Axiomatic Utility Theory Under Risk

The first attempts to develop a utility theory for choice situations under risk were undertaken by Cramer (1728) and Bernoulli (1738). Considering the famous St. Petersburg Paradox! - a lottery with an infinite expected monetary value -Bernoulli (1738, p. 209) observed that most people would not spend a significant amount of money to engage in that gamble. To account for this observation, Bernoulli (1738, pp. 199-201) proposed that the expected monetary value has to be replaced by the expected utility ("moral expectation") as the relevant criterion for decision making under risk. However, Bernoulli's 2 argument and particularly his choice of a logarithmic utility function seem to be rather arbitrary since they are based entirely on intuitively 3 appealing examples. Not until two centuries later, did von Neumann and Morgenstern (1947) prove that if the preferences of the decision maker satisfy cer tain assumptions they can be represented by the expected value of a real-valued utility function defined on the set of consequences. Despite the identical mathematical form of expected utility, the theory of von Neumann and Morgenstern and Bernoulli's approach have, however, IFor comprehensive discussions of this paradox cf. Menger (1934), Samuelson (1960), (1977), Shapley (1977a), Aumann (1977), Jorland (1987), and Zabell (1987). 2Cramer (1728, p. 212), on the other hand, proposed that the utility of an amount of money is given by the square root of this amount.

The need to understand the theories and applications of economic and finance risk has been clear to everyone since the financial crisis, and this collection of original essays proffers broad, high-level explanations of risk and uncertainty. The economics of risk and uncertainty is unlike most branches of economics in spanning from the individual decision-maker to the market (and indeed, social decisions), and ranging from purely theoretical analysis through individual experimentation, empirical analysis, and applied and policy decisions. It also has close and sometimes conflicting relationships with theoretical and applied statistics, and psychology. The aim of this volume is to provide an overview of diverse aspects of this field, ranging from classical and foundational work through current developments. Presents coherent summaries of risk and uncertainty that inform major areas in economics and finance Divides coverage between theoretical, empirical, and experimental findings Makes the economics of risk and uncertainty accessible to scholars in fields outside economics

1. PROGRESS IN UTILITY AND RISK THEORY At the First International Congress of Utility and Risk Theory in Oslo 1982 (FUR-82) it appeared to be a widespread feeling among the participants that the conference signalled something like a paradigm shift in the field. This does not necessarily mean that old truths were discarded and replaced by new ones, but rather that new theories and new empirical evidence were brought forth, compelling old theories to be critically analyzed from new angels. Some of the papers presented at FUR-82 have been published by Reidel in 1983 in a volume edited by Stigum and Wenst0p. The present volume contains com mentaries on a number of the papers presented at the conference together with broader outlines of current views on the theory. The observation that utility and risk theory now appears to be in a state of rapid change has prompted us to choose the title PROGRESS IN UTILITY AND RISK THEORY for the book, in the belief that science always moves from poorer to more advanced paradigms or from weaker to more forceful theories. In other words, change is usually progress, even though intermediate stages in a para digm shift may be bewildering, to say the least.

The Conference on "Utility: Theories, Measurements, and Applications" met at the Inn at Pasatiempo in Santa Cruz, California, from June II to 15, 1989. The all-star cast of attendees are listed as authors in the Table of Contents of this book (see p. V), except for Soo Hong Chew and Amos Tversky. The purpose of the conference, and of National Science Foundation Grant No. SES-8823012 that supported it, was to confront proponents of new generalized theories of utility with leading decision analysts com mitted to the implementation, in practice, of the more traditional theory that these new theories reject. That traditional model is variously iden tified in this book as expected utility or subjectively expected utility maximization (EU or SEU for short) and variously attributed to von Neumann and Morgenstern or Savage. I had feared that the conference might consist of an acrimonious debate between Olympian normative theorists uninterested in what people actually do and behavioral modelers obsessed with the cognitive illusions and uninterested in helping people to make wise decisions. I was entirely wrong. The conferees, in two dramatic straw votes at the open ing session, unanimously endorsed traditional SEU as the appropriate normative model and unanimously agreed that people don't act as that model requires. (These votes had a profound impact on my thinking; detail about them and about that impact is located in Chapter 10.

In this valuable book, Paul Schoemaker summarizes recent experimental and field research that he and others have undertaken regarding the descrip tive validity of expected utility theory as a model of choice under uncer tainty. His principal message is that this paradigm is too narrow in its con ception and misses some of the important elements of a descriptive model of individual choice. In particular, Schoemaker calls attention to the impor tance of individual differences, task effects, and context effects as they influence behavior. The expected utility hypothesis has come under scrutiny in recent years from a number of different quarters. This book brings together these many studies and relates them to the large body of literature on individual de cision making under risk. Although this paradigm may be appropriate for describing behavior under many conditions of uncertainty, Schoemaker presents convincing evidence that it does not do well with respect to protec tion against low-probability events. For example, he shows that the insur ance purchase decision is influenced by the way information is presented to the client, as well as by the statistical knowledge of the respondents.

