Frontiers in Business, Economics and Management ISSN: 2766-824X | Vol. 5, No. 3, 2022 201 The Influence of Government Background Key Customers on Enterprise Information Quality Renping Zeng1, * 1School of Accounting, Anhui University of Finance and Economics, Bengbu, Anhui 233030, China * Corresponding author: Renping Zeng (Email: 2432235746@qq.com) Abstract: Based on the customer information data of listed companies in China from 2011 to 2020, this paper explores the influence of government background key customers on enterprise information quality from the perspective of supply chain. The results show that the big customers with government background help to improve the information quality of enterprises. Compared with the enterprises with poor audit opinions, the information quality of enterprises with good audit opinions is better promoted than those in the government background, they can improve the information quality of enterprises with large industry competition. And put forward relevant suggestions. Keywords: Customer relationship, Enterprise information quality, Resource effect. 1. Introduction The most direct way for enterprise stakeholders to understand the company's operation of the company is to obtain the information disclosed by the company. The quality of enterprise information affects the decisions of shareholders, creditors and investors. In recent years, the company's information fraud has emerged in an endless stream, and Kangmei Pharmaceutical, Kangdexin and other enterprises have been caught in the fraud case, which reflects that the overall information quality of Chinese enterprises needs to be improved. How to improve the information quality of enterprises to promote social and economic development is still a key topic. The outbreak of COVID-19 has made the economy of many industries enter a weak period. Government procurement behavior will undoubtedly provide "a timely help" to enterprises and inject new vitality into their business activities. Its most obvious effect is to increase the operating income of the enterprise, and then improve the performance, and promote the development of the enterprise. Another means of government support is government subsidy, which can solve the "urgent need" of enterprises without paying any price, and alleviate the capital problems caused by operating difficulties or research and development investment by obtaining a large amount of funds. Temporary subsidy can only suspend the plight of the enterprise, and cannot maintain the development of the enterprise. Once the enterprise loses the support of the local government, its performance will immediately "change its face" (Chen Xiao et al., 2001)[1]. Zhao Yujie et al. (2018) [2] found from the perspective of surplus management motivation that the company uses negative surplus management to improve government subsidies, which reflects that government subsidies will reduce the information quality of enterprises. The rapid expansion of government procurement investment in recent years shows that the government is increasingly relying on this behavior for macro-control of social economy. Therefore, it is worth exploring whether government procurement will not be as effective as government subsidies. Based on the manual arrangement of the top five customer information disclosed by listed companies in China from 2011 to 2020, the study found that big customers with government background help to improve the information quality of enterprises; and further inspection also found that these phenomena are more obvious in the audit opinions, non- "four major" audit enterprises. This research helps to discover the influence of the customer structure on the enterprise information quality. The research contributions mainly include: (1) combine the customer structure and the enterprise information quality to enrich the relevant literature; (2) the government procurement is the macro policy, the enterprise information quality is the micro phenomenon; (3) broaden the research perspective of the enterprise information quality for the enterprise operation and governance, the government macro control and the investment decision of small and medium investors. 2. Related Literature Review 2.1. On the Economic Consequences of Customer Relations on The Business All the time, customer relations have an important impact on the operation of the enterprise. Customer relationship is a kind of social relationship capital, which can bring valuable social resources to enterprises[3] . At the company level, supplier / customer relationship transactions can explain the earnings management behavior to some extent[4] . The existence of big customers is conducive to resource integration, upstream and downstream supply chain to build trust, make enterprise rent-seeking cost reduction, reduce costs, and increase the existence of the enterprise income, reduce the business risk, and send a positive signal to the market, reduce the audit risk, therefore, the auditor no longer use more audit procedures, reduce the audit cost, so as to reduce the audit cost[5] . Thus reducing the cost of equity of enterprises, but in an uncertain environment, the customer concentration of enterprises negatively affects its equity capital cost[6] . From the level of managers, the existence of key customers alleviates the pressure of performance, brings performance to managers, stable key customers bring continuous and stable