Prospect theory (Kahneman & Tversky, 1979) is a descriptive model of risky choice — not how people should decide, but how they do. It replaced expected utility's assumption that only final wealth states matter.
Reference dependence
Outcomes are coded as gains and losses relative to a reference point (status quo, goal, expectation) — not absolute wealth. Jack elated at $5M after $1M; Jill miserable at $5M after $9M. Bernoulli's logarithmic utility missed this; Anthony and Betty choose differently on the same gamble.
Three core features
-
Loss aversion: losses loom ~2× larger than equivalent gains. Explains risk aversion for gains (take $900 sure over 90% of $1000) and risk seeking for losses (gamble to avoid sure $900 loss).
-
Diminishing sensitivity: marginal value decreases for both gains and losses — concave gains, convex losses.
-
Probability weighting: overweight small probabilities (lottery tickets, insurance), underweight moderate/high probabilities.
Extensions in the book
Endowment effect (mugs study), fourfold pattern (certainty effect, possibility effect), framing reversals, mental accounting. Foundation for Thaler's nudge and behavioral policy.