Does Managerial Power Explain the Association between Agency Costs and Firm Value? The French Case

Abstract

This paper demonstrates whether the impact of agency costs on firm value depends on the level of managerial power using the fraction of capital held by the manager, as well as their level of voting rights. Focusing on a sample of 120 non-financial French firms incorporated in the CAC All-Tradable Index for the period 2008–2022, the first empirical analysis provides strong evidence that agency costs of equity, as measured in terms of operating expenses, administrative expenses and the agency cost of free cash flow, exert a negative impact on the firm’s market value. In a second empirical analysis, we split our sample into three sub-samples with the aim of capturing the effect of managerial power. The findings lead us to believe that the association between the agency cost measurement and the firm’s market value depend on the level of managerial power. This paper challenges prior studies by strengthening our understanding of managerial behavior (incentive, neutral, and entrenchment) in relation to shareholder wealth. Furthermore, it contributes to the recent literature by enabling a better knowledge of the disparity related to studies conducted in other countries with different governance models.

1. Introduction

Firm value presents a major concern for various stakeholders, especially when addressing the conflicts of interest between managers and shareholders. This conflict is called a principal–agent problem, where managers act as agents working for the principals (shareholders). Moreover, shareholders can control managers only through the board of directors. These agency problems of type 1 result in the emergence of specific costs, which occur when managers do not strive to boost the firm’s market value and when shareholders incur expenses to supervise and restrict the managerial behavior (Abdullah and Tursoy 2023). According to the previous literature in corporate finance, several empirical studies demonstrate that firm value is significantly affected by the presence of agency conflicts that can arise both before and after contractual agreements. They stem from information asymmetry and the challenges of creating comprehensive contracts due to limited rationality and uncertainty (Aben et al. 2021), which give rise to pre-contractual costs. Once the agency relationship is established, post-contractual risks emerge in the form of “moral hazard”. This risk arises from the principal’s lack of information about the agent’s actions, as the agent may not uphold its commitments or deliver the expected level of performance agreed upon during the formation of the principal–agent relationship. (Jensen and Meckling 1976).

Prior studies focused on the position of corporate governance in resolving agency problems of type 1. It is generally agreed that the implementation of a good system of corporate governance may guarantee that shareholders’ pockets are close to managers’ hearts. According to the literature, corporate governance presents a combination of monitoring and incentives that attempt to keep their interests congruent (Liu 2023). According to several empirical studies, there are conflicting views regarding the role of managerial ownership. From one perspective, the fraction of capital held by the manager is considered as a tool for addressing agency conflicts between managers and shareholders and, consequently, for improving firm performance. From another perspective, it can be a tool for managerial entrenchment (Jensen and Meckling 1976; Charreaux 1991; Bouras and Gallali 2017). This paper also focuses on the importance of managers’ control of voting rights in accounting for firm value. Some research reveals that voting rights form a crucial tool of corporate governance, while other research confirms that this mechanism can enhance equity-holders’ wealth (Stulz 1988; Harris and Raviv 1988; Wang 2017; Scherrer and Fernandes 2021; Demirtas 2023). Consequently, the novelty of this paper consists in providing a link between these two key elements of corporate governance through the construction of our variable of managerial power level as a means to investigate their impact on the relationship between agency cost and shareholders’ wealth. Therefore, we attempt to analyze which level of managerial power validates the concepts of the convergence of interests, neutrality, and entrenchment hypotheses. This objective allows us to better recognize the dynamics of corporate governance relying on the French case. The choice of this context is linked to the nature of the ownership structure. Indeed, the high concentration of capital, the weak role of market discipline, and the strong presence of controlling shareholders may generate higher agency costs, which play a significant role in determining the value of French firms. Thus, we chose this area of investigation given that management shareholding, in French firms, is permitted to increase their control against their equity stakes. Specifically, they may take advantage of regulations covering share classes and multiple voting rights, leading to violating shareholders’ interest. Thus, controlling shareholders have both the incentive and authority to effectively monitor managerial decision (Faccio and Lang 2002). Additionally, studying the interaction between French firms’ managerial power, agency costs, and market value can contribute to the recent literature by providing insights into the country’s specificities. It also enables a better knowledge of the disparity related to studies conducted in other countries with different governance models.

This paper is structured as follows. Section 2 reviews the relevant theory and develops our research hypothesis. Section 3 outlines the methodology. Section 4 reports the results of the empirical part. Finally, the last section provides the concluding remarks and future research avenues.

