Received date: March 18, 2014; Accepted date: April 21, 2015; Published date: April 28, 2015
Citation: Saanoun IB, Riahi Y, Othmani S, Arab MB (2015) Private Benefits of Control: Contestability versus Coalition. Bus Eco J 6:151. doi:10.4172/2151-6219.1000151
Copyright: © 2015 Saanoun IB, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Business and Economics Journal
Despite the importance of private benefits of control in France and the persistence of concentration of ownership, the link between private benefits of control and ownership structure is still a frequently mentioned and addressed issue in the literature. Our study contributes to the literature by providing an empirical analysis of the harmonies between shareholders and their effect on private benefits of control. In this paper, we concentrate on this relationship by developing a model where the coalition or the contestability of control influences the size of private benefits of control. On a sample of 44 French listed companies for the period 2001- 2011, the empirical analysis produces some interesting results. It demonstrates that the concentrated ownership is a factor that promotes divergence of interests between shareholders. Controlling shareholders hold the most important part of property rights compared to the minority who will further expropriate corporate resources. We conclude that shareholders' agreement is not a good Corporate Governance mechanism because it could be destructive of the firm’s value. In other words, if there were no agreement between the two largest shareholders, control and mutual monitoring of each other help to protect minority interests and can positively influence the firm’s value.
Private benefits of control; Ownership structure; Contestability; Coalition
In modern finance, the private benefits of control are a subject of growing interest at the empirical and theoretical level. Several questions arise about the influence of the controlling shareholders on the private benefits and the means used to extract them.
Prior studies found that private benefits of control are due to the opaque portion of the informational environment. Insiders benefit from access to internal information which allows them to create channels to obtain profits . These advantages tend to increase in countries with weak investor protection and when large shareholders possess voting rights greatly exceeding their rights to cash flows [2-4]. Based on the agency theory, the principal is the controlling shareholder and the agent is the smaller holder of capital. In this relationship principal/agent, the first is supposed to take decisions in the interest of the second, seeks to provide various benefits at the expense of minority interests and benefits from their privileged position in the firm. Controlling shareholder receives additional revenue which can be divided into two categories: pecuniary as Excess salaries, perks and larges bonuses and nonpecuniary as the prestige and social status, the ability to employ family members and to appoint them on the board [5,6], the independence from superiors and even personal relationships [4,7].
Private benefits of control are presented as materialization from unfair treatment of shareholders . They can be divided into the benefits of the ownership and other related to the control. From this decomposition, we could designate beneficiaries who are the holders of blocks and managers [8,9]. They are able to affect a number of important corporate decisions, such as investment, the growth, debt and executive compensation .
Numerous recent studies have so far suggested that large shareholders control a significant number of firms in many countries, including developed countries [11,12]. In the economies of these countries, controlling shareholders by focusing their property tend to monitor managers and to intervene directly in the decisions of the company. Meanwhile, controlling shareholders can also divert corporate resources for private gains at the expense of other shareholders under the separation of voting and cash flow rights .
This situation is particularly important in the case of weak protection of minority rights. In such situation, controlling shareholders whose voting rights greatly exceed their rights to cash flows are frequently encouraged to seek personal benefits at the expense of minority shareholders [2,4,14]. In fact, concentrated ownership often appears to alleviate the agency problems between shareholders and managers .
Some researchers point out that when the ownership concentration is above a certain level, large shareholders are motivated to use their voting rights, or to exploit the resources of companies or to enjoy solely of companies advantages in spite of the minority shareholders and managers [2,4,7,14,15]. Controlling shareholders transfer the resources and the interests to other companies under their control . These transfers can also take place for taxation purposes . In fact, by increasing their shareholding, shareholders may practice a more active control and take more important management decisions. They can consequently benefit from shared or private higher benefits of control.
In fact, concentrated ownership often appears to alleviate agency problems between shareholders and managers (La [11,18] endorse that the strong participation of the entrepreneur is associated with a higher valuation of the company and a weakened expropriation of minority shareholders.
The literature on corporate governance has recently begun to analyze the strategic interaction among multiple large shareholders. This approach goes beyond the more traditional “large shareholder” framework, where the ownership structure is made up of one dominant owner and many small dispersed investors. The recent Report presented by the European Commission  includes agreements among the Control Enhancing Mechanisms (CEMs) employed by European companies to deviate from the “proportionality” (or “one-share-onevote”) principle. In the sample of 464 companies analyzed in the Report, 44% have one or more CEMs, the most common of which are pyramid structures (27% of occurrences of CEMs), followed by multiple voting rights shares (21%) and by shareholders’ agreements (14%). The later have a quite significant presence in Italy and Belgium, followed by France and Spain: 23% of the Italian companies included in the sample have shareholders’ agreements; the percentage is 31% for Belgium, 18% for France and 13% for Spain.
