1
The Effects of Contestation on Investment in Foreign Assets
Master Track Strategy Thesis
Student: Pascal Marcus Reinkingh Student №: 11819790
Supervisor: Panikos Georgallis
Statement of Originality
This document is written by Student Pascal Marcus Reinkingh who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document are 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
Inhoudsopgave
Abstract... 4
This study examines the relationship between contestation and investment in foreign assets. A model is proposed that conceptualizes three types of contestation, scrutiny, event-stigma and core-stigma. These concepts are operationalized in three independent variables: Media Scrutiny, De-individuation based on contested industry participation and De-individuation based on negative events. They are all hypothesized to have a positive effect on investment in foreign assets. After analysis, only the positive effect of media scrutiny was confirmed. The study adds to the literature by assessing an underdeveloped strategy in response to contestation. ... 4
Introduction ... 4
Literature Review ... 6
Contested Industries ... 7
Contestation in Non-Contested Industries ... 9
Acquiring Foreign Assets ... 11
Research Design ... 15
Independent Variable: Media Scrutiny ... 16
Independent Variable: Core-Stigmatization... 17
Independent Variable: Event-Stigmatization ... 18
Dependent Variable: Investment in Foreign Assets... 20
Control Variables ... 20
Descriptive Statistics ... 21
Methods & Results ... 23
Hierarchical Regression ... 23
Discussion & Conclusions ... 26
Findings and Theoretical Implications ... 27
Practical implications... 28
Limitations ... 29
Suggestions for Further Research ... 29
References ... 31
Abstract
This study examines the relationship between contestation and investment in foreign assets. A
model is proposed that conceptualizes three types of contestation, scrutiny, event-stigma and core-stigma. These concepts are operationalized in three independent variables: Media Scrutiny, De-indi-viduation based on contested industry participation and De-indiDe-indi-viduation based on negative events. They are all hypothesized to have a positive effect on investment in foreign assets. After analysis, only the positive effect of media scrutiny was confirmed. The study adds to the literature by as-sessing an underdeveloped strategy in response to contestation.
Keywords: Contestation, Scrutiny, De-individuation, Stigma
Introduction
Within organizational research, institutional theory has become a widely discussed topic. Within this discussion, a strong emphasis is laid on the construct of legitimacy (Vergne, 2011). Legitimate or-ganizations are oror-ganizations whose values, strategies and behavior are in line with socially accepted standards (Vergne, 2011). Companies that successfully strive for and obtain legitimacy are rewarded. Scholars have argued that legitimacy enhances organizational survival (Deephouse & Suchman, 2008) and that it can be used to pursue organizational goals (Suchman, 1995).
Despite the fact the legitimacy brings about many upsides, it does not seem to be a necessary require-ment for organizational survival. One can think of several organizations that survive and even thrive even though their values, strategies and behavior are not congruent with socially accepted standards. Some examples are organizations in the tobacco industry, the arms industry or the gambling industry. These types of organizations are examples of organizations that face contestation. They are defined by Durant and Vergne as experiencing “settings characterized by social contestation and targeted scrutiny by hostile audiences” (Durant & Vergne, 2015).
It would seem hard for organizations to survive under these circumstances. Yet companies seem able to strategize against and successfully fight contestation. How organizations are able to do this, is a question that research has not been able to find a complete answer to. Several strategies to counter contestation have been proposed and worked out by scholars (e.g. Roulet 2015, Durant & Vergne 2015). Roulet gives a broad sense of strategizing and lays out three distinct ways of fighting against contestation. The first is distancing. In this instance, a firm is trying to distance itself from a stigma and claim that the stigma is wrongly placed upon them (Roulet, 2015). The second strategy is to
5 defend the underlying values that are under attack and to try to defend the activities that have caused contestation (Roulet, 2015). The last option that Roulet mentions is to simply accept the stigma. In spite of contestation pressing down upon them, firms seem to sometimes be able to survive without taking any action to disprove or distance themselves from it (Roulet, 2015).
Research in this last category has given several suggestions of how an organization that accepts con-testation tries to survive (e.g. Palazzo & Richter, 2005, Hsu & Grodal, 2015, Simons, Vermeulen & Knoben, 2016). All of these scholars have focused on strategies that are executed within the same geographic area as where it is facing contestation. Organizations may, however, decide to invest into areas where contestation is of a lesser concern. This type of strategizing has been overlooked by scholars.
This paper will attempt to show a positive relationship between contestation and investing into for-eign countries. It will define three types of contested organizations, operating either in contested or non-contested industries. The relationship between these types and foreign investments will then be assessed to create a broader knowledge on strategizing against contestation. We now know that there are different strategic paths that firms may undertake in coping with contestation. We do not know if acquiring assets overseas is one of these paths. We have theorized that it could be a possible strategy to face contestation. We do not, however, know if firms are currently using this strategy and if dif-ferent degrees of contestation will result in a higher usage of this strategy.
In order to expand our knowledge about this type of strategizing I will operationalize three types of contestation: scrutiny, event-stigmatization and core-stigmatization. Each of these are expected to have a different effect on investing in foreign assets. The independent variables used to operationalize these groups are media scrutiny and de-individuation. As a dependent variable, foreign assets com-pared to total assets in 2015 will be used to assess the foreign investment strategies over the course of a longer time period.
The result section will show that scrutiny does, in fact, have an effect on investment in foreign assets. Undergoing a stigma, however, does not seem to result in a higher percentage of foreign assets. In
the last section of the paper, the theoretical and practical implications of the findings will be argued. The paper adds to the existing literature by exploring an underdeveloped strategy against contestation and it digs into the definitions of contestation and stigma. The practical implications can be useful for the contested firms, giving it a underused strategic option against contestation. It can also aid the contested, in understanding and countering firms that pose a threat to the public.
In the following section, existing literature will be assessed, and the independent and dependent var-iables will be defined. At the end of the chapter, several hypotheses will be proposed.
Literature Review
The body research on legitimacy has been growing (e.g. Vergne, 2011, Deephouse & Suchman, 2008). As stated in the introduction, scholars have found a positive relationship between legitimacy and organizational survival (Deephouse & Suchman, 2008). Moreover, academic work in the field of strategic management has shown a positive influence of legitimacy on performance measures like IPO’s (negatively related with underpricing) (Cohen & Dean, 2005) and stakeholder support (the more cognitively legitimate stakeholders perceive, the greater the likelihood of stakeholder support) (Choi & Shepherd, 2005). Yet organizations seem to be able to survive without legitimacy. Compa-nies that face public disapproval or contestation overcome or counter the negative effects of not hav-ing legitimacy. The question that rises is what these companies do to fight contestation? How do they strategize against it? To be able to answer them properly, a clearer picture of these organizations, and the environment that they are in, is needed. In order to categorize different types of contestation amongst firms, three types of contested firms will be analyzed. The first type of companies consists of companies that operate in contested industries. These companies are core-contested companies. The other two types both operate in non-contested industries. These are event-stigmatized and scru-tinized firms. First, I will discuss the characteristics of contested industries and the firms that operate in them. Afterwards I will look at the two types of contested companies that operate in non-contested industries.
