Is Expected Utility a good theory for explaining how people make choices - Essay Example

? IS EXPECTED UTILITY A GOOD THEORY FOR EXPLAINING HOW PEOPLE MAKE CHOICES? by 15 January Is Expected Utility a Good Theory for Explaining How People Make Choices? Introduction How people make choices remains one of the most complex and controversial issues of economic and probabilistic analysis. Dozens of theories were developed to explain the rational and emotional underpinnings of making choices. The problem is in that a whole set of factors and elements influence the process of making choices. Therefore, it is virtually impossible to distinguish the most important factors from the least important ones. More often than not, economic theorists and scholars in statistics and probability choose to describe one single element of human decision-making and fail to account for a variety of subjective influences and objective circumstances that change the decision-making reality. Theory of Expected Utility is fairly regarded as one of the most challenging, controversial, and sophisticated theories of making choices. Theory of Expected Utility provides a brief insight into how individuals weigh and choose the anticipated utilities of different actions and decisions. Unfortunately, Theory of Expected Utility is too narrow to explain how people make choices: the theory exhibits unbelievable insensitivity to emotional and probabilistic factors and does not account for the natural human striving to preserve emotional and rational status quo by all possible means. Theory of Expected Utility (or Expected Utility Theory – EUT) is rightly considered as one of the most complex and controversial explanations to how people make choices. EUT suggests that “the decision-maker chooses between risky or uncertain prospects by comparing their expected utility values, i.e., the weighted sums obtained by adding the utility values of outcomes multiplied by their respective probabilities” (Davis, Hands & Maki 1997, p.342). Put simply, EUT hypothesizes that, when making choices, individuals will rationally weigh the expected utility values of different decision-making options and their relative probability. The result of multiplying anticipated outcomes by their respective probabilities will lay the ground for making the ultimate choice (Davis, Hands & Maki 1997). Apparently, EUT relies on the intrinsic striving by humans to be rational and objective in their choices and decisions. It should be noted, that the roots and origins of EUT date back to the middle of the 18th century, when the first solutions to the St. Petersburg paradox were developed (Cohen 1994). Daniel Bernoulli was the first to propose an idea that any expectation and decision-making that follows would be integrally linked to the notion of personal worth or personal utility, which individuals attributed to each particular option (Cohen 1994). Moreover, it was due to Bernoulli that the process of making choices was presented in numerical terms (Cohen 1994). With time, making choices came to exemplify a complex set of rational steps and acts, which laid the ground for making one specified preference over other stated options (Cohen 1994). As a result, EUT was developed to reflect a theoretical belief that any option could be assigned a numerical value, generally described as “utility”, with the process of making choices directed toward the option with the highest expected utility (Cohen 1994). Despite its theoretical contribution, EUT is too narrow to reflect, discuss, and predict how people make choices. The growing body of scholarly criticism does not leave any room for EUT. The fact is in that individuals are being governed by a variety of non-rational, emotional, subjective meanings that are beyond the scope of EUT. EUT does not account for the fact that individuals tend to make emotional choices and are willing to preserve their status quo by all possible means. Therefore, EUT can hardly be a relevant means of explaining how people make choices. Rather, it presents only one, narrow dimension of the complex process of taking the best and most worthy decision. To begin with, EUT treats utility as a numerical significance of each particular option multiplied by its probability (Davis, Hands & Maki 1997). Simply stated, utility is a combination of personal worth and the probability which individuals assign to the desired outcome. However, the current state of economic science refutes a EUT assumption that individuals make choices, based on the economic worth and probability of the desired outcomes. Rather, it is certainty that plays the dominant role on how people make choices. According to Kahneman and Tversky (1979), individuals tend to overweigh the outcomes and options that seem more certain compared with other alternatives. This is what Kahneman and Tversky (1979) label “the certainty effect”. Numerous experiments and mathematical examples illustrate that individuals continuously violate one of the basic premises of EUT: they choose the options, where the desired outcome (e.g., winning) is the most probable (Kahneman & Tversky 1979). These are the common human attitudes of toward risks, chances, and