Bennedsen  highlights that a controlling coalition of shareholders can have a larger cash flow stake than an individual owner, inducing it to internalize the consequences of its actions; the same incentive alignment argument points to a positive role for shareholders’ agreements. On empirical grounds, Volpin  finds that agreements in Italian listed companies lead to a better governance (measured by the sensitivity of managerial turnover to firm performance) and to a higher Q ratio. A contrasting view is provided by Gianfrate , analyzing a sample of Italian listed companies: he finds that the announcement of a new agreement is a “bad news” for the stock market; he also finds that agreements are typically used to get a leverage effect (measured as the ratio between board rights and voting rights), enjoyed mostly by the first shareholder.
Laeven  using a large sample of companies listed in 13 European countries, show that a more even distribution of voting/cash flow rights among several large shareholders exerts a positive effect on firm value.
A significant number of studies investigated the determinants of private benefits, but the analysis of the coalition as an explanatory factor has not been analyzed until now. For this reason we are trying to offer some insight into understanding this fill this relationship and to explain the effect of shareholders agreements on private benefits of control.
This work contributes to the literature by providing an empirical analysis of shareholders’ agreements, which play a significant role in the governance of European corporations. In contrast, we analyze the impact of control contestability over the last decade.
Thus, our theoretical review of the literature will attack particularly three points:
The effect of the control threshold;
The effect of the coalition;
The effect of contestability.
Empirically, we can distinguish few previous researches treating this issue. Our contribution is in the extensions works on the impact of the ownership structure on the private benefits of control. We formulate a model in which the ownership structure is dependent on the existence of a coalition. Indeed, if the coalition, represented by the shareholders' agreement between the two majority shareholders exists, the ownership structure will take the value of the pact. In the opposite case, it will be evaluated by the contestability of control, the difference between the property rights held by the controlling shareholders.
France provides a typical example of a civil law country.
Due both to its weak legal protection rules and to its inefficient law enforcement system, the French judicial environment is deemed not to adequately shield minority shareholders. Furthermore, most French listed firms show a highly concentrated ownership structure and are often controlled through pyramiding and/or double voting shares. Large block-holders either manage the firm directly or tightly monitor delegated managers. The work of Windolf  highlights the binary division of firms in France: half of them are being part of the State sphere, the other half being family-controlled companies. The largest shareholder owns control rights in excess of its cash flow rights . The disparity between voting rights and cash flow rights gives the largest shareholder incentives to seek private benefits at the expense of minority shareholders [12,25].
La Porta  stipulated that the French legal environment is not effective in protecting the rights of minority shareholders. Thus, the level of expropriation is particularly significant and private benefits are much higher than in other developed countries, often exceeding 28% of corporate values .
Our study is made in the period 2001-2011 to benefit from several reforms. The NRE law of 15 May 2001 and the Breton Act of 26 July 2005 reinforce transparency by requiring the inclusion in the annual report a description of the fixed, variable and exceptional components of compensation and benefits.
The rest of the paper is organized as follows: Section 1 covers a literature review of the impact of the existence of controlling shareholders on private benefits of control and discusses the coalition and contestability of controls effects. Section 2 addresses the relationship between block ownership and private benefits of control. Section 3 describes the data and presents the research design. Section 4 summarizes the empirical results. A brief conclusion follows with implications of the findings and suggestions for future research.
Level of control and private benefits of control
There is empirical evidence on the relationship of the separation of ownership and control and the expropriation to minority shareholders. This evidence has been shown to vary among countries. For example, Zingales  found that in Italy, expropriation is large and consistent with voting power. Similarly, Franks and Mayer  found that in Germany, private benefits of control arise when shareholders have higher voting than cash flow rights. By contrast, evidence from Sweden suggested that separation of ownership and control does not result in substantial expropriation by largest controllers . In the UK, the agency conflict caused by voting controls differs from that generally found in Continental Europe. In the latter, expropriation of minority shareholders might be the key agency problem related to ownership concentration. As there is extensive protection to minority investors in the UK, the agency problem originates from the lack of managerial ownership concentration, which requires codes to prevent managers from benefiting against shareholders .
The issue of sharing control has been the subject of a recent focus. From a theoretical perspective, some authors argue that a possession beyond 50% of the voting rights can dominate the society . Recommend that 50% of the control rights are really needed to actually control a corporation.
La Porta  studied the control of a sample of the 27 countries in 1995. They noted that in more than 70% of cases, 20% of the voting rights are sufficient enough to isolate controlling shareholder and to exercise control over the company. This level may even be reduced to 10%, and it is sufficient to control a large firm in countries like the United States, the United Kingdom or Japan. According to these authors, the level of current 20% in 75% of cases allow the controlling shareholder to accomplish all the decisions he wishes freely on condition that there is no other shareholders with more than 10% of the rights.