7 Contested Industries
Contested industries have been defined as industries that face widespread disapproval because some sectors of society find them “offensive, inappropriate, or harmful” (Davidson, 2003, p. 2). A useful framework to operationalize contested industries comes from Galvin et al. (2005). They pose that, due to media rhetoric, industry evaluations can be brought to the public. Society may then either accept or contest an industry on the bases of these evaluations (Galvin et al., 2005). They infer: “We build from this position to argue that questions regarding societal evaluations and appropriate-ness of industries show up as struggles over products, strategies, and standard practices, and bound-aries amongst actors. These can translate into questions about the “lawfulness” of an industry and its practices due to the ability to collectively mobilize contestation and the ability of regulatory actors to impose industry or field-level constraints in response to such contestation. (2005, 62)”
They hereby imply that disapproval of certain industry characteristics can lead to two interrelated responses. The first is the collective mobilization of contestation. The second is the regulatory re-strictions on the basis of this contestation. What’s even more important, however, is that they provide us with four industry characteristics that can be contested; the products, strategies, standard practices and boundaries amongst actors within an industry. These will be used to determine whether or not an industry is contested. To further clarify these characteristics, a framework is used that categorizes them into four levels: product feature level, industry participant level, industry-stakeholder relation level, and industry level (Porac et al. 2002). It is argued that negative evaluations of product features will lead to negative evaluations of industry participants. A negative evaluation of these two levels will in turn lead to a negative evaluation of industry-stakeholders. When they are all combined, the industry as a whole becomes negatively evaluated. I argue, following this line of reasoning, that a negative public evaluation of any of these levels serves as an indicator that an industry is contested. In the next paragraph, I will give an explanation of the process of an industry becoming contested along with several examples from the tobacco industry.
Public scrutiny starts are the product level. Product features are comprised of the characteristics, the use and the buyer/seller profiles that make a product or service unique. Changes in the evaluation of
product features, brought about by media rhetoric, impact the normative acceptability of a product and therefore lead to contestation (Galvin et al. 2005). When, for example, cigarettes are constantly negatively depicted in the media because they are unhealthy, this will ultimately lead to a negative evaluation of the product.
When the public has formed a stable opinion about the products or services that an industry supplies, it will start forming similar opinions about those that are associated with the product. This affects the industry participants, such as producers, regulators and consumers. Here, again, the media plays a crucial role by identifying these participants (Galvin et al., 2005). Again, we turn to the tobacco in-dustry to illustrate this phenomenon. When the media constantly scrutinizes tobacco companies be-cause they sell harmful products, public opinion will eventually turn against these firms. Likewise, when smokers are constantly scrutinized for the harm they cause others, due to second hand smoke, they will also become negatively evaluated by the public.
The next level, industry-stakeholder relationships, concerns relationships between industry incum-bents and other actors in the broader environment, like competitors, suppliers, customers, regulatory agencies and other stakeholders. They define appropriate forms of doing business. When people share negative opinions about the products that an industry provides and about those that use and provide it, they will eventually disapprove of the way that business is done within that industry. The media comes into play by emphasizing those industry relationships, that are not congruent with public stand-ards (Galvin et al., 2005). Once more, an example is provided by the tobacco industry. When the media constantly emphasizes ways in which tobacco firms try to cheat tar or nicotine regulations, eventually, the way business is done within that industry will face public scrutiny.
When the public shares negative evaluations about an industry at these three levels, the industry as a whole is said to be contested (Galvin et al., 2005). Because a negative evaluation of one level will eventually lead to negative evaluations of the others, I argue that any of them serves as an indicator that an industry is contested. In extent, when the media is constantly scrutinizing an industry on one
9 of these levels, the industry is contested. By using these conditions, a clear distinction is made be-tween contested and non-contested industries.
Within these contested industries we find the first type of contested firms: the core-stigmatized firm. Core stigma relies on a deeply integrated flaw. When an organization is discredited because their core attributes, such as products, routines, or customers are incongruent with socially accepted norm, it is said to face core-stigma (Hudson & Okhuysen, 2009). In other words, core stigma arises from public disapproval of the very nature of an organization, who it is, what it is and whom it serves (Hudson, 2008). As argued above, these same characteristics are used to define whether an industry is con-tested. I, therefore, argue that whenever an industry is contested, all of the firms operating in this industry face core-stigmatization.
Contestation in Non-Contested Industries
Now that we have looked at what defines contested industries and the firms that operate in these industries, we turn to contestation in non-contested industries. I will argue that firms operating in non-contested industries can face two types of contestation: scrutiny and event-stigmatization. To understand the difference between these two categories, first a look will be cast at the definition of contestation itself. Contestation is defined as: “settings characterized by social contestation and tar-geted scrutiny by hostile audiences” (Durant & Vergne, 2014). It simply implies that there are people that discredit an organization for their actions of their characteristics. This most simple form of con-testation is the first type of concon-testation that a firm may face. It arises whenever the media scrutinizes a firm. I will define this type of contestation as scrutiny, and the firms facing it as scrutinez firms. The second type of contestation in contested industries is event-stigmatization. There are two things that set scrutinized and stigmatized firms apart: the severity of the contested attribute and the de-individualization that a stigma causes. Scrutiny is aimed at a single organization. A stigma, in con-trast, links an organization to a group of organizations that all carry the same stigma (Devers et al., 2009). Durant and Vergne define stigmatization as a: “special instance of contested settings, wherein the persistence of a deeply discrediting attribute nurtures distancing between business insiders and outsiders” (Durant & Vergne, 2014). As said, it is suggested that stigmatization refers to a more
deeply-discredited characteristic. This is confirmed by Devers et al., who infer that an organizational stigma is “a label that evokes a collective stakeholder group-specific perception that an organization possesses a fundamental, deep-seated flaw that de-individuates and discredits the organization” (Devers et al., 2009). Next to confirming that fundamentally flawed characteristics are needed for a stigma to occur, they mention that a stigma de-individuates the organization. The organization is linked to a group of organizations that are all perceived as having values that are not socially accepted. The stigma ties the organization to a category that is deeply discredited (Devers et al., 2009). Putting these observations together, a stigmatized organization will be defined as belonging to a group of organizations that share a characteristic that is deeply discredited by society. Both core-stigmatized and event-stigmatized companies share this definition. The difference between the two lies in the origin of the stigmatization.
