Corporate governance and firm characteristics as explanatory factors of shareholder activism: Validation through the French context

Abstract

This study deals with the major determinants of company shareholder activism investigated with according to a set of SBF 120 listed firms. Based on a sample of 77 companies, observed over the period 2008–2012, we are led to conclude that some firm governance characteristics do appear to affect shareholding activism. In addition, it has been revealed that the presence of institutional investors, ownership concentration, leaders’ presence in the capital, control structure, leadership change, firm growth as well as leverage level appear to have a significant influence on the probability of activism to take place.

Keywords:

• shareholder activism
• corporate governance
• firm characteristics

Public Interest Statement

The main purpose of this article was to improve the explanatory factors that influence shareholder activism; the author’s objectives include highlighting the role played by corporate governance, and the characteristic firm. Our main findings reveal that: the presence of institutional investors, ownership concentration, presence of managerial ownership, control structure, change of leaders, indebtedness rate, and company growth prove to constitute the major significant determinants relevant for shareholder activism to take place. The results achieved by this study seem to be potentially useful to shareholders, outsider investors, managers as well as entrepreneurial policy-makers for the purpose of improving and further consolidating the corporate governance practices.

1. Introduction

Shareholder activism constitutes a relatively new research area within the corporate governance domain. Nevertheless, the majority of shareholders’ exerted activism lies in a positive pressure put on leaders, in a bid to solve the control problems persisting between shareholders and executives (Girard, 2011) and maximize shareholder value. Noteworthy, however, the activism-related debates date back to the controversy that emerged following the release of the famous publication of the works elaborated by Berle and Means (1932). These works, the authors highlight that the listed firms’ shareholders have lost control to the leaders’ benefit. Indeed, activism takes an interesting position in the current research works dealing with corporate governance. These conducted works predominantly interrogate the set of principles and rules likely to help restrict the actions taken by leaders (Charreaux, 1997). The activism designates mainly the investors’ undertaken acts and measures as available means whereby can appeal to their political rights to defend their values interest and achieve their objectives. On doing so, they would strive to change some of the firm characterizing practices, by bringing together a set of procedures already initiated by some outside shareholders. As a matter of fact, their aim behind all thus lies in altering some of the company features, practices, and policies to make them fit for their targeted expectations: financial, social, or ethical (Le Maux, 2004). To note, shareholder activism has made subject of several empirical works, worth mentioning among which the study elaborated by Gordon and Pound (1993). The history of shareholder activism originated initially in the United States in the 1930s. By that time, the United States’ investors decided to be involved in the company decision-making policies, through access to information about shares as well as all undertaken decisions and strategies. Previously, however, shareholders had not been very interested in being intermingled in company affairs. They used to be primarily concerned with having their dividends received or investment repaid.

Under agency theory, shareholder activism has been used as an alternative control mechanism whereby agency conflicts between leaders and shareholders can be resolved (e.g. Gillan & Stark, 2007; Girard, 2011; Goranova, Abouk, & Soofi, 2015; Goranova & Ryan, 2014). In this respect, Jensen and Meckling (1976) have devised a theoretical framework suggesting that shareholders’ active monitoring would constitute an effective means through which the interests of both executives and shareholders can be aligned. Such a framework can be defined as a series of measures likely to be taken by shareholders for the sake of influencing the corporate governance process and providing advice in this regard (Becht, Franks, Mayer, & Rossi, 2009).

The present study’s major objective does not consist in investigating the shareholder activism relevant impact or effectiveness. It rather lies in examining factors likely to help enhance shareholders to become more active as crucial determinants of activism. Indeed, the study tries to substantiate a contribution, based on a study sample made up of some French SBF 120 listed companies. Simultaneously, an attempt to test the activism determinants’ explanatory elements has been considered.

Using conditional logistic regression analyses, a positive relationship has been discovered to persist between institutional investors and shareholder activism. Such a finding appears to be well consistent with the presence of institutional investors. Their presence seems significantly and positively associated with the occurrence of shareholder activism. Besides, a positive relationship has also been discovered to prevail between ownership concentration and shareholder activism. Another important finding of our study lies in the negative relationship between managerial ownership and shareholder activism. In addition, a negative relationship between firm family character and shareholder activism has also been noted. This finding is consistent with a more certain probability for activism to take place within the family and heritage firm types rather than in the managerial ones. However, it has been discovered that change of director and shareholder activism appear to be positively correlated. Another finding is the negative relationship prevailing between leverage and shareholder activism. Finally, a negative relationship has been discovered to persist between growth and shareholder activism. Consequently, the presence of institutional investors, ownership concentration, managerial ownership, control structure, change of leaders, leverage rate along with company growth prove to constitute the major significant determinants relevant for shareholder activism to take place. This result achieved by this study seems to be potentially useful to shareholders, outsider investors, managers as well as entrepreneurial policy-makers for the purpose of improving and further consolidating the corporate governance practices. The paper adds to existing literature on corporate governance by establishing a relationship between shareholders activism and corporate governance and characteristic firms.

Worth highlighting, his work involves adoption of a hypothetic-deductive method, repartitioned over three sections as follows. The first section deals with a literature review and hypotheses’ development. It involves an investigation of certain exogenous factors likely to affect shareholder activism, which are considered as useful procedures that could improve the quality of our results perfectly well. As for the second section, it is conceived to delimit the study sample and the data collection method; it defines the study of endogenous and exogenous variables’ set, along with their corresponding measures. It is also designed to specify the theoretical model to be tested, and present the analytical method adopted. As for the achieved results analysis and discussion, they make the subject of the third section. Finally, the work is concluded with a depiction of the major attained findings and paves the way for prospective work horizons.

2. Literature review and hypotheses development

Based on previously conducted studies, we consider identifying certain factors likely to help influence shareholder activism. In this context, some governance-related factors will be examined, namely, the presence of institutional investors, ownership concentration, managerial ownership, firm family nature, and respect of good-governance criteria, along with some company characteristics, mainly its performance, change of leaders, leverage level, as well as the size and growth, affecting shareholder activism.

