# Expected utility hypothesis

The expected utility hypothesis is a popular concept in economics that serves as a reference guide for decisions when the payoff is uncertain. The theory recommends which option rational individuals should choose in a complex situation, based on their risk appetite and preferences. The expected utility hypothesis states an agent chooses between risky prospects by comparing expected utility values (i.e. the weighted sum of adding the respective utility values of payoffs multiplied by their probabilities). The summarised formula for expected utility is $\displaystyle{ U(p)=\sum u(x_k)p_k }$ where $\displaystyle{ p_k }$ is the probability that outcome indexed by $\displaystyle{ k }$ with payoff $\displaystyle{ x_k }$ is realized, and function u expresses the utility of each respective payoff.[1] On a graph, the curvature of u will explain the agent's risk attitude.

For example, if an agent derives 0 utils from 0 apples, 2 utils from one apple, and 3 utils from two apples, their expected utility for a 50–50 gamble between zero apples and two is 0.5u(0 apples) + 0.5u(2 apples) = 0.5(0 utils) + 0.5(3 utils) = 1.5 utils. Under the expected utility hypothesis, the consumer would prefer 1 apple (giving him 2 utils) to the gamble between zero and two.

Standard utility functions represent ordinal preferences. The expected utility hypothesis imposes limitations on the utility function and makes utility cardinal (though still not comparable across individuals). In the example above, any function such that u(0) < u(1) < u(2) would represent the same preferences; we could specify u(0) = 0, u(1) = 2, and u(2) = 40, for example. Under the expected utility hypothesis, setting u(2) = 3 and assuming the agent is indifferent between one apple with certainty and a gamble with a 1/3 probability of no apple and a 2/3 probability of two apples, requires that the utility of one apple must be set to u(1) = 2. This is because it requires that (1/3)u(0) + (2/3)u(2) = u(1), and (1/3)(0) + (2/3)(3) = 2.

Although the expected utility hypothesis is standard in economic modelling, it has been found to be violated in psychology experiments. For many years, psychologists and economic theorists have been developing new theories to explain these deficiencies.[2] These include prospect theory, rank-dependent expected utility and cumulative prospect theory, and bounded rationality.

## Antecedents

### Limits of the expected value theory

In the early days of the calculus of probability, classic utilitarians believed that the option which has the greatest utility will produce more pleasure or happiness for the agent and therefore must be chosen[3] The main problem with the expected value theory is that there might not be a unique correct way to quantify utility or to identify the best trade-offs. For example, some of the trade-offs may be intangible or qualitative. Rather than monetary incentives, other desirable ends can also be included in utility such as pleasure, knowledge, friendship, etc. Originally the total utility of the consumer was the sum of independent utilities of the goods. However, the expected value theory was dropped as it was considered too static and deterministic.[4] The classical counter example to the expected value theory (where everyone makes the same "correct" choice) is the St. Petersburg Paradox. This paradox questioned if marginal utilities should be ranked differently as it proved that a “correct decision” for one person is not necessarily right for another person.[4]

## Risk aversion

The expected utility theory takes into account that individuals may be risk-averse, meaning that the individual would refuse a fair gamble (a fair gamble has an expected value of zero). Risk aversion implies that their utility functions are concave and show diminishing marginal wealth utility. The risk attitude is directly related to the curvature of the utility function: risk neutral individuals have linear utility functions, while risk seeking individuals have convex utility functions and risk averse individuals have concave utility functions. The degree of risk aversion can be measured by the curvature of the utility function.

Since the risk attitudes are unchanged under affine transformations of u, the second derivative u'' is not an adequate measure of the risk aversion of a utility function. Instead, it needs to be normalized. This leads to the definition of the Arrow–Pratt[5][6] measure of absolute risk aversion:

$\displaystyle{ \mathit{ARA}(w) =-\frac{u''(w)}{u'(w)}, }$

where $\displaystyle{ w }$ is wealth.