Utility theory or, value theory in general, is certainly the cornerstone of decision theory, game theory, microecon~mics, and all social and political theories which deal with public decisions. Recently the American School of utility, founded by von N eumann Morgenstern, encountered a far-going criticism by the French School of utility represented by its founder Allais. The whole basis of the theory of decisions involving risk has been shaken and put into question. Consequently, basic research in the fundamentals of utility and value theory evolved into a crisis. Like any crisis in basic research, and this one was not an exception, it was very fruitful. One may simply say: Allais versus von Neumann-Morgenstern, or the French School of utility versus the American School, became one of the battlefields of scientific development which proved to be a most creative source of new advances and new developments in all those sciences which are based on evaluation of utilities.

In this volume we present some o~ the papers that were delivered at FUR-82 - the First International Con~erence on Foundations o~ Utility and Risk Theory in Oslo, June 1982. The purpose o~ the con~erence was to provide a ~orum within which scientists could report on interesting applications o~ modern decision theory and exchange ideas about controversial issues in the ~oundations o~ the theory o~ choice under un certainty. With that purpose in mind we have selected a mixture of applied and theoretical papers that we hope will appeal to a wide spectrum o~ readers ~rom graduate students in social science departments and business schools to people involved in making hardheaded decisions in business and government. In an introductory article Ole Hagen gives an overview o~ various paradoxes in utility and risk theory and discusses these in the light o~ scientific methodology. He concludes the article by calling ~or joint efforts to provide decision makers with warkable theories. Kenneth Arrow takes up the same issue on a broad basis in his paper where he discusses the implications o~ behavior under uncertainty for policy. In the theoretical papers the reader will ~ind attempts at de~initive Statements of the meaning o~ old concepts and suggestions for the adoption o~ new concepts. For instance, Maurice Allais discusses four di~ferent interpretations o~ the axioms o~ probability and explains the need ~or an empirical characterization o~ the concept of chance.

The aim of this encyclopedia is to provide a comprehensive reference work on scientific and other scholarly research on the quality of life, including health-related quality of life research or also called patient-reported outcomes research. Since the 1960s two overlapping but fairly distinct research communities and traditions have developed concerning ideas about the quality of life, individually and collectively, one with a fairly narrow focus on health-related issues and one with a quite broad focus. In many ways, the central issues of these fields have roots extending to the observations and speculations of ancient philosophers, creating a continuous exploration by diverse explorers in diverse historic and cultural circumstances over several centuries of the qualities of human existence. What we have not had so far is a single, multidimensional reference work connecting the most salient and important contributions to the relevant fields. Entries are organized alphabetically and cover basic concepts, relatively well established facts, lawlike and causal relations, theories, methods, standardized tests, biographic entries on significant figures, organizational profiles, indicators and indexes of qualities of individuals and of communities of diverse sizes, including rural areas, towns, cities, counties, provinces, states, regions, countries and groups of countries.

Economic analysis of choice under uncertainty has been dominated by the expected utility (EU) model, yet the EU model has never been without critics. Psychologists accumulated evidence that individual choices under uncertainty were inconsistent with the predictions of the EU model. Applied work in areas such as finance was dominated by the simpler mean-variance analysis. In the 1980s this skepticism was dispelled as a number of generalizations of EU were proposed, most of which were capable of explaining evidence inconsistent with EU, while preserving transitivity and dominance. Generalized expected utility is now a flourishing subfield of economics, with dozens of competing models and considerable literature exploring their theoretical properties and comparing their empirical performance. But the EU model remains the principal tool for the analysis of choice under uncertainty. There is a view that generalized models are too difficult to handle or incapable of generating sharp results. This creates a need to show that the new models can be used in the kinds of economic analysis for which EU has been used, and that they can yield new and interesting results. This book meets this need by describing one of the most popular generalized models -- the rank-dependent expected utility model (RDEU), also known as anticipated utility, EU with rank-dependent preferences, the dual theory of choice under uncertainty, and simply as rank-dependent utility. As the many names indicate, the model has been approached in many ways by many scientists and for this reason, consideration of a single model sheds light on many of the concerns that have motivated the development of generalized utility models. The popularity of the RDEU model rests on its simplicity and tractability. The standard tools of analysis developed for EU theory may be applied to the RDEU model, but since RDEU admits behavior inconsistent with EU, the field of potential applications is widened. As such, the RDEU model is not as much a competitor to EU as an extension based on less restrictive assumptions.

The Second International Conference on Foundations of Utility and Risk Theory was held in Venice, June 1984. This volume presents some of the papers delivered at FUR-84. (The First International Conference, FUR-82, was held in Oslo and some of the papers presented on that occasion were published by Reidel in the volume Foundations of Utility and Risk Theory with Applications, edited by Bernt P. Stigum and Fred Wenst~p). The theory of choice under uncertainty involves a vast range of controversial issues in many fields like economics, philosophy, psychology, mathematics and statistics. The idea of discussing these problems in international conferences has been successful: two conferences have been held and others will follow. The climate of the debate has changed in the meantime, partly as a result of these conferences. It is no more only a question of attacking or defending the neo-Bernoullian assumptions, but also of proposing wider generalizations and including new elements in the analysis of the decision process. For instance Amartya Sen - comparing the two current notions of rationality, internal consistency and self-interest pursuit introduces the concept of reasoning and considers the irrationality which may result from the failure of a positive correspondence between reasoning and choice or from a limited capacity of reasoning. Rationality is also considered with respect to the controversial axiom of strong independence. John C. Harsanyi introduces the concept of practical certainty, i. e.