The Impact of Power Concentration on Corporate
Diversification in Family-based Firms in China
Master Thesis
BOYAN GAO
University of Groningen: s2287560
Email:
gbygary@gmail.com
Supervisor: Dr. D.H.M. (Dirk) Akkermans
Assessor: Dr. Andre van Hoorn
2
Content
Abstract ... 3
Chapter 1.Introduction ... 5
1.1Background and objective ... 5
1.2 Researching content and approach ... 6
1.2.1 Research content ... 6
1.2.2 Research approach ... 6
1.3 Innovative point of research ... 7
Chapter 2 Theoretical Background ... 8
2.1 Family business ... 8
2.1.1 Defining family business in sociology and culture ... 8
2.1.2 Defining family business in ownership and controlling right ... 8
2.1.3 Issues in studying family business in China. ... 9
2.2 Research on Power Distribution in the Family Business ... 10
2.2.1 Defining power in business ... 10
2.2.2 Complexity of family power in business ... 11
2.3 Diversification ... 11
2.4 Summary ... 12
Chapter 3 Hypothesis Development ... 14
3.1 Hypothesis 1 ... 14
3.2 Hypothesis 2 ... 14
3.3 Hypothesis 3 ... 15
Chapter 4 Data Collection and Methodology ... 17
4.1 Data ... 17
4.2 Variables and model designing ... 17
4.2.1 Variables ... 18 4.2.2 Research Designing ... 22 Chapter 5 .Results ... 25 5.1 Empirical Studies ... 25 5.1.1 Descriptive Statistics ... 25 5.1.2 Correlation analysis ... 26 5.1.3 Regression Results ... 27
5.2 Empirical Studies of Consanguinity Concentration ... 29
5.2.1 Descriptive statistics... 30
5.2.2 Logistic regression results of consanguinity concentration ... 31
5.2.3 Results of the logistic model ... 32
Chapter 6 Conclusion and Suggestion ... 35
6.1 Conclusion ... 35
6.2 Suggestions... 36
6.2.1 Suggestions to power distribution in family-based firm ... 36
6.2.2 Suggestions to diversification strategy in family-based firms ... 36
Chapter 7 Limitation ... 38
Abstract
Family business is an important part of Chinese economy; it has made an indelible contribution to the rapid growth of Chinese economy. Chinese family-based firms prefer to choose diversification as a way of financing to expand their operation because of some historical and political reasons, so family-based firm has a higher diversification. Previous studies for the reasons of family-based firms‘ diversification often treat the members of the family as a whole, ignore that different power distribution in the family members will produce different diversification decision. In order to make up for this, this paper uses empirical research to establish contact between family members' power concentration and corporate diversification, thus it concludes the relationship between them. In this paper, the research is divided into two parts: theoretical research and empirical research. In the theoretical part, firstly the paper determines the definition of related concepts by reviewing the related literatures; Secondly it generalizes the present situation of family business diversification by data analysis and case analysis; Finally it classifies and analyzes the diversification reasons from family characteristics' perspective and finds out that most of present researches neglect the family members' power concentration, after that it establishes a theoretical analysis model and provides theoretical support for subsequent empirical research. In the empirical part, this paper collects and processes 146 family–based firms' data through 2008 to 2010, and analyzes the data by descriptive analysis, correlation analysis and regression analysis; finally this paper draws the main conclusion as follows:
Family members' ownership concentration is negatively related to corporate diversification;
Family members' control concentration is negatively related to corporate diversification
4
corporate diversification.
Based on the above conclusion, the author puts forward a series suggestion for family corporate governance. Finally, this paper still has some shortcoming because of the limitation of experience and resource, but believe this will get further improved in the future.
Chapter 1.Introduction
1.1Background and objective
In the process of economy transition in China, The private economy has been one of the most dynamic sectors. The rapid development of the private economy was one of the strong impetuses to boost China's high-speed development. As the carrier of the private economy, the private company is the key to accelerating China's private economy development. Most of private companies in China have the features of the family business. There is a vast amount of studies on private business in China, while some of them overlooked the great influence from family-based characteristics. It will be of great significance to analyze private economy issue in the aspect of family-based features.
6
1.2 Researching content and approach
1.2.1 Research content
This paper puts a lot of attention on how internal power concentration in family members influence business diversification. After browsing the past literatures, this paper found that most researchers look the family members as a whole to study diversification, which overlooked the influence from the internal governance within family. Power concentration, as an important form of power distribution, influences a lot on diversification. So a hypothesis has been made to connect power concentration and diversification on the basis of above. Then a select of listed family firms in China will be a sample to extract related information by reading the annual report and prospectus. Afterwards, the hypothesis connecting the power concentration and diversification will be tested to find out the relationship between them.
Detailed contents of this paper are the following:
Chapter 1: Introduction, in this chapter, the studying background, framework and studying approach will be introduced.
Chapter 2: literature review, in this chapter, concepts of family business and power disposition within family members and the present research on diversification strategy are introduced and summarized.
