Shareholder Activism and Its Impact on Profitability, Return, and Valuation of the Firms in India

Abstract

The paper’s prime objective is to understand the impact of Shareholder activism on firm performance. This study is conducted in a unique setup where traditional activist investors such as pension funds and hedge funds are not present. However, the activism cases are increasing yearly in an emerging economy like India. We have created a comprehensive shareholder activism index (sha index) using multiple activisms and corporate governance factors. To measure firm performance, we have used valuation (Tobin’s Q and Market capitalization), profitability (operating profit margin and net profit margin), and return ratios (Return on capital and return on equity). Panel data analysis (PDA) is employed for the current study as it overcomes the shortcomings of the time series analysis and cross-sectional studies. The sample comprises 37 listed firms’ data for FY2017 to FY2020. Chosen firms have experienced activism instances at least once during the 2017–2020 period. As per our analysis, shareholder activism has a significant negative impact on valuation measured in market capitalization and profitability estimated by operating profit margin. Activism primarily impacts the other four parameters negatively, but it is insignificant. India is in the nascent stage of activism, partly explaining the insignificance of the effects of shareholder activism on firm performance. Also, activist investors are targeting companies. These attacks are not fructifying desired outcomes as promoters own over 50% stake in the listed companies. The latest data for FY2021 has not been considered for the study as covid-19 impacted the businesses during the financial year. Also, we cannot capture activism instances that are not reported in regulatory filings. Unlike past research in this area, we have used a comprehensive activism index as a proxy of activism and have employed PDA instead of event studies to assess the impact on firm performance. Also, this is the first such empirical study conducted in an emerging economy setup where neither large hedge nor pension funds are present.

1. Introduction

Shareholder activism is defined as engaging with company management and influencing their behavior, advocating policy changes, and impacting their overall conduct. Adopting activist proposed strategies is expected to help shareholders maximize wealth (Gillan and Starks 2000; Sjöström 2008). Bernard Black (1990) defines it as formal or informal monitoring of the corporate management. Though wealth maximization is seen as the goal of activist shareholders, an activist can have socially motivated goals along with financially motivated goals (Judge et al. 2010). We have understood the categories of issues raised by activists, which can fall under board-related issues, executives-related issues, Ideal Cash Related issues, Financial Performance & valuation-related issues, and other miscellaneous issues (Shingade and Rastogi 2020b).

There are diverse ways where the shareholders express their grief. They often sell the stock and move away, but few resolve with management as it is difficult to sell off the stocks or have many more benefits via activism (Gillan and Starks 2000). Berle and Means (1932) orated that those managers of public ltd firms have more power than the real stakeholders. The modern corporate structure is shareholders holding major stake commandeer the minority shareholders, especially in developing countries where the legal framework is still budding (Selarka 2005). Shingade and Rastogi (2019) claim that Shareholders Activism is a new evolving concept in India but well known in developed countries across the globe. Though the instances of shareholder activism are rising, studies in the literature did not focus on assessing the impact of such activism. In the past, shareholding activism has been discussed by Indian policymakers and academicians, but currently, we see actual events happening across Indian companies. It is observed that rejections of proposals by institutional shareholders are increasing year on year (Shingade and Rastogi 2020a). These enhanced activities can be attributed to the corporate governance framework, improving leaps and bounds (Narayanaswamy et al. 2012).

India has been experiencing the transformation in its corporate governance framework since introducing the new companies act in 2013. Various SEBI (Securities and Exchange Board of India) regulations have improved the corporate governance framework. Regulatory framework and effective execution are the critical enabler of Shareholder activism (Gordon and Pound 1993; La Porta et al. 2000b). The new initiatives are taken by regulators such as e-voting facility, appointing a minimum number of independent directors, separation of Chairman and Managing Director (MD) role has given more power to the activist shareholders in India. Institutional Investor stake increased to 34% in 2018 from 21% in 2001, and This is one of the key drivers of shareholder activism. The higher stake is associated with higher voting power and control policies.

Indian regulatory authorities such as the Securities and Exchange Board of India (SEBI), Reserve Bank of India (RBI), and Insurance Regulatory and Development Authority of India (IRDAI) are enforcing the participation of institutional investors in corporate governance activities of the investee companies. The conducive environment for protecting investors’ rights is fuelling shareholder activism. As a result, we have seen shareholder activism in India transforming from passive involvement of institutional shareholders (Khanna and Varottil 2015) to active participation and rejecting the proposal put forth by management. Institutional shareholders now have confidence in raising their voice rather than taking exits from their ownership in Indian listed companies (Khanna and Varottil 2015). However, whether this, the heightened shareholder activism results in the desired outcome, whether it is beneficial to firms or investors is not studied empirically. These vicissitudes inspired us to carry out the current empirical study.

Several studies have studied the impact of shareholder activism on firm performance (Karpoff et al. 1996) and have quantified such implications. Firm performance is measured in returns, profitability, or valuations (Wahal 1996; Carleton et al. 1998; Brav et al. 2008). Altaf and Shah (2018) empirically proven that an inverted U-shape relation exists between the ownership and performance of the firms when the investor protection enacts as a moderating variable. According to Filatotchev and Dotsenko (2015), though the empirical results did not identify any substantial impact of activism on a firm’s performance, partial implications of abnormal stock-market returns vary vividly depending on investor nature proposals in the UK. Similarly, the verdicts on French corporations show that activism is an inefficient tool that does not affect accounting choices (Bouaziz et al. 2020). Guimaraes et al. (2018) Results demonstrate a negative association between efficiency scores and the activism index, implying that activist shareholders prefer to invest in less efficient companies. Thus, the study proves that activism does not impact the efficiency scores of Brazilian listed companies.

The authors also considered Indian studies on corporate governance, which focus on qualitative aspects such as the board of directors, CEO duality, and executives’ compensation (Shingade and Rastogi 2020a). However, we found limited empirical literature which studies the impact of shareholder activism (Shingade and Rastogi 2019) on firm performance. In the literature, we found multiple studies on the assessment of firm performance globally (Karpoff et al. 1996; Wahal 1996; Carleton et al. 1998; Brav et al. 2008; Med bechir and Jouirou 2021). There are studies on corporate governance (Sinha 2006; Sharma 2016; Islam 2020; Chauhan et al. 2016) in India, but none of the studies assess the impact of activism. Globally, activism influence is measured mainly through event studies (Carleton et al. 1998; Brav et al. 2008) by assessing abnormal returns or through accounting-based ratios (Karpoff et al. 1996). Globally, research in this area revolves around large pension funds and hedge funds which may not represent a holistic picture of shareholder activism. In India scale of pension and hedge funds are still smaller. Studying the phenomenon will help understand how activism impacts such economic conditions. Authors have used the sha index as a proxy and assessed its effect on profitability, return ratios, and firm valuation.

