Risk Perception and Expected Utility Theory

Affective Decision Making: A Behavioral Theory of Choice

Anat Bracha and Donald J. Brown
COWLES FOUNDATION FOR RESEARCH IN ECONOMICS
YALE UNIVERSITY

November 13, 2007

Abstract

Affective decision-making (ADM) is a refutable and predictive theory of
individual choice under risk and uncertainty. It generalizes expected utility
theory by positing the existence of two cognitive processes – the “rational”
and the “emotional” process. Observed choice is the result of their simultaneous
interaction. We present a model of affective choice in insurance markets, where
risk perceptions are endogenous.

Keywords: Affective Choice, Endogenous Risk Perception, Expected Utility
Theory, Insurance

"ADM is a behavioral theory of choice - a property shared by consumer demand analysis – see part one in Deaton and Muellbauer (1980), but not evident in other strategic models of choice behavior such as Gul—Pessendorfer (2001), Bernheim—Rangel (2004) or Fudenberg—Levine (2006)...
ADM is a game-theoretic model of individual decision-making under risk and
uncertainty, which generalizes expected utility, and where the probability weights –
perceived risk–are endogenous, as implied by optimism bias (Slovic 2000, Weinstein
1980). In our model of individual decision-making there are two distinct psychological processes that mutually determine choice. This approach is inspired in part by Kahneman (2003), who proposes two systems of reasoning that differ in several important aspects, such as emotion. We call these systems of reasoning the rational process and the emotional process...

The emotional process is where risk perception is formed. In particular, the agent selects an optimal risk perception to balance two contradictory impulses: (1) affective motivation and (2) a taste for accuracy. This is a definition of motivated reasoning, a psychological mechanism where emotional goals motivate agent’s beliefs, e.g., Kunda (1990), and is a source of psychological biases, such as optimism bias...

The systematic departure of the ADM model from the expected utility model
allows for both optimism and pessimism in choosing the level of insurance, and shows,
consistent with consumer research (Keller and Block 1996), that campaigns intended
to educate consumers on the loss size in the bad state can have the unintended
consequence that consumers purchase less, rather than more, insurance. Hence, the
ADM model suggests that the failure of the expected utility model to explain some
data sets may be due to systematic affective biases."