income to the enterprise, transmits the good information of the enterprise to the outside world, and reduces the motivation of the enterprise management to control the accounting information due to the performance 202 pressure. However, as a strong external governance mechanism, it transmits operating pressure to the company's management through market competition[7]. Especially once the enterprise customer relationship is unstable, it may lead to the enterprise income instability. Based on the performance pressure, position and reputation factors of managers, it is easy to induce the management to control the accounting information, so as to reduce the quality of enterprise information. Stable customers may send the signals of relatively stable business operation and stable profit to the outside world. On the contrary, the frequent changes of customers show that the enterprise faces greater operational risks[8] . From the market level, the government background customers can effectively improve the capital market valuations of enterprises, enhance investors 'attention to enterprises, and significantly improve the analysts of the company's stock rating[9] , the existence of the government background big customers help to reduce the enterprise audit costs, the stronger, the higher the higher the political level of government background big customers can better help enterprises get lower audit fees, and the relationship in financing constraint degree is higher[10] . 2.2. Influencing Factors of Enterprise Information Quality The most direct way for enterprise stakeholders to understand the operating conditions of the company is to obtain from the information disclosed by the company. The quality of enterprise information affects the decisions of shareholders, creditors and investors. The quality of enterprise information disclosure is also of positive significance to the good order development of the capital market and the sustained economic growth[11] . Therefore, the quality of enterprise information has always been studied by scholars. From the perspective of influencing factors, Zhu Jie[12] (2020) found that there was a U-shaped relationship between the salary incentive of independent directors and the information disclosure violations of listed companies. Too low salary incentive will make independent directors lose the motivation to pay attention to the quality of the company's information, and reduce their enthusiasm. Excessive salary incentive makes independent directors to obtain high remuneration, cater to enterprise violations, and damage their independence. Only with moderate compensation can they really play the ability of independent directors to promote the quality of enterprise information.. Zhao Hua et al. [13] (2022) found that Confucian culture is positively related to the quality of enterprise accounting information, and mainly significantly affects the quality of enterprise accounting information by inhibiting the intermediary effect of self- interest motivation, following behavior regulation, rational investment decision, and reducing agency cost. The quality of accounting information can improve the efficiency of resource allocation by improving the efficiency of credit allocation, optimizing the factor input and improving the market competition mechanism (Li Qingyuan et al., 2021)[14] . High-risk enterprises actively improve the quality of accounting information to reduce the information risk of banks, so as to obtain credit resources. When enterprises are facing serious financing constraints, when it is difficult for banks to obtain enterprise private information from alternative information channels, and when the competition of commercial banks in the areas where enterprises are located is more fierce, the liberalization of the upper and lower limits of loan interest rates has a more significant impact on the quality of accounting information of high-risk enterprises[15] . Enterprise digital transformation can improve audit quality by improving information transparency and reducing enterprise risk taking[16]. From the perspective of economic consequences, the level of enterprise environmental accounting information disclosure is negatively correlated with the short-term value of enterprises, and positively correlated to the long-term value, and the value effect is more significant for non-state-owned enterprises than for state-owned enterprises[17] . The higher the quality of accounting information, the lower the degree of enterprise financing constraints; the higher the supplier concentration degree and customer concentration degree, the more obvious the mitigation effect of accounting information quality on financing constraints [18]. The quality of environmental information disclosure interferes with the improvement of enterprise innovation investment share; the degree of environmental regulation and the enterprise innovation ability; under the regulation of environmental regulation, the quality of enterprise innovation ability[19] . 