POWER, MORAL IDENTITY AND WILLINGNESS TO ASSOCIATE Master Thesis
Student: Elma Heffels / student nr.: 11412070
MSc. in Business Administration, Strategy track
University of Amsterdam, Faculty of Economics and Business Supervisor: Dr. F. Bridoux
STATEMENT OF ORIGINALITY
This document is written by Student Elma Heffels 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.
CONTENT
POWER, MORAL IDENTITY AND WILLINGNESS TO ASSOCIATE ... 1
Master Thesis ... 1
Student: Elma Heffels / student nr.: 11412070 ... 1
ABSTRACT ... 5
INTRODUCTION ... 6
LITERATURE REVIEW ... 8
Stakeholders and morality ... 8
Morality and firm’s attractiveness to stakeholders ... 10
Stakeholders and competence ... 12
Morality/Competence trade-offs ... 13
Power as a potential moderator ... 14
Research gap and research questions ... 18
Figure 1: conceptual model ... 19
THEORETICAL FRAMEWORK ... 20
This chapter will introduce the hypotheses of this thesis. ... 20
Power and morality ... 20
Power and behavioural disinhibition ... 20
Power and heightened self-focus ... 22
Effects of power and willingness to associate ... 24
Moral identity as a potential moderator of power effects ... 25
Effects of moral identity, power and willingness to associate ... 25
METHODOLOGY ... 30
This chapter will explain the research approach and design of this paper. ... 30
Research procedure ... 30
Research methods ... 31
Sample ... 31
Vignettes ... 32
Dependent variable: willingness to associate ... 34
Moderating variables ... 35
Control variables ... 36
In this chapter, the results of this research will be reported. ... 37
Reliability analyses of scales ... 37
Manipulation and realism checks ... 38
Table 1: descriptive statistics and correlations. ... 39
Table 2: descriptive statistics and correlations (continued). ... 40
Hypothesis testing ... 41
Table 3: results of regressions analyses testing moderators. ... 43
Table 5: results of testing for multicollinearity in Model 1. ... 47
Table 6: results of testing for multicollinearity in Model 2a, 2b and 2c. ... 48
Cells are empty when variables relate to different set of vignettes ... 48
Table 7: results of testing for multicollinearity in Model 3a, 3b and 3c. ... 49
Cells are empty when variables relate to different set of vignettes ... 49
DISCUSSION ... 50
Findings, limitations and ideas for future research ... 50
Contributions of this study ... 55
CONCLUSION ... 57
REFERENCES ... 59
APPENDIX B: SURVEY MORAL IDENTITY ... 70
APPENDIX C: VIGNETTES ... 73
Example: ... 74
APPENDIX D: WILLINGNESS TO ASSOCIATE SURVEYS ... 75
Employee condition ... 75
Customer condition ... 76
APPENDIX E: SENSE OF POWER SURVEY ... 79
ABSTRACT
Stakeholder theory acknowledges that stakeholders are not merely interested in themselves, but also in what is good for others. As a result of this, a firm’s morality and competence interact in affecting primary stakeholders’ willingness to associate with (work for or buy from) a firm. When there is a trade-off, morality is generally preferred over competence. However, is this always the case? This thesis combines insights from social psychology with stakeholder literature to examine conditions that influence willingness to associate with a firm. In particular, the effects of the self-importance of moral identity and sense of power are examined. By means of a vignette study, it was found that the higher a potential customer’s self-importance of moral identity, the larger the distance between their willingness to buy from a firm that has high morality and low competence and their willingness to buy from a firm that has low morality and high competence. However, this moderation was not found for employees, suggesting that all employees prefer a moral over a competent firm regardless of the importance of morality to their identity. The research contributes to the stakeholder literature by combining insights from social psychology and stakeholder theory, examining moral identity and power in relation to the willingness to associate literature and further examining trade-offs and heterogeneity in stakeholder relationships.
INTRODUCTION
Recent research in stakeholder theory has acknowledged that human behaviour is not purely self-interested, but reciprocal (Bosse, Phillips & Harrison, 2009; Bridoux, Stofberg & den Hartog, 2016; Hahn, 2015). Primary stakeholders do not only care about what is good for them, but also about what is good for others (Bridoux, Stofber & den Hartog, 2016). As a result of this, a firm’s morality and competence interact in affecting primary stakeholders’ willingness to associate with (work for or buy from) a firm. The literature suggests that when there is a trade-off, morality is generally preferred over competence (van Prooijen &
Ellemers, 2015). However, is this still true when a primary stakeholder feels powerful? While both the literature on the relationship between corporate social responsibility (CSR) and willingness to associate with a firm as well as that on the relationship between power and morality are extensive, the two have not yet been brought together in a study. This is
remarkable, since power is a basic force in social relationships and is therefore likely to affect willingness to associate with a firm (Keltner, Gruenfeld & Anderson, 2003).
In this thesis, I develop and test a theory predicting that when morality and
competence are traded off, power makes primary stakeholders more willing to associate with a competent firm than with a moral firm. I hypothesize that since high-power individuals have increased self-interest and behavioural disinhibition, they prefer competence above morality when there is a trade-off between the two. Additionally, I propose that the moral identity of a stakeholder moderates the moderating effect of power on willingness to associate. Moreover, I test whether moral identity has a direct moderating effect: are individuals with a high self-importance of moral identity more likely to prefer morality above competence when there is a trade-off? These hypotheses are tested by means of a vignette study.
This study contributes to the existing literature in a number of ways. First, it
moral identity with insights from stakeholder theory on willingness to associate. It examines the concepts of power and moral identity in relation to willingness to associate. Second, this study contributes to the research in stakeholder theory on trade-offs. In particular, it further examines the trade-off between morality and competence, which has not yet received extensive attention in stakeholder theory. In doing this, it provides a link to the real world, where trade-offs are frequent. Fourth, it contributes to the body of research on heterogeneity in stakeholder relationships.
LITERATURE REVIEW
This chapter introduces current research on stakeholders that is relevant to the topic of this thesis.
Stakeholders and morality
Recent stakeholder literature has acknowledged that in general, stakeholders value morality. Morality is commonly defined as a normative code put forward by society, groups or
individuals regarding responsiveness to the needs and interests of others (Aquino, Freeman, Reed, Lim & Phelps, 2009; Lammers, Galinsky, Dubois & Rucker, 2015). Immoral behaviour occurs when individuals do not follow the existing norms and values in a certain context, while moral behaviour does conform to these (Reynolds & Ceranic, 2007). Morality is thus something that is perceived by someone who compares behaviour to the existing norm. Moreover, it is associated with other-regard, whereas immorality is associated with self-interest (Lammers et al., 2015).
Perceived morality of a firm is influenced by the firm’s corporate social responsibility (CSR) practices (Bridoux et al., 2016; Ellemers, Kingma, van de Burgt & Barreto, 2011; van Prooijen & Ellemers, 2015). Corporate social responsibility, which is sometimes addressed as corporate social performance (CSP), can be defined as the extent to which an organization is concerned with the impact of its activities on multiple stakeholders (Turban & Greening, 1997). CSR practices can be targeted at stakeholders within as well as outside of the firm (Aguilera, Rupp, Williams & Ganapathi, 2007). Examples could be initiatives to increase the wellbeing of employees, investments in schools and hospitals, attempts to reduce
environmental impact and initiatives to benefit minorities.