2. Literature Review and Research Hypothesis

2.1. The Impact of Agency Cost of Equity on Firm Value

Since the agency theory advanced by Berle and Means (1932), the association between the agency costs of equity and firm value has been greatly studied in the recent literature. Using 11 Iraqi listed firms for the period 2004–2024, Sdiq and Abdullah (2022) found a negative association between agency cost, as measured by operating expenses ratio, and asset utilization ratio on firm performance. This finding aligns with Hoang et al. (2019), who conducted a study on the case of 736 firms in Vietnam from 2010 to 2015. They demonstrate a negative correlation between agency costs and financial performance. Similarly, Vijayakumaran (2021) indicates that agency costs of equity directly impact investment decisions and hinder Chinese firms’ capacity to invest in profitable projects. Moreover, the findings of Kontuš (2021) demonstrate the absence of a significant relationship between the variations in agency costs, as measured by asset utilization ratio and operating expense ratio, and the firm’s market value.

For this study, we use three proxies for the costs of manager–shareholder agency conflict. Firstly, we retain the ratio of operating expenses which reflects the effectiveness of the firm’s management in controlling operating costs, such as excessive consumption of perks and other direct agency costs. As described by Wang (2010) and Chaudhary (2022), a firm is considered to be in a deficit when its generated cash inflows are insufficient to cover its operating expenses. Ang et al. (2007) demonstrate that a greater expense ratio results in a lower firm efficiency, which gives rise to a potential agency conflict. Thus, we expect that firms with a high ratio of operating expenses are presumed to experience high agency costs that oppose managers to shareholders, which, in turn, decreases firm value. Secondly, we retain the administrative expenses ratio. According to Nguyen et al. (2020), this ratio is associated with greater agency costs and, consequently, should significantly reflect the discretionary management power over the allocation of the firm’s resources. As a result, we expect that the existence of higher administrative expenses is responsible for intensifying the type 1 agency conflicts. Thirdly, we retain the agency cost of free cash flow, which reflects the presence of an over-investment problem. This measurement is recognized when a firm generates sufficient free cash flow while its market value is low (Moez 2018). Based on agency theory, the association between managerial entrenchment and the surplus of cash flow should be positive, which results in potential conflicts of interest (Jensen 1986). Additionally, Cariola et al. (2005) and Hsiao et al. (2023) argue that the issue of managerial overinvestment is rooted in the hypothesis that managers perceive the firm as a means to enhance their own human capital. Furthermore, they may utilize this surplus of cash flow for opportunistic purposes rather than distributing dividends. Building upon prior research, the current study defends a negative association between the three selected measures of agency costs and firm value.

Hypothesis 1.

Agency cost of equity is inversely related to firm value.

2.2. The Impact of Agency Cost of Equity on Firm Value under Managerial Power Level

In the present study, we discuss two indicators that were used to construct our variable of managerial power level. Firstly, we retain the fraction of capital held by the manager as an effective way to boost firm performance. Several researchers agree that management shareholding offers a potential remedy for addressing the type 1 conflicts of interest. Given its effective advantages in addressing principal–agent problems, the previous literature predicted a positive correlation between managerial ownership and firm value over some range, but once that range is exceeded, the relationship becomes negative. Overall, there are three contrasting hypotheses of managerial ownership documented in the literature, namely the convergence of interests, the neutrality, and the entrenchment hypotheses. Regarding the first hypothesis, management shareholding acts as a disciplinary process designed to reduce the agency costs of equity. For instance, Jensen and Meckling (1976) suggest that a high level of managerial shareholdings ensures the convergence of interests between managers and shareholders. Furthermore, the authors document that the concept of zero agency cost is associated with firms whose managers own 100% of the capital. By examining a sample of 1708 American small firms, Ang et al. (2007) documented that the divergence of interests changes inversely with managerial ownership. For their part, Hermalin and Weisbach (1988) highlight the convergence of interests hypothesis when managerial ownership ranges between 0% and 1% and between 5% and 20%. In the same year, Morck et al. (1988) made a significant contribution to the field. Through their investigation into the effect of managerial shareholdings on firm value, the authors observed that the Tobin’s Q ratio exhibits an increase specifically when managerial ownership falls within two specific ranges: 0 to 5 percent and above 25 percent.