Chapter 3: research design, in this chapter, the hypothesis is made, and the research approach is introduced
Chapter 4: tests and results, in this chapter, the hypothesis is tested in SPSS, and the results are analyzed
Chapter 5: conclusion and recommendation, some suggestions are raised about the family business governance
1.2.2 Research approach
model. Then some related information is collected by reading annual reports of a selection of family listed firms. After being processing and measuring, the collecting data are tested and analyzed by descriptive statistical analysis, correlation analysis and regression analysis to get the conclusion
1.3 Innovative point of research
There are two main innovative points in this paper. First, the past literatures mostly treat the family members as a whole, which overlooked the influence from internal power distribution to diversification. This paper studies diversification on the view of family's inner circle, combining with the latest family listed firms' annual reports to find out the influence of family internal power concentration to business diversification.
8
Chapter 2 Theoretical Background
2.1 Family business
Family business plays an important role in the world business. There is no common definition on family business in academic circles. The different concepts can be divided into two types in different defining angles, and one is from the aspect of sociology and culture, the other is from the aspect of ownership and controlling right (Xiang, 2005).
2.1.1 Defining family business in sociology and culture
Donnelly (1964) argued that so-called family business must be managed by at least two generation in one family, and both two generations had interactive influence on firm and family. Gallo (1998) argued culture was the core of the family business, and that's to say it is a family business when both family and firm share the same value and prospect. If cultural factors are considered as an ultra-long agreement which is not sensitive to price, then family business is the end of the agreement. Family business is a shared-governance to transaction from both formal agreement and cultural agreement (Yao & Pei, 2003).
2.1.2 Defining family business in ownership and controlling right
indirectly, this business can be called family business.
In this paper, family business can be understood like following: family business is one kind of economic owned and managed by family basing on blood relationship to chasing for biggest family interest. Its essential features include: 1. More than 50% ownership is held in family members; 2. Business is mainly managed by family members; 3.The family business cannot separate from the controlling family; 4. with the development of family and society, socialization of the family business happens gradually.
2.1.3 Issues in studying family business in China.
There are three different attitudes towards the family business: First, some scholars believe modern industry is contradictory to the family business, the greatest obstacle to raising modern corporate system in China is the nepotism breeding in the traditional institution. Family business is one inefficient form of organization (Peng, 2000); Second, family business is one efficient form of organization (Yu, 2002). They believe family members can easily strive for one common goal with high cohesion; Third, Wong (2006) argues so-called family business in China is household business strictly speaking, household business in China is efficient, but family business in China is inefficient.
In recent years, property right structure in big family-based firm drew a lot of attention, many scholars held different attitudes on one-single-large shareholder in one firm. Li (2001) argues ownership right and management right should be separated apart. Liu (2001) believes medium and small shareholders should be protected from the majority shareholder institutionally. Others argue enhancing management performance is the key to family business governance (Wang, 2002).
10
non-family business in the angle of efficiency, at the same time there is no perfect choice on whether the family business is succeeded by founder's children or professional managers. All the need for the founder is to take the best Pareto choice. Li (2000) argues the basic goal for the business founder in China is to secure it instead of maximizing the profit or scale, so that it is easy to understand why family business in China are all transfer the leadership from father to son.
2.2 Research on Power Distribution in the Family Business
2.2.1 Defining power in business
In this paper, power in business includes ownership and control. Zhang & Chen (2004) state ownership consist of two meanings: on the one hand, it is property ownership right, which refers individuals have the right to occupy, use and profit in their property; on the other hand, it is corporate ownership, which refers enterprise's residue and control power.
Ownership and control are reliant on each other. In the period of emphasizing private right, business owner is the controller, which means they are unified in one hand. As the development of technology and enterprise system, separation happened between ownership and control. Essentially, control right is the implementation of ownership. In another word, ownership provides the possibility to profit from investment and control right makes it true.
corporate's operation and strategy (Loss & Sehgrnan, 1982).
2.2.2 Complexity of family power in business
In the past, most research on family business considered family as a whole. However, it is normally not always true that there are exactly same interests and goals in one family. Like Freiman (1985) saying that almost every family business has the conflict of siblings become enemies. In many families, siblings will fight for succession after the founder died. It is of great danger to the family business if the conflict can't be solved considerably. On the other hand, siblings may have a conflict on position and reward within one company, which injured company's operating efficiency a lot. Power deployment in the family business influences a lot on business's operation and performance. Attention has been put in exploring the field of power deployment in the family business. After studying relationship between power deployment and finance, Schulze (2003) found there is an inverted ‗U' pattern relation between power dispersion and external financing. Kellermans (2003) finds out there is a strong relation between corporate performance and power concentration. Eddleston (2007) argues deployment of ownership and controlling influence internal conflict in the family business a lot after studying on ownership succession within the family. Some researchers in China made further investigation on family business in China. Xu and Zheng(2007) point out that most of private companies in China are managed in the hand of founders. Altruism in family will aggravate the agency problem between owner and manager (Wang & Zhou, 2006).
2.3 Diversification
12
concentric diversification, horizontal diversification, vertical diversification and conglomerate diversification. Rajan argues it is related diversification when activities sharing and technology- transferring happened. There are different views between diversification and corporate value. One is diversification disperses the operation risk and helps build an effective internal finance market to break the limit of external finance market, in order to get economic efficiency of scale. Some researchers argue that diversification disperses limited corporate resource into different business, which will cause internal managing problem. At the same time, the internal finance market may bring excessive or insufficient investment. Diversification itself has no such a problem of right or not, and it is the key that corporate makes good use of different internal and external conditions.
