Douglas Bernheim - US grants

This project attempts to expand our understanding of social and economic behavior in environments with imperfect information. The project consists of two main parts. The first part of the project examines clustering, while the second studies incomplete contracting. "Clustering" refers to situations in which people act similarly, even though their preferences are different. Examples include herd behavior (for example, the formation of cultural or sub-cultural standards of behavior), the common practice of dividing estates equally among heirs, and the proliferation of "me too" political candidates. This project explores a theory of clustering based on two key assumptions: decision-makers desire popularity or esteem, and actions are observable but true characteristics are not. In a class of mathematical models, these assumptions produce clustering of choices, even when true characteristics are not clustered. This part of the study will lead to novel predictions about consumption behavior in various markets. "Incomplete contracting" refers to the fact that contracts often fail to cover many contingencies. This project explores a theory of incomplete contracts based on two assumptions: certain actions are not observable by those who enforce contracts (for example, courts or arbitrators), and parties to a contract will interact again in the future. In a class of mathematical models, these assumptions imply that parties to an explicit contract may leave some provisions ambiguous in order to improve the implicit aspects of their agreement. This part of the study will provide insights concerning why contracts leave out certain contingencies even when including them does not seem very costly.

This research is supported by the Joint NSF/Private Sector Initiative and focuses on the following two closely related questions: Do individuals make reasonable financial decisions for savings and retirement? How do various forms of financial education affect financial behavior? The cooperating organization is the Smith Richardson Foundation. Numerous previous authors have documented the poor quality of personal financial decisions, and several studies have, in particular, found that Americans save too little for retirement. The adequacy of saving for retirement among younger Americans has, however, proven controversial, in large part beaause pertinent economic developments over the next 30 to 40 years are highly uncertain, and may profoundly affect retirement prospects. Previous research has also shed considerable light on the general public's level of economic and financial literacy. However, few determinants of economic literacy have been studies (aside from formal education among high school students). The determinants of financial knowledge among adults are essentially unexplored, and virtually nothing is known about the effects of knowledge on financial behavior. Finally, the existing literature has not systematically considered the effects of financial education on financial decisions; nor has it evaluated the various mechanisms (knowledge, comfort, or habituation) through which financial education may affect behavior. Thus, a central objective of this research is to examine financial education as an important dimension of human capital accumulation. The study will focus on the analysis of three extremely promising and previously unexploited data sources. The first is a series of household survey sponsored by Merrill Lynch; the second is a series of unique employer surveys undertaken by KPMG Peat Marwick; and the third is an extensive 401(k) database compiled by Hewitt Associates. The research is of direct relevance to private firms interested in using education to alter the behavior of participants in contributory retirement plans. Indeed, it bears on the continued viability of 401(k) plans, and the relative merits of defined benefit and defined contribution plans. In addition, the research has potentially important implications for national economic policy, including strategies for promoting national saving, the design and implementation of education curricula, pension policy, information policy, and securities regulation.

This grant provides renewed support for the annual Summer Workshops on Economic Theory conducted each summer by the Stanford Institute of Theoretical Economics (SITE). This six-week collection of workshops brings together prominent scholars as well as younger economists for presentation and discussion of their current research. This workshop series has just completed its 30th year and continues to serve as an international forum for the study of innovative topics in economic theory, mathematical economics, and related areas. The topics covered in SITE programs and the content of the sessions offer valuable and unique opportunities for economists to come together. Some topics, such as general equilibrium, microeconomic theory, and game theory, are long-term interests of SITE. However, the topics of SITE sessions have evolved, just as economic theory and mathematical economics have evolved in economics generally. Accordingly, sessions based on newer topics, such as empirical work based on theoretical models, computational methods, strategic interactions, theoretical public finance, and theoretical macroeconomics, represent new directions that SITE has taken in recent years and expects to continue in future years. The more applied sessions at SITE, such as computation and theory-based econometrics, focus on the relation to economic theory and mathematical techniques. The macroeconomics sessions are also more focussed on the theoretical and mathematical aspects of those topics than the typical macroeconomic conference.