3. Research Hypothesis 3.1. Government Background Key Customers and Enterprise Information Quality The previous literature found that key customers are conducive to the operating conditions of enterprises, which is specifically reflected in the continuous and large performance of big customers to the enterprise, alleviating the operating risks of enterprises, and promoting the development of the enterprise. But the outbreak has made the social economy even more unstable, which leads to instability in the supply chains. In this case, enterprises are eager to look for safe and stable customers. Because big customers with government background have attached support motivation, compared with ordinary enterprises, the benefits of big customers with government background are far greater than those of ordinary enterprises. First of all, based on the resource effect, the government background big customer revenue source for fiscal revenue, never go bankrupt, its business credit is good, at the same time, the government background big customer purchase order amount is larger, so can improve performance, reduce business risk, it also avoids the management in order to performance pressure and disclose untrue information. Second, based on the signal transmission effect, the government background big customer has natural reputation and supervision function, with government background big customer enterprise itself has passed the government test, it also for the good business signal, reduce enterprise information asymmetry, to improve enterprise goodwill, further reduce the business risk, and management is more willing to strictly control the enterprise information quality. To sum up, you can make a hypothesis: H1a: Government background of key customers can improve the quality of enterprise information. On the other hand, the social instability of economy makes enterprises always have survival pressure, which is also a test for enterprise management. Based on the particularity of the government background big customer identity, the quality of purchasing goods or service, the condition of the enterprise standard is higher, in order to attract the quality customers, does not conform to the procurement standards of enterprises 203 may through surplus management means hidden negative information, packaging into meet the requirements of the enterprise, in order to achieve the purpose of improving performance, leading to the enterprise information quality is reduced. At the same time, in a competitive industry, it is not only important to have safe and reliable customers, but also to maintain customer stability. Big customers in the performance benefits at the same time, also brought the "value plunder" crisis, strong big customer bargaining power, in order to retain big customers, enterprises have to make a compromise, lost the interests of the enterprise, to maintain the motivation of the good image of enterprise, enterprises choose surplus management behavior, lead to the enterprise information quality reduction, deceive stakeholders. In conclusion, the following assumptions can be proposed: H1b: Government background of key customers will reduce the quality of enterprise information. 3.2. Analysis of Heterogeneity Audit opinions have always been the first impression of investors, supervisors and creditors, which shows the importance of audit opinions. In recent years, the Chinese Association of Certified Public Accountants has had higher and higher requirements for auditors, and the quality of enterprise audit is also getting better and better. The audit risk of the enterprises that are issued with good audit opinions is higher, and the enterprises with poor audit opinions have lower audit risk. For low-risk customers, auditors adopt more detailed audit procedures in the audit process. Enterprises know that enterprises with government background tend to choose standard audit opinions and obtain non-standard audit opinions. Therefore, such enterprises have no incentive to promote the role of government background customers on enterprise information quality. For high-risk customers, auditors are more cautious to reduce litigation risk and reputation risk. However, the audit risk of the enterprises issued with standard opinions is high, and the sensitivity of the disclosure of information quality is high. Affected by the characteristics of customers, the enterprises have the incentive to improve the quality of information disclosure. Competitive pressure of industry, in the enterprise of the risk of instability, market saturation, in order to cope with the uncertainty risk, enterprises urgently need to enhance core competitiveness, so enterprises need stable and safe customer relations to bring a steady stream of cash flow, on the one hand, capital guarantee makes enterprise high quality development, on the other hand, the good image of the enterprise can attract more quality customers, and the quality of enterprise information is further improved. However, enterprises in industries with little competitive pressure rely on a moderate industry environment, with little survival pressure, and have a weaker impetus to improve enterprise information quality. Therefore, audit opinions are good, the greater the sensitivity of enterprises with greater industry competition pressure. Based on the natural supervision role of the government, the information quality of enterprises with large government background customers will be higher. In conclusion, the following assumptions can be made: H2: Compared with poor audit opinions, big customers with government background can help to improve enterprise information quality in enterprises with good audit opinions. H3: Compared with the industry competition pressure and the industry competition pressure, the enterprises with great competition pressure in the industry have a more significant impact on the quality of enterprise information. 