Considering the second hypothesis of neutrality, Demsetz and Villalonga (2001) argue that managerial ownership does not have any association with firm value. With respect to the entrenchment hypothesis, managerial ownership encourages managers to entrench themselves. The literature review identified four major types of entrenchment strategies adopted by executives, namely specific investments, information manipulation, resource control, and the development of relational networks. According to Stulz (1988), the possession of a substantial percentage of the capital enables managers to mitigate control mechanisms, resulting in a decline in firm value. More specifically, an increase in equity holdings by CEOs is associated with a decrease in shareholder rights or an increase in managerial entrenchment. In the same vein of research, McConnell and Servaes (1990) document that a manager becomes entrenched when their ownership of the firm’s capital reaches or surpasses the threshold of 50%. Through an analysis of the largest 500 US-listed firms from 1988 to 2015, Fabisik et al. (2021) report that firms which lack liquidity and are controlled by managers with higher ownership tend to suffer from significant information asymmetries and agency problems. This, in turn, may negatively affect the firm’s market value as calculated by Tobin’s Q ratio. In the same vein, Nguyen et al. (2020) report that when managers hold a higher level of ownership, they can take advantage of their position to give preferential treatment to their relatives by transferring lucrative projects. Using a sample of 281 listed firms in Vietnam from 2013 to 2018, the authors argue that the root cause of this conflict among shareholders is driven by self-interest. In the UK context, Allam (2018) mentions that a higher level of managerial ownership gives them more decision-making power. Thus, managers can neutralize the internal governance mechanisms. In a similar vein, Kouki and Said (2011) endorse this idea in the context of French firms. Using 246 French-listed firms for the period 1997–2007, the authors document that once the manager holds more than 80% of the firm’s capital, they tend to profit from their majority positions by maximizing their private profit instead of other stakeholders’ interests. Through their comparative study between a sample of 159 French-listed firms and 203 U.S. firms during the period from 2002 to 2010, Bouras and Gallali (2017) found a non-linear, inverted U-shaped relationship between managerial ownership and Tobin’s Q in the French context and a quadratic relationship for American firms. The authors argue that this divergence between governance models is explained by the discrepancy level in terms of ownership concentration.

Secondly, voting rights are considered to be the primary contractual right that shareholders own. In shareholder assemblies, voting is perceived as a crucial tool of corporate governance. However, this mechanism can enhance equity-holders’ wealth. The managerial voting power is observed when managers’ voting rights exceed their ownership percentage. According to Burkart and Lee (2008), divergences from the one-share-one-vote rule stimulate shareholders to prioritize their own interests, which reduces the effectiveness of corporate governance mechanisms. More specifically, Eckbo and Verma (1994) document that dividends payment decreases as the managerial voting power increases, which intensifies the agency costs of equity. This finding is consistent with Wang (2017) and Scherrer and Fernandes (2021), who indicate that the increment in the value of voting rights affect negatively firm value. This situation can facilitate managerial entrenchment by protecting them against market discipline. In addition, they gain resistance from hostile takeover and the managerial labor market. Based on the important contribution of Stulz (1988), the portion of the managerial voting rights is a significant component of the ownership structure which affects the firm’s market value. More importantly, the higher the fraction of voting rights, the lower the value of the outsider’s shares. Consequently, managers can benefit more from the probability of a takeover attempt by making it costlier for outside shareholders through the “golden parachute”. Similarly, Harris and Raviv (1988) document that managers may want to modify the firm’s capital structure to control a greater portion of the voting rights. This result is complementary to that of Ehikioya et al. (2021) in the context of Nigerian firms. The authors argue that managerial entrenchment is detectable when they increase their voting power and control.

From the previous discussion, the fraction of capital held by the manager, as well as their voting rights, are used to construct the variable of managerial power. The objective is to demonstrate whether the impact of agency costs on the firm’s market value depends on the level of managerial power. Particularly, we seek to explore if the entrenchment hypothesis, the interest-alignment hypothesis, and the neutrality hypothesis depend on the level of managerial power. Thus, we predict the following hypothesis:

Hypothesis 2.

The effect of agency cost of equity on firm value depends on the level of managerial power.

3. Methodology Framework

In the following sub-section, we report the sample used, the variables investigated, and the econometric frameworks used in the current study.

3.1. Data and Sampling

The initial sample contains all the 202 firms belonging to the CAC All-Tradable Index. Due to their different financial structures, 11 financial firms were removed. Furthermore, firms with missing data are also excluded. The final sample includes 120 French firm over the 2008–2022 period. We collect financial data from the Worldscope database. Data on managerial ownership structure are obtained from annual reports. Referring to the CAC All-Tradable Index,

Table 1. Sample distribution per sector.

	Number of Firms	Weight%
Technology	26	21.66
Consumer Discretionary	22	18.33
Industrials	20	16.66
Consumer Staples	15	12.50
Health Care	9	7.5
Utilities	8	6.66
Basic Materials	7	5.83
Real Estate	5	4.16
Energy	4	3.33
Telecommunications	4	3.33
Total	120	100

Source: https://www.euronext.com/en/for-investors/indices (accessed on 8 September 2024).

categorizes the final sample by sector.

3.2. Variable Definition

Table 2. Variable definitions and sources.