2.1. Institutional ownership

Several empirical researches underline the remarkable extent of institutional investors’ participation in business affairs noticed recently. Such investors with a great desire to influence company policies, have turned into activists by being allowed to submit certain proposals to the general assembly in a bid to eliminate the anti-OPA measures (Gordon & Pound, 1993). In this regard, Aggrawal, Saffi, and Sturgess (2013) have revealed that corporate governance is important to institutional investors and that the proxy process is an important channel for corporate governance. In a survey of institutional investors conducted by McCahery, Sautner, and Starks (2011), the authors have discovered that corporate governance proves to be crucially important to the institutional investors, and that several institutions have shown certain willingness to engage in shareholder activism (lee. Noteworthy, however, when the institutional investors actually sell their shares, the share prices would, most after certainly decline dramatically, resulting in significant losses to the institutions. As such, the institutions appear to be locked in their investments. This dilemma might well lead to an increased activism from the part of institutional investors who would consider appealing to the voting right as a means, whereby management could be influenced or replaced rather than selling their shares (Nielsen, 2000). Thus, institutional investors have become influential and gather more right to vote and have a say in company affairs. In this same line of thought, Sahut and Othmani Gharbi (2011a, 2011b) indicate that such actors would appear to have a remarkably favorable influence on firm performance should they prove to have a particularly active corporate behavior. Hence, the more significant their share in capital is, the more active they would turn out to be. This way seems worth advancing:

H1: institutional ownership helps positively influence shareholder activism. 

2.2. Ownership concentration

Capital concentration stands as a maintenance of good governance, though some authors emphasize the persistence of a non-monotonous relationship between shareholding distribution and financial information quality (Bozec, 2008; Labelle & Schatt, 2005). As for Shleifer and Vishny (1986), they state that the majority shareholders should play an active role in company governance. In the same line, Mayer (1996) has discovered that ownership concentration helps well in ensuring a more efficient company governance and control. In addition, David, Bloom, and Hillman (2007) highlight that a shareholder activist share size proves to have an impact on firms’ decisions-making. With respect to Oded and Wang (2010), ownership concentration has been discovered to help further enhance frequent monitoring. They have put forward a model explaining that ownership concentration helps well enhance and motivate shareholder activism for the sake of promoting business value, and that a high concentration level is reflected by high expenditure in matters of supervision. Hence, the following hypothesis may well be proposed:

H2: ownership concentration positively influences shareholder activism.

2.3. Managerial ownership

Concerning this subject, several researchers have examined the empirical relationship prevailing between managerial ownership and firm performance (Cho, 1998; Demsetz & Villalonga, 2001; Morck, Shleifer, & Vishny, 1988). In this respect, the entrenchment theory, inspired by Shleifer and Vishny (1989), advances the idea that managers would strive to in store an establishment strategy which, according to them, consists in reducing the risk of their being substituted as company heads: “they are rooted.” The interest of such an approach lies in explicitly highlighting the managers’ active behavior hypothesis in seeking to neutralize the disciplinary systems, in a bid to enlarge their discretionary latitude through various means. In the model devised by Shleifer and Vishny, establishment goes through achieving idiosyncratic investments (specific to leaders), while it is maintained by means according to Stiglitz and Edlin (1992). As for Ravid and Sudit (1994), it can well be attained by means of: controlled amounts of financial resource, blockage in market mechanisms, exploitation of regulatory loopholes, and directors’ board neutralization with only two levers being put at their disposal to discipline leaders: “remuneration and revocation” (Charreaux, 1994). Empirical studies examining the relationship between managerial ownership and shareholder activism reveal controversial findings as to the negative or positive aspect of the relationship. Indeed, the major supporters of the relationship linearity are Jensen and Meckling (1976), who maintain the idea that whenever managerial ownership tends to increase, divergence of interests between shareholders and leaders would turn out to drop. In fact, managerial ownership helps encourage leaders to act in compliance with the other shareholders’ interests and engage in firm-value maximization projects. Thus, according to the same authors, managerial ownership of part of the company capital, or a group of employees, would certainly increase their earnings interests while reducing agency costs. In this context, Bekiris (2013) and Cziraki, Renneboog, and Szilagyi (2010) stress that when managerial ownership reaches a high level, agency problem would turn out to be largely mitigated due to complete alignment between managers and shareholders. Consequently, higher the managerial ownership proves to be, the lower shareholder activism would be. At this level, the following hypothesis can be postulated: 

H3: managerial ownership negatively influences shareholder activism.

2.4. Firm family character

Most studies conclude that family firms are marked with higher performance as compared to non-family firms (Allouche, Amann, & Garaudel, 2007). This feature applies to both stock market performance (Anderson & Reeb, 2003; Villalonga & Amit, 2006) as well as to economic and financial performance (Allouche & Amann, 1995; Lee, 2004; McConaughy, Walker, Henderson, & Mishra, 1998; Sraer & Thesmar, 2007). Based on Jensen and Meckling (1976) interest-convergence theory, family firms’ performance appears to have its explanation in the agency costs economies, stressing the crucial function of control and supervision. As for Huynh (2010), he concludes that the minority shareholders’ activism proves to be rather strong with respect to firms’ naturally ruse through managerial control system. The author also stresses the fact that the managerial firms’ special character is closely associated with an increased risk of shareholder-activism as opposed to the tightly controlled family and heritage firms. At this junction, the following hypothesis can be formulated: 

H4: the firm family character negatively influences shareholder activism.

2.5. Respect of good governance criteria

Good governance rests particularly on the rules related to voting policy organizational structure as well as on the shareholder dialog approach. One way through which the shareholder could ensure maintenance of good governance lies in taking part in the company essential orientations, especially in its general assemblies. It is a well-known fact that investors above all the most demanding ones, usually seek to place their funds in well-governed companies. In France, for instance, the legislator held the principle of “respect or be justified” for the good governance criteria to be adopted with the aim of safeguarding business flexibility and avoiding the highly prescriptive character, as emphasized by Charreaux (2008).

