What Is Prospect Theory? How It Shapes Investor Decisions

Search Fintrev

Search our guides, reviews and glossary.

Prospect theory is a model of how people choose between risky options. It holds that people judge outcomes as gains or losses relative to a reference point, usually their current position, rather than by the final wealth an outcome produces. Losses weigh more heavily than gains of the same size, and probabilities are distorted rather than taken at face value. Psychologists Daniel Kahneman and Amos Tversky introduced it in 1979 as a descriptive account of real decisions, in contrast to expected utility theory, which describes how a fully rational decision-maker should choose.

Origins: Kahneman and Tversky

Kahneman and Tversky’s original paper, “Prospect Theory: An Analysis of Decision Under Risk,” appeared in Econometrica in 1979. It documented choices that expected utility theory could not explain and proposed an alternative built from observed behavior. Their 1992 paper in the Journal of Risk and Uncertainty introduced cumulative prospect theory, which fixed technical problems in the first version and supplied the parameter estimates still quoted today. Kahneman received the 2002 Nobel Memorial Prize in Economic Sciences, partly for this work.

How prospect theory works

The theory rests on four features. Each one changes a prediction that standard economics would make.

The reference point

People code outcomes as gains or losses relative to a benchmark. That benchmark is often the status quo, but it can be a purchase price, an expectation or a goal. A $5,000 year-end bonus feels like a gain to someone who expected nothing and like a loss to someone who expected $10,000, even though both end up with the same money.

Loss aversion

Below the reference point, the value curve is steeper than above it. Losing $100 produces a larger negative feeling than the positive feeling of gaining $100. People are said to be operating in the domain of losses when every option on the table is a loss relative to their reference point, and that is where behavior shifts most sharply.

Diminishing sensitivity

The value curve flattens as you move away from the reference point in either direction. The difference between gaining $100 and $200 feels larger than the difference between gaining $1,100 and $1,200. For gains, this produces risk aversion. For losses, the same flattening makes an extra loss feel less painful than the first one, which produces risk seeking.

Probability weighting

People do not treat a 1% chance as exactly 1%. Small probabilities receive more weight than they deserve. Moderate and high probabilities receive less weight than they deserve, so a 95% chance feels noticeably less secure than certainty.

The prospect theory formula

Cumulative prospect theory gives the value function a standard form:

• For gains (x ≥ 0): v(x) = x^α
• For losses (x < 0): v(x) = −λ(−x)^β

Here α and β control diminishing sensitivity, and λ is the loss aversion coefficient. Tversky and Kahneman’s 1992 median estimates were α = β = 0.88 and λ = 2.25. They also proposed a probability weighting function, w(p) = p^γ / (p^γ + (1 − p)^γ)^(1/γ), with estimates of γ = 0.61 for gains and 0.69 for losses. These are population estimates from experiments, not constants that apply to every individual.

Hypothetical worked example. Suppose you are offered a coin flip: win $150 or lose $100. The expected value is +$25 (0.5 × 150 − 0.5 × 100). To keep the arithmetic simple, assume both outcomes receive a decision weight of 0.5, which ignores probability distortion.

• Value of the gain: 150^0.88 ≈ 82.2
• Value of the loss: −2.25 × 100^0.88 ≈ −2.25 × 57.5 ≈ −129.4
• Prospect value: 0.5 × 82.2 + 0.5 × (−129.4) ≈ −23.6

The bet has positive expected value but negative prospect value, so the model predicts that most people will refuse it. Under the same assumptions, the gain would need to be about $251 before the flip breaks even in prospect value terms. That is roughly two and a half times the amount at risk.

The editing and evaluation phases

The 1979 paper describes decisions as happening in two stages.

In the editing phase, the decision-maker simplifies the problem. They choose a reference point and code outcomes as gains or losses. They combine identical outcomes, cancel components the options share, and may discard outcomes that seem extremely unlikely. These shortcuts make complex choices manageable, but they also introduce bias.

