Prospect theory is a behavioral economics framework that describes how people actually make decisions when facing risk and uncertainty. Developed by psychologists Daniel Kahneman and Amos Tversky in 1979, it replaced the long-standing assumption that people weigh outcomes rationally by showing that we evaluate potential gains and losses differently, often in predictable but irrational ways. The work was significant enough that Kahneman received the 2002 Nobel Prize in Economics for “having integrated insights from psychological research into economic science, especially concerning human judgment and decision-making under uncertainty.”
Why Traditional Economics Got It Wrong
Before prospect theory, economists relied on something called expected utility theory to predict how people make choices. The idea was straightforward: people calculate the value of each possible outcome, multiply it by the probability of it happening, and pick whichever option gives them the highest total. In this model, what matters is your final state of wealth. If you have $100,000 and could win or lose $1,000, the theory says you’d compare how useful $101,000 is to you versus $99,000, weigh each by its likelihood, and choose accordingly.
The problem is that people don’t think this way. Kahneman and Tversky ran experiments showing that decisions shift dramatically depending on whether a choice is presented as a gain or a loss, even when the math is identical. People routinely violate the predictions of expected utility theory in consistent, patterned ways. Prospect theory was built to describe what people actually do rather than what a perfectly rational person should do.
The Three Core Ideas
Reference Dependence
In prospect theory, people don’t evaluate outcomes based on their total wealth. Instead, they judge outcomes as gains or losses relative to a reference point, which is typically their current situation. This matters because the same outcome can feel like a win or a loss depending on where you started. A salary of $80,000 feels great if you were making $60,000. It feels terrible if you were making $100,000. The objective number is the same, but the experience is completely different.
Reference points aren’t always the status quo. They can also be based on expectations, aspirations, or past experience. If you expect a $5,000 bonus and receive $3,000, you process that as a $2,000 loss even though you’re objectively richer. This flexibility in reference points is part of what makes the theory so powerful at explaining real behavior.
Loss Aversion
Losses hurt roughly twice as much as equivalent gains feel good. In their original estimates, Kahneman and Tversky found that the pain of losing $100 was about 2.25 times as intense as the pleasure of gaining $100. A more recent meta-analysis put the ratio closer to 1.31, but the core finding holds: we are wired to feel losses more sharply than gains of the same size.
This asymmetry drives everyday decisions. It explains why people hold onto losing stocks too long (selling would mean locking in a loss), why they turn down fair coin-flip bets (the potential loss looms larger), and why free trials are so effective in marketing (giving up the product feels like a loss once you have it).
Probability Distortion
People don’t process probabilities accurately. We tend to overweight small probabilities and underweight large ones. A 1% chance of winning feels bigger than 1%, which is why people buy lottery tickets. A 99% chance of success feels less certain than it should, which is why people pay for insurance against unlikely disasters.
This distortion also helps explain why people are drawn to long-shot gambles while simultaneously overpaying for certainty. The jump from 0% to 1% (from impossible to possible) feels enormous, while the jump from 99% to 100% (from almost certain to guaranteed) also carries outsized psychological weight. Probabilities in the middle get compressed and underweighted.
The S-Shaped Value Function
Prospect theory’s most recognizable feature is its value function, an S-shaped curve that captures how we experience gains and losses. On the gains side, the curve bends downward (concave), meaning each additional dollar of gain feels a little less satisfying than the last. Going from $0 to $100 feels much better than going from $900 to $1,000. This is diminishing sensitivity, and it mirrors what most people intuitively understand about money.
On the losses side, the curve bends the opposite way (convex), meaning each additional dollar of loss also hurts a little less than the previous one. Going from losing $0 to losing $100 is agonizing. Going from losing $900 to losing $1,000 barely registers as different. And critically, the loss side of the curve is steeper than the gain side, which is the visual representation of loss aversion.
This shape has a direct behavioral consequence. Because the gains curve bends down, people tend to be risk-averse when things are going well. They’d rather take a sure $500 than a 50/50 shot at $1,000. But because the loss curve bends the opposite way, people become risk-seeking when facing losses. They’d rather gamble on a 50/50 chance of losing $1,000 or losing nothing than accept a guaranteed loss of $500.
The Reflection Effect
This flip in risk preferences between gains and losses has a name: the reflection effect. It’s one of the most reliable findings in behavioral economics. When people choose between a certain gain and a risky gamble with the same expected value, most prefer the sure thing. But mirror the scenario into losses, and the same people suddenly prefer to gamble.
Consider a concrete example. Most people, given a choice between receiving $100 for certain or flipping a coin for $200 or nothing, take the guaranteed $100. Now flip it: choose between losing $100 for certain or flipping a coin to either lose $200 or lose nothing. Most people take the gamble. The math is symmetric, but the psychology is not. When losing feels inevitable, people will take risks they’d never accept in pursuit of gains.
This pattern shows up everywhere, from how investors double down on bad positions to how patients choose between treatment options. Framing the same medical outcome as a survival rate versus a mortality rate can shift which treatment people prefer.
How Framing Changes Decisions
Because prospect theory revolves around gains and losses relative to a reference point, the way a choice is framed can completely change the decision. This has practical consequences for public health, policy, and communication.
In one series of experiments, researchers presented people with a hypothetical disease outbreak and asked them to choose between two response programs. One offered a certain outcome, the other an uncertain one. When the options were framed in terms of lives saved (gains), participants strongly preferred the certain option. When the same options were framed in terms of lives lost, preferences shifted toward the uncertain gamble. The underlying numbers were identical. Only the framing changed.
Health messaging researchers have explored this extensively. Gain-framed messages (“Getting screened early improves your chances of successful treatment”) tend to be more effective for prevention behaviors, where people see themselves as protecting a good outcome. Loss-framed messages (“Failing to get screened means you could miss a treatable condition”) can be more effective for detection behaviors, where people are confronting a potential threat.
Updates to the Original Theory
The 1979 version of prospect theory had a known limitation: it could produce irrational predictions when applied to gambles with three or more possible outcomes. In some cases, the math allowed for violations of a basic principle called stochastic dominance, where a clearly worse option could be rated higher than a clearly better one.
Tversky and Kahneman addressed this in 1992 with cumulative prospect theory, which changed how the probability weighting function combines with outcomes. Instead of weighting each probability individually, the updated version weights cumulative probabilities, which eliminates the dominance violations while preserving the core insights about loss aversion, reference dependence, and probability distortion. Cumulative prospect theory is now the standard version used in most research.
Where Prospect Theory Shows Up
The theory’s fingerprints are on a wide range of real-world phenomena. In investing, loss aversion explains the “disposition effect,” where people sell winning investments too quickly (to lock in a gain) and hold losers too long (to avoid realizing a loss). In negotiations, the reference point determines whether each side sees a proposed deal as a gain or a concession, which changes how much risk they’re willing to tolerate.
Insurance purchasing is another natural fit. People routinely overpay for insurance against low-probability events, exactly what the probability weighting function predicts. The small chance of a catastrophic loss gets inflated in the mind, making the premium feel worthwhile even when the expected value math doesn’t support it.
Pricing strategies rely on the theory too. A $20 discount on a $50 item feels more significant than the same $20 discount on a $500 item, because the value function is steepest near the reference point and flattens as amounts grow. Retailers who understand diminishing sensitivity can structure discounts, bundles, and surcharges in ways that maximize their psychological impact.