Fund Company’s Equity Structure Effects on Fund Performance

It is well known that the equity structure of listed companies have effects on the company performances to some extent. However, the fund company is one special kind of financial enterprise and the relationship between its unique equity structure and fund performance still remains in suspense. On the basis of theoretical conjectures and results of predecessors, the development history of equity structure of Chinese fund company is analyzed and the fund evaluation methods are discussed in this paper which is based on the risk adjustment of excess return. Through sieving and accounting data, we conducted an empirical study about the effects of equity structure and regional feature on fund performance. The conclusions can be drew as that the fund company with foreign capital shows no signs of highlights and we found that concentrated shareholding and national resources are beneficial to improving fund performance level. In addition, the operating time and the fund company’s location, investment fund’s average tenure also have influences on fund performance.


Background and Significance
Securities investment fund (hereafter referred to simply as fund) is a kind of independent and acceptable asset which combined the indeterminacy investor's funds through fund share selling. The fund is trusted by fund custodian, managed by professional fund manager and it also can be viewed as a collective investment behavior, which is benefit-sharing and risk-sharing, carried through investment portfolio. This kind of indirect investment is quite different from stock and bond investment. Principal-agent The problem of normal investor's lacking of professional knowledge is solved, and the unsystematic risk is dispersed under the principal-agent by balancing the scales of investment stock, bond and monetary fund with the assurance of reasonable return rate.
According to the Wind database statistics, by the end of March 2015, 96 fund companies have been established and 3088 fund products been issued in total in China. The net value and circulation development 1 of fund in the market are showed under the following figure. Behind the prosperity of overall numbers there still exist some inevitable problems such as principal-agent problem. For instance, Chinese stock market has stepped into a period of deep change due to the bad expectation of International market since the early 2011 [1] . By September 30 th 2011, Shanghai Composite Index has been shocked from 2852.65 to 2359.22, dropped 493.43(17.29%). At the same time, SSE Fund Index has dived an accumulative 14.53%, slightly lower the drawdown of stock market and consumer expectation. Another example shows that from March 3 rd 2014 to March 2 nd 2015, Shanghai Composite Index gained 60.76% and reached 1261.04, but at the same period SSE Fund Index reached 1856.36, only gained 49.29%.So that we can see the comprehensive achievement of fund even lower the whole market index. The aim of fund investment is for a better resource allocation which can achieve higher returns and lower risks. However, there still exist some gaps between the current Chinese fund performances and investor's expectations. The investment fund manager's incomes are still gainful whether the fund lost or lower than stock gains, which lead to the reveal of principal-agent problems.

Domestic and International Research Situation
1.2.1. Relevant researches of fund management company's equity effects on fund performance The main type of foreign fund company is corporate-type, while it's almost contract-model in China, which leads to a great difference between the methods and results of equity structure and fund performance researches. Even research object that all Chinese companies taken is the same, the conclusions from different studies are often inconsistent.
Generally speaking, a certain extent concentration of equity structure is beneficial to company's making decisions and improving performance level. Randall Morck, Masao Nakamura and Anil shivdasani (2000) carried out a study about Japanese listed company [2] . The result shows that within a certain range, the more bank holds company stock shares, the better the company performance is. But once beyond the range, with the increase of shareholding ratio, the company performance becomes lower. Considering the different governance structures between the fund companies and ordinary companies, Robert F. Radin and William B. Stevenson (2005) came up a new point which says that based on the fund company's features [3] . We should strengthen independent director's supervision and shareholder's function so that we can enhance achievement and improve performance. According to the research of Qi Chen, Itay Goldstein and Wei Jiang (2005), the equity structures of fund company are related to the fund performance evaluation results of forecaster(Xing Chen) [4] . They find that different equity structures will significantly contribute to different company performance. Besides, the concentration of equity structure, small-scale of board and a larger proportion of independent board can reduce the relative rates.
Domestic research results show different viewpoints about particularity of Chinese capital market and Chinese basic condition. Li Xuefeng and Zhang Jian (2008) at Nankai university found that the factors such as the number of fund company shareholders, and the equity ownership of the largest shareholder and financial companies all have a positive role on fund performance [5] . Further, they proposed an equity structure which can optimize fund performance based on the findings. At the same year, Zhang Bin and Lin Sheng (2008) showed that the performance of fund companies is better when it compared with holding company and they supposed that a moderate increase in foreign capital shareholders will play an active role in enhancing fund performance [6] . As for the state-controlled enterprises, Jiang Ping, Tian Shu and Cheung Yan-Leung (2011) build a panel data measure model and draw the conclusion that the fund performance of state-controlled enterprises and Chinese-foreign joint ventures are much more better, which means ' the helping hand of national resources' does really exit [7] . However, through the empirical analysis, Gao Yuan (2012) found that foreign shareholders are not favorable for enhancing fund performance [1] . What's more, with the increase of equity concentration and securities company's stock shares, fund manager's ability of market timing and stock picking decreased.
It is found that documents of researching fund companies are not only less than normal companies, but also some results of study of fund companies are inconsistent, after collection of research documents and materials. So, further studies about predecessors' results of study are carried on and real factors affecting performance of fund companies are also tried to be found in this paper.