The association binding corporate governance and institutional investors has been subject of a great deal of literature. In this regard, Aggarwal, Erel, Ferreira, and Matos (2011) have discovered that institutional ownership helps greatly in maintaining effective corporate governance. In turn, Bushee, Carter, and Gerakos (2010) have demonstrated that appropriation by sensitive governance institutions in the US is connected with prospective improvements shareholder rights. For Chung and Zhang (2011), the company shares’ fraction held by institutions appears to increase remarkably in parallel with governance quality. Similarly, McCahery et al. (2011) have found that corporate governance is of crucially importance to institutional investors, and that a large number of institutions are willing to engage in shareholder activism. Thus, the following hypothesis can be suggested: 

H5: low-level compliance with good-governance criteria helps positively influence shareholder activism.

2.6. Company performance

A remarkable faction of empirical works has undertaken to examine the relationship between institutional activism and company performance. In effect, different studies highlight the positive influence of the institutional investors as part of company shareholding, on firm performance. In this regard, Tsai and Gu (2007) have shown a positive effect of such investors on Tobin Q. Nesbitt (1994) has reached a similar finding in terms of stock market long-run profitability, while Opler and Sokobin (1997) have demonstrated that firms with prevailing institutional investors turn out to be over performing on the market. In turn, Sahut and Othmani Gharbi (2011b) stipulate that this favorable influence on firm performance reflects well institutional investor’s particularly active behavior. On the other hand, several other authors have revealed a negative effect associated with the institutional investors’ presence on firm performance. Worth mentioning among them, one can mention: Khanna and Palepu (1999) for the case of India, Woidtke (2002) concerning American public pension funds, Sahut and Othmani Gharbi (2011a) with respect to both the French and reducer shareholdings detention cases, in addition to Faccio and Lasfer (2000) regarding the London Stock Exchange-listed firms. Noteworthy, also, some conducted works have reached in neutral result as regard to this particular issue, among which one may cite: Mínguez-Vera and Martín-Ugedo (2007) concerning a Tobin Q based Spanish context study; Mizuno (2010) on a ROE-based study dealing with the Japanese context; Daily, Johnson, Ellstrand, and Dalton (1996) concerning an ROA- and ROE-conducted study; as well as Gillan and Starks (1997) Huynh (2010) and Gantchev, Gredil, and Jotikasthira (2015), elaborated works leading to absence of relationship between firm performance and shareholder activism. So, the following hypothesis may well be postulated: 

H6: low-level financial performance positively influences shareholder activism.

2.7. Alternative hypotheses

2.7.1. CEO succession

A major idea advanced by Jensen and Meckling (1976) stresses that agency conflict may well emerge at the moment when shareholders decide to delegate part of the decision-making power, enthusiastic about reaching wealth maximization objective for the interest of the entirety of stakeholders, in some cases, maximization of their personal interests. In this regard, Lautsen (2002) has shown that, regarding the specific case of Denmark, performance levels appear to have an influence on the CEO possible eviction, which denotes well that the leader’s perpetual of being dismissed, entices him to act in compliance with the shareholders’ profitable interests. With respect to a German case study, Leker and Salomo (2000) have been led to conclude that decrease in performance (measured by ROA) has proven to have an impact on the leadership teams’ rotation rate. Similar findings appear also in other works maintaining and confirming the idea of interrelatedness between decline performance and the increased probability for a management team to be substituted (see, for instance, Harrison, Torres, & Kukalis, 1988). In this respect, Finet and Labelle (2004) postulate that once institutional investors appear to be at the origin of a management team modification procedure, they would themselves emit a positive signal to the markets by buying massive amounts of the company shares under review (a detention-level positive effect). This fact has made Davidson, Jiraporn, Nemec, and Nemec (2004) to maintain the idea that earnings management is closely linked to leaders’ succession, and that the newly designated leaders would face high pressures to show an improvement in their performance, in a bid to convince the company shareholders that their hiring decision is proceeding in the right direction. In turn, Hadani, Goranova, and Khan (2011) state that the institutional investors’ activism is highly linked to the leadership shift change on managing earnings. Accordingly, the following hypothesis can be drawn:

H7: change of leaders positively influences shareholder activism.

2.7.2. Leverage

Strickland, Wiles, and Zenner (1996), Smith (1996) along with Karpoff, Malatesta, and Walking (1996) consider that the leverage rate is synonymous with applying active defense measures against public bids. In fact, a high debt rate would frighten shareholders from an increased company bankruptcy risk. From agency theory perspective, the recourse to debt may stand as a solution fit to result conflicts likely to occur between shareholders and managers (Jensen & Meckling, 1976). Indeed, due to an increasing bankruptcy risk, the debt situation would entice the leader, feeling threatened by the likelihood of losing remuneration and other benefits, to adopt a more effective management strategy rather consistent with the shareholders’ interests. In this sense, debt would constitute a means useful for disciplining leaders.

In this regard, Berger, Ofek and Yermack (2000), have elaborated a study to investigate the association persisting between the established leader and firm capital structure. Their attained results have firstly revealed that leaders most often seek to avoid debt. Secondly, they have indicated that the leverage level tends to be reduced whenever leaders do not feel any pressure exerted upon them from the part of shareholders, and when no financial incentives or active control are being imposed upon than. More recently, on discussing the free cash-flows theory, Guiomar Novaes (2003) has emphasized that the shareholders usually have recourse to debt as a means through which managers can be disciplined and firm value maximized. Thus, the hypothesis below can be put forward:

H8: high-level leverage positively influences shareholder activism.

2.7.3. Company size

Empirical literature dealing with this particular area indicates that activism helps create more value whenever the target proves to be large. In this respect, Eesley and Lenox (2006), Hendry Rowley and Berman (2000), Rehbein, Waddock, and Graves (2004) Cai and Walkling (2011), Karpoff et al. (1996) along with Renneboog and Szilagyi (2011) have postulated that large firms are more likely to be targeted by activists shareholders. For Smith (1996), the California CALPERS pension fund targets have a size positively correlated with the likelihood of being targeted. Indeed, the funds tend to consider that activism valorization seems to be easier in large-size companies. Smith justifies this finding by the fact that institutional investors turn out to be more predominant in big corporations, and vote more heavily than other shareholders groups in case of disagreement to the management. Similarly, in a study dealing with 269 target companies, Karpoff et al. (1996) confirm that targets are discovered to be too large in size due to the fact that their control sample is smaller with respect to 65.3% of cases. Noteworthy, however, they underline that the targeted firm is often the largest of its industrial sector, hence the great difficulty of finding a company of the same size to stand as a control sample. In parallel, they confirm the theory of Smith (1996) stipulating that target firms have a greater deal of institutional investors participating in their capital. Previously conducted researches, mainly those conducted by Demsetz and Lehn (1985), Mikkelson and Partch’s (1989) and Song and Walkling (1993), have shown a positive correlation to prevail between firm size and the presence of institutional shareholders. Hence, the following hypothesis can be formulated: 

H9: firm size positively influences shareholder activism.