LeMaux  analyzed a sample of French companies and used a level of 40% set for a majority stake. He concluded that the expropriation of minority shareholders by a majority coalition is confirmed. He also found that, to the predefined threshold, rather than monitoring managers on behalf outsiders, the controlling shareholders use the powers which they dispose to extract a high level of private benefits.
At this stage, we test the following hypothesis:
H1: The concentrated ownership increases the private benefits of control
The literature on corporate governance has recently begun to analyze the strategic interaction among multiple large shareholders.
Contestability and private benefits of control
Ever since the publication of Berle  there has been agreement that the main governance problem in listed corporations, given their dispersed ownership, is the separation of ownership and control in the sense that management controls the firm without any ownership stake. Jensen and Meckling  modeled the main agency problem arising from this separation. In the 1980s, the theoretical research of Demsetz and Lehn  established a second agency problem in listed corporations that arises between a large shareholder and a widely dispersed ownership.
How a large shareholder affects a firm’s performance and valuation has elicited two opposite views. Some have argued that large shareholders have an incentive to seek private benefits of control and expropriate minority shareholders (tunneling). Others claim that incentive large block holders have yields to monitor management good governance since their ownership stakes are high and most of the benefits of good governance will be transferred to their own pockets, with the gains of better performance more than compensating for the costs of monitoring management.
Bloch and Hege also claim that multiple block holders can compensate for the poor legal protection for minorities. They argue that the relevant concept of control is the contestability of the incumbent shareholder’s position and that corporate control is contestable if the incumbent cannot increase the level of control rents without losing in a control contest. In their model, the presence of two large block holders acts to limit private rent extraction and attracts the votes of the minority shareholders when proposals are contested.
The non-cooperative competition among the controlling shareholders (the two most important controlling shareholders) plays an important role in mitigating the agency conflict because of the fear of losing control. These authors concluded that private benefits increase with the difference between the proprietary rights held by the two largest shareholders and then the firm value must be assigned [34-36].
According to Burkhart and Lee , the contestability of control and partial ownership concentration are two mechanisms for reducing conflict between internal and external shareholders. Thus, the possibility that an outsider person of the society might adversely affect the control could have a disciplinary effect on the controlling shareholders. Into such circumstances, controlling shareholders have an incentive to keep of private benefits of control and to reduce the wealth of the minority and the value of the firm.
Attig et al.  have arrived to conclude on a sample de11 65 companies belonging to thirteen countries of Western Europe and eight from Asia, that the existence of several controlling shareholders alleviates agency conflict sand reduces information asymmetry.
In the same sense, Dhillon and Rossetto provide that the presence of a second controlling shareholder can only be beneficial whether it exercise sits oversight role and guidance the company to ward profitable projects, through these votes into General Assembly.
Guti'errez and Pombo  also concluded from a study conducted on a sample of 233 non-financial companies in Colombia for a period ranging from 1996 to 2004 that a strong contestability of control between the controlling shareholders helps to limit the extractions of the private benefits.
On a sample of 110 French companies,  concluded that firms with more than one controlling shareholder have an average benefit of control lower than firms with no second large shareholder. These results seem to confirm the expectations that the ability of the controlling shareholder to extract private benefits is attenuated in the presence of a second large shareholder. In fact, sharing the control between several large shareholders might reduce the conflict of interests between large and minority shareholders, and hence decrease wealth expropriation [35,38].
Thus, we formulate our second hypothesis as follows:
H2: a strong contestability of control will lead to a decrease in private benefits of controls.
Coalition of control and private benefits
The effect of coalition over private benefits of control has been analyzed by several researchers in different contexts....
Le Maux  studied the relationship between the type of control of a listed company and private benefits on a sample of companies in the SBF 120 over three consecutive years (1998, 1999 and 2000). He defined coalition control or control group as an entity composed by managers and controlling shareholders who have access to all the tools and mechanisms for management and control (eg internal audit, strategic decisions ...) and benefit from large information in the controlled company. It examines the coalition control inside Board contesting that "the traditional view of the general assembly, which can be considered a place of strong protests" and confirms the hypothesis of the presence of majority control coalition allows to increase the amount of regulated agreements proofing that the highly concentrated structure significantly explains the private benefits. "
Acting in favor of this coalition, controlling shareholders create asymmetry information and make a barrier to the intrusion of new members to the Board and pay particular attention to the threat of taking external control.
Contrary to what has already been mentioned, Boubaker found in a sample of 510 French listed companies whose third are affected by a shareholders' agreement, that the dummy variable "pact" positively but not significantly influence performance of the company. A year later, in 2006, Roosenboom and Screamed proposed a fairly accurate description of shareholders' agreements on a sample of 299 French companies over the period 1993-1999 and confirm the hypothesis that the existence of an agreement between shareholders positively impacts the value of the firm. Similarly, Mancinelli and Ozkan  showed that Italian firms will distribute higher dividends when affected by a voting agreement between shareholders, which reflects on their lower expropriation of minority shareholders.
