In the spring of 2001, Benoit Monin and Dale Miller ran a series of experiments at Stanford University that produced one of the most uncomfortable findings in contemporary social psychology. Their core procedure was elegantly simple: some participants were first given an opportunity to disagree with a sexist statement — a chance to go on record as someone who held egalitarian views. Others were given no such opportunity. Then both groups were asked to recommend which job applicants should be hired for a physically demanding job that some people might consider better suited to men.
The results were unambiguous and unsettling. Participants who had already established their egalitarian credentials — who had, in effect, demonstrated that they were not sexist — were subsequently more willing to discriminate against female applicants. Their prior good act had functioned as a permission slip. Having proven they were not prejudiced, they felt freer to act in a prejudiced way. Monin and Miller called this effect "moral credentials," and their paper, published in the Journal of Personality and Social Psychology in 2001 under the title "Moral Credentials and the Expression of Prejudice," launched a decade of intensive research into one of behavioral science's most counterintuitive mechanisms.
Nine years later, Nina Mazar and Chen-Bo Zhong at the University of Toronto published a study in Psychological Science (2010) that demonstrated a structurally identical effect in the domain of consumer behavior. Participants who had been asked to shop in an online store that featured environmentally friendly "green" products were, in a subsequent ostensibly unrelated task, significantly more likely to cheat on a math problem for financial gain and more likely to steal money from an envelope left by the experimenter than participants who had shopped in a conventional online store. Buying green had, for many participants, conferred a sense of moral credit that made ethical violations in an entirely different domain more psychologically available. The green shopping had been morally licensing.
These two studies, separated by nearly a decade, illustrate the full scope of the phenomenon: moral licensing operates across domains, across acts of commission and omission, and across contexts ranging from hiring decisions to petty theft. The mechanism is neither trivial nor exotic. It is, as subsequent research has shown, one of the most reliable ways that a sincere commitment to being good actually undermines the behavior that commitment was supposed to produce.
"Past good deeds can liberate individuals to engage in behaviors that are immoral, unethical, or otherwise problematic." — Anna Merritt, Daniel Effron & Benoît Monin, 2010
What Moral Licensing Is
Moral licensing is the psychological phenomenon in which a prior virtuous act, commitment, or identity marker reduces a person's motivation to behave ethically in subsequent situations, because the earlier goodness generates a subjective sense of moral credit that temporarily satisfies the need to maintain a positive moral self-image.
Moral Licensing vs. Moral Consistency
| Dimension | Moral Licensing | Moral Consistency |
|---|---|---|
| Effect of prior good act | Increases willingness to behave badly in subsequent situations | Increases motivation to maintain ethical behavior going forward |
| Self-concept mechanism | Prior virtue satisfies the moral identity need, freeing the person to indulge | Prior virtue reinforces and activates the moral identity, making violations more threatening |
| Relationship between acts | Acts are substitutes: one good act "pays for" a future bad one | Acts are complements: good acts reinforce the identity that motivates further good acts |
| Temporal direction | Backward-looking: uses past credit to authorize present behavior | Forward-looking: uses past behavior as evidence of who one is going forward |
| Triggering condition | Prior act is perceived as sufficiently virtuous to establish credentials | Prior act is perceived as constitutive of a moral self-concept rather than merely credentialing |
| Likely outcome | Moral drift: gradual erosion of ethical standards over time | Moral consolidation: behavioral integrity strengthening across situations |
| Key predictor | Whether the prior act was experienced as "proving" one's goodness | Whether the prior act was experienced as "expressing" one's values |
The critical insight is that the same prior good act can produce either moral licensing or moral consistency depending on how it is construed. Sachdeva, Iliev, and Medin (2009), writing in Psychological Science, demonstrated this distinction directly: when participants wrote a self-affirming story using positive trait words, they subsequently donated less to charity than those who wrote a story using negative trait words — evidence for licensing. But when the prior act was framed as an expression of core identity rather than as evidence of credential, the licensing effect was substantially attenuated.
Cognitive Science: The Mechanism
Moral Self-Regulation as a Homeostatic System
The foundational theoretical framework was articulated by Sachdeva, Iliev, and Medin in their 2009 paper "Sinning Saints and Saintly Sinners: The Paradox of Moral Self-Regulation," published in Psychological Science (Volume 20, Issue 4, pages 523-528). The authors proposed that moral self-regulation operates through a homeostatic balance mechanism: people maintain a target level of moral identity, and deviations in either direction produce corrective pressure. When moral identity falls below the target — after a bad act — people experience guilt and increased motivation to behave well. When moral identity rises above the target — after a particularly virtuous act — the homeostatic pressure works in the opposite direction: the surplus of moral credit reduces the urgency of ethical behavior and makes unethical behavior more accessible.
