Believing you are more moral, honest, and virtuous than the average person, even without evidence to support it.
Imagine everyone in your class was asked to draw a picture of how good a person they are. Almost everyone would draw themselves taller and shinier than everyone else. Even the kid who just stole someone's crayon would draw himself as the nicest kid in the room. That's what our brains do—they make us feel like we're better, kinder, and more fair than everyone around us, even when we're really about the same.
The Illusion of Moral Superiority is a uniquely potent form of the better-than-average effect in which people systematically overestimate their own moral qualities—such as honesty, fairness, and principled behavior—relative to others. Research demonstrates that while people also inflate their competence and sociability, the inflation in the moral domain is the strongest, most widespread, and most resistant to correction. Critically, this moral self-enhancement persists even when controlling for rational inferences from limited information about others, meaning it reflects genuine irrational bias rather than a reasonable inference gap. The bias creates a dangerous feedback loop: because morality is both highly valued and inherently ambiguous in its behavioral indicators, individuals have unusually wide latitude to interpret their own actions as morally motivated while attributing less noble motives to others.
The same glitch looks different depending on the terrain. Finance, medicine, a relationship, a team — same mechanism, different costume.
Investors often believe they are more ethically principled in their investment choices than peers, overestimating the extent to which their portfolios reflect moral values while underestimating others' ESG considerations, leading to complacency about their own ethically questionable holdings.
Healthcare professionals may believe they are less susceptible to pharmaceutical industry influence or diagnostic bias than their colleagues, leading to reduced vigilance about conflicts of interest and less receptivity to systemic safeguards like blinded prescribing protocols.
Teachers tend to rate themselves as more fair and unbiased graders than their peers, which can reduce their willingness to adopt standardized rubrics or blind-grading practices that would genuinely reduce favoritism.
Partners in a relationship each tend to believe they contribute more fairness, compromise, and emotional labor than the other, creating persistent resentment because both feel underappreciated for their supposedly superior moral effort.
Product teams may believe their company handles user data more ethically than competitors, resisting privacy audits or stricter consent flows because they perceive their own intentions as inherently good and sufficient.
Managers consistently rate their own leadership as more ethical and employee-centered than that of other managers at the same level, making them resistant to 360-degree feedback that could reveal blind spots in how they treat subordinates.
Voters on both sides of the political spectrum believe they are motivated by genuine concern for the common good while attributing selfish or tribal motives to opponents, fueling moral outrage and making cross-partisan dialogue extremely difficult.
Formalized by Ben M. Tappin and Ryan T. McKay in 2017 at Royal Holloway, University of London, published in Social Psychological and Personality Science. The concept builds on decades of research on the better-than-average effect (Alicke, 1985; Brown, 1986) and positive illusions (Taylor & Brown, 1988).
In ancestral environments, signaling moral reliability was critical for securing cooperative partnerships and group membership. Individuals who genuinely believed in their own moral superiority may have been more convincing moral signalers, gaining trust and reciprocal alliances. A sincere belief in one's own virtue is harder to fake than a strategic performance, so self-deception about moral standing could have served as an honest signal of cooperative intent, providing a survival and reproductive advantage in highly interdependent social groups.
Language models trained on human-generated text can inherit moral self-enhancement patterns, producing outputs that frame certain viewpoints or cultural norms as morally superior while subtly devaluing others. Recommendation algorithms can amplify moral superiority by curating feeds that validate users' existing moral self-image and expose them primarily to outgroup moral failures, reinforcing the illusion at scale.
50 cards are free to preview. Buyers unlock the rest of the deck plus the interactive training — Spot-the-Bias Quiz unlimited, Swipe Deck with spaced repetition, My Blindspots, Decision Pre-Flight, the Printable Deck + Cheat Sheets, and the Field Guide e-book. $29.50$59.