Gianfrate  showed that the event “signature of shareholders Agreement’s result in a significantly negative abnormal return, the adverse event" end of a Shareholders Agreement“ could be translated as a significantly positive abnormal return.
In 2012, Carvalhal  conducted its study on a sample of companies in Brazil and analyzed the effect of agreements between shareholders in listed companies on the value of the company. Their results are that the agreements are intended to reduce conflicts between controlling shareholders and minority, which positively influences the value of the company.
In the same vein, Bargeron  highlighted the agreements between the shareholders and found that these agreements are associated with ownership concentration and higher managerial ownership. The results demonstrate that such agreements are not only to reduce information asymmetry; they are rather designed to increase shareholder value and reduce expropriations.
Thus, we will opt for the following hypothesis:
H3: the presence of a coalition decreases the private benefits of control.
Our goal is to analyze the impact of the ownership structure on the private benefits of control. To do this, we are dealing with:
• The existence of a controlling shareholder
• The coalition control whose formation is marked by a shareholders;
• The contestability of control which means, the power relationship between the first two shareholders.
Data sources and sample
Our sample is composed of 84 companies belonging to the SBF 250 index for a period of eleven consecutive years from 2001 to 2011, representing all sectors of the French economy. However, various adjustments were necessary to exclude commercial or industrial companies, which have experienced a significant change in scope, banks and insurance companies, with a particular accounting system. Data on the composition of the board of directors and shareholders have been obtained manually from the reference website (www.amffrance. org); otherwise we consult the annual reports.
Recall that the French context is particularly interesting to analyze for different reasons:
Governance mechanisms have seen developments, under the in pulse of reports. Thus listed companies have been motivated by the appointment of independent directors to their boards and multiplied the creation of specialized committees, including compensation committees.
NRE of 15 May 2001 and the Breton Act of 26 July 2005 reinforce transparency by requiring the inclusion in the report of a description of the management of fixed, variable and exceptional component salaries and benefits of the officers.
PBC: private benefits of control, CONC: refers to the existence of the controlling shareholder, TANG: tangible assets, ROA: return on assets, SIZE: the size of the company, DEBT: leverage, Q: growth opportunities.
To properly analyze and differentiate between controlling shareholders and those interacting in competition, we combine in a single model the dispersion, the concentration of ownership of controlling shareholders with or without coalition. We materialize the control coalition by the existence of a shareholder agreement and in this case the ownership structure will be measured by the ownership of two controlling shareholders who are recorded on pact as C12. Otherwise, we assume that the first two shareholders are competing and the ownership structure will be estimated by the difference between the properties rights of the first two controlling shareholders noted: C1C2.
Drawing on the work of Bloch and Hege, we can develop the following model:
If there is a shareholders' agreement between the controlling shareholders
Private benefits of controls are explained by the structure of ownership in a single equation as follows:
The indices i and t represent indicators for individuals and the time index. In our case, the individual index is associated to French companies and the time index presents the period of study (2001-2011).
μit: present the error term of the regression
Where PBC: private benefits of control, C1C2: the difference between the percentage of common shares owned by the two largest shareholders of French companies C12: the sum percentage of common shares owned by the two largest shareholders of French companies, CONC: refers to the existence of the controlling shareholder, TANG: tangible assets, ROA: return on assets, SIZE: the size of the company, DEBT: leverage, Q: growth opportunities.
Measurement of variables
In the following subsection, we will list the different variables used to test our hypotheses.
Measures of private benefits of control: To measure the extent of private benefits of control, we used the amount of related party transactions and excessive executive compensation. We deliberately chose these two heterogeneous measures to capture maximum illegitimate transfers and not to compare or combine all the results. These two measures of private benefits have the advantage of being observable through the annual reports of companies, due to a legal obligation.
Few studies have estimated the private benefits of control using more direct measures: the value of the voting rights [26,44] Control premiums [15,45], wages excessive executive compensation  and related party transactions [40,46].
a. Related party transactions: Le Maux and Dahya [8,46] used the amount of related party transactions as a direct measure of private benefits. In fact, it appears that through related party transactions, not only the leaders, associates with a significant proportion of the voting rights may also impose on the society of which they are shareholders prejudicial convention.
b. Excessive executive compensation: Excessive compensation goes directly to the leader, which is possibly the controlling shareholder of the company, expropriate the company and minority through high remuneration compared to its industry. We will follow the approach of Belanes  to estimate the private benefits of control by excessive salary that reflects the excess wages due to expropriation. To determine the amount of overpaid, we calculate the average earnings by sector and excessive compensation which is the difference between the total remuneration and the amount calculated before.