4. Research Design 4.1. Sample Selection and Data Source The information of the top five customers of the listed companies is obtained through the National Taian database, and then whether the top five customers are the major customers with the government background is judged through the national enterprise information disclosure system. Finally, the data of listed companies from 2011 to 2020 were selected for research. Other data were obtained from the Taian database. In order to ensure the reliability of conclusions and data integrity, the following procedures were made: (1) excluding delisted, ST or * ST companies to avoid the impact caused by abnormal financial conditions of the enterprise; (2) excluding sample enterprises with missing data; and (3) to avoid the impact of extreme values, all continuous variables are treated by 1% tail reduction. After the above screening, 633 listed companies, a total of 2 721 companies-annual observations. 4.2. Variable-definition 4.2.1. Government Background (Gov1) Referring to the practice of Cohen et al. (2020)[20] and Dou Chao et al. (2020) [8], the major government customers are identified based on the specific names of the top five customers voluntarily disclosed by the listed companies in China. 4.2.2. Accounting Information Quality (Quality) Drawing on the measurement method of Sun Chenggang et al. (2017)[21] and Yang Haiyan et al. (2012)[22] , the quality of accounting information disclosure is measured according to the evaluation level of the Shenzhen Stock Exchange, with excellent, good, passing and failing scores of 1,2,3 and 4 respectively. Transparency of information disclosure. The Shenzhen Stock Exchange Information Disclosure Evaluation (Quality) is used to express the overall quality of the disclosure. 4.2.3. Other Main Variables Are As Shown in Table 1 4.3. Model Specification Qualityi,t=β0+β1Gov1i,t+∑（αj,t×Controlj,t）+εi,t (1) In the above model, Quality is the enterprise information quality, ε is the random error term, and Control is the control variable, including all the previous control variables, and the specific definition is as shown in Table 1. If the regression coefficient of government background customers (Gov) in the empirical results of the model (1) is negative and can pass the significance test, then assume that H1 is confirmed. That is, big customers with government background can promote the improvement of enterprise information quality. 204 Table 1. A List of Variable Definitions and Measures Type of variable sign Variable name definition definition Quality Listed companies disclosure evaluation scores According to the evaluation level of the Shenzhen Stock Exchange, the quality of accounting information disclosure is measured, with excellent, good, passing and failing counting 1,2,3 and 4 respectively. explanatory variable Gov1 Big government customers dumb variable. Whether the first five customers in the t- period have government departments (party, government, military departments and government institutions at all levels) and state-owned enterprises, it is 1, otherwise it is 0. Group variable Edit audit opinion The third party accounting firm of the company issues a standard for the company is 1, otherwise it is 0. SOE Is it a state-owned enterprise dumb variable. If the listed company is a state-owned enterprise, it is 1, otherwise it is 0. controlled variable Size company size Natural logarithm of the total assets at the end of period t. Growth increase rate of business revenue (Phase t-t-1 operating income) / t-1 operating income. Lev asset-liability ratio Total liabilities at the end of phase t / Total assets at the end of t Roa all capital earnings rate Net profit amount / total assets. Inst Share-holding ratio of institutional investors Share ratio of institutional investors at the end of period t. Mshare Management shareholding Shareratio of the company's management at the end of period t. Risk financial leverage (Net profit + income tax expense + financial expenses) / (net profit + income tax expense) Cft Operating cash flow Net cash flow of operating activities in year t / total assets in year t. Occupy Capital funds occupied by major shareholders Total assets at the end of period t. 5. Empirical Results Analysis 5.1. Descriptive Statistics Descriptive statistical results are shown in Table 2, it can be seen that the average of enterprise information quality (quality) is 1.982, the median is 2, indicating that the sample data is uniform, the average of government background big customers (gov1) is 0.533, indicating that the sample has nearly half of the top five customers with government background, which shows that government procurement behavior has penetrated many enterprises, once again shows the influence of government background big customers on the enterprise has important practical significance. Table 2. Descriptive statistics N Min Max Mean Median SD Quality 2721 1,000 4.000 1.982 2.000 0,540 Gov1 2721 0,000 1.000 0,533 1.000 0.499 Gig4 2721 0.000 1.000 0,052 0.000 0,222 Edit 2721 0.000 1.000 0,977 1,000 0,149 Size 2721 19.299 25.765 22,436 22,314 1,274 Growth 2721 -0.562 2.473 0,149 0.079 0.402 Lev 2721 0.068 0.917 0,474 0,482 0,202 Occupy 2721 -0.000 0.140 0,016 0,008 0,023 Roa 2721 0.053 0.199 0,042 0,032 0,039 Cft 2721 -0,201 0.237 0,047 0,046 0,070 Inst 2721 0,008 0.870 0,442 0,458 0,218 Risk 