In recent years, economic scholars in a wide variety of specializations have become increasingly interested in questions of Political Economy. A central objective of this movement has been to supplement our understanding of the relationships between the political environment (including institutions) and the economic policies that are adopted. Though much progress has been made, virtually nothing is known about the strategic issues arising from the possibility that a legislature might choose to reconsider some provision of a law after its passage but prior to its implementation. Legislative reconsideration is a common occurrence in practice. Moreover, it gives rise to complex strategic issues. The adoption of a policy provision for some future period changes the political landscape. Further discussion of the issues prior to the law's implementation takes place against a different backdrop. Though the provision can be changed, repealed, or replaced before it becomes effective, its adoption alters the terms of the ensuing debate. The purpose of this project is to explore the effects of legislative reconsideration on economic policy selection. The analysis identifies and explores a tendency for reconsideration to create high concentration of political power (irrespective of whether policies are actually changed after adoption and prior to implementation). Using the tools of game theory, the project undertakes comparative institutional analysis, examining the role of reconsideration under a variety of alternative rules and procedures within the context of particular economic problems. It seeks to understand the mechanisms by which particular institutions control (or fail to control) the tendency toward high concentration of power, and to elucidate the implications for economic policy.

The findings of this research are relevant to a wide range of practical issues. Collective choice problems are pervasive not only in public policy making, but also in cooperative research, education, partnerships, networks, and many other forms of social activity. These problems are resolved through a wide variety of formal and informal procedures. By improving our understanding of the relationships between procedures and outcomes, the study of collective choice will provide constructive guidance concerning the choice of procedures and the design of institutions. Insights concerning the effects of various rules on the influence of underrepresented groups may prove particularly useful.

This award provides a three-year continuation of NSF support for the annual Summer Workshops on Economic Theory conducted each summer by the Stanford institute of Theoretical Economics.
For more than 30 years, these workshops have attracted top scholars from around the world, giving them opportunities to present their research results as well as collaborate on new research. This collection of workshops brings together prominent scholars as well as younger economists for presentation and discussion of their research. A special effort is made to identify young economists who would benefit from attending these workshops, and give them extra help to make the trip. This outreach to newer members of the profession gives them opportunities to meet senior scholars as well as other young researchers in their fields. These workshops have been particularly effective in providing an atmosphere for scholars to learn from each other and engage in collaborative research on innovative topics in economic theory, mathematical economics, and related areas.

This project will develop formal microeconomic models to address two important questions that cannot be answered with current models.

The first project will consider cooperation and enforcement in ongoing relationships. The goal is to develop explanations for important and poorly understood behavioral patterns in how partners punish bad behavior. For example, the criminal justice system is structured with more severe punishments for repeat offenses, a pattern we have observed in other relationships as well. Another example is seen in price-fixing cartels; firms refusing to adhere to a pre-existing cartel agreement by stealing business from cartel partners are often not punished by the cartel as a whole. Again, such non-punishment has also been observed in other contexts. While these patterns are widespread, standard economic theories of on-going relationships do not do an adequate job of predicting when we will observe such behavior. These theories predict instead that punishment should be of constant severity and should always be carried out. The principal investigators will develop a new theory to address these issues.

The second project attempts to explain why a "50-50" norm is influential in a wide range of practical circumstances - such as joint ventures among corporations, share tenancy in agriculture, the division of restaurant tabs among friends - as well as in a variety of laboratory settings. The leading economic theories, which attribute this behavior to preferences for fairness, fail to provide a complete explanation for existing experimental evidence. The investigators will develop and analyze a new theory that postulates that people like to be perceived as fair, as well as preferring fair outcomes.

The PI will incorporate recent research results from behavioral economics into the theory of welfare economics.

Interest in psychological approaches to economic questions has grown in recent years, stimulated largely by accumulating evidence that the standard economic model of consumer decision-making may provide an inadequate description of human behavior. "Behavioral" economic theories that incorporate non-standard components (motivated by psychological evidence) are increasingly finding their way into policy evaluation, which inevitably involves what economists call "welfare analysis" (that is, assessments of well-being). Yet these new theories fundamentally challenge our ability to formulate appropriate criteria for judging whether outcomes are "good" or "bad." If, contrary to the traditional assumption in economics, an individual's choices do not reflect coherent preferences, how can those choices provide economists with a basis for judging whether one outcome is better than another?

