The Challenge of Morality: The Effects of Rules and Ethical Leadership on Follower Compliance
Master thesis
MSc Human Resource Management,
University of Groningen, Faculty of Economics and Business
Marck R. Timmermans Student number: s2106272
Oude Ebbingestraat 54a 9712 HL Groningen tel.: +31 (0)651431984
E-mail: m.r.timmermans@student.rug.nl
Supervisor: L.B. Mulder Second corrector: H.J. van de Brake
Abstract
Firms try to prevent employees from engaging in immoral behavior by using compliance programs which consist of rules. Previous research by Mulder, Jordan and Rink (2015)
suggests that rule-framing may affect the effectiveness of a rule. They found that general rules are only mildly effective in shaping behavior but that they are applicable in various situations, whereas specific rules are very effective in shaping behavior in one particular situation. Due to the practicality of general rules for policy makers, it is important to find out if, and under what conditions, the effect of general rules can be strengthened. We hypothesized that ethical leadership should positively affect the effectiveness of the general rule due to the alignment of the messages of values and morality. A survey study was conducted using a 2 (leader: non- ethical/ethical) by 3 (rule: no/specific/general) between-subjects design. We found that the specific rule was an effective behavioral influencer. The general rule did not influence the behavior of employees. Surprisingly, when installed by an ethical leader, the specific rule had a spill-over effect on other moral situations. The general rule did not benefit from being installed by an ethical leader. These results imply that even though specific rules do not cover all thinkable situations, they may still be the most effective choice for practitioners.
Introduction
In 2015, investment bank Goldman Sachs was fined $50 million. An employee of the Federal Reserve had leaked confidential documents to an employee of Goldman Sachs who had formerly worked at the Federal Reserve. The employee of Goldman Sachs then proceeded to use the leaked documents for soliciting clients. Goldman Sachs uncovered the leak and took action against those involved. Although they took action, the company was still sanctioned the million dollar fine. The Federal Reserve reasoned that Goldman Sachs had not sufficiently supervised employees who had obtained the confidential documents. The Federal Reserve said that the firm needed to fix its shortcomings in its preventive policies that shape morality and had to enhance compliance programs (Hamilton, 2016). The Goldman Sachs example shows that compliance and morality are important governance issues. Companies are continually pressured to act as socially responsible agents that operate not only according to legal standards, but also to high moral standards. Organizations find multiple ways to do so.
For example, companies use formal methods to shape behavior.
Formal morality-shaping methods include ethics codes, training programs and rules that stimulate employees to engage in moral behavior (Valentine & Fleischman, 2008). Rules often form the core activity of a firm’s morality-shaping activities (Treviño, Weaver, Gibson,
& Toffler, 1999). The law differentiates between rules and standards (Covey, 2016). Rules use factual predicates to shape specific behaviors (e.g. “You may not drive faster than 50 km/h.”), whereas standards refer to objectives and require the subject (e.g. civilian), enforcer (e.g. police) or the interpreter (e.g. judge) to make an interpretation of the standard (e.g.
“Drive safely.”) (Schauer, 2001). Mulder et al. (2015) make a likewise distinction between specific and general rules. The specific rule (in law: the rule) is clear and enforceable, whereas the general rule (in law: the standard) is comprehensive, as it aims to evoke a broad scope of behaviors (Raiborn & Payne, 1990). The characteristics of a rule function as a
tradeoff between the breadth and depth of the effectiveness of that rule (Mulder et al., 2015).
General rules are only mildly effective in shaping behavior, but this mild effectiveness is spread among a broad scope of behaviors. On the contrary, specific rules are very effective in shaping the behavior it targets, but seem to be ineffective in shaping non-corresponding behaviors as it does not evoke a spill-over effect (Mulder et al., 2015).
Children are confronted with rules and compliance from an early age. As they develop, children will gradually move their compliance bases from being externally controlled, such as the fear of being punished, to a more effective internal behavioral self-regulation based on values (Gralinski & Kopp, 1993). Although it is not clear if employees in organizations go through a likewise process of moral development, it is clear that employees can react to moral issues using the same types of compliance bases. Employees might comply out of fear of getting caught and punished (Leal, Vrij, Nahari, & Mann, 2016), or they might internalize ethical standards and use value-based reasoning to come to moral decisions (Mulder et al., 2015). Firms should be especially interested in inducing compliance based on values instead of compliance based on the fear of getting caught or punished, because value-based
compliance is related to positive outcomes such as a reduction in unethical and illegal behavior in the firm, increased awareness of moral issues, increased ethical advice seeking within the firm, a greater willingness to deliver bad news or report ethical and legal violations to management, better decision making, and increased employee commitment (Treviño et al., 1999).
A compliance program that is only based on formal methods will not be sufficient to induce value-based compliance behavior (Adam & Rachman-Moore, 2004). The use of formal methods alone is related to outcomes such as apathetic and unwilling obedience (Fritz, Arnett, & Conkel, 1999). Previous research suggests that, as a prerequisite for effective compliance management, formal morality-shaping methods need to be combined with
informal morality-shaping methods, which consist of examples set by managers, cultural influencers and personal values of the employee (Fritz et al., 1999). Leaders can function as strong informal influencers. Several leadership theories define influence outcomes of leaders in terms of changing a follower’s values and beliefs (Yukl, 2013), which is important for effective compliance management. As most companies focus on compliance programs by using formal methods such as rules, it is interesting to see if, and under what conditions, leaders can strengthen or weaken the effects these rules have. Ethical leadership theories are of special relevance, as they emphasize leader influence in terms of moral behavior by employees (Bedi, Alpaslan, & Green, 2016). Therefore, in this research, the following research question will be examined:
Do ethical leaders (as opposed to non-ethical leaders) affect the effectiveness of rules that shape follower compliance?