The related researches of Performance Evaluation
method Sharpe (1966) used standard deviation to replace β of Treynor's index with the risk-adjusted basis [8] . However, the Treynor index may not be applicable when it comes to the situation of market descending tendency and this point was questioned by Jobson and Korkie(1981) [9] . In addition, Israelsen (2005) pointed that the fitting degree of traditional Sharpe index in bear markets is pretty low so that he corrected Sharpe index, replaced it with the product of excess earnings and risks and found it fits better than traditional index [10] .
In recent years, domestic scholars combined foreign experience with Chinese condition and made a variety of improvements. In 2002, some researchers used RAROC method in the evaluation of fund performance. Hui Xiaofeng and Chi Wei (2002) collected 33 closed-end funds and conducted empirical research in fund performance by four quarters of 2009 [11] . They found that there was a great difference between the research result and the direct evaluation of yields. Now that the researcher assumed that samples meet normal distribution condition, the experimental accuracy needs to be improved. Based on VaR, Xu Cuipin, Shi Qinghua and Shi Zhenghua(2007) built a fund performance evaluation model and further tested it, which showed that this model can overcome the defects of the traditional methods and has a better predicting ability [12] .

Empirical Analysis of Fund Company
Equity Structure Effect on Fund Performance

Selection of Period and Sample Data
In order to make research results more valuable, authors collect the five years data from 2010 to 2015 to be close the actual condition. Because the net asset value of fund is used in the following parameter settings contents and these data are mainly revealed in the fund company's quarterly reports, we chose the quarterly data as the research data. From the second quarter of 2010 to the first quarter of 2015, we collected 20 period data altogether. In practical application, we regarded the second quarterly data as the initial data since it's similar to yield calculation so that actually we got 19 periods.
In view of numerous of fund companies, fund companies are selected as sample size in the following criterion, namely, investment ways of fund and operation time of fund which better embody equity structure's impact on fund performance.
As it mentioned above, there are 96 fund companies are under operation in China and we selected the top 40 due to the Wind Information. The process of how the representative funds of each company are showed as the following flow chart. After the above process of selecting sample fund, there are 30 fund companies and its funds are chosen as the samples, which are showed as the following table:

The Selection and Definition of Variable
The samples include 30 funds and time span is 19 quarters, so authors think it necessary to provide the first fund company as example and make a critical analysis in this paper. Authors can provide other data in the following pages in case some readers need. Normally the number of fund company shareholder is less than 8, usually equal to 3 or 4. The autocorrelation exists if we select variable like the first big shareholder, second big shareholder, thirdly big shareholder or shareholder number. After referencing the definition of Herfindahl exponent, we introduced equity structure concentration system and built HHI index, which equal to the sum of squares of each shareholder's shareholding ratio. The index is between 0 and 1 and the more it approaches 1, the higher equity concentration it has. Size Fund company size The data come from Morningstar.com.

