Cardinal Revealed Preference, Price-Dependent Utility, and Consistent Binary Choice

34 Pages Posted: 24 Jan 2018 Last revised: 25 Jan 2018

See all articles by Victor Aguiar

Victor Aguiar

Western University (University of Western Ontario)

Roberto Serrano

Brown University

Date Written: January 17, 2018

Abstract

We present a new notion of cardinal revealed preference that exploits the expenditure information in classical consumer theory environments with finite data. We propose a new behavioral axiom, Acyclic Enticement (AE), that requires the acyclicity of the cardinal revealed-preference relation. AE is logically independent from the Weak Axiom of Revealed Preference (WARP). We show that the Generalized Axiom of Revealed Preference (GARP), which characterizes the standard rational consumer, is logically equivalent to AE and WARP. We propose a new notion of rationalization by means of a price-dependent utility function that characterizes AE, which in particular is suitable for welfare analysis. We also propose a consistency condition for preference functions that is equivalent to WARP. We use our axiomatic decomposition to show, in experimental and scanner consumer-panel data sets, that AE explains the majority of the predictive success of GARP. Moreover, AE taken alone is superior in predictive success to both WARP and GARP.

Keywords: Consumer Theory; Rationality; Revealed Preference Approach; Bounded Rationality

JEL Classification: C60; D10

Suggested Citation

Aguiar, Victor and Serrano, Roberto, Cardinal Revealed Preference, Price-Dependent Utility, and Consistent Binary Choice (January 17, 2018). Available at SSRN: https://ssrn.com/abstract=3103950 or http://dx.doi.org/10.2139/ssrn.3103950

Victor Aguiar

Western University (University of Western Ontario) ( email )

1151 Richmond
London, Ontario N6A5C2
Canada

Roberto Serrano (Contact Author)

Brown University ( email )

64 Waterman Street
Providence, RI 02912
United States
401-863-1036 (Phone)
401-863-1970 (Fax)

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
62
Abstract Views
606
Rank
696,161
PlumX Metrics