This research can help practitioners in several ways. First, a practical insight in the combination of formal and informal compliance methods will be given. As illustrated by the Goldman Sachs example, stakes in compliance management can be high. The majority of firms are now focusing on legal compliance using formal methods (Treviño et al., 1999). It is important to investigate whether these actions that organizations are taking will lead to the desired compliance behaviors. Second, this study can help organizations assess their current situation in regard to formal and informal compliance methods and decide which area needs more attention. For example, an insight can be given in the most effective ethics codes when current leaders possess certain leadership characteristics. On the other hand, as leadership characteristics can be developed (Yukl, 2013), organizations can opt for altering their
targeting, selection and training mechanisms for leaders in their organization in order to align leadership with existing ethics codes.
Theoretical background
Morality
Morality involves the judgment of situational “rights and wrongs,” and the response to such judgments (Fieser, n.d.). People should somehow encounter the moral characteristics in a situation and should be encouraged to base their behavioral decisions on moral reasoning (Jones, 1991). Moral reasoning is a process in which an individual identifies ethical dimensions of a situation, morally evaluates this situation and acts according to this evaluation (Sternberg, 2012). When an individual attends to these kinds of moral matters, their moral identity will become salient, which influences the self-identity (Weaver, 2006).
The self-identity consists of attitudes, values and beliefs (Ryder, Alden, & Paulhus, 2000) and influences behavioral intention (Terry, Hogg, & White, 1999). Identifying the moral
characteristics of a situation does not automatically mean that people will form a moral response. If other motivations are strong, such as the motivation to react selfishly, a conflict between the different motivations may arise (Baker, 2014). This conflict makes the moral character of a situation less dominant, which causes less personal resistance to violate moral standards (Tsang, 2002).
Specific and general rules
Firms mostly use formal methods to steer employees towards moral behavior when moral issues or dilemmas arise. As mentioned, one of the most important formal methods that companies engage in is the use of rules in ethics codes (Treviño et al., 1999). The
effectiveness of a rule may depend on its characteristics (Mulder et al., 2015). Raiborn and Payne (1990) describe three important characteristics of rules: comprehensiveness, clarity, and enforceability. These three characteristics help categorize rules on a continuum, with on one extreme a specific rule and on the other extreme a general rule (Mulder et al., 2015).
Specific rules are clear and enforceable. An example of a specific moral rule in a business setting could be: “Do not accept gifts from clients.” This rule communicates a clear and
enforceable message. It is indisputable that the employee cannot accept any gifts from clients.
The clarity of the rule makes it easy to enforce, as punishments for violations can be described and carried out. However, such a rule is not comprehensive, as it does not aim to evoke a broad range of behavior in employees, other than not accepting a gift. A rule can be framed comprehensively by formulating it in a more general way. An example of a general rule in a business setting could be: “Do not engage in behavior that could give rise to conflicts of interests.” This is a broader rule that not only addresses the problem of accepting gifts from clients, but also other issues like bribing, promoting family members, and other immoral behaviors.
Moral rationalizations
Although people are sometimes prone to act on immoral motivations, they will eventually realize that their behavior violated their moral principles. This creates a need to morally rationalize their behavior, because people have a strong need to view the self as moral (Tsang, 2002; Aquino & Reed, 2002). Mulder et al. (2015) found that specific rules will help reduce employees’ engagement in moral rationalizations, in turn strengthening the behavior inducing effect. Moral rationalizations help employees reason that their immoral behavior is morally permissible (e.g. “Accepting this gift will not change my interests.”). Under a specific rule, there is not much room to reinterpret immoral behavior, as the clarity and enforceability of the rule weakens this possibility (Ferrell & Gresham, 1985). On the contrary, a general rule leaves much more room for moral rationalizations because it does not target specific behavior.
Under a general rule, an employee will easier be able to reinterpret immoral behavior as being moral. For example, an employee that has befriended a client can easily reason that it is okay to accept a gift because it is a token of friendship and has nothing to do with their working relationship (if the general rule is: “Do not engage in behavior that could give rise to conflicts of interests.”).
In line with research from Mulder et al. (2015), we expect the specific rule to be more a more effective influencer of behavior than the general rule. The specific rule will help
decrease the possibility of moral rationalizations, whereas the general rule will leave more room to creatively justify immoral behavior. Also, we expect that the enforceability of the specific rule will be more effective in refraining people from breaking it out of fear of getting caught or punished (Raiborn & Payne, 1990; Leal et al., 2016). However, we think that the underlying, comprehensive objective of the general will function as a moral signal. This signal should stimulate some moral reasoning in employees (Velasquez & Rostankowski, 1985), causing the general rule to be more effective than when no rule is installed. Therefore, the following is hypothesized:
Hypothesis 1. A specific rule will induce more compliance than a general rule and a general rule will induce more compliance than no rule.
Breadth of induced moral behaviors
Although a specific rule seems to restrict the possibilities of moral rationalizations, it also has its limitations. Its effect depends strongly on the situation, as it is only applicable in a specific situation. The clarity and enforceability of the specific rule may cause apathetic obedience instead of commitment or goal internalization (Fritz et al., 1999). Employees will be willing to carry out a request, but they will not be enthusiastic about it and will refrain themselves to bare minimum effort (Fritz et al., 1999). Furthermore, a specific rule can narrow the moral perspective of an individual (Mulder et al., 2015). Instead of focusing on a broader scope of moral behavior, individuals merely focus on the behavior that is targeted by the specific rule. In this situation, following the rules becomes more important than the
underlying moral message. This inhibits the possibility of a spill-over of the moral message to other, similar situations. A general rule tries to influence behavior by modifying individual factors like values, intentions, and the self-concept. As such, a general rule should stimulate
ethical reasoning within employees, which should evoke a broad scope of moral behavior (Mulder et al., 2015). Therefore, the following is hypothesized:
Hypothesis 2. A general rule is mildly effective in various moral situations, whereas a specific rule does not have a spill-over effect.
Although the literature suggests that specific rules are more effective in shaping behavior, it is reasonable to assume that policy makers are more interested in using general rules. Creating endless lists of specific rules that cover any thinkable scenario is highly inefficient. Besides, by creating such long lists, the specific rule is bound to lose its power because it is easily overseen or forgotten due to the huge content of the resulting ethics codes.