This balancing model treats moral identity like a thermostat: the system works to maintain an equilibrium, not to maximize goodness. The implication is that people who have just done something admirable are not, by that fact, more likely to continue doing admirable things. They are, in a precise sense, less likely to, because their moral account is temporarily in surplus.
The Role of Moral Self-Concept Maintenance
Chen-Bo Zhong and Katie Liljenquist (2006), in a paper titled "Washing Away Your Sins: Threatened Morality and Physical Cleansing," published in Science (Volume 313, pages 1451-1452), demonstrated that moral and physical purity are processed through overlapping cognitive systems. Participants who recalled an unethical act of their own were more likely to prefer a cleansing product over a non-cleansing product. Participants given an opportunity to physically clean their hands after recalling the unethical act showed reduced motivation to volunteer for a subsequent charitable activity. Physical cleansing, like moral credentialing, temporarily discharged the motivational system that drives ethical behavior.
The implication, which Mazar and Zhong (2010) developed, is that moral self-regulation draws on a self-concept maintenance system that is relatively undiscriminating about how the maintenance is achieved. The system's goal is to preserve a positive moral self-image. A prior virtuous act achieves this goal. Physical cleansing achieves it to a detectable degree. What the system does not achieve, under licensing conditions, is motivate continued virtue.
The Credentials vs. Values Distinction
Anna C. Merritt, Daniel A. Effron, and Benoit Monin — in their landmark review paper "Moral Self-Licensing: When Being Good Frees Us to Be Bad," published in Social and Personality Psychology Compass in 2010 (Volume 4, Issue 5, pages 344-357) — synthesized the literature and identified a key moderating variable: whether a prior act functions as a moral credential or as an expression of moral values.
A credential is evidence — it proves something about who you are by providing an instance of behavior. A value expression is constitutive — it does not prove anything but rather enacts who you are. When people interpret a prior good act as providing evidence of their virtue (credential), licensing follows, because the evidence has been collected and the case is made. When they interpret the same act as an expression of values they hold (expression), consistency follows, because values are not "used up" by expressing them.
Effron and Monin (2009), writing in the Journal of Personality and Social Psychology (Volume 96, Issue 3, pages 513-525), tested this directly in the context of race and politics. White participants who had voted for Barack Obama in the 2008 presidential election — or who were asked to imagine doing so — were subsequently more willing to express preferences for white candidates over Black candidates in certain hiring contexts. The vote had provided a credential: it had established that the participant was not racially biased. That credential then licensed bias that the participant might otherwise have been more vigilant to suppress.
Goal Progress and Licensing: Fishbach and Dhar
Ayelet Fishbach and Ravi Dhar provided the theoretical bridge between goal research and moral licensing in their 2005 paper "Goals as Excuses or Guides: The Liberating Effect of Perceived Goal Progress on Choice," published in the Journal of Consumer Research (Volume 32, pages 370-377). Fishbach and Dhar found that when participants were reminded of progress toward a goal — having already exercised, having already eaten healthily — they were more likely to choose an indulgent option in a subsequent decision than participants who received no reminder. The key finding was that this licensing occurred only when goal progress was framed as an achievement signal rather than as evidence of ongoing commitment. Progress, when interpreted as an achievement ("I have already done X"), licenses departure. Progress interpreted as evidence of commitment ("I am the kind of person who does X") predicts consistency.
This study is important because it shows the licensing mechanism operating entirely within conscious goal-pursuit, without invoking identity or credentials at all. A person who exercises for health and then eats a large dessert is not being strategic or hypocritical. They are responding to a signal their goal-pursuit system is generating — a signal that the goal is sufficiently satisfied to permit a departure.
Intellectual Lineage
From Cognitive Dissonance to Credential Theory
The intellectual antecedents of moral licensing theory run through three distinct traditions in twentieth-century social psychology.
The most proximate ancestor is Leon Festinger's cognitive dissonance theory, introduced in his 1957 monograph A Theory of Cognitive Dissonance (Stanford University Press). Festinger proposed that inconsistency between held cognitions produces an aversive motivational state — dissonance — that drives the person to reduce the inconsistency. Elliot Aronson extended this in 1969, arguing in Advances in Experimental Social Psychology (Volume 4) that dissonance is most powerful when behavior threatens a person's central self-beliefs, particularly their belief in their own moral competence. The self-concept as a good person is confirmed by virtuous acts and threatened by moral failures. What Monin and Miller recognized was that this accounting runs in both directions. If moral failure creates pressure toward remediation, moral success creates a surplus that can be drawn on — a possibility Festinger's framework anticipated but did not develop.
