Government-enterprise Relation, Financing Plight and Enterprise Growth: Empirical Evidence from World Bank on Chinese Firms

Theoretically, the influence of government-enterprise relation on enterprise growth is divided into lubricating effect and friction effect. However, the government-enterprise relation has been discussed in only a few studies from the perspectives of enterprise financing and willingness. No unified conclusions have been formed regarding the influential consequences of government-enterprise relation. To investigate the influence mechanism of government-enterprise relation on enterprise growth, World Bank Survey of Chinese Enterprises data in 2012 was used in this study to empirically verify how the government-enterprise relation influenced enterprise growth, the mediating effect of financing plight via mediating effect models. The differentiated influences of governmententerprise relation on enterprise growth were explored through the grouping test of corporate ownership and scale. Moreover, the causal relation between governmententerprise relation and enterprise growth (the former facilitated the latter) was further tested and recognized through the instrumental variable method. Results show that the government-enterprise relation is capable of boosting enterprise growth by mitigating the financing plight faced by enterprises, thus proving the lubricating effect of governmententerprise relation. The findings still hold after the endogeneity problem is weakened by using the instrumental variable method and after the robustness test. In the grouping test, a good government-enterprise relation can promote the growth of private enterprises and small-sized enterprises by remitting the financing plight they face. The conclusions obtained from this study can provide a beneficial reference for expanding enterprise financing channels and constructing a complete financial system, and lay a theoretical foundation for probing the influences of government-enterprise relation.


Introduction
Small and medium-sized enterprises play an important role in the economy of China. An important government duty lies in creating a benign enterprise growth environment and in abating the institutional barrier and constraint faced in enterprise development. The current institutional construction matching China's market economy is imperfect, and government intervention in resource allocation forces numerous enterprises to take an active part in politics. To acquire the resources necessary for their development, enterprises need to establish a stable and close cooperative relation with the government. In many countries, enterprises strive for and hope to establish a solid long-term relationship with the government (Faccio, 2006). The unique Chinese characteristics in creating relations can answer why China's economy maintains rapid growth over the years (Allen et al., 2005). Putting the high-speed economic development aside, the deep problems created in the establishment of government-enterprise relation need to be urgently solved for the development and reform of China's economic system.
When enterprises make contact with the government and support their growth using government rights Goldman et al., 2009), the lubricating effect of government-enterprise relation is demonstrated. By contrast, the friction effect occurs when enterprises spend excessive costs in handling the government-enterprise relation, and these efforts go against enterprise growth (Fan et al., 2007;Boubakri et al., 2008). In addition, scholars believe that the government-enterprise relation is not simply positively or negatively correlated with enterprise growth.
According to the literature, enterprises seek to establish relations with the government to acquire government orders, fiscal support, as well as efficient government review, approval, and credit aid (Luo and Tang, 2009;Pan et al., 2013;Agudelo et al., 2019). Therefore, the government-enterprise relation is of enormous significance for the operation and growth of Chinese enterprises. The financing plight faced in enterprise growth can be effectively mitigated by a complete financial system and stable bank credit sources (Eisenhardt, 1990;Beck et al., 2006). According to traditional economics, the financing constraints of enterprises are caused by information asymmetry; but under the current situation, they are rooted in the national development strategies and institutional arrangement (Zhu et al., 2015). On the one hand, the allocation of key credit resources is still controlled by the government, and bank credit is also dominated by the government (Sapienza, 2004). Given the discrimination in bank credits, financial institutions within the system are affected by the government, and they will rationally reduce the loans granted to small and medium-sized private enterprises, which are always faced with intense external financing barriers (Zhang, 2000). On the other hand, relative to state-owned enterprises, other enterprises, especially small and medium-sized private enterprises, have limited resources. Bank loans can hardly support their growth because of few fixed assets and mortgages and small scale; hence, private enterprises are stuck in prominent financing plights. The capital market presents a policy inclination to large state-owned enterprises, which are given special treatment in listing review and approval procedures and bear fewer bond issuing limitations. By contrast, heavy barriers are imposed on the financing of small and medium-sized private enterprises in the capital market, thus bringing about financing difficulties to enterprise growth. On this basis, enterprises have intense motive to establish political connections (Morck and Yeung, 2004;Li et al., 2006) to acquire resources and overcome financing difficulties, and one of the important strategic decisions for them is to build a good government-enterprise relation (Zhang and Zhang, 2005). A good government-enterprise relation, which is a scarce resource, can serve as an informal institution that evades the adverse effects caused by the imperfect market mechanism to enterprise development (Chen et al., 2011a). More importantly, the government-enterprise relation is a form of signaling, which embodies the external reputation of an enterprise (Luo and Zhen, 2008). If the market mechanism is not completely established in an economic entity, then bank credit agencies will usually take the enterprise quality signal transmitted by a good government-enterprise relation as an important instrument for risk identification. Therefore, in a transitional economic entity, enterprises that enjoy a good government-enterprise relation can win good graces from credit agencies, and they are less probable to face financing plights.
On this basis, under the current institutional environment, a good government-enterprise relation can relieve financing plights faced in enterprise growth. However, theoretical hypotheses in existing research have been scarcely explained through empirical analyses. Based on the World Bank Survey of Chinese Enterprises data in 2012, how the government-enterprise relation influenced enterprise growth was empirically verified by using a mediating effect model. Our empirical analysis results showed that enterprise growth could be positively influenced by good government-enterprise relation. The subsequent mechanism verification manifested that government-enterprise relation could facilitate enterprise growth by relieving enterprises' financing difficulties. Moreover, the above conclusion remained prominent after the endogeneity problem was weakened by the instrumental variable method and the robustness test was carried out. Research on individual samples discovered that private enterprises, foreign-owned enterprises, and small enterprises gained growth by establishing relations with the government, and a proper government-enterprise relation could boost the growth of these enterprises by remitting their financing plights.
The main contributions of this study are as follows: (1) government-enterprise relation, financing plight, and enterprise growth have not been incorporated into a unified framework for analysis in most of the extant research. However, a core viewpoint held in this research is that measures can be taken to break through enterprise growth bottlenecks only when the action mechanisms of the three are clarified. On this basis, the influencing mechanisms of government-enterprise relation, financing plight, and enterprise growth are comprehensively explored, in an effort to provide new evidence for probing the corporate governance problem. (2) The mean value of government-enterprise relations established by enterprises of the same type has been mostly used as an instrumental variable to relieve the endogeneity problem. Unlike in the existing literature, the number of times that enterprises had been inspected or visited by government officials in the previous year is taken as an instrumental variable in this study to remit the endogeneity problem. (3) Whether any senior manager in the board of directors has taken any post in government sectors has been mostly used to measure government-enterprise relation in extant literature. In this study, government-enterprise relation is measured by the time spent by senior managers in handling rules and regulations according to the data structure of World Bank. In addition, the index is substituted to perform the robustness test.
The rest of this study is structured as follows: Section 1 is the literature review and research hypotheses. Section 2 is the research design, introduces the mediating effect models and variable selection, and presents the descriptive statistics. Section 3 shows the regression results. Section 4 expounds on the handling of the endogeneity problem and the robustness test. Section 5 presents the research conclusions.