In the evaluation phase, each edited option is scored by multiplying the value of each outcome by its decision weight, and the option with the highest total is chosen. Probability weighting applies here. Rare outcomes that survive editing tend to be overweighted, while those dropped during editing are ignored entirely.

Certainty, reflection and isolation effects

Certainty effect. People overweight outcomes that are certain compared with outcomes that are merely probable. In one of the original problems, most respondents preferred a sure 3,000 to an 80% chance of 4,000, even though the gamble’s expected value is 3,200.

Reflection effect. When the same choice is framed as losses, the preference flips. Most respondents preferred an 80% chance of losing 4,000 to a sure loss of 3,000, taking the gamble even though its expected loss is larger.

Isolation effect. To simplify choices, people disregard components that alternatives share and focus on the differences. Because the parts that get dropped depend on how the options are presented, two descriptions of the same choice can produce different decisions.

Together, these produce what researchers call the fourfold pattern: risk aversion for likely gains and unlikely losses (as when buying insurance), and risk seeking for likely losses and unlikely gains (as when buying lottery tickets).

How prospect theory applies to investors

Several well-documented investor behaviors fit the model:

• Disposition effect. Investors tend to sell winning positions quickly to lock in a gain and hold losing positions in the hope of breaking even. The purchase price acts as the reference point. A paper loss sits in the risk-seeking domain, so holding on and gambling on a recovery feels better than realizing the loss.
• Myopic loss aversion. Economists Shlomo Benartzi and Richard Thaler argued in 1995 that investors who check their portfolios often experience more frequent short-term losses, find stocks more painful to hold, and demand a higher return for owning them. This was offered as one explanation for the equity premium puzzle.

Hypothetical framing example. Two advisors describe the same fund. One says it has averaged a 10% annual return over three years. The other says it beat its benchmark over a decade but its returns have slipped recently. Prospect theory predicts that more listeners will invest after the first description, because it presents only gains.

To counter these biases, investors can set selling rules before buying, judge positions on forward-looking prospects rather than entry price, review portfolios on a fixed schedule instead of constantly, and restate any choice in both gain and loss terms before committing.

Limitations of prospect theory

• The reference point is not pinned down. The theory does not say exactly which benchmark a person will use, and different choices of reference point can produce different predictions.
• Parameters vary. The 1992 estimates are medians from experiments. Individual loss aversion differs widely, and some later studies report smaller asymmetries than 2.25, particularly for small or routine stakes.
• Laboratory evidence. Many of the classic results come from hypothetical questions with modest amounts. Behavior with large, real money and expert decision-makers can differ.
• Descriptive, not prescriptive. The theory explains how people choose. It does not say they should choose that way, and it does not predict prices or provide trading signals.

Prospect theory vs. related concepts

Loss aversion vs. prospect theory. These terms are often used interchangeably, but they are not the same. Loss aversion is one component of prospect theory, represented by λ. The full theory also includes reference dependence, diminishing sensitivity and probability weighting, which together explain the reflection effect and the fourfold pattern. Loss aversion alone cannot account for those.

Expected utility theory. Expected utility evaluates final wealth levels, weights outcomes by their actual probabilities, and assumes a consistent attitude toward risk. Prospect theory evaluates changes from a reference point, uses distorted decision weights, and allows risk attitudes to flip depending on whether the outcome is a gain or a loss.

Framing effect. Framing is a broader finding that presentation changes choices. Prospect theory provides one mechanism for it: a frame moves the reference point, and the reference point decides whether an outcome is coded as a gain or a loss.

The Bottom Line

Prospect theory matters to investors who want to understand why they hold losing positions, check portfolios too often or react to how returns are framed. The model describes behavior and offers no trading signals, so its practical use is self-correction. Setting selling rules before buying, judging positions on forward-looking prospects and restating each choice in both gain and loss terms can reduce its pull.