Diversification is one important business strategy. Four influencing factors including the institutional environment, financial characteristics, industry factors, and corporate governance are studied. Corporate governance, as a controllable factor, is the research focus. Amihud & Lev(1996) found enterprise with dispersive equity right preferred diversification more than enterprise with concentrated equity right. Yu (2007) states it is a U-shape relationship between share ratio in managers and level of diversification. Among the companies with separating ownership right and controlling right, managers have motivation to gain private benefit through diversified investment. When managers have a large proportion of share, they tend to use diversification to disperse personal risk while companies with majority shareholders do less irrelevant acquisition activities (Amihud, 1981).
2.4 Summary
diversity of corporate play a moderating role in power concentration within family (Yin & Zhou, 2010). A relative concentrate power within the family business improves corporate performance.
As a typical reflection of internal power structure within the family business, level of power concentration plays a significant role in corporate performance. Diversification, as an important business strategy, will be influenced by corporate governance in the family business inevitably. It has been found that besides corporate governance, characteristics of top management team and demographic factors of top leaders influence the diversification a lot. That is to say within a family-based enterprise, power structure will influence the diversification strategy inevitably. In other word, level of power concentration in the family business will influence corporate diversification
14
Chapter 3 Hypothesis Development
3.1 Hypothesis 1
Equity structure plays a significant role in corporate decision-making. Leech &Leahy (1991) argues level of power concentration has a negative relationship with corporate value. After studying influence of ownership concentration to corporate performance in China's listed firms, Xu found there is a significant positive linear relationship between them, and it exists among different types of shareholders (Xu, 2006). Researchers consider the family as a whole when studying family business, which overlooks the influence of internal equity structure of the family. Family members are not all altruists. There are different interests and value among different family members. Some members may use the information asymmetry to acquire benefit for themselves.
There are three types of equity structure in family-based firm: 1. concentrating in one founder's hand; 2. Dispersed in lineal relatives; 3.Dispersed in lineal and collateral relatives. This paper argues that the equity concentrating on a few family members will bring a series of problems. Because of its particularity, it is difficult to solve agency problems within family members through external market in family based firm. One possible way is to disperse equity relatively to other family members. Owning some equity in hand inspires family members working hard and sharing more resource to help firm develop. At the same time, dispersing equity within family members solves agency cost problem to reduce internal conflicts to some extent. Based on the above analysis, the first hypothesis comes up:
H1: There is a negative relationship between ownership concentration and level of
diversification in family-based firms in China.
Ownership and control are reliant on each other. Control right is the implementation of ownership. In another word, ownership provides the possibility to profit from investment and control right makes it true. Equally, the second hypothesis comes up:
H2: There is a negative relationship between controlling right concentration and the
level of diversification in family-based firms in China.
3.3 Hypothesis 3
In the past researches on internal governance in the family business, family members are put in equal place when facing agency conflicts and problems to analyze the agency relationship in the whole family. Not only family members and non-family members have conflicts in family-based firm, but core family members and non-core family members have conflicts. These conflicts will cause distrust among family members (Wong, 1988). This paper argues that assuming a same goal among family members simply is less rigorous. There are both convergence and difference among family members.
With the development of the family business, distant relatives join in business increasingly, as part of agents, not only limit in core family members. He (2008) divides family members of family-based firms into three types according to social learning theory. Core family member's type including founder's parents, children, brothers and sisters, non-core family member's type and compound type including both core family member and non-core family member. This paper uses it as a basis of phylogenetic concentration. In the above three types, core family type has the highest level of consanguinity concentration, and non-core family type has the least.
16
bad decision to the family business on behalf of their interests. Compound type is very complex in internal agency problem because agency problem in compound family type includes three different aspects, namely, problem in inter-core family members, problem in inter-non-core family members and problem in the core and non-core family members. This paper argues that different types of family members have different phylogenetic concentration, and their influence on family business will be different as well. So the third hypothesis comes up:
H3: There is a significant relationship between phylogenetic concentration and diversification in the family business in China.
Chapter 4 Data Collection and Methodology
4.1 Data
The sample includes the family-based firms listed on Shanghai and Shenzhen A-share Stock Exchange in 2011-2013. Though Chapter 2 has given a definition about family business, more quantified concept are needed. This paper uses Eddleston (2007)‘s definition:
1. Firm is owned by family members;
2. There are at least two family members in charge of the firm.
Basic information about China‘s private firms is downloaded from CSMAR database. Then annual reports of firms are found out according to stock code in Baidu and Sina. At last, three years‘ data of 146 selected firms in 2011-2013 are studied. Firms are excluded by the following principles:
Financial listed company. Because financial listed company usually has a different balance sheet structure significant. Scholars often exclude it.
Special Treatment Company. Their annual reports are not published in Stock exchange website.
Company related information is not fully available. Mainly there are three ways to collection data:
Downloaded from CSMAR (China Stock Market Accounting Research) database, including some control variables such as Asset Size, Liability Scale and ROA; Annual reports of sample companies downloaded from Shanghai Stock Exchange
website and Shenzhen Stock Exchange website, including majority shareholder, information about top management team and income;
Searching for information of the relationship among shareholders and managers through Google and Baidu to prepare for measuring power concentration.
