Shareholder activism and responses to
corporate abuse allegations
International Business and Human Rights: Evaluating Risk Factors and Outcomes of Corporate Human Rights Violations
Florian van Altena 11388102
Master Thesis MSc Business Administration: International Management
Supervisor: dr. M.K. Westermann-Behaylo 2016-2017
2
Statement of originality
This document is written by student Florian van Altena, who declares to take full responsibility for the contents of this document.
I declare that the text and work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3
Abstract
Increasing activities overseas of multinational enterprises cause growing concern for environmental and human rights abuse. Activist shareholders use proposals to express their discontent and signal management to take action. This research studied two models on how shareholder proposals can influence future corporate abuse allegations and how shareholder proposals can influence the remedy offered to such allegations. Financial slack of the firm is taken as a mediating variable in the first and second model, and degree of employee protection in the host country is taken as a mediating variable in the second model. A cross-sectional time-series regression analysis was conducted for the first model and found no relationship between shareholder proposals and the prevention of corporate abuse allegations, nor did financial slack have a mediating effect. A mediation style regression analysis was conducted for the second model and found a relationship between shareholder proposals and remedy offered, there was no mediating effect by employee protection. It was found that more shareholder proposals lead to more remedies, and specifically to more judicial and non-judicial remedies as the number of proposals increase.
Keywords: corporate abuse, shareholder activism, shareholder proposals, financial slack,
4
Acknowledgements
I express great thanks to my supervisor, dr. Michelle Westermann-Behaylo, for her support, patience and guidance to me while writing this research. Furthermore, I would like to thank my professors and my fellow MSc Business Administration students for their unconditional support and motivation.
5
Table of content
Introduction ... 6
Literature review ... 9
Corporate abuse allegations ... 9
Remedies ... 10
Shareholder proposals ... 10
Shareholder activism and proposals ... 11
Responses to shareholder activism ... 12
Financial slack ... 13
Degree of employee protection ... 14
Theoretical framework ... 15
Sample, data and methodology ... 19
Analytical Strategy ... 22 Results ... 24 Model 1 ... 24 Model 2 ... 25 Discussion ... 28 Limitations ... 30 Future research ... 31 Conclusion ... 32 References ... 33
6
Introduction
On April 20th, 2010, a key piece of safety equipment on a British Petroleum (BP) oil rig, Deepwater Horizon, failed to seal the drilling hole after a sudden surge of oil pressure (CSB, 2014). This caused a fire on the oil rig that burned for two days, killing and injuring multiple workers (CSB, 2014). Gas and oil flowed from the well for 78 days, causing one of the worst environmental disasters in the history of the United States (U.S.) (CSB, 2014). A year later, BP’s shareholders drafted a resolution (proposal) which called for an assessment of “principal, operational,
economic, environmental and social risks” and measurements to “reduce, manage and mitigate such risks” in future operations (Candor Solutions, 2010). BP’s shareholders came up with this
proposal after a series of health and safety violations by BP (Kropp, 2010).
BP’s shareholder proposal, focusing on the social aspects and consequences of the incident, is not unique in its existence. Another shareholder group from a mining company called Goldcorp Inc. submitted a proposal requesting an assessment of the impact on human rights of the firms’ activities in Guatemala after a series of incidents in 2009 (Dhir, 2012, p. 100). The human rights impact assessment report released by Goldcorp confirmed the violation of human rights of its employees and the indigenous population in Guatemala (Goldcorp, 2015). Goldcorp (2015) made ten commitments to ensure human rights would be respected in Guatemala. After publication of these commitments there have been no recorded human rights violations by Goldcorp in Guatemala.
The above examples of shareholder proposals on the subject of human rights and the environment are not exceptional, the amount of socially conscious shareholder advocacy in extractive industries have increased over the past years (Dhir, 2012, p. 100). With the expansion of overseas business operations of multinational enterprises (MNEs), concerns for human rights
7 and the environment have grown. Developing countries try to attract foreign direct investments (FDIs) from MNEs, often neglecting human rights and the environment by prioritising economic growth (Giuliani & Macchi, 2014, p. 479; Madies & Dethier, 2010). For example, the Nigerian government profited from the high revenues in the oil industry, ignoring the negative environmental and human rights consequences from the drilling operations in favour of these high revenues (Frynas, 1998). Countries provide incentives for MNEs to move their business, such as tax reductions, limited labour and/or environmental regulation, and end up in a regulatory race-to-the-bottom (Staiger & Sykes, 2011, p. 152).
Reactions from MNEs to shareholder proposals on these matters differ, due to various reasons. In the case of BP’s Deepwater Horizon incident, the board of BP approached the coalition of investors behind the proposal and urged them to withdraw, on the promise that BP would incorporate the lessons from this incident in its operations around the world (Kropp, 2011). The spill had cost BP and its shareholders three quarters of dividend, 10% of the R&D budget and forced BP to sell over 30 billion dollars in assets (Kropp, 2010). It is likely that this financial setback for BP made it less attractive to invest in safety measures, such as proposed by the shareholders. Goldcorp’s response to the shareholder proposal came after multiple incidents spread over years and repeated shareholder proposals (Dhir, 2012, p. 100), it is very likely that Guatemala’s relatively low worker protection influenced the response time of Goldcorp, as incidents in the host country may be less salient since they can be more common (OECD, 2013).
For many years after the oil spill, BP would have multiple lawsuits against them, most of them accusing BP of negligence and wilful misconduct (BHRRC, 2016). BP initially denied all allegations, accusing the manufacturer of the safety valve and the owner of the oil rig (which was rented by BP) of negligence (Guardian, 2012). BP later settled the lawsuit and was ordered to pay
8 damages to multiple organisations (BHRRC, 2016), there is no record of BP offering other remedies aside from the arrangements following from the lawsuits. Different factors influenced the decision of BP to settle, their financial situation was deteriorating due to the oil spill and several employees could be charged individually (Dhir, 2012; Guardian, 2012). It is very likely that the degree of employee protection and the financial situation of BP played a role in determining the remedy for the oil spill.
