A Rational Reconstruction of the Compromise Effect: Using Market Data to Infer Utilities

This article explores the possibility that consumers use market data to make inferences about product utilities. The argument is made by means of an example based on the "compromise effect" found in extant experimental data. This phenomenon is generally looked at as a manifestation of deviations from rationality in choice. However, assuming full rationality, I describe a decision rule that is based on consumers' inferences about their information about their own relative tastes. Through a number of examples, I will argue that consumers often use this or similar decision rules to make inferences about utility. I then show that the decision rule may generate compromise effects in experiments and that it may be sustainable. The compromise effect could therefore be seen as preliminary evidence that consumers make such inferences. Copyright 1995 by the University of Chicago.