Corporate Governance: A Major Factor in Shareholder Activism in Brazil

ABSTRACT

Objective:

This paper verified the relationship between shareholder activism and the elements of corporate governance and financial performance of selected Brazilian companies, articulating the agency theory and the voice concept from the exit, voice and loyalty theory.

Methods:

Activism was determined by a score constituted by the sum of its occurrences, observed in the minutes of the 2016 and 2017 general meetings of the 100 public companies in the sample. Governance and performance factors corresponded to 2015 and 2016, collected from Economática®. Quantile regressions (QR) were applied and the results were compared with ordinary least squares regressions (OLS).

Results:

there was a better adjustment of QR compared to the OLS method, suggesting that governance negatively affects activism and that its effects are heterogeneous in different parts of its distribution. Inconclusive results were obtained for performance variables. Robustness tests indicated better explanatory ability for activism observations in companies positioned in the last 20% of the activism indicator’s distribution.

Conclusion:

significant evidence of a relationship between activism and governance was found, which does not occur conclusively in relation to performance, revealing the predominance of the first variable.

Keywords:
shareholder activism; corporate governance; quantile regression; agency theory

RESUMO

Objetivo:

este trabalho verificou a relação entre o ativismo de acionistas e os elementos de governança corporativa e desempenho financeiro das empresas brasileiras selecionadas, articulando a teoria da agência e o conceito voice da teoria exit, voice, and loyalty.

Métodos:

o ativismo foi apurado por um índice constituído pelo somatório de suas ocorrências, observadas nas atas das assembleias de 2016 e 2017 das cem empresas de capital aberto da amostra. Fatores de governança e desempenho corresponderam aos anos de 2015 e 2016, coletados via Economática®. Aplicou-se a regressão quantílica (RQ) e os resultados foram comparados com a regressão por mínimos quadrados ordinários (MQO).

Resultados:

houve melhor adequação da RQ frente ao método de MQO, sugerindo que a governança afeta negativamente o ativismo e que seus efeitos são heterogêneos nas diferentes partes de sua distribuição. Resultados inconclusivos foram obtidos para variáveis de desempenho. Testes de robustez indicaram melhor capacidade explicativa para observações de ativismo em empresas posicionadas nos últimos 20% mais altos do índice.

Conclusão:

foram encontradas evidências significativas de relação entre ativismo e governança, o que não ocorre de forma conclusiva com relação a desempenho, revelando o predomínio da primeira variável.

Palavras-chave:
ativismo de acionistas; governança corporativa; regressão quantílica; teoria da agência

INTRODUCTION

The participation of shareholders in explicit movements that aim to influence corporate policies and practices in the companies in which they invest is an international phenomenon experienced by the capital markets, having been disseminated by the financial capital globalization process, which has broadened the investment frontiers of institutional investors (Gillan & Starks, 2003; Goranova & Ryan, 2014).

The amount of approximately US$ 65 billion employed in campaigns by global activist investors during 2018 attests to the potential of the market effects of this movement (Harvard Law School Forum on Corporate Governance and Financial Regulation [HLS], 2019). This amount was equivalent to the stock market of Portugal (US$ 62 billion) in that same year (World Bank Group, 2019). Among the outcomes of this effort to intervene in the management of companies, we have: increased representation on the board of directors; success in appointing CEO; and interruption of M&A processes. As a result of a total of 247 campaigns, these outcomes were not reflected exclusively in US companies: 43% of them (106) were directed to the global market, with four campaigns targeting three Brazilian companies (Stenco, Petrobras, and Oi) (HLS, 2019).

The presence of domestic companies among targets of international activist campaigns in specialized reports (Activist Insight, 2019; HLS, 2019) and the existence of guidelines that orientate voting in meetings in Brazil (Institutional Shareholder Service [ISS], 2018) issued by shareholder advisories may indicate the relevance of the phenomenon in Brazilian organizations. Added to these indications are the occurrences of conflict situations mediated by actions associated with the activism of shareholders reported in the specialized media, such as notes from the Valor Econômico newspaper on events that occurred in the last two years at CCR (Azevedo, 2019) and Qualicorp (Valenti & Schincariol, 2018), the deliberations and analyses of the Brazilian Securities Commission (CVM), and guidelines and communications from entities representing members of the capital market as well as those that promote good corporate governance (CG) practices.

The activism of shareholders in the Brazilian business environment constitutes a set of evidences sufficient to give rise to academic research that offers a better understanding of the phenomenon and its impacts. Among the possibilities allowed by the complex theme and aiming to contribute to a convergence in its understanding, this work verified the relationships between activism and: (a) the quality of GC and (b) the performance of the selected companies in the Brazilian capital market. The activism of shareholders observed in the general meetings (ordinary and extraordinary) in the years 2016 and 2017 were considered. It is noteworthy that the study did not aim to establish the unequivocal causal relationship between the variables described above, but rather to explore the relationships considering their differentiated effect on the sample companies, based on an innovative method (quantile regression) for the proposed theme, expanding the possibilities of analysis.

The theoretical foundation of shareholder activism has been agency theory (Jensen & Meckling, 1976) because it posits the classic shareholder-manager conflict and expands it, encompassing majority-minority shareholder relations. Activism articulates with the CG system as it uses its resources and structures to minimize conflicts and guarantee returns on its investments (La Porta, Lopez-de-Silanes, Shleifer, & Vishny, 2000; Shleifer & Vishny, 1997), and manifests itself through three possibilities of action (Hirschman, 1970): sale of shares (exit); active movement to guarantee rights (voice); and passive maintenance of shares (loyalty). The view on activism is broadened considering the motivation for social issues-stakeholder activism-, which coexists with the literature on the topic (Aguilera, Desender, Bednar, & Lee, 2015).

Empirical studies have approached the theme through several aspects, such as: motivation; background; processes, and results (Goranova & Ryan, 2014), showing, however, little consensus on cause and effect relationships, which may be associated with difficulties in relation to the heterogeneity of activists and their goals (Brav, Jiang, Partnoy, & Thomas, 2008; González & Calluzzo, 2019) and their different mechanisms of action (Aguilera et al., 2015), as well as their unobservable actions (McCahery, Sautner, & Starks, 2015).

Previous studies such as those by Gillan and Starks (2003; 2007) argue the improvement of CG mechanisms by the action of activist shareholders, just as McNulty and Nordberg (2016) understand that the active investor will produce the same effect. On the other hand, some studies show results sometimes with a weak relationship between such factors (Elst, 2012), sometimes in the opposite direction (Punsuvo, Kayo, & Barros, 2007; Vargas, Bortolon, Barros, & Leal, 2017).

With regard to shareholder activism and the performance of target companies, contrasting results were observed: while some studies reported an increase in the value of the firm when adopting better CG and in the presence of more investor activism (Brav et al., 2008; Cuñat, Gine, & Guadalupe, 2012; Denes, Karpoff, & McWilliams, 2017; González & Calluzzo, 2019), others, focused on the domestic environment, presented inconclusive results (Brandão & Crisóstomo, 2015; Pereira, 2010; Xavier, Marcon, Lana, & Silva, 2013).

To address the aspects of activism-governance and activism-performance, the quantile regression method (QR) was used, considering it to be more comprehensive in the analysis of the possible heterogeneity of the relationships between the variables (Conyon & He, 2017; Prazeres, 2018; Shawatari, Salem, Hussain, Alaeddin, & Thabit, 2016).

Unlike the ordinary least squares (OLS) regression model, the QR uses linear programming minimization (Koenker, 2005) and offers results that would allow answering the following hypotheses:

• Hypothesis 1: there is a significantly lower effect of shareholder activism in companies with a better CG than in those with a worse CG.

• Hypothesis 2: there is a significantly larger effect of shareholder activism in companies with worse performance than in those with better performance.

Once activism was evidenced through the calculation of an index (SAIM), we tried to relate this index to the CG and performance variables of the selected sample. The application of the OLS regression model was less acceptable, while the QR presented more adequate results, confirming the argument of the heterogeneity of the effects.

With regard to the CG element, the results suggest that the greater the number of independent members on the boards of directors, the lower the occurrence of activism, indicating the importance of this CG mechanism. Regarding the relationship between the phenomenon and the performance variables, it was not possible to confirm Hypothesis 2, since the results were inconclusive.

This research contributes to a better understanding of the relationship between activism of shareholders and CG in the national context, where there are still few studies. Additionally, it offers the business and regulatory community elements to verify the behavior of activists in view of the expansion of their rights to participate in meetings.

LITERATURE REVIEW

Corporate governance

Jensen and Meckling (1976) proposed their agency theory based on their studies about property rights, agency costs, and finance theory. It is the background of this study on which the aspects related to CG will be treated. This theory predicts that business owners have conflicting relationships with contracted administrators, generating potential costs that may be mitigated through monitoring and incentives. Thus, shareholders would have their interests and equity stakes protected (Goranova & Ryan, 2015).