This thesis will further investigate the relationship between shareholder activism (in proposals), future corporate abuse allegations and the remedies proposed by MNEs. Mediating variables that can influence this relationship are the financial slack of an MNE and the degree of worker protection in the host country. In the following section, a literature review of these variables will be presented and their hypothesized relationship. In the theoretical framework, two models will be presented and the relationship between variables explained. In the data section, the analytical strategy to research these models is discussed and the results are presented. The discussion section relates the findings to previous literature and presents the limitations of this study and gives recommendations for future research. The conclusion gives a brief overview of the findings in this study.
9
Literature review
In recent years, the literature on shareholder activism has grown and investigated different topics (Goranova & Ryan, 2014). To further understand the relationship between shareholder activism in the form of proposals, corporate abuse allegations, remedies offered by the firm and the role of mediating variables such as financial slack and employee protection, an overview of relevant literature will be discussed in this section.
Corporate abuse allegations
States have the responsibility to protect human rights, but many fail to do so because they lack the resources, suffer from corruption or are dependent on a firm for their investments (Amnesty International, 2017). Incidents involving unethical behaviour on behalf of a corporation (by the corporation or actors on behalf of the corporation) are corporate abuse. Fraud, negligence or scandalous behaviour towards employees or local communities are such cases. As mentioned before, countries (both less developed and developed) try to create a comfortable environment for MNEs, by limiting governmental regulation concerning human rights or the environment. The impact of this can be positive; increased life standards, FDI in infrastructure and community, job creation and access to new technology can improve the lives of people in the area (Dolan, Johnstone-Louis, & Scott, 2012).
However, there are many instances where MNEs have exploited host country regulation and had a negative effect on the population and environment (Amnesty International, 2017). These incidents happen frequently in the extractive industry, where human rights are violated and the environment is affected negatively (Amnesty International, 2017). Local communities often face unwilling officials or ineffective legal systems when seeking compensation (Amnesty International, 2017). This research will investigate environmental and human rights violations in the extractive industry.
10 Remedies
There is limited literature on remedies offered by firms to allegations, there is however some literature on state responses to human rights or environmental allegations (Cohen, 1996). State responses to human rights or environmental allegations can be categorised in the same way as corporate responses, they can deny, counterclaim or claim the allegation (Cohen, 1996).
There are different forms in each type of response (Cohen, 1996, p. 521). In the case of denial, allegations can be literally denied (it did not happen), interpretively denied (something else happened) or implicationally denied (there is justification in what happened) (Cohen, 1996, pp. 522–534). There are different reasons to why firms choose a type of denial; ranging from internal information asymmetry or the behaviour of other firms related to the same allegation (Cohen, 1996, pp. 530–534). Firms can also place counterclaims to the allegation, arguing that allegations are not legitimate or that they have been complying with regulations during the time of the allegation (Cohen, 1996). Shell denied allegations of environmental pollution in Brazil, claiming that they have been complying with regulation and that the police who collected the data were corrupt (ICIS News, 2001). The firm can follow counterclaims with judicial action. The final response type is acknowledgement of the allegation (Cohen, 1996). This is clear in the case of Goldcorp where after collecting information about the allegation, they settled the dispute with non-judicial remedies (Goldcorp, 2015).
Shareholder proposals
The shareholder proposal is a tool for shareholders to compel management to have a shareholder vote on a subject the sponsor deems relevant (Dhir, 2012, p. 101). Proposals can be used as formal sign of discontent towards managers about firm activities or to reach and/or gain support from other stakeholders in the firm, such as employees, or to signal agencies or customers (David, Bloom, & Hillman, 2007, p. 92; DeCapo, 1988; Turban & Greening, 1997). Managers are
11 not obliged to implement shareholder proposals, but the pressure from shareholders has resulted in greater awareness among different actors and pressure campaigns (Hoyer & MacInnis, 1997). It can therefore be seen that shareholder activism in the form of proposals has a disciplining function where managers are pressured into changing their strategy (David et al., 2007, p. 92).
Shareholder activism and proposals
The literature on shareholder activism is quite extensive (David et al., 2007; Goranova & Ryan, 2014; O´Rourke, 2003), ranging from shareholder influence to management responses on activism. The following section will discuss the literature on shareholder activism.
Shareholder activism literature is mostly focused on two different types of activism, financial and social activism (Goranova & Ryan, 2014). Financial shareholder activists focus on maintaining or improving the shareholder value of the firm and shareholder-governance related issues such as shareholder rights (Goranova & Ryan, 2014, p. 1242). Social shareholder activists are more concerned with stakeholder issues, corporate social performance (David et al., 2007) and political engagement of the firm (Goranova & Ryan, 2014, p. 1242). Different activist identities, ranging from pension funds to labour union funds, often lean more towards financial or social subjects (David et al., 2007; Goranova & Ryan, 2014, p. 1242; Wood, 1991).
However, financial and social interests are in reality often combined by activists (O´Rourke, 2003). There has been a movement of linking social and environmental subjects to the financial performance of a firm, claiming that these have a direct influence on shareholder value (O´Rourke, 2003, p. 230). An example of this is a recent Apple shareholder proposal for more diversity in top management because shareholders believe that low levels of diversity are a business risk (Guynn, 2017). Therefore, in this research both a social and financial mediating variable are considered.
12
Responses to shareholder activism
David et al., (2007, p. 97) found a negative relationship between shareholder proposals and corporate social performance (CSP). They find that shareholder activism is more likely signal management instead of it having a coercive effect (David et al., 2007, p. 97). Resources that can be used to improve CSP are re-allocated to combat further government regulation and external pressures rather than be used to change firm policy (David et al., 2007, pp. 97–98). David et al., (2007, p. 98) and other researchers find that responses from managers are not substantive, but more symbolic and that no changes to essential policies are made (Mitchell, Wood, & Agle, 1997; Porter & Van der Linde, 1995; Westphal & Zajac, 1997).
Perrault & Clark (2016) look further into shareholder status and firm responsiveness and find that high status shareholders and shareholders with a reputation to threaten the firm receive a higher amount of responses from management. They show that shareholders with negative characteristics can also achieve results from their activism (Perrault & Clark, 2016). Perrault & Clark (2016) only look at environmental shareholder activism.