For policy makers, it is important to consider (situational) characteristics that help general rules become more effective. Looking at recent leadership theories, we think ethical leadership could be such a characteristic.
Ethical leadership
Leaders heavily affect employees, and some leaders are more effective in doing so than others (George, 2000). Effective leaders engage in behaviors that help create a healthy and successful work environment: they can set collective goals, generate and maintain enthusiasm, convey the idea of the importance of work activities, and foster cooperative behaviors
(George, 2000). Scientific interest in ethical leader characteristics is growing as corporate social responsibility has become a global issue (Webb, 2008).
When leaders function as ethical guides for their followers they can be classified as ethical leaders (Brown, Treviño, & Harrison, 2005). The ethical leader has two roles (Brown
& Treviño, 2006). First, an ethical leader is a moral person. Ethical leaders are honest and trustworthy. They are also fair and principled decision-makers who care about people and the broader society. Second, the leader is a moral manager. This aspect is represented by the leader’s efforts to influence ethical and unethical behavior of followers. A moral manager
explicitly communicates values and morality by visibly and intentionally role modeling desired behavior. A moral manager also uses rewards and punishments in regard to morality to shape follower behavior (Brown et al., 2005). It is important that the moral personality of a leader closely resembles the moral management of that leader in order to be an effective ethical agent (Schminke, Ambrose, & Neubaum, 2005). Thus, the displayed morality of the leader as a person should match the leader’s moral management behaviors.
Ethical leaders make morality a core leadership message (Treviño, Brown, & Hartman, 2003), which helps induce desired ethical behavior among employees. This is because followers have a tendency to imitate leaders (Bandura, 1986). In a large web of
organizational, employer, and employee interests, conflicts with moral decision making may arise. Therefore, it is important that a leader makes the moral message salient in order to facilitate the imitation of moral behavior. However, followers will not blindly imitate leaders.
A leader also needs to be an attractive and credible role model (Bandura, 1986). Individuals learn by paying attention to attractive role models and copy their behavior, attitudes and values. According to Bandura (1986) power and credibility positively influence attractiveness.
Thus, leaders who are powerful and credible and, at the same time, display moral behavior are likely to induce moral decision making in followers. By influencing the attitudes and values of followers, an ethical leader is able to induce more than compliance among followers.
Followers can commit to the moral goals of the leader and internalize the leader’s values (Yukl, 2013), which stimulates followers to view morality as an important part of their own self-identity.
Because the alignment of values and messages is thought to positively influence follower behaviors (Abreu, Macedo, & Camarinha-Matos, 2009), we predict that the general rule should become more effective when installed by an ethical leader. The general rule and the ethical leader both try to make morality a salient part of the self-identity. A general rule
tries to evoke a broad scope of behaviors by addressing values and inducing moral reasoning (Mulder et al., 2015). The ethical leader makes morality a salient and important part of the leadership message (Treviño et al., 2013). By being an ethical leader, the leader will should credibility and power due to the alignment of the moral personality with moral management (Bandura, 1986; Schminke et al., 2005). Followers should then find the leader and the leader’s behaviors attractive, causing them to imitate moral behaviors (Bandura, 1986). At the same time, the leader should be able to influence the values and attitudes of followers, who will internalize the moral values and goals of the leader (Yukl, 2013). These positive effects of ethical leadership should align with the effects of the general rule, strengthening the salience of morality in the self-identity (Weaver, 2006). We think that because of these reasons the combination of the general rule and an ethical leader should increase the effectiveness of shaping follower compliance. On the contrary, a specific rule narrows the moral scope of the follower. Its clarity and enforceability divert attention away from the moral environment towards the specific situation. The clarity and enforceability of the rule may also instigate apathetic obedience because following the rule (instead of moral behavior) becomes the main objective (Fritz et al., 1999). Fear of getting caught or punished can also become the main behavioral influencer, because the specific rule can work as a deterrent that causes obedience (Leal et al., 2016). The dominance of the characteristics of the specific rule and their
contradictions to the influence mechanisms of the ethical leader have led us to believe that the specific rule will not be affected from being installed by an ethical leader. Therefore, the following is hypothesized:
Hypothesis 3. A general rule will induce more compliance when installed by an ethical leader than by a non-ethical leader. For a specific rule, the type of leader that installs it does not matter.
Method
Design and participants
A scenario study was performed with a 2 (leader: non-ethical/ethical) by 3 (rule:
no/specific/general) between-subjects design. Participants were randomly assigned to one of six conditions. The study took about 15 minutes to complete. The dependent variables and additional explanatory variables were all measured on a continuous scale.
A sample of 148 Dutch participants was used (54% male) that were at least 15 years old and had no work experience as an accountant. Participation was on a voluntary basis. All participants signed an informed consent form. The researcher’s own network was used to recruit the participants by means of email, social media, and personal requests. Participants received a link which redirected them to the online study. The mean age of the participants was 29.32 years (SD = 12.07). Seventy eight percent of the participants had received the highest possible education (WO), 20% the second highest education (HAVO, VWO, &
HBO). Sixty three percent of the participants were currently employed for more than 12 hours a week. Thirteen percent of the participants were not employed but were looking for a job at the time of the research. Forty eight percent of the participants signaled they were familiar with the content of the work of an accountant.
Procedure
First, the participants received a general questionnaire that measured demographics.
They were also asked to choose their most favorite day out of eight options (a sports game, a concert, a festival, the spa, a museum, a culinary tour, a cultural tour or a theme park). Their answer was later inserted into the scenario to create a personalized scenario. Participants then received a scenario in which both the leader and the presence of the rule were manipulated.
Following the scenario, the unethical intentions of the participants were measured. Next, multiple additional explanatory variables were measured. These measures included moral disapproval, moral rationalization, the perception of rule violation, punishability of act,
punitive ethical climate, and a rule fairness control. Finally, manipulation checks were presented.