18
4.2.1 Variables
Degree of Diversification (DEVI)
Level of diversification is one of the most important dimensions to measure corporate diversification. Ideal diversification should meet five conditions: 1.level of diversification is positively related to types of products; 2.level of diversification is negatively related to the degree of output uniformity; 3. Level of diversification is positively related to the difference of products. This paper uses Theil‘s entropy to measure the level of diversification in one corporate. Jacquemin & Berry (1979) applies Entropy of thermodynamics in to measure the level of corporate diversification. The value of entropy is positively related to the level of diversification. DEVI= i 1 N Pi
ln(1 𝑃𝑖)N is the number of business unit, Pi =business unit income/total income
Ownership Concentration level (VRD)
Ownership concentration level within family members can be measured by voting right. Since family-based firm is not totally owned by family members, in order to measure ownership concentration level, the voting right of family members should be a relative ratio. First, the voting right of the whole family in one firm is worked out. And then each family member‘s voting right can be worked out according to their shareholding ratio respectively. Then the ownership concentration level can be working out with the help of Herfindahl-Hirschman Index.
VRD=∑𝑁𝑖=1(𝑉𝑅𝑖)2 = ∑𝑁𝑖=1(𝑠ℎ𝑖/ ∑𝑁𝑖=1𝑠ℎ𝑖)2
N is the number of family members with voting right, VRi is the relative ratio of voting right of family members, shi is voting right of each family member.
Controlling Concentration level(MPD)
different positions. On the basis of reliability analysis and validity analysis, the average value of each position is worked out. The coefficient of each position is the value of the average value dividing the largest measured value. With the help of HHI, controlling concentration level can be working out in the following formula:
MPD=∑𝑁𝑖=1(𝑚𝑝𝑖)2 = ∑𝑁𝑖=1(𝑝𝑜𝑖⁄∑𝑁𝑖=1𝑝𝑜𝑖)2
N is the number of family member with controlling right, mpi is the relative ratio of controlling right of family members, poi is controllinging right of each family member.
Table 4-1 coefficient of different positions in firm
Position Coefficient Position Coefficient
Chairman with CEO 0.982 Vice-CEO 0.740 Chairman with other position 0.936 Financial Director 0.746
CEO with other position
0.905 Market Director 0.753
Vice-chairman with vice CEO
0.892 HR Director 0.656
Chairman 0.930 Secretary of CEO 0.649
Vice-Chairman 0.809
Director 0.706
CEO 0.832
Return on Equity(ROE)
Return on equity measures a corporation's profitability by revealing how much profit a company generates with the money shareholders have invested. The higher ROE, the more profit can obtain from the investment.
20
In the daily operation of a firm, corporate diversification will influence ROE inevitably. Corporate with low profitability prefers diversified business to get rid of the terrible situation. ROE is one of most important control variables in studying corporate diversification.
Scale of firm (SIZE)
Firm scale will influence the level of corporate diversification directly. Many empirics show there is a positive relationship between firm size and assets liability ratio. Comparing to small firm, big firm has more stable cash flow if other conditions are the same. More cash flow brings less operation risk of diversification. This paper uses the number of staff to evaluate firm scale.
Capital Structure(LEVEL)
Capital structure of one firm is usually measured by debt ratio. Debt Ratio is a financial ratio that indicates the percentage of a company's assets that are provided via debt. It is the ratio of total debt (the sum of current liabilities and long-term liabilities) and total assets. When there is a high degree of debt ratio, there will be a negative influence on corporate diversification.
Whether chairman of board and CEO in one firm is the same(DIRE)
Whether the corporate power is distributed dispersedly, one factor is to see whether the chairman of board and CEO in one firm is the same person. In general, firm with dispersed power concentration is more appropriate to operated diversified business because there are more managers in the firm. This paper assigns firm with the same person in Chairman of board and CEO to 1 and different persons to value 0
Industry factor(IND)
Firm in different industries will have different influence on corporate diversification. Many family-based firms in China are manufacture business, so in this paper, manufacturing firms of the sample are assigned 1 and non-manufacturing firms are assigned 0.
Family controlling ratio(SHARF)
members in one firm with a high level of family controlling ratio. This paper uses voting right to measure it.
Types of family-based firms(TYPE)
With different types of family members controlling in one firm, family-based firms can be classified into three types. Those are core family type, non-core family type and compound type. In the research of this paper, TYPE=0 represents firm controlled by core family members; TYPE=1 represents firm controlled by non-core family members; TYPE=2 represents firm controlled by compound type.
Table 4-2 Variables summarized in one table
Variables Sign Description
Level of Corporate Diversification DEVI DEVI= i 1 N Pi
ln(1 𝑃𝑖) The more DEVI is, the higher level of corporate diversification will be.Level of
Ownership Concentration
VRD VRD= ∑𝑁𝑖=1(𝑉𝑅𝑖)2 =
∑𝑁𝑖=1(𝑠ℎ𝑖/ ∑𝑁𝑖=1𝑠ℎ𝑖)2 The more value of VRD is, the higher level of ownership concentration will be. Level of Controlling Concentration MPD MPD= ∑𝑁𝑖=1(𝑚𝑝𝑖)2 = ∑ (𝑝𝑜𝑖 ∑𝑁 𝑝𝑜𝑖 𝑖=1 ⁄ )2 𝑁 𝑖=1
The more value of VRD is, the higher level of controlling concentration will be.
Return on Equity ROE Measure of firm
22
ROE=Net income/
Average equity
Firm Scale SIZE Measured by number of
staff
SIZE= number of staff
Capital Structure LEVEL Measured by Debt Ratio
LEVEL=DEBT/EQUITY Chairman with
CEO or not
DIRE Dummy variable
Value=1 when the same Value=0 when not
Industry IND Dummy variable
Value=1 when
manufacturing business Value=0 when not Strength of family
control
SHARF Measured that who share
ratio in one firm.
