Do Political Connections Affect Corporate Performance? Evidence from Listed Real Estate Companies in China

정치적인 배경이 기업 실적에 영향을 주는가? 중국의 상장된 부동산 회사들을 중심으로

Abstract

Political connection are widespread in many countries and are a hot topic in economic research right now. In China, companies are actively forming political connections in several ways. Based on the research model and theoretical analysis, this thesis puts forward the research hypothesis and constructs the research model of political connections and corporate performance using the financial data and corporate governance data of China A-share listed real estate companies in 2010-2014. The thesis mainly analyzes the political association of CEO. For the analysis result, this research uses the univariate regression analysis and multivariate analysis to carry out the robustness test. The empirical study analysis includes three parts: firstly, in Chinese listed real estate enterprises, CEO political connection has a significantly negative correlation with corporate performance. It means real estate enterprises whose CEO has political connections gain a lower performance than other enterprises; Secondly, from the personal characteristics of the CEO, age, sex and education level have positive relationship with the performance of the enterprise, the additional post situation negatively related to the firm's performance; Finally, the firm's scale liquidity ratio and capital asset rario have a negative relationship with corporate performance, and the number of top managers is positively related to corporate performance. This research made a study on the political connection in Chinese real estate industry, which could also provide beneficial references for the development of enterprises in this industry.

많은 나라에 정치적 연결이 널리 퍼져 있으며 현재 경제연구에서 화두가 되고 있다. 중국에서는 기업들이 여러 가지 방법으로 정치적 연결을 활발히 형성하고 있다. 본 논문은 연구모델과 이론분석을 바탕으로 2010~2014년 중국 A주 상장부동산의 재무자료와 기업지배구조 자료를 활용해 정치연계 및 기업성과의 연구 모델을 제시하고 구성한다. 논문은 주로 CEO의 정치 연관성을 분석한다. 이 연구는 분석 결과에 대해 일변량 회귀 분석과 다변량 분석을 사용하여 강건성 테스트를 수행한다. 실증적 연구 분석은 세 부분으로 구성되어 있다. 첫째, 중국 상장 부동산 기업의 경우 CEO 정치 연계는 기업 실적과 상당히 부정적인 상관관계를 가지고 있다. 이는 CEO가 정치적 연줄이 있는 부동산 기업이 다른 기업보다 낮은 성과를 얻는다는 것을 의미한다. 둘째, CEO의 개인적 특성에서 연령, 성별, 교육수준은 기업의 성과와 긍정적인 관계를 가지며, 추가적인 이후 상황은 기업의 성과와 부정적으로 관련되어 있다는 것을 의미한다. 마지막으로 기업 특성은 기업의 성과와 부정적인 관계를 가지며, 최고 관리자 수는 기업의 성과와 긍정적으로 관련되어 있다. 이 연구는 중국 부동산 산업의 정치적 연관성에 대한 연구를 수행했으며, 기업의 발전에 유익한 참고 자료를 제공할 수 있다.

I. Introduction

In Chinese traditional culture, the political connection is an integral part of “relation” culture, which can be traced back to ancient times. For this reason, in China, the political connection has become one of the most important and common factors, which should be considered during the processes of decision-making and governance within an enterprise, and it is also received extensive attention of domestic and foreign scholars in recent years. In the real estate industry, the political connection has always been reflected in some key steps, such as acquiring the usage right of state-owned land, construction approval and applying for the presale permission during the project development process. In addition, as a developing country, China’s current economic system is not perfect enough, in which there are no definite rules related to deal with the government-relation issues. In this case, some real estate enterprises will attempt to acquire the usage right of state-owned land and bank loan etc.

By establishing political connections for themselves, they can get interests and facilitate development process. And there are some main ways to build political connections, through which they are trying to make the president or general manager of the company to be selected as a deputy to the NPC(National People's Congress) or CPPCC(Chinse People's Political Consultative Conference) member, employing retired government officials and maintaining close contact with the officials already in government etc[1]. But it remains controversial about whether the business performance can be improved through establishing the political connection in the real estate enterprises.

