Evidence on the Determinants of Generosity
Paper Session
Tuesday, Jan. 5, 2021 10:00 AM - 12:00 PM (EST)
- Chair: Marta Serra-Garcia, University of California-San Diego
(When) Do Transfers Respond to Need? Evidence from Disasters and Media Coverage
Abstract
Communities affected by natural disasters often receive financial aid from both the government and private donors. If such aid is largely determined by need, then charitable donations and government disaster aid could become an increasingly important form of social insurance as extreme events become more frequent due to climate change. Media attention to a disaster has also been anecdotally identified as an important determinant of aid, but rigorous empirical evidence on this question is lacking. Using rich databases of news media coverage, of the universe of donations to the American Red Cross, and of US federal disaster aid, we estimate the effect of media coverage on aid in the aftermath of tornadoes and floods. Recognizing that media coverage is endogenous to unobserved disaster severity, we instrument for the former using the occurrence of other noteworthy stories that should not affect charitable giving or aid directly. We measure media attention to other events by calculating the share of content devoted to the top ten percent of news stories over the past week. We then show that this measure is negatively related to subsequent flood and tornado coverage over various time horizons, demonstrating that it permanently reduces media attention to these disasters. Second-stage estimates are forthcoming.Vice and Virtue Behaviours: Substitution and Non-Substitution Effects
Abstract
We examine how tax policy that encourages charitable giving spills over into other morally motivated behaviours. We examine behaviours that are seemingly unrelated to charitable giving, but that are easily classified as virtuous or vice behaviours—exercise and smoking—to study how people jointly engage in virtuous behaviours (charitable giving and exercise) as well as how people jointly engage in a mixed or vice-virtue bundle (charitable giving and smoking). Using the Panel Study of Income Dynamics, we estimate marginal tax rates for 17,181 individuals over 5 panels from which we can calculate the effective price of giving to charity. We then estimate the structural model developed by Dinardo & Lemieux (1992, 2001) to disentangle substitution effects (i.e., changes in behaviour that arise from changes in relative prices) from non-substitution effects (i.e., changes in behaviour that arise through behavioural channels). Contrary to reduced form results which confound substitution and non-substitution effects, we find that charitable giving and exercise are substitutes in consumption, but the positive non-substitution effect dominate, rendering a positive relationship between charitable giving and exercise. Charitable giving and smoking are also substitutes in consumption and the negative non-substitution effects contribute to a negative relationship between charitable giving and smoking. Together, we interpret both the dominance and the direction of the non-substitution effects as evidence consistent with moral consistency.Working for “Good”: How Prosocial Incentives Affect Effort and Cheating
Abstract
Prosocial incentives and other Corporate Social Responsibility initiatives, where worker's effort benefits a charitable cause, have been suggested as a potential tool for motivating workers. Prior research has shown that indeed prosocial incentives can motivate effort. Yet, they are often outperformed by traditional incentives. However, past work has typically only looked at one dimension of performance -- work quantity. In a novel experimental design, we capture an additional dimension – work quality– using a task that allows us to observe whether workers' output is achieved with or without using prohibited shortcuts (i.e., via cheating). In an online experiment, we study how prosocial incentives affect both work quantity and work quality, as compared to standard incentives. Looking only to work quantity, we observe a familiar pattern: individuals worked hardest and were most sensitive to incentive-size under the standard incentive scheme. Looking to work quality, we see a reversal: prosocial incentives produce higher-quality work, as workers' output is achieved with significantly fewer prohibited shortcuts. Our results suggest that linking workers' effort to a charitable contribution may be desirable in ways that were not captured by the existing literature.Discussant(s)
Rene Bekkers
,
Free University Amsterdam
Mark Ottoni-Wilhelm
,
Indiana University
Daniel Hungerman
,
University of Notre Dame
Marta Serra-Garcia
,
University of California-San Diego
JEL Classifications
- H4 - Publicly Provided Goods
- D0 - General