Government-enterprise relation and enterprise financing
Enterprises can win government resources by establishing relations with the government (Fisman, 2001;Xu and Liu, 2020). Good government-enterprise relation can influence enterprise financing directly and indirectly. The government is capable of directly providing financial support for enterprises and regulating their tax rate, thus influencing the financing. Using the panel data of 900 enterprises with and without relations with the government in 35 countries during 1997during -2002during , Faccio et al. (2006 deem that enterprises with political background influenced the possibility of government subsidies and the subsidy level, and companies could acquire government assistance and thus eliminate financial plights by the good government-enterprise relation under extreme cases. Taking private enterprises listed in China from 1993 to 2008 as the sample, Chen et al. (2011b) found that a good government-enterprise relation could help enterprises gain government financial aids, especially in regions with low economic marketization degree where the government had a powerful autonomous right in allocating economic resources. Moreover, they found that the rent-seeking behavior of government officials forced private enterprises to establish a relation with the government and maintain their interests. Kamasak et al. (2019) found that corporate social responsibility (CSR) and corporate political activities are complementary, and the coordinated management of corporate social responsibility and corporate political activities may lead to better firm performance.
Literature focusing on Chinese enterprise indicates that a good government-enterprise relation endows enterprises with diversified financial resources. Using Chinese private enterprises during 2002-2007 as the research samples, Yu et al. (2010) tested whether the government could give full play to the social effect, enterprise effect, and economic effect of fiscal subsidies if enterprises sought government-enterprise relation. The results showed that enterprises managing a good government-enterprise relation gained more access to fiscal subsidies, with obviously improved social benefits and enterprise benefits. Furthermore, this effect of acquiring fiscal subsidies was stronger in regions featured by low marketization degree, low level of rule by law, poor property rights protection, and many government-dominated resources. Further research has been carried out to discuss the influences of government-enterprise relation by taking companies in financial distress as the samples. For instance, Pan et al. (2009) sampled companies listed in Shanghai and Shenzhen during 2002-2007, which were suffering from abnormal financial status and operating status. They verified that companies in financial difficulties could acquire more government subsidies by building good government-enterprise relations. Moreover, private enterprises make more efforts to build a stable long-term relation with the government so that they can gain financial support from the government when financial plights arise. However, this effect is not significant among state-owned enterprise samples. Further analysis shows that government-enterprise relation can bring more resources to private enterprises only when they are located in regions with abundant local financial resources.
Besides, a good government-enterprise relation can lower the tax rate and relieve tax burdens on enterprises. In a research specific to Malaysia, Adhikari et al. (2006) analyzed