In theoretical field, there are two opposite opinions about the impacts on companies’ business performance caused by the political connections. The first one was put forward based on the Efficiency Theory. People holding this opinion believe that political connection is a kind of important social capital in one company, which can increase its capacity of acquiring and utilizing resources to achieve the improvement on the value and the business performance of that company. The second one is rent-seeking theory. They that enterprises must pay the government officials huge rent-seeking expenditures, which do not match the earning. For example, it does not only impact the social distribution and cause the loss of social welfare, but also causes the huge waste of the enterprise’s resources in the process of corporate’s rent-seeking or solving the problem of employment for government.

To sum up, against the background that the economic system is not perfect in China, different conclusions might be drawn according to different theories. On the one hand, by considering the Efficiency Theory, the political connections may elevate the enterprise’s business performance in Chinese real estate industry. On the other hand, by considering the rent-seeking theory, the political connections may impede the development of the enterprise.

Thus, the political connection is highly related to the business performance of an enterprise. In researching the impact of political connections, most of the existing studies may put the emphasis on the financial or the economic system background but seldom refer to the role played by the corporate governance in the government-relation and the influence of senior executives on the corporate performance. This article is aimed at revealing the impacts of the CEOs on the political connections-business performance in the present Chinese real estate industry. Based on the empirical analysis of some Chinese real estate enterprises, this article will make a further study on the political connection in Chinese real estate industry, which could also provide some beneficial references for the developments of enterprises in this industry.

Ⅱ. Literature Review

1. Concept of the Political Connections

Faccio (2006) pointed out that the political connections might universally exist within many countries[2]. Real estate companies are more keen to establish political relations with the government, especially in those countries, which have imperfect economic systems[3]. A politically related company is a company whose actual controller, major shareholder, or senior management has worked or now works in a government department, or a company that has a close relationship with an individual who has political power. Some scholars regard the circumstances as a political connection where the CEOs (CEOs) or general managers or independent directors and other executives have worked or now work in military or government departments, have been or now are NPC deputies. Faccio (2006) argues that if the company's major shareholders or executives (including the CEO, chairman, vice chairman, board member) now work or have been working in a government or parliament, or have a close relationship with the government, the company is a political association enterprise[2]. Fan, Wong and Zhang (2007) especially did some researches in Chinese companies and then draw a conclusion that if the company's CEO now works or has worked in the central government, local government or army, the company could be a political association[4].

Since CEO is the highest executive in China, this paper will use the government background of key managers (CEOs) to measure the company's political relevance.

2. Differences between the Political Connections and corruption

Unlike corruption and bribery, political connections are legal in law. As Faccio (2006) point out, political connections are perfectly legal at the legal level, and it also has the advantage that firms can influence government departments through political connections, which other firms do not have[2].

3. Differences between the Political Connections and Political Interference

Political connections and political interference are different and their action mechanisms are also completely different. Shleifer and Vishny(1994) found that The initiator of political intervention is the government who may influence the enterprise through various ways to achieve public and social goals or official private interests[5]. On the contrary, the initiator of the political connection is the enterprise who intends to establish political connections with the government for seeking a certain interest or other purposes and then influence the government and the market.

