Ownership Structure and Firm Performance
Evidence from the Chinese I.T. and Retail Industry
Thesis
Master of Science in Business Administration
Specialisation: Finance
University of Groningen
Faculty of Economics and Business
Author: Yidi Zhao
Abstract
In this paper, we aim to find the relationship between ownership structure and firm performance of listed firms from the I.T. industry and the retail industry on the Chinese stock market. We divide ownership structure into three dimensions: ownership concentration, managerial ownership and owner identity. We use both return on assets and market-to-book value of equity as proxies for performance and apply both OLS method and panel techniques to test the hypotheses. We find that ownership concentration has a significantly positive impact on performance and varies from the I.T. industry and the retail industry. We also find government ownership has a significantly negative impact on firm performance and institutional ownership has a significantly positive influence on performance. In addition, we find the relation between ownership concentration and performance is an inverted U-shape after the crisis and does not alter in these industries; while the optimal concentration level for I.T firms is 34.13% and is 43.11% for retail firms.
JEL classification: G01, G30
Keywords: ownership structure, ownership concentration, managerial ownership,
Table of content
1. Introduction ...4 2. Literature Review ...9 2.1 Theoretical Framework ...9 2.1.1 Corporate Governance ...9 2.1.2 Agency Theory ...9 2.2 Ownership Concentration ... 11 2.2.1 Two Hypotheses ... 11 2.2.2 Industry Effect ... 11 2.3 Managerial Ownership ... 12 2.4 Owner Identity ... 132.5 Share Structure in China... 14
2.6 Empirical Results ... 15
2.6.1 Ownership Concentration ... 14
2.6.2 Managerial Ownership... 17
2.6.3 Owner Identity ... 19
3. Hypotheses and Methodology ... 20
3.1 Hypotheses... 20
3.2 Equations and Measurements ... 22
3.3 Variables ... 23
4. Data and Descriptive Statistics... 28
1. Introduction
The study on the relationship between ownership structure and corporate performance has played a crucial role in the corporate finance research area for a long time. The origin of modern corporate governance theory could be traced back to Berle and Means (1932). They address the problems caused by separation of ownership and control in a typical modern corporation. In this situation, agency problems take place. Managers might seek value maximization of their own interests instead of shareholders’ interests, or even worse, at the expense of shareholders’ profits. On the other hand, it is acknowledged that managers possess exceptional expertise in maintaining a firm’s daily operation. The firm could perform better and more efficient so that shareholders could benefit. Jensen and Meckling (1976) define the former effect as the entrenchment effect and the latter one as the convergence-of-interest effect. Approximately fifty years after Berle and Means, Demsetz (1983, p.377) proposes that “ownership structure of the firm is an endogenous outcome of a maximizing process in which more is at stake than just accommodating to the shirking problems.” In fact, there are some studies support the endogeneity of ownership structure on firm performance (Himmelberg et al., 1999; Demsetz and Villalonga, 2001), while others ignore the endogeneity issue and find a certain relationship between ownership structure and firm performance (Morck et al., 1988; Han and Suk, 1998; Miguel et al., 2004).
performance. Different types of the largest shareholder may have various objectives and exercise their power in ways that would influence corporate strategy like dividend policy, capital structure choice etc. (McConnell and Servaes, 1990; Short, 1994). Therefore, the effect of owner identity on firm performance is of interest to study.
Moreover, corporate governance is not only a topic attractive to economists but also a hot issue to politicians. During the Asia financial crisis, tunneling appears to be a much more severe problem in emerging markets (Johnson et al., 2000b). Moreover, the scandal of Enron, WorldCom and Parmalat made corporate governance high on the policy agenda since the early 2000s. The US government even issued the Sarbanes Oxley Act of 2002, which makes explicit the responsibility of management regarding the quality of financial reporting, after the collapse of Enron and WorldCom Empire. In addition, the fall of Lehman Brothers in 2008 could, to a large extent, be attributed to failures and weaknesses in corporate governance mechanisms. In this paper, not only the relationship between ownership concentration and corporate performance, managerial ownership and performance, but also the effects of the identity of the largest shareholder on performance will be studied.
corporations operate more efficiently and effectively. However, owing to this, Chinese corporations suffer more from the recent global financial crisis than they did from the Asia financial crisis.
There are two stock exchanges in mainland China: Shanghai Stock Exchange (SHSE for short) and Shenzhen Stock Exchange (SZSE for short). SHSE was founded in 1990 and SZSE was established in 1991. The Chinese stock market has made rapid development and has become the fourth largest financial market in the world although it is still immature. The famous Yin Guang Sha1 event in 2001 was an astonishing lesson taught by the Chinese imperfect market mechanism. Accountants of Yin Guang Sha made false financial reporting and painted a prosperous picture about the firm. After this scandal was discovered by the public, the overall firm value shrank to one-fourth of its highest value and it prevented individual investors from investing in Chinese stock markets. Alongside with this, the insider transaction of Yan Bian Gong Lu2 by a securities trader, and the recently released scandals of pharmacy firms listed on SHSE and SZSE alarm that the corporate governance issue deserve attention from the Chinese government.
The general research question in this paper is: What is the relationship between ownership structure and corporate performance of listed firms in China? More specifically, five sub questions are: 1) What is the impact of ownership concentration on corporate performance? 2) Does the impact of ownership concentration on firm performance vary between the I.T. industry and the retail industry? 3) What is the impact of managerial ownership on corporate performance? 4) Do different identities of the largest shareholder have impact on corporate performance? 5) Does the global financial crisis have an effect on the relation between ownership concentration and corporate performance in these two industries? In this paper, percentage of shareholdings by the largest owner is used to represent ownership concentration of a
1
Yin Guang Sha is a listed firm on Shenzhen Stock Exchange. 2
corporation and percentage of shareholdings by the management team members in a firm is used to represent managerial ownership. In addition, the largest owners are classified into four types: family/individual, financial investors, non-financial companies and government (La Porta et al., 1999; Pedersen and Thomsen, 2003). Firms from the information technology (I.T. for short) industry and the retail industry will be studied. However, due to the data availability problem, this paper only studies the publicly listed firms from those two industries on China’s two stocks exchanges. Our data set contains 117 companies, 66 retail firms and 51 I.T. firms, from 2005 to 2010 and a total of 702 observations. We perform both ordinary least square (OLS for short) regressions and panel regressions and use two proxies for firm performance: asset returns and the market-to-book equity value to test the hypotheses for robustness aims. By answering the research questions empirically with these means, we try to provide a comprehensive description of influence by ownership structure on firm performance.
comparison could show a much clearer picture on the relationship between ownership level and firm performance in different industries. Fourth, we investigate the influence of the recent global crisis on the relationship between firm performance and the level of ownership. Since I.T. firms have considerably more firm-specific capital and investments than retailing firms, they might be affected differently by the crisis.
