top of page

# The Von Neumann-Morgenstern Utility Theorem: Insights and Implications for Finance and Investing

Updated: Feb 9

In the world of economics, finance, and decision theory, utility plays a pivotal role in shaping our understanding of human behavior. The concept of utility, which is a measure of the satisfaction or benefit derived by consuming a good or service, lies at the heart of the Von Neumann-Morgenstern (VNM) utility theorem. This theory has far-reaching implications for finance and investing and provides a mathematical basis for decision-making under uncertainty.

Understanding the Von Neumann-Morgenstern Utility Theorem

The Von Neumann-Morgenstern utility theorem is named after its creators, John Von Neumann and Oskar Morgenstern, who first introduced it in their seminal book, "Theory of Games and Economic Behavior," published in 1944. The theorem establishes a way of representing and measuring a decision-maker's preferences that satisfy certain rationality axioms.

The VNM utility theorem is based on four key axioms:

• Completeness: For any two lotteries (i.e., possible outcomes), a decision-maker can always decide whether they prefer one over the other, or are indifferent between them.

• Transitivity: If a decision-maker prefers lottery A over lottery B, and lottery B over lottery C, then they must prefer lottery A over lottery C.

• Continuity: If a decision-maker prefers lottery A over B and B over C, then there exists some probability mixture of A and C that the decision-maker will find equally attractive as B.

• Independence (or substitution): If a decision-maker is indifferent between two lotteries A and B, then they should be indifferent between any probability mixtures of A and C and B and C.

The theorem states that if a decision-maker's preferences for uncertain outcomes satisfy these four axioms, then there exists a utility function that can be used to rank these outcomes according to the decision-maker's preferences. This utility function is unique up to a positive linear transformation, meaning that it can be scaled and shifted but not changed in shape.

An Example: To illustrate how the VNM utility theorem works, consider an investor with \$10,000 to invest. They have two investment options:

• A risk-free government bond that will return \$10,500 in a year (a 5% return)

• A risky stock that has a 50% chance of returning \$11,000 (a 10% return) and a 50% chance of returning \$10,000 (no return)

Assuming the investor's preferences satisfy the VNM axioms, we can use a utility function to represent their preferences. Suppose the investor's utility function is U(x) = √x, where x is the final wealth.

• The expected utility of the bond is U(\$10,500) = √10500 ≈ 102.47.

• The expected utility of the stock is 0.5 * U(\$11,000) + 0.5 * U(\$10,000) = 0.5 * √11000 + 0.5 * √10000 ≈ 51.96 + 50 = 101.96.

Since the expected utility of the bond is higher than that of the stock, according to the VNM utility theorem, the investor should prefer the bond over the stock.

Implications for Finance and Investing

The VNM utility theorem has profound implications for finance and investing. It provides a theoretical foundation for expected utility theory, which is a cornerstone of modern economic and financial theory. Here are some of the main implications:

Risk-Aversion and Portfolio Selection: The VNM utility theorem is fundamental to understanding risk aversion, a key concept in finance. Investors typically prefer certain outcomes to uncertain ones, even if the uncertain ones have a higher expected return. This behavior is called risk aversion and can be represented by a concave utility function, which assigns higher utility to certain outcomes compared to uncertain ones with the same expected value. The VNM utility theorem provides a rigorous, mathematical basis for understanding and modeling this behavior.

In the context of portfolio selection, investors use their utility functions to determine the optimal mix of risky and risk-free assets. The optimal portfolio maximizes the investor's expected utility. The shape of the utility function, which reflects the investor's risk aversion, determines how much risk the investor is willing to bear.

Market Pricing and Valuation: The VNM utility theorem also underlies the fundamental theorem of asset pricing, which states that in a frictionless market with no arbitrage opportunities, the price of a risky asset must be its expected payoff discounted at the risk-free rate. The expected payoff is calculated using a risk-neutral measure, which is derived from the preferences of a hypothetical investor who is indifferent to risk (i.e., has a linear utility function). The VNM utility theorem provides a formal justification for using this risk-neutral measure.

Furthermore, the VNM utility theorem is crucial for the valuation of derivative securities, such as options and futures. These securities derive their value from the future price of an underlying asset, which is uncertain. By assigning utilities to different future prices and using the independence axiom to simplify the problem, investors can calculate the expected utility of a derivative security and hence its value.