4. Impacts of the Political Connections on Enterprise Performance

On the impact of political relevance on corporate performance, domestic and foreign scholars have done a lot of research, but did not reach a concerted conclusion. Thus, there are three conclusions about the impact of political connections on business performance:

Johnson and Mitton (2003) found that during the financial crisis, in which the government took control of the economy, the capital markets of these corporates who had political connections with the president performed significantly better than those without political relations. Besides, nearly 30% of the increase on the market of corporates with the political connections happened from the political connections[6]. Ma and Zhu (2016) studied the relationship between political connections and company performance from the perspective of social networks, and found that social networks will not affect the ability of political connections to help companies obtain more competitiveresources in China.[7].Li Jinglong (2014) also pointed out that as the market mechanism is imperfect, the government occupies a relatively important position in economy. If a company has strong political connection, it can obtain development support more efficiently[8]. The research of Deng Xinming (2016) found that political background can help companies enter into barriered markets easily, which is conducive to the improvement of corporate performance [9]. Shi Linlin (2018) found that political connections do not directly affect corporate performance, but indirectly affect corporate performance by affecting corporate capital structure. Companies with stronger political connections can more easily obtain financial support, which can optimize corporate performance. [10]

The enterprise political connections also play against the development of enterprise business performance. Faccio (2006) pointed out that although the political connections could bring some scarce resources, enterprises may have to pay lots of costs on rent-seeking during seeking the political connections[3]. Benjamin (2016), from the perspective of shareholders' interests, found that corporate political connections would inhibit the amount of dividends paid by companies[11]. Zhang Tianshu (2015) conducted that companies with more political connections would attract more venture capital into the company, which will affect company performance[12]. Xu Dan (2017) found that politically connected directors get higher salaries, but they cannot effectively improve the company's performance[13].

The enterprise political connections are little correlated with the development of enterprise business performance. Fisman (2006) visited the corporate, which had political connections with Richard, the current Vice President of the United States, he found that there was no significant correlation between the corporate value measured in share value and the political connections[14]. Coincidentally, Du(2009) found that the political connections of independent directors in Chinese listed private enterprises have no significant impact on the enterprise business performance[15]. Wang Zeming(2018) conducted a research on private enterprises in manufacturing industry and found that political connections have no significant relationship with corporate performance[16].

Ⅲ. Hypothesis & Data & Method

1. Hypothesis

In the real estate industry, the political connection has always been reflected in some key steps, such as acquiring the usage right of state-owned land, construction approval and applying the presale permission during the project development process. The political connection, which may spend a lot of time and money to maintain, could bring a huge interest to the entrepreneur. Faccio (2006) said that the politically connected enterprise’s executives, who have political background, might not be competent entrepreneurs[2]. From this aspect, enterprises may pay a lot of costs for their political rent-seeking activities, which could counteract the benefits bring by the political connections.

For example. in some state-owned enterprises in the Chinese real estate industry, their CEO’s are officials delegated by the government, whose initial work is letting the enterprises to meet the local government’s (SASAC’s) request， but not to achieve the maximization of economic benefit. In addition, because of the particularity of the state-owned enterprises, it is much more necessary for a state-owned enterprise to maintain the stability of market and also offer employment，social public services and products. In this paper, as the CEO is the top executive in Chinese enterprise, the CEO’s political connections will be researched next for pointing out the impact of the political connections on the enterprise business performance. For this reason, the hypothesis was come forward:

The CEO’s political connections are correlated with the development of enterprise business performance.

2. Data

In this paper, 147 listed real estate companies in A-share market were selected as initial research samples. According to the research practices, the samples were screened through the following standards: (1) eliminate the 21 companies who were dealt with by the CSRC(China Securities Regulatory Commission) using ST(Special Treatment), PT(Particular transfer), GST(Special Treatment for Reforming Stocks), *ST(Special Treatment Warning), or G*ST(Special Treatment Warning for Reforming Stocks), from 2010 to 2014. As there would be some serious deviations happened in the research including these companies, whose business performance is too bad to represent the average level of the industry. (2) eliminate the eight companies, whose data were missing during the period from Jan 1st, 2010 to Dec 31st. According to the above two standards, combining the CSMAR database and the financial data from the annual reports of the listed corporates, 118 available samples were obtained in this paper. Most real estate listed companies that issue A shares are engaged in land developing, real estate selling and renting. There are 46 state-owned companies and 72 private companies among 118real estate listed companies.

The main source of research data is CSMAR database.