The Arrow–Pratt measure of relative risk aversion is:

$\displaystyle{ \mathit{RRA}(w) =-\frac{wu''(w)}{u'(w)} }$

Special classes of utility functions are the CRRA (constant relative risk aversion) functions, where RRA(w) is constant, and the CARA (constant absolute risk aversion) functions, where ARA(w) is constant. They are often used in economics for simplification.

A decision that maximizes expected utility also maximizes the probability of the decision's consequences being preferable to some uncertain threshold.[7] In the absence of uncertainty about the threshold, expected utility maximization simplifies to maximizing the probability of achieving some fixed target. If the uncertainty is uniformly distributed, then expected utility maximization becomes expected value maximization. Intermediate cases lead to increasing risk aversion above some fixed threshold and increasing risk seeking below a fixed threshold.

### The problem of interpersonal utility comparisons

Understanding utilities in term of personal preferences is really challenging as it face a challenge known as the Problem of Interpersonal Utility Comparisons or the Social Welfare Function. It is frequently pointed out that ordinary people usually make comparisons, however such comparisons are empirically meaningful because the interpersonal comparisons does not show the desire of strength which is extremely relevant to measure the expected utility of decision. In other words, beside we can know X and Y has similar or identical preferences (e.g. both love cars) we cannot determine which love it more or is willing to sacrifice more to get it.[30][31]

## Recommendations

In conclusion Expected Utility theories such as Savage and von Neumann–Morgenstern have to be improved or replaced by more general representations theorems.

There are three components in the psychology field that are seen as crucial to the development of a more accurate descriptive theory of decision under risks.[25][32]

1. Theory of decision framing effect (psychology)
2. Better understanding of the psychologically relevant outcome space
3. A psychologically richer theory of the determinants

### Mixture models of choice under risk

In this model Conte (2011) found that behaviour differs between individuals and for the same individual at different times. Applying a Mixture Model fits the data significantly better than either of the two preference functionals individually.[33] Additionally it helps to estimate preferences much more accurately than the old economic models because it takes heterogeneity into account. In other words, the model assumes that different agents in the population have different functionals. The model estimate the proportion of each group to consider all forms of heterogeneity.

### Psychological expected utility model:[34]

In this model, Caplin (2001) expanded the standard prize space to include anticipatory emotions such suspense and anxiety influence on preferences and decisions. The author have replaced the standard prize space with a space of "psychological states," In this research, they open up a variety of psychologically interesting phenomena to rational analysis. This model explained how time inconsistency arises naturally in the presence of anticipations and also how this preceded emotions may change the result of choices, For example, this model founds that anxiety is anticipatory and that the desire to reduce anxiety motivates many decisions. A better understanding of the psychologically relevant outcome space will facilitate theorists to develop richer theory of determinants.