The second major lineage runs through Roy Baumeister's work on self-regulation, particularly his 1998 paper "Ego Depletion: Is the Active Self a Limited Resource?", published with Bratslavsky, Muraven, and Tice in the Journal of Personality and Social Psychology (Volume 74, pages 1252-1265). Baumeister proposed that self-control draws on a finite resource that can be depleted through use. Moral licensing is related but distinct: it is not primarily about resource depletion through effortful resistance, but about the symbolic meaning a prior act confers. The mechanism is not exhaustion but permission. Later replication failures of ego depletion effects (Hagger et al., 2016, Perspectives on Psychological Science) did not touch the core licensing mechanism, which depends on self-concept rather than resource availability.
The third lineage runs through Daryl Bem's self-perception theory (1967, Psychological Review, Volume 74, pages 183-200). Bem proposed that people infer their own attitudes and internal states in much the way they infer others': by observing their own behavior and drawing conclusions from it. If you see yourself donating to charity, you conclude you are a generous person. This self-attribution process is the mechanism through which moral credentials are established: prior good behavior creates the attribution "I am a good person," which then licenses less careful subsequent behavior.
Historical Antecedents: The Indulgence as Psychological Structure
The phenomenon also has deep roots in moral philosophy and religious practice, though the psychological literature rarely acknowledges this directly. The Catholic practice of purchasing indulgences — remission from purgatorial punishment in exchange for prayers, pilgrimages, or donations — against which Martin Luther nailed his Ninety-Five Theses in October 1517, represents perhaps the most elaborate institutional expression of moral credit-spending in Western history. Within that theological framework, a pilgrim who walked to Santiago de Compostela, or a donor who financed the construction of St. Peter's Basilica, was understood to have accumulated spiritual merit that modified the moral ledger. The psychological structure is identical to what Monin and Miller identified six centuries later: prior virtue generates transferable credit that authorizes subsequent departure from ideal conduct.
Claude Steele's self-affirmation theory (1988, Advances in Experimental Social Psychology, Volume 21, pages 261-302) established that the mind seeks to maintain a global sense of self-integrity rather than behavioral consistency in any specific domain. When one aspect of the self is threatened, affirming an entirely different aspect can reduce the motivation to defend the threatened area. This logic runs through the licensing literature: prior moral acts affirm global self-integrity in a way that reduces motivation to maintain specific ethical standards.
The 2001 Inflection Point
Monin and Miller's 2001 paper is the clear empirical origin of the modern literature. Before it, the phenomenon had been observed in scattered applied contexts but had not been named, systematically studied, or theoretically integrated. The moral credentials framework they provided was parsimonious and generative: it identified the mechanism (credential-based self-assessment), the psychological function (establishing that one is not prejudiced, selfish, or immoral), and the counterintuitive consequence (subsequent behavioral latitude in exactly the domains where goodness has been proven).
The decade that followed produced a rapid expansion of the research program: from prejudice to consumer behavior (Khan and Dhar, 2006), from consumer behavior to honesty and theft (Mazar and Zhong, 2010), from single-act credentials to identity-level licensing (Merritt, Effron, and Monin, 2010), from behavioral to political domains (Effron and Monin, 2009), and from individual to organizational contexts (Kouchaki, 2011).
Empirical Research: Key Studies and Findings
Monin and Miller 2001: The Founding Experiments
Monin and Miller's original paper in the Journal of Personality and Social Psychology (Volume 81, Issue 1, pages 33-43) reported four experiments. In the first, participants were or were not given the opportunity to disagree with a prejudiced statement, and were then asked to recommend a job candidate. Those who had established their egalitarian credential showed significantly more willingness to favor a white male candidate over an equally qualified female candidate.
In a subsequent experiment, Monin and Miller showed that the licensing was domain-sensitive: credentials established in the racial domain licensed subsequent behavior in a racial context more than in a gender context, and vice versa. This finding ruled out simple "global positivity" accounts of the effect. The credential needed to be relevant to the domain in which the licensed behavior operated, suggesting that the mechanism ran through the specific self-assessment — "I am not racially biased" — rather than through a diffuse "I am a good person" self-evaluation.
The mechanism Monin and Miller proposed was specifically about social presentation and self-concept, not merely attitudes. Having established that you are not biased, you have less reason to fear that any particular subsequent act will be interpreted as bias. The credential functions as a buffer against the inference others — or you yourself — might draw from your behavior. You can afford to make the decision that coincidentally favors the majority candidate, because your established identity as a fair evaluator means it will not read as bias, even if, statistically, it is.