3. Model Specification and Selection of Variables

According to the previous classic literatures that analysis the factors affecting corporate performance, we conduct that profitability, debt, and asset characteristics can affect corporate performance. The purpose of this article is to investigate the impact of CEO political connections on Chinese listed real estate companies, so on this basis, the CEO's personal characteristics and political connections are taken into consideration. So this paper constructs a linear regression model which shown in Model 3.1（i means i company, t means in t year)

\(\begin{aligned} R O A=& \alpha+\beta_{1} P C_{i, t}+\beta_{2} g e n d e r_{i, t}+\beta_{3} a g e_{i, t}+\\ & \beta_{4} \text { degree }_{i, t}+\beta_{5} \text { both }_{i, t}+\beta_{6} \text { number }_{i, t}+\\ & \beta_{7} \text { share }_{i, t}+\beta_{8} \text { finance }_{i, t}+\beta_{9} L e v_{i, t}+\\ & \beta_{10} T A G R_{i, t}+\beta_{11} \text { Ln } \text { size }_{i, t} \end{aligned}`\)       (1)

Dependent variable: Corporate performance is measured by using ROA index.

Return on assets (ROA)= Net profit/ Total balance of assets;

Total balance of assets= (Total Balance of assets at the end of the period + Total Balance of assets at the end of the same period of last year) /2.

Independent variable: Key variable is political connections (PC). As the CEO system was introduced to China relatively late, most of the listed companies have not set up the position of CEO. But considering the role and status of the CEO at the company’s management and implementation level, CEO refers to the general in Model 3.3.1, to examine the impact of CEO political relations on listed real estate manager of one company. The limitation of Political connection: the CEO is the NPC deputies or CPPCC members of the city, province and nation, or the person who has or had the government working background. In this paper, dummy variable method is adopted that “the CEO has political connections' value of 1, otherwise 0. CEO gender, age, number of top executives, current ratio, asset-liability ratio, firm size, etc. The definition of the relevant variables and the calculation methods are shown in [Table 1].

Table 1. VARIABLE

Ⅳ. Empirical Results

1. Data description

[Table 2] shows the descriptive statistics of the corporate performance (ROA), political Connections (PC), CEO gender, age, CEO degree, both, the number of top managers, shareholding ratio of first company shareholder, liquidity ratio, capital asset ratio (Lev), total assets growth rate (TAGR) and company size (LNsize). The analysis mainly includes the data of these indexes, which are the observations, means, standard deviation, minimum and maximum values of the variables.

Table 2. DESCRIPTIVE STATISTICS OF VARIABLES

From [Table 2], it can be seen that the corporate performance’s average value is 0.0429, standard deviation is 0.0528, the minimum value is 0 and the maximum value is 0.79, which indicates that the Chinese listed company performances fluctuate so much; CEO political connections’ average value is 0.2220, indicating that the CEOs in 22% of corporates have political connections; the gender, age, education degree and both of the CEO have the respective averages of 0.8729, 47.4881, 2.5610 and 0.1288, indicating that the CEOs of the sample company are mostly male at the age of 48, who generally have bachelor degrees, and mostly take both the role of chairman and general manager; the number of top managers and the shareholding ratio of first company shareholder respectively have the average of 6.3170 and 39.9646, the standard deviations of 2.3373 and 18.3234, the minimum values of 2 and 7.12, and the maximum values of 18 and 89.41, indicating that the number of top managers in the sample companies and the shareholding ratio of first company shareholder are relatively high but also perform differently in different companies; the liquidity ratio’s average value is 2.0859, the standard deviation is 1.7435, the minimum is 0.03 and the maximum is 29.713, indicating that the sample company corporates’ current liabilities are relatively high; the capital asset ratio (Lev) average is 0.64 also reflecting that the sample companies have large current liabilities. Considering the above descriptive statistics, this paper argues that political connections do affect the company capital asset ratio and performance level.