## References

1. Conte, Anna; Hey, John D.; Moffatt, Peter G. (2011-05-01). "Mixture models of choice under risk" (in en). Journal of Econometrics 162 (1): 79–88. doi:10.1016/j.jeconom.2009.10.011. ISSN 0304-4076.
2. "Stanford Encyclopedia of Philosophy". Reference Reviews 15 (6): 9. June 2001. doi:10.1108/rr.2001.15.6.9.311.
3. (in en) Expected Utility Hypotheses and the Allais Paradox. Dordrecht: Springer Netherlands. 1979. doi:10.1007/978-94-015-7629-1. ISBN 978-90-481-8354-8.
4. "The theory of risk aversion". Aspects of the Theory of Risk Bearing Reprinted in Essays in the Theory of Risk Bearing. Chicago, 1971: Markham Publ. Co.. 1965. pp. 90–109.
5. "Risk aversion in the small and in the large". Econometrica 32 (1/2): 122–136. January–April 1964. doi:10.2307/1913738.
6. Castagnoli and LiCalzi, 1996; Bordley and LiCalzi, 2000; Bordley and Kirkwood
7. "On the St. Petersburg Paradox" (in en). Scandinavian Actuarial Journal 2001 (1): 69–78. January 2001. doi:10.1080/034612301750077356. ISSN 0346-1238.
8. Martin, Robert (16 June 2008). "The St. Petersburg Paradox". Stanford Encyclopedia of Philosophy.
9. "Ramsey's Representation Theorem". Dialectica 58 (4): 483–498. 2004. doi:10.1111/j.1746-8361.2004.tb00320.x.
10. [natrep_ramsey_and_the_ethically_neutral_proposition.pdf "Ramsey and the Ethically Neutral Proposition"]. Australian National University. http://www.edwardjrelliott.com/uploads/7/4/4/7/74475147/[natrep]_ramsey_and_the_ethically_neutral_proposition.pdf.
11. "The Theory of Statistical Decision". Journal of the American Statistical Association 46 (253): 55–67. March 1951. doi:10.1080/01621459.1951.10500768. ISSN 0162-1459.
12. "The foundations of statistics (second edition), by Leonard J. Savage. Pp xv, 310. £1·75. 1972 (Dover/Constable)". The Mathematical Gazette 57 (401): 220–221. September 1973. doi:10.1017/s0025557200132589. ISSN 0025-5572.
13. "1. Foundations of probability theory", Interpretations of Probability (Berlin, New York: Walter de Gruyter): pp. 1–36, 2009-01-21, doi:10.1515/9783110213195.1, ISBN 978-3-11-021319-5
14. "Attitudes to Uncertainty in a Strategic Setting" (in en). The Economic Journal 127 (601): 809–826. 2017-05-01. doi:10.1111/ecoj.12486. ISSN 0013-0133.
15. Theory of Games and Economic Behavior (Third ed.). Princeton, NJ: Princeton University Press. 1953.
16. "A note on uncertainty and indifference curves". Review of Economic Studies 36 (1): 1–4. January 1969. doi:10.2307/2296336.
17. "A characterization of the distributions that imply mean-variance utility functions". Journal of Economic Theory 29 (1): 185–201. 1983. doi:10.1016/0022-0531(83)90129-1.
18. "On the class of elliptical distributions and their applications to the theory of portfolio choice". Journal of Finance 38 (3): 745–752. 1983. doi:10.2307/2328079.
19. "One-switch utility functions and a measure of risk". Management Science 34 (12): 1416–24. December 1988. doi:10.1287/mnsc.34.12.1416.
20. "Prospect Theory: An Analysis of Decision under Risk.". Econometrica 47 (2): 263–292. 1979. doi:10.2307/1914185.
21. Subjects changed their beliefs faster by conditioning on evidence (Bayes's theorem) than by using informal reasoning, according to a classic study by the psychologist Ward Edwards:
• Kleinmuntz, B, ed (1968). "Conservatism in Human Information Processing". Formal Representation of Human Judgment. Wiley.
• Daniel Kahneman, Paul Slovic and Amos Tversky, ed (1982). "Conservatism in Human Information Processing (excerpted)". Judgment under uncertainty: Heuristics and biases. Cambridge University Press.
• "Chapter 6: Conservatism in a simple probability inference task (Journal of Experimental Psychology (1966) 72: 346-354)". A Science of Decision Making:The Legacy of Ward Edwards. Oxford University Press. October 2008. p. 536. ISBN 978-0-19-532298-9.
22. "von Neumann Morgenstern preferences". Journal of Mathematical Economics 33 (1): 109–122. February 2000. doi:10.1016/s0304-4068(99)00004-x. ISSN 0304-4068.
23. "Rationality, the Bayesian standpoint, and the Monty-Hall problem". Frontiers in Psychology 6: 1168. 2015-08-11. doi:10.3389/fpsyg.2015.01168. PMID 26321986.
24. "Reversals of preference between bids and choices in gambling decisions". Journal of Experimental Psychology 89 (1): 46–55. 1971. doi:10.1037/h0031207.
25. "Economic Theory of Choice and the Preference Reversal Phenomenon". American Economic Review 69 (4): 623–638. 1979.
26. "Preference Reversals and the Independence Axiom". American Economic Review 76 (3): 508–515. 1986.
27. "List C. Are interpersonal comparisons of utility indeterminate?". Erkenntnis 58 (2): 229–260. 2003. doi:10.1023/a:1022094826922. ISSN 0165-0106.
28. "Simulation theory and interpersonal utility comparisons reconsidered". Synthese 191 (6): 1185–1210. April 2014. doi:10.1007/s11229-013-0318-9. ISSN 0039-7857.
29. Experiments on Decisions under Risk: The Expected Utility Hypothesis. 1980. doi:10.1007/978-94-017-5040-0. ISBN 978-94-017-5042-4.
30. "Mixture models of choice under risk" (in en). Journal of Econometrics 162 (1): 79–88. May 2011. doi:10.1016/j.jeconom.2009.10.011.
31. "Psychological Expected Utility Theory and Anticipatory Feelings" (in en). The Quarterly Journal of Economics 116 (1): 55–79. 2001-02-01. doi:10.1162/003355301556347. ISSN 0033-5533.