Dhir (2012) performed a case study research into the effects of the shareholder proposal mechanism in a Canadian extractive firm human rights violation case. He finds that the shareholder proposal mechanism is effective but only when certain conditions are met, the proposal must not undermine political goals of the government and the communities whose rights are violated should be included in the proposal (Dhir, 2012, pp. 111–112). However, it should be noted that there were external pressures forcing the firm to take action, such as the government, who was ready limit this firms’ activities in the country (Dhir, 2012, p. 104). Due to methodological limitations of a case study the results of this research cannot be generalized for the whole extractive industry, but it is
13 important to note that Dhir’s (2012) research provides valuable insight in the shareholder proposal mechanism and highlights the importance of external effects.
Financial slack
Literature on financial slack provides different definitions but agree on the idea that it is a resource (usually cash) ‘that exists in excess of some minimal level of foreseeable need’ (Bourgeois & Singh, 1983; Gulati & Nohria, 1996; Lewis, 2013, p. 8; March & Shapira, 1987; Mishina, Pollock, & Porac, 2004). Bourgeois & Singh (1983) introduce three kinds of slack resources: direct available slack, recoverable slack and potential slack, each differs in degree of difficulty to obtain. In line with Lee (2012, pp. 8–9), this research will focus on available slack which is the ‘easiest’ type of slack to obtain and most direct available. The most common calculation is the current ratio (current assets divided by current liabilities; the firm’s ability to pay short-term obligations) (Singh, 1986). To be able to best generalize the findings of this research this ratio will be used.
Lee (2012) researched the relationship between financial slack and the effect on the relationship between corporate governance and firm performance. He finds that there is a positive relationship between financial slack and firm performance. Bourgeois & Singh (1983, p. 44) found that management’s political behaviour is influenced by the potential gain of slack. Managers are politically more active when there is future potential slack to be acquired, this is because the long-term orientation of firms is of subject (Bourgeois & Singh, 1983, p. 46). On the other hand, available slack had no relationship with political behaviour (Bourgeois & Singh, 1983, p. 46). There is no literature expanding on financial slack and corporate abuse. But for this research the assumption will be made that positive financial slack influences firm performance positively.
14 Degree of employee protection
International labour standards differ across countries. The International Labour Organization (ILO) has adopted over 185 conventions covering labour standards related to but not limited to child labour, forced labour, freedom of association and more (ILO, 2014). The ILO sets, monitors and assists countries with these standards but does not have the authority to enforce these standards. Employee protection therefore differs in every country as governments’ effectiveness of enforcing these standards is not consistent in countries.
Literature on MNEs and human rights find evidence of a negative relationship between the two (Giuliani & Macchi, 2014, p. 485). Herrmann (1995) argues that globalisation is an unbalanced phenomenon in which MNEs cause inequality and legal injustice. Bilateral and multilateral mechanisms to ensure accountability or compliance with human rights and international environmental law fail and are further weakened by state-MNE relationship (Herrmann, 1995). Countries provide incentives for MNEs to move their business there, such as tax reductions, limited labour and/or environmental regulation, and end up in a regulatory race-to-the-bottom (Staiger & Sykes, 2011, p. 152).
This research seeks to answer the following two research questions:
1. To what extent does shareholder activism (in form of shareholder proposals) influence future corporate abuse allegations?
2. To what extent does shareholder activism (in form of shareholder proposals) influence the
remedy offered by the firm to corporate abuse allegations?
The research will furthermore consider the mediating effect of financial slack on the first research question, and the mediating effects of both financial slack and the degree of employee protection on the relationship in the second research question.
15
Theoretical framework
In the following section, the theoretical framework of both research questions will be discussed. Relationships between variables will be discussed and hypotheses presented. The section closes with a visual representation of the relationships in two models.
Model 1
In model 1, the relationship between shareholder activism (in form of proposals) and future corporate abuse allegations, is presented. The mediating factor that can influence this relationship is financial slack.
The relationship between shareholder proposals and future corporate abuse allegations finds fitting in the agency theory. Shareholders are the principals that monitor and create incentives for management (agents) for them to maximize shareholder value (Goranova & Ryan, 2014, p. 1241). By submitting shareholder proposals, the principals express their discontent with the agents’ strategy or performance and demand change (Becht, Franks, Mayer, & Rossi, 2010; Gantchev, 2013). The first hypothesis is as follows;
Hypothesis 1 (H1): a high number of submitted proposals leads to a lower likelihood of future
corporate abuse allegations.
The first hypothesis looks at the direct relationship between the independent and dependent variable from model 1.
Financial slack can influence this relationship, when a firm has more financial slack managers are more likely to improve corporate governance (Lee, 2012). It is assumed that in line with agency theory, shareholders provide managers with the right incentives to maximize shareholder value.
16
Hypothesis 1a (H1a): a relative high amount of financial slack in a firm positively affects the
relationship in H1 and reduces the likelihood of future corporate abuse allegations.
Model 2
In model 2, the relationship between shareholder activism (in form of proposals) and remedies offered by firms, is presented. Mediating variables that can influence this relationship are financial slack and the degree of employee protection in the host country.
Despite research showing that shareholder activism can have a positive influence on corporate policies (Dhir, 2012; Perrault & Clark, 2016), most research finds a negative relationship between shareholder proposals and changes in policy by management (David et al., 2007; Mitchell et al., 1997; Porter & Van der Linde, 1995; Westphal & Zajac, 1997). It is therefore expected that firm remedies to CAAs are less policy-based and lead more often to judicial action, since in light of shareholder proposals resources are used to combat regulation and external pressures (David et al., 2007, pp. 97–98). The second hypothesis is as follows;
Hypothesis 2 (H2): a high number of submitted proposals leads to a higher likelihood of judicial
remedies.
In model 2, it is also expected that financial slack has a mediating effect on this relationship, as managers are more likely to take action in favour to shareholder requests (Lee, 2012). It is therefore expected that financial slack increases the chance for non-judicial remedies.