Behavioral Finance: The VNM utility theorem also has implications for the field of behavioral finance, which studies how psychological biases and irrational behavior affect financial markets. While the VNM utility theorem assumes that investors are perfectly rational and always maximize expected utility, empirical evidence suggests that this is not always the case. For example, many investors exhibit loss aversion, meaning they are more distressed by potential losses than they are pleased by equivalent gains. This behavior violates the VNM axioms and cannot be represented by a standard utility function.

However, by relaxing some of the VNM axioms, researchers have developed alternative models that can account for these behavioral anomalies. For example, prospect theory, developed by Daniel Kahneman and Amos Tversky, assumes that investors evaluate gains and losses relative to a reference point and are more sensitive to losses than to gains. This theory can explain many observed deviations from expected utility theory and has opened up new avenues for research in finance and investing.

The Von Neumann-Morgenstern utility theorem has played a critical role in shaping modern economic and financial theory. By providing a rigorous, mathematical framework for understanding and modeling preferences under uncertainty, the theorem has led to significant insights into risk aversion, portfolio selection, market pricing, and valuation. Moreover, by highlighting the limitations of the expected utility theory, the theorem has spurred the development of alternative models that better reflect observed investor behavior. As such, the Von Neumann-Morgenstern utility theorem continues to exert a profound influence on the field of finance and investing.

AI systems and the Von Neumann-Morgenstern Utility Theorem

Artificial Intelligence (AI) has been at the forefront of financial innovation, particularly in areas such as portfolio management, risk modeling, and algorithmic trading. The Von Neumann-Morgenstern (VNM) utility theorem, with its robust mathematical framework for decision-making under uncertainty, has potential applications in these areas, empowering AI systems to make more informed and rational financial decisions.

AI in Portfolio Management: In portfolio management, AI algorithms can use the principles of the VNM utility theorem to optimize portfolio selection. Given an investor's utility function, an AI system can compute the expected utility of different portfolios and select the one that maximizes the investor's expected utility. For instance, a machine learning model might be trained on historical data to predict the returns of different assets. Using these predictions, the AI system can calculate the expected utility of various portfolio compositions and identify the optimal mix of assets. The VNM utility theorem provides a mathematical basis for this decision-making process, allowing the AI system to handle the inherent uncertainty in financial markets.

AI in Risk Modeling: Risk modeling is another area where the VNM utility theorem can be applied. AI systems can leverage this theorem to quantify and manage financial risks. Using the VNM utility theorem, an AI system can assign utilities to different outcomes and determine a decision-maker's risk tolerance based on their utility function. This information can be used to develop risk models that reflect the decision-maker's preferences and attitudes towards risk. For example, an AI risk management system might use the VNM utility theorem to evaluate the riskiness of different trading strategies. By assigning utilities to the potential outcomes of each strategy, the system can calculate the expected utility and choose the strategy that best aligns with the decision-maker's risk tolerance.

Limitations and Future Directions: While the VNM utility theorem provides a powerful tool for AI in finance and investing, it is important to note its limitations. The theorem assumes that decision-makers are perfectly rational and always maximize expected utility, which may not always be the case in reality. Moreover, accurately estimating utilities and probabilities can be challenging, particularly in complex and dynamic financial markets. AI systems must be carefully designed and trained to handle these challenges, and to adapt to changing market conditions.

Despite these challenges, the integration of the VNM utility theorem and AI holds promise for the future of finance and investing. By combining the mathematical rigor of the VNM utility theorem with the predictive power of AI, financial institutions can develop sophisticated systems that make more informed and rational decisions, ultimately leading to better financial outcomes. Research in this area is ongoing, and as AI technology continues to evolve, the applications of the VNM utility theorem in finance and investing are likely to expand and become more sophisticated.

Interesting fact: The Von Neumann-Morgenstern utility theorem has its roots in the field of game theory, not finance. John Von Neumann, one of the theorem's namesakes, is considered one of the pioneers of game theory, a branch of mathematics that deals with the study of strategic interactions among rational decision-makers. Von Neumann and economist Oskar Morgenstern originally developed the utility theorem to solve problems in game theory. However, their work quickly found applications in various other fields, including economics, political science, and finance. In particular, the theorem's framework for decision-making under uncertainty has proven incredibly valuable in finance, where uncertainty is a fundamental aspect of almost all activities, from investing in stocks to valuing derivative securities. Despite not originally being intended for finance, the Von Neumann-Morgenstern utility theorem has become a cornerstone of modern financial theory.