From the literature, it is evident that most of the researchers have used qualitative aspects such as changing CEO (Helwege et al. 2012) or appointment of representative director on board or changes in policies (Subramanian 2017), as a success criterion for activism. Another set of studies focuses on the influence of enhanced corporate governance on firm performance. Taking cues from these comprehensive studies, authors got motivated to conduct the study considering accounting and market-based variables (Das et al. 2009; Croci et al. 2012) to assess the long-term effect of activism on targeted firms. The current study tries to bridge the gap identified by empirically proving the association of activism (sha index) and performance for the selected firms.

To the best of our knowledge, this study stands unique as it explores the relationship between shareholder activism (sha index) and firm performance for an emerging economy like India. Activism is considered legal, but it has yet to be empirically tested in such a setting. The study will help various stakeholders to understand whether activism has a tangible impact on performance in emerging economies. It will provide empirical evidence to activist shareholders from a firm performance point of view, which will help them devise proper strategies. Research concentrates on accounting and market-based measures only as we feel these measures cover and represent firm performance.

The remainder of the paper is schematized as follows. Section 2 delves into the existing literature on shareholder activism, related topics, and hypothesis formulation. The data and approaches utilized in model construction are presented in Section 3. The study’s findings are reported in Section 4, followed by a discussion of the findings in Section 5. The concluding paragraph of the document brings the paper to a close in Section 5.

2. Literature Review

Research on shareholder activism and its impact on firm performance is carried out across developed and developing countries. There are mixed outcomes of such studies. Some studies suggest positive impact while others have indicated negative or no effect.

According to Smith (1776), self-interest comes first for managers rather than serving the owners. Berle and Means (1932) discussed agency cost theory, which explains the relationship between stakeholders on modern corporations. Agency costs cannot be eliminated but can be reduced and kept under control through various measures (Berle and Means 1932; Jensen and Meckling 1976). Modern corporation structure has three layers—many owners or shareholders, directors appointed by these owners, and managers who run the corporation’s day-to-day operations (Fama and Jensen 1983). We found instances where firm ownership, majority of independent directors, CEO duality, and the legal framework in the country are used as control variables in studies (Smith 1996; Gillan and Starks 2000). Various issues raised by activist falls into (Shingade and Rastogi 2020b) one of the below categories—Financial performance and valuation-related issues, Idle cash and capital structure-related issues, Board of Director-related issues, Executive-related issues, and other miscellaneous issues.

2.1. Connecting Dots between Shareholder Activism and Firm Performance

Shareholder activism uses ownership position to influence the management and, eventually, firm policies and behavior (Sjöström 2008). Shareholder activism is discussed in connection with corporate governance, or we can say primarily issues in corporate governance are the drivers of shareholders activism (Carleton et al. 1998; Bizjak and Marquette 1998). Therefore, it is crucial to understand the linkages between these two phenomena. The connection between shareholder activism and corporate governance can be established theoretically through agency cost (Berle and Means 1932). Key corporate governance parameters which affect the firm performance are ownership stake and type, CEO duality, independence of the directors, the appointment of directors, and legal systems. For a single owner, it is easier to monitor the management and replace if not satisfied with the performance, which is not possible in the case of dispersing ownership structure (Rock 1990; Heard 1987). As per Gillan and Starks (2000), non-coordinated activism, i.e., individual shareholders, does not yield desired results. At the same time, Arfa et al. (2017) found a positive effect of institutional holing on firm performance on listed companies in France. Shingade and Rastogi (2019) found that India’s extent and type of ownership are widely discussed.

Abundant literature on the executive or management-related issues and their connection with firm performance is available. Activists also encourage the firms to have performance-based compensation for executives, which is expected to positively affect the firm’s performance (Croci et al. 2012; Fortin et al. 2014; Hartzell and Starks 2003). Cronqvist and Fahlenbrach (2008) also found the effect of activist shareholders on the firm’s compensation policies and other vital policies; This, in return, affects the firm performance positively (Cai and Walkling 2011; Fortin et al. 2014). Regulators across the globe are pushing corporates to have separate CEO and Chairman (of Board of Director) positions. The rationale behind the separation is that when the CEO becomes chairman of the board, he acts more from an executive perspective and is a chairperson. He influences the decision-making and monitoring functions of the board (Jensen and Meckling 1976; Fama and Jensen 1983). As per Arfa et al. (2017), shareholder activism has a significant relationship with CEO duality, BoD independence, firm operating performance, and size. Baliga et al. (1996) also opine that separation (Brickley et al. 1997) of CEO and chairman position does not have a meaningful impact on firm performance measured in both short and long term. Gafoor et al. (2018) found no significant relationship between CEO duality and performance for listed banks. From the banks’ performance perspective, board composition and size are important factors.

Boards of directors are at the center of shareholder activism as they monitor on behalf of owners (Berle and Means 1932; Jensen and Meckling 1976; Dalton et al. 2007). Therefore appointment, compensation, and independence of board matters are essential while assessing the impact of shareholder activism on firm performance. In multiple studies, it is found that board independence has no positive impact on firm performance, and even independent directors are mainly appointed by considering the CEO’s view Dalton et al. (2007). Even those directors have chosen independently without considering the CEO’s opinion align themselves with the CEO (Hermalin and Weisbach 1998). As per Adithipyangkul and Leung (2018), activist shareholders try to have performance-based compensation policies for non-executive directors. Morck et al. (1988) studied the relationship between directors’ stake in the company and its impact on firms’ performance using a set of 500 companies. Morck et al. (1988) used Tobin’s Q to measure firm performance; Tobin’s Q increases with an increasing stake of directors reaching a peak and starts declining.

Other than the above discussed corporate governance-related issues, idle cash and capital structure Jensen and Meckling (1976), firm size Smith (1996), and legal framework La Porta et al. (2000a) are also discussed in shareholder activism literature in the context of firm performance. As per Aguilera et al. (2015), there is an internal and an external mechanism for corporate governance. These mechanisms, independently or in conjunction, affect the governance and, thereby, firm performance. The external mechanism consists of the country’s legal system external market availability for resources such as executives’ external stakeholders such as statutory auditors, rating agencies, and media. A strong legal framework helps investors monitor the managers and control the firm policies La Porta et al. (2000b).

2.2. Shareholder Activism and Valuation

Activist shareholders target the companies showing abysmal performance and try to improve the performance by actively engaging with them. This involves getting a board seat to buy back offers to LBOs. All these actions are expected to help the firm improve operational efficiencies and the shareholder’s wealth. Firms with book value less than the market value get targeted quite often (Karpoff et al. 1996; Brav et al. 2008). Tobin’s Q is a widely used variable for valuation in shareholder activism literature and corporate finance (Bartlett and Partnoy 2020; Yin et al. 2018). In certain studies, the price to book value ratio (Pb) is used instead of Tobin’s Q to assess the valuation of the firms. In our study, we used Tobin Q variables as Pb ratios represent shareholders’ net worth, whereas Tobin Q represents total capital employed by the firm.