There are also several implications of this paper. First, there is a demand for the Chinese government to protect minority shareholders. Second, the objectives of the largest owner differ with their identity; thus, the identity of the owners should be taken into account in evaluating firm performance. Third, the ownership structure of Chinese SOEs needs to be reformed by decreasing the governmental shareholdings and other types of controlling shareholders should be encouraged. Finally, a certain level of ownership concentration is vital for Chinese firms to overcome the negative effect of the global financial crisis on firm performance. The optimal concentration level of these industries differs: 34.13% for I.T. firms and 43.11% for retail firms.
2. Literature Review
In this section, theories about corporate governance including agency theory, ownership concentration, managerial ownership, and owner identities are provided. Information about the industry effect on ownership concentration is also included. A brief description of the share structure in China is introduced. Empirical results of these ownership dimensions are also provided in the last section.
2.1 Theoretical Framework 2.1.1 Corporate Governance
“Corporate governance deals with the ways in which suppliers of finance to corporations assure themselves of getting a return on their investment (Shleifer and Vishny, 1997, p.737).” Corporate governance could be separated into a corporation’s perspective and a public policy perspective. In a corporation, corporate governance is to maximize firm value under a corporation’s financial constraints and other legal obligations. From a public policy perspective, corporate governance is about fostering firms as well as ensuring the accountability when shareholders exercise their power and rights. These two perspectives provide a general framework for corporate governance worldwide (Iskander and Chamlou, 2000). According to Denis and McConnell (2003), mechanisms of corporate governance could be categorized into two aspects: external and internal. The primary internal mechanisms are about the relationship among the key players in the corporation, namely the board of directors and the equity ownership structure of the firm; while the external ones are about the external forces including external market for corporate control and the legal regulatory system. The interaction between the internal and external aspects governs the behavior and performance of the firm.
2.1.2 Agency Problems
another person (the agent) to operate the firm on their behalf by signing a contract with the agent (Jensen and Meckling, 1976). The classical agency problem depicts the conflict of interests between managers and shareholders caused by the separation of ownership from control (Chen, 2001). However, there are different types of stakeholders, besides shareholders, that may act as agents in a firm: directors, executives and creditors. Thus, potential agency conflicts are classified into three categories: conflicts between shareholders and managers; conflicts between blockholders and minority shareholders; conflicts between shareholders and creditors (Mulburt, 2009). Eisenhardt (1989) generalizes these conflicts into two agency problems. The first problem takes place when the goals of the two parties differ from each other while it is difficult or expensive for the principal to monitor the agent’s behavior. Agents may misbehave rather than aim at maximizing firm value, sometimes even at the expense of the principals (Howorth and Westhead, 2003; Denis and McConnell, 2003). On the other hand, shareholders could reduce their investment risks by diversification and might prefer more risks than manages. The problem of risk sharing, the other agency problem, arises when these two parties have different attitudes toward risk (Eisenhardt, 1989).
2.2 Ownership Concentration 2.2.1 Two Hypotheses
Ownership concentration is regarded as a mechanism to solve agency problems (Hart, 1995). There are two hypotheses on ownership concentration: the monitoring hypothesis and the expropriation hypothesis. Since shareholders prefer a diversified investment portfolio to reduce their risks, firms might have dispersed ownership structure. However, dispersion of the ownership structure could create opportunities for free-riders and makes manager monitoring difficult. As a result, the monitoring hypothesis suggests a positive relation between ownership concentration and firm performance (Miguel et al., 2004). In contrast, Shleifer and Vishny (1997) argue that, in some countries, the agency problem comes from the conflict between the large shareholder and minority shareholders. This situation is later defined as “tunneling” which comes into two forms (Johnson et al., 2000a). The simple form of tunneling depicts a situation where the largest shareholder legally transfers firm’s resources outside the firm. This could be achieved through self-dealing transactions such as loan guarantees and transferring price advantage to the largest shareholder. A more complicate form of tunneling depicts a situation where the largest shareholder increases shareholdings of the company without transferring any firm’s assets. This is often achieved by means of diluting share issues, which discriminates against minority shareholders and enhances the power of the largest shareholder in the corporation. As a result, the expropriation hypothesis predicts a negative effect of ownership concentration on firm performance.
2.2.2 Industry Effect
ownership structure and strategy of a firm, also vary among industries (Demsetz and Lehn, 1985). Pound and Zeckhouser (1990) distinguish two types of industries in evaluating firm performance: industries with highly firm-specific assets and investments and those without. They argue that industries with firm-specific investments and resources are difficult for shareholders to monitor and hard for large shareholders to intervene in management’s investment decisions. In addition, information about those industries is less transparent to shareholders which create barriers for shareholders to evaluate firm performance. Therefore, the owner-management agency problems are more prone to take place and the (positive) monitoring effect of concentrated ownership on firm performance is reduced in those industries.
2.3 Managerial Ownership
depends on the tradeoff between those two effects (Dennis and McConnell, 2003).
2.4 Owner Identity
Alongside with ownership concentration and managerial ownership, the identity of the largest owner is an essential but often neglected dimension of ownership structure in evaluating firm performance (McConnell and Servaes, 1990; Short, 1994). Pedersen and Thomsen (2000) argue that characteristics of different owner types vary, which has a direct impact on the firm's goal and strategy and indirectly influence firm performance. There four representative types of the largest owner: family/individual, financial investors, non-financial companies and government (La Porta et al., 1999; Pedersen and Thomsen, 2003). The costs and benefits of each owner identity are presented as a benchmark to assess their potential impacts on corporate performance.
Family/individual ownership has a unique characteristic, as family/individual owners have exceptional concerns over the firm’s survival and have strong incentives to monitor management. Consequently, the possibility of misalignment between manager and owner interests could decrease (McConaughy et al., 1998; Andres, 2008). However, concentrated ownership by a family or an individual means fewer possibilities for those investors to diversify their investments. Firms owned by this type of owner may less possibly issue shares, have less debt and pursue risk reduction strategies (Pedersen and Thomsen, 2003; Andres, 2008). In some cases, the family/individual plays a dual role within a corporation: owner and manager. However, the combination of management and control cannot prevent tunneling problems (Fama and Jensen, 1985).
knowledge, thus function as efficient monitors in solving agency problems. Moreover, the potential tunneling problem is less likely to take place, as they are regulated and supervised by government (Delios et al., 2006).
Non-financial companies often hold shares of other companies. Cross shareholdings could take place between corporations and in a business group structure (Kester, 1992). On the one hand, this form of ownership could increase firm value. It could serve as an anti-takeover device, optimize resources distribution, avoid information asymmetries and could take advantage of tax benefits of debt financing (Williamson, 1975; Lins and Servaes, 1999). On the other hand, the potential tunneling problems could not be avoided. There might be conflicts between different divisions and integrating into a business group can be costly in spite of resources sharing benefits (Jensen, 1986; Pedersen and Thomsen, 2003).