decision-making, which EUT can never capture (Kahneman & Tversky 1979). In a similar vein, EUT has little to do with the risks of losses, which individuals may attribute to various decision-making options and outcomes (Kahneman & Tversky 1979). These, however, are not the only problems with EUT and its implications for decision-making. The fact is in that individuals are willing to preserve their status quo by all possible means. Therefore, utility and probability are not the only factors affecting ultimate choice. Rather, the utility and probability of the desired outcome must be significant enough to make individuals give up their stability and make a choice. This is what Kahneman, Knetsch and Thaler (1991) call the “endowment effect”. The endowment effect was first described in 1980 by Thaler, who wrote that people often want too much to give up something they already possess. Moreover, people naturally tend to want more for giving up an object they already possess than they are willing to pay for acquiring a new object (Thaler 1980). These asymmetries are beyond the scope of EUT but they make the picture of making individual choices more sophisticated and complete. Take a simple example: individuals are given a lottery ticket or its equivalent of $2.00 (Kahneman, Knetsch and Thaler 1991). After some time, those having a lottery ticket are offered to exchange it for a certain amount of cash (Kahneman, Knetsch and Thaler 1991). Surprisingly or not, only few agree to give up the tickets they already possess for a definite sum of money (Kahneman, Knetsch and Thaler 1991). Even free market conditions and ample learning opportunities do not reduce the intrinsic human tendency toward the status quo (Kahneman, Knetsch and Thaler 1991). Therefore, when making choices, individuals will not simply choose an option with the greatest utility, but will also consider whether it is worth making a decision at all. Given potential losses, it is possible that individuals will choose to stay with what they already have, rather than rushing for something that is new to them. Unfortunately, EUT does not imply that individuals may choose to stay with what they already have. On the contrary EUT is about choosing from the multiple options, all of which are equally new but different in the amount of utility and probability they are being assigned. Needless to say, most decisions are being taken in the conditions of uncertainty. More importantly, it is uncertainty that makes the process of making choices particularly difficult and virtually unpredictable. Objective and subjective reality are influenced by so many factors, that it is almost impossible to weigh and determine relative significance of the each. Unfortunately, and contrary to EUT, individual choices and decisions are only partially rational. Moreover, it would fair to assume that individual choices and decisions are rational only to a small extent, with the rest of the decision-making process being governed by uncertain predictions and misconceptions as for the relevance, validity, and probability of the desired outcome. Insensitivity to prior probability is one of the principal mistakes, which individuals make when taking the best decision and an aspect, which EUT fails to take into account (Tversky & Kahneman 1974). In simple terms, insensitivity to prior probability is a form of individual negligence toward prior assumptions, beliefs, and situations. The problem of insensitivity to prior probability originates in the difficulties, which individuals experience while judging the worthiness and relevance of available evidence (Tversky & Kahneman 1974). The current research proves that, contrary to EUT, individuals fail to objectively evaluate the worthiness of available evidence and its relevance to the desired outcomes. On the contrary, individuals are inclined to judge any available evidence as worthy rather than have no evidence at all (Tversky & Kahneman 1974). Likewise, individuals display a tendency to compare anticipated outcomes to similar situations and similar outcomes. The likelihood of each result is evaluated, compared with the results in similar situations. However, rationality aside, individuals fail to anticipate and account for the differences in corresponding samples (Tversky & Kahneman 1974). These issues with representativeness of different samples and their comparative value are beyond the scope of EUT and its implications for decision-making. Individuals forget about the importance of the sample size, blinded by the essence of the problem itself and its possible effects in the individuals’ future (Tversky & Kahneman 1974). More often than not, individuals fall short of rationality and objectivity, of which EUT speaks. They lack coldness and reason needed to evaluate relative utility of each decision and the desirability of its outcomes. The situation is further complicated by the so-called illusion of validity, when invalid evidence and information is taken for valid and relevant. This misconception about human rationality is easy to explain. Tversky and Kahneman (1974) write that people tend to choose the outcome which is most representative of the evidence and information they have