Kahneman & Tversky (1979) - Prospect Theory: Decision Under Risk

Overview

Kahneman and Tversky's (1979) prospect theory is the most influential behavioral model of decision under risk. It earned Kahneman the 2002 Nobel Prize. The model challenges expected utility's claim that people evaluate gambles by their final wealth — instead, people evaluate outcomes as gains and losses relative to a reference point, and they treat those two directions very differently.

The Anomalies Expected Utility Cannot Explain

The paper opens with a series of choice experiments exposing systematic violations of expected utility. The certainty effect: people overweight sure outcomes relative to merely probable ones. The reflection effect: preferences reverse between the gain domain and the loss domain. The isolation effect: people decompose gambles in ways that change their decisions. Together these show that the framing of a gamble, not just its probabilities and payoffs, determines the choice.

The Value Function

Prospect theory replaces the utility-of-wealth function with a value function defined on deviations from a reference point: concave in the gain domain, convex in the loss domain, and steeper for losses than for gains. The asymmetry is loss aversion — a loss of a given size hurts about twice as much as an equivalent gain pleases. The shape explains risk aversion for gains, risk seeking for losses, and the extreme reluctance to realize losses in portfolios.

The Probability Weighting

Decisions weight probabilities non-linearly: small probabilities are overweighted (lotteries and insurance), while large probabilities are underweighted. This distortion explains demand for long-shot bets and for insurance against low-probability disasters, both of which expected utility struggles to justify.

Why It Matters

Prospect theory is the behavioral foundation underneath every market anomaly driven by sentiment and framing: disposition effects, the equity premium puzzle's preference component, and momentum. For practitioners it is the formal license to take investor behavior seriously as a force that moves prices, not a footnote to rational models.

Key Takeaways

• Reference dependence beats final-wealth maximization as a description of actual choice.
• Loss aversion (~2x) drives the disposition effect — selling winners, riding losers.
• Probability weighting explains both lottery demand and insurance demand — two puzzles, one mechanism.