Dependent Variables	Symbol	Definition	Sources
Firm Value	Tobin’s Q	ln ((total assets +market capitalization − total common equity)/total assets)	Worldscope
Independent Variables	Symbol	Definition	Sources
Operating Expenses	OPE	Total operating expenses divided by total sales	Worldscope
Selling, General and Administrative Expenses	SGA	Total administrative expenses divided by total sales	Worldscope
Risk of Free Cash Flow	RFCF	Dummy variable that takes 1 if, simultaneously, Tobin’s Q is lower than the sector average and free cash flow is positive, and zero otherwise.	Authors’ calculations
Managerial Power Level 1	MPL1	Binary variable that takes 1 if, simultaneously, the managerial ownership ranges from 0 and 20 percent and the managers’ control of voting rights surpasses their ownership percentage, and zero otherwise.	Author’s calculations
Managerial Power Level 2	MPL2	Binary variable that takes 1 if, simultaneously, the managerial ownership ranges from 20 and 50 percent and the managers’ control of voting rights surpasses their ownership percentage, and zero otherwise.
Managerial Power Level 3	MPL3	Binary variable that takes 1 if, simultaneously, the managerial ownership ranges from 50 and 100 percent and the managers’ control of voting rights surpasses their ownership percentage, zero otherwise.
Control Variables	Symbol	Definition	Sources
Dividends	DIV	Cash dividends/(Net income before preferred dividends − preferred dividend requirement)	Worldscope
Sales Growth	SG	(total sales (N) − total sales (N − 1))/total sales (N − 1)	Worldscope
Return on Assets	ROA	Net income divided by total assets	Worldscope
Firm Size	SIZE	The logarithm of total assets	Worldscope

Source: Own construction based on Worldscope data and annual reports.

clearly presents the measurement and definitions of the variables used in this study. As dependent variable, we used Tobin’s Q to capture the firm’s market value. Several studies consider that this measure demonstrates the firms’ future growth and plays an important role in explaining managerial performance. Additionally, it serves as an ex ante performance measure as it is assessed based on the current market value. Gatzert and Schubert (2022) demonstrate that this ratio serves as a measure of investor confidence. A higher Tobin’s Q value reveals that the firm has larger potential and development prospects. Based on Bjuggren and Wiberg (2008), when the Tobin’s Q exceeds 1, the firm’s return on equity covers the value of its minimum rate required by investors. Contrarily, a Tobin’s Q value below 1 indicates that the firm’s future performance is not perceived favorably by investors. In this case, the market expects a decline in the firm value. Thus, Tobin’s Q is appropriate for assessing agency cost of equity under varying levels of managerial ownership. (Short and Keasey 1999). Concerning agency cost of equity ratio, we retain, firstly, the ratio of discretionary operating expenses scaled by total sales. Secondly, we retain the administrative expenses ratio which includes salaries, commissions received by the CEO to facilitate transactions, travel expenses, advertising and marketing costs, rents, and other utilities. Thirdly, we retain the risk of free cash flow, which presents the risk related to a positive free cash flow and a low market value.

Regarding the nature of managerial ownership variables, the American thresholds are not applicable. Thus, we rely on the distribution of managerial ownership as presented by Charreaux (1991) in the French context. The author categorized the ownership held by the manager into three thresholds, 0%, 20%, and 50%, and observed that beyond the threshold of 50%, the managerial decision-making is conducted to the detriment of shareholders’ interests. In our empirical part, we first define the three managerial ownership levels. Then, we retain for each level the cases where the managers’ control of voting rights is greater than their ownership percentage. Based on the statistics of the French firms selected for this study, we recognize that French firms are characterized by a significant dissociation between the level of managerial ownership and the percentage of voting rights. This evidence confirms the finding of Kremp and Bloch (1999), who argue that voting power is very high in France. Thus, the choice of constructing the managerial power variable helps to gain better comprehension of the impact of agency conflicts on firm value. Thus, managerial power is used to split our sample into three sub-samples.

With regard to control variables, we retain the most important variables that are believed to impact firms’ market value. We specifically focus on dividend payments, firm size, sales growth, and firm performance. As one of the three fundamental strategic decisions of the firm, the role of dividend payments (DIVs) in determining firm value shows conflicting outcomes. According to agency theory, signaling theory, and the bird-in-hand theory, dividend payments are supposed to be positively correlated with firm value. Based on a sample of 111 firms listed on Borsa Istanbul from 1995 to 2017, Abdullah et al. (2023) provide evidence that Tobin’s Q is positively associated with payouts. Similarly, Budagaga (2017) examine the same Turkish business environment context through a sample of 44 listed firms for the 2007–2015 period. The authors suggest that cash dividend payments constitute a corporate governance mechanism that used to mitigate agency cost of equity. More specifically, this disciplinary mechanism averts managers from deploying the free cash flow in unprofitable projects. Other researchers suggest that when firms prioritize high dividends over the interests of creditors, this can impact their perceived market value and affect the confidence of investors and creditors. This, in turn, can harm firm value. Likewise, Rozeff (1982) theorizes that, because of the financing costs of issuing debt, higher levels of dividend payments are associated with lower firm value.