2. Univariate Correlation Analysis

After the descriptive statistical analysis of each variable, to avoid that the model test results are influenced by serious collinearity between the variables, this paper further analyzes the correlation coefficient between the various research variables of the sample companies.

As can be seen from [Table 3], the correlation coefficient between firm performance and political connections is –0.06. Before controlling other variables, this result is consistent with the hypothesis that the corporate performance is negatively correlated with political connections; the correlation coefficient of capital asset ratio is 0.01, between which there is a significant positive correlation, proving to some extent that the enterprise can improve the level of corporate debt when political connection exists; the capital asset ratio is negatively correlated with the corporate performance, which correlation coefficient is –0.05; the capital asset ratio is also negatively correlated with political connections. Political connections are positively related to monopolies, with a correlation coefficient of 0.37, which is significant under the 10% level.

Table 3. UNIVARIATE CORRELATION ANALYSIS

Note: The lower triangle of the table is the Pearson correlation coefficient.

In addition, the political connection is negatively correlated with the liquidity ratio, with a correlation coefficient of -0.04, which shows that the lower the liquidity ratio of the firm, the easier it is for the company to build a strong political connection with the government. The correlation coefficients in the table are all under 0. 5, indicating that if all the variables are putting into the same model, serious multiple collinearity problems would not happen among the variables.

3. Multivariate Regression Analysis

While dealing with the selected panel data, firstly Hausman test is used for determining whether a fixed effect or a random effect model should be used. as the Hausman test shows that the panel data is at the 5% significance level, which is against the hypothesis that random effect model is better. Therefore, the fixed effect model is used for regression, the results of which showed in [Table 4].

Table 4. THE REGRESSIN RESULTS OF FIXED EFFECT MODEL

Note: ***, **, * indicate the index is at 1%, 5% or 10% significance level

As can be seen from [Table 4], the dependent variable of the fixed effect model is the corporate performance (ROA). The regression coefficient of the ROI and the political connections(PC) are negative at the 5% significance level, indicating that the CEO’s political connection caused a significant reduction on the corporate performance; The CEO’s gender is positively but not significant correlated with the corporate performance (ROA); the age of the CEO is positively related to business performance (ROA) significantly at 5% level; the number of top managers in the firm is positively correlated with the firm's performance (ROA) significantly at the 10% level; the shareholding ratio of first company shareholder and the total asset growth rate (TAGR) are positively correlated with firm performance (ROA) significantly at 1% level; Lnsize is negatively correlated with firm performance (ROA) significantly at 1% level; both, the liquidity ratio and the capital asset ratio (Lev) are also negatively but not significantly correlated with firm performance (ROA).

In particular, The regression coefficient of business performance (ROA) and CEO political association (PC) is –0.0196, which means CEO with political background will decrease the corporate's performance. The values of t and p are respective -2.18 and 0.03, which are significant at the level of 5%. the age of the CEO has a positive effect on the corporate performance and the influence coefficientis 0.0014, of which T value is 2.34 and P value is 0.020, also significant at 5% level; the correlation of the executive number and corporate performance (ROA) is also positive, with a t value of 1.71 and a p value of 0.089, which are at 10% significance level; With the increasing of share and TAGR, the corporate performance would also increase; While the company size (Lnsize) has a negative influence on the corporate performance. In addition, the regression results of other variables are not significant. The gender and the education level of CEO are positively correlated with the corporate performance (ROA), and the regression coefficients separately are 0.0033 and 0.0052 but not significant; both, the liquidity ratio and the capital asset ratio (Lev) are also negatively correlated with the corporate performance (ROA), with regression coefficients of –0.0047, -0.0013 and -0.0122, respectively. Thus, the regression model can be put forward(i means i company, t means in t year):