Stakeholder theory has argued that stakeholders are not only concerned with how firms treat them, but also with how others are treated by the firm (Bosse, Philips & Harrison,
2009; Bridoux, Stofberg & den Hartog, 2016). From the perspective of the stakeholder, firms can engage in two types of CSR practices: self-directed CSR, which involves CSR towards the stakeholder’s own group, and other-directed CSR, which is directed towards other stakeholders (Bridoux, Stofberg & den Hartog, 2016). Due to resource constraints, firms generally face a trade-off when it comes to stakeholder interests (Bridoux, Stofberg & den Hartog, 2016). It has been found that stakeholders do not only react to self-directed CSR, but also to other-directed CSR (Bridoux, Stofberg & den Hartog, 2016; Rupp, 2011).
These reactions can be explained by the concept of stakeholder reciprocity, which has been discussed elaborately in the literature (Bosse, Philips & Harrison, 2009; Hahn, 2015). Reciprocity entails that people react more positively to friendly actions, and more negatively to hostile actions than pure self-interest would suggest (Fehr & Gächter, 2000). The original assumption that humans are strictly self-interested and want to maximize their own utility in the first place thus needs to be relaxed to account for this (Bosse, Philips & Harrison, 2009). Reciprocity makes that people are boundedly self-interested in general, because the desire to maximize one’s own utility is bounded by norms of fairness (Bosse, Philips & Harrison, 2009). Reciprocal stakeholders wish to decrease outcomes for a firm if they perceive its behaviour as unfair or immoral (Bridoux, Stofberg & den Hartog, 2016; Hahn, 2015).
Reciprocal behaviour is based on and driven by the norms by which a stakeholder evaluates firm behaviour (Hahn, 2015). In other words, stakeholder reciprocity is elicited by moralistic evaluations of a firm (Quintelier et al., 2017). Quintelier et al. (2017) distinguish between moralistic and strategic evaluations by stakeholders. When relying on strategic evaluations, stakeholders assess the firm’s behaviour according to its use for them personally, rather than on the basis of moral norms (Quintelier et al., 2017). These strategic evaluations are triggered by a profit image of the firm (Quintelier et al., 2017). Moralistic evaluations are,
in turn, elicited by a stakeholder-image of a firm, which implies that a firm cares about the norms concerning the firm’s impact on stakeholders (Quintelier et al., 2017).
As a result of reciprocity, stakeholders value other-directed CSR even when it comes at a cost to themselves (Bridoux, Stoelhorst & den Hartog, 2016). However, it is important to note that stakeholder reactions to the treatment of others cannot be completely separated from self-interest (Rupp, 2011). Firm behaviour towards others can function as an indication of how stakeholders might be treated themselves (Rupp, 2006; Rupp, 2011). More specifically, CSR activities could serve as an indication for a fair treatment in the future (Rupp, 2011).
Morality and firm’s attractiveness to stakeholders
Because stakeholders are reciprocal, they generally prefer stakeholder-oriented firms above profit-oriented firms (Quintelier et al., 2017). That is, stakeholders prefer firms that are other-oriented and attempt to attend to all stakeholders’ needs and interests (Quintelier et al., 2017). For one thing, Mayer, Davis and Schoorman’s (1995) have argued that integrity, which can be linked to morality, influences organizational trustworthiness. As a result, engaging in
corporate social responsibility activities can contribute to firm attractiveness or stakeholders’ willingness to associate with a firm (Greening & Turban, 2000). Willingness to associate with a firm is commonly examined by looking at willingness to work for or buy from a firm
(Quintelier et al., 2017; van Prooijen & Ellemers, 2015). In other words, willingness to associate with a firm has been researched from the perspective of two types of stakeholders: (potential) employees and consumers. An alternative group of stakeholders of which
willingness to associate is examined less often is investors. This has to do with the fact that investors are less immediate stakeholders, having a smaller impact on firms’ stakeholder management (Aguilera et al., 2017). An additional more practical reason is that potential employees and consumers can be easily selected from one participant pool.
Applicants as well as current employees of a firm have been found to be concerned with the perceived morality of a firm (Ellemers et al., 2011; Rupp, 2006; Rupp, 2011). Turban and Greening (1997) argue that CSR activities contribute to a firm’s reputation and
attractiveness as an employer. An additional explanatory mechanism behind this stems from social identity theory, which claims that members of a group draw value for themselves from the characteristics of their group (Ellemers et al., 2011; Greening & Turban, 2000). Since morality can be a source of group value, the perceived morality of a firm influences how an individual, for example an employee, feels about belonging to the firm (Ellemers et al., 2011). In particular, Ellemers et al. (2011) found that perceived organizational morality has an effect on willingness to work for a firm because it contributes to a sense of pride in associating with the organization. Thus, not only morality itself but also the feeling of belonging to a moral group is valued by stakeholders. Additionally, perceived CSR activities of a firm are positively related to individual-level outcomes for employees such as job commitment and satisfaction (Rupp, 2006; Ellemers et al., 2011).
Furthermore, consumers also react to CSR activities of a firm (Ellen, Webb & Mohr, 2006; White, MacDonnell & Ellard, 2011). CSR activities can influence willingness to buy both indirectly, through consumer attitudes about the firm, and directly, through product purchase intentions (Sen & Bhattacharya, 2001). White, MacDonnell and Ellard (2011) state that consumers’ purchase intent of fair trade goods (a form of CSR practice) is influenced by the perceived potential to restore justice of that purchase. Research has further suggested that there is a positive relationship between CSR activities and consumer attitudes towards a firm (Murray & Vogel, 1997; Sen & Bhattacharya, 2001). However, evaluations of a company are influenced by the perceived congruence between consumer and company, as well as whether a consumer supports the CSR domain (Sen & Bhattacharya, 2001). Additionally, consumer responses are influenced by attributions about motives (Ellen et al., 2006; Quintelier et al.,
2017). Consumers do seem to realize that firms also engage in CSR practices because it is in their self-interest (for example, for their reputation) and not only because firms want to help third parties (Mohr, Webb and Harris, 2001).
Stakeholders and competence
Apart from valuing morality, stakeholders are also interested in a firm’s competence (van Prooijen & Ellemers, 2015). Competence has been connected to effectiveness in task
performance or economic achievements (van Prooijen & Ellemers, 2015). It is associated with organizational success and self-enhancement (Ellemers et al., 2011; Schwartz, 1992). For applicants and current employees, competence-related aspects of a firm include opportunities for personal and career development (van Prooijen & Ellemers, 2015).
The stakeholder literature has paid little explicit attention to the relationship between competence and willingness to associate. For example, Mayer, Davis and Schoorman (1995) have conceptualized ability as a dimension of organizational trustworthiness. Although ability can be linked to competence and trustworthiness influences willingness to associate, this is not a direct examination of the relationship (Mayer, Davis and Schoorman, 1995). However, it has been stated that stakeholders wish to maximize their own utility within the norms of fairness (Bosse, Philips and Harrison, 2009). As a result of this, stakeholders are likely to look for relationships with firms that are best suited to maximize utility. In this sense, valuing competence can be linked to self-interest.