Government-enterprise Relation, Financing Plight and Enterprise Growth: Empirical Evidence from World Bank on Chinese Firms
how the government-enterprise relation influenced the effective tax rate of enterprises using the data of local listed companies during 1990-1999. They found that the actual tax rate of enterprises striving to establish a relation with the government was significantly lower than that of other samples in the research. By surveying over 500 listed private companies in China during 19992004, Wu et al. (2009) found that a favourable government-enterprise relation could bring tax preferences to enterprises in provinces that impose heavy tax burdens. Shen and Zou (2017) used the survey data of the World Bank on the investment environment of Chinese enterprises to explore how the government-enterprise relation influenced the tax rate of enterprises. They discovered that a good government-enterprise relation usually existed in large-scale enterprises, and it could bring preferences to enterprises in actual tax rate only at the local tax level but not at the national tax level.
The indirect influence of government-enterprise relation on enterprise financing is discussed mainly from the perspectives of enterprises acquiring bank loans and solving the information asymmetry between enterprises and government. From the theoretical level, the establishment of government-enterprise relation means that the government provides an implicit guarantee for external financing of enterprises. Financial institutions or other investors deem that the risk of lending or investing on these enterprises is low; hence, the financing convenience or financing availability of enterprises striving for relations with the government is enhanced (Johnson and Mitton, 2003;Claessens et al., 2008). In empirical research, enterprises enjoying such government-enterprise relation need fewer guarantees but more government-endorsed letters of credit for borrowing compared with other enterprises (Charumilind et al., 2006). Yu and Pan (2008) explored whether the loan effect of government-enterprise relation established by listed private companies would vary significantly with the institutional environment. They found that when seeking relations with the government, private enterprises could obtain the credit resources necessary for their development by relying upon the influence of government officials on state-owned banks. Moreover, they could reduce or overcome the ownership discrimination problem of credits. A good government-enterprise relation exerts a more significant loan effect in regions with low financial development level, low level of rule by law, and severe infringement upon property rights. Enterprises hoping to occupy a dominant information position are also pleased to establish relations with the government. Unlike individual enterprises, the government can master the development status, credit supply, and use of funds of enterprises within the industry, acquire the information that cannot be mastered by enterprises from the macroscopic level (Lin et al., 2010), and guide the decisions of individual enterprises and judge their investment opportunities. Many scholars think that enterprises that have established government-enterprise relations can increase their debt ratio according to the acquired government support . Furthermore, they can obtain loans from banks, with lower interest expenses and financial expenses (He, 2011). In the end, they gain access to more short-term debt financing from the aspect of loan structure (Li et al., 2016). On this basis, the following hypothesis is proposed:  Hypothesis 1: Enterprises that strive for government-enterprise relations have stronger financing abilities than enterprises without government-enterprise relations.
Private enterprises can keep and continuously arouse the vitality of the market economy. However, the "ownership discrimination" imposed by financial institutions on the credit financing of enterprises is a common phenomenon in China. On the one hand, state-owned enterprises are under a dominant position in the market by support and protection of  (Zhang, 2007). In the form of invisible subsidies for state-owned enterprises, the government will lend bank credits to state-owned enterprises at an interest rate lower than that in the market (Liu, 2007). On the other hand, private enterprises are exposed to financing difficulties due to the relatively backward legal system and property rights protection. With the unsound creditor's interest protection mechanism, creditors are forced to protect their interests through other channels. Failing to establish a natural relation with the government, state-owned enterprises firmly believe that the debt financing of state-owned enterprises can win additional government guarantees out of the trust in the government; hence, creditors like banks or other financial institutions grant more loans to state-owned enterprises (Pan et al., 2013). However, creditors may begrudge loans to private enterprises with small scale and high risks, as they are worried about the possible failure to recover the loans.
A primary reason for the prominent financing plight of private enterprises is their congenital advantages in enterprise financing (Luo and Zhen, 2008). Hence, more private enterprises expect to enhance their financing abilities by forming sound governmententerprise relations (Yu et al., 2012). Taking the top 100 private enterprises in Zhejiang Province in 2004 as the research samples, Hu (2006) discussed the relation between political identity of private entrepreneurs and financing convenience of private enterprises, and found that the political signal transmitted by private entrepreneurs increased the possibility for financial institutions to select private enterprises as their clients. This effect was remarkable under the institutional background of an imperfect judicial system. Accordingly, the following research hypothesis is raised:  Hypothesis 2: Government-enterprise relation exerts a more significant effect on the financing ability of private enterprises relative to state-owned enterprises.