Khan and Dhar 2006: Virtuous Consumer Choice Licenses Indulgence
Uzma R. Khan and Ravi Dhar published "Licensing Effect in Consumer Choice" in the Journal of Marketing Research in 2006 (Volume 43, Issue 2, pages 259-266). Their studies demonstrated that participants who had made or considered making a virtuous consumer choice — such as purchasing a product associated with charitable giving or choosing a low-calorie food option — were subsequently more likely to choose indulgent, hedonic options in an unrelated choice.
In one study, participants who had spent time considering a product described as beneficial to society were subsequently more likely to choose a luxury item over a more practical one. The mere contemplation of virtue, not just its enactment, was sufficient to generate licensing in the consumer domain. The licensing effect required that the virtuous choice be freely made — consistent with the interpretation that it was the self-relevant meaning of the act, not the act itself, that did the work. This finding has significant implications for cause-related marketing campaigns: brand associations with charitable giving or environmental responsibility may inadvertently license consumers to make less conscientious choices in adjacent decisions.
Mazar and Zhong 2010: Green Products and Theft
The Mazar and Zhong study, published in Psychological Science (Volume 21, Issue 4, pages 494-498) under the title "Do Green Products Make Us Better People?", used a behavioral economics paradigm that directly measured actual ethical violations rather than attitudes or preferences.
In their primary experiment, participants were randomly assigned to shop in an online store featuring either conventional products or green (environmentally friendly) products. After shopping, they participated in a matrix task — a standard cheating paradigm — in which they could overreport the number of math problems they had solved to earn money. They also participated in a coin-allocation task in which they had the opportunity to take money from an envelope.
The green-shopping condition produced significantly higher rates of both cheating and theft. Participants who had made green purchases showed a 36% higher rate of dishonesty on the matrix task and were more likely to steal from the envelope than those who had shopped in the conventional store. This is among the most striking findings in the licensing literature: not just reduced motivation to behave well, but increased willingness to commit what, in any other framing, would be recognized as theft.
The authors' interpretation follows the self-concept maintenance model: purchasing green products had temporarily satisfied the moral self-regulation system, creating surplus moral credit that reduced the inhibitory motivation against small-scale dishonesty. The participants who stole money after buying green products were not cynically trading environmental credits for cash. They were operating through a cognitive mechanism they did not know was running.
Sachdeva, Iliev, and Medin 2009: Sinners and Saints
The Psychological Science paper by Sachdeva, Iliev, and Medin (Volume 20, Issue 4, pages 523-528) provided the most direct test of the homeostatic balance model. Participants were asked to write a brief story about themselves using either positive traits (generous, caring, friendly, hard-working, honest) or negative traits (greedy, mean, dishonest, selfish, lazy). After writing the story, they were asked whether they would like to donate a portion of their $10 participation payment to charity, and if so, how much.
Participants who had written a self-affirming story using positive traits donated an average of $1.07 to charity. Participants who had written a self-diminishing story using negative traits donated an average of $2.71 — more than two and a half times as much. The prior self-affirmation licensed reduced generosity; the prior self-diminishment motivated compensatory generosity.
A crucial additional finding: when the positive-trait writing task was framed not as a self-affirmation exercise but as an expression of authentic values — emphasizing that the traits were genuinely the participant's own — the licensing effect was significantly attenuated. Participants in this framing condition donated amounts closer to those of the negative-trait group. This is the empirical demonstration of the credentials-versus-values distinction: the same prior behavior produces licensing or consistency depending on whether it is experienced as proving something about the self or as expressing something constitutive of the self.
Effron and Monin 2009: The Obama Licensing Effect
The political domain study, published in the Journal of Personality and Social Psychology (Volume 96, Issue 3, pages 513-525), was conducted in the immediate aftermath of the 2008 U.S. presidential election. Effron and Monin tested the hypothesis that the act of supporting Barack Obama's candidacy — which, for white voters, functioned as a public signal of racial open-mindedness — would license subsequent racially discriminatory choices.
White participants who were asked to recall or imagine their vote for Obama subsequently showed greater willingness to endorse policies that disadvantaged African Americans and showed less discomfort with racial stereotyping than participants who had recalled or imagined voting for the opposing candidate. In a direct behavioral measure, participants who had established their Obama-supporting credentials were more willing to recommend that a qualified black applicant be passed over for a job in favor of a white applicant — framing the recommendation as performance-based rather than race-based.
Effron and Monin were not arguing that supporting Obama caused racism. They were documenting a specific psychological mechanism through which a signal of progressive racial attitudes — widely experienced as an affirmation of non-racism — could, in certain contexts, reduce the cognitive vigilance that normally suppresses racially discriminatory responses. The credential "I voted for the black candidate" was functioning as permission to not think carefully about race in subsequent decisions.