Measures of interest variables
CONC: binary variable that takes 1 if there is a shareholder who holds more than 40% of the capital (the average for our sample of French listed companies), 0 otherwise. This is our measure ewe use for ownership structure.
LeMaux used a dichotomous variable which takes 1 if there is a shareholder holding a proportion more or equal to 40% of the company capital. We will expect that ownership concentration increases the amount of private benefits of control.
C1C2: The difference between the proportions of ownership owned by respectively the first and second controlling shareholder. Bloch and Hege consider that two large shareholders in competition will limit their consumption of private benefits and they came to the conclusion that private profits are increasing with the difference in ownership of the first and second largest shareholder and the value of the firm must be affected [34-36]. Thus, the existence of a second shareholder with a lower proportion compared to that held by the first increases the information asymmetry and the ability to extract the wealth of the firm.
CAOL: A binary variable that takes1 if there is a shareholder agreement between the first and the second controlling shareholder, 0 otherwise. We will expect that the impact of the presence of coalition control materialized by the existence of a shareholder agreement on the private benefits of control is positive. A highly concentrated structure significantly explains the private benefits. We only work on specific and declared agreements to stock exchange authorities by the signatory shareholders [1,8].
C12: This is the proportion of capital held by the first and second largest shareholders. Le Maux, Weifeng et al., and Belanes et al., demonstrated that the private benefits of control are important in firms with concentrated capital. As contestability index is calculated using the proportions of the two largest shareholders, in assessing the ownership structure, we have chosen the ownership of the two largest shareholders.
Measures of control variables: Tangible assets: Rated "Tang" is defined as the ratio between fixed assets and total assets. For holders of the blocks, it is more difficult to divert resources from the company when the assets are observable such as tangible assets [8,15,47].
Debt: Rated "Debt" and measured by total debt to total assets. This is an internal control mechanism. Under the assumption of free cash flow, debt reduces the agency costs related to cash flow available for internal and therefore, private benefits of controls used by controlling shareholders .
Growth opportunities: Noted "growth": measured by Tobin's Q that is defined as the ratio between the total market value of the company with the accounting value of the debt by the sum of the book value of equity and the book value of debts. We expect that the controlling shareholders will not encourage increasing their wealth in private benefits when growth opportunities are high .
Performance: Noted "ROA": we chose the return on assets ratio as an indicator of the wealth produced, which is the ratio of operating income to total assets (ROA). We assume that higher allows the controlling shareholder of appropriate resources to benefit the firm thus a higher level of private benefits .
Size: We chose Neperian logarithm of book value of the total assets of the firm as a size company indicator. We will expect that the relationship between the amount of private benefits of control and firm size is positive insisting that big companies offer more pecuniary and non-monetary benefits, [8,15].
Descriptive statistics and univariate analysis
The following Table 1 provides the descriptive statistics (mean, standard deviation, minimum and maximum) of each variable used in our analysis.
|private benefits of control PBC||Related party transactions RPT||The amount of related party transactions divided by total assets|
|Excessive managerial compensation EXCC||Excessive managerial compensation divided by total assets managé|
|ownership structure||CONC||A binary variable thattakes 1 if there is shareholderwhoholds more than 40% of the capital (the average for oursample of French listedcompanies), 0 otherwise.|
|Coalition||CAOL||A binary variable thattakes 1 if there is a shareholder agreement between the first and second controllingshareholder, 0 otherwise.|
|Ownership Concentration||C12||the proportion of capital held by the first and second largest shareholder.|
|Contestability of control||C1C2||The difference between the proportions of ownership owned by the first and second controlling shareholder respectively.|
|Tangible assets||TANG||Fixedassets / total assets|
|Growth opportunities||Q||(Market value of equity + book value of debt) / (book value of equity + book value of debt)|
|Leverage||DEBT||total debts to total assets|
|Return On Assets||ROA||Ratio of profit beforeinterest and tax to total assets|
|Size of the company||SIZE||Neperian logarithm of book value of the total assets of the firm.|
Table 1: Summary table of variables used.
Table 2 describes the values of the variables measuring the amount of private benefits of control and the variables that could have an impact on them. Excessive executive compensation (not divided by total assets) has an average value of 1125. It varies between -1238.11 and 12656 as the minimum and maximum value respectively. Thus, our sample is characterized by leaders who have over their sectors pay. Regarding the second variable, the average value of the regulated agreements is 19,416 with a minimum of 0 and a maximum of 150644.