Government-enterprise relation and enterprise growth
Enterprises come into contact with the government and support their growth owing to the government power Goldman et al., 2009). Faccio et al. (2006) thought that enterprises could increase their value and boost their growth by establishing political relations, and this effect was evident in countries or regions with weak property rights protection. Furthermore, enterprise decisions are significantly affected by the legal system, judicial operation efficiency, property rights protection, and financial system in a country or region (Yuan et al., 2015), and corporate reputation, resource commitment also affect enterprise decisions (Hesari et al., 2021). The political resource is also one of the important company resources (Boubakri et al., 2008). On the one hand, government-enterprise relations can support enterprises to expand their credit channels and increase government subsidies (Johnson and Milton, 2003;Claessens et al., 2008). On the other hand, a good government-enterprise relation can reduce the equity financing and credit financing costs needed for enterprise growth (Boubakri et al., 2012;Liu et al., 2013). Further analysis shows that a good government-enterprise relation can facilitate enterprises to master social capitals and exert the effect of relation capital (Li and Huang, 2010). After identifying the relations among society, government, and property rights, Tian and Zhang (2013) used the data of listed companies during 2001-2008, proposed hypotheses from three aspectssociety, government, and property rightsand carried out an empirical test. Their results showed that the long-term development of Chinese listed companies could be boosted by government-enterprise relations.

Government-enterprise Relation, Financing Plight and Enterprise Growth: Empirical Evidence from World Bank on Chinese Firms
According to scholars, enterprises spend excessive costs in handling government-enterprise relations, which is beneficial for enterprise growth (Fan et al., 2007;Boubakri et al., 2008). First, resource allocation is a focus in the market economy. When establishing the government-enterprise relation out of their interests, enterprises will reduce the resource allocation efficiency and enterprise investment efficiency (Asiedu and Freeman, 2009). Second, if the government-enterprise relation is inappropriately handled, an actual corruptive behavior may arise, which will be followed by other market subjects, thus aggravating social corruption (Liu and Peng, 2012).
Of course, scholars deem that the government-enterprise relation does not simply have a positive or negative correlation with enterprise growth. They support a view that government-enterprise relation presents a nonlinear relation with enterprise growth or the two are not correlated. Park and Luo (2001) indicated that the government-enterprise relation does not always influence enterprise growth. In developed Western economies, the marketization mechanism could effectively inhibit government rent-seeking and corruption and weaken the loan effect of government-enterprise relation. Taking China in the transitional period as the research object, Zhang (2013) pointed out the "Janus bipolar" in China, investigated the motives of enterprises with different scales and ownership patterns in seeking for the government-enterprise relation and the effect differences. The results showed that the age and scale of private enterprises had a nonlinear relation with government-enterprise relation, and only medium-sized and middle-aged private enterprises enjoyed faster growth through frequent interaction with the government. A possible reason is that the government has development intention and rent-seeking motive. By comparison, enterprises have two objectives: a good government-enterprise relation can prevent government rent-seeking and lower business operation costs and risks, and enterprises can acquire more preferential resources and policies through government relations.
In this study, the government-enterprise relation in China differs from what is expounded in Western literature. Therefore, the definition and measurement of government-enterprise relation need to be localized. In developed Western economies, enterprises generally influence government decisions by institutional means through political action committee, donations to political campaigns, and lobbying (Keim and Zeithaml, 1986). However, in emerging economies with imperfect capital market and loopholes in legal norms, enterprises will choose to acquire resources and opportunities by keeping good relations with the government (Khanna et al., 2005). Chinese enterprises, which have prominent performance in political behaviors, directly establish relations with government officials but not by means of influencing public policies, to acquire preferences in policies and regulations (Zhang and Zhang, 2005). Therefore, the Chinese characteristics should be considered in the empirical research on the promoting effect of government-enterprise relation on enterprise growth. Hence, hypothesis 3 is proposed:  Hypothesis 3: Compared with enterprises without government-enterprise relations, those striving for a good government-enterprise relation can facilitate their growth, and this effect is more significant among small and medium-sized private enterprises.

Sample selection and data sources
The data used in this study is World Bank Survey of Chinese Enterprises, including 25 Chinese cities 2,848 enterprises in 2012, which came from 7 service sectors and 11 manufacturing sectors. Compared with the survey data of World Bank on Chinese enterprises in previous years, this survey data is recent and more novel and unique for measuring the government-enterprise relation. The GDP and urban population size are derived from China City Statistical Yearbook.