Merritt, Effron, and Monin 2010: The Review That Mapped the Field
Anna C. Merritt, Daniel A. Effron, and Benoit Monin's 2010 review in Social and Personality Psychology Compass (Volume 4, Issue 5, pages 344-357), "Moral Self-Licensing: When Being Good Frees Us to Be Bad," synthesized the first decade of empirical work and distinguished three varieties of the phenomenon.
Behavioral licensing occurs when a specific prior good act generates credit that is spent in a subsequent act. This is the core experimental paradigm. Credential-based licensing occurs when a prior act provides evidence of a moral identity — not just that one has behaved well but that one is the kind of person who would never behave badly — and that evidence reduces vigilance in subsequent decisions. Anticipated virtue licensing occurs when a person plans to do something good in the future, and that anticipated goodness licenses present indulgence. The tripartite structure the review identified — past behavior, established identity, anticipated action — organized subsequent research and identified anticipated virtue licensing as the least studied but potentially the most consequential form, given how easily human planning can generate licensing without requiring any actual virtuous behavior.
Four Named Case Studies
Case Study 1: Carbon Offsets and Climate Behavior
The purchase of carbon offsets — voluntary payments to fund emission-reducing projects intended to compensate for one's own carbon footprint — has been studied as a naturalistic moral licensing environment. A 2007 paper by Michael Vandenbergh, Jack Barkenbus, and Jonathan Gilligan in the Tennessee Law Review examined why individual carbon offset purchasers showed, in consumer surveys, reduced likelihood of undertaking direct behavioral changes — increasing household efficiency, reducing air travel, changing dietary patterns — compared to non-purchasers who expressed similar levels of environmental concern.
The mechanism is direct: the offset purchase functions as a purchased moral credential. Having paid for the equivalence, the moral account is balanced, and the homeostatic self-regulation system relaxes its pressure toward behavioral change. The paradox is that carbon offsets — designed to amplify climate-positive behavior by making emissions offsetting accessible — may, through the licensing mechanism, partially substitute for the behavioral changes they were intended to supplement. A consumer who purchases annual carbon offsets and uses that purchase as a basis for not reconsidering their driving, flying, or dietary habits is not being consciously cynical. They are behaving in a way that the moral licensing literature predicts precisely: the credential has been established, and the regulatory pressure it was generating has been discharged.
The practical implication is one of the more troubling in the applied licensing literature: voluntary environmental programs may be most effective at changing behavior in populations where concern is high enough to motivate action but where no licensing credential yet exists. Once the credential is established — through offset purchase, organic food buying, or green product acquisition — it may actively reduce motivation toward the more consequential behavioral changes that environmental problems actually require.
Case Study 2: Corporate Social Responsibility and Regulatory Violations
Matthew Kotchen and Jon Jungwoo Moon published "Corporate Social Responsibility for Irresponsibility" in the B.E. Journal of Economic Analysis and Policy in 2012 (Volume 12, Issue 1). The study matched data on corporate social responsibility scores from the KLD Research and Analytics database against data on regulatory violations, legal settlements, and fines across a sample of U.S. companies between 1991 and 2005.
Their finding was statistically robust and directionally consistent with a licensing interpretation: companies with higher CSR scores showed a higher incidence of regulatory violations. The correlation survived controls for industry, firm size, and profitability, suggesting it was not an artifact of larger companies both engaging in more CSR and facing more regulatory scrutiny. The authors considered multiple interpretations, including the possibility that companies proactively invest in CSR to build credential reserves for anticipated regulatory trouble — what might be called strategic pre-licensing. Both the causal interpretation (CSR generates licensing that produces laxity) and the strategic interpretation (companies anticipate needing the credential) are consistent with the moral licensing framework. The finding does not prove causation, but it is precisely what the licensing model predicts when applied to institutional actors.
Maryam Kouchaki extended this to the individual level within organizations. Writing in the Journal of Applied Psychology in 2011 (Volume 96, Issue 6, pages 1273-1283), Kouchaki demonstrated that employees who identified strongly with an organization they perceived as highly ethical showed greater willingness to engage in small-scale workplace dishonesty: padding expense reports, misrepresenting working hours, taking office supplies. The organization's ethical reputation had functioned as a collective moral credential that licensed individual moral slack.
Case Study 3: Diversity Programs and Individual Discrimination
Sonia Kang, Katherine DeCelles, Andreas Tilcsik, and Sora Jun published "Whitened Resumes: Race and Self-Presentation in the Labor Market" in Administrative Science Quarterly in 2016 (Volume 61, Issue 3, pages 469-502). Their audit study sent matched pairs of resumes — one with a racially identifiable name, one with a whitened name — to job advertisements posted by companies with and without explicit diversity commitments, diversity mission statements, or formal equal-opportunity employer designations.