The way in which companies manage to structure themselves in order to minimize agency problems is called Corporate Governance. Despite the diversity of definitions of CG, for the purposes of this investigation, it is understood as the system that deals with the means through which shareholders ensure the return of their investments (Shleifer & Vishny, 1997). Widely studied in academia, it is understood that such a system has mechanisms that respond to external and internal pressures. Once efficiently employed, these mechanisms should produce alignment between managers and shareholders, leading to an improvement in the company’s performance (Aguilera et al., 2015). In this way, the process that assumes feedback from monitoring, reward, and return actions is materialized.

Among the studies associated with issues related to external pressures, there are those dealing with the legal and regulatory environment and their role in structuring CG profiles and levels of investor protection (Berglöf & Claessens, 2006; La Porta, Lopez-de-Silanes, Shleifer, & Vishny, 1998), which are complemented by more recent ones that seek to discuss CG in comparative terms at the level of firms, nations, and the relations between them (Aguilera, Marano, & Haxhi, 2019; Schiehll, Ahmadjian, & Filatotchev, 2014; Schiehll & Martins, 2016). The role of the business environment is investigated by research from the perspective of the market for corporate control (García-Castro, Aguilera, & Ariño, 2013; Shleifer & Vishny, 1997) and the role of institutional investors (Agrawal, 2012; García-Meca, López-Iturriaga, & Tejerina-Gaite, 2017) seeking to verify the effects on organizations and their respective performances.

Aspects related to internal CG mechanisms, corresponding to the pressures originating within organizations, have been studied and frequently focus on themes such as the structure of shareholding ownership (Carvalhal, 2012; González & García-Meca, 2014), boards of directors (Conyon & He, 2017; Santos, Orso, Lizote, & Marcon, 2018), and incentives for administrators (Pinto & Leal, 2013). In addition, there are studies that seek to measure the levels of CG in view of possible outcomes (Gompers, Ishii, & Metrick, 2003; Leal, Carvalhal-da-Silva, & Iervolino, 2015) through longitudinal studies, which intend to offer a measure of the evolution of this system in companies.

Academic works seek to provide answers to questions of causality and endogeneity in CG (Aggarwal, Erel, Ferreira, & Matos, 2011; García-Meca et al., 2017; Silveira, Leal, Carvalhal-da-Silva, & Barros, 2010). Aguilera, Desender, Bednar, and Lee (2015) present the discussion about CG as an exogenous or endogenous factor. On the other hand, equally important studies (Conyon & He, 2017; Cuñat et al., 2012; García-Castro et al., 2013), not necessarily seeking to evince such a relationship, approached the topic from other angles, formulating theoretical models or using different methods (the last three articles cited used, respectively: quantile regression, fuzzy, and discontinued regression).

In Brazil, attentive to the dissemination of CG, researchers have produced studies in order to verify the multiple characteristics and effects of this system on the domestic business environment. One of the main characteristics of Brazilian companies is their concentration of ownership. Several studies on this issue have been carried out, such as those by Carvalhal (2012), Brandão and Crisóstomo (2015), Saito and Silveira (2008), Sonza and Kloeckner (2014), and Valadares e Leal (2000), who show-among other conclusions-the conflicts between majority and minority shareholders.

Shareholder activism

Understood as an external governance mechanism, activism reflects a pressure to influence companies’ policies and practices, being played by both minority shareholders (shareholder activists) who use their corporate position to achieve their goals (Goranova & Ryan, 2014), as well as by other groups that do not own shares, but that have some interest in the target companies (stakeholder activists) (Aguilera et al., 2015). This study proposes to investigate the activism of the minority shareholders, relating it with the antecedent elements of CG and performance.

The preceding concepts refer to different factors such as: the types of actions carried out, the different objectives, the heterogeneous characteristics of the actors, and the different regulatory and business environments. They also bring, in their core, elements common to CG, namely: the need to monitor managers and provide incentives to them in order to maximize shareholder value.

There can be multiple approaches to the study of shareholder activism, given its complexity. Goranova and Ryan (2014) structure decades of research on the theme, pointing out: (a) their background, as characteristics of companies, the environment, and the activists themselves (Gillan & Starks, 2003; Judge, Gaur, & Muller-Kahle, 2010; La Porta et al., 1998); (b) processes, such as actions against administrators and activist tactics (Iliev, Lins, Miller, & Roth, 2015; McCahery et al., 2015); and (c) the outcomes regarding performance, CG, and reputation practices (Agrawal, 2012; Cuñat et al., 2012).

In a more recent study, McNulty and Nordberg (2016) expand Goranova and Ryan’s (2014) understanding when dealing with institutional investors, establishing the concept of active and passive ownership. In this first role, these investors would be engaged continuously, having a long-term perspective in relation to the invested company, in contrast to passive ownership, in which investors would keep their shares, negotiate, and eventually vote, but in a more casual way.

The choices of shareholders to act and safeguard their interests have traditionally been understood as exit, voice, and loyalty, referred as Hirschman’s theory (1970) (Gillan & Starks, 1998; Goranova & Ryan, 2014; McCahery et al., 2015), a study on the recovery mechanisms that can be used by economic actors to rescue dysfunctional situations of organizations and society as a whole. In a situation analogous to the loss of product quality, in the event of a drop in company performance and dissatisfaction as to the way in which it is conducted, shareholders could choose to sell their shares (exit) or speak out (voice), as presented by the author:

Voice is here defined as any attempt at all to change, rather than to escape from, an objectionable state of affairs, whether through individual or collective petition to the management directly in charge, through appeal to a higher authority with the intention of forcing a change in management, or through various types of actions and protests, including those that are meant to mobilize public opinion (Hirschman, 1970, p. 30).

Thus, the actions taken at the shareholders’ meetings are seen as a measure of the presence of activism, since they constitute a representative means of demonstrating dissatisfaction regarding performance and desire to change existing conditions (Elst, 2012; Iliev et al., 2015; McNulty & Nordberg, 2016), showing the involvement of shareholders in monitoring managers and maintaining their rights.

Compared to more quantitative studies related to CG, shareholder activism lies in a more difficult context, considering the relative novelty of the topic, especially when it is intended to study companies and/or countries where capital markets are not mature and where there is no critical mass of data that allows further analysis. Thus, it is possible to observe studies that approach the theme from the perspective of exploratory analysis with descriptive statistics and simpler regression models, with researchers careful to clarify their intention of not making inferences of causality between the terms of analysis (Iliev et al., 2015; McCahery et al., 2015).

The Brazilian academic literature on the subject, although scarce, shows recent contributions that can lead to a better understanding of the phenomenon. The works of Crisóstomo and González (2006), Punsuvo, Kayo, and Barros (2007), and Xavier, Marcon, Lana, and Silva (2013) focus on the analysis of the possible activism of pension funds as shareholders, highlighting their shareholding concentration, without, however, obtaining conclusive results.

A few studies broadened the analysis of shareholder activism in the Brazilian market, among which we highlight those by Almeida (2017) regarding the performance of institutional investors in the face of expropriation risks and Vargas (2013), Vargas, Bortolon, Barros, and Leal (2017), and Collares (2018), who produced a shareholder activism index. Guimarães, Leal, Wanke, and Morey (2019), in turn, use this evidence of shareholder activism to verify its impact on the efficiency of the target companies and find a negative relationship between these variables.

QUANTILE REGRESSION (QR)

QR is a method for estimating conditional quantile functions (Koenker & Hallock, 2001; Koenker, 2005; Wooldridge, 2010). Unlike regressions using the OLS method-the usual mean forecasting model (Hair, Black, Babin, Anderson, & Tathan, 2009) with which the method developed by Koenker and Bassett in 1978 is often compared (Wooldridge, 2010)-, QR allows the discussion of the relationship of the variables taking into account the possible heterogeneity in the responses associated with the covariates (Koenker, 2005). Such heterogeneity can be observed when a set of quantile regressions is performed (more easily visualized from graphs) and when comparisons are made between quantiles, selected according to the researcher’s interest.

The most expressive advantage of using this method over the traditional OLS estimation is that it offers a broader understanding of the relationship between dependent and independent variables (Conyon & He, 2017), since it presents robust results even in databases that contain outliers, residues with heteroscedastic characteristics, and non-Gaussian probability distributions (Prazeres, 2018).

