The Impact of Bank Loan on Executive Perquisites under the Background of Post-Split-Share Reform

Executive Perquisites are agency cost and harm the interests of shareholders. After the split share reform, the first largest shareholder has decreasing his shares and the problem of executives’ expropriation has been more prominent. This paper uses the data of Chinese listed firms during 2006 to 2017 and investigates the corporate governance effect of bank loan on executive perquisites. The results show that bank loan can constraint the executive perquisites which have corporate governance effects; the corporate governance effect of bank loan has more prominent in Non-SOEs than in SOEs; and the corporate governance effect of bank loan has more prominent in the listed firm with higher proportion of share of the first largest shareholder. So the conclusions provide empirical evidence not only for the discussion of commercial bank reform but also for the corporate governance reform of Chinese listed firms under the financial systems of main bank financing.

between shareholders and managers. Therefore, this paper selects managers' executive perquisites as the research object, discusses the first type of principal-agent problem between traditional shareholders and managers, and examines the corporate governance effects of bank loan on such principal-agent problems.
The executive perquisites are the monetary benefit of managers in the performance of their duties, including the enjoyment of luxury offices, possession of special planes or special vehicles, and catering and entertainment consumption.
It harms the interests of shareholders and constitutes the first type of agency cost for managers. Focusing on how to reduce this type of principal-agent problem, domestic and foreign scholars carried out research [1]- [6]. However, these research results are mainly based on the developed markets of the United Kingdom and the United States. The research on the executive perquisites of managers in China is mainly based on the experience of developed markets from the perspective of ownership structure and manager incentives. Considering China's special financial system with bank financing as the main body, this paper chooses bank loan as the corporate governance effect of creditor's rights, and studies its impact on managers' executive perquisites.
As a professional financial institution, banks have the advantage of solving the information asymmetry and moral hazard. They supervise the funds and management of the borrowing enterprises before, during and after the borrowing. However, the State-owned attribute of Chinese banks, through the market-oriented reform such as listing, whether it can give play to the corporate governance effect, there is a great deal of controversy. Cerqueiro et al. [7] and Lin et al. [8] use Swedish and US data to find that banks have a supervisory role on borrowing companies. Bailey et al. [9], Qian and Yeung [10] used data from Chinese listed companies and found that China's bank loan did not play a supervisory role for listed companies, even lowering the level of supervision of the entire financial market, including the equity market . However, the authors found that these research papers were mainly used during the data sample period, 1999-2004, 1995-2009, 2007-2008, and 2003-2014. This may not take into account the achievements of the market-oriented reform of China's commercial banks. Therefore, this paper examines whether the bank loan of listed companies in China has a restraining effect on the manager's executive perquisites, and there have two contributions. First, this paper expands the research on the literature related to executive perquisites. Previous studies have focused on the discussion of the economic consequences of executive perquisites and the reduction of executive perquisites from the perspective of equity and manager incentives. According to the author's inquiry, there is no discussion of manager's executive perquisites from the perspective of creditor's rights. This paper empirically tests the influence of bank loan on executive perquisites, which is a supplement to the literature research on executive perquisites. Second, it expands the related literature research on the agency cost of listed companies in China under the background of post-share split reform. Domestic and foreign scholars are very concerned about China's split share reform, which is China's unique institutional background. However, the principal-agent problem between the controlling shareholders and the small and medium shareholders in the governance of listed companies in China was generally existed before. After the split share reform, now it becomes a principal-agent problem between the controlling shareholder and the minority shareholders and the principal-agent problem between shareholders and managers. According to the author's inquiry, the former scholars lack to consider the influence of the gradual dilution of the controlling shareholder's equity. In this paper, the change of shareholding of controlling shareholders highlights the problem of principal-agent relationship between shareholders and managers, examines the difference in the proportion of shareholdings of controlling shareholders and the impact of bank loan on managers' on executive perquisites.

Institutional Background
China is a country with a transitional economy and has special institutional characteristics. It is significantly different from other developed countries in terms of marketization, ownership of financial institutions, degree of government regulation, and characteristics of financial systems. China is a financial system dominated by indirect financing of banks, and the corporate bond market has not yet developed. However, until China joined the WTO, China's commercial banks showed the characteristics of regional division, business division, and serious government administrative intervention. After joining the WTO in 2001, China's commercial banks have a five-year protection period. After 2006, China's banking industry needs to be fully open to foreign banks. In order to meet the needs of the WTO competition rules, China's commercial banks have accelerated the pace of reform. The Commercial Bank Management Law promulgated in 1995 requires commercial banks to strengthen credit risk management and gradually reduce bank loan issued to state-owned enterprises by government administrative orders. This law has achieved obvious results after implementation [11]. In addition to enacting laws, China has also adopted state-owned banks to inject capital, set up special regulatory agencies, implement new accounting standards, and reduce state-owned shares and other reform measures.  Chinese banking reforms from cost and profitability, bank loan scores, Loan officer and credit manager loan accountability and the ease with which state-owned  [15].
The split share is formed when China sets up the securities market. In order to attract investors to participate in stock trading, the listed major shareholders are mainly state-owned major shareholders, who promised that their shares are not in circulation. This stabilizes the investment confidence of small and medium-sized investors. However, with the development of the securities market, the drawbacks caused by the inability of major shareholder shares to flow are obvious.
The main manifestation is that the major shareholder does not care about the value of the enterprise and even directly encroaches on the company's interests.
In order to solve the problem of split shareholding and maintain the long-term have implemented the split share reform. However, after the implementation of the split share reform, non-tradable shares of major shareholders holding more than 5% of the shares shall not exceed 5% of shares within 12 months and shall not exceed 10% of shares within 24 months. There are still quite a few shares of major shareholders in the lock-up period, which is often referred to as post-split-share reform. In the process of continuous circulation of major shareholder shares, on the one hand, the interests of major shareholders and minority shareholders are more consistent, reducing the agency cost of principal-principal between major and minority shareholders. On the other hand, the majority shareholder's shares are converted from non-tradable shares to tradable shares, usually accompanied by a continuous decline in the shareholding ratio of major shareholders, which is conducive to large shareholders to diversify their investment risks. Figure 1 shows the trend of the largest shareholder's share with the different period, and we can the mean, median, p25 or p75 are all decline over time which means that the decrease of the controlling shareholder's shares, so the controlling shareholder's supervision of the manager is reduced, and the principal-agent cost is more prominent.

Literature Review and Research Hypothesis
As Jensen and Meckling [16] pointed out, when banks act as external entities to supervise borrowing companies, agency costs are reduced. Banks will manage the value of borrowing companies not only when they have defaulted. They will take advantage of their professional advantages, various information resources and management methods to implement supervision, or adjust bank loan terms and actively participate in corporate governance activities of enterprises [17].  Cerqueiro et al. [7] and Lin et al. [8] obtained empirical support by using mature market data. For example, Cerqueiro et al. [7] used Swedish specific data to find that the design of bank loan collateral clauses supervised borrowers; Lin et al. [12] used US data to find that loan banks need a larger loan spread, higher loan fees, shorter loan terms, smaller loans Scale, more stringent loan agreements and more loan terms when there is greater control gains for borrowers.
In China's capital market, there are widespread problems such as weak legal system, not strict market management and untimely information disclosure, which may affect the efficiency between Chinese banks and borrowing enter- East (2017) found that China's bank loan can not play a corporate governance role [9]- [16]. For example, Bailey et al. [13] found that the event announcement effect of Chinese listed companies was significantly negative when they obtained bank loans, which means that companies with bank loans have worse performance in the future; Qian and Yeung [14] found that the listed companies with controlling shareholders' high benefit expropriation continue to acquire new bank loan.
Jensen and Meckling [22] argue that executive perquisites are agency costs that undermine corporate value [18]. In the United States, the market reaction to declare or disclose executive perquisites is negative [1] [2] [3]. Luo et al. [5]  Jensen's [19] free cash flow hypothesis argues that debt financing is a hard constraint that can reduce the manager's free volume ruling and constrain the manager's interest encroachment behavior. As a professional financial institution, the bank can avoid the problem of "free rider", supervise managers in a market-oriented system, restrain the interests of managers from benefit expropriation, and reduce the executive perquisites of business managers. Therefore, the first research hypothesis of this paper is proposed: Research hypothesis H 1 : Bank loan reduces the executive perquisites of enterprises.
According to Lin et al. [20] and Lin and Tan [21], China's state-owned enter- 1 According to the executive perquisites indicators Perks1 and Perks2, the high and low two subsamples are divided. In the two subsamples, there is a significant difference in the average of ROA. Among the two subsamples of Perks1, the average ROA of the subsamples with high executive perquisites was 3.64%, which was significantly lower than the average of the low executive perquisites subsamples' ROA, 4.10%. In the two subsamples of Perks2, the average ROA of the subsamples with high executive perquisites was 3.48%, which was significantly lower than the average ROA of the low executive perquisites subsamples, 4.25%.
prises not only bear social burdens such as social employment and social stability, but also undertake the strategic burden of economic transformation. Therefore, state-owned enterprises have soft budget constraints. Lin Yifu and others proposed the concept of the viability of state-owned enterprises. They believed that only by stripping these two kinds of burdens would it be possible to solve the soft budget constraints of state-owned enterprises in China, and state-owned enterprises could become real competitors in the market. China's state-owned enterprises have started market-oriented reforms since the mid-1980s, but so far these two types of burdens, especially the strategic burden, have not yet been completed, and bank loan is difficult to play a supervisory role under the soft budget constraints of state-owned enterprises. On the contrary, non-state-owned enterprises are the participants of market competition. Under the background of market-oriented reform of commercial banks in China, they do not have the problem of soft budget constraints. They must obtain bank loans through their own advantages. Bank loan plays a supervisory role among non-state-owned enterprises. Therefore, the second research hypothesis of this paper is proposed: Research hypothesis H 2 : Compared with state-owned enterprises, bank loan has a more significant inhibitory effect on the executive perquisites of non-stateowned enterprises.
After China's split-share reform, the non-tradable shares of the original major shareholders gradually turned into tradable shares, and the major shareholders gradually reduced the number of shares they held to disperse their own risks. At this time, the interests of the major shareholders and the interests of the company gradually become consistent, and with the decrease in the number of shares of the controlling shareholder, the incentives for the supervision of listed company managers are reduced, and the agency cost of managers has gradually become prominent. Therefore, the supervision effect of bank loan on the manager's executive perquisites is more obvious. Therefore, the third research hypothesis of this paper is proposed: Research hypothesis H 3 : Compared with listed companies with high shareholding ratio of controlling shareholders, the bank loan of listed companies with low shareholding ratio of controlling shareholders has more obvious restraining effect on executive perquisites.
To get the empirical results, we do OLS regression which controls the year effect and industry effect to empirically test the impact of bank loan on executive perquisites, and further consider the impacts of different property rights and the proportion of the first largest shareholder holding and divide two sub-samples to test. Finally, we do a robustness test. Z-score is used as instrumental variable to deal with the endogeneity problems; considering the impact of liability and equity balance.