Leader manipulation
Participants received the following scenario that was adapted from Mulder et al. (2015):
“Imagine you are an accountant at a big accounting firm. Your main duty is approving the financial statements of other companies. It is important that you work accurately, as you need to spot irregularities and inaccuracies to provide financial assurance to the managers,
shareholders and tax authorities.”
In the ethical leader condition, this was followed by an extra paragraph about an ethical leader: “Your boss is a trustworthy person. He makes fair and balanced decisions, listens to what employees have to say, and defines success not just by results but also the way that they are obtained. He often discusses business ethics and values with employees and you know that he conducts his personal life in an ethical manner. He is the kind of person that, when making decisions, asks himself “What is the right thing to do?” He is all about setting the right example in terms of ethics and morality, and he is known to discipline employees who violate his ethical standards.” The traits of this leader were based on the ethical leadership scale developed by Brown et al. (2005). In the non-ethical leader condition, no information about the leader was provided.
Participants were asked to imagine working as an accountant at this firm for the remainder of the experiment.
Rule manipulation
In the specific rule condition, participants received the following rule: “Your boss recently installed a new code of conduct. Among others, your boss has specified the following rule: ‘Accountants should not accept gifts from clients.’”
In the general rule condition, participants received the following rule: “Your boss recently installed a new code of conduct. Among others, your boss has specified the following rule: ‘Accountants should not engage in conflicts of interests.’”
In the no rule condition, participants did not receive a rule.
All participants were then reminded of the responsibilities of an accountant: “Your most important task is to give accountancy declarations to companies. It is important that you work independently and precisely, as it is important that you spot inaccuracies and irregularities in order to provide management, shareholders and tax authorities with financial security.”
Participants then received the following scenario (Mulder et al., 2015): “One day, while in the process of approving a client’s declaration, the company representative you are
working for offers you, the accountant, two tickets for your favorite [selected day out was inserted here] as a “thank you” for the years of good work done for him, the client.”
Dependent variable measures
Unethical intention. Participants’ unethical intentions were measured by asking them to indicate on a 7-point Likert scale whether they would accept the tickets that were offered by the client (1 = certainly not, 7 = certainly).
Spill-over unethical intention. Participants then received the following scenario (Mulder et al., 2015): “One day, while working on a declaration, the client walks into your office. He offers you a prestigious position in his supervisory board. This would certainly give you the career boost you had been waiting for.” Participants were then asked to indicate how they would react on the same 7-point Likert scale as in the unethical intention measure.
Other measures
Moral disapproval. To assess to what extent participants morally disapproved of accepting the tickets and the position in the supervisory board, a scale adapted from Mulder and Nelissen (2010) was presented twice, once per each unethical behavior. The scale for
accepting the two tickets (α = .95) consisted of six items that all started with: ”Accepting the two tickets…” and ended with: (1) “…is not done,” (2) ”…I morally disapprove of,” (3) ”…is something I do not object to” (reverse coded), (4) ”…is morally incorrect,” (5) ”…ought not to be done,” and (6) ”…is not so bad” (reverse coded). The scale for accepting the position in the supervisory board (α = .96) consisted of the same six items, but started with: “accepting the position in the supervisory board…” followed by the same endings of statements one through six (e.g. (1) “…is not done”). Participants were asked to rate the items on a 7-point Likert scale (1 = totally disagree, 7 = totally agree).
Moral rationalizations. To assess whether participants would morally rationalize accepting the two tickets or the position in the supervisory board, participants were presented with 6 items on a 7-point Likert scale (1 = totally disagree, 7 = totally agree). Participants were presented the scale twice, once per each unethical behavior. The scale for accepting the tickets (α = .85) consisted of the following items: (1) “By accepting the tickets I do not harm anybody,” (2) “Accepting the tickets will strengthen the relationships with the client, which is good for the office I work for,” (3) “The client’s organizations needs to give the tickets away anyway, so I might as well accept them,” (4) “I can accept tickets without big consequences for society,” (5) “The tickets are merely a small token of appreciation, so it is okay to accept them,” and (6) “If I accept tickets my objectivity is not at stake.”
The scale for the spill-over unethical intention (α = .79) consisted of the same six items except it focused on the behavior of accepting the position in the supervisory board (e.g.
(1) “By accepting the position in the supervisory board I do not harm anybody.”).
Perception of rule violation. Participants were asked to indicate on a 7-point Likert scale whether the acts of accepting the tickets and the position in the supervisory board were in favor or against the rules (1 = violation of the rules, 7 = according to the rules).
Punishability of act. Participants received a scale of five items to assess if they felt that their behavior would be punished in the accounting firm. Again, this scale was presented twice, once per each unethical behavior. The scale for accepting the tickets (α = .79) consisted of the following five items: (1) “If I accept the tickets my boss will surely find out,” (2)
“Accepting the tickets will not go unnoticed,” (3) “If I accept the tickets I will probably get into trouble,” (4) “If I accept the tickets it will probably be problematic,” (5) “If I accept the tickets I will probably be punished.”
The scale for the spill-over unethical intention (α = .75) consisted of the same five items except it focused on the behavior of accepting the position in the supervisory board (e.g.
(1) “If I accept the position in the supervisory board my boss will surely find out.”).
Punitive ethical climate. Participants were then asked to assess the ethical climate at the accounting firm by rating four items (α = .81) adapted from Kaptein (2008) on a 6-point Likert scale (1 = totally disagree, 6 = totally agree). All four items started with: “In my
immediate working environment at the accounting firm…“ and ended with: (1) “…employees will be disciplined if they behave unethically,” (2) “…even small ethical violations will be punished,” (3) “…only people with integrity are considered for promotion,” and (4)
“…measures against unethical behavior are applied consequently.”
Fairness. To test whether participants who received a rule perceived it to be fair, four items (α = .76) were presented on a 7-point Likert scale (1 = totally disagree, 7 = totally agree). The participants were asked to judge whether their rule was: (1) “Fair,” (2)
“Righteous,” (3) “Clear,” and (4) “Reasonable.” This scale was only presented to participants that received a general or specific rule.