The quantile regression model is given by a quantile function, indexed by the quantile τ ∈ (0,1), where Q_y (τ | x) denotes the quantile (τ) of y as a function of x, according to the following expression, whose linear function minimization task is a linear programming problem (Koenker, 2005):

(1) ${\mathit{Q}}_{\mathit{y}}\left(\mathit{\tau }\mid \mathit{x}\right)={\mathit{x}}_{\mathit{i}}^{\mathit{T}}\mathit{\beta }$

Quantile regression in empirical research has become more popular, since it seeks to expand the analysis of the effects of estimates in different segments of an analyzed database (Wooldridge, 2016). In this sense, we have identified some studies that, although dealing with topics such as CG, finance, and control from different angles, have in common the interest in explaining the different relationships between their dependent and independent variables, when the latter, according to their distribution, affect the former differently.

Using this method, the study by Conyon and He (2017) allowed a new understanding regarding the effects of gender diversity on the board of directors of corporations with shares listed in the US market, since the results, in addition to suggesting that the female presence positively impacts performance, indicate that this effect occurs differently throughout the distribution. Yensen, Yulu, and Paoyu (2017), in turn, investigated the relationship between CG and performance variables on changes in the shareholding of foreign companies in the Thai market.

Dang, Houanti, Le, and Vu (2018) applied the QR finding that the percentage of independent board members negatively impacts-however without significant difference among the quantiles studied-the performance (evaluated by the Returns on Assets-ROA) of the companies present on the Vietnamese stock exchange, while this difference is present when dealing with performance versus duality of the CEO. Still in order to ascertain the heterogeneity of the effects of CG mechanisms on the market value of Kuala Lumpur’s capital market companies, Shawatari, Salem, Hussain, Alaeddin, and Thabit (2016) used the method in question and obtained statistically significant results in different quantiles of the sample.

Chi, Huang, and Xie (2015) seek to reconcile conflicting results in the literature comparing the conventional method (OLS) and QR, with respect to the cost of bank loans and CG variables and argue that the heterogeneity of the relationships found in their investigation potentially explains the theoretical divergences.

As far as it is known, the use of quantile regression for studies of shareholder activism is non-existent, with Audretsch, Hülsbeck, and Lehmann (2013) touching on the theme in their study of families as active monitors of performance in family businesses. It would also be worth mentioning Prazeres (2018), who used the method in his study regarding property structure and conditional conservatism. Thus, the use of the method in a shareholder activism study constitutes an opportunity to verify the relationship of this phenomenon and the CG and performance, considering that there are different effects on the results.

METHOD

Data and variables

For this study, a database was used that includes the Shareholders’ Activism Index at Meetings (SAIM) of one hundred publicly traded companies, listed in the Brazilian exchange called Brazil, Bolsa, Balcão (B3), which in 2016 presented the highest trading liquidity. The selection criteria of the sample were due to: (a) the understanding that the set selected was representative of the business environment, considering that, at the time, they corresponded to approximately 90% of the total market capitalization of all companies in the Brazilian stock exchange and (b) the feasibility factor, considering the volume of documents analyzed, demanded by documentary research.

The SAIM is the sum of the frequency of occurrences of shareholder activism by company-year (receiving a value of one if the phenomenon is verified and zero in its absence). The frequencies were identified by reading the 315 minutes of the ordinary and extraordinary meetings of the sample companies, held between January/2016 and April/2017. To this material, collected through access to the CVM website, the content analysis method with mixed grids was applied (Neuendorf, 2012), with the support of the Atlas.ti software.

The indicator consists of nine items that correspond to shareholder actions to: (a) elect members on boards of directors and fiscal boards; (b) reject proposals from management; (c) vote against proposals from managers; (d) send their proposals before the meeting; (e) present their proposals during the meetings; (f) request the institution of fiscal boards; (g) request elections by cumulative or separate vote; (h) approve their own proposals; and (i) publish vote of dissension. Considering that the index is the sum of the events of meetings that are held in different quantities by each company each year, it may exceed nine per company-year.

The activism categories that make up the index are theoretically based on the academic literature on the topic (Gillan & Starks, 1998; 2003; Goranova & Ryan, 2014; McCahery et al., 2015; Vargas et al., 2017), by the relevant legislation to corporations (Law No. 6,385, 1976; Law No. 6,404, 1976; Law No. 10,303, 2001), and regulations (CVM instructions No. 480 and No. 481, dated December 7 and 17, 2009, respectively). The SAIM is the dependent variable of the econometric analysis proposed herein.

Financial and corporate information of the selected companies were also included in the database, referring to the fiscal years of 2015 and 2016, obtained by accessing the Economática® database. The criterion of data lag of one year (t) corresponds to the assumption that activism actions are reactive to the performance of companies in the year (t-1), with their eventual dissatisfaction expressed in meetings after the release of financial reports. Thus, SAIM is associated with the year of the meeting (t), and governance and financial indicators are associated with the previous year (t-1).

The data were treated in the form of a panel. The explanatory variables selected for this study were listed in Table 1, which presents description, theoretical basis, and expectation of relationship with the activism index (SAIM):

Table 1
Description of the explanatory variables.

Code	Sign	Description	Basis
Premium	-	Dummy that receives a value of one for companies that are listed in the premium trading segments of B3 called Novo Mercado and Level 2, and zero for other trading segments.	Companies with poor governance quality motivate shareholder activism actions (Gillan & Starks, 2003; Vargas et al., 2017).
Ind-BOD	-	Percentage of independent directors in the board of the sample companies.	Elst (2012) notes a weak relationship be-tween activism and elections for the board of directors (BOD). Vargas et al. (2017) find a negative relationship without statistical significance.
LN_SIZE	+	Net Revenue Logarithm. Since there are financial companies in the data-base, for them, the calculation was based on total assets.	Judge, Gaur, and Muller-Kahle (2010) observed a positive relationship between activ-ism and size of target companies when activism was social-driven; when financial-driven, there were no significant results.
QTOBIN	-	Ratio between the sum of the market value of the shares and debt, divided by total assets (equation simplified by Chung & Pruitt, 1994).	Pereira (2010) and Xavier et al. (2013) suggest, without conclusive results, greater activism if the indicator is lower than expected by shareholders, indicating loss of value in the company.
LEV	-	Ratio between total liabilities and total assets.	Gillan and Starks (2003) and Vargas et al. (2017) suggest that there is a possible replacement of shareholder monitoring actions by companies' creditors.

• Note. Prepared by the author

Model specification

Following the objectives of this study, multivariate data analysis was performed, having the SAIM as a dependent variable (by company-year, indexed by i, t), initially using linear regressions (OLS), expressed from the equation:

(2) ${\mathrm{SAIM}}_{\mathit{i},\mathit{t}}=\alpha +{\mathit{\beta }}_1{\mathit{PREMIUM}}_{i,t-1}+{\beta }_2\mathrm{IND}\_{\mathrm{BOD}}_{i,t-1}+{\beta }_3\mathrm{LN}\_{\mathrm{SIZE}}_{i,t-1}+{\beta }_4{\mathrm{QTOBIN}}_{i,t-1}+{\beta }_5{\mathrm{LEV}}_{i,t-1}+{\epsilon }_i$

The quantile regression to be estimated presents the same explanatory variables as Equation 2-being indexed by τ, the quantile to be estimated-and is expressed by the equation:

(3) $Q_{\tau }\left({\mathrm{SAIM}}_{i,t}\right)={\propto }_{\mathit{\tau }}+{\mathit{\beta }}_{1\mathit{\tau }}{\mathrm{PREMIUM}}_{i,t-1}+{\beta }_{2\tau }\mathrm{IND}\_{\mathrm{BOD}}_{i,t-1}+{\beta }_{3\tau }\mathrm{LN}\_{\mathrm{SIZE}}_{i,t-1}+{\beta }_{4\tau }{\mathrm{QTOBIN}}_{i,t-1}+{\beta }_{5\tau }{\mathrm{LEV}}_{i,t-1}+{\epsilon }_i$

To verify the different impacts of the explanatory variables on shareholder activism, quantile regressions were estimated, for the interval from the 10th to the 90th percentile, at each step of 0.1, thus characterizing the distribution of the sample in 10 deciles. With the resources offered by the Gretl software, graphs were generated for each variable in Equation 3, allowing to see the importance of the results found in the 80th percentile.

From that observation, quantile regressions were performed at Q50 (median of the distribution), Q20 (point of the distribution that has 20% of the observations below its SAIM), and Q80 (point of the distribution that has 20% of the observations above its SAIM), in order to verify if the differences between the quantiles were statistically significant in symmetric intervals.

RESULTS

The first result obtained referred to the calculation of the total SAIM in the two periods observed, as a result of the dichotomous treatment of the occurrence of activism, based on the reading of the minutes and accessory documentation pertinent to the meetings of the hundred selected companies. Table 2 shows the sum of the occurrences by the nine items associated with the theme, with the highest occurrence being found in ATIV3 and ATIV1, respectively votes against the management proposal and for the election of members for the board of directors and fiscal council.