Government ownership is thought as leading to a negative effect on firm performance in general. Theoretically, the ultimate target for government could be, for instance, to reduce price, induce consumption, protect infant industry, gain public support and increase employment rate but rarely to maximize profits (Boycko et al., 1996; Delios et al., 2006). Shleifer (1998) posed another view on government ownership and argued that government employees lack incentives to maximize efficiency in reducing cost and innovating quality, which leading to the inefficient government ownership. However, benefits of government ownership could be low cost because of mass production and it most of the time plays a role as institutional alternative to regulation authorities (Pedersen and Thomsen, 2000).
2.5 Share Structure in China
same voting rights in theory, but differ in the ways of transactions (Qiang, 2003). State shares, legal person shares and tradable A-shares are three major types of shares outstanding in a typical listed company. Most of the listed companies do not issue foreign shares or employee shares, if there are any, the amount is no more than 10% of total outstanding shares (Delios et al., 2006).
2.6 Empirical Results
2.6.1 Ownership Concentration
Theoretically, the monitoring hypothesis and the expropriation hypothesis together suggest the possibility of a positive or negative, or non-linear relation between ownership concentration and performance such as U-shaped, inverted U-shaped and piecewise linear. Empirical studies on the effect of ownership concentration on performance across countries vary and some even contradict to the others. Table 1 presents the methodology and significant findings by various authors in evaluating the effect of concentrated ownership on firm performance in different countries.
Table 1: Ownership Concentration and Performance, Overview of Literature
Author Year Country Methodology Relationship Breakpoint
Earle, Estrin 1996 Russia OLS positive
Gedajlovic, Shapio 1998 US
Germany
OLS U-shaped 43%
OLS U-shaped 70%
Mudambi, Nicosia 1998 UK OLS/ WLS negative
Claessens, Djankov 1999 Czech OLS/random
effects
positive LauterBach,
Vaninsky
1999 Israel DEA negative
Pedersen, Thomsen 2000 EU OLS inverted
U-shape
83%
Morck et al. 2000 Japan OLS positive
Chen 2001 China OLS positive
Filatotchev et al. 2001 Russia OLS negative
Pivovarsky 2003 Ukraine OLS positive
Miguel et al. 2004 Spain GMM inverted
U-shaped
Edwards, Weichenriede
2004 Germany OLS positive
OLS: ordinary least square method; WLS: weighted least square method; random effects: random effects panel data method; DEA: data envelope analysis; GMM: generalized method of moments. If the relationship is nonlinear, the breakpoint level is also provided.
All the relationships are significant
Obviously, the relationship between ownership concentration and firm performance differ among countries. Shleifer and Vishny (1997) argue that large block holders could monitor managers’ behavior in countries that have weak investor protection and less developed stock markets. However, empirical results from transition economies do not explicitly confirm this positive relationship between ownership concentration and performance. While evidence from Czech, Chinese and Ukrainian corporations demonstrate a positive relationship between ownership concentration and firm performance, there appear an inverted U-shaped relationship in Spanish firms and a negative relationship in Israel corporations (Claessens and Djankov, 1999; LauterBach and Vaninsky, 1999; Chen, 2001; Pivovarsky, 2003; Miguel et al., 2004). Moreover, evidence from different developed stock markets differ and even contradict to each other. There is a positive relation between block holders and firm value in Japanese corporations, while the UK evidence proposes a negative effect by ownership concentration on performance. Furthermore, the US evidence signals a U-shaped ownership effect on performance, while evidence from the largest European firms shows an inverted U-shaped relation between ownership and performance (Gedajlovic and Shapiro, 1998; Mudambi and Nicosia, 1998; Pedersen and Thomsen, 2000; Morck et al., 2000).
In addition, there are only a few studies that compare and investigate the industry effects. Most of the empirical studies listed in Table 1 cover one industry or different industries within a country by using industry dummies. Evidence from the same country but different industries do confirm the existence of industry effects. Gedajlovic and Shapiro (1998) find a U-shaped ownership effect on performance in German medium to large-sized listed firms, while Edwards and Weichenriede (2004) find, from German nonfinancial industry, a positive relationship between ownership concentration and performance. Earle and Estrin (1996) analyzes listed industrial firms in Russia and find a positive effect of concentrated ownership on performance, while Filatotchev et al. (2001) find a negative relation between ownership concentration and firm performance among privatized manufacturing firms in Russia.
2.6.2 Managerial Ownership
Table 2: Managerial Ownership and Performance, Overview of Literature
Author Year Country Methodology Relationship Breakpoint
Morck et al. 1988 US OLS cubic 5%, 25%
McConnell, Servaes 1990 US OLS inverted U-shaped 49% Mudambi, Nicosia 1998 UK OLS/ WLS cubic 11%, 25%
Han, Suk 1998 US OLS inverted
U-shaped
41.8%
Selarka 2006 IND OLS cubic 45%, 63%
OLS: ordinary least square method; WLS: weighted least square method; IND: India. If the relationship is nonlinear, the breakpoint level is also provided. All the relationships are significant.
2.6.3 Owner Identity
Based on the costs and benefits analysis of the four types of largest owner, the effect of different owner identities on firm performance is expected to be different. Table 3 presents the significant findings of the effect of owner types on performance in different countries. Most of the empirical results are consistent with the theory, and only the corporation ownership shows different effects on performance among different countries.
Table 3: Owner Identity and Performance, Overview of Literature
Author Year Country Methodology Identity Relationship
McConnell, Servaes 1990 US OLS INS positive
Chaganti, Damanpour 1991 US OLS INS positive
Lichtenberg, Pushner 1994 JPN OLS INS positive
COR negative
Han, Suk 1998 US OLS INS positive
McConaughy et al. 1998 US OLS FAM positive
Lins, Servaes 1999 UK
JPN
OLS COR negative
COR negative
Xu, Wang 1999 CHN OLS GOV negative
INS positive
Dewenter, Malatesta 2001 US OLS GOV negative
Anderson, Reeb 2003 US fixed effects FAM positive
Pedersen, Thomsen 2003 EU 3SLS GOV negative
COR positive
Andres 2008 GER OLS/random
effects
FAM positive
JPN: Japan; CHN: China; GER: Germany; OLS: ordinary least square method; fixed effects: fixed effects panel data method; 3SLS: 3 stage least square method; random effects: random effects panel data method; FAM: family/individual ownership; INS: financial institution ownership; COR: corporate ownership; GOV: government ownership. All the relationships are significant.
3. Hypotheses and Methodology
3.1 Hypotheses
The question addressed by this paper is the effect of ownership structure on firm performance in China. Based on theories and the previous empirical studies on ownership structure and firm performance, this paper has some hypotheses. First, we believe there is an inverted U-shaped relationship between ownership concentration ratio and the performance of Chinese firms. When ownership is widely dispersed, free-rider problems may happen and monitoring costs are high. In that situation, the monitoring effect dominates the expropriation effect. When ownership becomes more and more concentrated, tunneling problems may more possibly appear. In that case, the (negative) expropriation effect is going to offset the (positive) monitoring effect. Therefore the hypothesis is:
H10: There is no relationship between ownership concentration and firm
performance.
