It has been previously verified that there is a positive relationship between the geographical distance of ACC and earnings quality. Therefore, in order to verify whether the internal control quality is mediated by the geographical location and earnings quality effect of ACC, the regression model (7) and (8) are constructed here. We select the natural logarithm of "Dibo Enterprise Internal Control Index" as the proxy variable (IC) of internal control.

Conclusion
We took the sample of A-share listed companies in Shanghai and Shenzhen from 2007 to 2018, and focused on the geographical distance between ACC's workplace and the operating place of the listed company. We have empirically tested the impact of the geographical distance of ACC on earnings quality of the company, as well as the internal mechanism of the inhibitory factors and influence. The conclusions are as follow : (1) the geographical distance of ACCs can significantly reduce earnings quality; (2) adopt Big4 auditors can significantly suppress the impact of geographic distance on earnings quality; and (3) internal control quality has some intermediary effect in the aggravating effect of geographic distance on earnings quality. The implications of our findings are: (1) It confirmed that the geographical location of ACC had a significant impact on earnings quality. Therefore, geographical distance can be took into consideration as a factor when appointing ACCs, hiring proximal directors, in order to promote better performance of ACC's duties, and help to create the higher-quality audit and supervision functions; (2) At the same time, Based on our findings, company can adopt Big4 auditors for external auditing to mitigate the impact of ACC, reduce the manipulation of earnings, and then improve the authenticity and transparency of financial information; (3) The geographical distance of ACC has a direct impact on internal control quality. Strengthening supervision over internal control can affect earnings quality. Therefore, the company can improve earnings quality by improving the quality of internal control. Thus, the company has more countermeasures to improve the quality of financial report information, so as to protect shareholders and minority investors to a greater extent.
After more than a year of exploration of academic accumulating, my personal growth is significant. The accumulation to our essay has made me more patient and careful. At the same time, I also feel that I will be more familiar with and smooth in the future academic study. This essay is on the audit committee as the main theme, the influencing factors of the performance of the audit committee chair as the main direction. I am very grateful for finding this interesting field that I am very interested in. As a person who can be highly inspired by passion, it is very exciting to have the honor to encounter the direction that I am interested in. Therefore, I am always very cheerful in the process of gradually improving our essay. Furthermore, starting with the selection and collation of the testing data on this research direction and the existing research, I become increasingly interested in. 979 Note: *, * *, and * * * are significant at the 10%, 5%, and 1% levels, respectively, with a p-value in parentheses Up to now, our essay has been almost fully completed. In a year or so in preparation, all I want to say is thank you now. I am grateful to my professor for helping me find the topic selection and research direction that I am very interested in. I want to say thank you to my partners for cleaning and sorting out the data together with me, and to my teacher for his countless meticulous help to me. This essay cannot complete so well without every time of help and instruct. This is the achievement for everyone together.