Hypothesis 2a (H2a): a high degree of financial slack in a firm increases the chance for non-judicial
remedies and thus negatively affects the relationship in H2.
Due to the state-MNE relationship and regulatory incentives provided by states to attract MNE FDI, often a combination of limited labour protection laws and environmental laws, it will
17 be assumed that a low degree of employee protection positively influences the relationship in hypothesis 2 and increases the change of a judicial remedy since firms are more likely to be in a favourable legal position.
Hypothesis 2b (H2b): a low degree of employee protection in the host country positively affects
the relationship in H2 and increases the chance for a judicial remedy.
Hypothesis 2a and 2b consider mediating variables that could possibly influence the relationship described in hypothesis 2.
Model 1: Shareholder activism and corporate abuse allegations
Shareholder Activism (proposals) Future corporate abuse allegations Financial slack (current ratio)
18
Model 2: Shareholder activism and remedy
Shareholder Activism (proposals) Remedy Financial Slack (current ratio)
19
Sample, data and methodology
The sample population is based on publicly traded firms in the extractive industry (oil, gas and minerals) that are registered in the Corporations & Human Rights Database (CHRD) and thus have been involved in a company abuse allegation (CAA), and the ‘Standards & Poor’s 500 index’ over the period 2000-2013. This time frame is chosen to (a) obtain a large enough sample to create panel data and (b) to analyse data over time to determine company responses to shareholder proposals and remedies to allegations. A firm needs at least one CAA in this time frame and a response to the allegation. These criteria ensure firms in this study to be similar in size and degree of internationalisation, and were active in the period of 2000-2013.
Following the criteria above, a panel of 13 international company’s active in the extractive industry was identified. Data on the corporate abuse allegations and remedies were gathered from the CHRD. It should be noted that the sample size is relatively small compared to the number of firms in the CHRD, this is because shareholder information of the firms needs to be publicly accessible. The availability of shareholder proposal data and the involvement of extractive industry companies in corporate abuse provided a suitable context to investigate the phenomenon of shareholder activism.
Variables and measures
Dependent variable 1. Company abuse allegation (CAA) was measured as a case in which
an individual and/or group accuses a firm of human rights abuse (Olsen, 2014). There are several types of abuse registered in the CHRD; physical abuse, health abuse, environmental abuse, labour abuse, and development and poverty abuse. Certain allegations fall within multiple types of abuse, therefore some allegations counted as more than one allegation. For both models, no distinction
20 was made between different types of abuse. There can be multiple CAAs in one year per firm. Data on te CAAs was gathered from the CHRD.
Dependent variable 2. Remedy was measured as the response/solution offered by the firm
to the injured party of an allegation or to the allegation and was classified as judicial, non-judicial, both judicial and non-judicial or no response. To identify if a firm had claimed, denied or counterclaimed the CAA, it was monitored if firms had, respectively, offered a non-judicial remedy, had no response or replied with judicial action. Data about company responses was gathered from the CHRD, where responses were classified into the mentioned categories.
Independent and mediating variables. Shareholder activism (in form of proposals) was
measured as shareholder proposals submitted to the board of a company (obtained from the Institutional Shareholder Services (ISS) database (formerly RiskMetrics)). Shareholder proposal data from 2000 to 2013 was used to match the data from the CHRD. Shareholder proposals did not necessarily have to be accepted by the board to have a positive effect on the relationship, proposals can also be informative or disciplining (David et al., 2007, p. 92). Therefore, all socially responsible investment proposals that were withdrawn, voted on or omitted were considered to influence the relationship. For model 1, a panel data index was created where data from the dependent variable and independent variable are placed so that a time-related conclusion can be made. For model 2, data was listed according to CAAs, this will later be further explained.
Mediator variable 1. Financial slack. In this study, it has been argued that a relevant aspect
of shareholder proposal success and the choice of remedy was the financial performance of a company (Lee, 2012). Financial slack was measured as the yearly current ratio of a firm. The current ratio was calculated as follows:
𝑐𝑢𝑟𝑟𝑒𝑛𝑡 𝑟𝑎𝑡𝑖𝑜 = 𝑐𝑢𝑟𝑟𝑒𝑛𝑡 𝑎𝑠𝑠𝑒𝑡𝑠 𝑐𝑢𝑟𝑟𝑒𝑛𝑡 𝑙𝑖𝑎𝑏𝑖𝑙𝑖𝑡𝑖𝑒𝑠
21 Current ratio was chosen as a measure because this is the most widely known measure of financial slack (Lee, 2012). Return On Assets (ROA) was a possible alternative for measuring the financial performance of a firm, but it does not properly reflect the financial slack a firm has (Lee, 2012, p. 5). Data for financial slack was gathered from the OSIRIS database offered by Bureau Van Dijk, this database contained extensive financial information from multinational enterprises, including firms in the extractive industries.
Mediator variable 2. Degree of employee protection. In this study, it has been argued that
employee protection was expected to influence firm response to shareholder proposals and remedies to allegations (Staiger & Sykes, 2011). The degree of employee protection in the host country (the country where the allegation occurred) was measured as the amount of legal protection employees in the private sector had. Data was gathered from the Employment Protection Database from the Organisation for Economic Co-operation and Development (OECD).
22
Analytical Strategy
The study is based on two different models to analyse the effect of shareholder proposals on CAAs: the relationship between shareholder proposals and future corporate abuse allegations (model 1), and the relationship between shareholder proposals and the remedy offered to a CAA (model 2). Both models have mediating hypotheses, namely current ratio for model 1 and current ratio and the degree of employee protection for model 2.
To test the main hypothesis for model 1, a panel data regression analysis, with a time-series (year) variable, was performed so that the influence of variables over time could be observed. This resulted in 182 yearly observations on 13 firms for 14 years, the data was balanced, meaning that there are measurements for all time periods. A Hausman specification test was conducted to determine if the data was a random or fixed effects model, results indicated that the data was a random effects model (Park, 2011, p. 10). The random effects model allows to generalize the results beyond the sample used in this model, as individual effects do not influence the predictor variables (Park, 2011, p. 10). Panel studies are by nature longitudinal studies where the unit of analysis is followed over time (Bryman, 2012, p. 63). Panel data allowed to capture the complexity of the dynamic relationship over time, thus allowing to make stronger conclusions about the dynamics of a relationship (Bryman, 2012, p. 63; Hsiao, 2007). By having aggregated data, predictions are stronger than studies where predictions are based on individual cases (Bryman, 2012, p. 63).