H1a: There is an inverted U-shaped relationship between ownership
concentration and firm performance.
In addition, we agree with Pound and Zeckhouser (1990) and believe there is an industry effect on the relationship between ownership concentration and firm performance. We select the I.T. industry in China as one representative industry with abundant firm-specific assets and the retail industry as one that without. As I.T. firms produce products or provide services with high-technology, there are more firm-specific assets compared to retailing firms. Hence, it is difficult for the largest owner to monitor the management and influence the management decisions. The monitoring effect of concentrated ownership on firm performance is reduced in the I.T. industry. Therefore the hypothesis is:
H20: There is no industry effect on the relationship between ownership
concentration and firm performance.
performance in the I.T. industry.
H2b: There is an inverted U-shaped relationship between ownership
concentration and firm performance in the retail industry.
Third, we believe managerial ownership has impact on firm performance. When the managerial ownership is low, increase of managerial ownership aligns the interests between the agent and the principal and reduces agency problems. In that case, the (positive) convergence-of-interest effect dominates the (negative) entrenchment effect of management ownership. However, when the managerial ownership reaches a certain level, managers might have enough power to secure their position and thus pursue personal interests. In that case, the entrenchment effect dominates the convergence-of-interest effect. Therefore the hypothesis is:
H30: There is no relationship between managerial ownership and firm
performance.
H3a: There is an inverted U-shaped relationship between managerial ownership
and firm performance.
Fourth, we believe owner types do have an effect on firm performance. Empirical evidence from the literature review section predicts a positive effect of family/individual ownership, a positive effect of institutional ownership and a negative effect of government ownership on firm performance. Though the results on corporation ownership differ, we the corporation ownership in China affects firm performance negatively. In fact, cross shareholdings of Chinese corporations are controlled and dictated by the government for strategic aims, such as to compete on the international market, rather than for efficiency aims. Therefore the hypothesis is:
H40: There is no relationship between the identity of the largest owner and firm
performance.
H4a: There is a positive relationship between family ownership and firm
H4b: There is a positive relationship between institutional ownership and firm
performance.
H4c: There is a negative relationship between company ownership and firm
performance.
H4d: There is a negative relationship between government ownership and firm
performance.
3.2 Equations and Measurements
In this study, the ordinary least square approach (OLS for short) is used to test the hypotheses. There are two reasons: the sample size in this paper is relatively small, and the OLS method is widely used in previous studies of ownership structure and corporate governance (Morck et al., 1988; McConnell and Servaes, 1990; Xu and Wang, 1999; Migual et al., 2004). We first run regressions for ownership concentration (Equation 1), managerial ownership (Equation 3) and owner identity (Equation 4) on the whole sample using both ROA and MBV as proxies for performance. Then we do separate regressions on ownership concentration to examine the potential industry effect (Equation 2). Moreover, we split the data into two sub samples: pre-crisis and post-crisis samples, and run OLS regressions to investigate the potential crisis effect on the relation between ownership concentration and performance, and whether it varies in different industries. We also perform year-by-year OLS regressions to offer a clear picture of the change of the relationship between ownership concentration and firm performance. Moreover, since our data comprise time series and cross-sectional elements, estimating the equations with simple OLS is not sufficient. Panel techniques are also used to investigate the hypotheses for robustness aim.
Model 1: Ownership concentration (testing hypothesis 1)
Pi,t = α + β1*CONi,t + β2*CONSQi,t + β3*INDi,t + β4*GROWTHi,t + β5*LEVi,t +
For each industry (testing hypothesis 2)
Pi,t =α + β1*CONi,t + β2*CONSQi,t + β3*GROWTHi,t + β4*LEVi,t + β5*SIZEi,t +
εi,t………...(2)
Model 2: Managerial ownership (testing hypothesis 3)
Pi,t = θ+ γ1*MANi,t + γ2*MANSQi,t + γ3* GROWTHi,t + γ4* LEVi,t + γ5*SIZEi,t +
εi,t………...(3)
Model 3: Identity (testing hypothesis 4)
Pi,t = µ1*FAMi,t + µ2*INSi,t + µ3*CORi,t + µ4*GOVi,t + µ5* GROWTHi,t + µ6*LEVi,t +
µ7*SIZEi,t + εi,t………..……….(4)
Where i and t denotes the firm and time subscripts; P=return on assets or market-to-book value of equity; CON= percentage of shares the largest owner has; CONSQ= square of CON; MAN= percentage of shares owned by the management team; MANSQ= square of MAN; IND dummy= 1 when it is the I.T. industry; IND dummy= 0 when it is the retail industry; FAM dummy=1 when the largest owner is family, if not FAM=0; COR dummy =1 when the largest owner is corporation, if not COR=0; INS dummy=1 when the largest owner is institution, if not INS=0; GOV dummy=1 when the largest owner is government, if not GOV=0; SIZE= natural logarithm of the book value of total assets in the end of the year; GROWTH= (sales this year/sales last year) -1; LEV= book value of all liabilities in the end of the year/book value of total assets in the end of the year; ε= error term.
3.3 Variables Dependent Variable
the book value of total assets, which is a popular measure of performance which has been used by many authors (Morck et al., 1988; Chaganti and Damanpour, 1991; Gedajlovic and Shapio, 1998; Han and Suk, 1998; Demsetz and Villalonga, 2001; Andres, 2008). It is an accounting measure of the efficiency with which corporate assets are managed. It reflects the profitability of a firm in the short run and is not influenced by tax effects. Moreover, Demsetz and Villalonga (2001) argue that it is more sensible to evaluate the relationship between ownership and performance based on what “has been” done, not based on what “will be” done. MBV equals share prices on the last trading day of each year multiplied by the number of total outstanding shares divided by the book value of equity. It is a sufficient market indicator of future return on equity and used by many authors (Xu and Wang, 1999; Thomsen and Pedersen, 2000; Edwards and Weichenriede, 2004; Selarka, 2006). As Selarka (2006) points out that MBV is directly attached to the confidence of the investors, and hence could be applied as a capital market assessment of a firm. By using both ROA and MBV to represent firm performance, we try to deliver our regression results in a robust way.
Independent Variables
The second independent variable is square of ownership concentration (CONSQ for short). It is widely used to capture the potential quadratic relationship between ownership and firm performance (Stulz, 1988; McConnell and Servaes, 1990; Anderson and Reeb, 2003; Miguel et al., 2004).
The third independent variable is managerial ownership (MAN for short). It is the percentages of the shares directly held by all the members of the corporate board, the CEO and top management (Morck et al., 1988; Demsetz and Villalonga, 2001).