at hand. That is, individuals judge the quality and relevance of the existing match between the desired outcome and the information they possess about it (Tversky & Kahneman 1974). In these situations, little or no attention is given to the lack of accuracy in available information, leaving no room for objectivity and rationality (Tversky & Kahneman 1974). Individuals in these situations violate all possible principles and underlying meanings of EUT. The major problem of EUT is in its unbelievable, even unreasonable rationality, which is neither possible nor realistic. The more rational EUT seems the more unrealistic it becomes. In the meantime, individuals find themselves in the midst of a broad illusion of validity, which persists even when individuals are well aware of the factors that limit the accuracy and relevance of available information (Tversky & Kahneman 1974). Why people tend to be confident that available evidence is the most reliable and relevant and can be used to predict the desired outcome is difficult to explain. Most probably, the human nature, intrinsically irrational and inherently unreasonable, makes the process of making choices too difficult to describe in numerical, objective terms. In this context, emotions often become the greatest barrier to taking reasonable decisions. Emotions are believed to impede the development and implementation of reasonable, objective models of decision-making and do not leave any single chance to evaluating the probability of each particular decision-making outcome. Emotions are a frequent topic of discussion in the context of individual decision-making. The process of making choices necessarily involves an emotional element, which is absent from EUT. EUT is limited to numerical categories and treats individuals as rational beings, leaving emotions and their implications for decision-making beyond the scope of the economic analysis. However, the current state of economics, statistics, and probability displays the growing concern about categories and factors other than numbers. Emotions have already become a vital element of human decision-making in conditions of uncertainty. Neither non-additive nor non-probabilistic theories of decision that lay at the heart of EUT can explain how emotions affect individual decisions. In the meantime, individuals exhibit extraordinarily high susceptibility to emotional factors of decision-making in the conditions of uncertainty. The example of a “hungry shopper” is extremely demonstrative, showing how the emotional and mental state of the individual can impede his (her) ability to take rational decisions and make rational choices. A hungry shopper enters a shopping mall – his feeling of hunger affects his mental and emotional state which, consequentially, predetermines and predicts the intuitive choices, which the shopper makes for or against each particular product. A hungry shopper asked to taste a spoonful or a yogurt or any other food product cannot be reasonable in predicting the effects of eating the whole cup (Kahneman & Thaler 2006). Nor can they anticipate that their product and shopping experiences will change as soon as they reach the point of saturation (Kahneman & Thaler 2006). Hungry shoppers tasting a spoonful of a plain yogurt erroneously believe that their perceptions of eating the full cup of the same yogurt will be the same as eating one single spoonful of the same product (Kahneman & Thaler 2006). However, they do not account for the fact that “ingesting a substantial amount of a disliked substance is distinctly worse than a single swallow” (Kahneman & Thaler 2006). The case is particularly relevant to the situations, when emotional and moral choices are at stake. The emotional changeability of the human nature and individual inability to control many or most of emotions make the prediction of emotional and hedonic choices extremely difficult (Kahneman & Thaler 2006). Hungry and emotionally aroused, individuals will naturally misinterpret the utility and probability of their emotional responses to each particular decision-making outcome. Contrary to EUT, individuals that experience strong emotions lack rational abilities to predict and evaluate utility and probability of each decision-making option. They fail to anticipate the changes in their emotional state that will follow and invariably affect their emotional responses to various decisions. This is one of the solid reasons why hungry shoppers make choices different from those made by shoppers who do not experience the feeling of hunger (Kahneman & Thaler 2006). In a similar way, individuals that experience special, strong emotions, toward one specific choice will tend to make this choice, irrespective of the reasonable arguments that either support or refute such a choice. Unfortunately, EUT is too narrow to account for human emotions. Yet, the current research treats emotions as one of the most reliable factors of human decision-making in the conditions of uncertainty. Therefore, it is at least incorrect to assume that EUT can reasonably explain how choices are usually made. A wide range of factors and influences other than those discussed by EUT make the theory extremely