2721 0.000 9,131 1,605 1,177 1,354 5.2. Correlation Analysis The results of the correlation analysis are shown in Table 3. According to the correlation analysis results, the correlation coefficient between Gov1 and Quality is-0.094 respectively, and it is significantly correlated at the 1% level, which preliminarily shows that big customers with government background can promote the improvement of the quality of enterprise accounting information. The correlation of the control variables is shown in the figure below, but it is not repeated here. 205 Table 3: Correlation analysis Quality Gov1 Size Growth Lev Occupy Roa Cft Inst Risk Quality 1 Gov1 - 0.094*** 1 Size - 0.169*** 0.197* ** 1 Growth 0.012 -0.019 -0.006 1 Lev 0.057** * 0.089* ** 0.519** * 0.058** * 1 Occupy 0.128** * 0.034* 0.003 0.052** * 0.174** * 1 Roa - 0.161*** - 0.049** - 0.120*** 0.164** * - 0.361*** - 0.079*** 1 Cft - 0.115*** 0.012 0.035* -0.005 - 0.133*** - 0.155*** 0.394** * 1 Inst - 0.153*** 0.045* * 0.396** * -0.032* 0.151** * - 0.078*** 0.117** * 0.128** * 1 Risk 0.080** * 0.003 0.169** * - 0.069*** 0.366** * 0.024 - 0.331*** - 0.072*** - 0.013 1 5.3. Regression Analysis 5.3.1. Government Background Key Customers and Enterprise Accounting Information Quality The regression results of government background key customers (Gov1) on enterprise information quality (Quality) are shown in the Figure. (1) The coefficient of Gov1 is significantly negative at 1%, indicating that government background key customers have a positive role in promoting the quality of enterprise accounting information. It shows that the resource effect and supervision effect of government background big customers can dilute the "value plunder" of big customers, and the hypothesis of H1 is verified. The regression results of the control variables in the model (1) were basically as expected. 5.3.2. .Government Background Key Customers and Enterprise Accounting Information Quality in The Heterogeneity Analysis By the figure (3), (4) column, was issued by the standard audit opinions, government background big customer influence on enterprise information quality is not significant, and issued standard opinions of the enterprise, both significantly negative at 1% level, also, in (5), (6) column, the industry competition pressure of enterprise government background big customer influence on enterprise information quality on the level of 10%, but the industry competition in enterprises government background big customer influence on enterprise information quality on the level of 5%. This shows that in the enterprises with high audit risk and great competitive pressure, enterprises are willing to cater to the preferences of big customers and improve their own defects. Hypothesis 2 and 3 were verified. Table 4. Regression Results variable (1) Edit SOE Gov1 -0.0625*** (-2.80) 0.0638 (0.23) -0.0579*** (-2.63) -0.0451 (-1.33) -0.0876*** (-3.01) Size -0.0991*** (-8.72) -0.1065 (-1.02) -0.0922*** (-8.12) -0.1035*** (-5.83) -0.0961*** (-6.44) Growth 0.0299 (1.18) 0.4386** (2.78) 0.0254 (0.99) 0.0572 (1.60) -0.0055 (-0.16) Lev 0.2312*** (3.19) -0.7899 (-1.07) 0.1968*** (2.70) 0.3294*** (3.18) 0.2508** (2.42) Occupy 2.3461*** (4.83) 3.9225 (1.41) 1.6063*** (3.24) 3.0529*** (4.41) 1.3699** (2.02) Roa -1.6506*** (-5.12) -0.5286 (-0.19) -1.8224*** (-5.68) -2.5766*** (-5.62) -0.3340 (-0.71) Cft -0.3544** (-2.18) -1.9471 (-1.19) -0.2687* (-1.66) -0.0682 (-0.28) -0.5462*** (-2.61) Inst -0.2162*** (-4.09) 0.8917 (1.13) -0.1972*** (-3.79) -0.0271 (-0.33) -0.2880*** (-3.86) Risk 0.0064 (0.78) -0.0859 (-1.43) 0.0056 (0.68) -0.0076 (-0.48) 0.0134 (1.47) Year&Industry YES YES YES YES YES R-squared 0.138 0.912 0.152 0.223 0.206 Number of obs 2645 62 2648 1318 1392 Note: * * *, * * and * indicate the significant levels at 1%, 5% and 10%, respectively, and the t-value after heteroscedastic adjustment (Robust) is in brackets, the same as. 206 6. Robustness Test 6.1. Tool Variable Method Although this paper controls the year and the industry in the regression, there is still no guarantee that all the influencing factors are considered. Therefore, to make the above test results more robust, the instrumental variable method was used to solve the endogeneity problem. The sample was regressed with the explained variable as the instrumental variable, and the influence of the government background of the big customer on the quality of the enterprise information was significant at 5%, indicating that the relationship between the two was still robust after considering the endogeneity. Based on the limited space, the chart is not shown. 6.2. Replace is The Explanatory Variable The enterprise information quality of the manipulated accrued profit (accearn) was calculated by the modified Jones model, and the sample regression was repeated. The results showed that the Gov1 coefficient was negative and significant at the 5% level, and the test results were robust. 6.3. Replace Explanatory Variables Similarly, the sales ratio of customers with government background among the previous five major customers, as the big customers with government background as the explained variable, was regressed on the sample again, and the results were still robust. 