Prospect Theory for Beginners: The Core Framework
Prospect Theory for Beginners: The Core Framework
Prospect theory is the mathematical and psychological framework that explains how real people make decisions under uncertainty, especially decisions involving money and risk. Unlike classical expected utility theory—which assumes rational actors who calculate expected values and choose accordingly—prospect theory describes how people actually decide, accounting for psychological biases, mental accounting, and the framing of choices. Since its introduction by Daniel Kahneman and Amos Tversky in 1979, prospect theory has become the foundation of behavioral finance and the theoretical anchor for understanding why investors do what they do with portfolios.
The core insight of prospect theory is deceptively simple: decision-making is not about objective outcomes but about how outcomes are perceived and framed relative to a reference point. A $10,000 loss feels catastrophic to someone who views their current position as a reference point. But to someone who frames it as moving from $1 million to $990,000, it is barely a dent. The same loss. The same absolute wealth. Completely different emotional and behavioral consequences. Prospect theory formalizes this observation and builds a practical model that predicts investor behavior.
Quick definition: Prospect theory is a behavioral economics framework describing how people evaluate uncertain outcomes by assessing potential gains and losses relative to a reference point, weighted by subjective probabilities, rather than by the objective probabilities assumed in rational choice theory.
Key takeaways
- Prospect theory replaces expected utility theory with a more realistic model that accounts for how people actually weight probabilities and feel losses versus gains.
- The theory has two stages: editing and evaluation, where decisions are first mentally simplified and then assessed relative to a reference point.
- Reference points are everything: the same $50,000 outcome feels like a gain if the reference point is $30,000 but a loss if the reference point is $70,000.
- People overweight small probabilities and underweight large probabilities, causing them to overestimate the chance of rare events and underestimate the likelihood of common ones.
- The S-shaped value function embeds loss aversion, where losses hurt roughly twice as much as equivalent gains feel good.
- Prospect theory explains real portfolio behavior far better than rational models and provides a foundation for disciplined investing strategies that account for bias.
The historical context: Why prospect theory was revolutionary
Before Kahneman and Tversky's work, expected utility theory dominated economics. In this framework, a rational decision-maker assigns a probability to each outcome, assigns a utility value to each outcome, multiplies probability by utility for each possible outcome, and chooses the option with the highest expected utility. This model is mathematically elegant, powerful, and almost completely wrong about how humans actually decide.
The classical model assumes that people:
- Assess probabilities accurately, without systematic bias
- Evaluate outcomes objectively, based on total wealth or final states
- Are indifferent between mathematically equivalent choices, regardless of how those choices are framed
- Treat all money as fungible, with equivalent dollars providing equivalent utility
None of these assumptions holds in practice. People systematically misbias their probability estimates. They evaluate outcomes relative to reference points, not absolute wealth. They are highly sensitive to framing. They mentally separate money into different accounts with different rules.
Prospect theory was revolutionary because it accepted these "irrational" observations as fundamental facts and built a model around them rather than dismissing them as anomalies. In doing so, Kahneman and Tversky provided a better description of human decision-making and, by extension, a better explanation of financial markets.
The two stages of prospect theory: Editing and evaluation
Prospect theory breaks the decision process into two stages.
Stage 1: Editing. Before formally evaluating an option, people simplify and mentally organize it. They might cancel common elements across options, rounding the probabilities and outcomes, or clustering similar choices together. They also edit outcomes into gains and losses relative to a reference point. A lottery might be edited as "a 50% chance to win $200 and a 50% chance to lose $100" rather than as "a 50% chance to end up with $300 and a 50% chance to end up with $100" in absolute terms.
The editing phase explains why framing is so powerful. A doctor's description of a medical treatment as "90% of patients survive" triggers different editing and evaluation than "10% of patients die," even though the objective information is identical. In investing, a portfolio showing a 15% annual gain is edited differently from one showing a 5% annual loss, even if the net outcome and historical volatility are identical.