Manipulation checks
First rule check. In the first rule check, all participants (including those that did not receive a rule) were presented with an explanation about the differences between specific and
general rules (with the example for a specific rule: “The maximum speed limit is 50 km/h.”
and for a general rule: “Drive safely.”). Participants were then asked to think about the rules at the accounting firm and were asked to indicate on a 7-point Likert scale (1 = very specific, 7
= very general) whether the rules at the firm were specific or general.
Second rule check. The second rule check was presented only to the participants who received a general or specific rule. In order to check whether these participants comprehended their rule and could make the correct distinction between a specific and a general rule they were once more presented with the rule that their boss had installed in the beginning of the study. They were then asked to rate this rule on the same 7-point Likert scale as in the first rule check.
Ethical leader check. The manipulation check of the leader installment consisted of 9 items (α = .85) that were assessed on a 7-point Likert scale (1 = totally disagree, 7 = totally agree). Participants were asked to rate their boss in the study on nine ethical leader
characteristics adapted from Brown et al. (2005). Examples of these characteristics are
“considerate,” “compassionate,” and “honest.”
Results Data analysis
For the dependent variables (unethical intention, spill-over unethical intention) two 2 (leader) x 3 (rule) ANOVAs were conducted. The assumptions were checked (independence, normality, homogeneity of variances). Unfortunately, both ANOVAs failed the assumption of normality, showing a W-shaped distribution for which no transformation was possible. The main effects of the rule and the leader, and the interaction effect leader*rule on the targeted unethical intention were tested for significance. Post hoc Scheffe tests and pairwise
comparisons were used to investigate individual group differences. The same 2 (leader) x 3 (rule) ANOVAs were used to test for the effects of the other explanatory variables and to
perform the manipulation checks (after checking the assumptions). The perception of rule violation was identified as a mediator for the unethical intention by an SPSS PROCESS bootstrapping macro. Moral disapproval was identified as a mediator for the spill-over
unethical intention by an SPSS PROCESS bootstrapping macro. For all tests an alpha level of .05 was used.
Manipulation checks
Ethical leader check. The 2 (Leader) x 3 (Rule) ANOVA on the ethical leader scale yielded only a significant main effect of the leader on the ethical leader scale, F(1,117) = 9,08, p = .003, η2 = .07. Participants in the ethical leader condition judged their boss to be more ethical (M = 5.15, SD = .73) than in the non-ethical leader condition (M = 4.72, SD = .87). No main effect of the rule or interaction effect was found. The results indicate that the leader manipulation was successful.
First rule check. The 2 (Leader) x 3 (Rule) ANOVA on the rule manipulation check yielded no significant main or interaction effects. Participants in the different rule conditions did not rate the rules at the accounting firm differently. Unfortunately, a successful rule manipulation cannot be confirmed, although it may be that the results of this check are distorted due to factors such as inattention of the participants, information density or a wrong manipulation control question.
Second rule check. The 2 (Leader) x 2 (Rule) ANOVA on the comprehension of the rules yielded only a significant main effect of the rule, F(1,100) = 12.02, p = .001, η2 = .11.
When presented their rule, participants could make the correct distinction between the specific (M = 2.88, SD = 1.70) and the general rule (M = 4.98, SD = 1.67).
Unethical intention: accepting two tickets to the favorite day out
The 2 (Leader) x 3 (Rule) ANOVA on the unethical intention showed a significant main effect of the rule on the unethical intention, F(2,140) = 4.93, p = .009, η2 = .06. Post hoc
Scheffe testing showed that in the specific rule condition (M = 2.76, SD = 1.47) participants were less prone to accept the two tickets than in the no rule condition (M = 3.55, SD = 1.53, p
= .05) and the general rule condition (M = 3.67, SD = 1.62, p = .013). No main effect of the leader or interaction effect was found.
Moral rationalizations. The 2 (Leader) x 3 (Rule) ANOVA on the moral
rationalizations of accepting the tickets yielded no significant main or interaction effects.
Moral disapproval. The 2 (Leader) x 3 (Rule) ANOVA on the moral disapproval of accepting the tickets yielded no significant main or interaction effects.
Punishability of act. The 2 (Leader) x 3 (Rule) ANOVA on the punishability of act when offered the tickets showed a significant main effect of the leader, F(1,124) = 7.46, p = .007, η2 = .06. The ethical leader resulted in a stronger feeling of being punished for accepting the tickets (M = 4.25, SD = 1.12) as opposed to the non-ethical leader (M = 3.71, SD = 1.12).
No main effect of the rule or interaction effect was found.
Perception of rule violation. The 2 (Leader) x 3 (Rule) ANOVA on the perception of rule violation when accepting the tickets showed a significant main effect of the rule, F (2,134) = 10.45, p < .001, η2 = .13. Post hoc Scheffe testing showed that in the specific rule condition (M = 2.52, SD = 1.70) participants felt that the behavior of accepting the tickets was more against the rules than in the no rule condition (M = 3.70, SD = 1.51, p = .003) or in the general rule condition (M = 3.91, SD = 1.72, p < .001). No main effect of the leader or interaction effect was found. An SPSS PROCESS bootstrapping macro for simple mediation was performed which showed that the perception of rule violation mediated the effect of the rule on the unethical intention, z = .69, Boot SE = .12, 95% CI [.59, .79].
Spill-over unethical intention: accepting a position in the supervisory board
The 2 (Leader) x 3 (Rule) ANOVA on the spill-over unethical intention yielded no significant main effects, but a significant interaction effect of the rule and the leader on the
spill-over unethical intention was found, F(2,140) = 3.24, p = .042, η2 = .04. Table 1 shows that the specific rule significantly lowered the spill-over unethical intention as compared to the general rule when installed by an ethical leader. However, when the non-ethical leader was present, the rules did not decrease the spill-over unethical intention.