As said before, stakeholder willingness to associate has been examined in the social identity literature with the concept of group value (Ellemers et al, 2011; Greening and Turban, 2000). Competence can be a source of group value as well (Ellemers et al., 2011; Leach, Ellemers & Barreto, 2007). According to Schwartz (1992), achievement is one of the personal values, meaning that it is a criterium by which people select, justify and evaluate actions.
Thus, similarly to the case of morality, stakeholders can be attracted to firms because their competence contributes to a positive social identity (van Prooijen & Ellemers, 2015).
Morality/Competence trade-offs
The stakeholder literature has not only paid relatively little attention to competence, but also to the fact that in real life, there is often a trade-off between morality and competence in stakeholder relationships (van Prooijen & Ellemers, 2015). For example, Mayer, Davis and Schoorman (1995) recognized that organizational trustworthiness depends on both the ability and integrity of the trustee, but paid little attention to trading off these dimensions. Due to resource constraints, a firm cannot invest unlimitedly in either perceived morality through CSR activities or in becoming more competent. However, it was discussed above that
willingness to associate with a firm depends on whether a firm is moral, as well as on whether a firm is competent.
More generally, the stakeholder literature has recognized the need for trade-offs, but contains little empirical research on them. In particular, balancing diverse and conflicting stakeholders’ interests has been a topic of research (Harrison & Freeman, 1999). One empirical research in this area is that of Reynolds, Schultz and Hekman (2006), which examined balancing the needs of different groups of stakeholders at the individual,
managerial level. Little is also known about stakeholder reactions to trade-offs. Auger, Burke, Devinney and Louviere (2003) found that consumers are willing to pay more if a product has ethical features. A later study added to this that consumers do value ethical features, but are unwilling to give up functionality for it (Auger, Devinney, Louviere & Burke, 2008). Folkes and Kamins (1999) found that superior product quality benefits firms involved in CSR practices more than unethically behaving firms.
Up to date, only van Prooijen and Ellemers (2015) examined stakeholder willingness to associate in the case of a trade-off between the morality and competence. They found that when firm morality and competence are traded off, stakeholders generally prefer morality above competence. Since both characteristics contribute to firm attractiveness, it is important to further study the relationship between the morality/competence trade-off and willingness to associate and conditions that may influence it. Thus, the question for further research is: is it always true that stakeholders prefer morality above competence when they consider if they want to associate with a firm? This thesis will examine the moderating effects of power and the self-importance of moral identity on this relationship: do primary stakeholders still prefer morality above competence if they feel powerful? And is this preference moderated by a highly self-important moral identity of a primary stakeholder?
Power as a potential moderator
In stakeholder theory, various scholars have accounted for individual-level heterogeneity when it comes to stakeholder behaviour (Greening & Turban, 2000). Likewise, DeCelles DeRue, Margolis and Ceranic (2012) have argued that (self-interested) behaviour is a function of both moral identity and power. In the existing literature, having power is commonly
defined as having superior access to valued resources in a social relationship, making it possible to punish others and modify their well-being by withholding these resources
(Galinsky, Magee, Inesi & Gruenfeld, 2006; Keltner, Gruenfeld & Anderson, 2003; Lammers et al., 2015). In other words, power can be seen as an individual’s ability to influence others (Galinsky, John & Keltner, 2002). Although related, the concept of power is not the same as that of status. According to Keltner, Gruenfeld and Anderson (2003, p. 6), status is “the evaluation of attributes that produces differences in respect and prominence, which in part
determines the allocation of resources within groups and by implication, each individual’s power”. Power does not automatically imply status and vice versa.
Another distinction that is of importance for research is that between the actual possession and the subjective feeling of power (Lammers & Stapel, 2009). The mere feeling that one is powerful is sufficient for people to think and act differently than they would have if they were to feel powerless (Anderson, John & Keltner, 2012; Galinsky et al., 2006; Lammers et al., 2015; Lammers & Stapel, 2009). In fact, an individual’s beliefs about their power shape their actual influence over others (Anderson, John and Keltner, 2012). These transformative effects on human behaviour and psychology make power an important concept to study. Moreover, power is a basic force in social relationships (Anderson, John & Keltner, 2012; Keltner, Gruenfeld & Anderson, 2003; Myllykangas, Kujala & Lehitma’ki, 2010). Thus, every stakeholder relationship deals with power to some extent, which makes having insights in its effects useful to understand both these and subsequent stakeholder behaviour.
A concept of interest in stakeholder theory that is related to power is bargaining power, which was defined as “the ability of one party to a contract to be able to influence the terms and conditions of that contract or subsequent contracts in its own favour” by Argyres and Liebeskind (p. 55, 1999). Thus, whereas power in a general sense relates to the ability to influence others, bargaining power is more specified to agreements. Coff (1999) stated that bargaining power for stakeholders is determined by 1) their capability for unified action, 2) accessibility to key information, 3) replacement costs to the firm and 4) switching costs for themselves. Together, these elements can give an employee more power when renegotiating the terms of employments for example. In his five forces model, Porter (1979) names bargaining power of suppliers and customers as one of the five factors influencing competition in an industry. More recent research in stakeholder theory has found that the
balance of bargaining power in a relationship influences stakeholder strategies (Frooman, 1999).
Social psychology has found that feeling powerful affects morality (Lammers et al., 2015). In particular, powerful people have been found to display behavioural disinhibition and have an increased focus on the self, instead of on others (Galinksy et al., 2006; Lammers et al., 2015). This indicates that feeling powerful might also influence the trade-off between morality, which is linked to other-regard, and competence, which is linked to self-interest, that stakeholders face when deciding to associate with a firm.
Moral identity as a potential moderator on willingness to associate and power effects As DeCelles et al. (2012) name moral identity as the other determinant of behaviour besides power, moral identity is another interesting concept to examine in relation to the
morality/competence trade-off and stakeholder willingness to associate. Shao, Aquino and Freeman (2008) have identified two perspectives on the conceptualization of moral identity: the character perspective and the social-cognitive perspective. The former is associated with the work of Blasi (1983; 1984; 2005) on the Self Model. This model holds that apart from making a moral judgement, people also make a judgement of responsibility when
contemplating the moral way to act (Blasi, 1984). This means that they consider whether they are responsible to act on their judgement (Blasi, 1984). Moreover, the Self Model states that the criteria for making moral judgements are the result of a person’s moral identity, the degree to which being moral is central to the sense of self (Aquino and Reed, 2002; Blasi, 1984). According to the character perspective, moral identity is a component of the self and something that is consistent over time (Shao, Aquino & Freeman (2008). However,
limitations of the perspective have prompted some researchers towards the social-cognitive conceptualization of moral identity (Shao, Aquino & Freeman, 2008). This perspective
addresses behavioural inconsistencies and automated moral behaviour, which remain unexplained by the character perspective, by adopting theories from social cognition. For example, scholars from this perspective have argued that the moral self-concept can be activated and deactivated by situational cues (Aquino & Reed, 2002; Aquino et al., 2008 in Shao, Aquino & Feeman, 2008; Markus & Kunda, 1986). Moreover, moral identity is seen as dependent on knowledge accessibility (Aquino et al., 2008 in Shao, Aquino & Freeman, 2008).