Descriptive Statistics
From the table above we can see that from 2010 to 2015, the abnormal return ratio is 0.271% and after the risk adjustment, the average quarterly abnormal return ratio of representative fund is -1.93% which means that Chinese fund investment level reminds to be improved. There are two ways to improve the level, whether reduce the risk degree or increase the yield degree. As for the fund equity structure and its characters, 90% of the companies are state holding enterprise and 53% of them are joint venture enterprise. The average value of equity concentration is 0.408 and the standard deviation is 0.11, which means the equity concentration level is pretty high and also proves the direct observation results. Most fund companies were established before ten years and the fund manager's average tenure is 904 days (nearly two and half years) which shows fund manager's job is pretty stable during the sample period, meanwhile, the average size of fund company reaches 100 billion. The city gets degree as 1.8 which means that the probability of a fund company choose Beijing and Shanghai as the company location is higher than other cities; Besides, the average fund number of one company is 46.

Equity impact on fund performance model
Model Building: In order to study fund company equity structure's effects of fund performance, we build above model for examination. RAROC i,t =α+β 1 *cehbl i,t +β 2 *flow i,t +β 3 *HHI i +β 4 *W i +β 5 *G i t presents time and i presents fund order RAROC is variable being explained, and cehbl (excess return), flow (capitals flow), HHI (equity concentration), W (fund companies are controlled by state) and G (foreign capital inflow) are explaining variables which can measure degree of relationship between them.
Since we choose the quarterly panel data as analytical sample, the same section funds which belong to same company must have some common characteristics and this will lead to cross sectional heteroscedasticity. We select the Eviews 8.0 as measuring software and we can choose the cross sectional weights at the Weight column, which can reduce the heteroscedasticity influences of cross-sectional data by the use of generalized least squares method (GLS).

Results analysis
The square R is equal to 0.731608, which is bigger than 0.5 and this means the multiple regression is acceptable and proves the explaining ability of the model.
Firstly, we can find that the excessive return ratio has a strong positive correlation with explained variable RAROC. After check the table we can know that when each group's included observations equal to 19, while the t value is bigger than 2.1, so it gets a better explanatory power. However, the t value of this group equal to 9.07 and the p value nearly equal to 0 which means the fitting degree between excessive return ratio and RAROC are very high. Since the excessive return ratio is directly deriving from the fund company's evaluation part of Wind database, high fitting degree means that RAROC's ability of evaluating fund performance is quite strong and feasible.
Secondly, as for the net capital flow, t value is less than 2.1 while the p value is bigger than 0.1 which means that explaining variable has connection with fund performance but does not have enough explanatory power. If only regarding this as the explanatory variable, using the time-changing variable to explain RAROC may lead to a wrong result and this is the reason why we introduce two groups of explanatory variables which may be related to RAROC.
Thirdly, the result of equity structure concentration is nearly accorded with the expecting demand. Although the t value is not as high as excessive return ratio but 2.3 is bigger than 2.1, the results show slight strong explanatory power. Meanwhile, the p value is equal to 0.047, which is smaller than 0.05 and this is also acceptable. The HHI index coefficient in this model is 0.016 which shows a positive correlation and this means that the higher the equity structure concentration is, the better the fund performance is. This kind of result is agreed well with the theoretical analysis above. The reason behind back is that high concentration of equity structure can reduce the phenomenon of free riding. Considering their own benefits and long term popularity, the main shareholders will improve the management level and increase some incentive means. What's more, the concentration of equity structure is also benefit for making decisions and lowering management cost, which is also a way to improve fund performance level.
Fifthly, the result of introducing foreign capitals is the most unexpected part. The t value equal to 3.7 and the p value equal to 0.0002, which both show that the variables have influence on fund performance and the most unexpected result is coefficient, which equal to -0.02. Although we have been discussed in the theoretical part that the introducing of foreign capitals may not applicable at domestic market according to its immaturity and characteristic, but the huge coefficient is still beyond calculation. However, this kind of result shows the current situation as the total amount of fund companies grows every month, the amount of joint venture company have decreased. On the one hand, foreign capitals can't adapt Chinese investment environment. On the other hand, the guidance of government plays a more important role in the current Chinese funds (or mean financial market) than international management experience, since it's still under the adaptive phase.