To test the mediation hypothesis, a Baron & Kenny (1986) style mediation analysis was conducted to assess if financial slack mediated the relationship between shareholder proposals and CAAs.
To test the main hypothesis for model 2, a new subsample dataset from the dataset of model 1 was created to conduct a multinomial logistic regression analysis to predict the multiple categories of the outcome variable remedy (Field, 2013). Furthermore, the variable shareholder
23 proposals was recoded into three groups that are representative of the sample, respectively zero proposals, one proposal or two or more proposals submitted, this was done to ensure the reliability of the variable. A Chi-square test of association was performed to determine the relationship between the independent and dependent variable, before conducting the regression analysis. A Fisher’s Exact test (Field, 2013, p. 876; Fisher, 1922) was conducted to increase the robustness of the Chi-square test and to combat the limitation of expected frequencies below 5 due to the relatively small sample size (Field, 2013, p. 876).
To test the mediation hypotheses, a Baron & Kenny (1986) style mediation analysis was used to assess if financial slack and degree of employee protection mediated the relationship between shareholder proposals and remedy.
24
Results
Model 1
Table 1 displays the descriptive statistics and correlations of the variables from model 1. It shows that CAA was negatively correlated with financial slack (-.21, p < .01). Table 1 also shows that shareholder proposals was negatively correlated with financial slack (-.17, p < .05). There is no indication that CAA and shareholder proposals are correlated. The Hausman specification test indicated that the data is a random effects model (.60, p > .05). The standard deviation (SD) values in parentheses are the SD values calculated over time regardless of firm.
Table 1: Means, Standard Deviations, Correlations for model 1
Variables M SD 1 2 3
1. CAA ,41 ,74 (,60) -
2. Shareholder proposals 1,96 2,78 (1,48) 0,116 -
3. Financial slack 1,61 ,85 (,61) -,211** -,171* -
Note. Hausman specification test = 0.6006. Numbers in parentheses, (), are SD values for within statistics (time = 14).
**. Correlation is significant at the 0.01 level (2-tailed). *. Correlation is significant at the 0.05 level (2-tailed).
Table 2 shows the results of the mediation regression analysis, including time as a mediator. As can be seen in table 2, shareholder proposals significantly predict financial slack, b = -,053, t = -2,33, p = ,021. The results for time with different predicters is presented as well, however these are not relevant for this research and therefore will not be discussed. The analysis shows that CAAs are not significantly predicted by shareholder proposals, b = ,021, t = 1,07, p = ,285, and time, b = ,004, t =,32, p = ,749. Financial slack is a significant predictor of CAA, b = -,174, t = -2,68, p = ,008.
25 Table 2: Results of the mediation regression analysis
Consequent
FS (M) Time (M) CAA
Antecedent Coeff. SE p Coeff. SE p Coeff. SE p
SHP (X) a1 -,053 ,023 ,021 a2 ,138 ,110 ,210 c1' ,021 ,020 ,285 FS (M) - - - a3 ,480 ,358 ,182 b1 -,174 ,065 ,008 Time (M) - - - - b2 ,004 ,014 ,749 Constant im1 1,714 ,076 <,001 im2 6,458 ,714 <,001 iv ,613 ,156 <,001 R2 = ,0294 R2 = ,0159 R2 = ,0517 F(1,180) = 5,4489, p<,05 F(2,179) = 1,4419, p>,05 F(3,178) = 3,2377, p<,05
Table 3 displays the direct, total and indirect effect results. Both the direct and total effect have a p > ,05 and are not significant predictors of CAA. The bias corrected confidence interval for the indirect effect does indicate a significant relationship as it does not contain zero (BC95 = [0,0031, 0,0204]). There is an indication of a mediation effect, however, there is no indication of such an effect looking at the total effect.
Table 3: Direct, Total and Indirect effect results
Effect SE p Direct effect c1' ,021 ,020 >,05 Total effect c1' ,031 ,020 >,05
Boot SE Boot LLCI Boot ULCI Indirect effect a1b1b2 ,010 ,004 0,0031 0,0204
Model 2
Table 4 displays the descriptive statistics and correlations of the variables from model 2, and table 5 presents the results of the Chi-square test and the descriptive statistics for the independent and dependent variable. Table 4 shows that remedy was positively correlated with shareholder proposals (.29, p < .01), remedy was also positively correlated to degree of employee protection (.35, p < 0.05). Financial slack was not a statistically significant predicter of remedy in this analysis.
26 Table 4: Means, Standard Deviations, Correlations for model 2
Variables M SD 1 2 3 4
1. Remedy ,84 ,99 -
2. Shareholder proposals 1,00 0,86 ,299** -
3. Financial slack 1,27 ,45 0,03 0,01 -
4. Degree of employee protection 1,85 0,66 ,35** 0,09 -0,1 - **. Correlation is significant at the 0.01 level (2-tailed).
*. Correlation is significant at the 0.05 level (2-tailed).
The Chi-square test of association results indicated that the number of shareholder proposals is statistically associated with the type of remedy, χ2 (df = 6, n = 75) = 13.326. However, the Chi-square assumption that all expected frequencies are a minimum of five or higher was violated, so a Fisher’s Exact test result was conducted to provide more accurate results. The Fisher Exact test allows for the expected frequencies to be less than five (Fisher, 1922). The Fisher Exact test also indicates a statistically associated relationship (.045, p < 0.05). A firm with no socially responsible investment shareholder proposals is more likely to have no remedy for a CAA, one proposal is more likely to have a judicial remedy, and 2 or more proposals is more likely to have a non-judicial remedy.