Table 2
Shareholder Activism Index at Meetings (SAIM) by criteria (sum and percentage).

Code	Activism Requirements	SAIM	%
ACTIV1	Shareholders elect members of the Board of Directors or of the Fiscal Board	111	20%
ACTIV2	Shareholders reject (Management proposal does not pass)	2	0%
ACTIV3	Shareholders vote against, but fail to reject Management's proposal	257	47%
ACTIV4	Shareholders submit proposals during meetings	20	4%
ACTIV5	Shareholders submit proposals before meetings	3	1%
ACTIV6	Shareholders require the institution of a fiscal board	43	8%
ACTIV7	Shareholders require cumulative voting to elect board members	67	12%
ACTIV8	Shareholders approve their own proposals	5	1%
ACTIV9	Shareholders publish dissenting vote	43	8%
	Total SAIM of the companies analyzed	551	100%

• Note. The shareholder activism index at meetings (SAIM) was calculated from the analysis of the minutes of the meetings and is a dichotomous variable that assumes a value of one in the event of an item above. The SAIM column is the sum of the scores for all companies, in the two years, in the item and the column "%" is the percentage of this score in relation to the total .

By grouping the types of activism in actions associated with the themes of representativeness in boards (ACTIV1, ACTIV6, ACTIV7), actions on the proposals (own approval: ACTIV4, ACTIV5, and ACTIV8; rejection of those issued by the administration: ACTIV2, ACTIV3), and manifestation of dissension (ACTIV9), we have the following score, respectively: 221 (40%), 287 (52%), and 43 (8%).

There is no balance between the first two access channels to increase the influence of activist shareholders (elect directors and act actively in the rejection/approval of proposals). Strictly considering the items ACTIV1 and the sum of the items ACTIV2 and ACTIV8 as effective actions-binding in the sense of mandatory administrative action (unless the elected board members do not comply with legal requirements, they will be part of the BOD and winning proposals will be executed)-, they represent 111 (20%) and 7 (1%) of occurrences respectively. The results, consistent with those found by Vargas (2013), suggest that access to the board of directors and the fiscal board may have been favored by changes in legal statutes and regulations, since they allow minority shareholders to require different types of voting (by multiple or separate vote), changing the dynamics of the electoral process.

Figure 1 shows the frequency distribution of the companies’ SAIM, where it is possible to observe its asymmetry: on the right there are 20 companies that concentrate 38% of the points (212 points, average of 10.60) while 80 companies add 62% (339 points, average of 4.24).

Figure 1
Frequency distribution of companies’ SAIM (100).

From this point onwards, aiming at the calculation of econometric models, the SAIM started to be analyzed according to its occurrence each year (200 observations).

As can be seen in Table 3, the SAIM variable has a maximum value of 14 (Petrobras-2017) and a minimum of zero (Gol-2016 and Pão de Açúcar-2016). IND_BOD, which corresponds to the percentage of independent directors in the board, has a minimum value of zero in 32 observations (corresponding to 21 companies for the 200 observations in the two years analyzed). The companies with the highest proportion of independent directors are Valid and BMFBovespa (0.71). The QTOBIN performance variable shows negative values in 38 observations, which leads to an average of 0.22. No collinearity was observed between the model variables, since the correlation coefficients were less than 0.90 (Hair et al., 2009), as can be seen in Table 4.

Table 3
Descriptive statistics.

Statistics	SAIM	PREMIUM	Ind_BOD	LN_SIZE	LEV	QTOBIN
Minimum	0.00	0.00	0.00	12.89	0.10	-0.66
1st quartile	1.00	0.00	0.14	14.55	0.48	0.06
Median	2.00	1.00	0.28	15.82	0.60	0.24
Average	2.76	0.74	0.29	15.83	0.61	0.22
3rd quartile	3.00	1.00	0.43	16.67	0.77	0.40
Maximum	14.00	1.00	0.71	21.08	1.42	1.18
Standard Deviation	1.90	0.44	0.20	1.71	0.22	0.26
Observations	200	200	200	200	200	200

• Note. SAIM is the indicator of shareholder activism at meetings defined in Table 2. The other variables were defined in Table 1.

Table 4
Pearson's correlation between selected variables.

	SIAM		Ind_BOD		LN_SIZE		LEV		QTOBIN
SIAM	1.000
Ind_BOD	-0.318	***	1.000
LN_SIZE	0.249	***	-0.169	**	1.000
LEV	0.054		-0.029		0.485	***	1.000
QTOBIN	0.230	***	-0.240	***	0.346	***	0.646	***	1.000

• Note. SAIM is the indicator of shareholder activism at meetings defined in Table 2. The other variables were defined in Table 1. *** and ** indicate significance at the 1% and 5% levels, respectively.

Following the goal of this study, regressions were carried out to investigate the relationships between shareholder activism and elements of governance and performance. In this sense, sequentially, (a) OLS regression, (b) simultaneous quantile regressions, and (c) interquantile regressions were performed. The objectives of each procedure corresponded, respectively: to verify if the OLS method would be adequate to the model; to verify if there would be a differentiated impact between the variables; and how it would be configured in the set of QRs performed simultaneously and, from this behavior, to select interquantile intervals for analysis. In addition, to verify reverse causality, the OLS regression model (2) was incorporated into the study, in which the role of the variables SAIM and QTOBIN was switched as, respectively, explanatory and dependent.

Table 5 presents the results of the OLS regressions and, in advance, the results of the quantile regressions (Q20, Q50, and Q80), which will be discussed after the presentation of the results of the simultaneous QRs, following the logical sequence of the study.

Table 5
Results of OLS and quantile models.

				Dependent variable for quantile regressions: SAIM
Variables	Sign	OLS (1) Depend. var.: SAIM	OLS (2) Depend. var.: QTOBIN	Q20			Q50			Q80
					CI 90%		Coef.	CI 90%		Coef.	CI 90%
Constant		1.422	-0.077	2.000	-0.800	2.000	3.000	0.549	3.000	4.424	0.246	7.093
		(0.344)	(0.642)
PREMIUM	-	-0.587	-0.031	-1.000	-1.000	-0.321***	-1.000	-1.000	-0.230***	-1.114	-2.468	-0.296***
		(0.104)	(0.437)
Ind_BOD	-	-1.765**	-0.222***	-0.000	-1.254	0.000***	-0.000	-2.757	0.000***	-1.615	-3.499	-0.571***
		(0.015)	(0.006)
LN_SIZE	+	0.184*	-0.010	0.000	0.000	0.327	0.000	0.000	0.179	0.088	-0.079	0.290
		(0.056)	(0.373)
LEV	-	-1.618**	0.782***	-0.000	-1.701	0.000***	-0.000	-2.147	0.000***	-1.722	-3.392	0.130
		(0.047)	(0.000)
QTOBIN	-	1.604**		0.000	0.000	0.913	0.000	0.000	2.282	1.959	0.868	3.451
		(0.013)
SAIM	-		0.020**
			(0.013)
R2		0.185	0.487
R2 adjust.		0.164	0.473
# observ.		200	200	200			200			200
Breusch-Pagan
p-value		(0.005)	(0.031)
Chi-square				112.961			112.961			47.195
p-value				(0.000)			(0.000)			(0.000)
Anova
	Pr(>F) Q80 and Q20									(0.002)***
	Pr(>F) Q80 and Q50						(0.001)***
	Pr(>F) Q50 and Q20			(0.1)

• Note. Models estimated with ordinary least squares (OLS) and quantile regressions. The SAIM is the indicator of shareholders' activism at meetings defined in Table 2. The OLS (1) and (2) models have, respectively, the SAIM and the QTOBIN as dependent variables. The quantile regression models consider the quantiles Q20, Q50, and Q80 of the SAIM. Q20 corresponds to the point with 20% of the observations with the lowest SAIM. Q50 corresponds to the median of observations ordered by SAIM. Q80 corresponds to the point with 20% of the observations with the highest SAIM. The other variables were defined in Table 1. CI is the confidence interval. Breusch-Pagan is a heteroscedasticity test whose null hypothesis is that the variances of the residuals are equal (homoscedasticity). The null hypothesis of the Chi-Square test is that the distribution of residuals in the quantile regression models is normal. The Anova test verified that the effects of the regressors are uniform across the selected quantiles, with the significance indicating that they are not. ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively.

As can be seen, the OLS regression (1) presents most of its coefficients significant and the signs of these relationships within the expectations of this study, shown in Table 1, except for the relationship between shareholder activism and QTOBIN. The coefficient of determination (R²) indicates that the model explains 19% of the variance, but the p-value (0.005) resulting from the application of the Breusch-Pagan test rejects the hypothesis of homoscedasticity, revealing the inadequacy of the OLS model to the data set.

