In 1979, Daniel Kahneman and Amos Tversky published a paper in Econometrica that quietly dismantled one of the foundations of modern economics. [1] The paper was called Prospect Theory: An Analysis of Decision Under Risk, and its central claim was simple enough to be stated in a single sentence: people do not evaluate outcomes in absolute terms but relative to a reference point, and losses from that reference point hurt roughly twice as much as equivalent gains feel good.

Standard expected utility theory — the dominant framework since von Neumann and Morgenstern — holds that a rational agent maximises the expected value of a utility function defined over final wealth states. [2] Kahneman and Tversky showed, through a systematic series of experiments, that this is not how people actually behave. The deviations are not random noise; they are structured, replicable, and theoretically coherent. They are features, not bugs, of human cognition.

The Value Function

The core of prospect theory is the value function $v(x)$, which maps gains and losses relative to a reference point into subjective value. Its canonical form is:

$$v(x) = \begin{cases} x^{\,\alpha} & \text{if } x \geq 0 \\[4pt] -\,\lambda\,(-x)^{\beta} & \text{if } x < 0 \end{cases}$$

where $\alpha, \beta \in (0,1)$ capture the diminishing sensitivity to outcomes (the function is concave for gains and convex for losses), and $\lambda > 1$ is the loss aversion coefficient. From experimental data, Kahneman and Tversky estimated $\alpha \approx \beta \approx 0.88$ and $\lambda \approx 2.25$: a loss feels about 2.25 times as bad as an equivalent gain feels good. [3]

v(x) x 0 GAINS LOSSES reference point +gain −loss (≈ 2.25×)
The prospect theory value function. Both curves pass through the reference point (origin). The gains side is concave (diminishing marginal value); the losses side is convex but steeper, reflecting loss aversion. Equal objective distances from the origin produce unequal subjective impacts.

The Coin Toss

The asymmetry becomes visceral with a simple gamble. Suppose someone offers you a coin toss: heads, you win £X; tails, you lose £100. Most people require roughly £200–250 to find this bet attractive — more than double the potential loss. Expected value theory says the fair price of the bet is zero at $X = 100$; prospect theory explains why it feels deeply unappealing at £120, £150, even £180.

This is not irrationality in any simple sense. Thaler showed that the same asymmetry structures consumer markets through what he called the endowment effect: once you own something, you value it more than you did before you owned it. [4] In a famous experiment, participants randomly given a coffee mug demanded roughly twice as much to sell it as others were willing to pay to acquire one. Ownership shifts the reference point, and anything below that point registers as a loss.

Probability Weighting

Prospect theory contains a second insight, less celebrated but equally important: people do not treat probabilities linearly. They systematically overweight small probabilities and underweight large ones. The probability weighting function $w(p)$ proposed by Tversky and Kahneman takes the form: [5]

$$w(p) = \frac{p^{\,\gamma}}{\bigl(p^{\,\gamma} + (1-p)^{\gamma}\bigr)^{1/\gamma}}$$

with $\gamma \approx 0.65$ for gains. This single parameter generates the characteristic inverse-S shape: a 1% chance is treated as though it were perhaps 3–4%, while a 95% chance feels closer to 85–90%. The practical consequences are everywhere. Lotteries are purchased not despite negative expected value but because the small probability of a large win is psychologically magnified. Insurance is bought against vanishingly unlikely catastrophes for the same reason, running in the opposite direction.

Status Quo as Reference Point

One of the most consequential extensions of prospect theory concerns how the reference point is set. In most cases, the reference point is the status quo — the current state of affairs. This means that any departure from the current state is framed as a loss or gain, and since losses are weighted more heavily than gains, there is a systematic bias in favour of inaction. Samuelson and Zeckhauser documented this in a series of elegant experiments, calling it status quo bias. [6]

The implications for policy are direct. Default options are extraordinarily powerful: whether employees are automatically enrolled in pension schemes or must actively opt in, whether organ donation is presumed or must be explicitly chosen — these framings, which should be neutral in a world of rational agents, are decisive in the real world precisely because the default defines the reference point.

A Note on Legacy

Kahneman received the Nobel Memorial Prize in Economic Sciences in 2002 — Tversky had died in 1996, and the Prize is not awarded posthumously. Kahneman's book Thinking, Fast and Slow brought prospect theory to a general audience, along with the broader programme of behavioural economics they had helped to found. [2] The book is a rare thing: a piece of genuine science that can be read on a train.

What stays with me, though, is not the applications but the original observation — the naked, uncomfortable fact that the same quantity of money can feel like two completely different things depending on whether you are gaining it or losing it. The universe does not know the difference. We do.