Have you ever wondered how people make decisions when the outcome is uncertain? Whether you’re choosing a career path, deciding on an investment, or even picking between two job offers, you’re constantly making choices without knowing exactly what will happen. This is where Expected Utility Theory comes in-a fundamental framework in economics that helps explain how rational decision-makers should behave when facing risk and uncertainty.
Table of Contents
What is expected utility theory?
At its core, Expected Utility Theory is an approach to rational decision-making under uncertainty. Rather than simply looking at the face value of different outcomes, this theory suggests that people should evaluate uncertain prospects by considering both the value of possible outcomes and their probabilities. The fundamental principle is straightforward: choose the option with the highest expected utility.
Think of it like this: imagine you’re offered two options. The first guarantees you ₹50,000. The second gives you a 50% chance of winning ₹120,000 and a 50% chance of getting nothing. Which should you choose? Expected Utility Theory provides a systematic way to evaluate such choices by calculating the weighted average of utilities, where each outcome’s utility is multiplied by its probability of occurring.
The theory originated from Daniel Bernoulli’s 18th-century work, when he sought to solve the famous St. Petersburg Paradox. Bernoulli recognized something crucial: the same amount of money doesn’t have the same value to everyone. An extra ₹10,000 means much more to someone earning ₹50,000 annually than to someone earning ₹5 million. This insight-that utility differs from monetary value-became foundational to the theory.
The mathematical foundation of expected utility
Expected utility theory rests on a mathematical framework. When a decision-maker faces different options (called “lotteries” or “prospects”), each option has multiple possible outcomes, each occurring with a certain probability. The expected utility of an option equals the sum of each outcome’s utility multiplied by its probability.
For instance, consider a simple lottery: 70% chance of winning ₹1,000 and 30% chance of winning nothing. If we assign a utility of 10 to ₹1,000 and 0 to nothing, the expected utility would be (0.7 × 10) + (0.3 × 0) = 7. According to the theory, a rational person would prefer this lottery over any option with an expected utility less than 7.
What makes this approach powerful is that it incorporates individual risk preferences into decision-making. Two people facing the same lottery might make different choices based on their attitudes toward risk-and both could be acting rationally.
The crucial independence axiom
For Expected Utility Theory to work, certain assumptions about rational preferences must hold. Among these, the independence axiom stands as perhaps the most important-and controversial.
Understanding the independence axiom
The independence axiom states something that initially seems intuitive: if you prefer lottery A to lottery B, then you should also prefer a mixture of A with some third lottery C over a mixture of B with that same lottery C, assuming both mixtures use identical probabilities. In simpler terms, your preference between two options shouldn’t change if both options are modified in exactly the same way.
Imagine you prefer ice cream to gravy. Now suppose I offer you two new choices: a coin flip that gives you either ice cream or celery versus a coin flip that gives you either gravy or celery. According to the independence axiom, you should prefer the first mixed option, because the only difference between the two is whether you get ice cream or gravy when the coin lands heads-and you already prefer ice cream.
This axiom is critical because it ensures that preferences can be represented linearly, allowing us to use the mathematical structure of expected utility. Without independence, the elegant mathematical framework of the theory falls apart.
Why the independence axiom matters
The independence axiom creates the linear structure that makes expected utility calculations work. It ensures that indifference curves in probability space are parallel straight lines, which in turn guarantees that we can find utility numbers for prizes such that one lottery has higher expected utility than another if and only if it is preferred.
However, this axiom has faced significant criticism. The famous Allais Paradox demonstrates that many people make choices that violate the independence axiom, even when those choices seem intuitively reasonable. In Allais’s experiment, people often prefer a certain ₹10 million over an 89% chance of ₹10 million plus a 10% chance of ₹50 million. Yet these same people prefer a 10% chance of ₹50 million over an 11% chance of ₹10 million-choices that mathematically contradict each other under expected utility theory.
Rationality and utility functions
Expected Utility Theory assumes that if decision-makers have rational and continuous preferences, these preferences can be represented by a utility function. But what does “rational” mean in this context?
Rational preferences must satisfy several conditions. They must be complete, meaning you can compare any two options and determine which you prefer or whether you’re indifferent. They must be transitive: if you prefer A to B and B to C, you must prefer A to C. And crucially, they must satisfy the independence axiom we discussed earlier.
When preferences meet these conditions, mathematicians can construct a utility function that represents those preferences. This function assigns numerical values to outcomes in a way that preserves the decision-maker’s preference ordering. Higher utility numbers correspond to more preferred outcomes.
Applications in everyday decision-making
Expected utility theory isn’t just an abstract concept-it has practical applications across various domains. Insurance companies use it to calculate premiums, considering both the probability of claims and the utility losses to policyholders. Investment advisors apply the framework to help clients choose portfolios that match their risk tolerance. Even public policy decisions about health interventions and safety regulations often rely on expected utility calculations.
Consider someone deciding whether to start a new business. They might estimate a 30% chance of significant success (high utility), a 50% chance of modest success (medium utility), and a 20% chance of failure (low or negative utility). By calculating the expected utility and comparing it to the utility of their current stable job, they can make a more informed decision that accounts for both their risk preferences and the probabilities involved.
Limitations and critiques
Despite its theoretical elegance, Expected Utility Theory faces important limitations. Research in behavioral economics has shown that real people often violate the theory’s predictions. The Allais Paradox is just one example; the Ellsberg Paradox and other experimental findings reveal systematic deviations from expected utility maximization.
These observations led to alternative theories like Prospect Theory, developed by Daniel Kahneman and Amos Tversky, which better describes how people actually make decisions under risk. These alternatives maintain some insights from expected utility theory while relaxing the strict independence axiom to accommodate observed behavior.
Furthermore, the theory struggles with scenarios involving extremely low probability events with high stakes, such as catastrophic risks. It also faces challenges in situations where probabilities are unknown or difficult to estimate-a common occurrence in real-world decision-making.
What do you think? Can a mathematical theory ever fully capture the complexity of human decision-making under uncertainty? And when your choices violate the independence axiom, does that necessarily mean you’re being irrational, or might there be good reasons for such deviations?
References
- https://plato.stanford.edu/entries/rationality-normative-utility/
- https://www.economicshelp.org/blog/glossary/expected-utility-theory/
- https://www.economicsonline.co.uk/definitions/expected-utility-theory.html/
- https://en.wikipedia.org/wiki/Expected_utility_hypothesis
- https://en.wikipedia.org/wiki/Allais_paradox
- https://corporatefinanceinstitute.com/resources/economics/expected-utility/
Leave a Reply