Multiple research papers have used event study methodology. Activist shareholder’s intervention date is chosen as an event date (Brav et al. 2008; Smith 1996; Karpoff et al. 1996; Carleton et al. 1998; Gillan and Starks 2000; Opler and Sokobin 1995; Bizjak and Marquette 1998; Cronqvist and Fahlenbrach 2008; Cai and Walkling 2011). A window of a certain period (7 days, 15 days, 30 days) is chosen as an event window, and a more considerable period (180 days, 365 days, and in some cases more than a year) is used to calculate the average share price return. Abnormal returns are calculated around the event window and compared with the average return, which shows the positive or negative impact of shareholder intervention. Brav et al. (2008) conducted an event study on a large data set from 2001–2006. As per him, hedge funds activism affects the firm performance positively. Hedge fund activism is found to positively impact the share price return (Brav et al. 2008; Smith 1996; Del Guercio and Hawkins 1999). Renneboog and Szilagyi (2011) also found a positive impact of shareholder activism on the share prices (Boyson and Mooradian 2011). In contrast, Prevost and Rao (2000) found a significant negative effect of shareholder proposals on share price return. Proposals that involve Poisson pills, Rejection of greenmail, adoption of antitakeover policies, rules related to director appointment, were found to have no significant impact on valuation ratios (Bizjak and Marquette 1998; Karpoff et al. 1996). Over the period, shareholder activism can have a varied effect on firm performance (Gillan and Stark, 2000). As most of the literature on valuation is around event studies, we attempt to assess the impact of activism using a comprehensive activism index on valuation. We are studying the below hypothesis using panel data analysis.

Hypothesis 1 (H1).

Shareholder Activism has an impact on Tobin’s Q of the firm.

Hypothesis 2 (H2).

Shareholder Activism affects the market capitalization of the firm.

2.3. Shareholder Activism and Profitability

In developed countries, we found research papers studying the impact of shareholder activism on firm performance (Karpoff et al. 1996). The effect of the sha index is quantified in terms of returns, profitability, or enhanced valuations (Wahal 1996; Carleton et al. 1998; Brav et al. 2008; DeHaan et al. 2019). Often poor firm performance attracts activist shareholders (Gillan and Starks 2000; Boyson and Mooradian 2011), and activist investors use various measures of firm performance. One widely used measure set is accounting-based ratios such as operating profit margin, net profit margin, and return ratios. It is found that the presence of activist shareholders tends to affect over-investment-related decisions and affect the firm performance positively Richardson (2006).

Brav et al. (2008) used accounting ratios such as return on asset (ROA), book value, cash on the book, and debt ratios of the targeted company to show the positive impact of activism on the company’s operating performance. Boyson and Mooradian (2011) studied the hedge fund activism effect over a longer period from 1994 to 2005 and found activism is associated with the positive operating performance of the firm. As per Kedia et al. (2016), operating performance improves significantly for firms targeted by activist investors.

Smith (1996) found no significant impact on the operating performance of the firms targeted by the CalPERS fund. Del Guercio and Hawkins (1999) studied the shareholder proposals of large and most active pension funds and interviewed the fund managers at these funds. They found no relationship between operating performance and activism. In line with this, Strickland et al. (1996) could not find any impact of shareholder proposals on operating performance measured by roa. As per Prevost and Rao (2000), firms targeted only once by activist investors have a statistically insignificant impact on operating performance, but firms targeted multiple times have a significant negative impact on operating performance over a longer period. As per the study conducted by DesJardine and Durand (2020), shareholder activism affects the firm performance positively but only for the short term. Over the longer term, it is associated with a decline in operating performance.

In some cases, roa generated by the targeted firms are on a lower side than non-targeted firms (Wahal 1996; Opler and Sokobin 1995). DeHaan et al. (2019) found that the firm’s size measured by market value drives the impact of shareholder activism on operating performance. Firms having a market value less than $40mn have positive long-term returns, whereas larger firms have no significant impact of activism on returns.

Most existing studies have focused on a single corporate governance variable, as discussed in the above sections. Also, these studies are conducted in the developed market and using firms targeted either by hedge funds or by large pension funds. We are using a comprehensive activism index to study the effect of shareholder activism on profitability measured using margin ratios and return ratios. The current study is the first empirical study on shareholder activism that employs panel data analysis to study the relationship between activism and profitability conducted in India. This will help understand the activism phenomenon in emerging markets where the pension and hedge funds are minimal. Figure 1 explains the strategy adopted for the study to achieve the formulated hypotheses.

Hypothesis 3 (H3).

Shareholder Activism has an impact on the Operating Profit Margin of the firm.

Hypothesis 4 (H4).

Shareholder Activism has an impact on the Net Profit Margin of the firm.

Hypothesis 5 (H5).

Shareholder Activism affects the return on capital employed of the firm.

Hypothesis 6 (H6).

Shareholder Activism affects the Return on Equity of the firm.

3. Data and Methodology

3.1. Data

This paper uses the data for the cross-section of 37 firms for four years (2017–2020). We have finalized those companies which have experienced activism events during FY2017–FY2020. Rejection of shareholder proposals by institutional investors is considered an activism event. We have gone through the filings of these companies with stock exchanges to finalize the sample of companies. We have finalized 37 companies as these companies’ data was available for all the variables under study across the chosen period. The secondary data is sourced from prowess CMIE database. The availability of enough sample size and observations for proper analysis and reliable results is the main criteria for selecting the firms and time. Additionally, the study’s most recent time has been considered to provide fresh evidence. The variables for which the data is fetched are explained in

Table 1. Variable List.

Parameters	Abbreviations	Definition	Formulae
Total sales of the company	sales	Total annual turnover of the company	-
Operating profit margin	opm	One of the widely used measures of profitability. It shows how efficiently the company is operating	EBITDA/Sales
Net Profit margin	npm	This is a profitability indicator that measures the net profitability of the company	PAT/Sales
Market Capitalisation	mc	Market capitalisation in Rs. Millions of the company. It shows the size of the company	Total no. of shares × share price
Tobin’s Q	Q	The widely used market-based performance measure	(mc + bv of debt)/(bv of total assets)
RoE	roe	Return on equity is a profitability measure that helps in assessing how are the returns given to shareholders by the company	PAT/Net Worth
RoC	roc	Return on capital measures how efficiently the company uses capital (equity + debt) to generate profits	EBITDA/Capital Employed
Institutional shareholding %	iii	Institutional Shareholder in the company holds the stake. One of the key variables related to shareholder activism	-
Activism Index	sha	Various corporate governance-related parameters are used in creating the index. It covers parameters such as Board of Directors, Executive compensation, Auditor.	-

Note: Table 1 describes the definition and formula of the variables used for the study.

