Liquidity constraints: Allocating clean water to the rural poor

How to cite

Burlig, Fiona; Jina, Amir; