The Impact of Fund ’ s Heavy Position on Company Performance — Empirical Evidence of China ’ s Main Board from 2007 to 2014

In China, developing institutional investor, especially securities investment fund, becomes more and more important. In order to improve the level of company governance by bringing in institutional investor, China Securities Regulatory Commission supports the development of institutional investor by legislation and other polices. Papers about the relationship between fund holding and company performance are still not enough and a lot of researches don’t consider the effect of some moderator variables. This article mainly researched for the influence of shares heavily held by fund on the performance of company, using data of A-share listed companies in Shanghai and Shenzhen during 2007 to 2014. It also introduced two moderator variables to discuss whether there existed effects on company performance if the fund was the second largest shareholder or the holding company was state-owned. Actually, the empirical result showed that, the percentage of stocks held by fund had a positive relationship with company performance. If the fund was the second largest shareholder, it would strengthen the relationship between fund heavily holding stock and corporate performance. If the company whose stocks were held by fund was stateowned, the influence on company performance would weaken.


Introduction
The issue of "the Law of Chinese Security Investment Fund" in 2004 has set a standard for regulating the opera-* Corresponding author.tion of the fund industry.After this, with the support of a variety of policies, the China's fund industry develops at a steady speed.It is well known that securities investment fund, on behalf of institutional investors, plays an important role on helping the development of institutions, promoting the management of listed companies and other aspects.Through heavily holding shares of listed companies, fund makes itself an important shareholder of listed company.As an important external manager, fund can have significant impact on the governance of listed company.Jinguan Yang et al. [1] deemed that fund held the shares of listed companies in order to obtain investment returns and the fund's income was largely derived from capital gains.Since the operation performance of holding companies would affect its stock price in the market (Daniel [2]; Dai Xiaojuan [3]), thereby affected the fund's investment returns.With the stocks held by the fund increasing, the stocks' liquidity is reduced.In order to ensure that the fund could get significant benefits before the exit of listed companies, the fund has a strong incentive to participate in the governance of listed companies.In the case of inability to quickly exit, the fund would gradually change the passive shareholder strategy in the past.Jinyan Shi [4] thought that, on the other hand, it would alter to the identity of important external governance, actively involved in the internal governance of listed companies, and focused on corporate performance and long-term value, which would affect the performance of listed companies.In recent years, the researches about institutional ownership affecting the performance of company have been the academic focus.Thus, how the fund holding affects corporate performance?
Based on the China's A-share market, we firstly studied the impact of fund shareholding ratio and fund's heavy position on the performance of listed companies.Secondly, we explored how the fund affected the company performance when it was the second largest shareholder of the company.Finally, we studied when the fund holding company was state-owned, how the impact of fund's heavy position on the company performance would change.

The Literature Review and Research Hypothesis
The current studies about the influences of fund holding on corporate performance are mainly expanded around the effects of institutional investors' ownership on corporate performance.
Jensen [5] thought institutional investors may supervise the management of listed companies, and potentially control the company governance issues.With the stake increasing, their motivations to monitor the company governance and management behaviors become more intense.Demsetz et al [6] believed that compared with individual investors, institutional investor owned more shares of listed companies.In order to obtain more holding incomes, the motivation of its regulatory to listed companies would be stronger; at the same time, institutional investor had greater voting rights, which guaranteed the power of their participation in corporate governance.Through researching, the authors found that the institutional investor shareholding proportion had a significantly positive impact on company performance.Moreover, Smith [7] found that maintaining the independence of the external institutional governance could improve internal governance, Helwege et al. [8] considered that institutional investor through active participating in corporate governance, continuously improved the company governance structure and optimized the equity structure of the company, thus to improve company performance.Gillan and Starks [9] found that institutional investors had been major shareholders of most US companies.Institutional investors had become active participation in corporate governance and could have an effective oversight of the company management.
From the domestic point of researches, Weian Li and Bin Li [10] through empirical method found that institutional investor effectively enhanced the level of corporate governance, and reduced agency costs of listed companies.Institutional investor shareholding had a significantly positive correlation with corporate performance and market value.Haifeng Fan et al. [11] from the heterogeneity point of institutional investor found that increasing the fund shareholding ratio would facilitate fund's monitoring to listed companies, thus increasing the market value of listed companies.Zhonghai Li and Dixin Zhang [12] thought that, especially for securities investment fund, there was a positive correlation between the proportion of the fund holding and the company operating performance.The higher the stake was, the better the holding company operating performance would be.Jinyan Shi et al. [13] using 2005-2011 mainboard listed companies as the study sample, through the empirical analysis of the panel data, with the increasing proportion of the fund shareholding, the fund's power of involving in corporate governance and supervising of the largest shareholder would continue to enhance, thus improved corporate governance, and improved corporate performance.Xianzhi Zhang, Xingfei Jia [14] used empirical analysis to find that different ownership stake and time range of social security fund had impacts on the value of listed companies.The empirical results showed that increased stake would significantly improve the company value.The longer the holding period was, the higher company value increased.The same with the holding stake.What's more, the social security fund holding in general could play a positive role in strengthening supervision, weakening the agency problem, etc., thereby enhancing the value of listed companies.Weimin Xie [15] used 2005-2011 mainboard listed companies as the research sample, and found that share proportion held by social security fund had a positive effect on company performance after controlling other variables' influences.Yajing Huang and Yueming Zeng's [16] study showed that China's institutional investor holding did play a positive role on the performance of listed companies.High share proportion held by fund was expected to improve corporate performance.While, low share proportion held by securities companies, QFII, insurance fund, social security fund and trust companies did not have significant influences on enterprise performance.
Based on the above analysis, we can find that, along with an increase in fund shareholding ratio, the motivation and ability of fund to participate in company governance are also growing.This power can play an important role in the company incentive and supervision, and also decision making.When the fund shareholding ratio reaches a certain percentage, earnings through participation in corporate governance outweigh the cost of their participation in corporate governance generated, which further promotes the fund companies to participate in corporate governance, thereby enhancing the company performance level.Especially when the proportion of fund holding is more than 10%, it is possible to send a director to play a more important role in improving corporate governance.When share proportion held by fund is more than 10%, this article would define it as fund's heavy position.This paper will investigate the effect of the fund's heavy position on the company performance, we raised the proposed hypothesis 1: Hypothesis 1: fund holding proportion has a significantly positive correlation with corporate performance, and so does fund's heavy position.
Because of the special China's background, the shareholding ratio of the largest shareholder generally occupies too high.The dominant shareholder usually tends to occupy the interests of minority shareholders, which is not conducive to improve corporate performance.Check-and-balance ownership structure can help improve the level of corporate governance.Yuanpei Lin and Wen Chen [17] selected the listed companies in growth enterprise market as sample, they found that the share proportion held by the first largest shareholder had a significantly negative relationship with company performance through empirical analysis.Jianxin Tang et al. [18] employed non-financial companies from 2003 to 2010 in China as the sample and analyzed whether equity ownership structure and the characteristics of board were conducive to tunneling.These results indicated that the first largest shareholder tunneled the listed company by acquiring equity control, which was harmful to company performance.Check-and-balance ownership structure could inhibit this tunneling behavior.Deping Chen and Yongsheng Chen [19] found that a high degree of shareholder balance in the company would prove to have a higher operating performance.Normally, the second largest shareholder holding ratio is higher than the sum of ratio of third to the fifth shareholders.And therefore to balance the first largest shareholder, it will need the second largest shareholder or the cooperation of other shareholders and the second largest shareholder.Fund as one of institutional investors, has the advantage of large scale funding, professional quality and other advantages.If the fund is the second largest shareholder of the company, the fund will actively involve in corporate governance, effectively supervise the first largest shareholder, and play professional advantages, thus improve corporate performance.So we put forward the hypothesis 2: Hypothesis 2: The fund as the second largest shareholder will improve the listed company performance and will strengthen the positive impact the fund's heavy position will have on company performance.
Most of our country's listed companies are converted from state-owned enterprises.The ultimate control is in the hands of the government, so the state-owned enterprises will take some non-economic tasks.Their goals are often influenced by political and social factors.While the government departments, in order to safeguard their own interests, would adopt some administrative intervention to state-owned companies, which is not conducive for the fund to play a role in corporate governance.Xinmin Dai et al. [20] used the date of state-owned listed companies of China's manufacturing industry from 2006-2011 as sample, they found that there existed negative relationship between the property of state-owned companies and companies performance.Hanjun Li et al. [21] empirically examined the relationship between the state-owned enterprises ownership structure and corporate performance from 2007 to 2012.The results indicated the implementation of equity diversification of stateowned enterprises, especially the institutional investors, could improve corporate governance structure to pro-mote its performance.Xing Liu and Xianchong Wu [22] found that the nature of state-owned had a negative impact on the fund's effective of improving the company performance.Accordingly, the proposed hypothesis 3: Hypothesis 3: When the listed company is state-owned, its performance is lower than the non-state-owned holding company.And this nature will weaken the positive impact the fund's heavy position will have on company performance.

Sample Treatment
The study sample in this paper is selected from listed company in Shanghai and Shenzhen Stock Exchange from 2007 to 2014.The data for dependent variables, independent variables and control variables is selected from Wind Financial Database.Corresponding financial data is from the CSMAR financial analysis index database.In this paper, primary sample is processed according to the following procedures: (1) Excluding financial and insurance companies.(2) Excluding ST, PT companies.(3) Excluding insolvency companies or related data incomplete companies.After the above dispositions, a total of 11,805 samples obtained.

Variable Description
Table 1 shows the type, name, symbol and measure method of variables.

Model Specification
To test whether hypotheses 1 and 2 is established, we set current return rates of total assets ROA as the dependent variable, the last period fund holding ratio HP as explanatory variables.At the same time, we added some appropriate control variables, and built the multiple regression Equation (1) for authentication.When HP regression coefficient is significantly positive, indicating that the higher the proportion of shares the fund holds, the more powerful effect the fund has on improving corporate performance, which verifies the hypothesis 1.To further test the hypothesis 2, we introduced two dummy variables in the regression Equation (1).One was whether the fund was in a heavy position, and another was whether the second largest shareholder of the company was fund.As the second largest shareholder has balance power to the first largest shareholder.If the fund is the second largest shareholder of its holding company, it will strengthen the impact of the fund's heavy position on improving company performance.For this reason, we introduced the interaction term HH * FS and investigated the fund's regulatory role as the second largest shareholder in its holding company.When β 3 , β 4 are significantly positive, hypothesis 2 was verified.In order to control the influence of the first largest shareholder of the company performance, we introduced the first largest shareholder holding ratio Fir_Share as control variable into the

∑
ROA i,t represents the company's return on total asset in period t.HP i,t−1 represents in the t − 1 period, the share proportion held by fund, taking the hysteretic nature of the fund's heavy position to the improvement of corporate performance into consideration.So introduce the lag one period variable.HH i,t−1 *FS i,t represents the FS i,t as the moderator variable introduced to model to study when the company's second largest shareholder is the fund, how the impact of the fund's heavy position will have on the company performance.
To test the hypothesis 3, we added state-owned dummy variable in the regression model (2), and examined the impact of the nature of state-owned on performance.Then we introduced an interaction term HH * SO, and studied whether the property of state-owned would weaken the fund's positive impact on corporate performance.Regression model ( 2) is as follow:

