Choices, Values, and Frames (book)

Daniel Kahneman and the late Amos Tversky are two of the Godfathers of behavioural economics - and their body of work since the 1970s gave a really firm yank on the rug of rationality that had thus-far supported traditional theories of human economic behaviour. The "rational consumer" we met in Economics 101 is actually a bundle of conflicts and paradoxes - and this volume is a colleciton of 42 essays that explain why people can make seemingly illogical decisions when confronted by choices, or risks.

The papers, some of them densely mathematical and others written in cogent plain English, demonstrate through experiments and some quite entertaining observations the various layers of logic we use when we make trade-off decisions.

One human skew comes from the way we tend to weigh up uncertainty and risk: and our tendency to vividly imagine downsides while having only a loose picture of upsides.

A second family of biases comes from our tendency to contextualise decisions and choices in different ways: to "reframe" the way we see problems and thus skew our outcomes.

A third family of human biases comes from our many perceptual biases, for example in the way we recall standout features from one sample, and then act as if these standouts (a couple of famous names for example)are representative of the whole sample.

Taken together we get a rich picture of human decision behaviour that is applicable in economics, marketing or - (for example) in the specific worlds of medical professionals who must evaluate data, frame each situation and make critical professional decisions.

This text is important, and provides rich reading for researchers in a wide number of fields. In the last 100 years of the history of human psychology, the emphasis has generally been on the background the nature/nurture of the subject. Kahneman and Tversky and their good company of researchers demonstrate that if we shine the spotlight on the decision-making processes of humans we can uncover many, many answers to why we behave as we do.

I took a star off for those hyper-mathematical papers. These weren't at all user friendly, and while I use statistical analysis each day in my work, I'm old-fashioned enough to think that a book still must communicate verbally. But that's my skew. Don't let it put you off the importance of this collection of extremely interesting studies.

New theme, not always easy to understand but so interesting, a must for a financier.

This collection of 42 essays,most of which were first published as articles in academic journals ,demonstrates a shocking amount of ignorance on the part of the essay authors concerning the technical contributions made by J M Keynes to decision theory in his 1921 book,A Treatise on Probability(TP).This is most likely due to the great influence on economists and philosophers of the error filled reviews of the TP made by Frank P Ramsey in 1922 and 1926.Ramsey based his review of the TP primarily on chapters 1-4 of the TP plus several pages taken at random from the other four parts of the TP.Ramsey totally misread and misinterpreted Keynes's initial,elementary discussion of the measurement of probabilities in chapter 3 of the TP.Ramsey misinterpreted the meaning of the terms"nonnumerical" and "nonmeasurable" to mean that Keynes was arguing that it was ,in general,not possible to use numbers to quantify the probability relationship.Thus,Ramsey concluded in both essays that Keynes's nonnumerical probabilities were "mysterious"entities that did not satisfy the probability calculus.Exactly the opposite is the case.Keynes warned the reader of chapter 3 of the TP on p.37 that only after Part II of the TP had been covered would Keynes's approach to quantification be complete.In chapters 15 and 17 ,as well as in chapters 20 and 22,Keynes made it quite clear that by nonnumerical he meant not by a single numeral(number).Thus,it took TWO numerals(numbers)to estimate a probability in general.Keynes is the founder of the interval estimate approach to probability.Thus,nonnumerical is defined explicitly to mean"...between numerical limits"on page 160 of the TP.The reader should note that Keynes emphasized the word "between".Keynes called his approach "inexact numerical APPROXIMATION".Keynes is the founder of the interval estimate approach to probability.A reader of this book,however,might not even know that Keynes ever lived.Keynes is the first scholar in history to specifically define an index,w, to measure the weight of the evidence(what Ellsberg called the ambiguity of the evidence and Savage called the vagueness of the evidence).On p.315 of the TP ,w is defined to lie on the unit interval [0,1].Thus,0<=w<=1.This is 40 years before Ellsberg defined rho on the same unit interval.Keynes developed a decision formula that incorporated variables that allowed a decision maker to include w(or rho)and nonlinear probability preferences(nonadditive sub and super probabilities)into a conventional coefficient of weight and risk,c.The use of the c coefficient solves all of the paradoxes of subjective expected utility theory as well as accounting for all of the effects discussed over the years by Kahneman and Tversky,such as the certainty,reflection,isolation and preference reversal effects.Keynes would certainly agree with essay 11 by M Rabin,titled"Diminishing Marginal Utility of Wealth Cannot Explain Risk Aversion",since this exact conclusion is derived by Keynes on p. 315 of theTP. Despite the glaring omission of any discussion of Keynes's contributions to decision theory,I still recommend that a specialist buy this book.It contains many valuable articles.

This should be the foundation of anyone's Ph.D. reading list in IO Psych or Judgment and Decision Making. Most of the classics.

Kahneman and Tversky's compilation of articles in this book is an outstanding exposition of recent advances in cognitive psychology, especially advances associated with prospect theory. The work presented in this volume is largely responsible for the authors being awarded the Nobel Prize (Tversky died before receiving it).

The text is somewhat dense at parts, being aimed at economists and psychologists with some mathematical familiarity. However, the portions of the book that require much mathematics can safely be bypassed without losing much of the substance of the text. This text is the most credible presentation of an alternative theory to the rational actor theory usually assumed in economics. For example, some of the articles help explain the magnitude of the equity return premium, or help show how people make choices differently in similar situations based simply on the way the situation is presented.

I would highly recommend this book to anyone interested in decision making theory, especially as it relates to consumer behavior. It is a brilliant volume that includes the most important articles by the leading mind in the field.