Data Sources
We consider the Background of Post-Split-Share Reform, selecting the period of

1) Dependent variable
Executive perquisites is measured by two methods: indirect method and direct method: define executive perquisites as deducting management expenses, total management compensation, bad debt provision and amortization of intangible assets; define that executive perquisites is a possible item in the "other cash related to operating activities" item in the notes to the company's annual report.
The sum of the eight types of expense items related to executive perquisites of executives: office expenses, travel expenses, business hospitality, communication fees, training abroad, board fees, car fares and conference fees [6]. Executive perquisites Perks1 is defined that the executive perquisites is divided by operating income; executive perquisites Perks2 is based on the executive perquisites defined by Zhang et al. [6], which is divided by operating income.

2) Independent variable
Bank loan is measured by according to the short-term and long-term borrowings in the balance sheet: Define Loan = (short-term borrowings + long-term borrowings) ÷ total assets.

Model
First, the following empirical model is proposed to examine the impact of bank loan on executive perquisites: Perks is executive perquisites, Loan is bank loan, X is the control variable, Year is the annual dummy variable, and Industry is the industry dummy variable. If the regression coefficient is significantly negative, then the study hypothesis H 1 is empirically supported.
Secondly, to examine the impact of the nature of property rights on the relationship between bank loan and executive perquisites, we can use the sub-sample of state-owned enterprises and the sub-sample of non-state-owned enterprises to use the empirical model (4-1) for measurement analysis and the following em-   Table 2 describes the statistical characteristics of the main variables. In Table 2, the mean and median of in-service Perks1 were 9.32% and 6.72%. The mean and median consumption of Perks2 were 1.83% and 0.46%, slightly lower than the average median Perks2 statistics of 1.945% and median 0.977% [9] of Zhang et al. (2015); the mean of bank loan Loan and The median was 22.13% and 17.84%. Table 3 is the Pearson correlation coefficient between variables. It can be seen from Table 3 that the absolute value of the correlation coefficient of each variable is less than 0.5, and there is no significant multicollinearity; the correlation coefficient between the executive perquisites Perks1 and Perks2 and the bank loan Loan is significantly negative, in other words the higher the bank loan of the listed firms, the lower the managerial consumption on In-service. Table 4 shows the results of the regression of the full sample data into the empirical model (4-1). As can be seen from    Note: *, ** and *** indicate passing tests at 10%, 5% and 1% significance levels, respectively.  Year executive perquisites, but the bank loan of non-state-owned listed companies can be significant. In addition, the regression coefficient of the intersection term Loan*SOE in the full sample data is 0.1107, and the significance level is 5%. That means, compared with state-owned enterprises, the impact of non-state-owned bank loan Loan on executive perquisites the Perks1 is more significant. In the Perks2 column of executive perquisites, the regression coefficient of bank loan Loan in state-owned listed companies is −0.0091, the significance level is 10%, and the regression coefficient of non-state-owned listed company bank loan Loan is −0.0324, the significance level is 1%. Additionally, there is a significant difference in the regression coefficient of bank loan Loan between state-owned and non-state-owned listed companies. Chow Test = 34.56, corresponding p value is 0.000. In addition, the regression coefficient of the intersection term Loan*SOE in the full sample data is 0.0191, and the significance level is 1%. That means, compared with the state-owned enterprises, the non-state-owned bank loan Loan has a more significant impact on the executive perquisites Perks1.