Bringing together elements of both perspectives, Aquino and Reed (2002, p. 1424) define moral identity as “a self-conception organized around a set of moral traits”. Their definition implies that moral identity is conceptualized by traits, as well as linked to a social construction of what a moral person is like (Aquino & Reed, 2002). According to Aquino and Reed (2002) one’s moral identity can be invoked by tapping into a subset of moral traits. In addition to this, Erikson (1964) theorized that an identity has two dimensions: a public self and a private self. Being one of many possible social identities, this distinction holds for moral identity as well (Aquino & Reed, 2002). The degree to which moral traits are reflected in public is referred to as symbolization (Aquino & Reed, 2002). This dimension reflects a sensitivity one has for conveying an image of having moral characteristics to the world, since it is about expressing commitment to moral traits in a social context (Aquino & Reed, 2002). On the other hand, internalization is the private dimension or the degree to which moral traits are related to the self-concept (Aquino & Reed, 2002).
In sum, moral identity is the extent to which an individual places importance on morality (DeCelles et al, 2012). In general, research has found that a high self-importance of moral identity leads to reduced self-interest, moral behaviour and appreciation of morality (Aquino & Reed, 2002; Hardy, 2006; Reed & Aquino, 2003; Reynolds and Ceranic, 2007). These effects indicate that the self-importance of moral identity is also likely to influence the
trade-off between morality and competence that stakeholders face when deciding to associate with a firm. Furthermore, following the argument of DeCelles et al. (2012) that
(self-interested) behaviour is a function of both moral identity and power, the effect of power on the relationship between the morality/competence trade-off and willingness to associate could be moderated by the moral identity of a primary stakeholder. This because the self-importance of moral identity differs across individuals and thus can be expected to lead to varying
reactions of stakeholders who feel powerful and face a trade-off between morality and competence. Examining the role of moral identity will contribute to the research on individual-level heterogeneity in stakeholder theory.
Research gap and research questions
In sum, stakeholders value both morality and competence. The latter, as well as trade-offs between the two, have not yet been researched extensively. Only one research has examined this trade-off and found that in general, stakeholders prefer morality above competence when there is a trade-off (van Prooijen & Ellemers, 2015). Specific conditions under which this outcome holds and individual-level differences have not been studied yet.
This thesis is concerned with three questions. First, what is the effect of feeling powerful on primary stakeholder willingness to associate with a firm when there is a morality/competence trade-off? Second, how does the self-importance of moral identity influence stakeholder willingness to associate with a firm in the case of a
morality/competence trade-off? Third, the research will test whether this effect is moderated by the moral identity of primary stakeholders. The research can be conceptualized as depicted in figure 1.
Figure 1: conceptual model
Morality/ Competence
trade-off
Willingness to associate with a firm
Power
THEORETICAL FRAMEWORK
This chapter will introduce the hypotheses of this thesis.
Power and morality
In social psychology, the alleged corruptive influence of power has been discussed
elaborately (Kipnis, 1972). Over the years, research has found that sense of power indeed has a transformative effect on human behaviour and psychology (Galinsky et al., 2006; Lammers et al., 2015; Lammers & Stapel, 2009;). Negative as well as positive behaviour has been associated with this effect. On the one hand, power has been found to increase stereotyping and dehuminazation (Fiske, 1993; Lammers & Stapel, 2011). On the other hand, power has been found to increase interpersonal sensitivity (Schmid Mast, Jonas & Hall, 2009). Below, I will discuss transformative effects of power and their relationship to morality.
Power and behavioural disinhibition
Keltner, Gruenfeld and Anderson (2003) formulated the approach/inhibition theory of power. This theory proposes that possessing power influences the relative activation of two broad and fundamental behavioural systems: the behavioural approach and inhibition systems. The former is concerned with regulating behaviour related to rewards and opportunities, while the latter is concerned with affective states as a result of threats: it works as an alarm system. High power activates the approach system because powerful people have increased access to rewards and are less likely to feel hindered by others in pursuing these rewards. Low power activates the inhibition system, as low power individuals are more subject to threats and punishment. They are more sensitive to the evaluations of others and potential consequences. As a result of activating the behavioural approach system over the inhibition system, power makes people more attentive to rewards than to social concerns. This preference is associated
with automatic social cognition instead of deliberative reasoning. Moreover, it makes powerful people less sensitive to social inhibitions and thus making them more likely to display disinhibited social behaviour.
The theory proposed by Keltner, Gruenfeld and Anderson was empirically validated by the work of Anderson and Berdahl (2002), who tested its hypotheses in two studies. Elaborating further on the approach/inhibition theory, Anderson and Galinsky (2006) found empirical evidence that powerful people perceive risks with more optimism and are more likely to take risks. As a potential conclusion of this, they suggest that their immoral behaviour might not be the result of the corruptive influence of power per se, but of its disinhibiting characteristics. People that feel powerful are more likely to engage in immoral behaviour because are more likely to ignore the social inhibitions against these behaviours (Lammer et al., 2015). In addition to this, disinhibition tends to reduce negative emotions associated with immoral behaviour (Lammers et al., 2015).
Moreover, on the basis of the approach/inhibition theory of power, Van Kleef, Oveis, van der Lowe, LuoKogan, Goetz and Keltner (2008) found empirical evidence that elevated social power is associated with diminished reciprocal emotional responses (distress) and diminished compassion to another person’s suffering. As the most plausible explanation for this, they suggest that high power individuals are less dependent on other people and therefore less motivated to respond to others. Likewise, Galinksy, Magee, Gruenfeld, Whitson and Liljenguist (2008) found high power individuals to be less influenced by social cues such as reputation. A lack of concern for the emotions of others can encourage immoral behaviour.
In contrast, power-related disinhibition has also been argued to influence morality positively (Lammers et al., 2015). Inhibitions to express disapproval when witnessing immoral behaviour may be overcome. The overcoming of reservations to disrupt social interactions and connections makes powerful people judge more strictly than powerless
people (Lammers & Stapel, 2009). In some experiments, powerful people were even found to be hypocrite (Lammers, Stapel & Galinsky, 2010).Moreover, pressure by others to make immoral decisions can also be overcome when feeling powerful (Lammers et al., 2015). Finally, effects of power-related disinhibition may differ across individuals, depending on their innate tendencies towards moral behaviour (Chen, Lee-Chai & Bargh, 2001; DeCelles et al., 2012; Lammers et al., 2015; Magee and Smith, 2013).
