This is a summary of the presentation given by Professor Daniel Kahneman at the ISPOR 10th Annual International Meeting First Plenary Session, May 16, 2005, Washington, DC, USA. Professor Kahneman is Eugene Higgins Professor of Psychology and Professor of Public Affairs at the Woodrow Wilson School, Princeton University, Princeton, NJ, USA.

Decision Utility versus Experience Utility
The concept of utility has had two quite different meanings in its history. Originally, Jeremy Bentham [1] introduced the concept of utility as the measure of pain and pleasure as the sovereign masters that can govern our conduct. This meaning of the term utility was used through the 19th century. Around the beginning of the 20th century, the development of certain philosophies of science known as positivism began. In psychology, positivism took the form of behaviorism, and the concept of utility as the experience of pain and pleasure was discredited because experience is private and therefore not amenable (so it was argued) to scientific study. What was substituted was a concept of utility as expressed in decisions. It is no longer what you experience or what you enjoy, it is what you want. Wants are much more observable directly than enjoyment.

They are observable because choices are observable. Utility became the degree to which one wants something. This concept has dominated economic and decision theory. Economists then derived from this concept their view of well-being. For economists, well-being of societies is measured by gross domestic product (GDP) or GDP per capita. The theory is that the more choices you have, the better off you are. If you have more opportunities to choose, then you cannot be worse off. Thus, to economists, income is a good proxy to the number of choices you have. In that sense, GDP is a measure of social welfare. Nevertheless, many people (including economists) are dissatisfied with this concept of utility. Now, there is a trend to return to the “Jeremy Bentham” concept of utility.

Decision utility is what is studied when we observe choices, and in the theoretical context, decision utilities infer some choices and are used to explain choices. Experienced utility is the utility that people experience. It is what makes life in its various aspects good and bad. In the early 20th century, when utility became decision utility, the standard model of economics assumed a rational agent. If a rational agent  does what is best for him/her, then the quality of the outcomes of that agent does not have to be measured, because the outcomes are as good as they could be. So if the agent is a maximizing agent, and the wants of the agent are known, there is no need to investigate whether the agent enjoys the consequences of her choices. But behavioral economists now believe that people are not necessarily always rational and if people are not rational, they do not necessarily always choose what is best for them. And in that case, it becomes useful to measure consequences, independently of people’s wants. So interest in the experienced utility grows.