Document Type

Article

Publication Date

4-28-2016

Abstract

We propose a comparative model of decision making under risk, uncertainty, and time, in which large differences in payoffs and probabilities or dates of receipt are perceived as salient and overweighted in the evaluation process. The predictions of the model depend on what differences are compared across alternatives which, in turn, depends on how the choice is framed. We formalize a class of matrix-based frames which applies to decisions under risk, uncertainty, and time, and we specify two important types of frames within this class: minimal frames which provide the simplest representation of choice alternatives, and transparent frames which make the normative appeal of the classical rationality axioms more transparent. We also propose two simple and natural assumptions regarding the perceived salience of differences in numerical magnitudes. We show that the model predicts systematic framing effects in which people will exhibit major violations of rational choice theory (the Allais paradox, common ratio effect, Ellsberg paradox, present bias, and violations of stochastic dominance) when the options are represented in a minimal frame but will behave more consistently with the classical axioms when the same choices are presented in a transparent frame. The model employs the same salience-based decision algorithm across the domains of risk, uncertainty, and time, thus providing a unified approach to explaining choice anomalies as decision errors. Moreover, because it maintains the assumption that preferences obey expected and discounted utility, it facilitates traditional welfare analysis.

Comments

Working Paper 16-08

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.