Stage 2: Evaluation. After editing, people evaluate prospects using a value function. This value function has two distinctive features. First, it is defined on gains and losses relative to a reference point, not absolute wealth. Second, it is asymmetric: losses are weighted more heavily than gains of equal magnitude. The value function is not linear. It is concave in the domain of gains (more wealth brings less additional pleasure) and convex in the domain of losses (greater losses feel progressively worse, but at a decelerating rate).
The mathematical form of the value function is:
Value = w(p) * v(x) + w(q) * v(y)
where:
v(x) = x^alpha if x >= 0 (gain)
v(x) = -lambda * (-x)^beta if x < 0 (loss)
w(p) = decision weight on probability p
alpha, beta typically around 0.88
lambda (loss aversion coefficient) typically 2 to 2.5
This formalization captures the key behaviors: losses loom larger than gains, small probabilities are overweighted, and diminishing sensitivity applies to both gains and losses.
Reference points: The foundation of prospect theory
A reference point is the mental anchor against which outcomes are compared. It is often the current state (status quo), but it can also be an aspiration level, a benchmark, or a recent previous state. The same portfolio outcome feels entirely different depending on the reference point.
Example: Three investors hold the same stock that rises from $100 to $120. Investor A bought at $130 and views the reference point as $130. The stock is down $10, and Investor A feels she has experienced a loss. Investor B bought at $100 and views the reference point as $100. The stock is up $20, and Investor B feels he has captured a gain. Investor C bought at $80, views the reference point as $80, and is thrilled with the $40 gain. All three are holding the exact same security at the exact same price, but they interpret the outcome completely differently.
This explains a vast array of investor behaviors. A fund that is up 8% year-to-date but trailing its benchmark by 2% will experience a loss in prospect theory terms because the reference point is the benchmark return, not the absolute gain. The fund manager, feeling the loss, might make aggressive repositioning moves that would not be rational if the reference point were simply "did I lose money in absolute terms?" The answer is no—but in terms of the benchmark, yes, and that is what the prospect theory investor feels.
Reference points are also sticky. Studies show that when an investment has declined, investors often set their reference point at the peak value of the investment, not at current wealth. This is why investors speak of being "down 40% from the high" even though in absolute terms they might still be ahead of where they began. The reference point clings to the peak, making the decline feel like a persistent loss.
Probability weighting: Why small probabilities feel bigger
People do not weight probabilities accurately. Instead, they apply a probability weighting function that distorts objective probabilities. The weighting function has a characteristic shape: small probabilities are overweighted, moderate probabilities are roughly accurate or slightly underweighted, and large probabilities are underweighted.
Practical consequence: A 1% chance of a 50% portfolio loss feels much worse than 1% of a 50% loss, which would be 0.5% in expected value terms. People mentally amplify the 1% probability, making it feel like a meaningful threat. This is why investors buy portfolio insurance and demand risk premiums far higher than standard models predict. This is why a market that has a 2% historical probability of a 20% drawdown in a given month still causes most investors to hedge and demand compensation.
Conversely, a 99% probability of a 5% gain is often underweighted. Investors treat it as less certain than the math suggests, introducing unnecessary caution.
The probability weighting function is particularly pronounced at the extremes. People overweight the difference between 0% and 1% (never happens vs. might happen) and between 99% and 100% (probably happens vs. definitely happens). The difference between 50% and 51% is barely noticed.
The value function and diminishing sensitivity
The value function describes how people evaluate outcomes in terms of gains and losses. It has three key properties.
1. Reference dependence. Outcomes are evaluated as gains or losses relative to a reference point, not in absolute terms. A portfolio worth $500,000 is either a gain or loss depending on whether the reference point is $400,000 or $600,000.
2. Loss aversion. The slope of the value function is steeper in the loss domain than in the gain domain. A $10,000 loss causes more disutility than a $10,000 gain causes utility. The ratio is roughly 2:1 for most people, though it can be as high as 2.5:1 for experienced investors.