As argued before, core-stigmatized firms receive their stigma because they operate in a contested industry and their core attributes, such as products, routines, or customers are incongruent with so-cially accepted norms (Hudson & Okhuysen 2009). They face de-individuation because they are linked to the other firms in the contested industry.
Event stigma, in contrast, is caused by discrete, anomalous events (Hudson, 2008). Examples of these types of events are bankruptcy, oil spills or significant product defects (Hudson & Okhuysen, 2009). The de-individuation that these firms face is, therefore, not based on industry participation but based on participation in a negative event.
After having differentiated between scrutiny, event-stigma and core-stigma, the different types of contested firms can then be grouped along two dimensions: whether they are scrutinized or stigma-tized and whether they operate in a contested or non-contested industry. This is illustrated in Table 1. The reason why there are no firms in the left bottom corner is because firms that operate within a contested industry are automatically facing the characteristics of severity and de-individuation that defines a stigma. They cannot face the less severe type of contestation that scrutiny implies. Nor can they face contestation that does not de-individuate the organization.
11 Table 1: Three Types of Contestation
Scrutiny Stigmatization
Non-Contested Industry Scrutinized Firm Event-Stigmatized Firm
Contested Industry - Core-Stigmatized Firm
Acquiring Foreign Assets
After having defined the independent variables we return to the questions posed in the beginning of this chapter: What is it that companies do to fight a stigma? How do they strategize against it? Roulet names three ways in which organizations may choose to react to stigmatization.
The first is isolation from the contested category. A firm may do this by trying to show that they are different, and better, than the other firms in a contested industry (Roulet, 2015). An example of this body of research comes from Hargie and colleagues, who take a look at the responses by four large banks to the banking crisis. They find a pattern of organizational dissociation from the events that are scrutinized (Hargie et al., 2010). This is underscored by Riaz et al., who finds that banks focus on finding others to blame and on appealing to expert authority, thereby isolating themselves from the contestation (Riaz et al., 2011). A second way of dealing with contestation, according to Roulet, is to actively fight it. An example given by Roulet himself, is that of a bank trying to defend big bonuses given to top management based on the need to draw top talent (Roulet, 2015).
The last strategy proposed by Roulet is the acceptance of the contestation. Without fighting or deny-ing contesation, companies can survive and sometimes even prosper (Roulet, 2015). Puttdeny-ing these strategies into simplified categories, according to Roulet a contested firm can choose to deny, fight or accept contestation.
The strain of research on this last category, of accepting contestation, has been growing over the last decade (e.g. Palazzo & Richter, 2005, Hsu & Grodal, 2015, Simons, Vermeulen & Knoben, 2016). Scolars look at strategies that organizations use to survive, whilst being contested. All of them have,
however, looked at strategies implemented within the country in which the contestation is happening. Organizations may, however, choose another path of action in order to survive contestation. Contes-tation may force an organization out of the areas in which they are contested. They may find that operating in their home country is becoming less profitable and decide to invest in assets in other parts of the world, where they are not publicly contested. In this paper, I will test for this relationship between contestation and the investment in assets in foreign countries. The first argument put forth in this paper is that investing in assets overseas is an overlooked strategy against contestation. I argue that contested companies will choose to follow this path of investing overseas. I pose that there is a positive relationship between contestation and the acquiring of foreign assets. It is important to stress that I am not necessarily arguing for a transfer of assets. When following the strategy of investing overseas, a firm may simply choose to lower investments in their home country and increase invest-ments in foreign countries. This does not imply that assets are being transferred.
Hypothesis 1: There is a positive relationship between contestation and investment in foreign assets
There are three forms of contestation that have been conceptualized in the theoretical framework: scrutiny, event-stigmatization and core-stigmatization. All these forms of contestation imply media scrutiny. For scrutinized companies, media scrutiny is the only form of contestation that they are facing. Event-stigmatized and core-stigmatized firms are, next to media scrutiny, also facing de-in-dividuation into stigmatized groups (Devers et al., 2009). Event-stigmatized firms are associated with other firms that are contested on the basis of their participation in a negative event. Core-stigmatized companies are associated with a group of companies that are contested on the basis of their partici-pation in a contested industry. I argue that this de-individuation increases the level of contestation because firms are not solely responsible for the disapproval that they are facing. When another firm in the group becomes the target of public scrutiny, there will be a negative spillover to the other firms in the group. Think of the contestation that oil companies face after several oil spills. These spills
13 may not be caused by every single oil company, yet the public associates all oil companies with the negative events. Hence, next to the individual media scrutiny that these companies are facing, they are also experiencing group-based scrutiny. Media scrutiny and de-individuation, thus, are both driv-ers of contestation.
In line with the argument that there is a positive relationship between contestation and investment in foreign assets, I argue that media scrutiny and de-individuation both positively influence investment in foreign assets.
Hypothesis 2: There is a positive relationship between media scrutiny and investment in foreign as-sets
Hypothesis 3: There is a positive relationship between de-individuation caused by a stigma and in-vestment in foreign assets.
The final hypothesis of this paper is based on the difference between core-stigmatization and event-stigmatization. For both types of stigma, contestation comes from media scrutiny and de-individua-tion. I argue, however, that the contestation resulting from de-individuation is more severe for core-stigmatized companies than for event-core-stigmatized companies. This difference lies in the basis on which the de-individuation is happening. For event-stigmatized companies, this is based on negative events. For core-stigmatized companies, this is based on contested industry participation. In this last case the characteristics on which the de-individuation is based are far more defining for the company. The people associate a company with other companies based on negative evaluations of their prod-ucts, other industry participants and industry-stakeholder relations. For event-stigmatized firms, de-individuation is not based on all of these characteristics but only on an event or a series of events. For example, in the case of event-stigmatization, a company may be associated with other companies due to a product malfunction. This is, however, only one characteristic of the company that is contested.
In the case of core-stigmatization, a range of characteristics is contested. I, therefore, argue that con-testation that is resulting from de-individuation is stronger for core-stigmatized firms than for event-stigmatized firms. Because contestation is said to have a positive relationship with the investment in foreign assets, hypothesis 4 is:
Hypothesis 4: De-individuation based on contested industry participation has a stronger effect on investment in foreign assets, than de-individuation based on negative events
In summary, the model that is proposed, consists of three independent variables and one dependent variable. The three independent variables are media scrutiny, de-individuation based on contested industry participation and de-individuation based on negative events. All forms of contestation are based on one or two of these variables. The dependent variable is investment in foreign assets. The model is illustrated in model 1.