Table 1
Means and standard deviations for the spill-over unethical intention in the 2 (Leader) x 3 (Rule) design
Ethical Leader Non-ethical Leader
No rule 3.27ab (1.86) 3.35ab (1.63)
Specific Rule 2.97a (1.43) 3.84b (1.52)
General rule 3.88b (1.64) 3.17ab (1.47)
Note: means with non-overlapping superscripts differ significantly (p < .05) from each other according to pairwise analyses.
Moral rationalizations. The 2 (Leader) x 3 (Rule) ANOVA on the moral
rationalizations of accepting the position in the supervisory board yielded no significant main or interaction effects.
Moral disapproval. The 2 (Leader) x 3 (Rule) ANOVA on the moral disapproval of accepting the position yielded no significant main effects, but a marginal interaction effect was found F(2,134) = 2.97, p = .055, η2 = .04. Table 2 shows that when the ethical leader was present, the general rule resulted in less disapproval of accepting the position in the
supervisory board than the specific rule. Surprisingly, the general rule did not result in such a decrease of moral disapproval when installed by the non-ethical leader. An SPSS PROCESS bootstrapping macro for moderated mediation was performed which showed that the moral disapproval mediated the effect of the rule moderated by the leader on the spill-over unethical intention, z = -.73, SE = .05, 95% CI [-.82, -.64].
Table 2
Means and standard deviations for the moral disapproval of accepting the position in the supervisory board in the 2 (Leader) x 3 (Rule) design
Ethical Leader Non-ethical Leader
No rule 4.39ab (2.01) 4.42ab (1.82)
Specific rule 4.60b (1.53) 4.05ab (1.68)
General rule 3.61a (1.96) 4.78b (1.69)
Note: means with non-overlapping superscripts differ significantly (p < .05) from each other according to pairwise analyses.
Punishability of act. The 2 (Leader) x 3 (Rule) ANOVA on the punishability of act when offered the position in the supervisory board yielded no significant effects.
Perception of rule violation. The 3 (Leader) x 2 (Rule) ANOVA on the perception of rule violation when accepting the position in the supervisory board yielded no significant main or interaction effects.
Punitive ethical climate
The 2 (Leader) x 3 (Rule) ANOVA on the punitive ethical climate showed a main effect of the leader, F(1,140) = 6,31, p = .013, η2 = .05. Under the ethical leader, participants felt that the ethical climate was more punitive (M = 4.10, SD = 1.00) than under the non- ethical leader (M = 3.66, SD = .98). No main effect of the rule or interaction effect was found.
Discussion
The aim of this research was to find out if, and under what conditions, ethical
leadership would influence the effects that rules have on follower compliance. This study has shown that only the specific rule benefited from being installed by an ethical leader. The following three hypotheses were investigated. First, it was hypothesized that the specific rule would be more effective in lowering the targeted unethical intention than the general rule, and
that the general rule would be more effective in doing so than when no rule was installed.
Second, it was hypothesized that the general rule would be mildly effective in various moral situations and that the specific rule would not have a spill-over effect. Third, it was
hypothesized that the general rule would become more effective when installed by an ethical leader than by a non-ethical leader, and that the type of leader would not influence the effectiveness of the specific rule.
The first hypothesis was partly verified. Like previous research by Mulder et al.
(2015) suggested, people were less prone to behave immorally when a specific rule was installed. However, it was also hypothesized that the general rule would be more effective in lowering the unethical intention than when no rule was installed, for which no evidence was found. Installing a general rule did not weaken the tendency to behave immorally. Previous literature had suggested that the specific rule would be effective because it would lower the opportunity to morally rationalize immoral behavior (Mulder et al., 2015). This claim cannot be verified by this research. All participants in this study had the tendency to morally
rationalize their behavior, no matter what the rule was or what kind of leader installed that rule.
The second hypothesis partly was verified. Overall, when people came across a moral situation that was different than the behavior described by the specific rule, there was not a spill-over effect of the specific rule. Thus, the specific rule created the expected narrow moral scope (Mulder et al., 2015). However, when zooming in on the different leaders that installed the rule, we found that a spill-over effect of the specific rule did take place when it was
installed by an ethical leader. This is strange, as it contradicts the notion of Raiborn and Payne (1990) about the lacking comprehensiveness of the specific rule. When an ethical leader installs a specific rule, a broad set of behaviors may be targeted. It was also hypothesized that
the general rule would be mildly effective in various situations. This was not found to be true.
The general rule did not yield any effect in any condition of this study.
The third hypothesis was not confirmed. It was predicted that the general rule would become more effective when installed by an ethical leader due to the alignment of the core moral leadership message with the objectives and mechanisms of the general rule. However, the general rule did not become more effective in lowering the unethical intention when installed by an ethical leader. Furthermore, when an ethical leader installed a general rule, people disapproved less of immoral behavior. As previously mentioned, when the specific rule was installed by an ethical leader its effect did surprisingly spill over to the non-related moral situation, effectively lowering the unethical intention. In this study, people were given the rule not to accept gifts from clients. When this was installed by the ethical leader, people were not only less prone to accepting the two tickets to their favorite day out, but they were also less prone to accept the position in the supervisory board. It seemed that the specific rule, when installed by an ethical leader, communicated a message of morality that exceeded merely the behavior which the rule targeted.
Rule effectiveness
Altogether, we think that the clarity and enforceability of the specific rule, instead of the comprehensiveness of the general rule, were most influential in shaping follower
compliance. These characteristics of the specific rule are thought to provide followers with a clear warning signal. We think that the comprehensiveness of the general rule causes it to lose its power. Followers may not see its function as a warning function and may become
inattentive to its objectives. This indicates that, in this study, following the rules may have been more important than the actual function of the rules. If this was the case, participants approached obedience as a utility function. Whenever the costs outweighed the benefits of breaking the rule, individuals were likely to abide. This line of thinking is based on the
inherent self-interest of the individual (Clague, 1993). It may be that the selfish motivation is too strong compared to the moral motivation in a hypothetical scenario study. We recommend that further research devotes to removing selfishness from the equation to see if the moral motive can prevail in such a study. By doing so, a less distorted observation of morality can be made. Furthermore, we suggest that research needs to aim to find possible moderators or mediators that could strengthen the effect of the general rule, because general rule are of great convenience for policy makers.