3. Diminishing sensitivity. Additional gains provide less additional utility as total gains increase. A gain from $0 to $1,000 feels much better than a gain from $100,000 to $101,000. Similarly, losses show diminishing sensitivity: the difference between a $100,000 loss and a $101,000 loss is barely noticeable, but the difference between a $0 loss and a $1,000 loss is acute. This explains why investors panic over small realized losses but are oddly calm about large underwater positions—the marginal pain diminishes.
Prospect theory in portfolio management
Prospect theory predicts several portfolio behaviors that are widely observed in practice.
Prediction 1: The Disposition Effect. Investors will sell winning positions and hold losing positions because selling winners locks in a gain (which feels good) while holding losers keeps alive the hope of avoiding the loss. Prospect theory's value function embeds loss aversion, making this behavior rational within the theory's framework, even though it is suboptimal in terms of long-term wealth.
Research on individual traders and active managers confirms this. A study of the Taiwan stock exchange found that retail investors were 48% more likely to sell a winning stock than a losing stock. Professional fund managers show the same pattern, though attenuated.
Prediction 2: Reference Point Attachment and the Endowment Effect. People demand more to sell an asset they own than they would pay to buy the same asset. This endowment effect arises because selling is framed as a loss relative to the reference point (current ownership) while buying is framed in terms of absolute value. Prospect theory predicts this: the loss aversion embedded in the value function makes people reluctant to realize losses.
Prediction 3: Hedging Demand and Portfolio Insurance. Investors will pay substantially more for downside protection than rational models predict because the probability weighting function overweights tail risks. A 1-in-100 drawdown scenario feels more likely than the 1% probability suggests. Consequently, investors demand and pay premium prices for puts and portfolio insurance.
Prediction 4: The Equity Risk Premium Puzzle. The historical equity risk premium—the extra return investors demand for holding stocks over bonds—is far higher than what standard models predict given the measured volatility of stocks. Prospect theory explains this: the loss aversion coefficient embedded in the value function makes investors reluctant to hold equities unless they anticipate substantially higher returns.
Real-world examples
Example 1: Benchmark-driven selling in down markets. A large-cap equity fund is down 12% year-to-date while its benchmark is down 10%. Even though the fund has underperformed its benchmark, the reference point (the fund's starting value) has moved down 12%. The loss is real in prospect theory terms. Fund managers often respond by aggressively repositioning to reduce the "loss" relative to the benchmark. This often means selling at depressed prices precisely when value is greatest—a classic prospect theory misdecision.
Example 2: IPO lockup expiration volatility. On the day that a company's IPO lockup expires and founders and early employees can finally sell their shares, stock prices often plummet. Why? Founders and employees evaluate their shares relative to the IPO price (the reference point), and after months of being unable to sell, they often view any price above the IPO price as a gain to be locked in. Prospect theory predicts this selling pressure, and it materializes reliably.
Example 3: Dividend cuts and stock crashes. When a company cuts its dividend, the stock often experiences a much larger decline than the dividend yield alone would justify. Shareholders had set their reference point at the previous dividend level, and the cut is framed as a loss. If the company reduced dividends by $0.10 per share on a $50 stock (a 0.2% reduction in annual income), prospect theory predicts a much larger reaction than a 0.2% decline because the loss aversion multiplier is embedded in the value function.
Common mistakes in applying prospect theory
Mistake 1: Confusing prospect theory with an excuse for irrational behavior. Prospect theory is descriptive, not prescriptive. It describes how people actually decide, not how they should decide. Using prospect theory to justify biased decisions ("I am just being prospect theory rational") is a misapplication. The point is to recognize the bias and design systems to limit its damage.
Mistake 2: Oversimplifying the value function. The value function is not simply "losses are 2x gains." The multiplier varies across individuals and contexts. Experienced traders show lower loss-aversion coefficients than novices. High-stakes decisions show different coefficients than low-stakes decisions. And the shape of the value function itself—how diminishing sensitivity applies—is not constant. A crude 2x multiplier is useful for intuition but insufficient for precise predictions.