Definitions of main variables
Enterprise growth (Growth). Developing potential of enterprise can be an index used to measure enterprise growth. In the existing literature, the measurement indexes for enterprise growth include gross sales level, profit level, and growth rate of employees, among others. As enterprise employees are influenced by the growth of labor productivity and replacement of labor with machine, the measurement of enterprise growth by the number of employees is deficient (Zhang et al., 2016). In many empirical researches, enterprise growth is measured by sales volume and sales growth rate (Chrisman et al., 2005;Zhao et al., 2012), as well as profit growth rate (Zahra et al., 2002). According to the data structure, enterprise growth was measured in this study by selecting the natural logarithm of gross sales in the previous year.
Government-enterprise relation (Relation). In many studies, government-enterprise relation takes into consideration that senior executives of a company take posts or once took posts in government sectors or NPC and may have served as CPPCC representatives (Khwaja and Mian, 2005;Du et al., 2009). However, this index cannot effectively measure government-enterprise interaction, and quantifying the time and money spent by enterprises in maintaining government relations is difficult. In this study, government-enterprise relation was measured using the natural logarithm of the time spent by a company's senior managers in handling government rules and regulations. If the senior managers of an enterprise can spend time in reading and studying the rules and regulations of different departments and interact with the government accordingly, then this enterprise makes effort to establish relations with the government.
Financing plight (Finance). On the basis of the research methods of Drakos and Giannakopoulos (2011) and Gou and Huang (2014), the loan information of enterprises was investigated from two dimensions: loan demand and actual loan acquisition, to discriminate whether an enterprise was faced with any financing plight. If an enterprise applying for loans failed to obtain approval from the credit department, it was faced with a financing plight; otherwise, it was not. Enterprises not applying for any loan had enough cash flow and broad financing channels; hence, they were free of financing plights. By contrast, enterprises that did not apply for any loan due to more mortgages required by the credit department and high interest rates were also faced with such financing plights. In the end, the samples were divided into enterprises facing financing plights and those that are not according to the two loan measurement dimensions. Enterprises without financing plights had stronger financing abilities. This variable was taken as 1; otherwise, it was 0.
Enterprise age (Age). The relation between enterprise age and enterprise growth is unclear due to the differences in the resources owned and the level of creativity between mature

Government-enterprise Relation, Financing Plight and Enterprise Growth: Empirical Evidence from World Bank on Chinese Firms
and young enterprises. In this study, enterprise age was measured by taking the natural logarithm after 2011 was deducted from the year of enterprise registration.
Experience of senior manager (Exper). Enterprise growth is closely related to the operation and management of senior managers. The talents of senior managers, the leading figures of an enterprise, play a significant role in enterprise development. In general, the more the working years of senior managers, the stronger their management abilities, and the better they can boost enterprise growth. In this study, the experience of senior managers was measured by taking the natural logarithm of the number of years they have worked at their posts.
Enterprise export (ExpDum). Enterprises expand their market scope and broaden their marketing channels through exports (Batra et al., 2003;Du and Guo, 2012), which can increase their main business income and facilitate their growth. In this study, this index was measured according to whether the export value of an enterprise was greater than 0; if yes, the variable value was 1, otherwise it was 0.
Industry competition (Competition). A fierce competitive environment will increase the credit needs of enterprises. Under unchanged capital supply, the above circumstance is a disadvantage in enterprise growth. Here, enterprise competition was measured by the quantity of enterprise competitors in the industry. In the questionnaire survey, if the observed value of this variable was greater than 0, then enterprise competition existed, and it was taken as 1; otherwise it was 0.
Per capita gross domestic product (GdpPop). Enterprise growth is not only related to an enterprise's financial status but is also influenced by the economic development level of the city where the enterprise is located. Under a high urban economic development level and a complete financial system, an enterprise can acquire more resources to boost its growth. In this study, the per capita GDP was measured by dividing the urban GDP by the urban population.

Models
The following mediating effect models were set to explore how the government-enterprise relation influenced enterprise growth: random error terms.
In Equation (1), coefficient β can be regarded as the total effect of enterprise growth, coefficient β 2 in Equation (3) represents the direct effect, and coefficient α 3 in Equation (2) represents the mediating effect of the mediating variable.

Descriptive statistics
Table no. 1 presents the descriptive statistics of the main variables. The mean value and median of government-enterprise relation were 0.507 and 0, respectively. Averagely, senior managers of enterprises spend 0.66% of their total time handling government rules and regulations. The median and minimum values of the variable measuring governmententerprise relation were 0, and the data were left-skewed, indicating that most enterprise samples have not established government-enterprise relation. The mean value of Finance was 0.692, whereas the median and maximum values were 1, presenting a right-skewed form. This result showed that most enterprise samples were troubled by a few financing plights; nonetheless, they had good financing abilities. Table no. 2 displays the correlation coefficients of the main variables. Governmententerprise relation was positively correlated with enterprise growth, and the correlation coefficient passed the significance test at the 1% level. This result indicated that enterprises enjoying a good government-enterprise relation had a high growth level. The correlation coefficient between financing ability and enterprise growth was significantly positive at the 1% level, and this result showed that better financing ability facilitated faster enterprise growth. The correlation coefficients of enterprise age, experience of senior managers, enterprise competition, and urban per capita GDP with enterprise growth were positive at least at the 5% significance level.