Model of regional feature's influence on fund performance
Model Establishment In order to find out whether some fund company's features can affect its fund performance or not, we use the above model to conduct the tests. RAROC i,t =α+β 1 *cehbl i,t +β 2 *city i +β 3 *age i +β 4 *size i +β 5 *jjs i + +β 6 *rq i t presents time and i presents fund order Considering that the flow value in the above model can't match the RAROC value very well so that in this model we only select the abnormal return ratio whose fitting degree is high as one of the parameters. Also, we can use the feasible generalized least squares method (GLS) to reduce the heteroscedastic influence of cross-section data.
Empirical Results

Result analysis
The square R is equal to 0.731608, which is bigger than 0.5 and this means the multiple regression is acceptable and proves the explaining ability of the model.
Firstly, regression test results of model show that the fund company's total size and fund number do not have bigger impact on fund performance since t value and p value are both rejected. To analyze the reason, one conjecture is that the fund performance itself has no connection with the fund company's total size. With the increasing of scale expansion, the popularity of industries may rise but it will diffuse more attention at the same time; Another conjecture is that the increasing of fund company's total size is connected with fund number increase, which means the fund performance may be influenced by each fund's average size instead of total size or total number. Secondly, the city brings higher mark. Since the t value is equal to 3 and the p value is equal to 0.0012, the city factor becomes one of the influence factors of fund performance. From this result we can see that one city's financial environment, business's hardware and the concentration degree are all the extra points of fund product's performance.
Thirdly, fund manager's average tenure plays an active role in the whole advancement of fund performance improvement and as we mentioned in the descriptive statistics part, the average tenure of fund manager is two and half a year. Although the t value is exactly equal to 2.1 and p value is between 0.05 and 0.1, the result still didn't show any strong correlation. But the result also showed a positive correlation as the longer tenure the fund manager has, the better they are good at making long term investment and management decisions, which is easier to adapt personal investment style with fund style.
Fourthly, it's found that the establishment date of fund company was insignificantly negative correlated with fund performance. One reason is that Chinese newborn funds may hire some strong management teams to achieve excellent performance in short term, to create Chinese brand and enhance its competitiveness. Another reason is that current old domestic fund company's management fall behind. Although it may have some marketing influential power, what investor identities is the ability of fund management so that the old fund companies should improve the management level by using others' management measures as references to enhance their own ability.

Conclusions and Enlightenments
After the theoretical and empirical study, the research conclusions are summarized as follows.
Firstly, the equity structures do have an impact on the company's fund performance and the effect can be interpreted as an approximate systematic effects. The concentration of equity structures have a positive effect on improving fund performances, and it's beneficial to making decisions, reducing management costs and free-riding behaviors so that fund company can concentrate more on developing business and enhancing performance level.
Secondly, the fund companies whose final largest shareholder is state have higher performance level than non-state owned holding fund companies and this is the manifestation of 'the helping hand of national resources'.
Thirdly, the performances of Chinese-foreign joint venture companies are lower than normal Chinese companies, which mean that foreign capital does not exert expectation effects. The reason behind the phenomenon is that current Chinese funds is in the stage which still under powerful control of government, foreign capital's management concepts advantages are not fully exerted.
Fourthly, as for the related company characteristics, the total fund size and fund number has no effect on fund performance. On the contrary, the fund company's management method and investment philosophy may fall behind the newborn company if it's founded early, which will cause the fund performance level lower than others.
Fifthly, the longer the fund manager's average tenure is, the higher the fund performance level is. Of course, short tenure and high mobility is not helpful in making investment strategy and accumulating experiences.
Sixthly, the performance of fund companies is also affected by financial environments which lead to the obvious fact that the location will have an impact on fund performance. As the finance center of China, Shanghai has the highest performance level and the second is Beijing which is the political center.
The results of this research have certain guideline for general investors, fund industries and regulators. Through the facts which are explored in this paper, general investors can define investment scope and improve investment effects by some visual standards. Some conflicts of equity structure can be dealt with by combination of fund companies' merits and demerits, which can improve control power of joint venture company and make long-term profits together. As for the regulators, the research results can help them improve macroscopic management level, accelerate the localization process of foreign capital, adapt to Chinese investment environment and make Chinese funds meet the international standards.