The finding was counterintuitive from the perspective of naive credentialism: applicants with racial minority names were no less likely to face discrimination when applying to companies with explicit diversity commitments, and in some conditions faced marginally higher rates of callback differences. The organization's public commitment to diversity had not reduced individual hiring managers' reliance on racial bias. The licensing interpretation is direct: the organization's credential — its formal commitment to fairness — had created precisely the psychological condition that licensing predicts. Individual decision-makers operating within an organization that is publicly committed to non-discrimination may experience reduced pressure to be vigilant about their own biases, because the institutional credential absorbs what individual vigilance was previously required to maintain.
This finding has generated substantial debate in the organizational behavior literature, and the effect size is contested. But the directional finding — that diversity programs may not reduce, and may sometimes not improve, outcomes for minority applicants — has been replicated across multiple audit studies and is consistent with the licensing mechanism as one contributing factor.
Case Study 4: Medical Ethics, Disclosure, and Pharmaceutical Relationships
Sunita Sah and George Loewenstein published a series of papers between 2010 and 2015 examining the "disclosure paradox" — the finding that requiring physicians to disclose financial relationships with pharmaceutical companies sometimes increased rather than decreased the influence of those relationships on prescribing behavior.
The mechanism they identified was directly analogous to moral licensing. Disclosure, rather than motivating more conservative prescribing, sometimes led physicians to recommend more aggressively the products of companies with which they had disclosed relationships. The disclosure had functioned as a moral credential: having been transparent about the relationship, the physician felt less need to compensate for it through behavioral restraint. The act of disclosure — itself virtuous, a fulfillment of a professional and legal obligation — had licensed the very bias it was designed to counteract.
A 2013 paper by Sah, Loewenstein, and Daylian Cain in Psychological Science (Volume 24, Issue 2, pages 252-259) tested this in a controlled laboratory setting, finding that advisors who disclosed conflicts of interest subsequently gave more biased advice than advisors who did not disclose, and felt less guilty about doing so. Disclosure, the authors concluded, works at the systemic level — it allows patients and regulators to discount biased advice — but fails at the individual level because the disclosing physician experiences the act of disclosure as moral credential that licenses rather than restrains subsequent bias. This finding has been influential in medical ethics debates about whether disclosure requirements, absent behavioral constraints, may be insufficient or actively counterproductive.
Limits and Nuances
The Replication Landscape
Like much of social psychology, the moral licensing literature was affected by the replication crisis of the 2010s. A pre-registered meta-analysis by Irene Blanken, Niels van de Ven, and Marcel Zeelenberg, published in Personality and Social Psychology Bulletin in 2015 (Volume 41, Issue 4, pages 540-554), analyzed 76 studies and found significant heterogeneity in effect sizes. The mean effect was d = 0.31 — small to moderate — and substantially smaller than several high-profile original studies had suggested. Publication bias was detected, meaning that the true population effect is likely smaller than the published literature as a whole implies.
The Blanken et al. meta-analysis also identified important moderators: licensing effects were stronger when the prior act was explicitly framed as moral or virtuous; when the temporal gap between the prior act and the test was short; when participants were low in trait moral identity; and when the study design gave participants a clear sense that their virtuous act was noted or recorded. Studies that lacked these features showed weaker or no licensing effects. This pattern of moderation suggests that moral licensing is not a universal automatic process but a context-sensitive one that depends on how moral acts are psychologically experienced.
Domain Specificity and Transfer
The original Monin and Miller studies showed that licensing was domain-sensitive: a racial credential did not license gender discrimination as effectively as a gender credential did. Subsequent studies showed cross-domain transfer in some conditions. The conditions that predict cross-domain transfer — versus domain-specific effects — have not been fully worked out.
Daniel Effron has proposed, in work at the London Business School published between 2012 and 2018, that the relevant unit of licensing is the global moral self-concept rather than domain-specific credentials. When a prior act activates a global "I am a good person" self-assessment, transfer is broad. When it activates a domain-specific self-assessment ("I am not racially biased"), transfer is narrower. The degree to which individuals hold a unified moral self-concept versus a domain-structured one — which may vary with culture, trait moral identity, and context — should therefore predict the breadth of licensing transfer.
When Prior Virtue Produces Consistency, Not Licensing
Moral consistency — the finding that prior good acts motivate subsequent good acts — is also a real and documented phenomenon. The conditions that produce consistency rather than licensing are theoretically important. Merritt, Effron, and Monin's 2010 review identified three primary conditions under which consistency is predicted over licensing.