Regression results
Regression was performed by using the least-squares method according to Equations (1) -(3). The result was corrected by using the robust standard company clustering error to solve the heteroscedasticity problem.  In Column (1) of Table no. 3, the regression coefficient of government-enterprise relation for enterprise growth was significantly positive at the 1% level, proving that enterprises could achieve their growth by establishing a relation with the government. Their business income could be increased by 0.32% if they spent additional 1% of their total time studying government rules and regulations and the policy interaction with the government. Enterprises could acquire government resources to promote their growth by establishing a good government-enterprise relation. Therefore, Hypothesis 3 was verified.
In Column (2), the regression coefficient of government-enterprise relation for financing ability was significantly positive at the 10% level. Thus, a good government-enterprise relation remitted the financing plights of enterprises and improved their financing abilities. Hypothesis 1 was verified. In Column (3), the regression coefficients of governmententerprise relation and financing plight for enterprise growth were significantly positive at the 1% level, and the partial mediating effect of financing ability between governmententerprise relation and enterprise growth was verified.

Grouping test
 Regression of sample division according to enterprise ownership system As different ownership systems exert different effects in enterprise production process and enterprises with different ownership systems establish different relations with the government (Fan et al., 2007), the samples were divided into state-owned enterprises,  Columns (1) to (3) in Table no.4 show that the regression coefficient of governmententerprise relation for enterprise growth was significantly positive at least at the 5% level among private enterprises and foreign-owned enterprises. This result indicated that private enterprises and foreign-owned enterprises could achieve their growth by establishing government-enterprise relations. In Columns (4) to (6), only the regression coefficient of government-enterprise relation established by private enterprises for financing ability was positive at the 10% significance level. Thus, private enterprises could enhance their financing abilities through good government-enterprise relation; hence, Hypothesis 2 was verified. In Column (7)-(9), the regression coefficient of government-enterprise relation established by private enterprises for enterprise growth was significantly positive in Column (8), and so was the financing ability of private enterprises for enterprise growth. Therefore, the partial mediating effect of financing ability of private enterprises between government-enterprise relation and enterprise growth was verified.

 Regression of sample division according to enterprise scale
This section investigates whether government-enterprise relation exerted heterogeneous effects on the growth of enterprises with different scales. As different enterprise scales had different influencing mechanisms and effects on enterprise growth and financing abilities, large enterprises were generally not faced with financing plights, thanks to capital adequacy. By contrast, small and medium-sized enterprises, which were characterized by high risks and weak financing abilities, tended to improve their financing abilities by keeping close contact with the government. On this basis, the samples were divided into large and medium-sized enterprises and small enterprises for the regression. In accordance with the related standard of the National Bureau of Statistics, enterprises with less than 300 employees were defined as small enterprises, and those with over 300 employees were large and medium-sized enterprises. Table no.5 displays the regression results.   Note: ***, **, and * represent the significance levels of 1%, 5%, and 10%, respectively; the estimated coefficient values obtained through the regression are displayed in the above table, and the values in brackets are the standard clustering errors corrected based on the company clustering effect; Constant denotes the constant term, and N is the sample number.
In Column (1)-(2) in Table no.5, the regression coefficient of government-enterprise relation for enterprise growth was significantly positive at the 1% level among small samples. Thus, the growth of small enterprises could be facilitated by establishing a good government-enterprise relation. In Column (3)-(4), only the government-enterprise relation established by large and medium-sized enterprises showed a significant regression coefficient for financing ability. In Column (5)-(6), the regression coefficient of government-enterprise relation and financing ability of small enterprises for enterprise growth was significantly positive at least at the 5% level. Sobel test should be performed for the regression coefficient of the government-enterprise relation of small and mediumsized enterprises and their financing abilities, which was not significant as seen in Column (2), according to Wen (2014)'s method. The Sobel test results satisfied the significance test requirements. The statistical Z value of Sobel test was 2.025 (p=0.0428), indicating that the mediating effect exceeded the 5% significance level and accounted for 24.9% of the total effect. Therefore, the partial mediating role played by the financing ability of small enterprises between government-enterprise relation and enterprise growth was verified. This finding further proved that small enterprises were under high operating risks and narrow financing channels; hence, establishing a stable government-enterprise relation could expand their financing channels, remit their financing difficulties, which can ultimately boost their growth.

