Street Charity

How to cite

Sharma, Nishtha; Malik, Samreen. 2024. "Street Charity." AEA RCT Registry. AEARCTR-0012650. https://doi.org/10.1257/rct.12650-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Nishtha Sharma* , Samreen Malik

* Primary investigator

Registry ID
AEARCTR-0012650
DOI
https://doi.org/10.1257/rct.12650-1.0
First registered
Last updated
Status
On going
Trial dates
2023-10-30 – 2024-04-15
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/12650

Abstract

Helping a stranger' is the most common form of charitable giving behavior worldwide. Nearly half of the world's adults, or two and a half billion people, have donated to a stranger over the past decade (Charities Aid Foundation, World Giving Index [2019]). Helping a stranger often takes the form of giving to a beggar, a needy person who solicits money in public places. Begging is a global urban phenomenon, and its eradication is an often-cited political agenda. Enforcing policies like banning beggars from public spaces or mandating their removal from such arenas are not sustainable solutions. Effective policies to mitigate beggary necessitate an understanding of the underlying market from both the beggar's and the givers' side.

This project has two aims: first using surveys and experimental methods we uncover the perceptions and attitudes of givers who donate to beggars; the background, motivations, and behavior of individuals resorting to begging; and the dynamics governing interactions between givers and beggars. The second aim is to understand the role of fairness in informal charitable donations. In particular, we develop a theoretical model and append it with our primary data collected from surveys and experimental methods, the project aims to test whether people have fairness considerations while making charitable decisions towards beggars.

Our field of study is Delhi, India. The first part of the project design is based on four surveys. The first survey will document interactions between beggars and givers passers-by to measure charitable transfer amounts by beggar type (Interactions Survey). The second survey involves detailed interviews with beggars and includes their preference elicitation using behavioral/experimental games (Beggars Survey). The third survey involves a belief elicitation experiment of the general population to understand the mechanisms underlying the charitable behavior measured in the interactions survey (General Population Survey). Finally, the fourth survey is an observational survey of crowded areas in Delhi and people's meritocracy to study the composition and rate of successful interactions by beggar type in areas with high and low meritocracy (Composition survey).

The second part of the project explores whether people have fairness considerations while making charitable decisions toward beggars. To understand the role of fairness preferences and beliefs in charitable behavior, we exploit a natural variation in begging styles found in most urban areas. In addition to those who only solicit charity (beggars without items), another kind of beggars found on the urban streets is those who offer trivial items such as pens, stickers, and flowers while soliciting charity (beggars with items). We propose and test a novel application of the signaling theory to examine the effect of begging with an item on people's perception of beggars' deservedness and consequent charitable donations.

In our model, meritocratic people care whether the beggar is poor and begging because they are unwilling to work and earn (choice) or unable to work and earn (luck). Since procuring an item to offer is costly for beggars, it signals a willingness to exert effort and participate in the labor market, attracting higher donations. Under the signaling framework, the perceived probability of a beggar's willingness to do paid work and overall perception of the beggar's deservedness for charity is higher for beggars with items than without, which drives higher donations towards them.

Alternatively, the feeling of reciprocity towards beggars offering items also predicts higher donations towards them than the beggars without items. However, under this reciprocity model, there is no difference in beliefs about their willingness to work or other notions of deservedness. Finally, if social preferences such as fairness concerns and reciprocity do not interact with altruism, then the charitable behavior towards beggars with and without items does not differ.
Using the theoretical model and our primary data, we seek to identify the causal impact of begging with items on the general population's beliefs about their willingness to work and their deservedness for charity. Our primary hypothesis is that begging with an item has a signaling value, i.e., people's perception of the beggar's deservedness of charity improves due to the offering of an item while begging.

Intervention

No intervention, see experimental design section.

Experimental design

The design consists of four surveys. In the first survey, we will document interactions between beggars and passers-by, aiming to measure charitable transfer amounts by beggar type (Donations Survey). The second survey involves detailed interviews with beggars and includes their preference elicitation using behavioral/experimental games (Beggars Survey). The third survey involves a belief elicitation experiment of the general population to understand the mechanisms underlying the charitable behavior measured in the interactions survey (General Population Survey). Finally, the fourth survey is an observational survey of crowded areas in Delhi and people's meritocracy to study the composition and rate of successful interactions by beggars' type in areas with high and low meritocracy (Composition survey).

Design details

Randomization method
Randomization for colony/neighborhood selection for a general survey of givers is done by stata from an entire list of colonies in Delhi by income strata. Randomization for treatment assignment (whether belief elicitation about beggar with or without items is done first) is done internally by the survey software upon the participant's consent.
Randomization unit
Individuals
Clusters
There is no clustered sampling in this study.
Observations
1200 beggars for beggars' survey and lab-in-the-field experiment. 1200 people/households for survey of general population and belief elicitation experiment. 600 charitable interactions between beggars and givers 80-90 areas for observational survey with 540 passers-by on the streets for meritocracy survey
Arms
The only survey involving experimental variation is the belief elicitation survey among the general population (described in the experimental design section). We randomize the treatments, i.e., whether they report beliefs about the beggars with items (treatment 1) or beggars without items (treatment 2) at the individual level. Each respondent will be equally likely to be assigned to either treatment. Therefore, we expect a roughly equal sample size across the two treatments, i.e., 600 in each treatment arm.
Primary outcomes
People's beliefs about the preferences of beggars with and without items over: 1. working for money 2. free-riding 3. truth-telling
Secondary outcomes
1. Share of INR 100 allocated to a beggar with an item 2. Donations amount to beggars

Cited by

No citation data available for this registration yet.

Keywords