Table 5: Results of Chi-square Test and Descriptive Statistics for Shareholder Proposals by Remedy Remedy
None Jud Non-jud Both
Shareholder proposals No proposals Freq. 17(12,3) 5(6,3) 2(4,3) 1(2,0) 1 proposal Freq. 9 (9,9) 8(5,1) 1(3,5) 2(1,6) 2 or more proposals Freq. 11(14,8) 6(7,6) 10(5,2) 3(2,4) Total Freq. 37(37,0) 19(19,0) 13(13,0) 6(6,0)
Note. χ2 = 13.326*, df = 6, Fisher’s Exact test = 0.045*. Numbers in parentheses, (), are expected frequencies. Freq. = frequency.
*p<.05
Table 6 shows the results of the mediation regression analysis. Due to missing data 17 cases were dropped, the n = 58. The analysis indicated that shareholder proposals does not significantly
27 predict the degree of employee protection, for .506, p > 0.05, t = 0,67. The coefficient for a1 is also
very close to zero, indicating that difference in one unit of shareholder proposals is a difference of 0,070 in degree of employee protection. The R2 indicated that shareholder proposals explained 0,8% of the variance in degree of employee protection. The analysis further shows the results of remedy predicted by both shareholder proposals and degree of employee protection. It indicates that shareholder proposals significantly predict remedy, even with degree of employee protection in the model, b = 0,314, t = 2,12, p = 0,039; degree of employee protection also significantly predicts remedy, b = 0,495, t = 2,70, p = 0,009. The R2 indicated that the model explains 19% of the variance in remedy. The positive b for shareholder proposals and degree of employee protection indicate that when either increases, remedy increases as well. However, the bias corrected confidence interval in table 7 does not indicate a significant relationship as it contains zero (BC95 = [ -0,085, 0,2205]). There is no mediation effect.
Table 6: Results of the mediation regression analysis
Consequent
Emp (M) Remedy
Antecedent Coeff. SE p Coeff. SE p
SHP (X) a1 ,070 ,107 ,506 c1' ,314 ,148 ,039
Emp (M) - - - b1 ,495 ,183 ,009
Constant i1 1,791 ,126 <,001 i2 -,389 ,370 >,05
R2 = ,0079 R2 = ,1904
F(1,56) = ,4477, p > ,05 F(2,55) = ,003, p < ,05
Table 7: Direct, Total and Indirect effect results Effect SE p
Direct effect c1' ,314 ,148 <,05
Total effect c1' ,349 ,156 <,05
Boot SE Boot LLCI Boot ULCI Indirect effect a1b1 ,036 ,075 -0,085 0,2205
28
Discussion
This research sought to answer to what extent shareholders could influence the number of corporate abuse allegations a firm is involved in and to what extent they could influence the type of remedy offered to such an allegation. For model 1, there were two different hypotheses. In hypothesis 1, it was expected that a high number of shareholder proposals would reduce the likelihood of a firm being involved in a CAA. In hypothesis 1a, it was expected that a high ratio of financial slack would further reduce the likelihood of a firm being involved in an CAA.
The results indicate that there is no significant relationship between shareholder proposals and the number of CAAs a firm is involved in. Table 3 shows that the total effect is slightly larger than the direct effect, however, the level if significance is above ,05. Previous research has showed that managements’ responsiveness to shareholder proposals is often symbolic (David et al., 2007). The results of this research indicate that shareholders have not steered managerial resources towards preventing CAAs. The findings are in line with previous literature stating that there is a negative relationship between shareholder activism and corporate social performance (David et al., 2007; Mitchell et al., 1997; Porter & Van der Linde, 1995; Westphal & Zajac, 1997).
On the other hand, Perrault & Clark (2016) have stated that external factors such as shareholder status play an important role in management response to shareholder proposals. Dhir (2012) also stressed the importance of external factors in management response, such as government action against firms. A limitation of this study was the aggregation of all socially responsible shareholder proposals. While all proposals in this study were socially responsible, they did not all have the same intentions, as some proposals may be oriented on past abuse allegations, while others focused on taking preventative measures.
29 Financial slack, as an external factor, does not provide a mediating effect for the relationship between shareholder proposals and CAAs. However, the direct relationship between financial slack and CAAs has a significance level below ,05. With every unit growth in financial slack, there is a slight decline in number of CAAs. This is in line with Lee’s (2012) findings on financial slack and corporate performance, however, Lee’s research is not related to corporate abuse. More research into the relationship between financial slack and corporate abuse is suggested as this research showed a significant relationship between the two variables. There is no support for hypothesis H1 and H1a.
Model 2 considered shareholder proposal effects onto remedy offered and it considered the mediating effects of financial slack and the degree of employee protection. It was hypothesized that (H2) a high number of shareholder proposals would lead to more judicial remedies, and that (H2a) a high financial slack would negatively mediate this relationship leading to more non-judicial remedies and that (H2b) a low degree of employee protection would positively influence this relationship leading to more judicial remedies.
The results presented in table 5 indicate that there is a relationship between the number of shareholder proposals submitted and the remedy offered. Firms without socially responsible shareholder proposals are more likely to have no remedy for a CAA, there is no reason for managers to take action as shareholders have not shown their discontent. When there is one proposal, managers are more likely to issue judicial remedies. When there are two or more proposals, managers are more likely to issue a non-judicial remedy or have both. Hypothesis 2 finds support in these findings, as proposals do lead to more judicial remedies but also to more non-judicial remedies as the number of proposals increase. Table 6 shows a significant relationship between shareholder proposals and remedy, with a relatively high level of significance and a coefficient
30 showing that every increase in shareholder proposals increases towards a judicial and/or non-judicial remedy with ,314. The results reflect findings in previous literature, proposals are used by shareholders as a formal sign of disgruntlement and continued use is a sign of a pressure campaign to force management into action (DeCapo, 1988; Hoyer & MacInnis, 1997; Turban & Greening, 1997).