Model 1: The Effect of Contestation on Investment in Foreign Assets
To study the relationship between contestation and foreign investment, these variables need to be operationalized and a methodology for acquiring the data needs to be put forth. The following section will address this.
Media Scrutiny De-individuation based on Negative Events De-individuation based on contested Industry Participa-tion Investment in Foreign Assets
15
Research Design
This paper will attempt to show a relationship between contestation and investing into foreign coun-tries. I have defined three types of contested organizations, operating either in contested or non-con-tested industries. These groups all face contestation in some form. They all face media scrutiny and next to this, the event-stigmatized and core-stigmatized companies also face de-individuation. The first question, then, that needs answering is whether media scrutiny has an effect on investment in foreign assets. Consequently, I will assess the effects of the two aforementioned types of de-individ-uation on foreign asset investment.
As a sample, I will use all firms from the S&P 500 list. This list consists of the 500 largest companies in the US, based on their market capitalization. This sample was chosen for two main reasons. The first lies in the nature of contestation. Contestation is composed of scrutiny and de-individuation. Both can be seen as ways in which people evaluate a company. When evaluating a company, people look to certain values that they find important. These values may differ across different countries and cultures. In some countries, environmental considerations will be greatly valued. People in other countries may find ethical considerations more important. On the basis of these cultural values, com-panies are assessed and possibly contested. Looking at comcom-panies in one country makes it possible, to some extent, to control for these cultural differences. The second reason is that all companies share a relative similarity in size. One can imagine that a company that is small in size has less resources to invest in foreign assets. These companies may choose to take other paths in strategizing against contestation. By using only the largest 500 companies in the U.S., I am able, to a certain extent, to control for size differentials as an alternative explanation for investment in foreign assets.
The time period that will be investigated ranges from 2005 to 2015. This time period is chosen to ensure a complete look at strategy development and implementation. When a company faces contes-tation in one year, it is not likely to change its strategy immediately. Several years of contescontes-tation are expected to have a more defining influence on firm strategy. Next to this, the implementation of the strategy to invest overseas will also take time. This implies that there may be a time lag between the
strategic decision to invest overseas and the actual possession of foreign assets. To assess a complete picture of this entire process, I have decided to look at a ten-year time period of contestation.
Independent Variable: Media Scrutiny
The first step of data collection will focus on operationalizing the independent variables. The first and most important variable that all contested groups share is media scrutiny. To measure media scrutiny, a series of Asset4 variables were used that examine whether a company has been under the spotlight of the media due to controversy linked to a certain issue. In total, 20 of these issues were addressed, divided across six categories. The first category describes controversies linked to a com-pany’s corporate governance structure. An example of an issue in this category is whether a company has been under the spotlight of the media due to controversy linked to high executive or board com-pensation. The second category examines controversies linked to economic issues, such as insider dealings or customer dissatisfaction about products. The third category looks to environmental con-troversies, such as oil spills and biodiversity. Categories four to six are all related to social issues. The first of these social categories, product responsibility, refers to issues regarding a company’s ability to produce quality goods and services, whilst upholding customer health and safety. An ex-ample of an issue in this category is whether a company has received media attention caused by marketing unhealthy products. The second social category, society and community, analyses whether a company respects business ethics and public safety. An example of an issue in this category is whether a company has been under the spotlight of the media concerning corruption or bribery. The final social category, society and human rights, focuses on human rights issues. An example of an issue in this category is whether a company has received media attention due to child labor charges. A full list of the variables is given in the appendix(i). When a company has been under the spotlight
of the media in one year, it receives a score of one. Adding all the scores of all issues together, an overall score of media scrutiny was calculated for every company. For example, if a company faces scrutiny on two different issues in the year 2006, scrutiny on four issues in 2010 and scrutiny on one issue in 2013, it receives and overall media scrutiny score of 7. For scrutinized firms, these scores
17 core-stigmatization, event-stigmatization and investment in foreign assets, two variables are added that operationalize de-individuation.
Independent Variable: Core-Stigmatization
Next to scrutiny, another variable is needed to describe the situation which core-stigmatized compa-nies face. Not only is the company itself contested, it is linked to other firms in the industry, which is contested. This association satisfies the condition of de-individuation of a stigma (Devers et al., 2009). Not only does the core-stigmatized firm face media scrutiny, it also faces extra contestation that is projected onto the group, which it is a part of. This contestation may not be caused by the firm itself. Whenever another firm in the contested industry is linked to bad behavior, there is a spillover onto the other groups in the firm. It is, therefore, predicted that contested industry participation, next to media scrutiny, will also affect investment in foreign assets.
To operationalize industry contestation, a dichotomous variable is used. This variable defines whether a firm is experiencing de-individuation due to contested industry participation. Important to keep in mind is that the contested industry does not necessarily have to be the primary industry a firm is operating in. Whenever a company receives revenue from a contested industry, it will be marked down as a core-stigmatized company. Four industries were designated that satisfy the conditions of a contested industry (Galvin et al., 2005): the tobacco industry, the arms industry, the pornography industry and the gambling industry1. There were, however, no firms in the sample that obtained
rev-enue from the pornography industry so only the other three industries are represented in this study. In order to assess whether a firm from the sample is operating one of these industry, three Asset4 variables were used that assess whether the revenues, generated from contested industries, are larger than five percent.
1 The alcohol industry was not included as a contested industry because it does not satisfy all the characteristics of a
contested industry, put forth by Galvin et al. (Galvin et al., 2005). Products of the alcohol industry are, to some extent, contested. In contrary to large tobacco conglomerates like Phillip Morris, however, breweries do not seem to face exten-sive media scrutiny. This implies that industry participants are not contested and therefore I decided to leave the alcohol industry out of the contested industry group.
The companies that yielded a yes for any of these keywords received a score of one for the dichoto-mous variable de-individuation caused by contested industry participation and were listed as core-stigmatized companies. A full list of Asset4 variables is given in the appendix(ii)
Independent Variable: Event-Stigmatization
Like core-stigmatized and contested companies, event-stigmatized companies experience media scru-tiny. Yet, like core-stigmatized companies, the effect of contestation on foreign investment is not explained solely by this variable. A defining characteristic of event-stigmatized companies is that they are de-individuated into groups based on an event that triggers public disapproval (Hudson, 2008). This event allows the public associate the event-stigmatized company with a group of organ-izations, that all share the same flaw. This association will, next to media scrutiny, add to the severity of contestation. It is therefore hypothesized that it will have an effect on the investment in foreign assets.
