Re-Evaluating Assumptions about Behavior and Choice in Response to Public Assistance: A Behavioral Decision Analysis

WCPC Seminar Series on Poverty and Policy: Spring 2012

Presented by Crystal Hall
Assistant Professor of Public Affairs
University of Washington
Monday, May 14, 2012;  12:30 - 1:30 p.m., questions / discussion until 2:00 p.m.
Parrington Hall Commons, Room 308
University of Washington


Crystal Hall joined the Evans School faculty in 2008. She teaches courses on psychology for policy analysis, decision theory, and quantitative analysis.

While at Princeton University, her primary research focus was on decision making in the context of poverty. Specific topics in this research program include the influence of trust on the choice between financial contracts, the structure of mental accounting among low-income populations, and how simple interventions relating to self-affirmation and identity can influence behavior. Other previous research topics included the relationship between facial appearance and election outcomes, how the existence of extra information can impair judgment, and the exploration of an alternate method of implementing social judgment theory. She was previously a member of the interdisciplinary Fellowship of Woodrow Wilson Scholars.

In addition to her academic work, Hall has provided guidance to community organizations seeking to implement tools from psychology and behavioral economics into the design and delivery of their programs and services. She has worked with organizations in Central New Jersey and Philadelphia, and most recently with members of the Washington Asset Building Coalition.

Hall holds a Ph.D. and MA in Psychology from Princeton University. In addition, she holds a BS from Carnegie Mellon University in both decision science and policy and management.


Public "safety net" programs intended to help low-income households are based on a series of assumptions about human preferences and decision-making. Despite their importance to social service design and implementation, these assumptions are rarely stated explicitly or empirically tested. In some instances, key assumptions may reflect theoretical constructs carried over from an earlier era of social service delivery, or decision models for higher-income populations that do not hold in a low-income context. Shifts in the demography of US poverty since the 1970s and the increasing stigmatization of the poor may exacerbate the disconnects between assumptions and actual behavior--which may have important implications for program take-up and outcomes. Using a set of public assistance programs as case studies, this paper will examine the implicit and explicit assumptions about how households make decisions in response to public assistance, and identify whether empirical research regarding decision-making and choice supports these foundational assumptions. Despite the potential insights that a behavioral perspective can bring to policy design, the field has remained largely silent and no comprehensive review of the evidence on decisions in context of poverty has been conducted. We employ a behavioral decision-making framework to discuss decisions in the context of poverty, through a nuanced analysis of factors low-income households may consider when engaging in judgment and choice. This approach emphasizes the role of contextual factors in decision-making for individuals of any income, and argues that an applied behavioral view offers novel tools to help policymakers serve low-income populations. This perspective provides a unique and under-utilized framework useful to explain some behavioral puzzles, examine and predict the actions of individuals living poverty, and understand disappointing program outcomes. For example, why are take-up rates for public assistance programs often strikingly low? Why do low-income Housing Choice Voucher (HCV) recipients tend to remain in high poverty neighborhoods, even when "better" options should be available? Which approaches may encourage low-income households to save, or participate in formal banking? We argue that a striking disconnect exists between assumptions and evidence, potentially resulting in less effective policy design and implementation, at a substantial financial and social cost.


See the slides from this presentation here.