Prospect Theory: An Analysis of Decision under Risk
A loss hurts about twice as much as an equal gain — so we choose by change, not by final wealth.
Offered a sure $3,000 or an 80% shot at $4,000, most people grab the sure thing — then turn around and gamble when the very same numbers are losses. Two psychologists turned that quirk into a theory.
The big idea
Classic economics assumed people weigh choices by their final wealth and pick whatever offers the best average payoff. Daniel Kahneman and Amos Tversky showed that's not how minds work. We don't judge an outcome against our total bank balance; we judge it as a gain or a loss from wherever we happen to be standing — a reference point. And we don't treat the two evenly: a loss hurts roughly twice as much as the same-sized gain feels good.
That single asymmetry — loss aversion — together with the fact that we feel the first dollar of a change more sharply than the thousandth, bends our choices in predictable ways. We play it safe to lock in a gain, but we'll take risks to escape a sure loss. And we overreact to tiny probabilities, which is why the same person can buy a lottery ticket and an insurance policy in the same week.
How it came about
Kahneman and Tversky were Israeli psychologists who began working together at the Hebrew University of Jerusalem in 1969 and became one of the most famous partnerships in science — so equal that they sometimes tossed a coin to decide whose name went first. Through the 1970s they ran deceptively simple questionnaires: would you prefer this gamble, or that one? The answers kept breaking the reigning theory of rational choice.
In 1979 they laid out the pattern in Econometrica — deliberately publishing in an economics journal to reach the field they meant to change. Tversky died in 1996; Kahneman received the 2002 Nobel Memorial Prize in Economics, and said for the rest of his life that the prize was theirs together.
Why it mattered
It cracked open the assumption of a perfectly rational "economic man" and founded behavioral economics. Once you accept that people misjudge risk in systematic, predictable ways, you can explain behavior the old models couldn't — why markets panic, why we cling to losing investments, why the wording of a choice changes what we pick — and you can design better defaults, warnings and policies around how people actually decide.
A way to picture it
Think of a thermometer for money, zeroed not at absolute cold but at room temperature — wherever you are right now. Everything reads as "warmer" (a gain) or "cooler" (a loss) than now, never as an absolute number. And the cold half of the dial is stretched: drop ten degrees and it feels far worse than rising ten degrees feels good. Move your reference point — get used to a bigger salary, say — and the whole scale slides with you, so yesterday's luxury becomes today's ordinary.
Where it sits
For two centuries economics ran on expected utility — the idea, from Daniel Bernoulli and later John von Neumann, that people maximize the average usefulness of their final wealth; it underlies the Library's Nash (1950) and Markowitz (1952). Prospect theory is the empirical rebuttal: a careful map of how real choices swerve from that ideal, in the spirit of Herbert Simon's "bounded rationality". From here the road runs straight to Richard Thaler's "nudges" and to the behavioral public policy of today.
Decision making under risk can be viewed as a choice between prospects or gambles. A prospect (x₁, p₁; …; xₙ, pₙ) is a contract that yields outcome xᵢ with probability pᵢ.
[P]eople overweight outcomes that are considered certain, relative to outcomes which are merely probable — a phenomenon which we label the certainty effect.
[T]he value function is (i) defined on deviations from the reference point; (ii) generally concave for gains and commonly convex for losses; (iii) steeper for losses than for gains.