The Psychology of Fairness: Why Random Draws Feel More Just Than Voting

Procedural Fairness vs. Outcome Fairness

Psychologists distinguish between two types of fairness:

• Distributive fairness — whether the outcome itself is fair (did the right person win?)
• Procedural fairness — whether the process used to reach the outcome is fair (was the selection process unbiased?)

A foundational finding in social psychology, established by John Thibaut and Laurens Walker in the 1970s, is that procedural fairness often matters more to people than distributive fairness. People will accept outcomes they personally dislike — losing a competition, being passed over — if they believe the process was fair. Conversely, they will reject outcomes they would have preferred if they believe the process was rigged.

Random selection derives its perceived fairness from procedural properties, not outcomes. A lottery winner chosen by algorithm is perceived as legitimate even by those who lost; a winner chosen by a committee may be resented even when the choice is objectively defensible.

Why Randomness Signals Impartiality

Random selection is perceived as fair because it is structurally immune to the biases that corrupt human judgment. A random number generator cannot favour the tall, attractive, or socially confident. It cannot be swayed by personal relationships, implicit prejudices, or self-interest. It applies the same probability to everyone.

This structural impartiality corresponds to John Rawls's philosophical concept of the veil of ignorance: a procedure is fair if we would choose it before knowing which outcome would benefit us personally. From behind the veil — not knowing whether we will be the one selected — a random draw is one of the few procedures everyone would endorse, because each person has an equal chance.

Research by E. Allan Lind and Tom Tyler (1988) confirmed that perceived impartiality of procedure — the sense that no thumb was on the scale — is a key driver of fairness judgements. Random selection maximises perceived impartiality because its neutrality is not just a claim but an observable property of the mechanism.

When Random Selection Outperforms Voting

Voting — deciding by collective preference — seems more democratic than random selection. But research identifies several contexts where random selection produces better outcomes:

Social choice problems

Arrow's impossibility theorem (Kenneth Arrow, 1951) demonstrates that no voting system can simultaneously satisfy all reasonable fairness criteria when there are three or more options. Voting systems always distort preferences in some way — through strategic voting, cycling preferences, or the spoiler effect. Random selection among tied options avoids the social choice paradox by not claiming to aggregate preferences at all.

Jury selection

The deliberate use of randomness in jury selection (discussed further in a related article) exists precisely because human selection of jurors — even with good intentions — introduces biases. Random draws from the population produce more representative juries than any human-curated selection process.

Citizen assemblies

Ireland's Citizens' Assembly (2016–2018), which produced recommendations on abortion and marriage equality that were subsequently enacted in referendums, used sortition — random selection from the population — to assemble a representative group of citizens. The randomly selected assembly was widely perceived as more legitimate than a committee of politicians, and its deliberations produced recommendations that reflected the population's actual views more accurately than political debate had.

The Psychological Comfort of Random Outcomes

There is also a psychological benefit to random selection in personal decisions: it removes guilt and regret. When a coin flip or random picker decides, the chooser is not responsible for the outcome. If it goes badly, there is no one to blame — including yourself.

Research on counterfactual thinking (imagining what would have happened if you had chosen differently) shows that outcomes from our own choices generate more regret than equivalent outcomes from random processes. "I should have chosen differently" is a thought that only applies when we made the choice. When randomness chose, this thought does not arise with the same force.

This may partly explain why people turn to random selection for genuinely difficult decisions — the psychological offloading of responsibility is a real benefit alongside the procedural fairness.

Let randomness decide →