Regarding the OLS regression (2)-QTOBIN as a dependent variable-, we also observed significance in several explanatory variables, suggesting the existence of reverse causality, one of the classic problems of endogeneity, related to the theme. In this sense, the variables of interest (SAIM and QTOBIN) show significance in both models. Similarly to the first model, the result of the Breusch-Pagan test (0.031) rejects the hypothesis of homoscedasticity.

Still to verify whether the OLS method would be adequate for the analysis of the relationship between shareholder activism, GC, and performance, tests were carried out on the OLS model (1), to verify multicollinearity and adequacy of the choice of the random effects panel for data analysis. The variance inflation factor (VIF) test showed a small mean value of the variables (1.755) and the Hausman test resulted in a p-value of 0.7017, suggesting, respectively, the absence of multicollinearity and the adequacy of the random effects model.

In view of the objective of the study in verifying the differentiated behavior in the relationship of activism with governance and performance, regressions were applied to the quantiles (tau = τ) covering the intervals between 0.1 and 0.9, with increment of 0.1.

Such behavior can be better observed from its graphical representation, where the x-axes in Figures 2 and 3 show the quantile scale while the y-axes show, respectively, the effects of the variables Ind_BOD and QTOBIN on shareholder activism (SAIM). In both figures we have: (a) a dashed line that represents each estimated τ; (b) a shaded area corresponding to the 90% confidence interval (90% CI); and (c) a dotted line that represents the OLS regression coefficient.

Figure 2
SAIM and Ind_BOD: estimation of quantiles.

The y-axis shows the coefficients referring to Ind_BOD, the proportion of independent directors on the Board of Directors. The x-axis represents the quantile distribution according to the SAIM, the indicator of shareholders’ activism at meetings defined in Table 2.

Figure 3
SAIM and QTOBIN quantile estimation.

The y-axis shows the coefficients referring to QTOBIN, defined in Table 2. The x-axis represents the quantile distribution according to the SAIM, the indicator of shareholders’ activism at meetings defined in Table 2.

The estimates of a quantile point are interpreted as the effect of the change of independent directors in the board (Figure 2) and Tobin’s Q ratio (Figure 3) on the activism indicator, keeping each one of the other variables in the fixed statistical model.

The relationship between independent directors in the board and the activism indicator is decreasing in most of its points, indicating that the increase in the number of independent members reduces the occurrence of activism, being significant in Q40, Q50, Q60, and Q80 considering CI 90%. The results suggest that the greater presence of independent directors in the board can promote the improvement of CG, minimizing agency conflicts and, therefore, reducing activist action (Gillan & Starks, 2003; 2007; McCahery et al., 2015).

The relationship between performance (QTOBIN) and activism is positive and increasing, indicating that the improvement in performance may lead to an increase in shareholder activism, but with non-significant results (CI 90%). Although the direction of this relationship was not as expected, the results are compatible with those obtained by Brav, Jiang, Partnoy, and Thomas (2008) and by González and Calluzzo (2019).

Although the graph was not presented, the results for the dichotomous variable PREMIUM were evaluated and showed to be significantly relevant (90% CI) in selected quantiles (Q20, Q50, to Q80) and are negatively related to shareholder activism throughout the set of results, suggesting that companies with a better level of CG (according to the B3 listing level) tend to suffer less from activist investors, as observed by Leal, Carvalhal-da-Silva, and Iervolino (2015) and Vargas et al. (2017).

The other explanatory variables did not show differentiated behavior in the quantiles calculated.

The significant results in Q80, observed in Figures 2 and 3, led to their further analysis, comparing them with those resulting from the regressions Q50 (median) and Q20 (selected because it is the symmetric interval), as shown in Table 5.

In terms of differentiating the magnitude of the effect of the relationship between activism and Ind_BOD, Table 5 shows significant coefficients at the level of 1% for the regressions Q20, Q50, and Q80. This confirms the argument that this explanatory variable affects shareholder activism differently, and Hypothesis 1 about the negative relationship between activism and quality of CG practices cannot be rejected. Regarding the intensity of the performance effect on activism, coefficients without statistical significance are observed for the same selected quantiles, not supporting Hypothesis 2 about the negative relationship between activism and performance.

An attempt was then made to verify whether the effects of regressors are uniform across the selected quantiles, using the Anova criterion (Kleiber & Zeileis, 2008). The results showed a significant difference between Q80/Q50 and Q80/Q20 that does not occur between Q50/Q20.

As a way of assessing sensitivity, the data set (N = 49) with SAIM greater than that of the eighth quantile was applied to Equation 1, generating OLS (3), with the characteristics shown in Table 6. The same data set was used in OLS regression (4) with the stepwise backward method, which eliminates variables that do not contribute significantly to the best fit of the model (Hair et al., 2009). The results of the regressions led to an increase in the explanatory ability of the model, observed in the adjusted R², by 22.7% (OLS 3) and 24.7% (OLS 4), without rejecting the hypothesis of homoscedasticity of the Breusch-Pagan test, as shown in Table 6.

Table 6
Sensitivity tests.

		OLS (3)		OLS (4)		OLS (5)
Variables	Sign			(stepwise)		(stepwise)
		N(49)		N(49)		N(200)
Constant		-1.353		-0.950		-0.880
		(0.700)		(0.425)		(0.946)
PREMIUM	-	0.117
		(0.862)
Ind_BOD	-	-2.831	*	-2.766	*
		(0.067)		(0.062)
LN_SIZE	+	0.488	**	0.463	**	0.213	***
		(0.039)		(0.013)		(0.008)
LEV	-	-3.400	*	-3.332	*
		(0.064)		(0.060)
QTOBIN	-	4.410	***	4.405	***	1.171	**
		(0.008)		(0.007)		(0.025)
R2		0.308		0.307		0.085
R2 adjust		0.227		0.244		0.076
Observations		49		49		200
Breusch-Pagan
p-value		(0.062)		(0.071)		(0.001)

• Note. Models estimated by the method of ordinary least squares (OLS). The dependent variable is the SAIM, which is the indicator of shareholders' activism at meetings defined in Table 2. Models 3 and 4 were applied to the data set that contains 20% of the observations with the highest SAIM. Model 5 was applied to all observations in the sample. The other variables were defined in Table 1. CI is the confidence interval. Breusch-Pagan is a heteroscedasticity test whose null hypothesis is that the variances of the residuals are equal (homoscedasticity). ***, **, and * indicate significance at the 1%, 5%, and 10% levels, respectively.

Still testing the models, the OLS model (4) was applied to the total data set (N = 200), which best explains the regression applied to the data set described in the previous paragraph (N = 49). The last column of Table 6-OLS (5)-shows that this model loses its explanatory ability (adjusted R² = 0.76%) and assumes a p-value (0.001), not excluding the heteroscedastic characteristic of the data using the Breusch-Pagan test, which reinforces the argument of inadequacy of this regression method (OLS) for this database.

CONCLUSIONS

The use of content analysis (Neuendorf, 2012), associated with the set of shareholder activism categories supported by academic research (Gillan & Starks, 2007; McCahery et al., 2015) on documentation (minutes and respective annexes) of shareholders’ meetings, which are legal events that integrate and substantiate corporations, allowed the identification and measurement of the presence of activism in the selected sample (Collares, 2018; Leal et al., 2015; Vargas, 2013).

The scoring of occurrences of shareholder activism allowed the elaboration of an index (SAIM) that totaled 551 points in the database companies, showing that the distribution of points is not homogeneous: 20 companies answer 38% of the points (average of 10.60) while 80 of them concentrate 62% of the points (average of 4.56).

The evidences above give substance to the theories whose articulation was intended in this research, namely: the presence of conflicts in organizations as in agency theory (Jensen & Meckling, 1976), possibly mediated and minimized by the adoption of best CG practices (Gillan & Starks, 1998; 2003; McNulty & Nordberg, 2016), and the theory developed by Hirschman (1970) in which subjects can adopt attitudes of exit, voice, and loyalty, in an attempt to change the status quo (Goranova & Ryan, 2014). The activist action of manifestation of dissatisfaction regarding the conduct of the investee companies is understood as the voice element of Hirschman’s theory (1970) and, in this work, it was represented by the acts of the activists during the shareholders’ meetings.

Given the computation of the index, the next step was to analyze relative to governance and performance of the selected companies, verifying the most appropriate regression method, considering that such variables have different effects on the distribution of activism. Before that, however, it is important to comment that studies that deal with themes related to shareholder activism find recurring difficulties in evincing them, as the cause and effect relationships are not fully verifiable, nor are invisible actions (behind the scenes) (Gillan & Starks, 2007).