Leader effectiveness
It was expected that by making morality a core leadership message, people would become morally aware of the situation through contagion and imitation of the ethical leader (Brown & Treviño, 2006; Bandura, 1986). By doing so, it was expected that people would direct attention to moral reasoning in moral situations. However, results indicate that the ethical leader diverted attention away from moral reasoning. Under the ethical leader, people had a stronger sense of working in a punitive ethical climate in which ethical violations would be discovered and punished. Furthermore, when the general rule was installed by an ethical leader, people disapproved less of immoral behavior. The strong, clear and salient moral message of the ethical leader may have been a warning signal that resulted in followers focusing on getting caught and punished for moral violations, instead of focusing on morality.
As such, the combination of the general rule with the ethical leader may have been a
mismatch, as the general rule lacked the enforceability which the ethical leader emphasized.
Combining rules with ethical leadership
Different situations require different styles of leadership (Hersey & Blanchard, 1969), which may explain the ineffectiveness of the combination of the general rule with the ethical leader. Participants may have focused on behaviors that would obviously be punished, but under the general rule did not perceive their behavior to be punishable. This idea is illustrated
by the reaction of participants to the question if their behavior had violated the rules. Under the general rule, participants did not find any of the unethical behaviors to be rule violating.
Following this line of reasoning, we suggest that the spill-over effect of the specific rule was caused by a match with ethical leadership characteristics. The clarity and enforceability of the specific rule signaled a strong and punitive moral message, just like the ethical leader did. We think that the alignment of these characteristics caused participants to react cautiously in potential rule-violating situations, because they received warning signals from both the rule and the leader. By being cautious, the moral characteristics of the situation may have become salient even though the rule did not cover the specific situation. We think this caused the unexpected spill-over effect of the specific rule. The question that now remains is if rule- leadership alignment is an important prerequisite to cover all possible moral scenarios in an organization, and if so, which types of characteristics need to be aligned to facilitate positive results. Furthermore, due to the practicality of general rules, there is a desperate need for the identification of (leadership) characteristics that will positively influence the effects of the general rule.
Limitations
This study had some noteworthy limitations. The pool of participants was quite small and homogenous, most participants were young and higher educated, which could have caused a low variability of answers. In a real organizational setting there would most likely be a more diverse group of employees, all with different backgrounds, experiences, values and other characteristics that would base their behavioral decisions upon different cues. The participants had few (or none) experience with moral situations in a work context. Imagining such a situation and the consequences it can have may have been difficult for these
participants. This could have guided their behavioral intention differently than people who do have such experiences. Furthermore, this study did not measure real behavior, it merely
focused on a behavioral intention shaped by a hypothetical scenario. People were asked to picture themselves as an accountant working for a client. The results of this research depend highly on if, and how, participants pictured the situation. Furthermore, because this was a scenario study, there was nothing to be gained or lost. People may have presented themselves more morally to enhance their own self-image, or they may have taken greater risks than they would in a real life setting.
The leader manipulation seems to have worked, although the actual influence of the manipulation is hard to assess. For ethical leadership, it is important for someone to identify with a leader before imitation takes place (Bandura, 1986). In a real life setting, under a real ethical leader, people will probably be influenced to a much greater extent than by this short paragraph. Participants had difficulties recollecting the rule they had received, causing us to have difficulties assessing the strength of the rule manipulation. We think that due to the large amount of information, duration of the study, and the fact that people did understand the differences in general rules in our second rule check, we can reasonably say that our
manipulation did have an effect. We think that the manipulation check question we asked did not make participants attentive of their original rule, which is why we could not find a
connection between the condition and the manipulation check. However, we cannot conclude this with certainty.
Practical implications
This research has given an important signal to policy makers. The results indicate a situation-specific model of leader and rule effectiveness, in which both methods should somehow be aligned. Policy makers should take the organization-specific situation in account when formulating rules. Furthermore, although the focus of policy makers usually lies on formulating general rules, they should be very cautious in doing so, as they have shown to be weak influencers. The specific option has been proven to be very effective, and it even has the
potential to create a spill-over effect when installed by an ethical leader that has never before been illustrated. Furthermore, this research has shown that, countering intuition, ethical leadership does not always lead to morally engaged followers. In this study, we think followers had a tendency to focus on the punitive climate under the ethical leader, possibly because the ethical leader sent a strong warning signal not to break the rules.
Conclusion
In this research, the moderating effect of an ethical leader on the influence of rules on an immoral behavioral intention was investigated. General rules were again proven to be less effective in shaping moral behavior than specific rules. Furthermore, a new dimension of a specific rule has been discovered. In this research, the narrow moral scope of the specific rule was broadened when an ethical leader installed it, making it effective for other situations as well.
References
Adam, A., & Rachman-moore, D. (2004). The methods used to implement an ethical code of conduct and employee attitudes. Journal Of Business Ethics, 54(3), 225-244.
doi:10.1007/s10551-004-1774-4
Aquino, K., & Reed, A., (2002). The self-importance of moral identity. Journal of Personality and Social Psychology, 83(6), 1423–1440. doi: 10.1037/0022-3514.83.6.1423
Baker, D. F. (2014). When moral awareness isn’t enough: teaching our students to recognize social influence. Journal Of Management Education, 38(4), 511-532.
doi:10.1177/1052562913504922
Bandura, A. (1977). Social learning theory. Englewood Cliffs, N.J: Prentice Hall.
Bandura, A. (1986). Social foundations of thought and action. Englewood Cliffs, N.J:
Prentice Hall.
