Introduction

Over the past decade, there has been a growing and fruitful debate over the applicability of various models of limited or bounded rationality to economic decision making. In particular, in a wide variety of situations the claim is frequently made that the standard model of utility-maximization (SUM) provides an inadequate explanation for observed behavior and that some alternative behavioral model provides a better explanation. However, SUM models continue to guide applied economic research in market settings. There have not been, tomyknowledge, any empirical studies of actual market decision making that specify two competing models (oneSUMand one based on an alternative behavioral model) and test their relative strengths. This study contributes such a test in the context of residential energy consumption.

The alternative behavioral models emphasize the difficulty or the impossibility of obtaining and processing the information required to maximize utility. Although some maintain the concept of utility-maximization, they impose information and transactional constraints that alter the predicted behavior. Other models reject the concept of utilitymaximization and substitutesomeform of simplifying decision routine. For short I shall refer to each class of models, respectively, as limited utility-maximization(LUM) and bounded rationality (BR). A full review of both classes of alternative models is beyond the scope of this study. Two of their attributes are particularly relevant to note here. First, the existing tests of the alternative models almost always rely on data that are not normally observable in market settings.