Family Type TYPE Classify family-based
firms into 3 types
TYPE=0, core family type TYPE=1, non-core family type
TYPE=2, compound type
4.2.2 Research Designing
controlling right concentration level as independent variables ,and other control variables to build three multiple linear regression models.
To test how does ownership concentration influence corporate diversification(4.1) DEVI=β0+β1VRD+β2ROE+β3SIZE+β4LEVEL+β5DIRE+β6IND+ε
β0 is constant, β1 –β6 is coefficient of each control variable respectively, ε is the random error.
To test how does controlling concentration influence corporate diversification(4.2) DEVI=β0+β1MPD+β2ROE+β3SIZE+β4LEVEL+β5DIRE+β6IND+ε
β0 is constant, β1 –β6 is coefficient of each control variable respectively, ε is the random error
To test how does ownership concentration and controlling concentration influence corporate diversification(4.3)
DEVI=β0+β1VRD+β2MPD+β3ROE+β4SIZE+β5LEVEL+β6DIRE+β7IND+ε
β0 is constant, β1 –β7 is coefficient of each control variable respectively, ε is the random error
In the above models, DEVI is the level of corporate diversification in family-based firms in China; VRD is the level of ownership concentration; MPD is the level of controlling concentration; ROE is the asset earning rate; SIZE is the scale of firm; LEVEL is the capital structure of firm; DIRE reflects whether the is Chairman of Board and CEO the same; IND symbolizes whether the firm is manufacturing firm. Logic regression model of consanguinity concentration
This paper uses logic regression model to test whether there is a significant correlation between the degree of consanguinity concentration and corporate diversification. In this paper, three types of the family member represent different degree of consanguinity concentration. TYPE=0 represents core family type; TYPE=1 represents non-core family type; TYPE=2 represents compound type. The mode is like following(4.4):
ln𝑃𝑖
𝑃𝑗=β0+β1DEVI+β2ROE+β3SIZE+β4LEVEL+β5DIRE+β6SHARF+ε
24
Chapter 5 .Results
5.1 Empirical Studies
5.1.1 Descriptive Statistics
This paper carries out a descriptive statistics analysis on each variable and their minimum values, the maximum values, means, and standard deviations are in the following table. From the analysis we can see:
The minimum and maximum values of corporate diversification are 0.00 and 1.98 respectively, which shows that there is a great difference in level of diversification among different family-based firms. The average value of corporate diversification is 0.55, showing that the sample firms are in a moderate level of diversification.
Comparing the mean of the degree of ownership concentration (VRD=0.70) and mean of the degree of controlling concentration (MPD=0.55), this paper find out ownership is more concentrated than controlling in family-based firms.
Values of ROE, SIZE, LARGE and SHARF are relatively large in Standard Deviation, which shows that sample firms have large difference in these four aspects. According to the descriptive statistics, mean of LARGE equals 30.28% shows equity in sample firms is relatively concentrated, and Mean of SHARF equaling 40.95 proves that family members have dominant controlling advantage in family-based firms in China. Besides, mean of DIRE is 0.25 and Standard Deviation of DIRE is 0.43 shows that most chairman of the board in family-based firm are not CEO as well. Mean of IND is 0.71 shows most family-based firms in China are manufacturing business.
Table 5-1
variables Minimum Maximum Mean Std.
Deviation
Number of sample
26 VRD 0,12 1.00 0.70 0.24 146 MPD 0.17 1.00 0.55 0.26 146 ROE -51.45 395.32 13.93 27.49 146 SIZE 13.00 49040.00 2838.43 4764.51 146 LEVEL 0.06 1.98 0.48 0.25 146 DIRE 0.00 1.00 0.25 0.43 146 IND 0.00 1.00 0.71 0.45 146 LARGE 2.60 78.18 30.28 14.75 146 SHARF 2.60 100.00 40.95 18.80 146 PYRAM 0.00 1.00 0.71 0.45 146 5.1.2 Correlation analysis
Multicollinearity is a statistical phenomenon in which two or more predictor variables in a multiple regression model are highly correlated. It produces large standard errors in the related independent variables, which decrease the significance of T value. To test whether do variables in model 4.1-4.3 have multicollinearity, this paper carries Pearson correlation analysis to variables in the three models. The results can be found in Table 5-2
Table 5-2 Correlation matrix Variable
s
DEVI VRD MPD ROE SIZE LEVEL IND DIR
ROE -0.066 (0.171) 0.015 (0.763) 0.045 (0.353) 1 SIZE 0.079 (0.103) 0.097* * (0,044) -0.031 (0.517) 0.037 (0.441 ) 1 LEVEL 0.034 (0.479) 0.034 (0.475) 0.154** (0.001) -0.026 (0.585 ) 0.053 (0.268) 1 IND -0.006 (0.906) -0.057 (0.239) -0.094* * (0.049) 0.033 (0.486 ) 0.121* * (0.012) -0.054 (0.264) 1 DIRE 0.066 (0.169) 0.012 (0.796) 0.068 (0.153) 0.029 (0.542 ) -0.030 (0.531) -0.199* * (0.012) 0.031 (0.521 ) 1
Source: SPSS 13.0, *, **, *** represents 10%, 5%, and 1% significant level.