Power and heightened self-focus
Besides behavioural disinhibition, Lammers et al. (2015) describe heightened focus on self-interest as another major transformative effect of power. An example is that high power individuals tend to spend more on themselves than on others (Rucker, Dubois & Galinksy, 2011). Keltner, Gruenfeld & Anderson, (2003) give the increased sensitivity of powerful people to rewards over threats as an explanation for this effect. High power individuals assess what others can mean for them in achieving rewards and therefore view others through a lens of self-interest (Keltner, Gruenfeld & Anderson, 2003). In contrast, low power individuals see themselves through a lens of others’ interest, as they are more sensitive to threats (Keltner, Gruenfeld & Anderson, 2003). Over the years, there has been much empirical evidence in the social psychology literature supporting this line of reasoning. For example, Kipnis (1972) found that institutional power holders guide their actions towards others by their own goals and Gruenfeld, Inesi, Magee and Galinsky (2008) found that high power increases seeing others as instrumental.
In itself, heightened self-focus as a result of power has a number of consequences as well. Firstly, it can cause powerful people to perceive themselves as having greater value to society than others (Rucker, Dubois and Galinsky, 2011). In addition to this, it can create social distance from others (Lammers, Galinsky, Gordijn & den Otten, 2012; Magee & Smith,
2013). The self-sufficiency associated with power, which comes from the powerful not needing to rely on others and feeling that as well, increases the tendency to social distance (Lammers, et al., 2012). In other words, asymmetry in dependence on others raises
asymmetry in social distance (Magee & Smith, 2013). Lastly, heightened self-focus as a result of power reduces one’s ability to take perspective (Galinsky et al., 2006).
Through heightened self-focus and its effects, power can undermine morality. Placing the self above others can make powerful people care less about the suffering of others, as long as they are not harmed themselves (Van Kleef et al., 2008). This indifference is related to reduced perspective taking, which makes that people cannot even imagine the harm done to others (Galinsky et al., 2006). Reduced perspective taking and making people more closed (i.e., socially distant) to others are psychological processes that play a role in dehumanization as well (Lammers & Stapel, 2011). Dehumanization helps justify decisions harmful to others, and has been linked to power in previous research (Lammers & Stapel, 2011). Furthermore, using others instrumentally can be seen as immoral.
On the other hand, the literature describes reasons why heightened self-focus might encourage high power individuals to act morally. Dubois, Rucker and Galinsky (2015) state that powerful people might be more selfish, but not more immoral per se. Their research found that powerful people are more likely to engage in unethical behaviour when it benefits them, but not when that behaviour harms others. Relatedly, powerful people may facilitate moral behaviour when this is necessary in order to get the wished rewards (Lammers et al., 2015). In addition to this, because powerful people are more focused on themselves, they may be less attentive to immoral stimulants from others (Galinksy et al., 2008 in Lammers et al., 2015). This is related to the previously discussed relieved pressure to conform to social pressures.
Effects of power and willingness to associate
In sum, two major effects of power have been identified: 1) disinhibition and 2) heightened self-focus, which in turn can lead to estimating oneself over others, increased social distance and difficulties in perspective taking. These effects can undermine morality, but there have also been accounts of disinhibition and heightened self-focus having a positive effect on moral behaviour. The question is: what do these effects mean for stakeholder willingness to associate?
In general, stakeholders care for both morality and competence when deciding whether they want to associate with a firm, but prefer morality above competence (Van Prooijen & Ellemers, 2015). Feeling powerful may alter this preference for potential employees and customers. As a result of feeling powerful, they may overcome social inhibitions for both competence and morality. That is, they are more likely to ignore social opinions against or in favour of associating with a certain type of firm. Given that power creates heightened self-focus, it can cause potential employees and customers to focus more on what can be done for oneself than what can be done for others. Thus, the above discussed arguments suggest that CSR-practices directed to others may become less important to stakeholders that feel powerful. This effect is strengthened by the increased social distance and difficulties in
perspective-taking that powerful stakeholders will have. Valuing self-regard over other-regard can be linked to choosing competence over morality. Based on this line of reasoning, I
propose:
Hypothesis 1: The higher stakeholders’ sense of power, the smaller the positive difference in willingness to associate between a firm with a high morality and low competence trade-off and a firm with a low morality and high competence trade-off.
Moral identity as a potential moderator of power effects
In the section above, I have discussed that power-related disinhibition may have different effects across individuals regarding morality (Chen, Lee-Chai & Bargh, 2001; DeCelles et al., 2012; Lammers et al., 2015; Magee and Smith, 2013). In addition to this, there have also been accounts of a reduced focus on self-interest as a result of power (Chen, Lee-Chai & Bargh, 2001; Decelles et al., 2012). Chen, Lee-Chai and Bargh (2001) state that while some individuals may link power to self-interest, others will link it to social responsibility goals (communals vs. exchangers). In other words, the transformative effects of power and resulting behaviour are not homogeneous across people.
As one explanation for this, DeCelles et al. (2012) state that power might not directly impact behaviour, but might rather activate individuals’ underlying traits or attributes. This corresponds with the statement of Galinksy et al. (2008) that power heightens responses to internal states, while reducing the strength of a situation. In fact, high power individuals have been found to act more in line with their dispositions and values than low power individuals (Guinote, 2008; Magee and Smith, 2013). This because they are more guided by why to act than by how to act (Magee and Smith, 2013). As a particular trait that nuances the relationship between power and morality, DeCelles et al. (2012) bring forth the self-importance of moral identity. More specifically, they argue that self-interested behaviour is a function of both power and moral identity.
Effects of moral identity, power and willingness to associate
Reynolds and Ceranic (2007) have argued that together with moral judgement, moral identity influences moral behaviour. Likewise, Aquino and Reed (2002) state that the self-importance of moral identity is a predictor of moral cognition and behaviour. According to Shao, Aquino and Freeman (2008) there are a number of ways in which moral identity interacts with
behaviour. Firstly, research has shown a positive link between moral identity and pro-social actions (Aquino & Reed, 2002; Hardy, 2006; Reed & Aquino, 2003; Reynolds and Ceranic, 2007). Aquino and Reed (2002) found that individuals with greater self-importance on morality were more likely to have engaged in pro-social behaviour in the past, such as
volunteering at a homeless shelter. Moreover, participants with high self-importance on moral identity donated more to a food drive in the present (Aquino & Reed, 2002). Adding to these findings, research has also found that moral identity is negatively related to anti-social behaviour (Barriga, Morrison, Liau & Gibbs, 2001; Sage, Kavussanu & Duda, 2006). As the driving force that makes moral identity result in certain moral behaviour, Blasi (1984)
suggests the desire for self-consistency. A person with a certain self-definition is likely to feel obliged to act according to it (Blasi, 1984). Thus, a person with a highly self-important moral identity is likely to act morally. As another interaction, research in the social-cognitive perspective has suggested that temporarily increasing the accessibility of participants’ moral identities has a positive effect on willingness to participate in pro-social actions (Reed et al., 2007). Finally, moral identity has been found to affect behavioural mediators, in particular moral disengagement (Detert, Trevino & Sweitzer, 2008; Aquino et al., 2007)
As said before, DeCelles et al. (2012) have argued that self-interested behaviour, which has been opposed to moral behaviour, is a function of both power and moral identity. Thus, not only power, but also moral identity influences the focus of individuals on one’s own needs (Aquino et al., 2009; Reed & Aquino, 2003). Reed and Aquino (2003) argue that since moral identity is a social construct, it not only defines the self’s relation to others but also the psychological boundaries between out-groups and in-groups. Individuals with a high self-importance of moral identity, particularly those scoring high on the internalization dimension, are more likely to include strangers within their circle of moral regard (Reed & Aquino, 2003). That is, they feel obliged to morally consider (the needs of) a larger group of others. In
addition to this, individuals with a high self-importance of moral identity are less likely to accept harm to out-group members (Reed & Aquino, 2003). Explaining the relationship between moral identity and self-focus further, Aquino et al. (2009) have suggested an incompatibility between moral goals and values and self-interested goals and values. This gives a conflict when both one’s moral identity and a self-interested aspect of one’s identity are activated (Aquino et al., 2009). In general, the identity activated by situational factors then wins, which means that even if a person’s moral identity is highly self-important he or she can display self-interested behaviour if the situation provokes this (Aquino et al., 2009). This confirms the idea of DeCelles et al. (2012) that moral identity in itself is not enough to explain behaviour. According to them, it is complemented by power, which is situational as well.
