This award funds work to develop a new mathematical model of decision making with limited attention. The goal is to build a new link between research on decision theory and research on bounded rationality. The end result is new ways to analyze, predict, and understand how individuals make decisions. Numerous studies in psychology, finance, marketing, and other areas have demonstrated that people have limited attention for making choices. That is, rather than considering all possible options most people first compare and then choose from a much more limited set of options called a consideration set.
The project is designed to answer three broad kinds of research questions. First, how can social scientists infer preferences and consideration sets from datat that only records the choices people made, not the specific options considered? Second, what are the specific ways in which decision makers with this kind of limited attention behave in ways that cannot be explained by standard economic models? Third, how do people decide what options to consider? How do they form their consideration sets and what effect do outside factors like marketing campaigns or public policy have on these consideration sets?
Theis research will have broader impact by using methods and tools from microeconomic theory to answer questions of interest to a large and interdisciplinary research community. In addition, the research has the potential to guide public policy to new kinds of "nudge" interventions that might be able to guarantee that individuals consider the right set of options, a set that does not arbitrarily exclude excellent choices.
Project Outcome Report Limited attention has been widely studied in economics: neglecting nontransparent taxes (Raj Chetty, Adam Looney, and Kory Kroft (2009)), inattention to released information (Stefano DellaVigna and Joshua Pollet (2007)), costly information acquisition (Xavier Gabaix, et al. (2006)), and rational inattention in macroeconomics (Christopher A. Sims (2003)). This project aims to study decision making with limited attention by extending the revealed preference paradigm to accommodate the notion of limited attention. We focus on three main questions: (1) How can one infer preferences and consideration sets (those items that a decision maker pays attention to) from observed choice data? (2) How do decision makers with limited attention differ behaviorally from standard economic agents? (3) How are consideration sets formed? The distinction between a preference and an (in)attention is crucial. Hence, in a welfare analysis it is important to understand the underlying model of the decision-maker. The main of this project is to define a revealed preference ordering when there are multiple explanations for the same choice behavior. We provide a definition, which is very conservative. It says "a product is revealed to be preferred to another alternative only when all possible explanations agree on it." Our aim is to avoid any claims for which there may be doubt about. This conservative approach ensures that a social planner can always follow our welfare recommendations. Our main result states that one can deduce both the decision maker's preference and the alternatives that she pays attention to from observed behavior. Our result provides a guideline to identify both preferences and consideration sets. We then provide an example that highlights the importance of knowledge about the underlying choice procedure for welfare analysis. In our example model free welfare analysis approach leads to wrong conclusion about preferences. In essence, we point out that welfare analysis is a more delicate task than it looks. Our project has so far produced two published papers. One of them is published in a top general interest journal (American Economic Review) and the other is one of the top field journal (Theoretical Economics). We have also two working papers. The American Economic Review paper is cited well and recognized by many scholars from both in and outside of our field. Broader Impact: The interaction between economics, psychology and decision theory is extremely fruitful to all fields. We show how revealed preference theory can help us better understand the concepts of limited attention and consideration sets concepts, which are well-established in marketing and psychology. Most studies in marketing and psychology take consideration sets as primitives. We illustrate how to infer consideration sets via observed choice behavior, allowing scholars in these fields to test their models. As a result, scholars in marketing and psychology, in addition to those in economics, will benefit from our results. There are also benefits to policy makers. Knowing which options individuals pay attention to, and when, can be used to better structure choice settings (for example menus of retirement plans), so that individuals do not overlook options, while still allowing them the flexibility of choice. Given the increasing demand for behavioral economics courses both at undergraduate and graduate levels, the project has merit for teaching purposes. Our results are both intuitive and mathematically simple so that they can be taught to advanced undergraduates. Graduate students benefit from learning these topics and applying them to, among other field, marketing, public policy, and industrial organization.