The fourth independent variable is square of managerial ownership (MANSQ for short). It is an essential technique to capture the possible quadratic relationship between managerial ownership and firm performance (McConnell and Servaes, 1990; Han and Suk, 1998; Miguel et al., 2004).
The fifth independent variable is owner identity (ID for short). In this paper, types of owner identity are family, companies, financial investors and government. They are presented as dummy variables. If the largest owner is the government, GOV equals to 1, otherwise, GOV equals to 0; if the largest owner is another corporation, COR equals to 1, otherwise, COR equals to 0; if the largest owner is an institution, INS equals to 1, otherwise, INS equals to 0; if the largest owner is family or individual, FAM equals to 1, otherwise, FAM equals to 0.
The sixth independent variable is industry (IND for short). It is measured as a dummy variable. There are two industries studied in this paper: the I.T. industry and the retailing industry. The industry dummy equals to 1 if a firm is in the I.T. industry and equals to 0 if a firm is in the retailing industry.
Control Variables
logarithm of the book value of total assets of firms at the end of the year. On the one hand, free rider problems and agency problems are more possible to appear in large firms; on the other hand, those firms could enjoy economies of scale (Demsetz and Lehn, 1985; Himmelberg et al., 1999; Chen, 2001). We therefore follow many authors and treat size as a control variable (Han and Suk, 1996; Lauterbach and Vaninsky, 1999; Anderson and Reeb, 2003; Andres, 2008).
The second control variable is sales growth (GROWTH for short). It is calculated as sales this year divided by sales last year minus one. This measures the demand conditions facing the firm and the product life cycle effects on the firm. Lins and Servaes (1999) argue that firms in relatively fast-growing markets or in the growth phase of the product life are expected to have above-average profitability. We therefore follow many authors and control for a possible sales growth effect (Gedajlovic and Shapiro, 1998; Thomsen and Pedersen, 2000; Chen, 2001; Selarka, 2006).
The third control variable is leverage (LEV for short). It is calculated as the ratio of book value of all the liabilities of the firm to the book value of total assets at the end of the year. On one hand, debt is attractive since debt is tax deductable and it could generate tax shields. On the other hand, debt could lead to considerable financial distress costs. We therefore follow many authors and control LEV in our model (Morck et al., 1988; McConnell and Servaes, 1990; Thomsen and Pedersen, 2000; Chen, 2001; Anderson and Reeb, 2003; Miguel et al., 2004).
In brief, Table 4 provides the list of all the variables and their definitions in the model.
Table 4: List of Variables
Variables Definition
Market-to-book value of equity (MBV)
Share prices on the last trading day of each year multiplied by the number of total outstanding shares divided by the book value of equity
Ownership concentration (CON) Shares owned by the largest owner (%)
Square of ownership concentration (CONSQ)
Square of CON
Managerial ownership (MAN) Portioned sum of the shares directly held by
members of the corporate board, the CEO and top management (%)
Square of management ownership (MANSQ)
Square of MAN Ownership identity (ID)
dummy
FAM= family/individual; INS= financial institutions; COR= corporations; GOV= government Industry (IND)
dummy
IND=1, I.T. industry; IND=0, retail industry
Sales growth (GROWTH) Sales this year/sales last year-1
Firm size (SIZE) Natural logarithm of the book value of total assets in
the end of the year
Leverage (LEV) Ratio of book value of all the liabilities of the firm
4. Data and Descriptive Statistics
4.1 Data Collection
Since this study focuses on China's evidence, we collect the data from China’s two stock exchanges: SHSE and SZSE. They both provide complete and detailed information of the firm including their annual report. According to the China Securities Regulation Commission, the listed firms on both stock exchanges are obliged to disclose annual reports including capital structure, top ten shareholders, profit or loss, total assets etc. This paper studies listed firms in the I.T. industry and the retail industry from 2005 to 2010. There are 93 retail firms and 76 I.T. firms that went public before 2005; however, some of them were bankrupt and some have missing data in their annual reports. As a result, the data set contains 66 retail companies and 51 I.T. companies over six years and a total of 702 firm-year observations. The data of the percentage of the largest shareholding, the earnings before income tax, the book value of total assets, the number of total outstanding shares, the book value of equity, the sales volumes, and the book value of total liabilities are collected directly from the firms’ annual reports. We collect share prices on the last trading day of each year from the stock exchanges’ website. The proportion of managerial shareholding is added to the portion shareholdings of every member from the management team of the company. Moreover, we investigate the actual shareholdings of the largest shareholders to identify the ultimate owner identity of the firm. We apply this rule: if firm A is owned by firm B, but firm B is owned by firm C and firm C is not owned by other firms, then the identity of firm C is the identity of firm A.
family/individual compared to retail firms. Financial institutions tend to favor retail industry while corporation ownership is more common in the I.T. industry.
Table 5: Owner Identity in the I.T. Industry and the Retail Industry
FAM INS COR GOV Total
I.T. industry 60 12 24 210 306
Retail industry 24 30 18 324 396
Total 84 42 42 534 702
FAM= family/individual ownership; INS= financial institution ownership; COR= corporate ownership; GOV= government ownership.
4.2 Descriptive Statistics
Table 6 presents two panels of descriptive information for our sample of firms. Panel A provides means, medians, maximum and minimum values, and standard deviations for the variables in our sample. Panel B provides the correlation matrix for the variables in our sample.
Table 6: Descriptive Data (n=702 firm-year observations)
Panel A: Summary Statistics for the Full Sample
ROA MBV CON MAN GROWTH LEV SIZE
Mean 2.49% 4.37 31.60% 1.70% 21.42% 57.27% 21.25
Median 3.05% 3.41 28.69% 0.00% 11.80% 54.08% 21.23
Maximum 256.89% 72.04 70.49% 25.95% 1642.82% 733.15% 24.50
Minimum -385.61% -124.09 6.69% 0.00% -98.41% 3.73% 16.83
Std 19% 6.43 14% 7% 92% 49% 0.95
Panel B: Correlation Data
ROA MBV CON MAN GROWTH LEV SIZE
ROA 1 MBV 0.05 1 CON 0.05 -0.02 1 MAN 0.04 0.02 -0.12 1 GROWTH 0.05 -0.02 0.01 0.00 1 LEV -0.27 -0.13 -0.07 -0.05 0.23 1 SIZE 0.22 0.03 0.2 -0.01 -0.07 -0.19 1
of total assets; Std= standard deviation.