\(\begin{aligned} R O A=& 0.2108-0.0196 P C_{i, t}+0.0033 \text { gender }_{i, t}+\\ & 0.0014 a g e_{i, t}+0.0052 \text { degree }_{i, t}-\\ & 0.0047 \text { both }_{i, t}+0.0029 \text { number }_{i, t}+\\ & 0.0017 \text { share }_{i, t}-0.0013 \text { finance }_{i, t}-\\ & 0.0122 \operatorname{Lev}_{i, t}+0.0923 \mathrm{TAGE}_{i, t}-\\ & \text {0.0147Lnsize }_{i, t} \end{aligned}\)       (2)

4. Robustness Test

To further test the relationship between CEO political connections and corporate performance, the personal character variables of the CEO and affect variables of corporate are used as control variables to gradually return. In [Table 5], the dependent variables from model (1) to model (5) are also corporate performance (ROA). The regression coefficients of corporate performance (ROA) and political connections (PC) in five models are all negative, and all pass the test at 1%~5% significance level, indicating that the political connections between the CEO more significantly reduce the performance of enterprises.

Table 5. ROBUSTNESS TEST

Note: Words in blankets are t values, ***, **, * indicates the indexes are at 1%, 5% or 10% respectively

For example, among the sample corporates, the corporate performance of these companies with political connections reduced approximately 0.02 percent. From model(1) to model (5), a more consistent regression result is obtained that the corporate performance (ROA) is negatively correlated with the CEO's political connections (PC), of which the regression coefficients are all at 1%~5% significance level; Gender and degree are positively but not significant correlated with corporate performance (ROA); age and ROA are positively related at the 5% significance level; the number of top managers in the firm is positively correlated with ROA at the 10% significance level; the shareholding ratio of the first company shareholder and the total assets growth ratio (TAGR) are positively related to corporate performance (ROA) at 1% significance level; both, the ratio of flow and asset-liability (Lev) are also negatively but not significantly correlated with corporate performance (ROA).

According to the above regression results, we remove the CEO personal characteristic variables that have positive and negative effects to create model (1) and model (2), and then remove all the CEO personal characteristic variables to create model (3). Moreover, We exclude the number and share variables in the enterprise characteristic variables to create model (4) and remove finance variables create model (5).

In the model (1), the variable of gender, age, degrees are removed from the personal characteristic variables in the control variables to return. The regression results show that the regression coefficient between corporate performance (ROA) and CEO political connections (PC) is -0.0206, which means the political background of CEO would go against to corporate performance; the number of top managers is also positively correlated with the corporate performance at 10% significance level; and the increases of the shareholding ratio of the first company shareholder (share) and the total assets growth ratio (TAGR) would cause the increasing of the corporate performance and are at 1% significance level; while with the increase of LNsize, the corporate performance would decrease; both, finance and Lev are also negatively correlation with corporate performance (ROA) correlation, of which regression coefficients were -0.0031, -0.0015 and -0.0109 but not significant. In the model (2), model (3), model (4) and model (5), the regression analysis shows that the regression coefficients of corporate performance (ROA) and CEO political connections (PC) are respectively 0.0208, -0.0215, -0.0186 and -0.0175, and are all significant at the 5% level; the Lev is also positively correlated to the corporate performance (ROA) of respectively -0.0124, -0.0110, -0.0116 and .0.0105 influent coefficient, which are not significant; the total assets growth ratio is also positively correlated to the corporate performance (ROA) of respectively 0.0922, 0.0917, 0.0926 and 0.0920 influent coefficient, which are all at 1% significance level; in addition, the negative correlation coefficients between the CEO's covariance (both) and the corporate performance (ROA) from the model (1), model (4), model (5) are respectively -0.0031, -0.0027 and -0.0038, which are not significant; in model (1), model (2), model (3), and model (5), the number of top managers (number) is positively related to the corporate performance (ROA), and per person increasing on t performance (ROA) from the model (1), model (4), model (5) are respectively -0.0031, -0.0027 and -0.0038, which are not significant; in model (1), model (2), model (3), and model (5), the number of top managers (number) is positively related to the corporate performance (ROA), and per person increasing on the total number of executives may cause 0.0029, 0.0028, 0.0028 and 0.003 percentage increasing on the corporate performance at 10% significance level; from model (1) to model (4), the liquidity ratio (finance) and the corporate performance (ROA) are also negatively correlated. The regression coefficients are respective -0.0015, -0.0012, -0.0015 and -0.0005, which are also not significant.