Descriptive Statistics
Table 2 divides samples into two categories according to whether the fund heavily holds companies' shares and lists the consequence of variables' descriptive statistics.In Table 2 we can find that the company performance is better when its shares are heavily held by fund.Specifically, the mean of proxy variables of company performance such as ROA, Tobinq are higher, they respectively reach 0.0744, 2.1449.As to the median of ROA, Tobinq when shares heavily held by fund, and the numbers are 0.0641, 1.7010.They are still higher than the numbers of companies whose shares are not heavily held by fund.We can find that the standard deviations of ROA, Tobinq are 0.0601, 1.333w3 when HH remarks as 1.They are lower than the numbers of companies when HH remarks as 0, which shows that the companies' operations are more stable when the fund heavily hold their shares.The HP's mean is 0.2672 when fund heavily hold companies' shares, which is higher than the number of HP's mean when fund don't heavily hold companies' shares, and the numerical difference is big.

The Correlation Test
Table 3 shows the correlation between the variables by using Spearman correlation test, which can judge whether there is a serious multicollinearity problem between variables.It can be found from Table 3, fund shareholding proportion is significantly positive correlation with performance indicators ROA, Tobinq, so the hypothesis 1 is preliminarily verified.The coefficient of FS (whether the fund is the second largest shareholder of the company) is positive, which means company performance will be improved when fund is company's second largest shareholder.The property of state-owned is significantly negative correlation with corporate performance, which is consistent with the hypothesis 2, 3.The shareholding proportion of the first largest shareholder is significantly negative correlation with corporate performance.And other control variables are significantly correlated with corporate performance.
In the view of the correlation coefficients between the various explanatory variables, we can find that the correlation coefficient between HH and HP is 0.52, which is biggest.The correlation coefficients between the other variables are less than 0.45, and the numbers of VIF are all less than 10 through testing, so this part can indicate that above models don't exist serious multicollinearity problems.Also the correlation between Tobinq and ROA is significantly positive, so this paper took robustness test by using Tobinq as a substitute variable for ROA.

The Empirical Result Analysis
Through using the unbalanced panel data from 2007 to 2014, the Table 4 shows OLS regression result by using fixed effect model (Hausman Test rejected the null hypothesis that there was no systematic differences between fixed effect method and random effect method, so this study used fixed effect model).It can be found from Table 4, the coefficients of HP (fund shareholding proportion) in regression result (1), ( 2), (3) are positive (0.129, 0.103, 0.104), and being significant at level 1%, which shows that the higher the proportion of the fund shareholding , the more active the participation of corporate governance is.Because the fund's earnings are correlated with the stock price of holding company, and the stock price is affected by company governance.Thus, the fund will have motivations to play an important role in supervising company governance, and the company performance will be improved.These regression results confirm the hypothesis 1.The coefficients of HH (whether the fund heavily hold companies' share) in regression result (2), (3) are significantly positive at level 1%.Fund can give full play to the functions of their supervision when heavily holding companies' shares.Because the fund can send a director to the company, which can enhance company governance level.
The coefficient of FS in regression result (2) is positive, which is significant at level 1%.So we can know that company performance will be improved if the company's second largest shareholder is fund.Some reasons are as follow, on the one hand, high shareholding ratio gives the fund positivity to participate in company management.Fund can make good use of talents and have more chance to improve company governance.On the other hand, the fund can balance the power of the largest shareholder when being the second largest shareholder of the  company, which can cease the condition that a dominant controls the company and can inhibit the first largest shareholder's tunneling behavior.Therefore, the company performance will be improved.From the point of interaction item HHFS, the coefficient is significantly positive at level 1%, which shows that the promotional effect of fund's heavy position will be better when the fund is the second largest shareholder of the company.The above results confirm the hypothesis 2.
The coefficient of SO (whether the companies are controlled by government) in regression result (3) is significantly negative at level 1%, which shows that non-state-owned enterprises have better performance than state-owned enterprises.Because non-state-owned enterprises are less affected by administrative intervention, they can pursue the goal of maximizing business efficiency without government's mandatory intervention and need not consider too much social responsibility.So they can focus on improving company performance.While, the state-owned corporate may assume some social goal and help government to realize some functions, and maximizing company performance isn't its main purpose, which is not good for company development.The coefficient of interaction item HHSO in regression result (3) is significantly negative, which means that the state-owned characteristic of listed companies may limit the fund's role in corporate governance and weaken positive impact of the fund's heavy position on corporate performance.The above results confirm the hypothesis 3.
From the point of other control variables, the coefficients of Fir_Share (the shareholding proportion of the first largest shareholder) in regression result (1), ( 2), ( 3) are all significantly negative, which shows that the higher the proportion of shares held by the first largest shareholder, the worse performance of the companies is.The first largest shareholder may not be well supervised and consider its own interests too much, which will violate small shareholders' interests, and having bad impact on the operation of whole company.The coefficients of lnSize (the natural logarithm of the company size) in regression result (1), ( 2), ( 3) are all significantly positive, which means the larger the scale of company is, the better the performance of company will be.The company will pay more attention to governance when company size is large.The company will improve company performance by exploiting scale economies effect and enhancing the market share.The coefficients of Lev (leverage ratio) in regression result (1), ( 2), ( 3) are all significantly negative.This part shows that the higher the leverage ratio is, the worse performance of the company will be.This is because company will ensure a lot of interest when borrowing much money.At the same time, the company will face the risk of capital chain rupture, which will lead the company to bankruptcy.Thus, the company may face a lot of bondage and do harm to the company operation.
In order to get more reliable conclusions, this paper took robustness test by using Tobinq as an indicator to measure the performance of the company.The regression results are showed in Table 4, we can find that the coefficients of HP, HH, FS, HHFS are all significantly positive and the coefficient of SO is significantly negative at level 5% in regression result (4), (5).The coefficient of HHSO is negative, but not significant.Regression results are similar to the above researches, which shows that the models are robust.