5. Results
5.1 Regression Results Ownership Concentration
Table 7 shows the OLS results for the impact of ownership concentration level on firm performance. Column 1, 2 and 3 present results using an accounting performance measure: ROA; Column 4, 5 and 6 show results using a market measure: MBV. First, we examine the whole sample: Column 1 and Column 4. The coefficient estimate for CON is positive and significant, while that for CONSQ is negative and insignificant when both ROA and MBV are measures for firm performance. When the concentration level increases by 1%, the asset return of the firm could improve by 0.133% and the MBV value of the firm could improve by 3.007. Though the results are statistically significant but economically insignificant, the coefficient estimates together confirm a positive relationship between firm performance and ownership concentration ratio, which is consistent with the findings of Earle and Estrin (1996), Morck et al. (2000), Chen (2001), Edwards and Weichenriede (2004). This positive relationship confirms the monitoring hypothesis about ownership concentration that large shareholders may serve as a monitoring device and help evade potential free rider problems of minority shareholders, and thus help to increase firm value. There appears a positive relation between ownership concentration level and firm performance; therefore Hypothesis 1 should be rejected.
when evaluating firm performance with either ROA or MBV measure, the level of ownership concentration does have a significantly positive impact on performance in the retail industry, but the impact is not significant in the I.T. industry. As I.T. firms have more firm-specific assets and investments than retail firms, it is difficult for shareholders to monitor or to intervene in management’s decision making. The effect of the largest owner on firm performance is reduced in the I.T. industry comparing to the retail industry. Therefore Hypothesis 2 could be rejected, and there exists industry variation in evaluating firm performance.
Concerning the control variables, we find LEV is significantly negative related to firm performance in all cases though it is not economically significant. It seems that debt financing may constrain firm’s investing strategies and incur considerable financial distress costs, and hence lowers firm performance. In addition, debt financing could not solve the agency problems. The effect of SIZE is significant in the retail industry. However, the sign of SIZE by the ROA measure is positive and is negative by the MBV measure. Apparently, large firms could enjoy economies of scale and hence have higher profitability. Because of information asymmetric problems, the market does not favor large firms as the agency problems are severe in these firms. GROWTH is significantly positive related to firm performance in the whole sample and indicates that high-growth opportunities could help a firm to perform better. Furthermore, the significant IND variable suggests a negative industry effect on firm performance. What is worth mentioning is that, in spite of the significant effects of those control variables, they are, in fact, not economically significant.
Table 7: Relationship between Firm Performance and Ownership Concentration
ROA MBV
Total I.T Retail Total I.T. Retail
C -0.191 -0.011 -0.281*** 3.38 3.532 9.764**
CON 0.133* 0.516 0.121* 3.007* 4.918 3.001*
GROWTH 0.037*** 0.043*** -0.002 0.109* 0.011 0.123 LEV -0.204*** -0.223*** -0.093*** -2.038** -1.967*** -2.119*** SIZE 0.013* 0.003 0.016*** -0.11 0.077 -0.31* INDUSTRY -0.015* -0.7* Adjusted R2 0.286 0.297 0.163 0.081 0.122 0.079 Prob(F-statis tic) 0 0 0 0 0 0
Firm performance is represented by ROA and MBV; C= constant term; ROA= return on assets; MBV= market-to-book value; CON= ownership concentration ratio; CONSQ= square of CON; GROWTH= sales growth rate; LEV= leverage ratio; SIZE= natural logarithm of total assets; IND= industry dummy; number of observations is 702. *Significant at 10% level; **Significant at 5% level; ***Significant at 1% level.
Yearly Evidence
Table A, Table B, and Table C in the Appendix display yearly OLS regressions on the relation between the level of ownership and performance for the whole sample, for the I.T. firms and the retail firms respectively. As the results of ROA have much higher adjusted R2 than those of MBV, ROA is used as a proxy for performance.
We first investigate the whole sample. The results for the years 2005, 2006, 2008 and 2010 confirm the previous result and prove a significantly positive relationship between ownership concentration and performance. However, the regression result for the year 2007 proves no ownership concentration effect on performance, while the result for the year 2009 suggests a significant inverted U-shaped relation. In general, the results from Table A indicate that there is a yearly effect in evaluating the relation between firm performance and the level of ownership.
performance. In contrast, evidence from the retail industry presents a different picture. The results for the years 2005 and 2007 suggest no relation, for the year 2006 shows a significant inverted U-shape relation, and for the years 2008, 2009 and 2010 prove a significantly positive relation between ownership concentration and firm performance.
Third, we compare the industrial samples. The signs of CON are significantly positive for the years 2008, 2009 and 2010 in both industries. Compared to the results for the years 2005, 2006 and 2007, we could first confirm the crisis effect. However, evidence from the year 2009 shows a significantly inverse U-shaped impact of ownership on performance among the I.T. firms. Therefore, we cast doubt on whether the crisis has influenced the relationship in these industries differently.
Crisis Impact
In addition, we also aim to investigate the influence of the recent financial crisis (from 2007 to 2008) on the relation between the level of ownership and performance in different industries. We split our samples into two sub samples: the pre-crisis sample containing observations from 2005 to 2007 and the post-crisis sample containing observations from 2008 to 2010. Although the recent crisis took place in September, 2007, it has directly affected the US and EU market but indirectly influenced the Chinese market. In addition, the signs of CON are all significant positive after the year 2007 from the yearly evidence.
and the retail industry, while the breakpoints differ: the optimal concentration level for I.T firms is 34.13% and is 43.11% for retail firms. In addition, the coefficient estimates of CON and CONSQ are both statistically and economically significant in the I.T. industry. This suggests that a certain level of ownership concentration could contribute on increased I.T. firms’ performance considerably. In short, we could confirm the existence of crisis effect and its influence on the relation between firm performance and ownership level does not vary in these industries.
Managerial Ownership
Table 8 shows the OLS results of the effect of managerial ownership on performance with an accounting measure, ROA in Column 1, and a market measure, MBV in Column 2. Although the coefficients of MAN and MANSQ are not significant, they are positive for MAN and negative for MANSQ when firm performance is measured by ROA and MBV. Concerning the control variables, the effects of LEV and SIZE are consistent with the results in Table 7. In addition, their effects are statistically significant though economically insignificant. While the effect of GROWTH is the same as the previous result using ROA as a proxy; it is not significantly though positively related to firm performance with the MBV measure. In short, Hypothesis 3 cannot be rejected due to insignificant coefficients, and managerial ownership does not influence firm performance.
Table 8: Relationship between Firm Performance and Managerial Ownership
ROA MBV C -0.204 6.877 MAN 0.074 2.136 MANSQ -0.128 -2.488 GROWTH 0.037*** 0.12 LEV -0.202*** -2.1** SIZE 0.016** -0.06 Adjusted R2 0.283 0.043 Prob(F-statistic) 0 0.04
market-to-book value; MAN= managerial ownership; MANSQ= square of MAN; GROWTH= sales growth rate; LEV= leverage ratio; SIZE= natural logarithm of total assets; number of observations is 702. *Significant at 10% level; **Significant at 5% level; ***Significant at 1% level.
Owner Identity
Table 9 shows the OLS results for the relation between owner identity and firm performance with an accounting performance measure, ROA in Column 1, and a market measure, MBV in Column 2. We examine the effect of each identity in detail and offer possible explanation to those results.