Table 5-2 shows the correlation between different variables. The correlation coefficients of variables are all less than 0.4, so that it can be seen there is no multicollinearity among the variables in the three models. From table 5-2, the coefficient of ownership concentration level (-0.308) and controlling concentration level (-0.269) are all less than 0 and are significant under 1% level, which preliminary proves H1 and H2.
5.1.3 Regression Results
The results of regressions of whole sample firms are presented in Table 5-3.
28
concentration level (VRD) and corporate diversification level (DEVI) in family-based firms in China (H1). From model 2 in the table 5-3, coefficient of controlling concentration level (MPD) is -0.537 at 1% significant level, proving that there is a negative correlation between controlling concentration level (MPD) and corporate diversification level (DEVI) in family-based firms in China (H2). When VRD and MPD are substituted into model 3, the coefficient of VRD and MPD are -0.589 and -0.477 respectively at 1% significant level, proving that power concentration level is negatively correlated with corporate diversification level in family-based firm in China.
Table 5-3 Regression Results D variable s In-D variable s
DEVI DEVI DEVI
Model 1 Model 2 Model 3
IND -0.052 (-1.003) 1.025 -0.058 (-1.140) 1.029 -0.074 (-1.534) 1.027 DIRE 0.093* (1.808) 1.017 0.115* (2.212) 1.205 0.120** (2.403) 1.175 F 9.455*** 7.545*** 13.138*** R2 0.106 0.084 0.166 Durbin– Watson statistic 1.768 1.894 1.816
Source: SPSS 13.0, *, **, *** represents 10%, 5%, and 1% significant level The F-statistics in three models are all significant at 1% level, indicating that these three models are significant and successful generally. F-statistics of three models are all at 1% significant level. D-W test values of three models are 1.768, 1.894, and 1.816, showing that stochastic errors in the three models have no autocorrelation (Gujarati, 1978). Besides, when value of Variance Inflation Factor (VIF) in one model is more than 10, there is a great multicollinearity among independent variables, which may influence the least square value a lot. In table 5-3 we can easily found VIFs of three models are all less than 10, indicating that there is no multicollinearity in these three models.
According to results in table 5-3, control variables such as firm scale (SIZE), capital structure (LEVEL), and DIRE are significantly correlated with corporate diversification level. Coefficients of capital structure (LEVEL) are always more than 0, 0.086, 0.164 and 0.164 respectively. This shows that to some extent debt is positively correlated with corporate diversification level, namely more debt in one firm brings more diversification.
30
5.2.1 Descriptive statistics
Table5-4 Distribution of different family types
TYPE Number of sample Percentage (%)
0 (core family) 274 62.56
1 (non-core family) 23 5.25
2 (compound family) 141 32.19
TOTAL 438 100.00
Source: baidu.com and annual reports of firms
Table 5-5 corporate diversification level in different family types in family-based firms YEAR TPYE=0(CORE FAMILY) TYPE=1(NON-CORE FAMILY) TYPE=2(COMPOUND FAMILY) 2011 MEAN 0.4868 0.2099 0.6902 Std D 0.4282 0.2819 0.5426 2012 MEAN 0.5365 0.1917 0.6188 Std D 0.4596 0.1991 0.5389 2013 MEAN 0.5535 0.0987 0.6565 Std D 0.4489 0.1338 0.5951
From the results of Table 5-4 and Table 5-5,
Most family-based firms are core family type (62.56%); the least is non-core family type at a ratio of 5.25%; non-core family type stays in the middle.
Means of corporate diversification level in 20111-2013 show that among three types of firms, compound family type has the highest level of corporate diversification, followed by core family type, and non-core family has the least level.
core family type increased from 0.4868 (year 2011) to 0.5535 (year 2013). Diversification level in non-core family type decreased from 2011 to 2013 but not significantly.
5.2.2 Logistic regression results of consanguinity concentration
To test whether there is multicollinearity in the logistic model, this paper carries one Pearson coefficient test on the variables. Results are in the following table 5-6. In the table 5-6, all Pearson coefficients of variables are less than 0.4, there is no multicollinearity among variables. So that it is safe to substitute them into Logistic model. From the statistics results, corporate diversification level (DEVI) is correlated with family type at 5% significant level, which indirectly proves that there is a significant correlation between corporate diversification level and consanguinity concentration (H3). Besides, firm scale (SIZE), capital structure (LVEL) and SHARF are also correlated to family type in family-based firms, and their significance are 0.052, 0.023 and 0.009 respectively.
Table 5.6 correlation analysis on each variable in Logistic model
Variables TYPE DEVI ROE SIZE LEVEL DIRE SHARF
32 SHARF 0.124*** (0.009) 0.007 (0.880) 0.119** (0.013) -0.173*** (0.000) -0.071 (0,141) 0.072 (0.134) 1
Source: SPSS 13.0 results; *, **, *** represent 10%, 5%, 1% significant level.
5.2.3 Results of the logistic model
Likelihood and Wald statistics are usually used to test Logistic regression model. Likelihood functions play a key role in statistical inference, especially methods of estimating a parameter from a set of statistics. Since likelihood function value is usually a very little decimal, it is convenient to use it after applying natural logarithm and times -2(-2LL). -2LL is usually used to reflect the degree of fitting in one model. The more -2LL value, the less degree of fitting in that model.