Since our dependent variable incorporates growth opportunities, we integrate the sales growth ratio (SG) as a proxy for measuring firm growth. According to Nezami et al. (2018), sales growth is positively correlated with firm value, given that a higher firm valuation derives from greater sales. In such a case, it will be easier to attract new investors through revenue announcements, which are linked to market returns. Being considered as an indicator of accounting-based performance, the present paper includes the return-on-assets ratio (ROA). For instance, Bidaya et al. (2023) demonstrate that firm performance presents a guide for investors to invest their funds. Through an analysis of 17 Indonesian firms listed on the Indonesian Stock Exchange, the authors indicate that investors interpret the improved firm performance as a signal for future growth opportunities. Thus, the firm’s stock prices move up, which, in turn, increases its value. Lastly, we select the firm size (SIZE) as control variable. Several studies consider that size has a positive impact on firm value. According to Marsh (1982), large firms possess knowledge of tax-saving strategies and are more selective in their choice of CEO. Other points of view indicate that managerial entrenchment is associated with large firms (Jensen 1986). Likewise, Saona and San Martín (2018) report that larger firms seem to be more difficult to monitor. Because of diversification, these firms may operate in unprofitable industries, which can harm firm value.

3.3. Econometric Tools and Model Specification

The model estimation follows a static panel data approach. Prior to conducting the regression analysis, we employ the natural logarithm (ln) for variables that do not follow a normal distribution like Tobin’s Q. According to Khaoula and Moez (2019), this transformation can reduce asymmetry by reducing the influence of extreme observations. Using the Breusch–Pagan–Godfrey test, we confirm the presence of a heteroskedasticity problem. Additionally, we employ the Wooldridge test to detect autocorrelation problems. The outcome shows the presence of first-order autocorrelation in the error terms. To address these problems, it is recommended to estimate the regression model by using the Generalized Least Squares (GLS) method. The model is defined as:

$${Tobin\_Q}_{i,t}={\beta }_0+{\beta }_i{Agency Cost of Equity variables}_{i,t}+{\beta }_j{Control variables}_{i,t}+{\epsilon }_{it}$$

(1)

To analyze the potential impact of managerial power on the relationship investigated in the previous model, we build on the works of Charreaux (1991) in the French context. Thus, the present study divides the sample into three sub-samples of managerial equity ownership: from 0 to 20 percent, from 20 to 50 percent, and above 50 percent. Simultaneously, we consider the fraction of the manager’s voting rights in order to capture the level of managerial power.

4. Results

4.1. Descriptive Statistics and Correlation

According to the summary statistics in

Table 3. Descriptive statistics.

Variables	Observation Nbr	Average	Min	Max	SD
Tobin’s Q	1800	1.493	−0.271	6.987	0.825
OPE	1800	0.963	0.002	5.101	0.392
SGA	1800	0.334	0	1.709	0.275
DIV	1800	0.260	0	0.990	0.258
SG	1800	0.098	−0.999	0.984	0.206
ROA	1800	0.034	−1.430	0.550	0.109
SIZE	1800	8.873	4.453	11.27	1.062
Dichotomous Variables	Modality			Frequency	Percentage
RFCF	1: Tobin’s Q is lower than the sector average and FCF > 0. 0: otherwise.			182 1618	10.11 89.89
MPL1	1: Managerial ownership ranges between 0 and 20 percent and the ratio of manager’s voting rights divided by their ownership percentage is greater than 1. 0: otherwise.			723 1077	40.16 59.84
MPL2	1: Managerial ownership ranges between 20 and 50 percent and the ratio of manager’s voting rights divided by their ownership percentage is greater than 1. 0: otherwise.			135 1665	8.10 91.89
MPL3	1: Managerial ownership ranges between 50 and 100 percent and the ratio of manager’s voting rights divided by their ownership percentage is greater than 1. 0: otherwise.			201 1559	11.16 88.83

Source: Author’s own elaboration based on results from Stata 14 statistical analysis.