Quantile regression-which allows analysis of the results of the different regression quantiles (Koenker & Hallock, 2001; Koenker, 2005)-was more appropriate for the analysis, and the relationship between shareholder activism and the explanatory variables related to governance and performance was more representative after the eighth decile (Q80).

The negative and significant relationship found by means of quantile regressions between activism and CG variables-premium governance trading listings and proportion of independent board members-is in line with Gillan and Starks (1998; 2003) and Vargas et al. (2017), in the sense that the occurrence of activism decreases as the mechanisms that ensure better governance practices are effective, as shareholder dissatisfaction will be less. Among these CG variables, the proportion of independent directors in the board stands out for presenting significance in all models.

Regarding the results between performance and shareholder’s activism, it is observed that Tobin’s Q did not meet expectations regarding the direction of its relationship with shareholder activism and did not show significance, adding to the inconclusive results obtained by Pereira (2010) and Xavier et al. (2013) for Brazilian companies. Considering that QTOBIN has as a measure of good performance a value close to one and the sample median is 0.22, shareholders’ dissatisfaction with performance would be expected, which was not detected. Analysis by quantile regressions shows that the indicator affects shareholder activism differently, according to the observed quantile. Future works will be able to better study the performance from the segregation of the sample by groups that present certain levels of performance.

The fact that noticeable evidence was found in the relationship between shareholder activism and governance factors and that, at the same time, it was not possible to satisfactorily establish such a relationship with performance leads to the conclusion that CG is more important than performance for activism in the Brazilian context.

Like many, this study is also subject to limitations and caveats. The first is the lack of information regarding the events that occurred at the meetings: although there are legal requirements, in many companies analyzed the minutes and respective documents are not available. In addition, the vague nature of the writing of the minutes and the absence of a ballot paper may have led to an underestimated indicator (SAIM).

The second point is the limited empirical basis: the inclusion of other companies in the sample and a longer period to be analyzed could lead to more robust results and expand the power of explanation and generalization of the used model (QR), favoring the use of econometric models, including with regard to the search for possible causality. It is possible that there are activism actions among less liquid companies, for example.

Although the study has proposed to investigate the phenomenon of the actions that activist shareholders promote in the event that is structurally and legally representative of their power of action-the shareholders’ meeting-, this does not limit their possibilities of action. Opportunities for future research are the study of activism actions and respective effects in different channels, whether public (how activist campaigns use different media) or private (how activists make their “invisible” actions viable), to achieve their goals. The study of the effectiveness of the independent director in improving CG practices can be a natural way to advance the research presented here.

It is understood that this study has contributed to a better understanding of the phenomenon. The expansion of discussions and discoveries about activism in academia may be useful to practitioners and regulators, who, each in their own way, already face this issue as a reality.

• Data Availability
  All data and materials were made publicly available through the Mendeley platform and can be accessed at:
  http://dx.doi.org/10.17632/frxhy2hk94.1 Collares, Marta (2020), “Data for: "Corporate Governance: a major factor in shareholder activism in Brazil " published by RAC - Revista de Administração Contemporânea”, Mendeley Data, v1.
• Funding
  The author informed that there was no financial support for the research in this article.
• Copyrights
  RAC owns the copyright to this content.
• Plagiarism Check
  The RAC maintains the practice of submitting all documents approved for publication to the plagiarism check, using specific tools, e.g.: iThenticate.
• Peer Review Method
  This content was evaluated using the double-blind peer review process. The disclosure of the reviewers' information on the first page is made only after concluding the evaluation process, and with the voluntary consent of the respective reviewers.

ACKNOWLEDGEMENTS

The author is grateful to Ricardo P. Câmara Leal for his invaluable support for this research.