Table 5-7, table 5-8 and table 5-9 document the results of different family types correlated with corporate diversification level. In the table, likelihood value is significant at 1% level and value of goodness fit in this model (Nagelkerke R2) is 14.2%, indicating that this model has a higher forecast accuracy and better goodness fitting.
diversification in family-based firms in China according the analysis above (H3).
Table 5-7 Logistic regression results (using TPYE=2 as reference) In 𝑃𝑃0
2
variables Coefficient Wald Sig.
Intercept 0.679 2.532 0.112 DEVI -0.582*** 7.040 0.008 ROE 0.010 1.989 0.158 SIZE 0.000* 3.183 0.074 LEVEL 1.066** 5.099 0.024 DIRE -0.014** 5.808 0.016 SHARF 0.138 0.313 0.576 Likelihood test 648.270*** Chi-Square (sig.=0.000) Nagelkerke R2 14.2%
*, **, *** represent 10%, 5%, and 1% significant level respectively Table 5-8
In 𝑃𝑃1
2
variables Coefficient Wald Sig.
34
Nagelkerke R2 14.2%
*, **, *** represent 10%, 5%, and 1% significant level respectively
Table 5-9 In 𝑃1
𝑃0
variables Coefficient Wald Sig.
Intercept -1.148 2.476 0.116 DEVI -2.908*** 10.241 0.001 ROE 0.004 0.807 0.369 SIZE 0.000* 3.197 0.074 LEVEL -0.318 0.075 0.784 DIRE 0.005 0.118 0.731 SHARF 0.690 1.946 0.163 Likelihood test 648.270*** Chi-Square (sig.=0.000) Nagelkerke R2 14.2%
*, **, *** represent 10%, 5%, and 1% significant level respectively
Chapter 6 Conclusion and Suggestion
6.1 Conclusion
This paper studies how does the level of power concentration influence level of corporate diversification combining theory and practice in China. This paper comes up three hypotheses and carries empirical research on 146 sample firms through descriptive analysis, correlation analysis, multiple regression analysis and three-type logistic regression analysis. Some conclusions are drawn in the following:
First, there is large difference among listed firms in China, while level of corporate diversification in family-based firms is higher than other types firms averagely. On the one hand, leaders in family-based firms have a very strong enterprise; on the other hand, under the special institution in China, corporate diversification is one of limited ways to access finance.
Second, averagely, level of ownership concentration is higher than level of controlling concentration. It shows that in family-based firms, distribution of ownership right is more cautious than controlling right. It is easy to see the preference of leaders in family-based firm in power selection.
Third, there is a negative correlation between level of ownership concentration and level of corporate diversification in family-based firm. It shows relative dispersed ownership makes family members build convergent goal and help family-based firm grow more.
Four, there is a negative correlation between level of controlling concentration and level of corporate diversification in family-based firm. Like dispersed ownership right, relative dispersed controlling right inspires family members to participate decision-making more objectively, increasing executing efficiency invisibly.
36
(TYPE=2). Different types of family-based firms have different levels of corporate diversification. From the researching results, compound type has a higher level of core family type and non-core family type, proving that different levels of consanguinity concentration in family-based firms have different levels of corporate diversification.
6.2 Suggestions
6.2.1 Suggestions to power distribution in family-based firm
Power distribution among family members has a significant influence on the family business. Family members are one unity of the whole to family business. Capacity, resource, personal willingness of each family member should be considered into when processing power distribution to promote the common interest of family.
Conflict will occur inevitably during power distribution in family-based firm. Conflict is one of outstanding characteristics of the family business. It is necessary to build conflict processing mechanism to prevent from negative governance.
At present, corporate governance in family-based firms is not mature in China. Board of directors work not well independently, and independent directors cannot play a supervision role well. A lot of improvements need completing in family-based firms in China. One way is introducing external shareholders such as external funds and private equity. Other ways like share ownership encouragement and introducing professional managers also can make internal governance of family-based firms better.
6.2.2 Suggestions to diversification strategy in family-based firms
a diversified operation. One reason is that the ways of finance of family-based firm in China are very limited; the leaders often go on a diversified strategy to make good use of the limited financial resource. Other reason is that leaders of family-based firm in China usually enter into a new sector blindly after one success. There is a great risk in corporate diversification. When the firm chooses diversification strategy, the leader need to realize the core business of the firm is the foundation of expansion. According the former research, the higher level of business concentration a firm has, the better performance a firm has, indicating that stable core business is the profit source and foundation of diversified expansion. Entering into an unfamiliar area without necessary information and skill is full of risk especially at the beginning stage; more stable management power and financial resource are needed.