2.7.4. Firm growth or book-to-market ratio

Firm growth is usually regarded as a positive signal by shareholders. While the objective of every shareholder is to maximize wealth, businesses in high growth are characterized with a high level of uncertainty, making control function more complex. Consequently, high-growth company directors are usually supposed to be confronted with numerous and diverse problems, rendering the leaders’ performance, monitoring, and evaluation a rather difficult task, involving a complex work. In this regard, on using the book-to-market ratio as a proxy for growth opportunity, McNichols (2000) has reported a positive large accumulation associated with growth enterprises. In turn, Klein (2002) has reported a statistically significant relationship between book-to-market ratio and good corporate governance: the higher the ratio is the more effective governance the company will have. Similarly, Yan and Zhang (2009) consider that the company is regarded as a growing company once its book-to market ratio is high. Velury and Jenkins (2006) have demonstrated a persisting relationship between institutional investors and the quality. The authors examine the relationship between institutional ownership, company-growth, and abnormal accruals. In addition, Hadani et al. (2011) along with Goranova et al. (2015) examined the relationship between shareholders proposals, earning management and company growth measured by the book to market ratio. A low corporate growth helps encourage shareholders to put pressure on leaders in such a way as to manage accounting earnings and conceal them. So, the following hypothesis can be maintained:

H10: company growth negatively influences shareholder activism.

Once the theoretical framework is established, the different advanced assumptions will undergo an analytical testing stage. 

3. Data collection and research methodology 

3.1. Sample composition and data collection

The present study is focused exclusively on SBF 120 listed companies. After eliminating the financial specialized businesses (insurances and banks), the companies with no complete accounting and financial information, necessary to our conduct study, firms involving Dutch law, along with the shares financed companies, a final sample of 77 companies has been retained. These admitted businesses are going to be observed over a five-year period, ranging from 2008 to 2012.

As for the data collection procedure, different means have been applied. Most information has been obtained by consulting all companies’ annual reports and/or reference documents. An appeal has also been made to the databases: “infinancials” and “Thomson one Banker” concerning accounting data, and to the database “Datastream” relevant to stock data as well as to the concerned companies’ respective sites whenever certain information has not been available.

3.2. Research methodology 

For the sample-adapted character to be preserved, a conditional logistic regression (Hosmer, Lemeshow, & Sturdivant, 2000) has been applied given that our dependent variable (shareholder activism) is dichotomous, taking value 1 if activism does persist, and 0 otherwise. Hence, the following logistic regression model turns out to be worth estimating:$$\text{activ}={\mathit{\alpha }}_0+{\mathit{\alpha }}_1\text{inst}+{\mathit{\alpha }}_2\text{conc}+{\mathit{\alpha }}_3\text{mang}+{\mathit{\alpha }}_4\text{carafirm}+{\mathit{\alpha }}_5\text{indexgov}+{\mathit{\alpha }}_6\text{roa}+{\mathit{\alpha }}_7\text{changdir}+{\mathit{\alpha }}_8\text{lev}+{\mathit{\alpha }}_9\text{growth}+{\mathit{\alpha }}_{10}\text{size}$$

3.3. With

Activ: a dichotomous variable that takes value 1 if activism persists and 0 otherwise. Inst: the proportion of shares held by institutional investors. Conc: dichotomous variable that takes value 1 if there is a shareholder detaining more than 20% of capital and 0 otherwise. Carafirm: a dichotomous variable that takes value 1 if the managerial firm and 0 if the family firm or heritage one. Mang: a dichotomous variable that takes value 1 if there is a managerial ownership in the company and 0 if not. Roa: net income/total assets. Indexgov: assigns to every firm 1 point for each of the index 10 criteria respected by the firm. Changdir: a dichotomous variable that takes value 1 if there is a change of director and 0 if not. Lev: total debts divided by total assets. Size: natural logarithm of total assets. Growth: the ratio between the equity market value and book value. 

4. Empirical results

4.1. Descriptive and univariate analysis

Table 1 presents the descriptive statistics relevant to every single variable, correlation of the variable to explain with each of the explanatory variables separately, the Mann–Whitney U-test for the continuous variables, and the chi-square test for categorical variables.

Table 1. Total sample descriptive statistics (N = 385)

Part 1: Continuous variables
Variables	Group	Total sample		Total sample`		No target firms		Mean	Median	Std. dev.	Univariate correlation	z-stat
		Frequency	%	Frequency	%	Frequency	%
Inst	Presence of inst	308	0.8	119	0.844	189	0.7746	0.3321	0.2589	0.3077	0.0532	z = −1.201
	Absence of inst	77	0.2	22	0.156	55	0.2254				0.298	p = 0.2298
Indexgov	Indexgov1 (between 0 and 2) bad governance	12	0.0311	4	0.0284	8	0.0328
	Indexgov2 (between 3 and 5) acceptable governance	254	0.6598	89	0.6312	165	0.6762	4.8389	5	1.3443	0.0671	z = −1.010
	Indexgov3 (between 6 and 10) good governance	119	0.3091	48	0.3404	77	0.291				0.189	p = 0.3125
ROA	roa ≥ mean high performance	132	0.3429	55	0.39	71	0.291	0.0509	0.0393	0.1812	−0.019	z = −1.067
	roa ≤ mean weak performance	253	0.6571	86	0.61	173	0.709				0.7106	p = 0.2860
Lev	lev ≥ mean high levels of leverage	191	0.4961	62	0.3246	129	0.6754	0.1762	0.1705	0.1683	−0.0657	z = 1.757^*
	lev ≤ mean low levels of leverage	194	0.5039	79	0.4072	115	0.5928				0.1983	p = 0.0789
Size	size ≥ mean large firms	183	0.4753	70	0.3825	113	0.6175	6.8464	6.8006	0.6055	−0.0157	z = 0.912
	size ≤ mean small firms	202	0.5247	71	0.3515	131	0.6485				0.7589	p = 0.3617
Growth	growth ≥ mean high levels of growth	42	0.1091	11	0.2619	31	0.7381	3.1541	1.3928	9.8934	−0.1081^**	z = −1.335
	growth ≥ mean low levels of growth	343	0.891	130	0.379	213	0.621				0.0340	p = 0.1820