complicated and inapplicable in real-life situations. For example, individuals are increasingly susceptible to the ways the question is being framed. Framing the question largely predetermines the desired option, rather than rational beliefs about utility and probability, as presented by EUT. Hirshleifer and Riley (2000) describe an example, when individuals are given $200 and are asked to either choose additional $50 or a 30% chance to win additional $200. Or they can be given $400 and offered to either give up $150 or an 80% chance of losing one half of their sum (Hirschleifer & Riley 2000). In either case, individuals can choose to gain or lose nothing (Hirschleifer & Riley 2000). The experiments show that individuals additional $50 in the first problem and losing 80% of losing a half of their sum in the second problem, although the rational utility of both options is equal (Hirschleifer & Riley 2000). All these examples prove a simple truth: in no way can EUT explain how individuals make their choices. Moreover, the basic principles of EUT create an extremely complicated picture of individual decision-making, which is too rational to be true and too reasonable to be applicable in conditions of uncertainty. EUT is correct in that individuals pursue a complex procedure of making choices. However, EUT displaces the relevance of irrational components of uncertainty, which inevitably affect the essence and quality of the final decision. The principal problems with EUT include (a) the absence of an emotional angle of decision-making, (b) failure to anticipate the changes in individual emotional state over time, (c) individual failure to account for representativeness and prior probability, and (d) individual reluctance to give up the status quo even for the sake of an extremely valuable decision. The growing body of criticism in regards to EUT is logical and anticipated, given its theoretical and especially practical inconsistency. However, the value of EUT is in that it gives rise to the development of new, more objective and realistic theories of making individual choices. For the most part, EUT shows how decisions ARE NOT AND SHOULD NOT BE MADE. Conclusion Expected Utility Theory was developed to explain how people make choices. According to EUT, expected utility values serve the principal criterion of choosing between multiple options and taking the final decision. EUT assumes that while trying to take the best decision, individuals evaluate the expected utility of each particular option and its relative probability, compared to other outcomes. Unfortunately, EUT fails to reasonably explain how individuals make choices and does not account for a wide range of factors that affect individual decisions. To begin with, utility often gives place to certainty; in other words, individuals tend to overweight the utility of choices which are the most probable. Endowment effects and individual reluctance to give up their status quo further complicate the situation. More often than not, individuals want too much to give up the resources and possessions they already have; and even ample learning opportunities and free market conditions do not change the situation. Individuals exhibit extraordinary insensitivity to prior probability and extensively rely on their emotional state. All these factors are beyond the scope of EUT, which seems too reasonable, rational, and well-structured to reflect the reality of individual decision-making in conditions of uncertainty. The problem of EUT is in that it is too narrow to describe how people make choices. EUT does not account for emotionality and changeability of the environmental conditions that affect the process of taking individual decisions. How individuals make choices is being influenced by a wide range of factors, from their emotional state to the way questions and problems are presented. In light of everything said above and given the growing of body of criticism in regards to EUT, the latter looks like an ideal representation of individual decision-making in conditions of uncertainty. EUT exemplifies an interesting, measurable dimension of individual choices but in no way can it serve a reliable explanation to how individuals make choices in real-life conditions. References Cohen, BJ 1994, ‘Is expected utility theory normative for medical decision-making?’, Medical Decision-Making, vol.16, no.1, pp.1-6. Hirshleifer, J & Riley, JG 2000, The analytics of uncertainty and information, Cambridge University Press. Kahneman, D & Tversky, A 1979, ‘Prospect theory: An analysis of decision under risk’, Econometrica, vol.47, no.2, pp.263-292. Kahneman, D, Knetsch, JL & Thaler, R 1991, ‘Anomalies: The endowment effect, loss aversion and status quo bias’, The Journal of Economic Perspectives, vol.5, no.1, pp.193-206. Kahneman, D & Thaler, R 2006, ‘Anomalies: Utility maximization and experienced utility’, Journal of Economic Perspectives, vol.20, no.1, pp.221-234. Thaler, RH & Sustein, CR 2008, Nudge: Improving decisions about health, wealth, and happiness, Yale University Press. Tversky, A & Kahneman, D 1974, ‘Judgment under uncertainty: heuristics and biases’, Science, vol.185, no.4157, pp.1124-1131. Read More