In sum, most of the research has found that a highly self-important moral identity has a negative influence on self-interested behaviour. A stakeholder with a high self-importance of moral identity is likely to engage in and support pro-social behaviour (Aquino & Reed, 2002; Hardy, 2006; Reed & Aquino, 2003; Reynolds and Ceranic, 2007). In other words, for
stakeholders deciding to associate with a firm that have a high self-importance of moral identity, other regard and thus morality is of great value. As a result, I expect:
Hypothesis 2: The higher stakeholders’ self-importance of moral identity, the larger the positive difference in willingness to associate between a firm with a high morality and low competence trade-off and a firm with a low morality and high competence trade-off.
I expect this effect to present through both the symbolization and the internalization dimension. On the one hand, a preference for morality over competence when deciding on willingness to associate can be due to an intrinsic value (internalization). On the other hand,
willingness to buy from or work from a firm can reflect a message to the outside world (symbolization).
DeCelles et al. (2012) connect the effect of moral identity on self-interested behaviour to the concept of moral awareness. Reynolds (2006) has defined this as “a person’s
determination that a situation contains moral content and legitimately can be considered from a moral point of view” (p. 233) and argued that moral awareness differs across individuals. Moral awareness is negatively related to self-interested behaviour as a lack of recognition of a moral issue hinders initiation of the moral decision-making process (Reynolds, 2006). As one individual characteristic explaining heterogeneity in moral awareness DeCelles et al. (2012) bring forth moral identity, stating that an individual with a highly self-important moral identity has greater moral awareness. Moreover, they argue that this positive effect is enhanced by sense of power. Research on power has shown that when a person feels powerful, his or her individual traits increase availability of certain cognitive schemes and influence information processing (Chen, Lee-Chai & Bargh, 2001; Galinksy et al., 2003). An individual with a highly self-important moral identity will already have increased cognitive access to moral concepts, but this effect will thus be strengthened when feeling powerful. Reed et al. (2007) have suggested that temporarily increasing this accessibility has a positive effect on willingness to participate in moral behaviour. Conversely, a powerful person with a less self-important moral identity will be less aware of the moral implications of his or her actions (DeCelles et al., 2012).
On the basis of the arguments above, it can be expected that the self-importance of moral identity is heterogeneous across individuals. Therefore, varying reactions of
stakeholders who feel powerful and face a trade-off between morality and competence can be expected, depending on their moral identity. Following the line of reasoning proposed by
DeCelles et al. (2012), one’s moral identity gives an explanation for the heterogeneity of power effects. Therefore, I propose that:
Hypothesis 3: The self-importance of moral identity has a moderating effect on the moderating effect of power in the relationship between the morality/competence trade-off and willingness to associate.
METHODOLOGY
This chapter will explain the research approach and design of this paper.
Research procedure
The data for this study was collected in the lab of the Amsterdam Business School as part of a joint data collecting effort. The complete research design therefore contains elements that contribute to other studies. After describing the entire procedure of the lab sessions, I will discuss the research elements relevant to this thesis more elaborately.
Respondents started with participating in an interactive public goods game. After completing this, they were redirected to a Qualtrics questionnaire in which the present study was
embedded. This survey consisted of the following elements:
1. Intentions and fairness perceptions of the participant in the public goods game.
2. Demographics (age, gender, nationality, study field, study year and university)
3. Individual differences that would moderate in the experiments (communal and exchange orientation, propensity to trust, need for power, hexaco, moral identity, moral foundations, social value orientations)
4. Neutralizing component
5. Vignette based experiment
6. Questions company Alpha
7. Measure of the generalized sense of power
Used for this study were the elements in bold: the demographics, individual differences in self-importance of moral identity, the vignette experiment and the measure of generalized
sense of power. The vignette and parts of the questionnaire relevant to this study can be found in Appendix B, C, D and E.
Research methods
This study sought to examine stakeholder willingness to associate with a firm under varying conditions of morality and competence of that firm. Following prior research on this topic, we chose to do this with the help of vignettes (Bridoux, Stofberg & den Hartog, 2016; van
Prooijen & Ellemers, 2015). Vignettes can be used to explore normative issues in very specific situations (Alexander & Becker, 1978; Finch, 1987). By describing specific hypothetical situations, the context of survey questions is standardized across respondents (Alexander & Becker, 1978). Moreover, the possibility to alter details of the described
situation for different respondent groups allows for the manipulation of independent variables, while controlling for other factors (Aguinis & Bradley, 2014). Hence, this study made use of vignettes, embedded in a questionnaire. The used vignettes were inspired by the vignettes in the studies of Bridoux, Stofberg and den Hartog (2016) and van Prooijen and Ellemers (2015). The design of the research was experimental and between-subjects, with participants randomly assigned to conditions. The study was carried out with approval of the Ethics Committee Economics and Business (University of Amsterdam), which can be found in Appendix A.
Sample
I used a sample of 408 completed questionnaires for this research. Following Bridoux, Stofberg and den Hartog (2016), all participants were students, 36,6% male and 63,4% female. This had the advantage that we could select participants for both job search and purchase intent from the same pool, which helps increase comparability across the two. Most
were studying at the University of Amsterdam (92.4%), which is where the data was collected, and were students in the field of Business and/or Economics (59.3%). The conditions were divided relatively equally among respondents.
Vignettes
The objective of this research is to explain stakeholders’ willingness to associate with a firm. In line with prior research, two groups of stakeholders were included in the study, namely prospective employees and prospective customers, andwillingness to associate with a firm was conceptualized as both willingness to work for a firm and willingness to buy from a firm, or in other words job pursuit intent and purchase intent (Bridoux, Stofberg & Den Hartog, 2016; van Prooijen & Ellemers, 2015; Quintelier et al., 2017). Respondents were asked to imagine themselves as a potential future employee or customer of company ABC. As a reason for choosing a consumer electronics manufacturer, Bridoux, Stofberg and den Hartog (2016) state that this is a relevant situation for the participant pool, students.