The Influence of Property Rights
Therefore, this empirical result shows that bank loan has no significant corporate governance effect in state-owned enterprises, and has significant corporate governance effects in non-state-owned enterprises. This is related to the soft budget constraints of state-owned enterprises proposed by Lin. and others.
State-owned enterprises and banks are state-owned, while state-owned enterprises have a strategic burden of solving social burdens such as employment and industrial restructuring. State-owned enterprises have soft budget constraints, and corporate governance effects of state-owned enterprises are not significant.
That means H 2 is established. Therefore, the reform of China's commercial banks needs to be deepened, reducing government administrative intervention and increasing the independent status of participating in market economic activities. of listed companies on the relationship between bank loan and executive perquisites in the context of post-split-share reform. It can be seen from Table 6  Therefore, H 3 is established.

1) Endogenous discussion
Considering the possible relationship between bank loan and executive perquisites, this paper uses the instrumental variable method to re-examine the above conclusions. Since China's commercial banks generally adopt the scoring system to decide whether to issue loans, this paper chooses the Z value as the instrumental variable of bank loan. The empirical results are shown in Table 7.
The Z-value is calculated using the Z-score model proposed by Altman in where X 1 = net working capital ÷ total assets; X 2 = retained earnings ÷ total assets; X 3 = profit before interest and tax ÷ total assets; X 4 = total market value ÷ total liabilities; X 5 = sales revenue ÷ total assets.
It can be seen from Table 7 that the regression coefficients of bank loan Loan are significantly negative, both executive perquisites Perks1 and Perks2, and the main research conclusions remain unchanged.
2) The impact of liability It's probably not because the bank's supervisory mechanism works, but that debt as a hard constraint has a depressing effect on the manager's encroachment on the company's interests. We can see that the regression coefficient of the asset-liability ratio Debt is not significantly stable and negative from the previous empirical results of Tables 4-7. Therefore, it is not because the hard constraints Z. Q. Ye et al. of bank loan have a corporate governance effect on the manager's executive perquisites, but because the bank is a professional financial institution, which has the advantages of information, talents and technology, and can better overcome the problem of information asymmetry and adverse selection, playing a role in corporate governance.

3) Equity balance
Although the results in Figure 1 indicate that the shareholding ratio of the largest shareholder is still high, Figure 1 shows the fact that the shareholding ratio of the largest shareholder is declining, which may cause changes in the equity balance between major shareholders, influencing executive perquisites of managers. The degree of equity balance is high, so the incentives of the supervisors of the major shareholders are reduced, and the power of the managers increase. At this time, the effect of bank loan on manager supervision is stronger, and the effect of reducing managers' executive perquisites is stronger at the same time.
Therefore, this paper defines the equity balance degree as the ratio of the shareholding proportion of the second largest to the fifth largest shareholder to the shareholding proportion of the largest shareholder, and replaces shareholding ratio of the largest shareholder with the equity balance to discuss the impact of bank loan on executive perquisites in the context of split-share reform. The results can be seen from Table 8 that whether Perks1 or Perks2, the absolute value of the bank loan Loan regression coefficient is higher than the equity balance, supervisory effect of bank loan in the context of the split-share reform.

Conclusions
Different from the financial system based on direct financing such as Britain and the United States, show that bank loan can significantly reduce the manager's executive perquisites.
Compared with state-owned enterprises, bank loan has stronger corporate governance effects in private enterprises, and significantly reduces managers' executive perquisites. Compared with the listed companies with a high shareholding ratio of the largest shareholder, the corporate governance effect of bank loan with a low shareholding ratio of the largest shareholder is stronger, and the effect of reducing manager's executive perquisites is stronger.
The research conclusions of this paper not only provide empirical evidence for the discussion of China's commercial bank reform, but also determine the success of China's commercial bank reform, existing a strong corporate governance effect. However, the governance effect of state-owned listed companies is not significant, and it is required to further deepen the reform of China's commercial banks. For example, we should increase the proportion of foreign shareholdings in China's commercial banks, or expand the business scope of foreign banks in China, and further reform the internal governance mechanism of commercial banks. It also provides empirical evidence for the reform of listed companies in China under the banking-based financial system. The empirical research of the predecessors rarely considers the bank as an external corporate governance mechanism. This paper provides sufficient evidence to support the corporate governance of commercial banks in China.