Conclusions and Suggestions
This study used the Shanghai and Shenzhen's A-share listed companies from 2007 to 2014 as the research sample.Firstly, it researched whether the stocks proportion held by fund influenced the company performance.Secondly, it showed that how the condition would impact the company performance if the fund was the second largest shareholder.Thirdly, it analyzed that how the condition would impact the company performance if the company was state-owned.Through empirical test, this paper confirmed the hypothesis 1, 2, 3 and had the following conclusions: 1) the stocks proportion held by fund had a significantly positive influence on corporate performance.Because the more proportion the fund company holds, the much more actively the fund will participate in holding company governance, which will gradually promote the company performance by supervising company operation; 2) fund as the second largest shareholders would improve the holding company performance, and would strengthen the positive relationship between fund's heavy position and company performance.When the fund as the second largest shareholder of listed companies, the fund can effectively balance the first largest shareholder's rights and help the company to make better decisions, so the condition can improve company performance; 3) the state-owned property would have a significantly negative effect on corporate performance, and would weaken the positive relationship between fund's heavy position and company performance.When the actual controller of the company is government, the government will take account of political and social purposes and may intervene in the company governance.And this case will reduce company performance level.
In view of the above research conclusions, this study raises following suggestions: First of all, the government of China should take effort to develop securities investment fund and other institutional investors.And China Securities Regulatory Commission can help form a diversified, multi-level professional investment team.Secondly, listed companies should actively introduce fund investors and properly increase the proportion of stocks held by fund.As the fund company has the advantage of professional talents, listed companies should encourage fund investor to participate in company governance.Thirdly, government should further strengthen the reform of state-owned enterprises and gradually reduce the proportion of state-owned shares.As the actual controller of state-owned enterprises, the government should reduce administrative intervention in the listed companies and actively encourage fund investors to participate in the governance activities of listed companies.

Table 1 .
Variable description table.At last, the paper brought in annual dummy variables to control the impact of different years on corporate performance.Regression model (1) is as follow: _ _ ∑Referring to existing classic literature research methods, the paper chose Tobin's Q value (Tobinq) to replace the ROA to do robust test.

Table 2 .
The result of descriptive statistics.