The expected mediation effects of financial slack and degree of employee protection did not take place. Financial slack did not show a statistically significant relationship with any of the variables and degree of employee protection did not cause a statistically significant mediation on shareholder proposals and remedy. The findings for financial slack are similar to Bourgeois & Singh’s (1983, p. 46), available slack has little influence on management action towards political matters. As presented in table 6, the direct effect of degree of employee protection does show a statistically significant relationship with remedy, a high degree of protection does increase the likelihood of a judicial, non-judicial remedy or both. The coefficient is relatively large, indicating a relatively quick rise on the slope for remedy. These findings are similar to the arguments made by Herrmann (1995) and Staiger & Sykes (2011), governments offer MNEs favourable regulation allowing them to avoid judicial consequences. In countries with a higher degree of employee protection it is possible that the judicial action MNEs face are initiated by them employees and less by the firm. There is support for hypothesis H2, but no support for hypothesis H2a and H2b.
Limitations
A limitation of this study was the accessibility of shareholder proposal data. While the database on CAAs contained a large number of companies, of only a few the shareholder proposal data was available online. This reduced the number of the sample to 13 firms. To be able to generalise the findings as much as possible certain measures were taken during the data analysis.
31 For model 1, random effects modelling was applied to reduce the disturbance of individual effects on the outcome. For model 2, a Fisher’s Exact test was conducted to compute the exact probability of the chi-square even with a small sample size. However, even with these measures, it should still be taken into account that this study was conducted with a relatively small sample.
Future research
This research has showed a relationship between shareholder proposals and the type of remedy offered by firms. Future research could be directed towards finding factors that influence this relationship, Perrault & Clark’s (2016) shareholder status or issue salience (as certain allegations may get more media attention) are factors that can be researched, as their research is limited to environmental shareholder activism. Research questions that can be asked are ‘to what extent do highly salient allegations get more non-judicial remedies?’ or ‘how does shareholder status influence the type of remedy offered?’.
While it has proved to be difficult and time-consuming, further research should focus on acquiring a larger sample of firms in the extractive industry and more detailed information on shareholder proposals. By conducting a case-study, more insight can be gained into management response towards different types of shareholder proposals and the possible symbolic nature behind responses and proposals. With more research into environmental and/or labour-oriented proposals a better understanding of the influence shareholders have on management is gained.
32
Conclusion
Shareholder proposals have no influence on preventing firms being involved in company abuse allegations, but they can influence the type of remedy offered to an allegation. Shareholder proposals lead towards a remedy for allegations, while no proposals lead to no remedies offered. The more shareholder proposals submitted, the higher the chance for a non-judicial remedy over a judicial remedy and a higher change for both remedy types. While it was expected that a firms’ financial slack would influence the involvement into allegations and the type of remedy offered, no such relationship was found. Employee protection in the host country, contrary to what was expected, also does not provide a mediation effect on the type of remedy offered. However, a direct relation between employee protection and remedy was found.
These results contribute to the previous literature on shareholder activism and corporate abuse by further mapping the power of shareholders and the use of proposals as a tool for their power. It has been identified that shareholder proposals can be used to pressure management into taking action towards allegations, but it has limited reach as future allegations cannot be prevented. Furthermore, it was found that one internal factor, financial slack, and one external factor, degree of employee protection, have no influence on future allegations or the remedies offered. For future research other, internal or external factors, should be researched to determine how and what influences management to respond to shareholder activism.
33
References
Amnesty International. (2017). Corporations: Corporate Accountability. Retrieved March 16, 2017, from https://www.amnesty.org/en/what-we-do/corporate-accountability/
Baron, R. M., & Kenny, D. a. (1986). The Moderator-Mediator Variable Distinction in Social The Moderator-Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic, and Statistical Considerations. Journal of Personality and Social Psychology, 51(6), 1173–1182. https://doi.org/10.1037/0022-3514.51.6.1173
Becht, M., Franks, J., Mayer, C., & Rossi, S. (2010). Returns to shareholder activism: Evidence from a clinical study of the hermes UK focus fund. Review of Financial Studies, 23(3), 3093– 3129. https://doi.org/10.1093/rfs/hhn054.ra
BHRRC. (2016). US Deepwater Horizon explosion & oil spill lawsuits. Retrieved June 29, 2017, from https://business-humanrights.org/en/us-deepwater-horizon-explosion-oil-spill-lawsuits
Bourgeois, L. J., & Singh, J. V. (1983). Organizational Slack and Political Behavior Among Top Management Teams. Academy of Management Proceedings, 8(1), 43–47. https://doi.org/10.5465/AMBPP.1983.4976315
Bryman, A. (2012). Social Research Methods (4th ed.). Oxford: University Press.
Candor Solutions. (2010). BP facing shareholder resolution on oil spill. Retrieved March 11, 2017, from http://www.candorsolutions.co.za/bp-facing-shareholder-resolution-on-oil-spill/
Cohen, S. (1996). Government Responses to Human Rights Reports: Claims, Denials, and Counterclaims. Human Rights Quarterly, 18(3), 517–543. https://doi.org/10.1353/hrq.1996.0028
34 CSB. (2014). Deepwater Horizon Blowout Animation. U.S. Chemical Safety Board. Retrieved
from http://www.csb.gov/videos/deepwater-horizon-blowout-animation/
David, P., Bloom, M., & Hillman, A. J. (2007). Investor Activism, Managerial Responsiveness, and Corporate Social Performance. Strategic Management Journal, 28, 91–100. https://doi.org/10.1002/smj
DeCapo, T. A. (1988). Challenging Objectionable Animal Treatment with the Shareholder Proxy Proposal Rule. University of Illinois Law Review, 119–138. https://doi.org/10.1525/sp.2007.54.1.23.