.

3.2. Variables

The study has used six dependent variables representing firm performance, one explanatory variable, one control variable, and one instrument variable.

The dependent variable Tobin’s q (tq) is the measure of performance calculated by dividing the firm’s market value by the firm’s replacement cost (Lindenberg and Ross 1981; Fu et al. 2016). The m_cap (market capitalization) is the firm’s value measured as the multiplication of the number of shares of the firm and its price per share (Kumar and Shah 2009; Dias 2013). The dependent variable opm (operating profit margin) is another indicator of profitability computed as the ratio of operating income to net sales (Goel and Rekhi 2013; Goyal 2013). The npm measures firm profitability computed as the percentage net profit from sales (Khan 2011; Singh and Das 2018). The roe is another measure of firm profitability as the net income ratio to shareholder’s equity (Goyal 2013; Taani 2013). Similarly, roc is calculated as the ratio of operating income to employed capital (Goyal 2013; Taani 2013).

The explanatory variable is sha indicating shareholders’ activity. The sha is measured through an index that incorporates the regulatory measures for several CG practices following the Companies Act 2013. A greater index value shows a higher shareholders activity (Jewell and Mankin 2011; Nimer et al. 2015). Sales are taken as a control variable (instrument variable in 2sls) in terms of INR. It is measured by multiplying average sales price by the number of goods or services sold Kuzic et al. (2002). In 2SLS models, the current study has introduced institutional investors as the instrument variable, indicating the firm’s share of institutional investors as the percentage (Davis and Steil 2004; Song et al. 2020).

Construction of the Shareholder Activism Index

Shareholder activism denotes the involvement of the firm’s shareholders in corporate governance systems (CG). For the SHA’s evaluation, a customized SHA index is made. The current study has established the SHA index of the enterprises listed in India by adopting the studies Sarkar and Sarkar (2012), Vargas et al. (2018), and Islam (2020). This index is created by examining various CG activities and shareholders participating in such CG systems. A dichotomous methodology assigns the value “1” if the shareholders are involved in a specific activity; otherwise, the value “0” is assigned to that CG practice (Gillan and Starks 2000). The study only considers CG activities that adhere to the regulatory standards outlined in the Companies Act 2013 for Indian publicly traded companies.

3.3. Methodology

The panel data analysis (PDA) has been performed in the study to examine the assumed set of hypotheses. We have used the PDA due to its benefits over typical time-series or cross-sectional investigations. As per Baltagi (2008) and Hsiao (2007), the PDA reveals both the characteristics of time-series and cross-sectional analysis. Additionally, it also provides comparatively more information and variability. Panel data is helpful to study heterogeneity among individuals, and structure makes it possible to deal with certain types of endogeneity without the use of exogenous instruments. Extends the natural experiment framework to situations in which there may be endogeneity. Several studies (Bouaziz et al. 2020; Goranova et al. 2017; Hadani et al. 2011; Lee and Lounsbury 2011) support the usage of panel data analysis to better present regression results. STATA version 16 is used to get the regression results of the study.

The authors have included all firms having data accessible for all four years to avoid left-hand or right-hand censorship, resulting in balanced panel data. Panel data is valuable when dealing with unobservable unit heterogeneity bias (Wooldridge 2002; Hsiao 2007; Baltagi 2008). To deal with unobserved heterogeneity bias, researchers frequently select between two generally used estimate methods: fixed-effects and random-effects models. The main distinction is their assumption about the link between the unobserved unit effect and the observable explanatory factors. The random-effects model presupposes no association between them, but the fixed effect model makes no assumptions and allows for any correlation (Halaby 2004). In the present study, except for models one and six, all the other models are considered fixed effects as the best fit for the model by performing the Hausman test. The researchers have accomplished all the required diagnostics tests to assure that the methodologies adopted are accurate for the data analysis and empirical evidence presented are reliable.

We have developed six models based on six different dependent variables. We performed F-test and Breush-Pagan tests where both fixed and random effects were observed for Model 1, Model 5, and Model 6. Post this based on the Hausman test result; we used random effect regression in Models 1, 5, and 6. Due to the presence of endogeneity (confirmed by Durkin (score) Chi-square test and Wu Hausman test), we have used 2SLS regression for Models 2, 3, and 4. The following expressions are set for the mathematical model design of the PDA. The following econometric models are used in the study.

Methodology used for model 1—Random Effect

Model 1

$$t{q}_{it}=\alpha +{\beta }_{1}sh{a}_{it}^{ }+{\beta }_{2}l\_sale{s}_{it}+u_{it}$$

(1)

Methodology used for models 2, 3, and 4—Two-Stage least square

Model 2

$$l\_m{c}_{it}=\alpha +{\beta }_{1}sh{a}_{it}^{ }+{\beta }_{2}l\_sale{s}_{it}+u_{it}{}_{ }$$

(2)

$${\widehat{sha}}_{it}=\widehat{{\beta }_0}+\widehat{{\beta }_1}l\_sale{s}_{it}+\widehat{{\beta }_2}i{i}_{it}^{ }{}_{ } (1^{\mathrm{st}} \mathrm{stage})$$

(3)

$$l\_m{c}_{it}=\alpha +{\beta }_1{\widehat{sha}}_{it}^{ }+{\beta }_{2}l\_sale{s}_{it}+u_{it}{}_{ } (2^{\mathrm{nd}} \mathrm{stage})$$

(4)

Model 3

$$op{m}_{it}=\alpha +{\beta }_{1}sh{a}_{it}^{ }+{\beta }_{2}l\_sale{s}_{it}+u_{it}{}_{ }$$

(5)

$${\widehat{sha}}_{it}=\widehat{{\beta }_0}+\widehat{{\beta }_1}l\_sale{s}_{it}+\widehat{{\beta }_2}i{i}_{it}^{ }{}_{ } (1^{\mathrm{st}} \mathrm{stage})$$

(6)

$$op{m}_{it}=\alpha +{\beta }_1{\widehat{sha}}_{it}^{ }+{\beta }_{2}l\_sale{s}_{it}+u_{it}{}_{ } (2^{\mathrm{nd}} \mathrm{stage})$$

(7)

Model 4

$$np{m}_{it}=\alpha +{\beta }_{1}sh{a}_{it}^{ }+{\beta }_{2}l\_sale{s}_{it}+u_{it}{}_{ }$$

(8)

$${\widehat{sha}}_{it}=\widehat{{\beta }_0}+\widehat{{\beta }_1}l\_sale{s}_{it}+\widehat{{\beta }_2}i{i}_{it}^{ }{}_{ } (1^{\mathrm{st}} \mathrm{stage})$$