For the employee condition, the vignette started as follows: “In this study we are interested in students’ preferences for prospective employers. Please read the following description carefully: Imagine that you have (almost) finished your studies and that you are looking for a job. Company ABC is a manufacturer of consumer electronics, such as
television sets, DVD players and digital cameras. It has job openings in your area of expertise. These job openings seem attractive and you want to know more about Company ABC. A reputable research organization has recently investigated companies in the consumer electronics industry.”
For the customer condition, the vignette was introduced by the following text: “In this study we are interested in students’ preferences for companies to buy from. Please read the following description carefully: Company ABC is a manufacturer of consumer electronics,
such as television sets, DVD players and digital cameras. You are considering buying a product from Company ABC. A reputable research organization has recently investigated companies in the consumer electronics industry.”
The independent variable of this study is the morality/competence trade-off. After either the employee or customer condition, participants were shown a description of company ABC’s morality and competence. The vignette refers to information coming from a reputable research organization to avoid respondents doubting the credibility of the information
provided (Bridoux, Stofberg & den Hartog, 2016). The characteristics of company ABC followed the vignettes of Van Prooijen and Ellemers (2015) and were based on Mayer, Davis and Schoorman’s (1995) descriptions of the dimensions of perceived trustworthiness. Their definition of the integrity dimension gave the basis for moral characteristics, such as adhering to a set of principles and having a strong sense of justice (Mayer, Davis & Schoorman, 1995). Ability, which was defined as “that group of skills, competencies and characteristics that enable a party to have influence within some specific domain” (Mayer, Davis & Schoorman, p. 10), was linked to competence characteristics. In writing the vignettes, it was considered that participants would not perceive extremities in morality or competence as realistic (Bridoux, Stofberg & den Hartog, 2016).
The research included two conditions for the morality/competence trade-off: 1) low morality and high competence and 2) high morality and low competence. The low morality and high competence character of company ABC was described as follows in the vignette: “Its report shows that Company ABC is not a very principled company. It sometimes uses dishonest advertising campaigns, regularly breaks the promises it makes to employees, and is not very transparent in its annual report. It seems that the company’s actions do not reflect a strong sense of justice and are not primarily guided by what is right or wrong. The agency also reports that Company ABC is very effective at conducting its business. It has experienced
leaders, deep knowledge of the industry, and is very capable of producing and marketing its products. It seems likely that Company ABC will be commercially successful in the years to come.”
For the second condition, high morality and low competence, the vignette continued with the following text: “Its report shows that Company ABC is a very principled company. It uses honest advertising campaigns, fulfils the promises it makes to employees, and is
transparent in its annual report. It seems that the company’s actions reflect a strong sense of justice and are primarily guided by what is right or wrong. The agency also reports that Company ABC is not very effective at conducting its business. It has relatively inexperienced leaders, lacks deep knowledge of the industry, and is not always very capable of producing and marketing its products. It seems likely that Company ABC may struggle to be
commercially successful in the years to come.” In total, this means that four different
vignettes were shown to participants. We counterbalanced the presentation of the competence and morality information. Thus, half the respondents read the information about competence before morality, the other half reversed. Two realism checks with a 7-point Likert scale were included: “I think there are companies like this in real life” and “I had no difficulty
imagining myself in this situation”.
Dependent variable: willingness to associate
The dependent variable of this research is willingness to associate with a firm, which is operationalized as job pursuit intention for potential employees and purchase intent for customers. For the employee condition, a four-item scale was adopted from Greening and Turban (2000) to measure job pursuit intention, including the following statements: “I would put in a great deal of effort to work for ABC; I would be interested in pursuing a job
offer from ABC.” For the customer condition, a four-item scale was adopted from White, MacDonnell and Ellard (2012) to measure purchase intention. The following statements were included: “I would be willing to buy a product from ABC; I would exert a great deal of effort to purchase a product from ABC; I would likely make ABC one of my first choices in
consumer goods electronics; I would be likely to purchase a product from ABC.”
Respondents were asked to assess these statements on a Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree).
Moderating variables
To examine the moderating effect of power on the relationship between the
morality/competence trade-off and willingness to associate with a firm, general sense of power was measured. To establish the general sense of power, we used seven scale items that were based Anderson and Galinksy (2006) and Anderson, John and Keltner (2012). One example of an item is: “I can get others to do what I want.”
Additionally, self-importance of moral identity was measured. In order to activate respondents’ moral identity, they were asked to visualize a person that is caring,
compassionate, faire, friendly, generous, helpful, hardworking, honest and kind, before responding to ten statements using a 7-point Likert scale (1 = strongly disagree, 7 = strongly agree). The first five relate to internalization, which is the degree to which the moral traits are central to the self-concept (Aquino & Reed, 2002). An example is: “It would make me feel good to be a person who has these characteristics.” The other statements concern
symbolization, the degree to which the traits are reflected in the respondent’s actions in the world (Aquino & Reed, 2002). For example, one item was “I am actively involved in activities that communicate to other that I have these characteristics.”
Control variables
A number of control variables were included in this research, of which most were demographics (age, gender, nationality, study field, study year and university). Previous research has related gender to moral behaviour (Barriga et al., 2001; O’Fallon & Butterfield, 2005). Moreover, we wished to control nationality as culture may have an influence on one’s morality perspectives (Haidt, Koller & Dias, 1993). Age, study field, year and university were included because these variables may impact job pursuit intent. Apart from demographics we controlled for whether the respondent had already found a job for after graduation and
whether the respondent was interested in working for a consumer goods company, which might influence responses for the employee condition.
RESULTS
In this chapter, the results of this research will be reported.
Reliability analyses of scales
To assess the used measures, I first performed tests for reliability of scale. Cronbach’s Alpha’s are reported in table 1. The items on the scale for moral identity together had a Cronbach’s Alpha of 0.725. One problematic item might be the third, which is the only item with a corrected item – total correlations value of 0.078. Deleting the item would raise Cronbach’s Alpha to 0.742. The corresponding statement was ‘I would be ashamed to be a person who has these characteristics’. Apart from being reverse-coded, this item describes the complex concept of shame, which reflects a very strong negative emotion about
something (Lutwak, Ferrari & Cheek, 1998). This addition of another emotion may have been problematic for the item. However, since the change in Cronbach’s Alpha was not very substantial, I decided not to delete the item.
The job pursuit intent and purchase intent scales had high reliability, with Cronbach’s Alpha respectively 0.938 and 0.907. All corrected item - total correlations were above 0.30, indicating that all items have a good correlation with the total score of the scale. None of the items would raise reliability substantially if deleted. Sense of power had high reliability, with Cronbach’s Alpha 0.809, but one item with a corrected item – total correlations of 0.275. This item was the statement ‘my wishes do not carry much weight’, a reverse-coded item. Deleting it would increase reliability to a Cronbach’s Alpha of 0.829, which I judged as not substantial enough to remove the item.