In order to account for this effect, a dichotomous variable is used that states whether a firm is expe-riencing de-individuation on the basis of a negative event. In order for this type of de-individuation to happen, two factors need to be in place. The first is that companies must share some amount of similarities to be associated with each other in the minds of the public. For example, when a Coca-Cola and Pepsi are both simultaneously scrutinized by the media on the basis of a major product defect, it is unlikely that the public will project this scrutiny onto firms in the banking world. Other firms in the beverage business are, however, prone to experience negative consequences from this negative event. In order to address this first factor, the industry group classification level of the Thompson Reuters Business Classification index was used. I argue that within these industry groups, de-individuation is possible.
The second factor that needs to be accounted for is that firms are grouped together on the basis of events. To test for this condition, again, the list of Asset4 variables was used. Instead of adding all scores on the different issues together, however, only the scores of the issues per category were added.
19 Investigating media scrutiny categorically made it possible to look at de-individuation based on in-volvement in controversial events. In the following paragraph I will elaborate on the process of des-ignating which firms are experiencing de-individuation based on negative events.
First, all the firms were scored on each of the six categories: corporate governance, economic, envi-ronmental, product responsibility, society & community and society & human rights. For example, when a company was under the spotlight of the media due to controversy linked to high executive pay for two separate years and it was under the spotlight of the media for four separate years due to controversy linked to shareholder rights, it received a score of 6 on the category corporate govern-ance. Afterwards, a rise in industry group media exposure was measured per category. When multiple firms within an industry group are experiencing scrutiny based on one category, de-individuation may be taking place. To operationalize when an industry group is actually experiencing a rise in media scrutiny, an arbitrary condition is proposed. When two thirds of all the firms in an industry group have a score of at least 1 on the same category, the companies in this group are possible event-stig-matized companies2.
Not all of these firms, however, are used as event-stigmatized firms. As mentioned by Devers et al., the stigmatized organization is perceived to possess a fundamental, deep-seated flaw (Devers et al., 2009). It would therefore be an overgeneralization to list all the firms in an industry group that is experiencing a rise in scrutiny, as event-stigmatized. To address the characteristic of severity, again, an arbitrary line was drawn. Of all the firms that received a score of at least one, only those residing in top-quartile of scores will be marked as event-stigmatized firms3. Using these two arbitrary
condi-tions, a list of 96 Event-stigmatized companies was created. All of these firms received a score of one
2 The arbitrary decision that an industry group is experiencing a rise in media exposure when two-thirds of all
constitu-ents have a score of at least one on the same category is touched upon in the discussion as a possible limitation of the research design
3 The arbitrary decision is made due to the greater amount of variance that it creates between the three different groups
of contested firms. Three analyses were done, using all firms in the industry groups that were seeing a rise in media scrutiny, using the firms that resided in the top half of all scores and using the firms that resided in the top quartile of all scores. The best results were yielded by the latter category
on the dichotomous variable de-individuation caused by negative events and were listed as event-stigmatized firms.
Dependent Variable: Investment in Foreign Assets
In order to show a relationship between different levels of contestation and investment in foreign assets, a variable is used that compares the amount of foreign assets and the amount of homeland assets. Datastream was used to extract this data. The variable that was shows the percentage of foreign assets compared to total assets. It is a ratio variable ranging from 0 to 100 percent. This variable was chosen, opposed to using data that describe foreign investment patterns in the years following con-testation, to give a more complete insight into the relationship between contestation and the invest-ment in foreign assets. As invest-mentioned before, strategy will take time to be developed and to be imple-mented. I argue that looking at the differences in foreign asset percentages after a ten-year period of varying contestation, will show whether a company has undergone the entire process of recognizing contestation, choosing the strategy of investing in foreign assets and actually carrying out the strategy by purchasing assets overseas.
Control Variables
What is important to recognize is that the decision to purchase assets overseas can be influenced by wide range of factors besides contestation and thus there is a need for a set of control variables to control for these external effects. In choosing control variables I used existing literature on investment strategies (Biddle & Hillary, 2006). All the control variables were extracted from Datastream using descriptives that are listed in the appendix(iii)
The first variable is market value. It is measured in US dollars. I used this control variable to control for differences in size between the companies in the sample. The use of S&P 500 companies already accounts for some variation in size, but a control variable can still be useful to account for the differ-ences that remain. Size may influence investment behavior due to a difference in resources. Larger companies may be more equipped to invest overseas than smaller companies.
21 Next to size, other factors come into play when it comes to foreign investment. Companies that are large do not necessarily have the financial means to invest overseas. It is important to know if a firm is profitable and if it is able to generate cash from revenues. Taking this into account, two control variables are added: cash flow per share and Return on Assets. Cash flow per share depicts the after-tax earnings plus depreciation and is an indicator of a firm’s financial strength. It is expressed in US dollars. Return on assets depicts a company’s net income divided by its total assets. It is an indicator of how profitable company is compared relative to its assets. It is also expressed in US dollars. It is expected that firms with a low cash flow per share or a low return on assets are less likely to act on investment opportunities. The last variable that is added to the study is total debt, which is also ex-pressed in US dollars. If a firm has a high debt burden, it is less likely to invest in foreign assets because lowering debt may be of a higher priority.
Since the time period in which firms may strategize against contestation by investing in foreign assets ranges from 2005-2015, the control variables depict averages over the course of this period. For every year a value was extracted from Datastream, after which an average was calculated for every variable. Descriptive Statistics
After having defined and operationalized all the variables, we shortly assess their basic descriptive statistics. Using SPSS’ descriptive statistics, a brief overview of the data was made. The descriptives of the Independent variables, the control variables and the control variables are shown in Table 1.
As shown in the table, I was not able to extract all the data for all companies from Datastream. The amount of valid N is, therefore, brought down to 466 companies. What stands out mostly in the de-scriptives table is the high standard deviation of the dependent variable. When looking at frequency statistics, it shows that this standard deviation is caused by a small set of outlier firms that have a very high percentage of foreign assets. This may influence the analysis and will therefore be touched upon in the discussion.
Next to these descriptive statistics a correlation table is needed to assess the correlations between all the variables. The correlations are shown in Table 2.
The important things that stand out in the table are that two of the four control variables have a sig-nificant correlation with the dependent variable. These control variables are Return of Assets and Market Value. The other two control variables, Total Debt and Cash Flow per Share, are not signifi-cantly correlated with the dependent variable. Of the independent variables, only the total media scrutiny score is significantly correlated with the dependent variable. Both types of de-individuation do not have a significant correlation with the dependent variable.