(9)

$$np{m}_{it}=\alpha +{\beta }_1{\widehat{sha}}_{it}^{ }+{\beta }_{2}l\_sale{s}_{it}+u_{it}{}_{ } (2^{\mathrm{nd}} \mathrm{stage})$$

(10)

Methodology used for models 5 and 6—Random Effect

Model 5

$$ro{c}_{ }=\alpha +{\beta }_{1}sh{a}_{ }^{ }+{\beta }_{2}l\_sale{s}_{ }+u_{it}$$

(11)

Model 6

$$ro{e}_{ }=\alpha +{\beta }_{1}sh{a}_{ }^{ }+{\beta }_{2}l\_sale{s}_{ }+u_{it}{}_{ }$$

(12)

In the above model expressions, the dependent variables are tq (Tobin’s q), l_mc (log of market capitalization), opm (operating profit margin), npm (net profit margin), roc (return on capital), and roe (return on equity). All the dependent variables represent firm performance. The sha (shareholders activity index) is the explanatory variable, and l_sales (log of sales) is the control variable. $\alpha$ and $u_{it}$ represent the constant and the error term, respectively. Additionally, in 2SLS models (Models 2, 3, and 4), sha is instrumented using two instrument variables l_sales and ii (institutional investors). The variables under hats show the estimated values.

4. Result

The current section of the paper is devoted to the result of the study. Further, we provide a detailed description of the outcomes of the models designed for the investigation.

4.1. Descriptive Statistics

Table 2. Descriptive Statistics.

Correlation Matrix
	sha	opm	npm	tq	roe	roc	m_cap	sales	ii	Mean	SD
sha	1									0.8295	0.0635
opm	0.0259	1								0.2991	0.2493
npm	−0.0244	0.2856 *	1							0.0787	0.1446
tq	0.0721	−0.2206 *	0.3841 *	1						1.5916	1.3423
roe	−0.1227	−0.0365	0.5576 *	0.3785 *	1					0.1185	0.1875
roc	−0.0141	−0.1172	0.6043 *	0.6612 *	0.7591 *	1				0.0835	0.0940
m_cap	0.3749 *	−0.0049	0.2076 *	0.4337 *	0.1569	0.3795 *	1			264,590.6	674,294.2
sales	0.3446 *	−0.1491	−0.0719	−0.0376	−0.0685	−0.0441	0.3951 *	1		179,448.4	504,533.1
ii	0.4209 *	0.2186 *	0.0734	−0.0774	−0.1208	−0.0686	0.3312 *	0.1481	1	0.3308	0.1932

Note: The correlation matrix has the values of correlation coefficients. * Significance level at 5%.

shows the descriptive statistics for the variables and their correlation. The average sha value is 0.8295. It suggests that Indian companies have a high level of shareholder activity. The opm and npm have an average value of 29.91% and 7.89%, respectively, showing a reasonable operating margin and net profit margin level in these firms. The performance measure tq has the mean value of 1.5916, which is good enough for firms in India. The performance measure roe and roc have the mean value of 11.85% and 8.18%, indicating a moderate roe and roc (performance). On average, these firms have m_cap (market capitalization) of INR 264,590 (mn) and sales of INR 179,448 (mn). Both m_cap and sales are large enough and are representative of Indian firms. There is significant participation of institutional investors (ii) in firms operating in India as it shows on average, Institutional investors hold a 33.08% stake. Developed models do not face multicollinearity problems. The sales variable has a positive correlation with sha and m_cap. The instrument variable ii also positively correlates with sha, opm, and mcap.

4.2. Model 1

Model1 finds the link between the TQ (financial performance) and sha (shareholders activity). As per model diagnostics (

Table 3. Model 1.

Part A (Coefficient Analysis) Dependent Variable: tq
	Standard Errors (Normal)			Robust Standard Errors
Variable Name	Coefficient	SE	p-Value	Coefficient	SE	p-Value
Constant	3.764 *	1.904	0.048	3.764	2.927	0.587
sha	−1.64	2.198	0.455	−1.64	3.02	0.198
l_sales	−0.07705	0.1060	0.468	−0.07705	0.1069	0.471
Part B (Model Estimates)
F-test (Model)	2.28 * (0.0000)
R-Square	0.0411
σμi	0.6828			0.6828
F-test Fixed Effect	9.31 * (0.0000)
Breush-Pagan Test	91.05 * (0.0000)
Hausman Teat	4.22 (0.1215)
No of observations (n)	144
Degree of freedom	106
Wald test for Heteroscedasticity 1	16,672.36 * (0.0000)
Wooldridge Autocorrelation Test 2 AR (1)	88.306 * (0.0000)

Note: ¹ Wald test of heteroscedasticity has the null of no heteroscedasticity. ² Wooldridge test of autocorrelation in the panel has the null of no autocorrelation (with 1 lag). σ_iμ is the variance of individual effect (companies in this case). Robust estimates are estimated due to significant Heteroscedasticity and Autocorrelation. DV is tq (Performance). The higher value represents the greater performance. The sha is the explanatory variable and represents the shareholders’ activities. The l_sales is a log of sales and is a control variable. * Significance at 5%.

), The F-test and B-P test have significant p-values (i.e., 0.000 < 0.05), which shows both fixed as well as random effects are significant. Therefore, the Hausman test has been used to select the appropriate model. The Hausman test has p-values > 0.05, and the random effect regression model is used (Arellano and Bond 1991; Baltagi 2008). The Wald test confirms the presence of heteroscedasticity (significant p-value < 0.05), and the Wooldridge test confirms the existence of autocorrelation (significant p-value < 0.05) (Arellano and Bond 1991). Therefore, robust estimates have also been mentioned (see Table 3). Rho representing the variance in individual effect is 0.6828 (Baltagi 2008; Hsiao 2007.)

The explanatory variable sha has a negative coefficient (−1.6405), but its p-value is insignificant (>0.05). This indicates that there exists no significant linkage between sha and TQ. This means shareholders’ activities do not influence the firm performance measured in terms of TQ for Indian firms. The control variable l_sales is also insignificant. The robust estimates also confirm the outcome of a random effect regression model. Thus, the assumed hypothesis H1 is rejected.

4.3. Model 2

The result of Model2 appears in

Table 4. Model 2.