Table 9: Relationship between Firm Performance and Owner Identity
ROA MBV FAM -0.022 -0.48 INS 0.154* 2.087* COR -0.055 -1.332 GOV -0.022* -0.14* GROWTH 0.037*** 0.14 LEV -0.208*** -1.865** SIZE 0.008* 0.23*** Adjusted R2 0.283 0.063
Firm performance is represented by ROA and MBV; ROA= return on assets; MBV= market-to-book value; FAM= family/individual ownership; INS= financial institution ownership; COR= corporate ownership; GOV= government ownership; GROWTH= sales growth rate; LEV= leverage ratio; SIZE= natural logarithm of total assets; number of observations is 702.
*Significant at 10% level; **Significant at 5% level; ***Significant at 1% level.
Family/individual ownership has a negative but not significant impact on firm performance when both ROA and MBV are the proxies of firm performance. This is consistent with the predictions of Fama and Jensen (1985): the potential principal-agent conflicts are eliminated by the duo role of family ownership, but the potential tunneling problems cannot be avoided.
Pushner (1994), and Xu and Wang (1999). This owner type often acts as a monitoring device, and thus, the potential agency problems are relatively reduced. Moreover, as they are directly regulated and supervised by the government (Delios et al., 2006), the potential tunneling problems also decline.
Corporation ownership has a negative but insignificant relation with firm performance using both ROA and MBV as proxies for performance. This is consistent with the fact that cross shareholdings could realize relative economies to scale and avoid asymmetric information, but there are costs for different divisions to integrate. In fact, cross shareholdings of Chinese corporations are controlled and dictated by the government for strategic aims, such as to compete on the international market, rather than for efficiency aims.
Government ownership has a significant negative effect on firm performance and the results do not alter with different measures of performance. This confirms the findings of Dewenter and Malatesta (2001) and Thomsen and Pedersen (2003). Most Chinese government-owned firms do not aim at value maximizing, but most of the time, at increasing the employment rate and at providing public goods. Furthermore, the government often fosters infant industry by investing in firms within the industry regardless of their performance.
Concerning the control variables, the effects of LEV, GROWTH and SIZE are consistent with the results in Table 7 with the ROA measure. While the effect of LEV is the same as the previous result, SIZE is positively significant and GROWTH is insignificant with the MBV measure. The significantly positive signs of SIZE with both ROA and MBV measures predict that large firms could enjoy economies of scales and share useful resources internally.
measures of firm performance: ROA and MBV. Family/individual and corporation influence firm performance negatively though not significantly. Financial institutional has a significant and positive impact on performance. Government ownership is significantly negatively related to performance. Those effects are statistically significant but economically insignificant, which is consistent with the finding of Thomsen and Pedersen (2003). However, the adjusted R2 value shows that the regression results with the MBV measure should be coped with cautions. In sum, Hypothesis 4 should be rejected, and the type of owner does have an impact on firm performance.
5.2 Robustness Test
In section 5.1, we have estimated the models using the OLS method with both an accounting measure of performance and also a market measure of performance for robustness aim. However, as our data comprise time series and cross-sectional elements, estimating the equations with simple OLS is not sufficient. In this section, we follow Claessens and Djankov (2001) and Anderson and Reeb (2003) and use an alternative approach, the panel techniques, to investigate the hypotheses.
Ownership Concentration
Table E in the Appendix presents the panel regression results as well as the OLS results for the whole sample on the relation between firm performance and ownership concentration. Compared to the prior OLS results, the panel regression results do not alter with either the ROA measure or the MBV measure of performance which proves that firm performance is significantly positive related to ownership concentration. We could conclude that our results are robust with both simple OLS method and panel techniques. Hence, Hypothesis 1 should be rejected, and the level of ownership does influence firm performance positively.
Table F and Table G in the Appendix present the OLS and panel regression results of the impact of the level of ownership on performance in the I.T. industry and the retail industry separately. Table E shows that except the cross-sectional fixed effects result with the MBV measure, the other five regression results are consistent with others and indicate no impact of ownership concentration on I.T. firms. Table F shows that except the period fixed effects result using MBV as a performance measure, the other five regression results all suggest a positive relation between the level of ownership and firm performance in the retail industry. As most of the regressions confirm the previous results, we could conclude that our results are robust with both the simple OLS method and panel techniques. Hypothesis 2 should, therefore, be rejected and there is an industry effect in evaluating the influence of ownership concentration on firm performance.
Managerial Ownership
Hypothesis 3 could not be rejected because of the insignificant results.
Owner Identity
6. Conclusion
In this paper, we study the impact of ownership structure on performance in Chinese I.T. firms and retail firms. We divide ownership structure into three dimensions: ownership concentration, managerial ownership and owner identity, and study the relationship between these dimensions and firm performance. Our data set contains 117 companies, 66 retail firms and 51 I.T. firms, from 2005 to 2010 and a total of 702 observations. We apply not only an accounting performance measure, return on assets, but also a market performance measure, market-to-book value of equity, and use the simple OLS method and panel techniques to test our hypotheses. We compare the results of pooled regressions to those of panel regressions and find that our findings are robust. Our findings demonstrate that different ownership dimensions influence firm performance differently.
First, we find a significantly positive relation between the level of ownership and firm performance. This positive relationship confirms the monitoring hypothesis on ownership concentration that large shareholders help to monitor managers’ behavior. In addition, as there are more arbitragers in the Chinese market than in the developed markets, concentrated ownership also helps to reduce potential free rider problems of individual shareholders, and thus to increase firm value.
Third, we find a crisis effect that affects this relationship. There is a significant inverted U-shape relation between firm performance and ownership concentration after the crisis, while no relation is found before the crisis. In addition, this relationship is not altered between the I.T. industry and the retail industry.
Fourth, we find that managerial ownership does not influence firm performance. This might be explained by the fact that most of the sample firms are owned by the government, where managers are mostly paid according to their position and working years. Even if managers are compensated with firm shares, they are not dependent on firm performance. As a result, managers lack incentive to increase their pay by improving firm performance.
Last but not least, owner identity should be taken into consideration in evaluating firm performance. We find that financial institutions have a significant and positive impact on performance and often serve as a monitoring device. Government ownership is significantly negatively related to performance, as the central aim of the firms is not to maximize firm value but to pursue political objectives such as increasing the employment rate and providing public goods. However, we find no evidence of the impact by family/individual owner or corporations on firm performance.
Implications
strategy and firm performance varies. Third, other types of controlling shareholders should be encouraged, since most large firms are still controlled by the government, which are not operating efficiently. Finally, a certain level of ownership concentration is crucial to help firms to overcome the negative impact of the recent financial crisis on firm performance. In addition, the optimal concentration level of I.T. firms is 34.13% and it is 43.11% for retail firms.