First, when the prior act is framed as an expression of core values rather than as a credential or achievement, consistency follows. This has been demonstrated directly by Sachdeva, Iliev, and Medin and confirmed by Jordan, Mullen, and Murnighan in their 2011 paper in Organizational Behavior and Human Decision Processes (Volume 116, Issue 2, pages 169-180): framing recall of a past good act as evidence of a commitment rather than a credential reduced the subsequent licensing effect and in some conditions produced behavioral consistency.
Second, when the prior act activates a specific behavioral goal rather than a global self-assessment, the goal-pursuit system motivates continued goal-relevant behavior rather than licensing departure. This is Fishbach and Dhar's contribution: the same prior act can license or motivate continued virtue depending on whether it is coded as "progress achieved" or "commitment active."
Third, when social accountability is present — when the person expects their subsequent behavior to be observed and evaluated — licensing is attenuated. The credential provides permission primarily in contexts of low surveillance. Public commitments, behavioral tracking, and external accountability structures shift the balance toward consistency.
Individual Differences
Research on individual differences has identified several moderating personality variables. Karl Aquino and Americus Reed II, writing in the Journal of Personality and Social Psychology in 2002 (Volume 83, Issue 6, pages 1423-1440), developed the moral identity scale, distinguishing between internalization (moral values are central to one's sense of self) and symbolization (moral values are expressed through public behaviors). Individuals high in internalization show weaker licensing effects: for them, a prior good act expresses rather than proves moral identity, generating consistency rather than license.
Subclinical psychopathy — characterized by reduced affective response to moral violations — predicts weaker licensing in both directions: the homeostatic moral self-regulation system operates at lower intensity, so neither the compulsion toward virtue nor its licensing consequence is as strong.
Regulatory focus theory, developed by E. Tory Higgins in 1997 in the Psychological Review (Volume 104, pages 411-423), predicts that promotion-focused individuals (oriented toward gain and advancement) should show stronger licensing than prevention-focused individuals (oriented toward loss avoidance). Empirical tests have provided partial support: promotion focus amplifies the licensing effect, while prevention focus — which treats any moral lapse as a potential loss — attenuates it.
Organizational and Institutional Scale
When licensing operates at the institutional level, its effects are potentially larger and harder to detect than at the individual level. An organization that has publicly committed to ethical conduct — through mission statements, ethics programs, or CSR initiatives — has established a collective moral credential. Individual members of the organization who identify with it inherit the credential, and may inherit the licensing it confers, without having performed any virtuous act themselves. Kouchaki's 2011 research documented this "vicarious licensing" directly. The implication is that organizational ethics programs, if they strengthen institutional moral identity without creating individual behavioral accountability, may achieve the opposite of their intended effect.
Conclusion
The permission slip that the brain writes itself is rarely examined or acknowledged. The person who steals money after buying organic vegetables does not think: I have earned the right to take what is not mine. The executive who relaxes due diligence after signing a corporate ethics pledge does not think: my charitable giving covers this. The voter who endorses racial inequality after supporting a Black candidate does not think: I have proven I am not biased, so I can afford to be. The mechanism operates below the level of deliberate reasoning, running through the self-concept maintenance system that monitors moral identity and signals when the account is in sufficient surplus to tolerate a departure from ideal conduct.
Monin and Miller's 2001 experiments were the beginning of a systematic investigation into this mechanism. The findings that have accumulated since — from Khan and Dhar's consumer choice studies, to Mazar and Zhong's green product experiments, to Sachdeva, Iliev, and Medin's story-writing paradigm, to Effron and Monin's political domain analyses — describe a coherent and reproducible psychological process with implications that extend from individual dietary choices to the design of organizational ethics programs and the interpretation of symbolic political progress.
The uncomfortable implication that runs through the literature is that sincerity is not sufficient protection against the licensing mechanism. The participants who stole money after buying green products were not cynical. The employees who padded expense reports after working for an ethical company were not strategically calculating. The physicians who gave more biased advice after disclosing their conflicts of interest were, in many cases, genuinely committed to non-bias. The mechanism does not require bad faith. It requires only that a prior virtuous act be experienced as establishing a credential — and that the self-regulation system respond to the credential by relaxing the vigilance that credential was supposed to reflect.
Understanding moral licensing does not dissolve the phenomenon, but it identifies the leverage points. The shift from credential framing to value expression framing — from "I have proven I am good" to "I care about this" — is the single most consistent moderator the literature identifies. Behavioral structures that focus on specific outcomes rather than global identity, accountability systems that maintain vigilance independently of prior good acts, and interventions that activate commitment rather than achievement are the practical responses that follow. They are less inspiring than the hope that past goodness will motivate future goodness. But the research is consistent on this point: past goodness, when construed as credential rather than commitment, is at least as likely to license future badness as to prevent it.