Part A (Coefficient Analysis) Dependent Variable: l_mc
	Standard Errors (Normal)			Robust Standard Errors
Variable Name	Coefficient	SE	p-Value	Coefficient	SE	p-Value
Constant	14.4843 *	1.7563	0.000	40.9624	842.121	0.961
sha	−5.7581 *	1.523	0.000	−41.2307	1128.626	0.971
l_sales	−0.1118	0.1275	0.383	0.3914	8.9303	0.965
Part B (Model Estimates)
F-test (Model)	7.28 * (0.0011)
R-Square	0.1208
σμi	0.9384			0.9384
F-test Fixed Effect	20.13 * (0.0000)
Breush-Pagan Test	105.34 *(0.0000)
Hausman Teat	28.7 *(0.0000)
No of observations (n)	144
Degree of freedom	106
Wald test for Heteroscedasticity 1	3833.90 * (0.0000)
Wooldridge Autocorrelation Test 2 AR (1)	34.053 * (0.0000)

Note: ¹ Wald test of heteroscedasticity has the null of no heteroscedasticity. ² Wooldridge test of autocorrelation in the panel has the null of no autocorrelation (with 1 lag). σ_iμ is the variance of individual effect (companies in this case). Robust estimates are estimated due to significant Heteroscedasticity and Autocorrelation. DV is l_mc (log value of market capitalization representing performance). The higher value represents the greater performance. The sha is the explanatory variable and represents the shareholder’s activities. The l_sales is the log of sales and is the control variable. * Significance at 5%.

. This model investigates the connection between l_mc (valuation) and sha. The model follows the 2SLS regression due to the existence of the endogeneity (Table 9), as the F-test and B-P test are significant (p-value 0.000 < 0.01). Hence, no consensus can be made to choose from FE or RE. Therefore, the Hausman test is performed to confirm the suitability of the FE or RE model. The test indicates that the model is consistent with the fixed effect (Arellano and Bond 1991; Baltagi 2008).

The Wald and Wooldridge tests have been found significant. Thus, this indicates the presence of both heteroscedasticity and auto-correlation (Arellano and Bond 1991). Hence, the robust standard error estimates are presented (Baltagi 2008; Hsiao 2007) in Table 4. The coefficient of the sha is both negative (−5.7581) and significant at 5%. It indicates that the sha negatively influences the l_mc, supporting H2. This implies that the shareholders’ activities decrease the valuation (l_mc). Unlike the regular estimates, the robust estimates indicate no significant linkage between sha and l_mc. As in model1, the control variable l_sales are also insignificant in model2. Further, considering the issue of endogeneity bias, variable ii (institutional investors) appears to be the excellent instrument variable as it is not correlated with the error term but correlated with sha.

4.4. Model 3

Model 3 (given in

Table 5. Model 3.

Part A (Coefficient Analysis) Dependent Variable: opm
	Standard Errors (Normal)			Robust Standard Errors
Variable Name	Coefficient	SE	p-Value	Coefficient	SE	p-Value
Constant	−0.7462 *	0.3072	0.017	−47.892	1527.049	0.975
sha	−0.5269 *	0.2663	0.050	62.634	2045.784	0.976
l_sales	0.1406 *	0.0223	0.000	−0.3572	16.1334	0.982
Part B (Model Estimates)
F-test (Model)	20.84 * (0.0000)
R-Square	0.2823
σμi	0.959			0.959
F-test Fixed Effect	41.78 * (0.0000)
Breush-Pagan Test	160.56 * (0.0000)
Hausman Teat	20.72 * (0.0000)
No of observations (n)	144
Degree of freedom	106
Wald test for Heteroscedasticity 1	66,770.76 * (0.0000)
Wooldridge Autocorrelation Test 2 AR (1)	1.351 * (0.2530)

Note: ¹ Wald test of heteroscedasticity has the null of no heteroscedasticity. ² Wooldridge test of autocorrelation in the panel has the null of no autocorrelation (with 1 lag). σiμ is the variance of individual effect (companies in this case). Robust estimates are estimated due to significant Heteroscedasticity and Autocorrelation. DV is opm (operation margin representing performance). The higher value represents the greater performance. The sha is the explanatory variable and represents the shareholder’s activities. The l_sales is the log of sales and is a control variable. * Significance at 5%.

) examines the impact of sha on operating margin (opm). In model 3, we have used 2SLS due to endogeneity. Both F-test and B-P tests show fixed and random effects, respectively. Therefore, we have performed the Hausman test, which confirmed that the FE is a consistent model to go with (having a significant p-value < 0.01 (Arellano and Bond 1991; Baltagi 2008).

There exists heterogeneity as the Wald test has a significant p-value (<0.05). However, the null of autocorrelation is not rejected due to an insignificant p-value (>0.05). Hence, auto-correlation is not the problem in this model (Baltagi 2008). As heterogeneity exists, robust estimates are also observed (Baltagi 2008; Hsiao 2007).

The negative and significant coefficient (−0.5269 with p-value 0.05) of the sha reveals a negative influence of sha on opm. It means sha also deteriorates the operation margin in Indian firms. The hypothesis H3 is accepted here in this model. However, the robust estimates indicate no significant linkage of sha and opm as the endogeneity issue has also been the prime concern in this model. Therefore, instrument variables (lV) have also been introduced in model3, in which ii (institutional investors) has been found the good IV.

4.5. Model 4

Model4 is designed for investigating the effect of sha on npm.

Table 6. Model 4.

Part A (Coefficient Analysis) Dependent Variable: npm
	Standard Errors (Normal)			Robust Standard Errors
Variable Name	Coefficient	SE	p-Value	Coefficient	SE	p-Value
Constant	−0.1022	0.3063	0.739	−121.902	3961.655	0.975
sha	−0.3698	0.2656	0.167	162.804	5307.52	0.976
l_sales	0.0462 *	0.0222	0.040	−1.2399	41.861	0.976
Part B (Model Estimates)
F-test (Model)	2.89 (0.0602)
R-Square	0.0516
σμi	0.835			0.959
F-test Fixed Effect	11.64 * (0.0000)
Breush-Pagan Test	104.53 *(0.0000)
Hausman Teat	7.16 * (0.0279)
No of observations (n)	144
Degree of freedom	106
Wald test for Heteroscedasticity 1	1.4 × e6 * (0.0000)
Wooldridge Autocorrelation Test 2 AR (1)	4.676 * (0.0375)

Note: ¹ Wald test of heteroscedasticity has the null of no heteroscedasticity. ² Wooldridge test of autocorrelation in the panel has the null of no autocorrelation (with 1 lag). σiμ is a variance of individual effect (companies in this case). Robust estimates are estimated due to significant Heteroscedasticity and Autocorrelation. DV is npm (non-performing asset). The higher value represents the more outstanding performance. The sha is the explanatory variable and represents the shareholder’s activities. The l_sales is a log of sales, and ii (institutional investors) are the instruments variable. * Significance at 5%.

demonstrates the PDA result of Model4. The endogeneity test confirms the endogeneity issue in the explanatory variable (significant p-value < 0.05 rejects the null of no endogeneity). Therefore, the 2SLS model is applied by introducing the IV. Further, the Hausman test is performed to ascertain the consistency of the FE model (as the p-value of 0.025 is significant at 0.05) (Arellano and Bond 1991; Baltagi 2008).