• Foundations of Rational Choice Under Risk. Oxford: Oxford University Press. 1993. ISBN 978-0-19-823303-9.
• "Uncertainty and the Welfare Economics of Medical Care". American Economic Review 53: 941–73. 1963.
• "Probabilism: A Critical Essay on the Theory of Probability and on the Value of Science (translation of 1931 article)". Erkenntnis 31. September 1989.
• "La Prévision: ses lois logiques, ses sources subjectives". Annales de l'Institut Henri Poincaré. 1937.
"Foresight: its Logical Laws, Its Subjective Sources (translation of the 1937 article in French". Studies in Subjective Probability. 7. New York: Wiley. 1964. pp. 1–68.
• Theory of Probability. New York: Wiley. 1974.
• Andrew Schotter, ed (1976). "Some Reflections on Utility". Selected Economic Writings of Oskar Morgenstern. New York University Press. pp. 65–70. ISBN 978-0-8147-7771-8.
• "On Small Differences in Sensation". Memoirs of the National Academy of Sciences 3: 73–83. 1885.
• Martin Shubik, ed (1967). "Subjective Probability Derived from the Morgenstern-von Neumann Utility Theory". Essays in Mathematical Economics In Honor of Oskar Morgenstern. Princeton University Press. pp. 237–251.
• "Events, Utility and Subjective Probability". Theory of Measurement. Wiley. 1968. pp. 195–220.
• "Chapter 7 (specifically) and 8, 9, 10, (to show paradoxes to the theory)". The psychology of judgment and decision making. 1993.
• "Chapter VII: Truth and Probability". The Foundations of Mathematics and other Logical Essays. 1931.
• "The Expected Utility Model: Its Variants, Purposes, Evidence and Limitations". Journal of Economic Literature 20: 529–563. 1982.
• Decision-Making: An Experimental Approach. Stanford University Press. 1957.
• "On the St. Petersburg Paradox.". Scandinavian Actuarial Journal 2001 (1): 69–78. 2001. doi:10.1080/034612301750077356.
• "Normative Theories of Rational Choice: Expected Utility". The Stanford Encyclopedia of Philosophy. 2019.
• "Strange Expectations". Philosophy of Science 47 (4): 562–567. 1980. doi:10.1086/288956.
• "The time resolution of the St Petersburg paradox". Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 369 (1956): 4913–4931. 2011. doi:10.1098/rsta.2011.0065. PMID 22042904. Bibcode2011RSPTA.369.4913P.
• "Experiments on Decisions under Risk: The Expected Utility Hypothesis.". Experiments on Decisions under Risk. 1980.