Interpersonal Utility Comparison

The interpersonal comparison of utility is the problem of whether — and how — the welfare, happiness, or preference-satisfaction of one individual can be measured against that of another. It is the foundational question of welfare economics, social choice theory, and political philosophy. If my gain of ten utils cannot be compared to your loss of five, then the very concept of "social welfare" collapses into a mere aggregation of uncomparable individual metrics, and policies that trade one person's benefit against another's cost are flying blind.

Economists in the ordinalist tradition — notably Lionel Robbins in 1938 — declared interpersonal utility comparison impossible in principle, on the grounds that utility is a subjective mental state inaccessible to external measurement. This declaration was not a proof; it was a methodological boundary drawn to protect economics from ethical controversy. If utilities are incomparable, then economists can study efficiency (Pareto improvements) without touching distribution (who gains and who loses). The result was a field that became technically sophisticated and politically neutered, able to prove that a market is efficient but unable to say whether it is fair.

The declaration of impossibility has been challenged from multiple directions. The veil of ignorance framework, developed by John Harsanyi and popularized by John Rawls, proposes that behind a veil of ignorance about one's own position, rational agents would agree on a social welfare function — implying that interpersonal comparison is not impossible but merely requires a specific informational and institutional context. The capability approach of Amartya Sen shifts the metric from utility to functionings, arguing that capabilities are more observable and more ethically relevant than subjective mental states. More recently, neuroeconomics has attempted to measure welfare-relevant brain states directly, reviving the nineteenth-century aspiration of a scientific hedonometer.

The deeper issue is not methodological but political. The claim that interpersonal utility comparison is impossible functions as a liability shield: it prevents economists from making distributive judgments while allowing them to endorse efficiency-enhancing policies that systematically disadvantage specific groups. A social welfare function — a rule for aggregating individual utilities into a collective ranking — requires interpersonal comparison as a prerequisite. Without it, the entire apparatus of cost-benefit analysis, policy evaluation, and social choice theory rests on a convention that is simultaneously indispensable and officially disavowed.