Part 2: Binary variables
Variables	Group	Total sample		Target firms		No target firms		Mean	Median	Univariate correlation	X^2-test
		Frequency	%	Frequency	%	Frequency	%
conc	Concentrated ownership	100	0.2597	49	0.3475	51	0.209	0.2597	0	0.1522^***	X^2 = 8.915^***
	Dispersed ownership	285	0.7403	92	0.6525	193	0.791			0.0028	p = 0.003
carafirm	Managerial firms	270	0.7013	95	0.6737	175	0.7172	0.7012	1	−0.0457	X^2 = 6.073^**
	Family firms	115	0.2987	46	0.3263	69	0.2828			0.3708	p = 0.014
Manag	Presence of the manager in the capital	326	0.8467	111	0.7872	215	0.8811	0.8467	1	−0.1256^**	X^2 = 0.805
	Absence of the manager in the capital	59	0.1533	30	0.2128	29	0.1189			0.0137	p = 0.369
changdirec	Change of director	30	0.078	16	0.5333	14	0.4667	0.0779	0	0.1008^**	X^2 = 3.914^**
	No change of director	355	0.922	125	0.3521	230	0.6479			0.048	p = 0.048

Notes: Activ: a dichotomous variable that takes value 1 if activism persists, and 0 otherwise. Inst: the proportion of shares held by institutional investors. Conc: dichotomous variable that takes the value 1 if there is a shareholder detaining more than 20% of capital and 0 otherwise. Carafirm: a dichotomous variable that takes value 1 if the managerial firm and 0 if the family firm or heritage one. Mang: a dichotomous variable that takes value 1 if there is a managerial ownership in the company and 0 if not. Roa: net income/total assets. Indexgov: assigns to every firm 1 point for each of the index 10 criteria respected by the firm. Changdir: a dichotomous variable that takes value 1 if there is a change of director and 0 if not. Lev: total debts divided by total assets. Size: natural logarithm of total assets. Growth: the ratio between the equity market value and book value.

^* Significantly different from zero at the 0.1 level.

^** Significantly different from zero at the 0.05 level.

^*** Significantly different from zero at the 0.01 level.

The descriptive statistics displayed in the above table indicate that institutional investors are massively present in the French firms. The table also reveals that, on average, institutional investors hold 33% of capital and, therefore, have a say in the company affairs and can exert pressure on the leading team. Besides, institutional investors appear to be active ones with respect to more than 85% of the companies. As can be deduced, institutional investors are usually active and considered as rather control shareholders. One can also note that managers who detain shares prove to be inactive (215 firms compared to 326). In fact, whenever, ownership concentration exceeds 20%, the percentage of the target firms’ active dominant shareholders appears to be higher than the percentage of the non-targeted firms’ dominant shareholders. Noteworthy, however, ownership concentration appears to play a vital role in the provenance of the activism. Concerning the “firm character” variable it has been discovered that most companies in our sample are of a managerial type (70.13%), given that the companies of the SBF 120 index stand as the largest stock market capitalizations in France. Still, our sample is characterized with a low number of firms featured with bad governance (3%). Thus, in most companies (85%) the leaders prove to be shareholding managers. 

Regarding the preliminary results’ univariate correlation, one may well state, in advance, the persistence of a positive relationship between shareholder activism, and its relevant explanatory factors. The correlation between the variables “Presence of institutional investors” and “shareholder activism,” as reflected by the coefficient is positive (0.05) and statistically non-significant (0.29). The Mann–Whitney test confirms well such a relationship. Hence, one may well confirm that no significant differences appear to be noticeable with respect to activism occurring to members of both company types whether with or without presence of institutional investors. Yet, correlation between the concentration ownership at the 20% threshold and activism proves to be significant. At the threshold of 5% one can confirm that (X² = 8.915 and p = 0.003), denoting that when there are dominant shareholders other than the institutional investors, activism does occur. Nevertheless, the correlation coefficient between the firm managerial nature and activism is low, of a negative sign (−0.04) and statistically non-significant (0.37). Thus, Pearson correlation coefficient between governance index and activism is low, bears a positive sign (0.06), and statistically non-significant (0.18). Besides, the Mann–Whitney test does confirm such a relationship. The variable “managerial ownership” presents a significantly negative and strong correlation with the variable “shareholder activism.” The chi-square test appears to confirm well such a relationship. Moreover, the correlation coefficient (0.79) suggests a negative association between the variables “ROA” and “shareholder activism.” Still, this correlation is not significant, and the Mann–Whitney test does confirm this relationship. The correlation coefficient highlights a positive association, though very low (0.10), between the variables “change of director” and “shareholder activism.” Yet, this correlation is statistically significant (0.04) at the threshold of 5%. The correlation between the variables “leverage” and “shareholder activism,” as reflected by the coefficient, seems to be very minute, negative (−0.06) and statistically non-significant (0.19). The correlation between the variables “firm size” and “shareholder activism,” as reflected by the correlation coefficient, is very low, negative (around −0.01) and statistically non-significant (0.75). The correlation between the variables “firm growth level” and “shareholder activism,” as reflected by the correlation coefficient, is very low (−0.10) and statistically significant at the threshold of 5% (0.03). The Mann–Whitney test does well confirm such a relationship.