References
Monin, B., & Miller, D. T. (2001). Moral credentials and the expression of prejudice. Journal of Personality and Social Psychology, 81(1), 33-43.
Merritt, A. C., Effron, D. A., & Monin, B. (2010). Moral self-licensing: When being good frees us to be bad. Social and Personality Psychology Compass, 4(5), 344-357.
Khan, U. R., & Dhar, R. (2006). Licensing effect in consumer choice. Journal of Marketing Research, 43(2), 259-266.
Mazar, N., & Zhong, C.-B. (2010). Do green products make us better people? Psychological Science, 21(4), 494-498.
Sachdeva, S., Iliev, R., & Medin, D. L. (2009). Sinning saints and saintly sinners: The paradox of moral self-regulation. Psychological Science, 20(4), 523-528.
Effron, D. A., & Monin, B. (2009). Letting people off the hook: When do good deeds excuse transgressions? Journal of Personality and Social Psychology, 96(3), 513-525.
Zhong, C.-B., & Liljenquist, K. (2006). Washing away your sins: Threatened morality and physical cleansing. Science, 313(5792), 1451-1452.
Fishbach, A., & Dhar, R. (2005). Goals as excuses or guides: The liberating effect of perceived goal progress on choice. Journal of Consumer Research, 32(3), 370-377.
Kouchaki, M. (2011). Vicarious moral licensing: The influence of others' past moral actions on moral behavior. Journal of Applied Psychology, 96(6), 1273-1283.
Blanken, I., van de Ven, N., & Zeelenberg, M. (2015). A meta-analytic review of moral licensing. Personality and Social Psychology Bulletin, 41(4), 540-554.
Jordan, J., Mullen, E., & Murnighan, J. K. (2011). Striving for the moral self: The effects of recalling past moral actions on future moral behavior. Organizational Behavior and Human Decision Processes, 116(2), 169-180.
Kang, S. K., DeCelles, K. A., Tilcsik, A., & Jun, S. (2016). Whitened resumes: Race and self-presentation in the labor market. Administrative Science Quarterly, 61(3), 469-502.
Frequently Asked Questions
What is moral licensing?
Moral licensing is the psychological tendency for past virtuous behavior to increase the likelihood of subsequent unethical or self-indulgent behavior. Having established a positive moral self-image through a good act, people feel licensed to behave less virtuously without threatening their self-conception as a good person. Benoît Monin and Dale Miller introduced the concept in a 2001 Journal of Personality and Social Psychology paper showing that subjects who had first established moral credentials were more likely to subsequently express biased judgments.
What did the Monin and Miller 2001 study find?
Monin and Miller's foundational 2001 study had subjects either establish moral credentials by disagreeing with a sexist statement, or not. Subjects were then asked to recommend candidates for a job stereotypically suited to men. Those who had established moral credentials — by having previously appeared unprejudiced — were significantly more likely to recommend the male candidate. The credential acted as a license: having demonstrated they were not biased, subjects felt freer to make a biased recommendation without threatening their self-image as fair-minded people.
Do green products make people behave less ethically?
Nina Mazar and Chen-Bo Zhong's 2010 Psychological Science study tested this directly. Participants who purchased green products from a simulated store subsequently behaved more selfishly: they took more money in a subsequent allocation task and cheated more in a subsequent reporting task compared to participants who purchased conventional products. The green purchase established moral credentials that licensed subsequent self-interest. The effect was specific to actual purchasing, not mere exposure to green products.
How does moral licensing explain political behavior?
Daniel Effron, Jessica Cameron, and Benoît Monin's 2009 Journal of Experimental Social Psychology study showed that expressing support for Barack Obama's 2008 presidential candidacy subsequently licensed more racially biased decisions among white American subjects. Participants who had registered their support for Obama felt that this credential established their lack of racial prejudice, and were consequently more willing to express preferences that favored white candidates in other contexts. The study demonstrated that moral licensing operates with symbolic gestures as well as direct actions.
Can moral licensing be reduced?
Research suggests several strategies. Fishbach and Dhar's 2005 work on goal pursuit found that framing progress as a commitment to ongoing goals — rather than as an achievement that reduces obligation — reduces licensing. Identity-based interventions that frame virtue as a trait ('I am a healthy person') rather than an act ('I did something healthy') appear to produce positive spillover rather than licensing. Accountability mechanisms and explicit commitment to consistency can also counteract the effect, though no intervention has proven robust across all contexts.