Can Vouchers Reduce Elite Capture of Local Development Projects? Experimental Evidence from the Solomon Islands

How to cite

BenYishay, Ariel; Weber, Roberto; Grosjean, Pauline; D'Adda, Giovanna; Beath, Andrew. 2016. "Can Vouchers Reduce Elite Capture of Local Development Projects? Experimental Evidence from the Solomon Islands." AEA RCT Registry. AEARCTR-0001820. https://doi.org/10.1257/rct.1820-1.0

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

Ariel BenYishay* , Roberto Weber , Pauline Grosjean , Giovanna D'Adda , Andrew Beath

* Primary investigator

Registry ID
AEARCTR-0001820
DOI
https://doi.org/10.1257/rct.1820-1.0
First registered
Last updated
Status
Completed
Trial dates
2013-06-01 – 2013-11-01
Countries
Solomon Islands ()
Registry record
http://www.socialscienceregistry.org/trials/1820

Abstract

By reducing citizen oversight, external financing of public goods in developing countries can exacerbate elite capture and produce sub-optimal outcomes for communities. We test a novel mechanism channeling funds for local public goods via citizens. While control communities are offered block grants to fund local public goods, households in randomly-assigned communities are provided with vouchers that they may contribute to a public good or redeem at a discount for private consumption. Vouchers increase community participation in allocation decisions, satisfaction in decision outcomes, and the probability that allocations reflect the preferences of individuals who had not previously participated in local meetings, when different from those of leaders.

Intervention

In all villages, selected participants were informed that SBD 2,000 (approximately USD 300) had been allocated to fund the improvement of a local non-religious public facility, such as a school, health center, market, toilet, road, well, or irrigation system. The research team gathered a random sample of villagers in a meeting to discuss the type of project that they believed would most benefit the community, with the goal of reaching a consensus on which project to fund. In order to ensure the discussion resembled a community meeting, no structure was imposed on the form of the discussion or on the method of selection of the project. Sample villages were randomly allocated to either the control or treatment group, which differed in the mode of fund allocation. In villages assigned to the control group, facilitators informed participants that a block grant of SBD 2,000 would be allocated to fund the community project. In the treatment group and prior to the discussion, facilitators issued each of the 20 selected participants with 10 paper vouchers and explained that each voucher could either be redeemed for cash or contributed to the fund for the community project. If redeemed, vouchers would be worth SBD 5 each, whereas vouchers contributed to the project would be worth SBD 10 each. Following the discussion and project selection, participants in the treatment group were asked by the facilitator to indicate privately how many of the vouchers they wished to redeem and how many they wished to contribute to the community project.

Experimental design

Each of the 80 sample villages was randomly assigned into either a treatment arm and a control arm, with stratification by ward. As discussed in the Treatment description, while control villages were offered block grants to fund local public goods, households in treatment villages were provided with vouchers that they could contribute to a public good or redeem at a discount for private consumption.

Design details

Randomization method
Randomization was conducted in an office by a computer, with stratification by ward
Randomization unit
Villages
Clusters
80 villages
Observations
1,600 households
Arms
40 villages in treatment, 40 villages in control
Primary outcomes
Community members' participation in discussions about project funding, the agreement between community members' preferences and the selected projects, community members' satisfaction with the process and the selected project, and the timing and cost of implementation of the project.

Cited by

No citation data available for this registration yet.

Keywords