4.2. The multivariate analysis 

4.2.1. Matrix of Pearson and multicollinearity

In Table 2, we provide the Pearson’s rho correlations between the variables used in the logistic regression. The most important aspect in the logistic regression analysis lies in the multicollinearity problem predominating among the independent variables. According to this table, all correlation coefficients prove to be lower than 0.8, the threshold at which a serious multicollinearity problem starts to occurs. Furthermore, the “Variance Inflation Factors” VIF have been computed as a means to test the presence of collinearity among the explanatory variables. In all cases, the VIFS are discovered to be lower than 2, bearing in mind that the critical value is set a 10 (Tabachnick, Fidell, & Osterlind, 1996). Thus, one might well conclude that the multicollinearity problem is completely non-existent.

Table 2. Pearson correlations matrix and VIFs

N = 385	VIFs	inst	conc	mang	carfirm	indexgov	roa	changdir	lev	Size	Growth
inst	1.26	1	−0.0848^*	0.2115^***	0.1158^***	0.3373^**	0.016	−0.0023	0.1054^**	0.1257^**	−0.0482
			0.0968	0	0.0231	0	0.7421	0.9635	0.0387	0.0136	0.3453
conc	1.26		1	0.1262^**	0.1277^**	0.1990^**	0.082	0.0930^*	0.0186	−0.0984^*	0.2444^***
				0.0132	0.0121	^* 0.0001	0.1082	0.0684	0.7164	0.0536	0
mang	1.16			1	0.0532	0.0456	0.0555	0.0161	−0.0601	−0.2235^***	−0.0782
					0.2978	0.3718	0.2771	0.753	0.2396	0	0.1258
carfirm	1.16				1	0.0866^* 0.0898	−0.0060 0.9065	0.1262^** 0.0132	−0.1200^**	0.1450^***	−0.1346^***
									0.0185	0.0044	0.0082
indexgov	1.22					1	−0.0325	0.0349	−0.0033	0.1019^**	−0.0052
							0.5248	0.4952	0.9492	0.0456	0.9186
								0.1971
changdir	1.03							1	0.0092	0.0472	−0.0404
									0.8578	0.3556	0.4296
lev	1.13								1	0.2042^***	−0.1148^***
										0.0001	0.0243
Size	1.33									1	0.0256
											0.6168
Growth	1.15										1
Mean VIFs	1.18

Notes: Activ: a dichotomous variable that takes value 1 if activism persists, and 0 otherwise. Inst: the proportion of shares held by institutional investors. Conc: dichotomous variable that takes the value 1 if there is a shareholder detaining more than 20% of capital and 0 otherwise. Carafirm: a dichotomous variable that takes value 1 if the managerial firm and 0 if the family firm or heritage one. Mang: a dichotomous variable that takes value 1 if there is a managerial ownership in the company and 0 if not. Roa: net income/ total assets. Indexgov: assigns to every firm 1 point for each of the index 10 criteria respected by the firm. Changdir: a dichotomous variable that takes value 1 if there is a change of director and 0 if not. Lev: total debts divided by total assets. Size: natural logarithm of total assets. Growth: the ratio between the equity market value and book value.

^* Significantly different from zero at the 0.1 level.

^** Significantly different from zero at the 0.05 level.

^*** Significantly different from zero at the 0.01 level.

4.2.2. Regression analysis

Table 3, summarizes the logistic regressions’ set of shareholder activism explanatory occurrence factors of the French SBF 120 listed firms. The table involves seven models.

Table 3. regression’ recapitulative

Variables	Predicted sign	Model 1		Model 2		Model 3		Model 4		Model 5		Model 6		Model 7
		Coefficient	Wald X^2	Coefficient	Wald X^2	Coefficient	Wald X^2	Coefficient	Wald X^2	Coefficient	Wald X^2	Coefficient	Wald X^2	Coefficient	Wald X^2
Principal variables
inst	+	0.879^**	0.032	0.372	0.290	–	–	–	–	–	–	–	–	–	–
Conc	+	1.01^*	0.000	–	–	0.890^***	0.000	–	–	–	–	–	–	–	–
mang	+/−	−0.887^*	0.007	–	–	–	–	−0.890^***	0.003	–	–	–	–	–	–
carfirm	+/−	−0.629^**	0.016	–	–	–	–	–	–	−0.419^*	0.083	–	–	–	–
indexgov	+/−	−0.007	0.984	–	–	–	–	–	–	–	–	0.095	0.238	–	–
roa	+/−	−0.299	0.651	–	–	–	–	–	–	–	–	–	–	−0.169	0.803
Control variables
lev	–	−1.864^*	0.009	−1.151^*	0.086	−1.348^*	0.050	−1.221^*	0.072	−1.30^*	0.055	−1.087	0.105	−1.104^*	0.098
changdir	+	0.745^*	0.068	0.725^*	0.061	0.589	0.137	0.761^*	0.051	0.804	0.040	0.709^*	0.068	0.713^*	0.066
Size	+/−	0.012	0.955	−0.177	0.924	0.108	0.566	−0.093	0.622	0.06	0.32	−0.013	0.943	−0.006	0.975
Growth	–	−0.069^**	0.019	−0.048^*	0.092	−0.061^**	0.031	−0.056^*	0.052	−0.057^*	0.076	−0.048^*	0.085	−0.050^*	0.086
constant		0.469	0.756	−0.297	0.810	−1.193	0.350	1.115	0.406	−0.374	−0.30	−0.676	0.594	−0.247	0.850
Number of observations		385		385		385		385		385		385		385
R^2 de Nagelkerke (%)		14		5		9		7		6		5		5
LR X^2		41.510^***		13.65^**		25.10^***		21.18^***		15.52^***		13.94^**		12.60^**
Prob > X^2		0.000		0.0180		0.0001		0.0007		0.0084		0.016		0.0274

Notes: Activ: a dichotomous variable that takes value 1 if activism persists, and 0 otherwise. Inst: the proportion of shares held by institutional investors. Conc: dichotomous variable that takes the value 1 if there is a shareholder detaining more than 20% of capital and 0 otherwise. Carafirm: a dichotomous variable that takes value 1 if the managerial firm and 0 if the family firm or heritage one. Mang: a dichotomous variable that takes value 1 if there is a managerial ownership in the company and 0 if not. Roa: net income/total assets. Indexgov: assigns to every firm 1 point for each of the index 10 criteria respected by the firm. Changdir: a dichotomous variable that takes value 1 if there is a change of director and 0 if not. Lev: total debts divided by total assets. Size: natural logarithm of total assets. Growth: the ratio between the equity market value and book value.