It can be seen from the above analysis that comparing the regression results of model (1) to model (5) with the original model, it can be found that the variables’ regression coefficients and symbols in the five models are basically the same as those of original model, the difference between the regression coefficients of each variable in each model is small, and the significance levels are also consistent. The results of this paper have not changed, which indicates that the model setting and the conclusion are all robust.

Ⅴ. Conclusions and Prospects

1. Conclusions

By empirical analysis, we can draw conclusions as follows.

First, after controlling for the relevant characteristic variables of the enterprise or the CEO, it is found that the higher the degree of political connection(PC) of the enterprise or CEO, the worse the corporate performance (ROA). The result shows that the listed real estate companies pay more cost than profits to establish political connections. For the results, the explanations are as follow: Firstly, although political connections can bring more resources for companies, at the same time huge cost may be paid during getting resources. Moreover, as the CEO of an enterprise is mainly responsible for the company's daily operation and management activities, so the general manager needs more management experience and professional skills, otherwise it will easily cause the company's operation and management in low efficiency and reduce company performance.

In addition, from the personal characteristics of the CEO, age, sex and education level have positive relationship with the performance of the enterprise, the higher the age and education level of the CEO, the higher the corporation performance, enterprise also can get a higher corporate performance when the CEO is male. The additional post situation is negatively related to the firm's performance. The corporate performance of firms which CEO as chairman and general manager are worse than CEO only for chairman. Finally, from the enterprise characteristics, the number of top managers and the ratio of first company shareholder are positively related to firm performance, and the firm's scale, liquidity ratio and capital asset ratio are negatively correlated with firm performance.

2. Prospects

In this article, there are some insufficiencies in two aspects: First, the measurement of political relevance: the assignment method is used to measure the degree of political connections of real estate enterprises that whether the CEOs have ever served as government officials, whether the CEOs are former or the current NPC deputies or CPPCC members. From this aspect, this research is comparatively scientific. But in this article the two differences between "former" and "incumbent" were not distinguished; more importantly, limited to the difficulty of data collection, this article did not consider the CEO's relatives, friends, student relations and comradeship to the political connection. It is the common problem of existing research. Second, endogenous problems: the relationship between the political connections of CEO and the corporate performance is largely endogenous to the institutional environment of the company. In this paper, it has been considered to use the market-oriented index to control for the institutional environment, but it is difficult to obtain the nationwide business characteristics of these real estate companies listed at the same time so that the institutional environment and the financial environment have not been controlled, which might weaken my conclusion.

Based on the above research, the outlook is put forward as following: First, the definition of executives: there is no uniform standard of top managers in the existing research, which might lead to the unanimous conclusion on the influence of corporate political connections. In this paper, the former research results are used to study the political connections of real estate enterprises from the aspect of key top managers. But the definition of top managers should also be further discussed. And from the aspect of empirical aspects, the differences between different types of top managers can be distinguished by comparing the different impacts caused by different top managers in the system of "political relevance - corporate performance”. Second, the selection of control variables: to further discover the relationship between the political association of the real estate enterprise and the performance of the company, and also figure out its mechanism of action, the intermediate variables should be further explored. In the existing research, it is always analyzed from the aspect of the nature of the enterprise's ultimate controller and the institutional environment. However, it is also slightly inadequate to do research based on the characteristics of executives, internal governance and other endogenous factors of research. Therefore, the follow-up research can be further done from the aspect of endogenous variables and exogenous variables.