Mark Schneider

Research Abstracts

Mark has long been intrigued by how people make decisions. His first few research projects have expanded on this theme, ranging from how people consume with credit cards to how political parties allocate campaign resources. He is also interested in the fundamental issues in microeconomics and more broadly, how people systematically deviate from making "rational" decisions. Inspired by the work of George Loewenstein, Dan Ariely, Robert Shiller, and other pioneers in behavioral economics, Mark thoroughly enjoys investigating the foibles and idiosyncrasies of human behavior. Below are the abstracts for several of Mark's working papers.

• Campaign Spending under Rank Dependence: A Simple Heuristic

We consider a simple allocation rule for campaign resources based on an assumption of rank dependence which constrains the possible conditions of nature on Election Day. Prior work has argued that candidates should allocate resources in proportion to the closeness of the race in each state, or that allocations should be made in proportion to the number of electoral votes in each state. We view these two conclusions as complementary rather than contradictory. We propose that candidates allocate resources in proportion to the probability of a state being pivotal, where this probability depends on both the number of electoral votes in each state and the closeness of the race. Case studies from the 2004 and 2008 U.S. presidential elections were conducted to assess the validity of this allocation rule. The evidence presents strong initial support in favor of this heuristic.

• The Credit Card Effect on Consumption and Saving

There is no explicit economic theory of payment mechanism. Classical economics implicitly assumes that "how you pay" should not affect "how much you pay." In this paper I argue, using theory and evidence from both classical and behavioral economics, that credit cards can have a marked effect on consumption and saving, relative to other payment instruments such as cash or check. I conclude that the growth in the use of credit cards should be considered as part of the explanation for the recent U.S. consumption boom and the related two-decade decline in the U.S. personal saving rate.

• Context Theory: A Procedural Approach to Individual Choice

We develop a choice-based framework called context theory, which draws from the literatures on dual-system models, procedural models, and similarity relations. Our framework introduces two decision algorithms: The context-outcome heuristic chooses based on similarity relations over contexts and outcomes. The choice-simplicity heuristic selects the option that simplifies the decision process. We identify when each heuristic is likely to be used, and we characterize the corresponding boundary conditions on rational behavior. Context theory explains notable anomalies for decisions under risk, over time, and between products. The evidence suggests that human choices are often based more on similarity judgments than optimization principles.

• The Computational Efficiency of Inherent Preferences

This paper explores the possibility that inherent (context-independent) preferences may emerge from computational constraints. Two principles are used to identify a special set of human universal preferences. First, it is assumed that the ease of certain computational tasks can "leak" into the perception of the content being processed. That is, certain environmental features are preferred because they are easier to process. Second, we assume that computational constraints lead organisms to effectively compress information. The preferences we identify appear to be highly effective adaptations. Nevertheless, these preferences occasionally give rise to observed biases in behavior, just as energy-saving heuristics may cause biases in probability and belief. We note that the principles of efficient computation provide a unifying theme for identifying and analyzing inherent preferences. There is much work to be pursued in this area across cognitive science disciplines in order to develop a "science of preferences."