Source : Variables: RPT - related party transactions, RPT* : related party transactions divided by total assets, EXCC: Excessive managerial, EXCC*: Excessive managerial compensation divided by total assets manager, 0 otherwise. C12:the proportion of capital held by the first and second largest shareholder control, C1C2: The difference between the proportions of ownership owned by the first and second controlling shareholder respectively. TANG: tangible assets, ROA: return on assets, DEBT: leverage, Q: growth opportunities. SIZE : Neperian logarithm of book value of the total assets of the firm.
Table 2: Descriptive statistics of depend and independent variables.
It was found that the average percentage of common shares owned by the two largest shareholders of French companies is 41.41% and varies from 0% to 99.9%. Thus, on average the two largest shareholders hold half the capital. These statistics allow the proposition that the stock ownership is locked in the hands of the two largest shareholders. The ownership stock of the French firms is closely held by the block holders although the firms of the sample belong to the SBF250 index. In fact, it is argued that there is separation of the control rights from the cash flow rights when companies are listed on the stock exchange, and the non-controlling owners are represented by a large number of minority shareholders. However, this explanation is valid for only a tiny minority of large groups .
Size measured by the logarithm of total assets shows an average value of 21, 68 while the average of debts is 57% of total assets. The longterm debt can reach a maximum value of 100% of total assets. Tangible assets are on average of 16% with a minimum of 0 and a maximum of 99%. The rate of return on assets is on average of 5%. It varies between -27% and 59% shown in Table 3.
Source : Variables: CONC- A binary variable that takes 1 if a shareholder who holds more than 40% of the capital (the average for our sample of French listed companies), 0 otherwise, CAOL:A binary variable that takes 1 if there is a shareholder agreement between the first and second controlling shareholder.
Table 3: Descriptive statistics for quantitative variables.
For our sample composed only of French companies, there is at least one shareholder holding more than 40% in 71% of companies. This rate is similar to that found by Faccio et al., who showed that 64.75% of French companies were dominated by one large shareholder.
Coalitions of control in our study have been materialized by the existence of the pact in almost 38% of our sample. This result seems consistent with Boubaker who noted the presence of shareholder agreements in the third of French companies’ sample.
The correlation test shows that there is a positive correlation relationship between the two measures of private benefits of control and (CONC). Such result is compatible to Bebchuk’s study, which assumes that firms with high levels of private benefits of control are likely to have more concentrated ownership structure.
This relationship is significant at respectively the level of 1% and 5% for excessive executive compensation and related party transactions. The level of private benefits of control increases with the concentration of ownership. This result is consistent with those of [7,8,15,16].
Our finding contradicts a priori those of Jensen and Meckling who found that the major contribution to rights to benefits by the controlling shareholder increases its commitment to the distribution of dividends and reduces incentives for expropriation, which is likely to disprove our first hypothesis.
Table 4 also points out a negative correlation between the two measures of private benefits of control and assets tangibility. Such correlation is expected because we assume that firms with more tangible assets have lower private benefits of control.
Source : RPT - related party transactions divided by total assets, EXCC: Excessive managerial compensation divided by total assets manager, CONC: A binary variable that takes 1 if a shareholder who holds more than 40% of the capital (the average for our sample of French listed companies), 0 otherwise, CAOL:A binary variable that takes 1 if there is a shareholder agreement between the first and second controlling shareholder, 0 otherwise.C12:the proportion of capital held by the first and second largest shareholder control, C1C2: The difference between the proportions of ownership owned by the first and second controlling shareholder respectively.TANG: tangible assets, ROA: return on assets, DEBT: leverage, Q: growth opportunities. SIZE : Neperian logarithm of book value of the total assets of the firm. ***, **, * denote significance at 1%, 5%, and 10% level, respectively.
Table 4: Pairwise Correlations.
There is a negative correlation between leverage and profitability showing that the most profitable firms’ use less debt. However, the correlation between private benefits of control and performance is positive stipulating that most successful companies are those whose controlling shareholders benefit most from their resources.
The results presented in Table 4 show that the concentration of ownership weakens the large size of the firm. This result allows us to confirm the results of Villalonga and Qu that showed that great French companies have more dispersed ownership structures than small and medium companies.
After the realization of econometric tests: Pearson's correlation matrix and vif ’s test, test for the presence of individual effects, Hausman test and hetero scedasticity test, it would be wise to present the results of our models.
As already mentioned, we use two proxies for private benefits of control: related party agreements and excessive executive compensation both deflated by total assets. We estimate our empirical model by the method of generalized least squares. The estimate of the dependent variable related party transactions allowed us to be more attractive in terms for significance as those found using excessive executive compensation results.