, the average Tobin’s Q is 1.493. This indicates that, during the study period, French firms’ market value exceeded their replacement cost. For agency cost of equity, the SGA mean value is 0.334 with SD = 0.275. The OPE has a greater mean value (0.963) and a higher standard deviation (0.392). In addition, the ACFCF is observed in only 10.11% of our sample. Regarding the managerial power level, it can be seen that the first level (MPL1) is observed in 40.16% of our sample. In the range of 0 to 20%, the result proves the significant presence of managerial ownership, especially when managers’ control of voting rights is greater than their ownership percentage. In the same context, it is observed that 8.10% of firms have managers who simultaneously own between 20 and 50% of capital and their voting right is greater than their proportion of shares. Finally, MPL3 is observed only in 11.16% of our sample.

Concerning control variables, the results imply that the mean of dividend payments is 26%. In terms of firms’ performance, the descriptive statistics show an average rate of ROA of about 3.4%, with a standard error of 0.109 ranging from −1.430 to 0.550. Also, it is observed that the mean of the sales growth ratio (Growth) is 9.8%. Finally, the measure used for firm size varies from a small firm (4.45) to a big firm (11.27).

As reported in

Table 4. Correlation matrix.

	OPE	SGA	RFCF	DIV	GROWTH	ROA	SIZE
OPE	1
SGA	0.1720	1
RFCF	−0.0006	0.0287	1
DIV	−0.2389	−0.1812	0.0102	1
SG	−0.0054	0.0549	−0.1407	−0.1553	1
ROA	−0.3799	−0.1667	−0.0294	0.2085	0.1805	1
SIZE	−0.2645	−0.2835	−0.0624	0.3067	−0.1485	0.1024	1

Source: Author’s own elaboration based on results from Stata 14 statistical analysis.

, the Spearman correlation analysis shows that the greatest correlation coefficient is 0.37, specifically between firm performance and operating expense ratio. Thus, we confirm the absence of multicollinearity problems. In line with Gujarati and Porter (2003), this observation points out that the explanatory variables are relatively independent of each other, thereby boosting the credibility of the results.

4.2. Regression Results

4.2.1. Findings on the Impact of Agency Costs of Equity on Firm Value

Table 5. Results of FGLS estimates.

	Equation (1): Whole Sample
Variable	Coef	P > |z|
OPE	−0.133	0.000 ***
SGA	−0.138	0.015 **
RFCF	−0.115	0.000 ***
DIV	−0.012	0.632
SG	0.070	0.004 ***
ROA	0.079	0.042 ***
SIZE	−0.212	0.042 ***
Wald chi2 (7)	758.42
Prob > chi2	0.0000
Obs Nbr	1800

Note: **, and *** denote the significance level at 5% and 1%, respectively. Source: Author’s own elaboration based on results from Stata 14 statistical analysis.

displays the regression result for the first hypothesis without integrating the variables of managerial power level. Using the Two-Stage Least Squares regression analysis for the whole sample, the results provide strong evidence that agency cost variables, measured in terms of operating expenses, administrative expenses, and the agency costs related to free cash flow, showed a significant negative impact on Tobin’s Q. This leads us to accept H1. This aligns with the findings of Sdiq and Abdullah (2022) in the case of the Iraq Stock exchange; Hoang et al. (2019) in the case of Vietnam; and Vijayakumaran (2021) in the case of the China Stock Market. This evidence supports the entrenchment hypothesis that accentuates the potential conflict of interest between managers and shareholders in the context of French firms and its impact on firm value.

Regarding control variables, the coefficient on dividends payment is negative but has no significant impact on firm value. In addition, sales growth (SG) shows a significant positive relationship with the firm’s market value. This finding is in line with Nezami et al. (2018), who indicate that firms with greater sales may secure higher market valuations. Thus, they will be able to attract new investors through revenue announcements. The results also reveal that the association between firm performance and the firm’s market value is statistically significantly positive, meaning that the firm’s income-generating constitutes a key determinant of firm value. The results are complementary to those of Bidaya et al. (2023), who indicate that investors interpret the improved firm performance as a signal for future growth opportunities. Clearly, the results document that the increases in the stock prices lead to increases in firm value. Lastly, the coefficient of firm size looks negative and significant. This finding reinforces the empirical support of Jensen (1986) and Saona and San Martín (2018), who report that larger firms seem to be more difficult to monitor. Because of diversification, these firms may operate in unprofitable industries, which can harm firm value.

4.2.2. Findings on Impact of Agency Cost of Equity on Firm Value under Managerial Power Level

Table 6. Results obtained for FGLS estimates.