^* Significantly different from zero at the 0.1 level.

^** Significantly different from zero at the 0.05 level.

^*** Significantly different from zero at the 0.01 level.

According to Table 3, model 1 reveals that the institutional investors’ percentage coefficient is positive at a = 0,879, in accordance with the expected sign, and significant. Therefore, hypothesis 1 is confirmed. This finding corroborates well the results reached by Gordon and Pound (1993), Opler and Sokobin (1995), Del Guercio and Hawknis (1999), Karpoff et al. (1996), Smith (1996), Carleton, Nelson, and Weisbach (1998), Baudru and Lavigne (2001), Hadani et al. (2011), Hadani (2012), as well as Aggrawal et al. (2013) conforming that the presence of institutional investors is significantly and positively associated to the occurrence of shareholder activism. Actually, this relationship has its explanation in two major reasons. On the one hand, institutional investors are sometimes obliged to maintain the titles they hold, as they are unable to get rid of them without bearing output penalties (Hadani, 2012). In this regard, they turn out to be rather interested in closely managing the affairs of firms in which they participate and find themselves standing as captive “prisoners.” On the other hand, the institutional investors’ massive participation in the capital helps promote their monitoring capacities (Hadani et al., 2011). Indeed, they may accumulate enough means and competence skills whereby to influence the way in which the business is managed. Similarly, they tend to acquire the necessary resources enabling them to engage effectively in the control activities. In fact, they are able to exercise their control rights while ensuring that the benefits of monitoring do actually exceed the costs incurred. This conjunction between the ability and the duty to control represents an explanatory factor of major institutional activism. 

Thus, in model 1 the ownership concentration coefficient is positive at a = 1.01, simultaneously significant and conforming with the expected sign. This relationship is confirmed by model 3 through a univariate correlation. Hence, these results allow validating our hypothesis 2, corroborating the results attained by Shleifer and Vishny (1986, 1997), Mayer (1996), La Porta, Lopez-de-Sinales, Shleifer, and Vishny (1999), Alexander, Cici, and Gibson (2006), Gillan and Stark (2007), Hadani (2012), Labelle and Schatt (2005), Bozec (2008), Oded and Wang (2010). In this regard, the shareholders’ detention of majority shares in the company capital could well stand as an incentive to their shareholder activism. However, the majority shareholders who are simultaneously non-institutional and non-executive have enough means put at their disposal to intervene in the company affairs’ management and engage in the monitoring activities, e.g. through shareholder activism and implementation of voting rights.

With regard to the managerial ownership coefficient, it appears to be negative at a = −0.887 and to confirm with to the expected sign being significant with respect to both models 1 and 4. So, hypothesis 3 is confirmed. This result is also confirmed by Bekiris (2011), along with Jensen and Meckling (1976), who consider that managerial ownership helps encourage leaders to act in compliance with the other shareholders’ interests and to engage in projects likely to help in further maximizing the firm value. This finding is consistent with expectations that managerial ownership, acting as an incentive alignment mechanism, constrains managerial opportunism and thus generates less shareholder discontent. However, firms with higher executive pay are more likely to be targeted by activists (Ertimur, Ferri, and Muslu (2010), Ferri and Sandino (2009), hence leading to more shareholder resolutions. Self-serving managers can extract higher pay, but they are also more likely to attract shareholder discontent (Hillman, Shropshire, Certo, Dalton, and Dalton (2011).

In fact, the “firm family character” relevant coefficient is negative at a = −0.629 significantly confirming the expected sign. Besides, this relationship is confirmed though model 5. In this way, hypothesis 4 is confirmed, though not corroborating the finding revealed by Huynh (2010). As a matter of fact, the relationship between the managerial firm type and shareholder activism seems negative, i.e. the firms’ character (either family and/or heritage) that is discovered to have an effect on the probability of prevailing activism. Hence, there is a more certain probability for activism to take place in the family and heritage firm types rather than in the managerial types. Such a relationship may have its explanation in the fact that the concerned shareholders are themselves the businesses leaders, dominating in the directors’ board, and are more informed about the business matters, in contrast with the managerial type of firms, in which property is dispersed and t investors might well be passive and rather speculative shareholders (the myopic institutions theory¹and the portfolio theory).²

In effect, the governance index coefficient proves to be negative at a = −0.07, in conformity with the expected sign, but not significant. The model 6 sustains well such a relationship. As a result, hypothesis 5 appears to be rejected, corroborating finding of Huynh (2010), Yan and Zhang (2009) as well as Ruiz-Mallorquí and Santana-Martín (2011). Such a result is discovered to be in contrast with those attained by Thomas and Cotter (2007), Gillan and Stark (2007), Aggarwal et al. (2011), Chung and Zhang (2011), Bushee, Carter, and Gerakos (2009) along with McCahery et al. (2011). Indeed, the good governance criteria seem to have no effect on the activism probability. So, it is worth noting that such a relationship may well have its explanation in the majority of firms’ compliance with the good governance practices, over the last few years, has not appeared to yield the desired results.

Furthermore, the financial performance coefficient is discovered to be negative at a = −0,299, complying with the expected sign, but does not seem to be significant with respect to model 1, even on re-estimating it via a different model (model 7). Therefore, hypothesis 6 is considered to be rejected. This finding is in line with that released by Mínguez-Vera and Martín-Ugedo (2007), Mizuno (2010), Daily et al. (1996), Gillan and Starks (1997) along with Huynh (2010), who have discovered that ROA sounds to have no effect on the occurrence minority shareholders’ activism. Similarly, Karpoff et al. (1996), Gillan and Starks (1998), Kochhar and David (1996), Karpoff (2001) as well as Gantchev et al. (2015) have found that only little, or even no link appears to prevail between shareholder activism and company performance. Karpoff (2001) has attributed this significant disparity in empirical results to methodological differences on the part of the authors, i.e. in regard of definitions consecrated to the activism success-associated events and the characteristics. For instance, the private negotiations’ phase between the dissident shareholders and targeted management is likely to bias the results obtained (Carleton et al., 1998). Girard (2001) points out that rational expectations’ capacity and adaptation of control coalition allow, particularly, to mobilize and experiment with new governance mechanisms, thus reducing the stock market price impact on the release of shareholder activism practices.

