← The Casebook

The bench · equal acts, unequal verdicts

Moral Luck

The tendency to judge people more harshly for actions that lead to bad outcomes than for identical actions with neutral outcomes, even when the outcome was beyond their control.

Both pans hold the identical decision. Only the result differs — yet the beam tips toward the outcome.

Why the verdict bends

Multiple mechanisms have been proposed. The outcome bias account (Cushman 2008) suggests that bad outcomes directly increase blame independent of mental states. The hindsight bias account (Kneer & Machery 2019) argues that after learning of a bad outcome, people retrospectively judge the harm as more probable and the agent as more negligent. Young et al. (2010) found that belief justification mediates much of the effect—false beliefs are seen as less justified, driving increased blame. Recent evidence suggests outcome effects may be largely driven by information leakage in between-subjects designs.

Competing account — The rationalist account (Richards 1986; Rosebury 1995) holds that false beliefs genuinely are less justified, while the irrationalist account (Royzman & Kumar 2004) views this as a hindsight bias. Kneer & Machery (2019) challenge whether a genuine 'Difference Intuition' exists at all.

Cases brought to the bench

Case 01Bathroom scenario fMRI studypartial eta-squared = 0.83 for truth effect on blameworthiness

False beliefs contributed more to moral luck than bad outcomes. Agents with false beliefs were judged more blameworthy even when no harm occurred. Moral judgment differences were mediated by belief justification ratings.

Young et al. · 2010 · n = 24 (behavioral), n = 42 (behavioral 2), n = 19 (fMRI)

Case 02Within-subjects moral luck replicationeta-squared = 0.21 for moral luck effect in between-subjects

In within-subjects designs, the vast majority (85-90%) judged lucky and unlucky agents identically for wrongness, blame, and permissibility. Between-subjects designs showed outcome effects mediated by hindsight bias and probability ascriptions.

Kneer & Machery · 2019 · N = 2043 across 10 experiments

Case 03Moral luck in investment contextsnot reported

Profitable investments were rated more moral than unprofitable ones with identical decision processes. Participants changed moral evaluations and relabeled investments as 'speculation' or 'gambling' after learning of losses.

Max & Uhl · 2023 · not reported

tips under some framings

Evidence — mixed. The effect is robust in between-subjects designs (unlucky agents judged more harshly) but largely disappears in within-subjects designs where most participants judge lucky and unlucky agents identically. Kneer & Machery (2019) found that most people do not endorse the 'Difference Intuition' when explicitly asked, and their concrete case judgments align with abstract principles against moral luck. The philosophical 'Puzzle of Moral Luck' may therefore rest on a shaky empirical foundation.

Judged in the wild

  • law · 2003

    Ryan Holle felony murder conviction

    In 2003, Ryan Holle lent his car to a roommate who used it in a burglary that resulted in a death. Holle was convicted of felony murder and sentenced to life in prison despite having no direct involvement in the killing. Teehan (Princeton) uses this case to illustrate how legal systems inconsistently incorporate moral luck—Holle had 'bad luck' that his car was used in a homicide.

  • finance · 2023

    Moral luck in investment decisions

    Max and Uhl (2023) showed that identical financial decisions were judged less moral when they led to losses. Participants not only evaluated unprofitable investments more harshly but also reinterpreted their nature—relabeling them as 'speculation' or 'gambling' rather than 'investment' after learning of a loss.

To weigh the act, not the result

Use within-subjects comparison—explicitly compare the lucky and unlucky agent's decisions side by side. Research by Kneer & Machery (2019) shows that when people directly compare cases, about 85-90% judge them identically. Also focus on the decision process and agent's mental state (beliefs, intentions) rather than the outcome.

In within-subjects designs, the vast majority of participants (85-90%) judge lucky and unlucky agents as equally blameworthy, and their concrete judgments align with abstract principles against moral luck (Kneer & Machery, 2019).

Catch it in the act

Watch for situations where someone is blamed more for an identical decision because the outcome was worse—e.g., a drunk driver who hits a pedestrian vs. one who doesn't, or an investment that loses money vs. an identical strategy that profits.

Related judgment errors

File your own case

Open the same case on your own draft.

Paste a memo, a research draft, or a strategy argument. It is scored against all 175 cards, and the strongest two or three risks come back with the evidence quoted and one practical next check.

Open a case on your draft →