Table 5 shows that the effect of ownership structure is consistent with the univariate analysis and states that the existence of a shareholder holding more than 40% of capital "CONC" is positive and significant at the 1% level for both measures of the dependent variable. Therefore, the size of private benefits of control increases with the concentration of ownership beyond 40%. Controlling shareholders are expected to benefit from the business under their control to increase their wealth and transfer private benefits outside the company to offset the cost of ownership concentration.
|coefficient||Z||Value P>|Z|||coefficient||Z||Value P>|Z||
Source : RPT-related party transactions divided by total assets, EXCC: Excessive managerial compensation divided by total assets manager, CONC: A binary variable that takes 1 if a shareholder who holds more than 40% of the capital (the average for our sample of French listed companies), 0 otherwise, .TANG: tangible assets, ROA: return on assets, DEBT: leverage, Q: growth opportunities. SIZE : Neperian logarithm of book value of the total assets of the firm. ***, **, * denote significance at 1%, 5%, and 10% level, respectively.
Table 5: The impact of the ownership concentration on private benefits of control.
Thus, the concentration of ownership is a factor that promotes the interest divergence of shareholders holding the largest share of property rights over those of minority, increasing the risk of expropriation, there by affirming our first hypothesis.
These results match partly those of Belanes  who found that the concentration of ownership in the hands of the top three shareholders has a significant effect on private benefits of control. The results are compatible to the hypothesis that the higher the stock ownership is, the higher is the level of private benefits of control. Such result is already approved by the studies of [7,8,15,16].
Meanwhile, the larger the stock ownership is, the greater are the private benefits in terms of the frequent use of related-party transactions. In fact, when the stock ownership is concentrated enough, the control level is powerful enough. The shareholder is appealed to lonely extract corporate benefits in spite of both minority shareholders and managers [2,3,4,7,15]. Block-shareholders expropriate such private benefits either to increase their own wealth or to provide their controlled companies with necessary funds, or even for tax purposes [16,17]. But above all, block-shareholders are appealed to siphon off higher private benefits of control so as to safeguard their wealth and compensate the cost of concentrating their wealth within one firm and the risk to all lose. In fact, most French firms are family-owned. Founders are often concerned with the survival of their firms and aim at protecting their legacy for future generations and preserve the family wealth accumulated through decades as well.
This result is contradicted in other studies. A holder of blocks may be encouraged to make an effort in monitoring managers since the benefits (in proportion to their ownership of rights to capital), can reasonably offset the cost of this effort. La Porta concluded that the strong participation of the Contractor's rights to the benefits is associated with a higher valuation of the company and a decreasing expropriation of minority shareholders .
The estimation results also reveal the significance of tangible assets for both measures of private benefits of control. The presence of tangible assets negatively affects private benefits of control. Controlling shareholders are more difficult to divert the assets of the firm when they are easily observable. These results confirm those of [9,45] who found that the tangible are negatively related to private benefits of control.
Regarding the leverage ratio, it has a negative and significant coefficient for excessive executive compensation and a positive significant coefficient for related party transactions highlighting the influence of the firm leverage depends on the identity of the beneficiary of private profits.
The negative effect of debt on excessive executive compensation is explained by the fact that controlling shareholders choose financing by debts because loans do not change the ownership structure of the company. This option is particularly relevant to family not wishing the coming of new owners and the loss of control. For related party transactions, the positive relationship can be explained by the fact that through leverage, shareholders will try to transfer the maximum of resources to other firms under their control, through regulated agreements. So, the debt has no longer the disciplinary role of governance, on the contrary, it increases the private benefits of controlling shareholders. These results confirm whose main contribution is that the relationship between the impacts of debt policy on the private benefits of control depends on the identity of beneficiaries.
About growth opportunities, the coefficient of this variable is significant and negative for both proxy variables of private benefits of control, showing that the controlling shareholders try to capture and extract the wealth of the company when growth opportunities are low. However, the coefficient is not significant for excessive executive compensation and significant for related party agreements.
The coefficient of firm size is significantly negative. Such a result shows that the increase in firm size is associated with a decrease in the level of private benefits of control. According to Dyck and Zingales studies  we argue that large firms are likely to have a lower level of private benefits of control. In such a situation, the controlling shareholders use all the authority available to them to extract maximum wealth in their favor and at the expense of outside shareholders.
Finally, regarding the return on assets for our two measures of private benefits of control, the coefficient of the variable is significant and positive confirming our expectations. High profitability allows the controlling shareholders to appropriate the resources to benefit the company in many forms such as excessive salaries or related party transactions. This result corroborates those of Le Maux .
In previous sections, we investigated the impact of the ownership concentration of private benefits of control based on the existence of a controlling shareholder holding more than 40% of the capital. In what follows, we will enrich our study analyzing the effect of the interaction between the two majority shareholders.
Impact of the coalition or contestability in the private benefits of control
To analyze the effect of the formation of coalition between the controlling shareholders on the extent of private benefits of control, we will refer tomodèle3.
