Prospect Theory

Definition

Prospect theory offers an alternative to expected utility theory for describing how people actually choose among risky outcomes. It proposes three key deviations from rational choice: (1) people evaluate outcomes as gains or losses relative to a reference point (status quo), not as final states of wealth; (2) the value function is steeper for losses than gains, creating loss aversion (a loss of X dollars feels worse than a gain of X dollars feels good); and (3) decision weights distort probabilities, overweighting very low probabilities and underweighting moderate to high ones.

In the Book

Tversky and Kahneman develop this theory in response to empirical patterns in choice that expected utility theory cannot explain. The classic example is the "Asian Disease" framing experiment: when a choice is framed in terms of lives saved (gains), people are risk-averse and prefer a certain outcome. When the identical choice is framed in terms of lives lost (losses), people become risk-seeking. A gain of 200 lives out of 600 seems unattractive when sure, but the possibility of losing 400 lives feels worse—same outcome, opposite preference. The value function is concave for gains (diminishing sensitivity to increasing wealth) and convex for losses (further losses feel successively less painful), creating the pattern of risk-aversion for gains and risk-seeking for losses.

Tversky notes that decision weights—the impact of probabilities on choices—follow a nonlinear transformation that overweights small probabilities (people treat a 1% chance too much like certainty) and underweights large ones. The certainty effect further distinguishes risk from uncertainty: moving from impossible to possible has greater impact than moving from possible to probable.

Why It Matters

Prospect theory unifies decades of observed choice patterns under one framework: why people hold losing stocks too long, why defaults matter enormously, why identical policies produce opposite preferences when framed as gains versus losses, and why insurance exists despite actuarial unfairness. It reveals that preferences are not stable properties but are constructed anew based on how a decision is presented. This has profound implications for economics, policy (defaults in retirement savings, organ donation), law (how liability is framed), and negotiation (are we negotiating a gain or avoiding a loss?).