23
Methods & Results
In order to test the predictability of the independent variables, a hierarchical linear regression was used. The first model of this regression will assess the relationship between the control variables and the dependent variable. Afterwards, the independent variables will be added to the model. First, the independent variable media scrutiny was added. This allowed me to check if media scrutiny by itself can be a predictor of investment in foreign assets. Finally, both types of de-individuation were added to the model to assess if they enhance the predictive ability of the model.
Hierarchical Regression
First, there are a couple assumptions of a linear regression that need to be addressed. First of all, both the independent and dependent variable need to be continuous. This is the case since the independent variable media scrutiny and the control variables are interval variable and the dependent variable is a ratio variable. Secondly, the relationship between the independent variable and dependent variable needs to be linear. Since the relationship that I am predicted is linear, this assumption is satisfied. Next to continuous variables and linearity, a linear regression also assumes non-collinearity. This assumption can easily be checked by created in a scatter plot of the data. This plot shows no signs of collinearity. It is provided in the appendix(iv). The last two assumptions are homoscedasticity and
normality. Homoscedasticity means that for every value of the independent variable, the variance of the residuals of the dependent variable is constant, thereby implying that the predictive power of the model remains constant for high values as well as low values. Normality means that the residuals of the dependent variable are normally distributed with a mean of zero. To check for normality, a histo-gram and standardized p p-plot were made of the residuals of the dependent variable. These plots are shown in the appendix(v). Both show a distribution close to a normal distribution and the condition of
normality is therefore satisfied. To check for homoscedasticity, a scatter plot was made of the stand-ardized residuals and the standstand-ardized predicted values. This plot is shown in the appendix(vi). The
plot does not show an obvious pattern so there are no signs of homoscedasticity. The results of the hierarchical regression are shown in table 3.
In the first model, the predictors are the control variables: return on assets, market value, total debt and cash flow per share. This model is significant with a p-value smaller than 0,05. The regression coefficient of the first model is 0,201, meaning that there is a correlation between the independent and the dependent variables. This correlation is significant because the t-value is 0,001, which is smaller than 0,05. The R square shows that the predictive capacity of the basic model is 4 percent. The control variables, therefore, predict 4 percent of the variance of the dependent variable.
In the next model, model 2, total media scrutiny is added as an independent variable. The model is significant with a p-value that is smaller than 0,01. What we see is that R, the regression coefficient, has gone up to 0,223. The R square change is significant because the t- value is 0,031, which is smaller than 0,05. The R square value has also gone up to 0,05, which implies that the predictive capacity of the model is enhanced by adding media scrutiny. The model is now able to explain 5 percent of the variance of the dependent variable. Media scrutiny, thus, is able to predict 1 percent of the percentage foreign assets divided by total assets. This may seem like a small number, but it can actually amount to quite a large amount of assets. To illustrate this fact, the average amount of foreign assets of all the companies in the sample was estimated at 9.312.552, 67 Million US Dollars, in 2015. In the third model, de-individuation based on contested industry participation and de-individuation based on negative events were added. Although the model remains significant with a t-value that is smaller than 0,05, the R-squared change is not significant. This implies that de-individuation has not added to the predictive capacity of the model. From these models, we can conclude that media scru-tiny adds predictive ability and that de-individuation doesn’t.
Table 4 gives an overview of the coefficients of the variables that went into the three models. The first thing that is important to notice is that in the first model, only total debt and market value are
25 statistically significant predictors of the dependent variable. This will be touched upon in the discus-sion. The second important thing to point out is that media scrutiny, in the second model, has a beta of 0,119, which is significant. This would be expected, since adding media scrutiny to the model, yielded a significant R-squared change. The last thing that is important is that the B value of 0,172 tells us that for an increase of one unit of media scrutiny, the percentage of foreign assets divided by total assets increases with 0,172 percent. This confirms hypothesis 1, which states that there is a positive relationship between media scrutiny and investment in foreign assets.
Hypothesis 3, which predicted a positive relationship between de-individuation and investment in foreign assets, cannot be confirmed by the data. Hypothesis 4, which predicted a stronger influence of one type of de-individuation over the other, can therefore not be assessed. Having reported the results, I now move to the discussion and conclusions section, in which the results will be linked back to the theoretical framework.
Discussion & Conclusions
This paper set out to improve the theoretical understanding of strategic choices that firms make when they are experiencing contestation. Of the three paths, laid out by Roulet (Roulet, 2015), that firms may choose when the need to strategize against contestation, the one that entails accepting the stigma had been the most underdeveloped. Within this pathway, several sub areas of strategizing had already been explored by scholars (e.g. Palazzo & Richter, 2005, Hsu & Grodal, 2015, Simons, Vermeulen & Knoben, 2016). There was, however, one strategic direction that had been mostly overlooked. This entailed the strategic decision to invest in overseas areas where contestation is less prominent or where regulations resulting from public scrutiny are less prominent. In order to delve into this area of strategizing, the goal was set to investigate the relationship between contestation and the acquiring of foreign assets.
In order to investigate this relationship a model was proposed that predicted a positive effect of con-testation on the investment in foreign assets. In model, three independent variables were defined to describe the concept of contestation. Media scrutiny and two types of de-individuation allowed for the differentiation of three types of contestation. All three were hypothesized to have a positive effect on the investment in foreign assets.
The analyses assessed the strengths of three models. The first used only the four control variables and yielded a significant relationship. I the second model, media scrutiny was added. It was shown that this variable, indeed, had a significant r-squared change on 1 percent. The relationship between media scrutiny and investment in foreign assets turned out to be a positive one and hypothesis 2 was con-firmed. The two types of de-individuation were, however, not adding to the predictive ability of the model in a significant way and hypotheses 3 and 4 were rejected. Hypothesis 1, predicting a positive effect between contestation and the investment in foreign assets, can therefore only be confirmed for scrutinized firms and the model that is proposed in the theoretical framework is only partially con-firmed by the analysis.