38
Chapter 7 Limitation
Reference
Amihud, B Lev. Risk Reduction as a Managerial Motive for Conglomerate Mergers,
The Bell Journal of Economics, 1981,9 (12) : 371-392
Anderson,Lemmon,M.Corporate Governance and Firm Diversification, Financial
Management, 2000, 3 (29) : 5-22
Anderson, R.C.,Reeb, D. Founding Family Ownership and Firm Performance: Evidence from the S&P 500,Journal of Finance, 2003,2 (58) : 1301-1328
Ananda Mukheiji. The Relationship Between Prior Performance and Diversification: A Study of Three Industries, 33rd Meeting of the Easton, Academic of
Management,1996,5 (13) : 91-104
Ansoff P. Corporate Strategy: An Analytic Approach to Business Policy for Expansion, Harvard Business School, 1965, 7 (5): 99-110
Berry, C.H. Corporate Growth and Diversification, Princeton University Press, 1975,8 (4) : 374-392
Eddleston K.A., Kellermanns F.W. Destructive and Productive Family Relationships: A Stewardship Theory Perspective, Journal of Business Venturing, 2007, 22 (4) : 545-565
Fauver.L,Houston.J, Naranjo.A, Capital Market Development, International Integration,Legal Systems,and the Value of Corporate Diversification: A Cross-country Analysis, Journal of Financial and Quantitative Analysis, 2003, 7 (38) : 135-1570
F.M. Gollop,J.L. Monahan. A Generalized Index of Diversification; Trends in US Manufacturing,The Review of Economics and Statistics,1991,41 (5): 385-403
Gallo.M.A. A Financial Perspective on Structure, Conduct, and Performance in the Family Firms: an Empirical Study, Family Business Review, 1998, 8 (11) : 35-47
40
Hanbrick,D.C,Mason, P. Upper Echelons: The Organization As a Reflection of Its Top Managers, Academy of Management Journal, 1984,2 (14) : 401-418
Hambrick, D.C.' Theresa, C.' Ming-Jer.The Influence of Top Management Team Heterogeneity on Firm's Competitive Moves, Administrative Science Quarterly 1996, 41 (4): 659-684
Jensen, M. Agency Costs of Free Cash Flow,Corporate Finance and Takeover,
American Economic Review, 1986,8 (76) : 323-329
Jacquemin A.P,C.H. Berry. Entropy Measures of Diversification and Corporate Growth, Journal of Industrial Economics, 1979,2 (27) : 359-369
JoaoGomes. Dmitry Livdan Optimal diversification: reconciling theory and evidence,Journal of Finance, 2004,9(1): 507-535
Kellermanns F.W, Eddleston K.A. A Family Perspective on When Conflict Benefits Family Firm Performance, Journal of Business Research, 2007, 60 (10): 1048-1057 Khanna, T.,K. PalePu. Is Group Affiliation Profitable in Emerging Markets?—An Analysis of Diversified Indian Business Groups, Journal of Finance, 2000,7(55) : 867-891
Khanna.T.,K.PalePu. Why Focused Strategies May Be Wrong for Emerging Markets,
Harvard Business Review, 1997, 5 (75) : 41-49
Lang,L.H.P.,R.M. Stulz. Tobin's Q,Corporate Diversification and Firm Performance, Journal of Political Economy, 1994,12 (102) : 1248-1280
Leech.D,Leahy J. Ownership Structure, Control Type Classifications and the Performance of Large British Companies, The Economic Journal, 1991, 7 (3): 38-49 Leonard .H.Freidman. Generation to Generation, New York; Guilford, 1985, 4 (7): 90-107
Michael C.J. The Theory of Firm: Managerial Behavior, Agency Costs and Ownership Structure, Journal of Financial Economics, 1976, 12 (3): 305-360
Monigomery C. Corporate Diversification, Journal of Economic Perspectives, 1994, 9 (8): 163-178
Penrose Edith.T, The Theory of the Growth of the Firm, Academy of
Management,1959,9 (25) : 72-93
Peredo,A.M. Emerging Strategies Against Poverty: The Road Less Traveled, Journal
of Management Inquiry, 2003,12 (2) : 155-166
Phillip.I. Blumberg. Corporate Groups, Wolters Kluwer Law & Business, 1983,(6) : 10-19
P.Raj an Varadarajan,Vasudevan Ramanujam. Diversification and Firm Performance: a Reexamination Using a New Two-dimensional Conceptualization of Diversity in Firms, Academy of Management, 1987, 3 (30) : 380-393
Robert G. Donnelly. The Family Business, Harvard Business Review, 1964, 42 (4): 93-105,
Schulze W.S., Lubatkin M.H., Dino, R.N. Exploring the Agency Consequences of Ownership Dispersion among the Directors of Private Family Firms, Academy of
Management Journal, 2003,46 (2) : 179-194
Shleifer A., R. Vishny. Management Entrenchment: The Case of Manager-Specific Investment, Journal of Financial Economics, 1989,9 (5): 123-139
Saman Majd,Stewart C. Myers. Tax Asymmetries and Corporate Income Tax Reform,
University of Chicago Press,1983,7 (9) : 343-375
Stein,J. Internal Capital Markets and the Competition for Corporate Resources,
Journal of Finance, 1997,8 (52) : 111-133
Teece.D. Towards an Economic Theory of the Multi-product Firm, Journal of
Economic Behavior and Organization, 1982, 6 (3): 39-63
Wernerfelt B., C. Montgomery. TobirTs q and the Importance of Focus in Firm Performance, American Economic Review, 1988,12 (78) : 247-250
Whiteside M, Aronoff C E, Ward J L. How Families Work Together, Family
Enterprise Publishers, 1993,8 (4) : 92-113
Wilbur G. Lewellen. The Ownership Income of Management, National Bureau of
Economic Research,1971,5 (8) : 69-81
42
Wrigley.Divisional Autonomy and Diversification, Unpublished Doctoral Dissertation, Graduate School of Business Administration, Harvard Business School, 1970,9 (2) : 89-103
Wong, S.L. Emigrant Entrepreneurs: Shanghai Industrialists in Hong Kong, Oxford