The results presented in Table 6 shows that R2 Between is about 30.29% for excessive executive compensation and 40.13% for related party transaction and there for models have acceptable explanatory power for the total sample.
|coefficient||Z||Value P>|Z|||coefficient||Z||Value P>|Z||
Source : Variables: RPT: related party transactions divided by total assets, EXCC: Excessive managerial compensation divided by total assets manager,, CAOL:A binary variable that takes 1 if there is a shareholder agreement between the first and second controlling shareholder, 0 otherwise.C12:the proportion of capital held
by the first and second largest shareholder control, C1C2: The difference between the proportions of ownership owned by the first and second controlling shareholder respectively.TANG: tangible assets, ROA: return on assets, DEBT: leverage, Q: growth opportunities. SIZE : Neperian logarithm of book value of the total assets of the firm. ***, **, * denote significance at 1%, 5%, and 10% level, respectively.
Table 6: Estimation of the impact of the presence of a coalition and contestability in the private benefits of control.
The results in the Table 6 show the significance of the variable C12*COAL, with a positive sign which is contrary to our expectations, didn’t valid our H3 hypothesis which suggests that the presence of the coalition have a negative impact on private benefits and leaves omen the beneficial nature of the coalition as a good governance mechanism.
The coalition majority control there for uses all the powers available to it to extract maximum wealth in his favor and at the expense of outside shareholders. This result is in line with that of Le Maux  amount instead of making a reciprocal monitoring on behalf of outside shareholders the main shareholders use the powers they have to extract a high level of private benefits at the expense outside shareholders.
Our findings reinforce those of Gianfrate  that the signing of shareholders results in a significantly negative abnormal return, while the opposite event (end of a shareholders' agreement) results in a abnormal return significantly positif. No usalsocontradictVolpin  and Ozkan  that the shareholder agreements are mechanisms for good governance.
Moreover, if there is no agreement between the two largest shareholders, the value of the property that will be included in the model is the index of contestability. Its coefficient is significant and negative, which allows us to confirm our second hypothesis.
So when controlling, shareholders have rights to the cash flows with similar size, the contestability is important and the level of private benefits of control will be less. This could be explained by the fact that the controlling shareholders, who are not subject to an agreement and with relatively similar size, are subject to control and reciprocal monitoring each other to protect the interests of minority and positively influence the value of the firm. However, when the relative difference in size between the two shareholders is relatively large it makes the control of the first by the second doubtful. The risk wealth transfer is important and the value of the firm is lower. This outcome is similar to Blochand Hege model predictions and Bennedsen and Wolfenzon .
These results seem to confirm the findings of Gutierrez and Pombo  who states that strong control contestability between the holders of blocks limits the "tunneling" and extractions of the private profits.
These results confirm those of Blochand Hege, Maury and Pajuste  and Attig et al.,  which have shown that sharing control between controlling shareholders could reduce conflicts of interest between minority and majority. Thus, the contestability of control appears as a factor which limits the appropriation of earnings by controlling shareholders at the expense of minority interests.
Acting on behalf of the coalition, controlling shareholders create information asymmetry to appropriate the benefits of the company. However, the contestability of control allows reducing them.
This paper covers the recent empirical findings and theoretical reasoning for the relationship between ownership structure and private benefits of control. To test the validity of our hypotheses, we conducted an analysis in the French context with a sample of 44 French companies in the SBF 250 over a period of 11 years. Through analysis and detailed overview to the ownership structure of French companies, we could better understand the concept and extraction of private benefits of control channels. We have the advantage of using two direct proxies of private benefits of control: related party transactions and excessive executive compensation corresponding to blocks holder and managers. We contribute to the existing academic literature by demonstrating that the existence of an agreement between the controlling shareholders increases the private benefits of control. However, the contestability of control reduces the diversion of wealth from shareholders control. First, we have highlighted the impact of ownership concentration measured by a binary variable which takes the value1ifthere is a shareholder holding more than 40% of the capital on the private benefits of control for both measures. Our results confirm those of LeMaux  who found that the majority control increases the appropriation of earnings. However, we contradict the results of Jensen and Meckling , which stipulated that ownership concentration, is a factor that promotes alignment of interests and results in reduced levels of expropriation. Second, we have materialized the coalition by shareholder agreement and we expected that it accentuates the extractions. We concluded that shareholder agreement is a bad governance mechanism enhancing appropriation earnings of the company even by the controlling shareholder or the manager. However, if this latter two does not agree, the power struggle between the controlling shareholders positively affects private benefits of control with controlling and preventing the transfer of wealth. We examined the effect of ownership structure on the private benefits of control by focusing on the consequences of the existence of a shareholders' agreement materializing the formation of the coalition. The fact that most French companies are family firms, motivates us to enrich our study by examining the impact of the type of the controlling shareholders (if they are of the same type or not, institutional or families).
Make the best use of Scientific Research and information from our 700 + peer reviewed, Open Access Journals