The Wald test ascertains the presence of heteroscedasticity with a significant p-value at 5%. In addition to this, the Wooldridge test rejects the null of no autocorrelation, confirming the existence of autocorrelation Baltagi (2008). Therefore, the robust estimates are also presented (Baltagi 2008; Hsiao 2007) in Table 6.

On observing the coefficient of sha in this model, its value is −0.369 with a p-value of 0.167, showing an insignificant coefficient. Therefore, it indicates that sha does not significantly affect the npm of targeted firms. In the robust estimates, the coefficient is again found insignificant. As the 2SLS regression has been performed due to the endogeneity, the ii variable is found to be the good IV for the model as having no correlation with the error term but a significant correlation with the sha.

4.6. Model 5

The influence of sha on roc is examined in model5.

Table 7. Model 5.

Part A (Coefficient Analysis) Dependent Variable: roc
	Standard Errors (Normal)			Robust Standard Errors
Variable Name	Coefficient	SE	p-Value	Coefficient	SE	p-Value
Constant	0.1788	0.1351	0.186	0.1788	0.1319	0.175
sha	−0.1355	0.1571	0.389	−0.1355	0.1216	0.265
l_sales	0. 0016	0.0074	0.827	0.0016	0.0071	0.820
Part B (Model Estimates)
F-test (Model)	0.64 (0.5295)
R-Square	0.0118
σμi	0.6594			0.6594
F-test Fixed Effect	8.28 * (0.0000)
Breush-Pagan Test	88.33 *(0.0000)
Hausman Test	0.55 (0.7588)
No of observations (n)	144
Degree of freedom	106
Wald test for Heteroscedasticity 1	22,529.47 * (0.0000)
Wooldridge Autocorrelation Test 2 AR (1)	2.435 * (0.1276)

Note: ¹ Wald test of heteroscedasticity has the null of no heteroscedasticity. ² Wooldridge test of autocorrelation in the panel has the null of no autocorrelation (with 1 lag). σiμ is the variance of individual effect (companies in this case). Robust estimates are estimated due to significant Heteroscedasticity and Autocorrelation. DV is roc (Performance). The higher value represents the greater performance. The sha is the explanatory variable and represents the shareholder’s activities. The l_sales is a log of sales and is a control variable. * Significance at 5%.

presents the outcome of this model. The F-test and the B-P test are significant for this model (p-values are less than 0.05). We have conducted the Hausman test, which shows the presence of random effect (p-value > 0.05); hence RE regression model is used (Arellano and Bond 1991; Baltagi 2008). The B-P test for Heteroskedasticity rejects the null of constant variance. Hence, Heteroskedasticity exists in the model. Therefore, the robust standard error is also estimated (Baltagi 2008; Hsiao 2007). The Wooldridge test suggests that the null of no autocorrelation cannot be rejected; hence, the model has no autocorrelation (Baltagi 2008).

The value of the sha coefficient is found negative (−0.1355) but insignificant (>0.05 significance level). The finding indicates no significant association of sha to roc in Indian firms. A similar outcome is also found in the robust estimates. The result rejects the assumed hypothesis H5.

4.7. Model 6

Finally, the impact of sha on roe is tested in Model6. The outcome of the regression model is reported in

Table 8. Model 6.

Part A (Coefficient Analysis) Dependent Variable: roe
	Standard Errors (Normal)			Robust Standard Errors
Variable Name	Coefficient	SE	p-Value	Coefficient	SE	p-Value
Constant	0.3957	0.2444	0.105	0.3957	0.1986	0.046
sha	−0.3168	0.3073	0.303	−0.3168	0.2756	0.250
l_sales	−0.0013	0.0115	0.907	−0.0013	0.0117	0.907
Part B (Model Estimates)
F-test (Model)	0.51 (0.6043)
R-Square	0.0000
σμi	0.1768			0.1768
F-test Fixed Effect	1.79 * (0.0122)
Breush-Pagan Test	4.98 * (0.0128)
Hausman Test	1.35 (0.5097)
No of observations (n)	144
Degree of freedom	106
Wald test for Heteroscedasticity 1	1.0 × e6 * (0.0000)
Wooldridge Autocorrelation Test 2 AR (1)	0.147 * (0.7035)

Note: ¹ Wald test of heteroscedasticity has the null of no heteroscedasticity. ² Wooldridge test of autocorrelation in the panel has the null of no autocorrelation (with 1 lag). σiμ is the variance of individual effect (companies in this case). Robust estimates are estimated due to significant Heteroscedasticity and Autocorrelation. DV is roe (Performance). The higher value represents the greater performance. The sha is the explanatory variable and represents the shareholder’s activities. The l_sales is a log of sales and is a control variable. * Significance at 5%.

. The F-test and the B-P test exhibit significant p-value (>0.05 sig. level); therefore, we have performed the Hausman test. As per the Hausman test (p-value > 0.05), we have used the RE regression model (Arellano and Bond 1991; Baltagi 2008). The null of constant variance is rejected due to a significant p-value in the B-P test for Heteroskedasticity. Hence, robust estimates have arrived for the model (Baltagi 2008; Hsiao 2007), and the Wooldridge test shows no presence of autocorrelation.

In the present model, the coefficient of sha is negative (−0.3168) and insignificant (as the p-value is more than 0.05). It also indicates that the sha does not significantly influence the roe in Indian firms. The robust estimates also exhibit similar results. Hence, the finding rejects the assumed hypothesis H6.

4.8. Robustness and Endogeneity

Durbin Chi-square and Wu Hausman tests are performed to test the issue of endogeneity. The p-values of these tests are found significant for models 2, 3, and 4 at a significance level of 5% (

Table 9. Endogeneity test.

	Endogeneity Tests
Model	Durkin(score) Chi-Square Test	Wu Hausman Test
Model 1	1.2475 (0.2640)	1.2235 (0.2706)
Model 2	7.6257 * (0.0058)	7.8284 * (0.0059)
Model 3	11.1342 * (0.0008)	11.732 * (0.0008)
Model 4	4.0087 * (0.0453)	4.009 * (0.0472)
Model 5	0.8214 (0.3648)	0.8032 (0.3717)
Model 6	0.7937 (0.3730)	0.7759 (0.3799)

Note: Values in parentheses are p-values. * Is significant at 0.05.

). Therefore, the 2SLS regression model is adopted in models 2, 3, and 4 by introducing instrument variable institutional investors (ii).