Endogeneity and robustness test
 Endogeneity problem and instrumental variables Endogeneity problem should be eliminated to explore the government-enterprise relation.
The benign government-enterprise interaction contributes to enterprise growth, and enterprises with good growth momentum can usually acquire government support. Therefore, the endogeneity problem of reverse cause and effect exists between governmententerprise relation and enterprise growth, indicating that the estimation coefficients obtained through the regression in this study were biased.
To mitigate the endogeneity problem between government-enterprise relation and enterprise growth, the number of times that enterprises had been inspected or visited by officials was taken as the instrumental variable of government-enterprise relation. The value was estimated by using the two-stage least squares method. On the one hand, that an enterprise was visited by government officials could influence government-enterprise relation. Enterprises could construct a good communication channel with the government through an inspection (Liu, 2004). On the other hand, that an enterprise was visited by government officials was an internal decision made by the government. The number of official visits was relatively independent of enterprise growth, and government officials made autonomous decisions to visit target enterprises at their discretion (Pan et al., 2009). Therefore, the number of times that enterprises had been inspected or visited by officials was used as an instrumental variable. The influence of political connection on enterprises could be measured by activities of government officials, namely, inspection and visit, which were occasional events, thus remitting the endogeneity problem between government-enterprise relation and enterprise growth. On this basis, a two-stage regression model was constructed as follows: Stage 1:  0.264 −0.038 Note: ***, **, and * represent the significance levels of 1%, 5%, and 10%, respectively; the estimated coefficient values obtained through the regression are displayed in the above table, and the values in brackets are the standard clustering errors corrected based on the company clustering effect; Constant denotes the constant term, and N is the sample number.

Government-enterprise Relation, Financing Plight and Enterprise Growth: Empirical Evidence from World Bank on Chinese Firms
Column (1) of Table no.6 presents the first-stage regression results. The coefficient of the number of times that enterprises had been inspected or visited by government officials was positive at the 1% significance level. This finding indicates that the number of times that enterprises had been inspected or visited by government officials had a significant positive correlation with government-enterprise relation. Moreover, the t value of this instrumental variable was 4.833, thus passing the test of weak instrumental variable. Column (2) reports the second-stage regression results. The explained variable was enterprise growth, and the coefficient of government-enterprise relation was significantly positive at the 1% level, indicating that after the endogeneity problem was controlled, enterprise growth could be driven by good government-enterprise relation. In addition, the conclusions achieved from the basic regression were not perplexed by the endogeneity problem.

 Robustness test
In this study, the government-enterprise relation was better measured from two dimensions: enterprise acquisition of government contract and informal payment. The governmententerprise relation variable in the basic regression was replaced by establishment of government-enterprise relation (RelationDummy), whether the enterprise had any guarantee or attempted to acquire a government contract in the previous year (ContractDummy), the total amount of informal payment (InformalPay), and whether the presents or informal payment were required in the contact between enterprise and government official (OfficerInformalPayDummy), in an effort to perform the robustness test.  Table no. 7 reveals that after government-enterprise relation was substituted, the four substitutive indexes were positive at least at the 5% significance level, indicating that the results remained robust after the two-dimension indexesenterprise acquisition of government contract and informal paymentwere used to replace government-enterprise relation in the basic regression.

Discussion
The above analysis results show that Hypothesis 1, Hypothesis 2, and Hypothesis 3 are supported. According to the research findings, government-enterprise relation mitigates the financing constraints and difficulties faced by enterprises, thus promoting enterprise growth. Therefore, enterprises are motivated to establish close relations with the government and spend time reading and studying the department rules and regulations and interacting with the government. Given this finding, the government can drive enterprise development, give them preferential policies of a certain proportion such as subsidies, and promote enterprise growth. The grouping test shows that the financing plights of private