Limitations
There are some limitations in this paper. First, we only investigate listed firms from the I.T. industry and the retail industry on China’ two stock exchanges due to the availability of the data. The results in this paper might be biased and should be coped with cautions. Second, most of the adjusted R2 values are not high, implying that the results could not be conclusive enough and we might have omitted some variables in our analysis. Third, our proxies for firm performance could not be directly obtained. The ROA and MBV measures might not be suitable enough to investigate the effect of ownership structure on performance in China. Fourth, our data period might be too short to investigate the yearly effect, and accordingly the panel regression results might be biased. Fifth, we ignore the endogeneity issues of a firm’s ownership structure as most of our sample firms are controlled by the government. In addition, we do not examine the causal relation between ownership structure and firm performance.
Suggestions for Future Research
Reference
Agrawal, A. and Knoeber, C. (1996), “Firm Performance and Mechanisms to Control Agency Problems between Managers and Shareholders”, Journal of Financial and Quantitative Analysis 31 (3), 377-397.
Anderson, R.C. and Reeb, D.M. (2003), “Founding-Family Ownership and Firm Performance: Evidence from the S&P 500,” Journal of Finance 58 (3), 1301-1328.
Andres, C. (2008), “Large Shareholders and Firm Performance: An Empirical Examination of Founding-Family Ownership”, Journal of Corporate Finance 14, 431-445.
Berle, A. and Means, C. (1932), The Modern Corporation and Private Property, Macmillan: New York.
Boycko, M.A., Shleifer, A. and Vishny, R.W. (1996), “A Theory of Privatization”, Economic Journal 106 (453), 309-328.
Chaganti, R. and Damanpour, F. (1991), “Institutional Ownership, Capital Structure, and Firm Performance”, Strategic Management Journal 12 (7), 479-492.
Chen, J. (2001), “Ownership Structure as Corporate Governance Mechanism: Evidence from Chinese Listed Companies”, Economics of Planning 34, 53-72.
Claessens, S. and Djankov S. (1999), “Ownership Concentration and Corporate Performance in the Czech Republic”, Journal of Comparative Economics 27 (3), 498-513.
Delios, A., Wu, Z.J. and Zhou, N. (2006), “A New Perspective on Ownership Identities in China’s Listed Companies”, Management and Organization Review 2 (3), 319-343.
Demsetz, H. (1983), “The Structure of Ownership and the Theory of the Firm”, Journal of Law and Economics 26 (2), 375-394.
Demsetz, H. and Lehn, K. (1985), “The Structure of Corporate Ownership: Causes and Consequences”, Journal of Political Economy 14, 1155-1177.
Demsetz, H. and Villalonga, B. (2001), “Ownership Structure and Corporate Performance”, Journal of Corporate Finance 7, 209-233.
Dewenter, K.L. and Malatesta, P.H. (2001), “State-Owned and Privately Owned Firms: An Empirical Analysis of Profitability, Leverage and Labor Intensity”, American Economic Review 91 (1), 320-334.
Earle, J.S. and Estrin, S. (1996), “Privatization versus Competition: Changing Enterprise Behavior in Russia”, Working Paper 49, International Institute for Applied Systems Analysis.
Edwards, J.S. and Weichenrieder, A.J. (2004), “Ownership Concentration and Share Valuation”, German Economic Review 5 (2), 143-171.
Eisenhardt, K.M. (1989), “Agency Theory: An Assessment and Review”, Academy of Management Review 14 (1), 57-74.
Filatotchev, I., Kapelyushnikov, R., Dynomina, N. and Aukutsionek, S. (2001), “The Effects of Ownership Concentration on Investment and Performance in Privatized Firms in Russia”, Managerial and Decision Economics 22 (6), 299-313.
Gedajlovic, E.R. and Shapiro, D.M. (1998), “Management and Ownership Effects: Evidence from Five Countries”, Strategic Management Journal 19 (6), 533-553.
Han, K.C. and Suk D.Y. (1998), “The Effect of Ownership Structure on Firm Performance: Additional Evidence”, Review of Financial Economics 7 (2), 143-155.
Hart, O. (1995), “Corporate Governance: Some Theory and Implications”, Economic Journal 105 (430), 678-689.
Himmelberg, C.P., Hubbard, R.G. and Palia, D. (1999), “Understanding the Determinants of Managerial Ownership and the Link between Ownership and Performance”, Journal of Financial Economics 53, 353-384.
Howorth, C. and Westhead, P. (2003), “The Focus of Working Capital Management in UK Small Firms”, Management Accounting Research 14 (2), 94-111.
Iskander, M.R. and Chamlou, N. (2000), Corporate Governance: A Framework for Implementation, World Bank Group: Washington.
Jensen, M.C. (1986), “Agency Costs of Free Cash Flow, Corporate Finance, and Takeovers”, American Economic Review 76, 323-329.
Johnson, S., Boone, P., Breach, A. and Friedman, E. (2000a), “Corporate Governance in the Asian Financial Crisis”, Journal of Financial Economics 58, 141-186.
Johnson, S., La Porta, R., Lopez-de-Silances, F. and Shleifer, A. (2000b), “Tunneling”, American Economic Review 90 (2), 22-27.
Kester, C.W. (1992), “Industrial Groups as Systems of Contractual Governance”, Oxford Review of Economic Policy 8 (3), 24-44.
La Porta, R., Lopez-de-Silanes, F. and Shleifer, A. (1999), “Corporate Ownership Around the World”, Journal of Finance 54 (2), 471-517.
Lauterbach, B. and Vaninsky, A. (1999), “Ownership Structure and Firm Performance: Evidence from Israel”, Journal of Management and Governance 3, 189-201.
Lichtenberg, F.R. and Pushner, G.M. (1994), “Ownership Structure and Corporate Performance in Japan”, Japan and the World Economy 6, 239-261.
Lins, K. and Servaes, H. (1999), “International Evidence on the Value of Corporate Diversification”, Journal of Finance 54 (6), 2215-2239.
McConaughy, D.L., Walker, M.C., Henderson Jr., G.V. and Mishra, C.S. (1998), “Founding Family Controlled Firms: Efficiency and Value”, Review of Financial Economics 7 (1), 1-19.
McConnell, J.J. and Servaes, H. (1990), “Additional Evidence on Equity Ownership and Corporate Value”, Journal of Financial Economics 27 (2), 595-612.
Miguel, A., Pindado, J. and La Torre, C. (2004), “Ownership Structure and Firm Value: New Evidence from Spain”, Strategic Management Journal 25 (12), 1199-1207.
Morck, R., Nakamura, M. and Shivdasani, A. (2000), “Banks, Ownership Structure, and Firm Value in Japan”, Journal of Business 73(4), 539-567.
Morck, R., Shleifer, A. and Vishny, R.W. (1988), “Management Ownership and Market Valuation: An Empirical Analysis”, Journal of Financial Economics 20, 293-315.
Mudambi, R. and Nicosia, C. (1998), “Ownership Structure and Firm Performance: Evidence from the UK Financial Services Industry”, Applied Financial Economics 8 (2), 175-180.
