Association for Public Policy Analysis & Management

Panel: Financial Incentives, Poverty, and Behavior Change: Lessons from Recent Research

(Poverty and Income Policy)

Financial incentives have become an increasingly popular way to try to influence a range of behaviors, including school performance, weight loss, health practices, employment, and even charitable fundraising. Incentives have been used as a tool in anti-poverty and economic mobility programs, or initiatives to improve personal well-being or pro-social behaviors. But do they work? Evidence on their effectiveness has grown substantially, painting a mixed picture. This increases the importance of understanding what types of incentives work (or don’t), for what types of people, and under what conditions. This panel will share lessons from several very different approaches. Paper 1 will present longer-term findings on two related conditional cash transfer (CCT) programs evaluated with randomized trials. Both represent attempts to adapt Mexico’s CCT program to the US. The first, Opportunity NYC –Family Rewards (or “Family Rewards 1.0”) offered a comprehensive set of incentives to low-income families, with payments tied to benchmarks in children’s education, family health care, and parents’ employment. Based on early evidence from that experiment, a revised model (“Family Rewards 2.0)” was launched in the Bronx and Memphis. It had fewer incentives and included advisors to guide families on achieving incentivized outcomes. Impact findings now available provide a wealth of evidence on the ways these interventions did and did not work. This paper will consider what the findings from both studies suggest about adopting CCTs as a poverty-fighting strategy in the US. Paper 2 will present results of an exploratory analysis of the effects of work incentives on low-income individuals employment outcomes. It uses data from the NYC Work Rewards Demonstration, which tested three interventions for recipients of housing subsidies—one involving incentives alone, one combining incentives with services, and one offering services alone. The interventions using incentives had the largest earnings impacts, but these were concentrated among initially nonemployed participants, and most consistently among nonemployed lone parents with no other adults in the household. The paper discusses a theoretical framework for interpreting these findings, drawing on recent behavioral psychology literature on scarcity and executive skills, and offers insights for designing future interventions. Paper 3 will present findings from two randomized trials testing the effects of incentives on prosocial behaviors in the realm of charitable giving. The results support a hypothesis that the best advocates for a cause are those whose motives are unadulterated by extrinsic rewards. Without any incentive, a caring individual will express true concern for a cause in a way that appears sincere to others. When an incentive is introduced, the same action becomes noticeably disingenuous. The paper presents evidence that donors can sense the sincerity of an advocate’s pitch, and that this affects their donations. Moreover, when advocates are offered a financial incentive to increase the amount of donations they raise, donors detect less sincerity and reduce their contributions. The findings highlight the potential for incentives to have unintended consequences. The discussant will comment on the studies’ relevance for public policy, including the implications for maturing CCT programs in Mexico and other countries.