This project develops, applies, and refines a unified framework for policy evaluation that encompasses the new generation of behavioral economic theories. The framework developed here is a natural extension of standard welfare economics. The standard approach is based on choice. In its simplest form, it advises a policy maker to respect the choices an individual would make for himself. The guiding principle behind that approach is an extension of the libertarian deference to freedom of choice, which takes the view that it is better to give a person the thing he would choose for himself rather than something that someone else would choose for him.

The central premise of this project is that the same principle remains applicable even when people behave in ways that are inconsistent with standard economic assumptions. Although the guidance provided by people's choices may be ambiguous in some circumstances, it is typically unambiguous in others, and this partially ambiguous guidance can provide a foundation for rigorous policy evaluation. The project develops a general framework for policy evaluation and explores its implications for particular behavioral theories. It also reexamines an assortment of public policies, employing generalizations of the standard tools that economists conventionally use when evaluating the desirability of public policies.

The broader impact of this work will follow directly from its focus on developing appropriate criteria for evaluating public policies. Previous studies that have employed non-standard behavioral models in the context of public policy analysis have generally relied on ad hoc criteria for assessing the relative desirability of different outcomes. This project will revisit many of those policy issues, such as the design of capital income taxes and social security, to identify conclusions that are rigorously justified based on the application of the new framework. In addition, this work enables greater integration of economics, psychology, and neuroscience. Specifically, it allows economists to conduct rigorous welfare analysis using literally any model of behavior, including ones imported from other fields, regardless of whether those models depict behavior as reflecting optimization.

This award funds the development of a new method for inferring the choices people will make in possible situations from a variety of 'non-choice' data. Many practical problems in microeconomics call for the researcher to predict the distribution of households' choices in not-yet-observed situations. Economists prefer to draw such inferences from actual choices in a closely related domain. However, this approach is sometimes problematic in practice because of the practical limitations of choice data. For example, we may not have data about closely related choices, or we may have serious concerns about uncontrolled factors, selections, and the endogeneity of opportunity sets.

The new method uses non-choice data. This method goes beyond current 'non-choice' methods such as the use of survey responses to include possible measures of passive physiological and neurological responses that have been widely studied in behavioral science. The new approach has several features beyond the use of non-choice data. The researchers focus on non-choice responses that are likely to have context-independent meanings over large collections of choice problems. The researchers treat the decision problem as the unit of observation, and they treat the problem as one of optimal statistical prediction. Finally, they want to determine whether and when these new methods are better than the standard method using imperfect choice data.

Broader impacts are substantial. The research could result in a new valuation method that could be used in addition to or instead of such methods as contingent valuation. The result would be improvements in how microeconomics is used to analyze a broad array of practical questions.

This award funds doctoral dissertation research in Economics that uses controlled decision-making experiments to determine whether and why financial incentives for volunteering sometimes result in reduced volunteer effort.

Previous researchers have argued that receiving a financial incentive for volunteering or donations could mar one?s social image. If this is the case,then offering an incentive might actually reduce prosocial volunteer effort; the financial incentive means that volunteering no longer convinces society that you are a generous person. This idea has been incorporated into behavioral theories, but these theories are not are not able to forecast when one may expect incentives to encourage prosocial behavior. Similarly, no study has been able to precisely identify the negative image effect.

In carefully testing the mechanisms involved in volunteer behavior, these laboratory experiments answer these questions. Study 1 implements a novel treatment that identifies and measures the negative image effect as well as other main effects. Study 2 tests a new theory that can explain the results of past research in a straightforward way. The associated negative image effect of incentivizing volunteer behavior may well be smaller for someone with an established, as opposed to unestablished, good volunteer reputation, because those with established reputations are already known to volunteer for good, not greedy, reasons. Study 3 investigates the role of diminishing returns to incentives in the context of volunteer behavior.

In achieving identification of the negative image effect as well as proposing and testing novel mechanisms that may influence prosocial behavior, the results of this research provide new insight into prosocial behavior. The laboratory stands as an ideal tool for acquiring these results, since the observability of one's actions and associated reputation as well as the level of volunteer behavior and offered incentives can all be tightly controlled and measured.

Broader Impacts: In addition to contributing to interdisciplinary research in economics, psychology, and decision science, these experimental results will provide valuable information to government and nonprofit organizations about the encouragement of volunteer behavior. Expected follow-up field studies will engender partnerships between academia and non-profit or government organizations.