Moral Licensing and Racial Sensitivity: Does Being Good Make You Bad?

Introduction

Morality is a social construct, with individuals deriving their ethical standards from social and cultural frameworks. It should come as no surprise, therefore, that moral behavior is not determined solely by an individual's personal beliefs, but also by how that individual perceives his or her actions will be viewed by others. Moral licensing refers to the notion that prior behavior — either one's own or that of others in one's group — can influence an individual's subsequent behavior. "Moral self-licensing occurs because good deeds make people feel secure in their moral self-regard" (Merritt et al., 2012). This feeling of security appears to discourage people from questioning their internal moral impulses and comparing them to cultural norms and standards of behavior, which, rather than increasing their likelihood of engaging in moral behavior, actually increases their likelihood of engaging in immoral behavior.

Moral licensing is particularly evident when one examines bigotry and prejudiced behavior. The theory is that people are more willing — rather than less willing — to "express prejudiced attitudes when their group members' prior behavior has established non-prejudiced credentials" (Kouchaki, 2011). This effect is exacerbated when the individual identifies strongly with group members, and it occurs even when the individual was not personally involved in the original non-prejudicial decision-making process.

Aquino and Reed examined the social and cultural influences on moral psychology by investigating the associations between an individual's view of moral identity, their thoughts about morality, and their actual behavior. What they discovered was that moral self-concepts, shaped in part by social and cultural influences, can help explain individual moral conduct (Aquino & Reed, 2002). This is important when one considers how dominant ethnic groups respond to racist or non-racist treatment of minorities. Although prevailing social and cultural norms suggest that racism is unacceptable, they simultaneously reinforce stereotypes that portray minority members as inferior.

The result of these internalized stereotypes appears to be that majority group members feel virtuous when they refrain from racism. This seems to have two direct effects on individual behavior. First, people seek a moral license to act in negative ways if they have previously acted morally, particularly when they had the opportunity to behave immorally (Effron et al., 2012). Notably, the moral licensing effect seems to occur even when the audience is unaware of the individual's prior non-prejudiced behavior (Monin & Miller, 2001). Furthermore, when people are made to feel insecure about their morality, they not only point to prior moral behavior as justification but also tend to exaggerate the immoral alternatives that were available to them (Effron et al., 2012).

Research also suggests that explicit measures of an individual's moral personality cannot reliably predict specific moral actions (Perugini & Leone, 2009). Rather than relying on explicit measures, implicit measures of moral personality appear to have greater predictive power of actual moral behavior. To examine implicit measures, Perugini and Leone developed an Implicit Association Test (IAT) to measure moral versus immoral self-concept, and discovered that the IAT was highly predictive of actual moral behavior (2009).

Theoretical Background and Prior Research

This study examines the relationship between moral licensing, explicit moral self-concept, implicit moral self-concept, racism sensitivity, and pro-social behavior in non-Aboriginal students who have lived in Australia for at least one year. A total of 107 psychology students participated. Some participants were exposed to a moral licensing activity — a crime vignette in which they had the opportunity to make a non-racist judgment when identifying a suspect. These participants were then compared to a control group on four dependent variables: explicit moral self-concept, implicit moral self-concept, racism sensitivity, and pro-social behavior.

The belief underlying the study design was that, rather than encouraging pro-social and highly moral behavior, the opportunity to engage in a non-racist behavior would encourage participants to be less moral in subsequent tasks. The four hypotheses were as follows:

H1: Moral licensing will not impact explicit moral self-concept.
H2: Moral licensing will have a negative impact on implicit moral self-concept.
H3: Moral licensing will make participants less racially sensitive.
H4: Moral licensing will make participants less likely to volunteer than control participants.

Study Design and Hypotheses

This study examined the relationship between moral licensing and four distinct concepts: explicit moral self-concept, implicit moral self-concept, racial sensitivity, and volunteerism. Participants were assigned to either a control condition — in which both suspects in a moral licensing crime vignette were Anglo-Australian — or a moral licensing condition, in which one suspect was Anglo-Australian and the other was Aboriginal. In the vignette, participants had not seen the suspect's face but were able to provide a general description to police. Participants were therefore forced to choose between a suspect with no alibi, a history of petty theft, and significant property on his person at the time of detention, and a suspect who was at work during the crime, had no prior record, and had only a small amount of property on him. This design gave participants not only the opportunity to make a non-racist assumption about crime, but actually made the non-racist choice the more logical one. Making that choice then provided participants with the moral license to engage in less moral behavior in subsequent tasks.

The first hypothesis was that moral licensing would not impact explicit moral self-concept. This prediction was supported. Explicit moral self-concept focuses on how a person internalizes qualities considered moral, but these scores only reflect individuals' self-image of their morality, which may bear little relationship to how they actually behave. It is therefore unsurprising that the moral licensing exercise did not affect subjects' scores on the explicit moral self-concept measure.

Discussion of Findings

The second hypothesis concerned implicit moral self-concept, measured using the IAT. The IAT is believed to be highly predictive of actual behavior because it does not simply ask how a person self-identifies morally. Given the strong social and cultural pressure to self-identify as a moral person, self-report measures are susceptible to social desirability bias. The IAT instead measures reaction time: participants who respond more quickly to moral-compatible pairings than to immoral-incompatible pairings are understood to have a stronger implicit moral self-concept, since moral identification may be more intrinsic and automatic rather than deliberate. The results of this study demonstrated that moral licensing can have a negative impact on this implicit measure. Participants in the control condition scored higher on the IAT than those in the moral licensing condition, suggesting that moral licensing not only influences behavior but also affects how a person views his or her own moral standing.

The third hypothesis was that moral licensing would make participants less racially sensitive. Given that prior research has shown moral licensing to increase the likelihood of prejudiced behavior following a non-prejudiced act, it was assumed that participants would be less attentive to social norms about racial sensitivity after the licensing exercise. However, this hypothesis was not supported. No significant difference was found between the control condition and the moral licensing condition on the measure of racial sensitivity. This suggests that subsequent prejudiced behavior following moral licensing may be a matter of conscious choice rather than the neglect of societal norms.

The fourth hypothesis was that moral licensing would make participants less likely to volunteer. Volunteerism is widely regarded as a positive, moral, pro-social behavior — and importantly, it is a behavior unrelated to racial prejudice or culturally specific norms about race. This hypothesis was supported: participants in the moral licensing condition were less likely to volunteer than those in the control condition. This finding suggests that moral licensing has an umbrella effect, such that behavior seen as pro-social in one domain can lead to negative behaviors in entirely unrelated domains.

This research, like prior work in the field, challenges some deeply held assumptions about moral behavior. The prevailing cultural assumption has been that requiring people to engage in pro-social behavior will promote further positive behavior. On the contrary, this study suggests that requiring people to engage in pro-social behavior may actually encourage them to engage in negative behavior thereafter.

The implications could be significant. For example, many offenders are required to perform community service — in part on the theory that doing good will make them better people. What this research suggests is that mandatory community service may, paradoxically, encourage more negative behavior rather than less. More broadly, policies and programs that rely on one-time prosocial acts as a means of reforming attitudes or behavior may need to account for the moral licensing effect if they are to achieve their intended goals.

Aquino, K., & Reed, A. (2002). The self-importance of moral identity. Journal of Personality and Social Psychology, 83, 1423–1440.