27 Findings and Theoretical Implications
Going back to the theoretical framework, this study suggests that some of the characteristics about stigma can be added. The first concerns the de-individuation that a stigma brings about (Devers et al., 2009). Whether a company is linked to a group of firms due to their collective participation in a contested industry (core-stigmatized companies), or a to a group of firms across an industry group due to their collective participation in an event or series of events (event-stigmatized companies), these linkages do not result in a higher ratio of foreign assets. This can imply that firms are not faced by this effect as much as by scrutiny, that they underestimate the effect or that they don’t have enough knowledge on its implications. It can also mean, and this seems like a more plausible explanation, that firms that reside in these groups, are more likely to resort to other strategies to counter contesta-tion. An example of this comes from the banking industry. In the aftermath of the financial crisis, banks were grouped together based on their participation in this negative event. In order to counter this contestation, the majority of banks resorted to the distancing strategy (Roulet, 2015). It may lie in the very nature of stigmatization that firms do not accept it, because they feel like it is not solely caused by themselves. This may create a sense of unfairness about the severe consequences that the stigma has. Firms may feel like they are punished more than they deserve. This, however, remains a speculative assertion that can only be confirmed by further research.
The second characteristic of stigma that is challenged by this study is its severity. According to pre-vious literature, a difference in severity of contestation exists between scrutinized and stigmatized firms (Durant & Vergne, 2014). This assumption has not been proven by the model of this paper. A possible explanation may again be that stigmatized firms tend to strategize differently against public scrutiny than contested firms. These assertions can, however, only be confirmed by further research. This study, thus, calls for a better theoretical framework to describe the differences between the three forms of contestation
Having already touched upon several implications of this paper, a broader sense of its theoretical contributions is needed. The paper makes several additions to the existing literature. First of all, it provides insight into a relatively uncharted area of strategizing against contestation. By showing a
significant relationship, between media scrutiny and the acquisition of foreign assets, a previously overlooked strategy has been added to previous research.
The second addition that this paper makes concerns the difference in severity of contestation between scrutinized and stigmatized firms. The study does not show that core and event-stigmatized firms are more likely to acquire overseas assets than scrutinized firms, although this might have been expected due to the more severe nature of stigmatization they face. It seems that event-stigmatized and core-stigmatized firms do not view their situation too tough to survive. This may imply that the conse-quences of contestation and stigma are not so different. It may also imply that stigmatized firms are better equipped or experienced to face public scrutiny. In either case, a better definition of scrutiny and stigmatization is needed, in order to explain the lack of differentials found in this study.
Practical implications
The practical implications of the study can be useful for the contested firm as well as the contester. Contested firms may interpret the study in several ways. They can either view the limited use of foreign investment as a strategy against contestation as a sign that it is not a good strategy. They may also view it to be an underdeveloped or underused strategy and look into its compatibility. Another takeaway for firms is that operating in a contested industry or being grouped together into event-stigmatized groups has not led other firms to move away from the area in which they are contested. They might interpret this as a sign that, even though the contestation that they face may seem hard to overcome, there is no reason to leave the countries in which they are contested. This may strengthen a firms believe that survival is possible when other strategies are conducted.
On the side of the contester, whether that is the media, the public, the government, or all together, the study has some practical implications as well. Whenever it is needed to constrain a firm’s behavior due to its incongruence with normative standards, it is important to analyze its strategy under the circumstances of contestation. Knowing that some contested firms choose to move their operations elsewhere, may allow for some international cooperation in constraining bad firm behavior. An ex-ample is informing other countries about the consequences of the movement of certain firms, such as
29 Limitations
Having discussed the theoretical and practical contributions and implications of the paper, we now turn to its shortcomings. I believe the first shortcoming lies in the fact that I wasn’t able to fully the effects of event and core-stigma. According to previous work, a clear difference exists between con-testation and the two types of stigma but in this study, this has not resulted in a change of the depend-ent variable. Another limitation may lie in the choice of sample. By choosing to use 500 large com-panies, the effects of contestation on smaller companies are ignored. It would be interesting to see how contestation will affect a smaller company. Smaller companies may be more vulnerable to con-testation because they have less experience or resources in order to strategize against it. They may imply that they are even less able to move overseas. It may also imply that they choose to sell over-seas, without purchasing a great deal of assets there.
A third limitation lies in the arbitrary decisions made in defining the rise in industry scrutiny, used to define event-stigmatization. Using a higher arbitrary cut off point, may ensure that the companies in the event-stigmatized group are experiencing a higher amount of contestation.
The fourth limitation comes from the control variables, of which two did not have a significant cor-relation with the dependent variable. Using other control variable may enhance the strength of the basic model
The final limitation lies in the usage of a total score of media scrutiny. It may be interesting to look at the effects of scrutiny in a smaller time period. What does a company do in the year following contestation? This is something that new research may assess.
Suggestions for Further Research
Having touched upon the limitations of this study, I will now discuss some suggestions of how further research can carry this academic field forward and enhance our understanding of strategizing against contestation even further. I argue that the first thing that has to be clarified by further research, is the definition of stigma and contestation and their respective differences. I call for more qualitative re-search so that the operationalizing of variables that capture contestation and stigma, becomes more accurate and precise.
The second possible extension of research may lie in examining the relationship between contesta-tion, stigma and foreign investment behavior for smaller companies. This will give a better and more generalizable insight into the acquiring of foreign assets as a response to contestation.
A last suggestion is to investigate if different types of contestation and stigmatization consistently lead to the same strategic reaction amongst the recipient firm. This study has shown that event and core stigma do not lead to a higher percentage of foreign assets. Other strategies may be used to counter the higher severity of contestation that a stigma brings about.
31
References
Devers, C.E., Dewett, D., Mishina, Y., Belsito, C.A. (2009) A General Theory of Oganizational Stigma. Organization Science 20(1):154-171
Davidson, D.K. (2003): Selling Sin: The Marketing of Socially Unacceptable Products (2nd ed.). Westport: Praeger Publishers.
Durand, R., & Vergne, J. P. (2015). Asset divestment as a response to media attacks in stigmatized industries. Strategic Management Journal, 36(8), 1205-1223.
Roulet, T. (2015). “What good is Wall Street?” Institutional contradiction and the diffusion of the stigma over the finance industry. Journal of Business Ethics, 130(2), 389-402.
Hudson, B.A. (2008). Against All Odds: A Consideration of Core-Stigmatized Industries. The Ac demy of Management Review, 33(1), 252-266
Hargie, O., Stapleton, K., & Tourish, D. (2010). Interpretations of CEO public apologies for the banking crisis: Attributions of blame and avoidance of responsibility. Organization, 17(6), 721–742.
Riaz, S., Buchanan, S., & Bapuji, H. (2011). Institutional work amidst the financial crisis: Emerging positions of elite actors. Organization, 18(2), 187–214.
Palazzo, G. & Richter, U. (2005). CSR Business as usual? The case of the tobacco industry. Journal of Business Ethics, 61, 387-401.
Driffield, N., Jones, C. & Crotty, J. (2013). Regulation as Country-Specific (Dis-)Advantage: Smoking Bans and the Location of Foreign Direct Investment in the Tobacco Industry. British Journal of Management, 27(3), 464-478.