Mistake 3: Ignoring the editing phase. How a decision is framed—how it is edited—often matters more than the underlying objective facts. A portfolio allocation decision framed as "how much to allocate to stocks" versus "how much to allocate to bonds" will generate different responses even though they are mathematically identical. Ignoring the framing is to ignore a primary determinant of behavior.
Mistake 4: Assuming reference points are fixed. Reference points shift over time and in response to experience. A portfolio manager who has experienced a 20% drawdown will often reset their reference point downward, making a 10% drawdown feel less severe than it would have before. New investors often have unrealistically high reference points and take losses particularly hard.
Mistake 5: Applying prospect theory only to individual investors. Institutional investors, including professionals with decades of experience and formal training, are subject to prospect theory. The bias is not eliminated by education or experience. Professional traders still exhibit loss aversion, overweight small probabilities, and are susceptible to framing. The scale and speed might change, but the underlying psychology persists.
FAQ
Q: Is prospect theory the same as behavioral finance?
A: No. Prospect theory is one foundational framework within behavioral finance. Behavioral finance is the broader field that studies how psychology influences financial decisions and markets. Prospect theory is the most widely applied model, but behavioral finance encompasses many other frameworks—mental accounting, herd behavior, overconfidence, and so on.
Q: Can I use prospect theory to beat the market?
A: Prospect theory predicts how most investors will behave, which can reveal market inefficiencies. For example, if prospect theory predicts that loss-averse investors will avoid a particular type of asset, that asset might be underpriced. Exploiting such inefficiencies requires discipline and a contrarian willingness to act against the prospect theory instincts of the crowd.
Q: How does prospect theory apply to professional investors who have years of experience?
A: Experience does reduce the magnitude of some prospect theory biases, but it does not eliminate them. Studies of professional traders show that they exhibit lower loss-aversion coefficients and are less susceptible to framing than novices. However, they still show measurable loss aversion, overweight small probabilities, and are susceptible to framing. Expertise makes the behavior more sophisticated, not rational.
Q: If prospect theory is so important, why doesn't every fund use it?
A: Some sophisticated funds do design portfolios and decision processes with prospect theory in mind. However, many fund managers operate under constraints—quarterly performance measurement, client benchmarks, redemption risk—that align with prospect theory biases rather than against them. A manager seeking to minimize quarterly losses is inadvertently optimizing for prospect theory behavior rather than long-term wealth.
Q: Can prospect theory help me time the market?
A: Prospect theory describes systematic behavioral patterns but does not provide timing signals. What it does do is warn against making timing decisions in response to loss aversion. Many market timing mistakes arise from prospect theory motives: selling after losses to avoid further pain, buying after gains when confidence is highest. A prospect theory awareness might help you avoid these timing traps.
Q: Is the 2x loss-aversion coefficient universal?
A: Research suggests the coefficient is in the range of 1.5 to 2.5 across most populations, but it varies. Wealthier individuals show slightly lower coefficients. Experienced traders show lower coefficients. Cultural differences exist. And the coefficient depends on the magnitude and domain of the decision. A 2x coefficient is useful as a baseline, but precise predictions should account for individual and contextual variation.
Related concepts
- What Is Loss Aversion?
- Why Losses Hurt Twice as Much
- The Prospect Theory Value Function
- Reference Points and How We Judge Outcomes
- What Is Behavioural Finance?
Summary
Prospect theory is a mathematical framework that describes how people actually make decisions under uncertainty, replacing the expected utility theory that dominates rational choice models. The theory has two stages—editing and evaluation—where choices are first simplified and then assessed relative to reference points using a value function that embeds loss aversion and probability weighting. The core prediction is that people evaluate outcomes as gains or losses relative to a reference point, that losses feel roughly twice as painful as equivalent gains feel pleasurable, and that people systematically misbias their probability estimates by overweighting small probabilities and underweighting large ones. These principles explain a vast array of portfolio behaviors, from the disposition effect to the willingness to pay premium prices for portfolio insurance, and provide a foundation for understanding why rational models fail to predict real investor behavior.