Manipulation and realism checks
To determine the effect of the morality/competence manipulations, I conducted analyses of variance (ANOVA) for both the customer and employee conditions. These analyses show that the manipulated morality and competence characteristics of Company ABC significantly affected willingness to join (F = 33.879, p < 0.001) and willingness to buy (F = 62.784, p < 0.001). As for whether the vignettes were perceived as realistic, the mean of the realism checks across the entire sample was 5.25 on a seven-point scale. Perceived realism varied across conditions, with the employee condition having a mean of 5.139 and the customer condition having a mean of 5.313. However, an analysis of variance indicated that this difference was not significant. The low morality and high competence scenario had a realism mean of 5.668, while the high morality and low competence condition had a lower mean of 4.780. This difference was in fact significant with F = 68.336 and p < 0.001.
Descriptives:
Outliers of the sample were excluded for all averages except job pursuit and purchase intent (moral identity, symbolic moral identity, internal moral identity, sense of power). In total, this meant that 10 were excluded. Table 1 and 2 present descriptive statistics and correlations. Job pursuit and purchase intention had relatively more negative kurtosis than other variables, indicating a flatter distribution than normal. Sense of power had the highest positive skewness, meaning the most asymmetrical distribution. In line with previous research, a significant positive correlation can be seen between a high morality and low competence trade-off of a firm and both purchase intent and job pursuit intent (van Prooijen & Ellemers, 2015).
Table 1: descriptive statistics and correlations.
Variable N Min Max Mean SD Skewness Kurtosis 1 2 3 4 5 6 7 8
1 Purchase intention 200 1 7 3.7075 1.4116 0.049 -0.753 (0.907)
2 Job pursuit intention 198 1 7 4.0265 1.583 -0.179 -0.93 (0.938)
3 Moral identity 398 1 7 5.1555 0.6858 -0.251 -0.001 -0.025 -0.044 (0.725)
4 Moral identity - Symbolization 398 1 7 4.3899 0.9646 -0.259 -0.076 -0.026 0.007 0.85**
5 Moral identity - Internalization 398 1 7 5.9211 0.75 -0.652 -0.125 -0.012 -0.087 0.735** 0.268**
6 Sense of power 398 1 7 3.021 0.8138 0.399 0.268 0.046 -0.113 -0.165** -0.121* -0.146** (0.809)
7 Vig employee 0 1 0.015 0.029 -0.009 -0.015
8 Vig high morality and low competence 0 1 0.490** 0.384** -0.092 -0.135 0.004 -0.076 0.015
9 Female 398 0 1 -0.023 -0.05 0.147** 0.133** 0.098 -0.034 -0.014 -0.066
10 Age 398 0 1 -0.016 -0.02 0.01 -0.039 0.069 0.055 -0.064 0.01
11 Dutch 398 0 1 -0.111 -0.118 -0.059 -0.073 -0.013 0.06 0.025 -0.019
12 Business and Economics 398 0 1 -0.046 0.063 0.015 -0.003 0.032 -0.056 0.047 -0.055
13 Interest in consumer goods company 198 0 1 0.348** 0.04 0.068 -0.011 -0.03 -0.032 -0.032
14 Found job 398 0 1 -0.075 0.057 0.074 0.008 -0.06 -0.018 -0.097
15 University of Amsterdam 398 0 1 -0.045 -0.109 -0.32 -0.06 0.018 0.079 -0.078 -0.078
16 MSc 398 0 1 0.001 0.049 0.086 0.064 0.074 -0.046 -0.081 -0.071
*p < 0.05 **p < 0.01
Table 2: descriptive statistics and correlations (continued).
Variable 9 10 11 12 13 14 15
1 Purchase intention 2 Job pursuit intention 3 Moral identity
4 Moral identity - Symbolization 5 Moral identity - Internalization 6 Sense of power
7 Vig employee
8 Vig high morality and low competence 9 Female
10 Age 0.068
11 Dutch -0.117* 0.015
12 Business and Economics -0.079 -0.39 -0.199**
13 Interest in consumer goods company 0.66 -0.92 -0.248** 0.362**
14 Found job -0.015 0.307** -0.022 0.028 0.088
15 University of Amsterdam 0.059 -0.044 -0.093 0.17** 0.086 -0.108*
16 MSc 0.196 0.377 -0.148 0.026 0.05 0.107 0.087
*p < 0.05 **p < 0.01
Cells are empty when variables relate to different set of vignettes
The correlation is higher for purchase intention (0.490) than for job intention (0.384), suggesting a stronger relationship between willingness to buy from a firm and its morality/competence trade-off than between willingness to join a firm and its
morality/competence trade-off. In general, women scored higher on self-importance of moral identity and self-importance of symbolic moral identity than men, which is in line with previous research (e.g. Decelles et al., 2012). In addition to this, job pursuit intention and the control variable interest in a consumer goods company were positively correlated (0.348). Interest in a consumer goods company was also positively correlated with respondents being Dutch and studying in the field Business and Economics. Moreover, sense of power was significantly and positively correlated with moral identity (0.165).
Hypothesis testing
I tested hypothesis 1, which predicts that the higher stakeholder’s sense of power, the smaller the difference in willingness to associate between a firm with a high morality and low
competence and a firm with a low morality and high competence trade-off, by means of the PROCESS application (model 1) of Hayes (2012). The results of this analysis for both the customer and employee condition can be found in table 3 under model 1. While the main effect of the trade-off on willingness to buy was significant, there were no significant interaction effects (trade-off x power) for the customer or employee condition. Thus, the moderation of power on willingness to associate (purchase intent and job pursuit) does not take place. Therefore, hypothesis 1 is not supported.
Hypothesis 2 proposes that the higher self-importance of stakeholder’s moral identity, the larger the difference in willingness to associate between a firm with high morality and low competence and a firm with low morality and high competence. I used the PROCESS
application of Hayes (2012) (model 1) to investigate this hypothesis. The results of this are in table 3 under model 2a, 2b and 2c. Model 2a tested the moderating effect of moral identity as a whole, model 2b that of the symbolization component of moral identity and model 2c that of the internalization component of moral identity. No significant effects were found for the employee condition, thereby hypothesis 2 is not supported for potential employees. Although the main effects were not significant, all of the interaction effects were positive and
significant in the customer condition. The three models respectively explained 29.9%, 28.68% and 29.07% of the variance in purchase intent. For moral identity as a whole, the interaction effect (trade-off x moral identity) was 0.6909 and significant for a level of confidence of 95% with p = 0.0061 < 0.01. In other words, a high morality and low competence condition (trade-off = 1) causes purchase intent to increase with 0.6909 for an increase of 1unit in moral identity. For symbolization, the interaction effect was 0.3684 with p = 0.0394 < 0.05.
Internalization had a higher interaction term than symbolization, 0.5663, and was significant for a level of confidence of 95% with p = 0.0158 < 0.05. When probing the interactions, I found that the effects are significant at every value of moral identity. To better understand the nature of the interactive effects, I graphed them in Figures 2, 3 and 4. The graphs indicate that the three effects are similar, with individuals with a higher self-importance of moral identity, either as a whole or specified into symbolic or internal, reacting more strongly to differences in morality of a firm than individuals with a low self-importance of moral identity. In sum, hypothesis 2 is supported for the customer condition, meaning that the higher willingness to buy in reaction to a high morality low competence trade-off compared to a high competence low morality trade-off was stronger for participants high on self-importance of morality.