Government-enterprise Relation, Financing Plight and Enterprise Growth:
Empirical Evidence from World Bank on Chinese Firms enterprises and small enterprises can be remitted by establishing good governmententerprise relation. Enterprise growth can be further enhanced, specifically as follows: First, the regression coefficient of government-enterprise relation for enterprise growth is positive at the 1% significance level as seen in Table no.3, proving that enterprises can achieve their growth by establishing a connection with the government. Moreover, the regression coefficient of government-enterprise relation for enterprise financing ability is significantly positive at the 10% level, showing that a benign government-enterprise relation mitigates the financing plights of enterprises and strengthens their financing abilities. This finding accorded with the viewpoint of Faccio et al. (2006) and Yu and Pan (2008): enterprises can gain direct access to government fiscal subsidies and tax preferences by keeping contact with the government, and acquire more bank loans by virtue of government-endorsed letters of credit. In the end, the regression coefficients of government-enterprise relation and financing plight for enterprise growth are positive at the 1% significance level. Therefore, the partial mediating effect of enterprise financing ability between government-enterprise relation and enterprise growth is verified, which supports the lubricating effect hypothesis from the theoretical level. This finding coincides with the viewpoint of Boubakri et al. (2012) and Liu et al. (2013). Government-enterprise relation could help enterprises obtain more credit resources or government subsidies (Johnson and Milton, 2003;Claessens et al., 2008). Moreover, good government-enterprise interaction could facilitate lowering financing costs necessary for enterprise growth (Boubakri et al., 2012;Liu et al., 2013). Among the main control variables, the regression coefficient of enterprise age and experience of senior managers was positive at the 1% significance level, indicating that older enterprise age and richer experience of senior managers were better for enterprise growth. A possible reason is that enterprises founded earlier occupy more social resources, along with mature management mode and entrepreneurship, which can improve enterprise performance. The regression coefficient of enterprise export was also positive at the 1% significance level, indicating that the increase in enterprise export value expanded the market scope, increased the sales channels, and further boosted enterprise growth (Batra et al., 2003;Du and Guo, 2012).
Second, the regression coefficient of government-enterprise relation for enterprise growth is significantly positive at the 5% level among private enterprises and foreign-owned enterprises as shown in Table no.4. This result demonstrates that the growth of these enterprises can be promoted by establishing government-enterprise relation. The regression coefficient of government-enterprise relation established by private enterprises is positive at the 10% significance level, showing that private enterprises enhance their financing abilities through a benign government-enterprise relation. The regression coefficient of financing ability of private enterprises for enterprise growth is significantly positive at the 1% level; hence, the mediating effect of the financing ability of private enterprises between government-enterprise relation and enterprise growth is validated. Relative to state-owned enterprises, private enterprises may bear more financing limitations; hence, they seek to establish a stable relation with the government to expand their financing channels (Yu et al., 2012) and to relieve their financing difficulties owing to small scale, few mortgages, and high risks, so as to boost their growth.
Third, the regression coefficient of government-enterprise relation for enterprise growth among small enterprises is significantly positive at the 1% level as shown in Table no.5. Therefore, small enterprises can achieve vigorous growth by establishing good government-enterprise relation. The regression coefficients of government-enterprise relation and financing ability for enterprise growth are significantly positive at the 5% level, and the Sobel test results satisfy the significance test requirements. These results prove that the mediating effect passes the significance test at the 5% level. At the same time, referring to Wondirad (2020), this study also recommends that deepening financial outreach to the small enterprises, countries need to initiate and encouraging competition in their microfinance market. Along with this, policy regulators also need to design a policy framework that creates a sound competitive microfinance market.
Fourth, the number of times that enterprises had been inspected or visited by government officials was selected as the instrumental variable of government-enterprise relation. The first-stage regression results show that the coefficient of the number of times that enterprises had been inspected or visited is positive at the 1% significance level. In the second-stage regression results, the coefficient of government-enterprise relation is significantly positive at the 1% level. This result indicates that after the endogeneity problem is controlled, enterprise growth can be facilitated by establishing good government-enterprise relation. Moreover, the basic regression conclusions are perplexed slightly by the endogeneity problem. Table no.7 shows the robustness test after the substitution of the government-enterprise relation variable. The result indicates that government-enterprise relation exerts a significant positive influence (over 5%) on enterprise growth.

Conclusions
Using the World Bank Survey of Chinese Enterprises data in 2012, mediating effect models were utilized to empirically test how government-enterprise relation influenced enterprise growth. The following conclusions were drawn: (1) Good government-enterprise relation positively influences enterprise growth.
(2) Government-enterprise relation can facilitate enterprise growth by mitigating enterprises' financing difficulties. This conclusion remains significant after the endogeneity problem is weakened by the instrumental variable method and the robustness test is carried out.
(3) In the sample-specific research, private enterprises, foreign-owned enterprises, and small enterprises seek for their growth by establishing a connection with the government; a good government-enterprise relation can relieve them from financing plights and boosted their growth.
This study provides the following managerial implication: (1) this study reports that political ties have a positive effect on enterprise growth, whereas business ties have a positive effect on enterprise growth. This means that when an SME can obtain policy and financial support from the government, it should pursue the enterprise growth strategy.
(2) This study indicates if managers of SMEs want to effectively utilise good managerial ties to improve their growth, the role of social network and political ties cannot be ignored.
(3) This study shows that governments in China play a very important role in the implementation of growth strategies by SMEs. it is necessary for SMEs to focus on the national key development direction (such as digital economy, blockchain and artificial intelligence) when making enterprise growth plan.
Although this study reveals the influencing mechanism of government-enterprise relation on enterprise growth to some extent, some problems remain to be solved. (1) World Bank Survey of Chinese Enterprises data is used to unfold this study. However, the institutional background and industry structure in China are significantly different from that in Western countries. Therefore, whether the conclusions apply to the market environment in developed countries and regions needs to be tested.
(2) the government-enterprise relation is measured by using the time spent by senior managers in handling government rules and regulations, and the robustness test of government-enterprise relation is implemented by using the following variables: whether enterprises establish the government-enterprise relation, whether they are guaranteed or attempt to acquire government contract, the total amount of informal payment, and whether presents or informal payment are required in the communication with the government. Nevertheless, the government-enterprise relation can be measured from diversified dimensions. It can be comprehensively analyzed from the perspectives of social network, corporate shareholders, and management layer. In the follow-up research, the government-enterprise relation can be profoundly depicted, and its influences on enterprises may be further discussed.