Bernoulli’s Errors

Kahneman opens the section on choices by examining the theory he and Tversky set out to replace: the expected-utility model rooted in Daniel Bernoulli's eighteenth-century insight. Bernoulli solved a famous puzzle by proposing that people value not money itself but its psychological utility, which diminishes as wealth grows — the first hundred dollars means far more to a pauper than to a millionaire. This elegant idea explained risk aversion and dominated economics for nearly three hundred years.

The diminishing-utility idea is genuinely powerful: because each additional dollar adds less happiness, a sure amount is often preferred to a gamble with the same average value, which is why people buy insurance and decline fair bets. Bernoulli's model gave economics a rigorous account of rational choice under uncertainty and underwrote the 'rational agent' at the heart of the discipline. Kahneman gives it full credit before delivering the critique that motivates everything to follow.

Bernoulli's fatal flaw, Kahneman argues, is that his theory assigns utility to states of wealth while ignoring the reference point from which those states are reached. His model has no place for history — for whether you arrived at your current wealth by gaining or by losing. Yet the same final state can be a triumph or a tragedy depending on where you started, and a theory blind to that difference must misdescribe how people actually feel and choose.

His decisive illustration contrasts two people who both end the day with two million dollars. Anthony started with one million and has doubled it; Betty started with four million and has lost half. Bernoulli's theory says they are equally satisfied because their final wealth is identical — but obviously Anthony is elated and Betty is devastated. Their happiness depends on the change from their reference point, not the absolute state, and Bernoulli's model cannot capture this because it has no concept of a reference at all.

The applied takeaway is that evaluations of outcomes are inescapably relative. People do not experience wealth, health, or status in absolute terms but as gains or losses against a baseline of expectation or recent experience, so the same objective situation can feel wonderful or terrible depending on the comparison. Any account of choice or well-being that ignores the reference point — as much of conventional economics did — will systematically mispredict how people actually respond.

Kahneman's deeper observation is the phenomenon he calls 'theory-induced blindness': once a theory is accepted and woven into a field, its flaws become nearly invisible, and counterexamples that should have refuted it are explained away or simply not seen. Bernoulli's omission of the reference point was glaring in hindsight, yet generations of brilliant economists worked within his framework without noticing it. The chapter is thus both a critique of a specific model and a warning about how a successful theory can blind even experts to its own obvious limitations.