The “change director” variable-related coefficient is positive at a = 0.745 in harmony with the expected sign, and very significant, corroborating the works elaborated by Parrino, Sias, and Starks (2003), Finet and Labelle (2004), Davidson et al. (2004) and Hadani et al. (2011). Therefore, fail to reject the hypothesis 7. This result has its explanation in the shareholders’ commitment to put pressure on the newly appointed leader to align their interests and manage earnings in a certain way. Hence, the activists’ shareholders can influence corporate decisions, specifically the change director, by taking an active role in the decision-making process rather than selling their shares. (Parrino et al. (2003)). Also, a new CEO will bring new competences for the firm which are beneficial to company and shareholders.

Concerning the “leverage” variable, the coefficient is negative at a = −1.864 to the expected sign, and is also significant. Indeed, hypothesis 8 is confirmed. This corroborates the results achieved by Strickland et al. (1996), Smith (1996), Karpoff et al. (1996) along with Edmans, Fang, and Zur (2013). This finding has its justification in the agency theory, highlighting that by increasing bankruptcy risk, debt would prompt the leader, feeling threatened to lose his remunerations and other similar benefits, to adopt a rather effective management, conforming more to the shareholders’ interests. Hence, debt would constitute a means whereby leaders could be disciplined, which makes investors behave in a passive way. Furthermore, low indebtedness may indicate available slack resources, potentially allowing greater managerial discretion in addressing shareholder activism (Goranova et al. (2015)). In addition, higher levels of debt can enforce fiscal discipline and constrain managers’ ability to engage in self-serving strategies (Hart, 1993), so less leveraged firms are more likely to be targets of hedge fund activism (Klein and Zur (2009)), although the reverse holds for governance-related activism (Ferri and Sandino (2009), Karpoff et al. (1996)). Again, hedge fund activism deviates from governance-oriented activism, as it tends to target more profitable and financially healthy firms (Brav, Jiang, Partnoy, & Thomas, 2008; Klein and Zur (2009).

The “firm size” coefficient is positive at a = 0.0120 in accordance with the expected sign, though not significant. Therefore, hypothesis 9 proves to be rejected. This result does not match with the results achieved by Smith (1996), Eesley and Lenox (2006), Hendry and Rowley and Berman (2000) as well as Rehbein et al. (2004). Usually, companies having a big size are characterized by more transparency, less information asymmetry, an access easier to the financial market, and consequently they are less put into debt. Indeed, the firms, subjects of our sample, are of the same size, a fact which appears to weaken the variable effect making it ineffective to yield the expected results. An alternative explanation, hedge funds are less likely to target larger firms (Brav et al., 2008; Klein and Zur (2009)), perhaps reflecting the difficulty of building a substantial block of shares in major companies (Edmans et al. (2013).

As for the “firm growth or Book to market” coefficient it is negative at a = −0,208, in harmony with the expected sign and is also significant as well. In this way, hypothesis 10 is confirmed. This finding does actually meet the results attained by Velury and Jenkins (2006), Yan and Zhang (2009), Hadani et al. (2011), as well as Goranova et al. (2015). It has its explanation in low corporate growth which enhances shareholders to put pressure on leaders to specifically manage the accounting earnings and conceal them so as to display a positive signal and image on the market.

5. Conclusion

The present study has been focused on a thorough examination of shareholder activism-associated determinants. To this end, a model-constructs compilation based on six research assumptions has been devised. This model construct deals mainly with the institutional investors’ participation in the capital, ownership concentration, presence of managerial ownership, control structure, respect of the good governance criteria and performance of four alternative hypotheses pertaining to leadership substitution, leverage, firm size, and firm growth.

Relying on a sample of 77 French SBF 120 listed firms, a logistic regression estimate has been undertaken. A positive relationship has been discovered to persist between institutional investors and shareholder activism. Such a finding appears to be well consistent with the presence of institutional investors. Their presence seems significantly and positively associated with the occurrence of shareholder activism. Besides, a positive relationship has also been discovered to prevail between ownership concentration and shareholder activism. Another important finding of our study lies in the negative relationship between managerial ownership and shareholder activism. In addition, a negative relationship between firm family character and shareholder activism has also been noted. This finding is consistent with a more certain probability for activism to take place within the family and heritage firm types rather than in the managerial ones. However, it has been discovered that change of director and shareholder activism appears to be positively correlated. Another finding is the negative relationship prevailing between leverage and shareholder activism. Finally, a negative relationship has been discovered to persist between growth and shareholder activism. Consequently, the presence of institutional investors, ownership concentration, managerial ownership, control structure, change of leaders, leverage rate along with company growth prove to constitute the major significant determinants relevant for shareholder activism to take place. 

Despite the crucial importance of the achieved results, our study bears certain limitations. Firstly, the size of our study sample is quite small (77 firms). Secondly, only the corporate governance and firm characteristic-related variables have exclusively been investigated. Thirdly, the media role has not been recognized and remained unknown owing to information unavailability. Despite these limitations, the results achieved by this study seem to be potentially useful to shareholders, outsider investors, managers as well as entrepreneurial policy-makers for the purpose of improving and further consolidating the corporate governance practices. The paper adds to existing literature on corporate governance by establishing a relationship between shareholders activism and corporate governance and characteristic firms.

Additional information

Funding

Funding. The authors received no direct funding for this research.

Notes on contributors

Siala Bouaziz Souha

Souha Bouaziz Siala, PhD, is an associate researcher at LARTIGE (Laboratory of Research) in the University of Sfax, Tunisia. She is the corresponding author and her research interests are corporate governance, earnings management, and accounting. Anis Jarboui is Professor of Finance and Accounting at the University of Sfax, Tunisia. His research interests are in corporate governance, financial literacy, and behavioral finance.

Notes

1.  The myopic institutions theory: it provides that the shareholders are "speculative shareholders." (Jarboui & Olivero, 2008).

2.  The portfolio theory: it considers that investors are required to pursue some diversification strategies in order to minimize the portfolio risk. Ben M’Barek (2003).