REFERÊNCIAS

• Activist Insight. (2019). The activist investing annual review 2019. Retrieved from https://corpgov.law.harvard.edu/wp-content/uploads/2019/02/AIAnnual_Review2019.pdf
  » https://corpgov.law.harvard.edu/wp-content/uploads/2019/02/AIAnnual_Review2019.pdf
• Aggarwal, R., Erel, I., Ferreira, M., & Matos, P. (2011). Does governance travel around the world? Evidence from institutional investors. Journal of Financial Economics. 100(1), 154-181. https://doi.org/10.1016/j.jfineco.2010.10.018
  » https://doi.org/10.1016/j.jfineco.2010.10.018
• Agrawal, A. K. (2012) Corporate governance objectives of labor union shareholders: Evidence from proxy voting. The Review of Financial Studies. 25(1), 187-226. https://doi.org/10.1093/rfs/hhr081
  » https://doi.org/10.1093/rfs/hhr081
• Aguilera, R. V., Desender, K., Bednar, M. K. & Lee, J. H. (2015). Connecting the dots: Bringing external corporate governance into the corporate governance puzzle. Academy of Management Annals, 9(1), 483-573. https://doi.org/10.5465/19416520.2015.1024503
  » https://doi.org/10.5465/19416520.2015.1024503
• Aguilera, R. V., Marano, V., & Haxhi, I. (2019). International corporate governance: A review and opportunities for future research. Journal of International Business Studies, 50, 457-498. https://doi.org/10.1057/s41267-019-00232-w
  » https://doi.org/10.1057/s41267-019-00232-w
• Almeida, A. L. P. (2017). Ativismo de investidores institucionais em empresas de capital aberto no Brasil (Master’s Thesis). Universidade Federal do Rio de Janeiro, Instituto COPPEAD de Administração, Rio de Janeiro, RJ, Brazil.
• Audretsch, D. B., Hülsbeck, M., & Lehmann, E. E. (2013) Family as active monitors of firm performance. Journal of Family Business Strategy, 4(2), 118-130. https://doi.org/10.1016/j.jfbs.2013.02.002
  » https://doi.org/10.1016/j.jfbs.2013.02.002
• Azevedo, R. (2019, April 06). Pagamento da CCR a delatores é contestado. Valor Econômico, Caderno Empresas, Infraestrutura, p. B3.
• Berglöf, E., & Claessens, S. (2006). Enforcement and good corporate governance in developing countries and transition economies. The World Bank Research Observer. 21(1), 123-150. https://doi.org/10.1093/wbro/lkj005
  » https://doi.org/10.1093/wbro/lkj005
• Brandão, I. F., & Crisóstomo, V. L. (2015). Ownership concentration and governance quality of brazilian firm. Brazilian Review of Finance. 13(3), 438-469. http://dx.doi.org/10.12660/rbfin.v13n3.2015.45739
  » http://dx.doi.org/10.12660/rbfin.v13n3.2015.45739
• Brav, A., Jiang, W., Partnoy, F., & Thomas, R. (2008). Hedge fund activism, corporate governance, and firm performance. The Journal of Finance. 63(4), 1729-1775. https://doi.org/10.1111/j.1540-6261.2008.01373.x
  » https://doi.org/10.1111/j.1540-6261.2008.01373.x
• Carvalhal, A. (2012). Do shareholder agreements affect market valuation? Evidence from Brazilian listed firms. Journal of Corporate Finance. 18(4), 919-933. https://doi.org/10.1016/j.jcorpfin.2012.04.003
  » https://doi.org/10.1016/j.jcorpfin.2012.04.003
• Chi, W., Huang, H., & Xie, H. (2015). A quantile regression analysis on corporate governance and the cost of bank loans: A research note. Review of Accounting and Finance. 14(1), 2-19. https://doi.org/10.1108/RAF-12-2012-0126
  » https://doi.org/10.1108/RAF-12-2012-0126
• Chung, K. H. & Pruitt, S. W. (1994). A simple approximation of Tobins’s q. Financial Management. 23(3), 70-74. https://doi.org/10.2307/3665623
  » https://doi.org/10.2307/3665623
• Collares, M. (2018) Ativismo de acionistas e transações com informação privilegiada no mercado brasileiro de capitais: uma possível relação (Master’s Thesis). Universidade Federal Fluminense, Niterói, Rio de Janeiro, Brazil.
• Conyon, M. J., & He, L. (2017) Firm performance and boardroom gender diversity: A quantile regression approach. Journal of Business Research. 79, 198-211. https://doi.org/10.1016/j.jbusres.2017.02.006
  » https://doi.org/10.1016/j.jbusres.2017.02.006
• Crisóstomo, V., & González, E. V. (2006). Possível estratégia de ativismo de fundos de pensão no Brasil. Revista de Economia Contemporânea. 10(1), 139-155. http://dx.doi.org/10.1590/S1415-98482006000100006
  » http://dx.doi.org/10.1590/S1415-98482006000100006
• Cuñat, V., Gine, M., & Guadalupe, M. (2012). The vote is cast: The effect of corporate governance on shareholder value. The Journal of Finance. 67(5), 1943-1977. https://doi.org/10.1111/j.1540-6261.2012.01776.x
  » https://doi.org/10.1111/j.1540-6261.2012.01776.x
• Dang, A., Houanti, L., Le, N. T., & Vu, M.-C. (2018). Does corporate governance influence firm performance? Quantile regression evidence from a transactional economy. Applied Economics Letters. 25(14), 984-988. https://doi.org/10.1080/13504851.2017.1390309
  » https://doi.org/10.1080/13504851.2017.1390309
• Denes, M. R., Karpoff, J. M., & McWilliams, V. B. (2017). Thirty years of shareholder activism: A survey of empirical research. Journal of Corporate Finance. 44, 405-424. https://doi.org/10.1016/j.jcorpfin.2016.03.005
  » https://doi.org/10.1016/j.jcorpfin.2016.03.005
• Elst, C. V. D. (2012). Shareholder rights and shareholder activism: the role of the general meeting of shareholders [Working Paper No. 188/2012]. European Corporate Governance Institute (ECGI) http://dx.doi.org/10.2139/ssrn.2017691
  » http://dx.doi.org/10.2139/ssrn.2017691
• García-Castro, R., Aguilera, R.V., & Ariño, M. A. (2013). Bundles of firm corporate governance practices: A fuzzy set analysis. Corporate Governance: An International Review. 21(4), 390-407. https://doi.org/10.1111/corg.12024
  » https://doi.org/10.1111/corg.12024
• García-Meca, E., López-Iturriaga, R., & Tejerina-Gaite, F. (2017). Institutional investors on boards: Does their behavior influence corporate finance? Journal of Business Ethics. 146(2), 365-382. https://doi.org/10.1007/s10551-015-2882-z
  » https://doi.org/10.1007/s10551-015-2882-z
• Gillan, S., & Starks, L. T. (1998). A survey of shareholder activism: Motivation and empirical evidence. Contemporary Finance Digest, 2(3), 10-34. Retrieved from https://ssrn.com/abstract=211409
  » https://ssrn.com/abstract=211409
• Gillan, S., & Starks, L. T. (2003). Corporate governance, corporate ownership, and the role of institutional investors: A global perspective. Journal of Applied Finance. 13(2). Retrieved from https://ssrn.com/abstract=480983
  » https://ssrn.com/abstract=480983
• Gillan, S., & Starks, L. T. (2007). The evolution of shareholder activism in the United States. Journal of Applied Finance. 19(1), 55-73. https://doi.org/10.1111/j.1745-6622.2007.00125.x
  » https://doi.org/10.1111/j.1745-6622.2007.00125.x
• Gompers, P., Ishii, J., & Metrick, A. (2003). Corporate governance and equity prices. The Quarterly Journal of Economics. 118(1), 107-156. https://doi.org/10.1162/00335530360535162
  » https://doi.org/10.1162/00335530360535162
• González, T. A., & Calluzzo, P. (2019). Clustered shareholder activism. Corporate Governance: an International Review. 27(3), 210-225. https://doi.org/10.1111/corg.12271
  » https://doi.org/10.1111/corg.12271
• González, J. S., & García-Meca, E. (2014). Does corporate governance influence earnings management in latin american markets? Journal of Business Ethics. 121, 419-440. https://doi.org/10.1007/s10551-013-1700-8
  » https://doi.org/10.1007/s10551-013-1700-8
• Goranova, M., & Ryan, L. V. (2014). Shareholder activism: A multidisciplinary review. Journal of Management. 40(5), 1230-1268. https://doi.org/10.1177/0149206313515519
  » https://doi.org/10.1177/0149206313515519
• Goranova, M., & Ryan, L. V. (2015). Shareholder empowerment: An introduction. In M. Goranova & L. V. Ryan (Eds), Shareholder empowerment: A new era of corporate governance (Cap. 1, pp. 1-35). London, UK: Palgrave Macmillan.
• Guimarães, P., Leal, R. P. C., Wanke, P., & Morey, M. (2019). Shareholder activism impact on efficiency in Brazil. Corporate Governance. 19(1), 141-157. https://doi.org/10.1108/CG-01-2018-0010
  » https://doi.org/10.1108/CG-01-2018-0010
• Hair, J. F. Jr. , Black, W. C., Babin, J., Anderson, R. E., & Tathan, R. L. (2009). Análise multivariada de dados Porto Alegre: Bookman.
• Harvard Law School Forum on Corporate Governance and Financial Regulation. (2019). Lazard’s Review of Shareholder Activism - Q1 2019 Retrieved from https://www.lazard.com/media/450943/lazards-q1-2019-review-of-shareholder-activism.pdf
  » https://www.lazard.com/media/450943/lazards-q1-2019-review-of-shareholder-activism.pdf
• Hirschman, A. O. (1970) Exit, voice and loyalty: Responses to decline in firms, organizations and states Cambridge, USA: Harvard University Press.
• Iliev, P., Lins, K. V., Miller, D. P., & Roth, L. (2015). Shareholder voting and corporate governance around the world. The Review of Financial Studies. 28(8), 2167-2202. https://doi.org/10.1093/rfs/hhv008
  » https://doi.org/10.1093/rfs/hhv008
• Institutional Shareholder Service (2018). Brazil proxy voting guidelines: Benchmark policy recommendations Retrieved from https://www.issgovernance.com/file/policy/active/americas/Brazil-Voting-Guidelines.pdf
  » https://www.issgovernance.com/file/policy/active/americas/Brazil-Voting-Guidelines.pdf
• Instrução CVM nº 480, de 07 de dezembro de 2009 (2009). Dispõe sobre o registro de emissões de valores mobiliários admitidos à negociação em mercados regulamentados de valores mobiliários. Com as alterações introduzidas pelas Instruções CVM nº 488/10, 509/11, 511/11, 520/12, 525/12, 547/14, 552/14, 561/15, 567/15, 568/15, 569/15, 583/16, 584/17, 585/17, 586/17 E 588/17. Retrieved from http://www.cvm.gov.br/legislacao/instrucoes/inst480.html
  » http://www.cvm.gov.br/legislacao/instrucoes/inst480.html