Dhir, A. A. (2012). Shareholder Engagement in the Embedded Business Corporation: Investment Activism, Human Rights, and TWAIL Discourse. Business Ethics Quarterly, 22(1), 99–118. https://doi.org/10.2139/ssrn.1416198
Dolan, C., Johnstone-Louis, M., & Scott, L. (2012). Shampoo, saris and SIM cards: seeking entrepreneurial futures at the bottom of the pyramid. Gender & Development, 20(1), 33–47. https://doi.org/10.1080/13552074.2012.663619
Field, A. (2013). Andy Field - Discovering Statistics Using SPSS: and sex and drugs and rock “n”
roll. SAGE Publications (4th editio). SAGE. https://doi.org/10.1111/j.1365-2648.2007.04270_1.x
Fisher, R. A. (1922). On the Interpretation of Chi-square from Contingency Tables, and the Calculation of P. Journal of the Royal Statistical Society, 85(1), 87–94. https://doi.org/10.2307/2340521
Frynas, J. G. (1998). Political instability and business: Focus on Shell in Nigeria. Third World
35 Gantchev, N. (2013). The costs of shareholder activism: Evidence from a sequential decision model. Journal of Financial Economics, 107(3), 610–631. https://doi.org/10.1016/j.jfineco.2012.09.007
Giuliani, E., & Macchi, C. (2014). Multinational corporations’ economic and human rights impacts on developing countries: A review and research agenda. Cambridge Journal of Economics,
38(2), 479–517. https://doi.org/10.1093/cje/bet060
Goldcorp. (2015). Goldcorp Human Rights Policy, (July), 1–3.
Goranova, M., & Ryan, L. V. (2014). Shareholder Activism: A Multidisciplinary Review. Journal
of Management (Vol. 40). https://doi.org/10.1177/0149206313515519
Guardian, T. (2012, September 5). Deepwater Horizon: US ramps up rhetoric on BP over oil spill.
The Guardian. Retrieved from
https://www.theguardian.com/business/2012/sep/05/deepwater-horizon-us-bp-oil-spill
Gulati, R., & Nohria, N. (1996). Is Slack Good or Bad for Innovation? The Academy of
Management Journal, 39(5), 1245–1264.
Guynn, J. (2017). Apple shareholder proposal seeks more diversity at top. Retrieved March 16, 2017, from http://www.usatoday.com/story/tech/news/2017/02/01/apple-shareholders-push-diversity-executives-board-members/97375208/
Herrmann, P. (1995). Human environmental crisis and the transnational corporation: The question of culpability. Human Ecology, 23(2), 285–289. https://doi.org/10.1007/BF01191653
Hoyer, W. D., & MacInnis, D. J. (1997). Consumer Behavior. Boston, MA: Houghton Mifflin.
36
Investigación Operativa, 0(0), 1–63. https://doi.org/10.1007/s11749-007-0046-x
ICIS News. (2001). Shell Chemical Brazil denies revived environmental claims. Retrieved June 27, 2017, from https://www.icis.com/resources/news/2001/08/01/144090/shell-chemical-brazil-denies-revived-environmental-claims/
ILO. (2014). Guide to International Labour Standards. Turin, Italy.
Kropp, R. (2010). Shareowner Resolution Addresses Risk Management at BP. Retrieved March 11, 2017, from http://www.socialfunds.com/news/article.cgi/article3106.html
Kropp, R. (2011). Investor Coalition Withdraws Resolution at BP. Retrieved March 12, 2017, from http://www.socialfunds.com/news/article.cgi/article3134.html
Lee, S. (2012). Corporate Governance, Financial Slack and Firm Performance: A Comparative Study between US and UK. Seoul Journal of Business, 18(1), 3–23.
Lewis, T. (2013). A Review and Analysis of the Effects of Financial Slack on Firm Innovation.
University of Wisconsin-Milwaukee, (August).
Madies, T., & Dethier, J. J. (2010). Fiscal competition in developing countries : a survey of the theoretical and empirical literature. The World Bank, Policy Research Working Paper Series:
5311, 2010, (May), 2–33. https://doi.org/10.1142/S1793993312500135
March, J. G., & Shapira, Z. (1987). Managerial Perspectives on Risk and Risk Taking. Management
Science, 33(11), 1404–1418.
Mishina, Y., Pollock, T. G., & Porac, J. F. (2004). Are more resources always better for growth? Resource stickiness in market and product expansion. Strategic Management Journal, 25(12), 1179–1197. https://doi.org/10.1002/smj.424
37 Mitchell, R. K., Wood, D. J., & Agle, B. (1997). Toward a Theory of Stakeholder Identification and Salience : Defining the Principle of Who and What Really Counts. Academy of
Management Review, 22(4), 853–886. https://doi.org/10.5465/AMR.1997.9711022105
O´Rourke, A. (2003). A new politics of engagement: Sahreholder activism for corporate social responsibility. Business Strategy and the Environment, 50(1), 6–27. https://doi.org/10.1002/bse.364
OECD. (2013). OECD Indicators of Employment Protection. The Organisation for Economic
Co-operation and Development. Retrieved from
http://www.oecd.org/employment/emp/oecdindicatorsofemploymentprotection.htm
Olsen, T. D. (2014). CAA CODING GUIDE CORPORATIONS & HUMAN RIGHTS DATABASE PROJECT. Corporations & Human Rights Database.
Park, H. M. (2011). Practical Guides To Panel Data Modeling : A Step by Step Analysis Using
Stata. Public Management & Policy Analysis.
Perrault, E., & Clark, C. (2016). Environmental Shareholder Activism: Considering Status and Reputation in Firm Responsiveness. Organization & Environment, 29(2), 194–211. https://doi.org/10.1177/1086026615571939
Porter, M. E., & Van der Linde, C. (1995). Green and Competitive: Ending the Stalemate. Harvard
Business Review, 73(5), 120–134.
Singh, J. V. (1986). Performance , Slack , and Risk Taking in Organizational Decision Making.
The Academy of Management Journal, 29(3), 562–585.
38 Regulation. The Journal of Legal Studies, 40(1), 149–203.
Turban, D. B., & Greening, D. W. (1997). Corporate Social Performance and Organizational Attractiveness to Prospective Employees Author ( s ): Daniel B . Turban and Daniel W . Greening Source : The Academy of Management Journal , Vol . 40 , No . 3 ( Jun ., 1997 ), pp . 658-672 Published by : Aca. The Academy of Management Journal, 40(3), 658–672.
Westphal, J. D., & Zajac, E. J. (1997). Substance and Symbolism in CEOs Long-Term Incentive Plans. Administrative Science Quarterly, 39(3), 367–390.
Wood, D. J. (1991). Corporate Social Performance Revisited. The Academy of Management