	Equation (2): MPL1		Equation (3): MPL2		Equation (4): MPL3
Variable	Coef	P > |z|	Coef	P > |z|	Coef	P > |z|
OPE	0.049	0.000 ***	0.034	0.109	−0.376	0.008 ***
SGA	−0.019	0.247	−0.057	0.000 ***	0.052	0.070 *
RFCF	−0.129	0.000 ***	−0.063	0.001 ***	−0.072	0.000 ***
DIV	−0.011	0.279	−0.050	0.059 *	−0.063	0.058 *
SG	0.050	0.000 ***	0.169	0.000 ***	0.069	0.118
ROA	0.141	0.000 ***	0.212	0.170	0.030	0.853
SIZE	−0.005	0.278	−0.044	0.000 ***	−0.075	0.000 ***
Wald chi2 (7)	277.46		104.00		80.28
Prob > chi 2	0.0000		0.0000		0.0000
Observation Nbr	716		130		200

Note: *, and *** denote the significance level at 10% and 1%, respectively. Source: Author’s own elaboration based on results from Stata 14 statistical analysis.

reports the regression result for the second hypothesis by taking into consideration the three intervals of managerial power. Thus, we split the hole sample into three sub-samples. According to the first position of managerial power (MPL1), the result in Equation (2) shows that operating expenses exert a significant positive impact on firm value (coefficient = 0.049 and p < 0.01). This finding denotes that managerial ownership and voting rights at this level are associated with incentive alignment. More specifically, the manager tries to signal to the investor that the firm is well managed by maintaining a low operating expenses ratio through effective cost management and operational efficiency. By moving to the second level (MPL2), the impact of operating expenses on the firm’s market value does not appear to be significant and supports the neutrality hypothesis. This indicates that managerial power at level two does not affect this relationship. Nevertheless, the findings in Equation (4) indicate that managers become entrenched through operating expenses when they own more than 50% of equity shares and simultaneously hold significant voting rights (MPL3). Especially when the managerial power goes beyond that level, the agency problem seems to press down the firm’s market value. In such a case, the decrease in Tobin’s Q supports the entrenchment hypothesis. This result agrees with the prediction of Stulz (1988) regarding the negative association between the higher level of managerial ownership and the firm’s market value. Additionally, this result conforms to the findings of Scherrer and Fernandes (2021), who provide evidence that the firm’s market value is negatively impacted by the increase in managerial voting rights.

In terms of the selling, general, and administrative expenses ratio, we find that the entrenchment effect only appeared when the percentage of equity shares possessed by the manager ranges between 20 and 50 percent and the ratio of manager’s voting rights divided by their ownership percentage is greater than 1 (MPL2). At this level of power, managers become unmonitored. They can take advantage of free cash flow through excessive spending on administrative costs, strengthen their relationships with partners, and enlarge their power. In Equation (4), the results exhibit that SGA expenses (coef. = 0.052, t-stat = 0.070) positively influence firm value when the manager holds more than 50% of the ownership and simultaneously holds significant voting rights (MPL3). These results indicate that, beyond this threshold, the manager tends to maximize shareholder value. Thus, this level is associated with incentive effects that result in a convergence of interests. This evidence is in line with the predictions of Jensen and Meckling (1976), who suggest that a high level of managerial ownership ensures the alignment of the interests between managers and shareholders. Likewise, this finding is complementary to that of Stulz (1988), who supports that the portion of managerial voting rights is a significant component of the ownership structure that affects the firm’s market value. In reality, strong managerial voting rights can foster coherence with shareholders by reducing the excessive spending on administrative expenses.

Regarding agency costs linked to free cash flow, the coefficients in all the regressions demonstrate a significant negative association with firm value, which indicates that managers are trying to achieve bonuses using the free cash flow aside from their power level. In particular, managerial power in the three levels gives managers the opportunity to retain the control of free cash flow in order to entrench their own power and position. Particularly, they expand firm size to prioritize a higher remuneration, rather than distribute it to shareholders as dividend payments (Hsiao et al. 2023). This finding is in line with Jensen (1986), who supports the theory linked to the free cash flow forms and confirms the explanation of the managerial entrenchment hypothesis.

Among control variables, all the regressions in Table 6 indicate that dividend payments have a negative association with firm value. However, the coefficient is significant only at the levels of managerial power thresholds MPL2 (coefficient = 0.049 and p < 0.01) and MPL3 (coefficient = 0.049 and p < 0.01). The results signify that when the managerial power goes beyond the second level (MPL2), dividend payout increases. Therefore, the firm value appears to be pressed down because of the higher transaction costs of financing dividends. In particular, dividend payments at this level are unable to restrict managerial discretion.