• Instrução CVM nº 481, de 17 de dezembro de 2009. (2009). Dispõe sobre informações, pedidos públicos de procuração, participação e votação a distância em assembleias de acionistas. Com as alterações introduzidas pelas Instruções CVM nº 552/14, 561/15, 565/15 e 594/17. Retrieved from http://www.cvm.gov.br/legislacao/instrucoes/inst481.html
  » http://www.cvm.gov.br/legislacao/instrucoes/inst481.html
• Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, 3(4), 305-360. https://doi.org/10.1016/0304-405X(76)90026-X
  » https://doi.org/10.1016/0304-405X(76)90026-X
• Judge, W. Q., Gaur, A., & Muller-Kahle, M. I. (2010). Antecedents of shareholder activism in target firms: Evidence from a multi-country study. Corporate Governance: An International Review. 18(4), 258-273. https://doi.org/10.1111/j.1467-8683.2010.00797.x
  » https://doi.org/10.1111/j.1467-8683.2010.00797.x
• Kleiber, C., & Zeileis, A. (2008) Applied econometrics with R. New York: Springer-Verlag
• Koenker, R. & Basset, G., Jr. (1978). Regression quantiles. Econometrica. 46(1), p. 33-50. Retrieved from https://people.eecs.berkeley.edu/~jordan/sail/readings/koenker-bassett.pdf
  » https://people.eecs.berkeley.edu/~jordan/sail/readings/koenker-bassett.pdf
• Koenker, R., & Hallock, K. F. (2001). Quantile regression. Journal of Economic Perspectives. 15(4), 143-156. https://doi.org/10.1257/jep.15.4.143
  » https://doi.org/10.1257/jep.15.4.143
• Koenker, R. (2005). Quantile regression Cambridge, UK: Cambridge University Press.
• La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. W. (1998). Law and finance. Journal of Political Economy. 106(6), 1113-1155. https://doi.org/10.1086/250042
  » https://doi.org/10.1086/250042
• La Porta, R., Lopez-de-Silanes, F., Shleifer, A., & Vishny, R. W. (2000). Investor protection and corporate governance. Journal of Financial Economics. 58(1-2), 3-27. https://doi.org/10.1016/S0304-405X(00)00065-9
  » https://doi.org/10.1016/S0304-405X(00)00065-9
• Leal, R. P. C., Carvalhal-da-Silva, A., & Iervolino, A. P. (2015). One decade of evolution of corporate governance practices in Brazil. Brazilian Review of Finance. 13(1), 134-161. http://dx.doi.org/10.12660/rbfin.v13n1.2015.50904
  » http://dx.doi.org/10.12660/rbfin.v13n1.2015.50904
• Lei nº 6.404, de 15 de dezembro de 1976 (1976). Dispõe sobre as sociedades por ações. Diário Oficial da República Federativa do Brasil. Brasília, DF.
• Lei n° 6.385, de 07 de dezembro de 1976 (1976). Dispõe sobre o mercado de valores mobiliários e cria a comissão de valores mobiliários. Diário Oficial da República Federativa do Brasil. Brasília. DF.
• Lei n° 10.303, de 31 de outubro de 2001 (2001). Altera e acrescenta dispositivos na Lei no 6.404, de 15 de dezembro de 1976, que dispõe sobre as Sociedades por Ações, e na Lei no 6.385, de 7 de dezembro de 1976, que dispõe sobre o mercado de valores mobiliários e cria a comissão de valores mobiliários. Diário Oficial da República Federativa do Brasil. Brasília, DF.
• McCahery, J. A, Sautner, Z., & Starks, L. T. (2015). Behind the scenes: The corporate governance preferences of institutional investors. The Journal of Finance. 71(6), 2905-2932. https://doi.org/10.1111/jofi.12393
  » https://doi.org/10.1111/jofi.12393
• McNulty, T., & Nordberg, D. (2016). Ownership, activism and engagement: Investors as active owners. Corporate Governance: An International Review. 24(3), 364-358. https://doi.org/10.1111/corg.12143
  » https://doi.org/10.1111/corg.12143
• Neuendorf, K. A. (2012) The content analysis Guidebook (2nd ed.). Los Angeles: Sage.
• Pereira, F. (2010). Ativismo dos investidores institucionais, valor, liquidez e governança corporativa: o caso do BNDES no Brasil (Master’s Thesis). Universidade Federal do Rio de Janeiro, Instituto COPPEAD de Administração, Rio de Janeiro, RJ, Brazil.
• Pinto, M. B., & Leal, R. P. C. (2013). Ownership concentration, top management and board compensation. Revista de Administração Contemporânea. 17(3), 304-324. http://dx.doi.org/10.1590/S1415-65552013000300004
  » http://dx.doi.org/10.1590/S1415-65552013000300004
• Prazeres, R. V. (2018). Relações entre a concentração de propriedade e o conservadorismo condicional nas companhias abertas brasileiras: Uma análise através do modelo de regressão quantílica. Revista Evidenciação Contábil & Finanças, 6(3), 21-36. https://doi.org/10.22478/ufpb.2318-1001.2018v6n3.36941
  » https://doi.org/10.22478/ufpb.2318-1001.2018v6n3.36941
• Punsuvo, F. R., Kayo, E. K., & Barros, L. A. (2007). O ativismo dos fundos de pensão e a qualidade da governança corporativa. Revista de Contabilidade e Finanças. 18(45), 63-72. https://doi.org/10.1590/S1519-70772007000400006
  » https://doi.org/10.1590/S1519-70772007000400006
• Schiehll, E., Ahmadjian, C., & Filatotchev, I. (2014). National governance bundles perspective: Understanding the diversity of corporate governance practices at the firm and country levels. Corporate Governance: An International Review. 22(3), 179-184. https://doi.org/10.1111/corg.12067
  » https://doi.org/10.1111/corg.12067
• Schiehll, E., & Martins, H. C. (2016). Cross-national governance research: A systematic review and assessment. Corporate Governance: An International Review. 24(3), 181-199. https://doi.org/10.1111/corg.12158
  » https://doi.org/10.1111/corg.12158
• Shleifer, A., & Vishny, R. W. (1997). A survey of corporate governance. The Journal of Finance. 52(2), 737-783. https://doi.org/10.1111/j.1540-6261.1997.tb04820.x
  » https://doi.org/10.1111/j.1540-6261.1997.tb04820.x
• Saito, R., & Silveira, A. D. M. da (2008). Governança corporativa: Custos de agência e estrutura de propriedade. Revista de Administração de Empresas. 48(2), 79-86. http://dx.doi.org/10.1590/S0034-75902008000200007
  » http://dx.doi.org/10.1590/S0034-75902008000200007
• Santos, R. C. dos, Orso, L. É., Lizote, S. A., & Marcon, R. (2018). Board of directors: The perspective of independents in the performance of private organizations? Evidences in Brazil. Revista de Administração Mackenzie. 19(4), eRAMF180003. http://dx.doi.org/10.1590/1678-6971/eramf180003
  » http://dx.doi.org/10.1590/1678-6971/eramf180003
• Shawatari, F. A., Salem, M. A., Hussain, H. I., Alaeddin, O., & Thabit, O. B. (2016). Corporate governance characteristics and valuation: Inferences from quantile regression. Journal of Economics, Finance and Administrative Science. 21(41), 81-88. https://doi.org/10.1016/j.jefas.2016.06.004
  » https://doi.org/10.1016/j.jefas.2016.06.004
• Silveira, A. D. M. da, Leal, R. P. C., Carvalhal-da-Silva, A. L., & de Barros, L. A. B. de C. (2010). Endogeneity of Brazilian corporate governance quality determinants. Corporate Governance. 10(2), 191-202. https://doi.org/10.1108/14720701011035701
  » https://doi.org/10.1108/14720701011035701
• Sonza, I. B., & Kloeckner, G. de O. (2014). Governança em estruturas proprietárias concentradas: Novas evidências para o Brasil. Revista de Administração. 49(2), 322-338. https://doi.org/10.5700/rausp1149
  » https://doi.org/10.5700/rausp1149
• Valadares, S. M., & Leal, R. P. C. (2000). Ownership and control structure of Brazilian companies. Revista Abante, 3(1), 29-56. Retrieved from https://ideas.repec.org/a/pch/abante/v3y2000i1p29-56.html
  » https://ideas.repec.org/a/pch/abante/v3y2000i1p29-56.html
• Valenti, G., & Schincariol, J. (2018, October 03). Acionistas buscam anular acordo com fundador. Valor Econômico, Caderno Empresas, p. B2.
• Vargas, L. H. F. (2013). Ativismo de acionistas no mercado acionário brasileiro: Evidências e determinantes em empresas listadas na BM&FBovespa (Master’s Thesis). Universidade Federal do Espírito Santo, Centro de Ciências Jurídicas e Econômicas, Vitória, Espírito Santo, Brazil. Retrieved from http://www.cienciascontabeis.ufes.br/pt-br/pos-graduacao/PPGCC/detalhes-da-tese?id=6527
  » http://www.cienciascontabeis.ufes.br/pt-br/pos-graduacao/PPGCC/detalhes-da-tese?id=6527
• Vargas, L. H. F, Bortolon, P. M., Barros, L. A., & Leal, R. P. C. (2017). Recent activism initiatives in Brazil. Relatórios COPPEAD, 432. Retrieved from https://pantheon.ufrj.br/bitstream/11422/2351/1/432.pdf
  » https://pantheon.ufrj.br/bitstream/11422/2351/1/432.pdf
• Wooldridge, J. M. (2010). Econometric analysis of cross section and panel data (2nd ed.). Massachusetts, USA: The MIT Press.
• Wooldridge, J. M. (2016). Introductory econometrics: A modern approach (6th ed.). Boston: Cengage Learning.
• World Bank Group. (2019). DataBank - World Development Indicators Retrieved from https://databank.worldbank.org/source/world-development-indicators?savedlg=1&l=en#
  » https://databank.worldbank.org/source/world-development-indicators?savedlg=1&l=en#
• Xavier, W. G., Marcon, L., Lana, J., & Silva, A. (2013, September). Fundos de pensão, ativismo e governança corporativa. Anais do Encontro Nacional da Associação Nacional de Pós-Graduação e Pesquisa em Administração, Rio de Janeiro, RJ, Brasil, 37.
• Yensen, N., Yulu, L., & Paoyu, H. (2017). Foreign institutional investors, shareholding change, and corporate governance. Emerging Markets Finance and Trade. 53(4), 764-775. https://doi.org/10.1080/1540496X.2015.1105634
  » https://doi.org/10.1080/1540496X.2015.1105634

Edited by

Editor-in-chief: Wesley Mendes-Da-Silva (Fundação Getulio Vargas, EAESP, Brazil) https://orcid.org/0000-0002-5500-4872

Reviewers: Antonio Zoratto Sanvicente (Fundação Getulio Vargas, EESP, Brazil) https://orcid.org/0000-0002-2888-8318

Valcemiro Nossa (Fucape Business School, Brazil) https://orcid.org/0000-0001-8091-2744

Publication Dates

• Publication in this collection
  24 June 2020
• Date of issue
  Sep-Oct 2020

History

• Received
  12 Oct 2019
• Reviewed
  07 Apr 2020
• Accepted
  07 Apr 2020