7. Conclusions In recent years, the state has been aware of some disadvantages of the direct government subsidies, and the procurement intensity has become more and more intense, which affects the production and operation, strategic planning, development prospects and enterprise behavior of the downstream enterprises in the supply chain. The data of government background customers from 2011-2020 were manually collated, and the influence of government background big customers on enterprise information quality was explored from the perspective of supply chain. It is found that the key government customers can improve the information quality of suppliers; and the information quality of good audit opinions is more obviously promoted by the key customers with poor government background than those with poor industry competition. Conclusion provides reference for policy formulation, enterprise management and investor investment decision. Based on the conclusion of the above research, the following suggestions are made: (1) increase procurement and try to replace government subsidies through procurement; (2) regulatory authorities should strengthen supervision, standardize market mechanism and improve enterprise information transparency; (3) small and medium-sized investors can consider whether their key customers have government background to assist in making decisions before making investment decisions. Acknowledgment The study was funded by Anhui University of Finance and Economics Graduate Innovation Fund Project (ACYC2021457). References [1] X Chen, J Li. Analysis of the role of local government financial behavior in improving the performance of listed companies [J]. Accounting Research, 2001 (12): 20-28 + 64. [2] YJ Zhao ,ML Liu . Surplus management, executive motives, and government grants [J]. Securities Market Herald, 2018 (04): 42-51. [3] XY Shi ,XZ Zhang . Big customer is a resource or a risk —— is based on the IPO pricing scenario test [J]. Journal of Shanxi University of Finance and Economics, 2021,43 (08): 112-126. [4] HX Fang, Y Zhang . Supplier / Customer Relations Transactions, Surplus Management, and Auditor Decisions [J]. Accounting Research, 2016 (01): 79-86 + 96. [5] XY Wang , P Wang , JP Zhang. Customer concentration and audit expenses: customer risk or supply chain integration [J]. Audit research, 2014 (06): 72-82. [6] J Chen , XY Wang , Peng Xuan. Environmental uncertainty, customer concentration, and equity capital costs [J]. Accounting Research, 2015 (11): 76-82 + 97. [7] DM Kong , SS Liu , and YN Wang . Market competition, Property Rights and government subsidies [J]. Economic Research, 2013,48 (02): 55-67. [8] XY Wang , KJ Gao . Such: underwriters, key customers and corporate bond issuance pricing [J]. Management World, 2017 (09): 42-59 + 187-188. [9] C Dou , X Chen , and XZ Li . Government background customers, market cognition and investment opportunities — — is based on the perspective of supply chain integration [J]. Management Review, 2020,32 (08): 13-28. [10] C Dou , M Yuan , and X Chen . —— is based on the perspective of supply chain risk transmission [J]. Accounting Research, 2020 (03): 164-178. [11] JZ Du , ZY Wu . Research on the Impact of Digital Economy on the Quality of Enterprise Information Disclosure [J]. Friends of Accounting, 2022 (20): 72-78. [12] J Zhu . Compensation incentive of independent directors and information disclosure violation of listed companies [J]. Audit and Economic Research, 2020,35 (02): 77-86. [13] H Zhao , WF Li . Research on Enterprise Accounting Information Quality [J]. Friends of Accounting, 2022 (11): 47- 54. [14] QY Li , XS Liu . Accounting information quality and resource allocation —— comes from the empirical evidence of industrial enterprises above designated size in China [J]. Accounting Research, 2021 (08): 3-21. [15] LJ Ma , WY Zhang , and L Xie . Interest rate liberalization and enterprise accounting information quality ——. A quasi- natural experiment based on loan interest rate deregulation [J]. Accounting Research, 2022 (04): 3-21. [16] HY Zhai , QR Li . Has the digital transformation of enterprises improve the audit quality? The —— is based on the empirical test of a multi-time-point dual difference model [J]. Audit and Economic Research, 2022,37 (02): 69-80. [17] JF Shen , JJ Li . Research on the influence of enterprise environmental accounting information disclosure quality on enterprise value under the goal of "dual-carbon". —— is based on the empirical analysis of electric power enterprises [J]. Wuhan Finance, 2022 (04): 55-63. 207 [18] XJ Li , X He . Supply chain Relations, Accounting Information Quality and Financing Constraints —— is based on the empirical study of Chinese A-share listed companies [J]. Technical economy and Management Research, 2022 (02): 80- 84. [19] XM Cui , SB Wen , and BB Li . Environmental information disclosure and enterprise innovation: Promote or crowding out —— based on the "Porter hypothesis" [J]. Accounting and Communications, 2021 (18): 30-35. [20] Cohen D A,Li B. Customer-base concentration,investment,and profitability:The U.S. government as a major customer [J]. The Accounting Review,2020,95(1),101−131. [21] CG Sun , XB Huang . Accounting information quality and enterprise investment Efficiency —— is based on the empirical evidence of Shenzhen Stock Exchange listed companies [J]. Economics and Management Research, 2017,38 (08): 136-144. [22] HY Yang , DH Wei , Sun Jian. Can institutional investors' holdings improve the quality of accounting information of listed companies?—— and on the differences of different types of institutional investors [J]. Accounting Research, 2012 (09): 16-23 + 96.