The results also reveal that sales growth (SG) maintains the same positive impact on firm value. However, from MPL2, its impact is positive but not significant. This result complies with the coefficients of free cash flow and dividend payments. Any increase in sales will be affected to offset the free cash flow, as well as the reliance on costly borrowing to pay dividends. Thus, investors do not take into consideration the level of sales growth because, in the presence of managerial entrenchment at MPL3, variations in sales do not guarantee the expected level of return. Firm performance highlights a positive correlation with firm value only when the manager holds less than 20% of the ownership and simultaneously holds significant voting rights (MPL3). This result argues that managers at MPL3 act towards maximizing the shareholder wealth. Lastly, the coefficient of size maintains the same sign as that of the result of the whole sample in Table 5. Thus, we identify that the coefficient increases with the increase in managerial power. (MPL1: coefficient = −0.005/MPL2: coefficient = −0.044/MPL3: coefficient = −0.075). This finding indicates that by maintaining large-size firms through diversification, managers can benefit from increased prestige, along with greater control. In turn, this may amplify the negative association between the firm size and its market value.

Through an overall observation of the results, our findings confirmed that the impact of agency cost of equity on firm value depends simultaneously on the level of managerial ownership and its fraction of voting rights held by managers. The results comply with the theory of Morck et al. (1988), which supports the non-linear relationship.

Table 7. Summary of findings.

Agency Cost Variables	Equation (1): Whole Simple	Equation (2): MPL1	Equation (3): MPL2	Equation (4): MPL3
	Observable Effect	Observable Effect	Observable Effect	Observable Effect
OPE	Entrenchment	Incentive	Neutrality	Entrenchment
SGA	Entrenchment	Neutrality	Entrenchment	Incentive
RFCF	Entrenchment	Entrenchment	Entrenchment	Entrenchment

Sources: own construction.

summarizes the full results obtained in terms of the manager’s incentive, neutrality, and entrenchment effects.

5. Conclusions

Despite the existence of a permanent interest in the importance of corporate governance mechanisms in reducing the agency cost of equity and maximizing the firm’s market value, empirical surveys on the role of managerial behaviors in this relationship remain inconclusive. For empirical analysis, we apply data from French-listed firms for the period 2008–2022. We provide evidence on how the association between agency conflict type 1 and firm value varies with the level of managerial power. To achieve the goal of this study, we construct three measures of agency costs of equity utilizing the discretionary operating expenses ratio, the administrative expenses ratio, and the risk associated with free cash flow. We construct a binary variable of managerial power using the fraction of capital held by the manager, as well as their level of voting rights. To capture the effect of managerial power, we relied on managerial equity ownership and the fraction of manager’s voting right. This paper differs from previous academic research by creating a new measurement that may can specify the managerial power. In the first empirical analysis, the findings argue that agency cost of equity has a significant negative effect on Tobin’s Q. These conflicts of interest are likely to cause additional costs of surveillance and control, as well as an increase in the cost of capital. This in turn weakens firm value.

By dividing the whole sample into three levels using managerial power, we present strong evidence that the behavior of the manager (neutral, entrenched, or aligned) depends on their power within the firm’s ownership structure. Similarly, the present findings provide a notable difference as compared to those of Hermalin and Weisbach (1988), Stulz (1988), and Charreaux (1991). More precisely, the use of the operating expenses ratio enables us to argue that the incentive effect appears only when managerial power is at the first level (MPL1). The manager tends to maintain a low operating expenses ratio through effective cost management and operational efficiency. However, the neutrality effect becomes visible at the second level (MPL2), and finally, at a high level of managerial power, the entrenchment effect becomes dominant (MPL3). Regarding the selling, general, and administrative expenses ratio, the results demonstrate that the neutrality effect appears only when managerial power is at the first level (MPL1), the incentive effect becomes visible at the second level (MPL2), and finally, at a high level of managerial power, managers tend to maximize shareholder value (MPL3). Lastly, we document that agency costs related to free cash flow form an excellent means for managerial entrenchment regardless of its power level.

The present research paper may contribute to the academic literature by advancing the knowledge of the existing relationship between agency cost of equity, firm value, and managerial power. Specifically, we tried to further clarify that each level of managerial power is connected to a different strategy (incentive, neutral, and entrenchment), which in turn affects the firm’s market value. This input could help French corporate management establish effective corporate governance mechanisms designed to reduce equity agency conflicts. Nevertheless, the present study has some limitations. First, we did not take into consideration some sources of managerial power through the board of directors’ characteristics. In particular, the introduction of variables such as CEO duality, directors’ tenure in the same firm, and board independence may better explain the behavior of managerial power in relation to agency conflicts and firm value. Therefore, future research could take into consideration these variables to construct an index of managerial power. Second, other measures of agency costs related to managers and shareholders could be used to better explain the purpose of this study.