|
Field
Abstract
|
Before
We study how social and behavioral factors drive charitable giving. The context for this project is an ongoing large-scale campaign in India, where the government aims to improve the targeting of fuel subsidies by encouraging citizens to voluntarily donate their subsidy benefits. We implement a direct-marketing intervention on top of this existing campaign in seven major cities in India, where we send out solicitation mails to 100,000 households. The design is a cluster-randomized trial with individual-level variation, within treated clusters, of the nature of the marketing message. The marketing message varies in several dimensions -- social distance, information on the value of the subsidy, signaling, and moral suasion. This study aims to measure the spillover effects of charitable giving using a relative saturation design which randomly varies the fraction of households who receive a direct-marketing solicitation in each cluster from 0% to 100%. At the same time, the individual-level treatment arms on the content of the solicitation are designed to separate several reasons for spillovers in giving. We use administrative and survey data to measure the effects of the experimental interventions and to interpret households’ motivations. To complement our analysis, we conduct an in-person charitable solicitation (for a different charity) during surveys.
.
|
After
Social norms about who is deserving may help target public benefits, by making sure people in need receive benefits or by excluding people with lesser need to reduce program costs. The Government of India launched a campaign to use norms to improve the targeting of a cooking gas subsidy used by 220 million beneficiary households. The campaign, called Give It Up, asks beneficiary households to voluntarily forfeit cooking gas subsidies so the money can be used to subsidize the poor instead. We propose a field experiment to test the efficacy of self-targeting as a policy instrument and to measure whether the campaign induces spillovers in giving across households. The main policy outcome, usage of the cooking gas subsidy, is measured with high-quality administrative data in a sample of 1.2 million gas-using households. The cluster-randomized design will provide the first experimental estimates of the strength of spillover effects in charitable giving.
|