Bedi, A., Alpaslan, C., & Green, S. (2016). A meta-analytic review of ethical leadership outcomes and moderators. Journal Of Business Ethics, 139(3), 517-536.
doi:10.1007/s10551-015-2625-1
Brown, M. E., Treviño, L. K., & Harrison, D. A. (2005). Ethical leadership: A social learning perspective for construct development and testing. Organizational Behavior & Human Decision Processes, 97(2), 117-134. doi:10.1016/j.obhdp.2005.03.002
Brown, M. E., & Treviño, L. K. (2006). Ethical leadership: A review and future
directions. Leadership Quarterly, 17(6), 595-616. doi:10.1016/j.leaqua.2006.10.004 Clague, C. 1993. Rule obedience, organizational loyalty, and economic development. Journal
of Institutional and Theoretical Economics, 149(2): 393-414.
Covey, R. D. (2016). Rules, standards, sentencing, and the nature of law. California Law Review, 104(2), 447-496. doi:10.15779/Z386V8W
Ferrell, O. C., & Gresham, L. G. (1985). A contingency framework for understanding ethical decision making in marketing. Journal Of Marketing, 49(3), 87-96
Fieser, J. (n.d.). Ethics. Internet Encyclopedia of Philosophy, Retrieved from http://www.iep.utm.edu/ethics/
Fritz, J. M. H., Arnett, R. C, & Conkel, M. (1999). Organizational ethical standards and organizational commitment. Journal of Business Ethics, 20, 289-299.
George, J. M. (2000). Emotions and leadership: The role of emotional intelligence. Human Relations, 53(8), 1027-1055.
Gralinski, J.H., & Kopp, C.B. (1993). Everyday rules for behavior – mother requests to young children. Developmental Psychology 29(3), 573-584. doi: 10.1037//0012-1649.29.3.573 Hamilton, J. (2016). Goldman Sachs fined $36 million by Fed over leaked documents [Press
release]. Retrieved from https://www.bloomberg.com/news/articles/2016-08- 03/goldman-sachs-fined-36-3-million-by-fed-over-leaked-documents-irf0sjm4 Hersey, P., & Blanchard, K. H. (1969). Life cycle theory of leadership. Training and
Development Journal. 23(5), 26–34.
Jones, T.M. (1991). Ethical decision making by individuals in organizations: An issue- contingent model. Academy of Management Review, 16(2), 366-395.
Kaptein, M. (2008). Developing and testing a measure for the ethical culture of organizations:
the corporate ethical virtues model. Journal Of Organizational Behavior, 29(7), 923- 947.
Leal, S., Vrij, A., Nahari, G., & Mann, S. (2016). Please be honest and provide evidence:
deterrents of deception in an online insurance fraud context. Applied Cognitive Psychology, 30(5), 768-774. doi:10.1002/acp.3252
Mulder, L., & Nelissen, R. (2010). When rules really make a difference: the effect of cooperation rules and self-sacrificing leadership on moral morms in social
dilemmas. Journal Of Business Ethics, 9557-72. doi:10.1007/s10551-011-0795-z
Mulder, L. B., Jordan, J., & Rink, F. (2015). The effect of specific and general rules on ethical decisions. Organizational Behavior & Human Decision Processes, 126, 115-129.
doi:10.1016/j.obhdp.2014.11.002
Raiborn, C., & Payne, D. (1990). Corporate codes of conduct – A collective conscience and continuum. Journal of Business Ethics, 9(11), 879–889. http://dx.doi.org/10.1007/
BF00382911.
Ryder, A. G., Alden, L. E., & Paulhus, D. L. (2000). Is acculturation unidimensional or bidimensional? A head-to-head comparison in the prediction or personality, self- identity, and adjustment. Journal of Personality and Social Psychology, 79(1), 49-65.
doi:10.1037/0022-3514.79.1.49
Schauer, F. (2001). The convergence of rules and standards. Regulatory Policy Program Working Paper RPP-2001-07. Cambridge, MA: Center for Business and Government, John F. Kennedy School of Government, Harvard University.
Schminke, M., Ambrose, M. L., & Neubaum, D. O. (2005). The effect of leader moral development on ethical climate and employee attitudes. Organizational Behavior &
Human Decision Processes, 97(2), 135-151. doi:10.1016/j.obhdp.2005.03.006
Sternberg, R. J. (2012). A model for ethical reasoning. Review Of General Psychology, 16(4), 319-326. doi:10.1037/a0027854
Terry, D.J., Hogg, M.A., & White, K.M. (1999). The theory of planned behaviour: Self- identity, social identity and group norms. British Journal of Social Psychology, 38(3), 225-244. doi: 10.1348/014466699164149
Treviño, L. K., Weaver, G. R., Gibson, D. G., & Toffler, B. L. (1999). Managing ethics and legal compliance: what works and what hurts. California Management Review, 41(2), 131-151.
Treviño, L. K., Brown, M., & Hartman, L.P. (2003). A qualitative investigation of perceived executive ethical leadership: perceptions from inside and outside the executive suite.
Human Relations, 56(1), 5-37.
Tsang, J. (2002). Moral rationalization and the integration of situational factors and
psychological processes in immoral behavior. Review Of General Psychology, 6(1), 25- 50. doi:10.1037/1089-2680.6.1.25
Valentine, S., & Fleischman, G. (2008). Ethics programs, perceived corporate social responsibility and job satisfaction. Journal of Business Ethics, 77(2), 159-172.
doi:10.1007/s10551-006-9306-z
Velasquez, M. G., & Rostankowski, C. (1985). Ethics: Theory and practice. Englewood Cliffs, NJ: Prentice-Hall.
Weaver, G.R. (2006). Virtue in organizations: moral identity as a foundation for moral agency. Organization Studies, 27(3), 341-368.
Webb, T. (2008). Why ethical leadership matters more than ever in a downturn. Retrieved from http://www.ethicalcorp.com/business-strategy/why-ethical-leadership-matters- more-ever-downturn
Yukl, G.A. (2013). Leadership in organizations (8th ed.). Essex, England: Pearson Education.