Hsu, G., & Grodal, S. (2015): “Category taken-for-grantedness as a strategic opportunity: The case of light cigarettes, 1964 to 1993.” American Sociological Review, 80 (1), 28–62.
Simons, T., Vermeulen, P. A., & Knoben, J. (2016). There’s No Beer without a Smoke: Community Cohesion and Neighboring Communities’ Effects on Organizational Resistance to
Antismoking Regulations in the Dutch Hospitality Industry. Academy of Management Journal, 59(2), 545-578.
Oliver, C. (1991). Strategic Responses to institutional Processes. Academy of Management Review, 16(1), 145-179
Vergne, J.P. (2011). Toward a New Measure of Organizational Legitimacy: Method, Validation and Illustration. Organizational Research Methods, 14(3), 484-502
Suchman, M. C. 1995. Managing legitimacy: Strategic and institutional approaches. Academy of Management Review, 20, 571–610.
Deephouse, D. L., & Suchman, M. C. (2008). Legitimacy in organizational institutionalism. In R. Greenwood, C. Oliver, K. Sahlin, & R. Suddaby (Eds.), The SAGE handbook of
33 Cohen, B. D., & Dean, T. J. 2005. Information asymmetry and investor valuation of IPOs: Top
management team legitimacy as a capital market signal. Strategic Management Journal, 26: 683–690.
Choi, Y. R., & Shepherd, D. A. 2005. Stakeholder perceptions of age and other dimensions of newness. Journal of Management, 31, 573–596.
Porac, J., Ventresca, M. & Mishina, Y. (2002). Interorganizational Cognition. In Baum, J. (Ed), Companion to Organizations (579-598). Oxford: Blackwell
Galvin, L.T., Ventresca, M.J. & Hudson, B.A. (2005). Contested Industry Dynamics. International Studies of Management and Organization. 34, 56-82
Biddle, G.C. & Hilary, G. (2006). Accounting Quality and Firm-Level Capital Investment. The A counting Review. 81(5), 963-982
Appendix
(i) Asset4 Keywords
Code Category Meaning
CGCPO08(S or V) Corporate Govern-ance
Is the company un-der the spotlight of the media because of a controversy linked to high execu-tive or board com-pensation?
CGSRO07(S or V) Corporate Govern-ance
Is the company un-der the spotlight of the media because of a controversy linked to sharehold-ers rights?
ECSLO14(S or V) Economic Is the company
un-der the spotlight of the media because of a controversy linked to insider deal-ings and other share price manipulations? ECSLO17 (S or V) Economic Is the company
un-der the spotlight of the media because of a controversy linked to aggressive or non-transparent accounting issues? ECCLO12 (S or V) Economic Is the company
un-der the spotlight of the media because of consumer com-plaints or dissatisfac-tion directly linked to its products or ser-vices?
ECCLO13(S or V) Economic Is the company
un-der the spotlight of the media because of a controversy linked to anti-com-petitive behavior (e.g., anti-trust and monopoly), price-fix-ing or kickbacks?
35 ENERO02 (S or V) Environmental Is the company
un-der the spotlight of the media because of a controversy linked to biodiver-sity?
ENERO20(S or V) Environmental Is the company di-rectly or indidi-rectly (through a supplier) under the spotlight of the media because of a controversy linked to the spill of chemicals, oils and fuels, gases (flaring) or controversy relat-ing to the overall im-pacts of the com-pany on the environ-ment?
ENPIO21(S or V) Environmental Is the company un-der the spotlight of the media because of a controversy linked to the environ-mental impact of its products or ser-vices?
ENRRO13(S or V) Environmental Is the company un-der the spotlight of the media because of a controversy linked to the environ-mental impact of its operations on natural resources or local communities?
SOPRO08(S or V) Social Is the company
un-der the spotlight of the media because of a controversy linked to market with-drawal (closing of branches), retreating or failing to serve specific markets or customers?
SOPRO09(S or V) Social Is the company
un-der the spotlight of the media because
of a controversy linked to the compa-ny's marketing prac-tices, such as over marketing of un-healthy food to vul-nerable consumers?
SOPRO11(S or V) Social Is the company
un-der the spotlight of the media because of a controversy linked to its products or services quality and responsibility?
SOCOO06(S or V) Social Is the company
un-der the spotlight of the media because of a controversy linked to activities in critical, undemocratic countries that do not respect fundamental human rights or to disrespecting the rights of indigenous people?
SOCOO08(S or V) Social Does the company
report on crisis man-agement systems or reputation disaster recovery plans to re-duce or minimize the effects of reputation disasters?
SOCOO09(S or V) Social Is the company
un-der the spotlight of the media because of a controversy linked to public health or industrial accidents harming the health & safety of third parties (employees and non-customers)?
SOCOO10(S or V) Social Is the company
un-der the spotlight of the media because of a controversy
37 corruption, political contributions, im-proper lobbying, money laundering, parallel imports or any tax fraud?
SOHRO02(S or V) Social Is the company
un-der the direct or indi-rect (through suppli-ers) spotlight of the media because of a controversy linked to freedom of associa-tion?
SOHRO03(S or V) Social Is the company
un-der the direct or indi-rect (through suppli-ers) spotlight of the media because of a controversy linked to child labour?
SOHRO04(S or V) Social Is the company
un-der the direct or indi-rect (through suppli-ers) spotlight of the media because of a controversy linked to general human rights issues?
SODOO05(S or V) Social Is the company
un-der the spotlight of the media because of a controversy linked to workforce diversity and oppor-tunity?
SOEQO13(S or V) Social Is the company
un-der the spotlight of the media because of a controversy linked to the compa-ny's employees, con-tractors or suppliers due to wage, layoff disputes or working conditions?
SOHSO04(S or V) Social Is the company
un-der the spotlight of the media because of a controversy
linked to workforce health and safety?
(ii) Keywords Contested Industry Participation & List Core Stigmatized Companies
SOPRDP064: Are revenues generated from gambling activities larger than 5% of the total net
reve-nues?)
SOPRDP065: Are revenues generated from tobacco production larger than 5% of the total net
reve-nues?
SOPRDP062: Are revenues generated from armaments larger than 5% of the total net revenues? SODRDP067: Does the company produce cluster bombs?
(iii) Control variable Datamstream Codes
WC05501: Cash Flow per Share WC03255: Total Debt
WC18100: Market Value ROA1FD12: Return on Assets
39
Histogram Residuals Dependent Variable & P-P Plot Residuals Dependent Variable (v)