4.9. Summary of the Results

Even the concept of Shareholder activism is very prominent Globally. The effects of the SHA are not similar around the globe. Literature is evident of the mixed empirical results. Shareholder activism is not always a practical means of corporate strategy practices. Sometimes it is detrimental from a corporate perspective (Fox and Lorsch 2012). Even though studies (Arfa et al. 2017; Croci et al. 2012; Fortin et al. 2014; Cronqvist and Fahlenbrach 2008; Cai and Walkling 2011) had presented a positive association of SHA towards firm performance. In contrast, studies (Holderness and Sheehan 1988; McConnell and Servaes 1990; Mehran 1995; Song and Szewczyk 2003; Guimaraes et al. 2018; Bouaziz et al. 2020) have empirically proven that shareholder activism does not hold any association in improving the valuation, efficiency or accounting profits of corporate firms. There is evidence of criticism over the costs of activism and its inefficiency (Hoffman et al. 2016; Fox and Lorsch 2012) in the literature.

To summarize the current study results, India is an emerging country with diverse corporate firms since activism is thriving. The promoters of the firms host the majority of the ownership control. Even shares are scattered in smaller stakes with institutional investors, facing collective action problems. The empirical results are insignificant for all the six models (proxies) performed for the study. Implies that shareholder activism has provided no value-added or tangible benefits for the targeted companies (sample of 37 firms). The obtained results show that SHA is effective and beneficial for many aspects, like the appointment/removal of the director from the board. However, for a firm’s financial performance, it is not found significant hence it has nothing to significantly with financial performance. The results align with the studies (Guimaraes et al. 2018; Bouaziz et al. 2020).

5. Discussion

The current paper has used six different performance measures as dependent variables; hence six different models have been developed. In all the models, a negative relationship has been found between performance and stockholders’ activity, but only in two cases (i.e., models 2 and 3), these associations are found significant. Existing literature on shareholder activism has different views on firm performance. In some instances, the positive impact (Boyson and Mooradian 2011; Brav et al. 2008; Prevost and Rao 2000) of activism has been discussed, whereas there are studies that show negative impact (Smith 1996; Wahal 1996). We also want to highlight that studies that even show no relationship (Karpoff et al. 1996; Del Guercio and Hawkins 1999) exists between activism and firm performance.

The primary reason for the insignificant negative relationship is that activism is still nascent, and promoters hold majority stakes in the listed Indian companies. On average, institutional investors hold ~33% stake in these companies and have little control in decision-making. Even holdings are dispersed across multiple institutional shareholders having lower stakes and thus facing collective action problems (Rock 1990). Lack of presence of activist hedge funds in India could be another reason. In the developed world, firm performance is positive when the activist shareholder is hedge funds (Del Guercio and Hawkins 1999; Cronqvist and Fahlenbrach 2008). Pension funds (Heard 1987) are other categories of investors which are associated with positive outcomes, but in India, the pension industry is very small to make a significant impact on companies. Also, as per Gillan and Stark (2000), shareholder activism could have varied impacts on firm performance over the period.

As per model 2 outcomes, there is a significant negative (Smith 1996; Wahal 1996) relationship between shareholder activism and valuation measured in market capitalization (m_cap). Also, model 3 shows that activism affects the operating performance measured in terms of operating margins (opm), whereas net profit margin (npm) has an insignificant relationship. Net profit margins are post-finance cost, whereas Opm is a profitability measure before finance cost; hence are not affected directly by the presence of debt which works as a monitoring mechanism (Jensen and Meckling 1976) and helps in agency cost reduction.

In emerging economies like India, listed firms are family-owned and also occupy CEO /MD positions which directly affect the opinion of directors (Saini and Singhania 2018). The literature shows that the market views shareholder activism as a relevant external control device. The stock price effects are driven by the target firm’s prior performance and governance quality (Saini and Singhania 2018; Renneboog and Szilagyi 2011). Firms receiving shareholders’ proposals tend to have better operating performance due to the pressure of proposals (Karpoff et al. 1996). Past study evidence show activism affects the governance structure, but performance measured in terms of operating profitability of valuation was insignificant to shareholder activism (Smith 1996). Our empirical results indicate that the relationship between shareholder activism and firms’ performance is not influential in India. The prime reason for insignificant results may be that firms strictly do not practice activism as it is still an emerging stage in India. Shareholder activism is not a short-term phenomenon. The effect of activism may be seen when investors start paying heed to activism in the upcoming years. Another reason may be that mutual fund investors are extensive investors. Also, they have regulatory requirements to participate in voting exercises and file the same with the regulator. Essentially, they do documentation work and do not get actively involved in management decision-making, and they flow with a management decision.

Shareholder activism fascinates extensive attention from researchers and policymakers (Denes et al. 2017). It helps improve the performance and valuation of the firms in developed nations (Del Guercio and Hawkins 1999; Cronqvist and Fahlenbrach 2008). The study practically will aid in the monitoring of the impact of shareholder activism on the firm’s valuation and financial performance using a variety of proxies. Investors, researchers, and policymakers will benefit from this research. The study reveals that governance reforms incentivize corporate enterprises to adopt better governance practices. This study contributes to the limited literature on shareholder activism in India, and the findings can pave a path for future research in the area (Shingade and Rastogi 2019). This is the first study that uses the sha index to interrogate the relationship between shareholder activism and firm performance. In emerging markets like India, investors need to have a different approach or regulatory framework to support minority shareholders. This study sets the base for further deep dive into the area. Current research is expected to ignite all-around thinking in the investor community and regulatory fraternity.

6. Conclusions

The current study intends to present the impact of shareholder activism on the performance of the targeted corporate firms. The study findings show an insignificant relationship between shareholder activism and firm performance parameters regarding valuation, profitability margins, and returns. This could be due to multiple factors, explicitly lower institutional stake in the targeted firms, lack of effective implementation of regulatory practices, and promising stage of activism in India. The study’s findings conclude that the targeted firms have received no value addition in profitability or valuation from shareholder activism. Results obtained from the study are similar to the existing studies (Holderness and Sheehan 1988; McConnell and Servaes 1990; Mehran 1995; Guimaraes et al. 2018; Bouaziz et al. 2020).

The present study is not free from the limitations; one of them is the lack of data from an activism point of view. We have seen instances of increased activism, but we had to restrict ourselves with the samples size of 37 firms due to the lack of availability of all the parameters across the four years. Another shortcoming is that we could trace a limited number of activism cases through filings, as in many cases, there would be direct investor–management communication (Becht et al. 2009).

We conducted the study over four years (FY2017–FY2020). Due to covid-19, the financial year 2020–2021 was impacted, and hence we could not use the latest data samples in our panel data analysis. As activism is currently in a nascent stage in India, we can test these models in different periods. A similar study can be conducted in countries in various stages of activism. Countries with a developed framework and other emerging economies have activism and allied regulations in place.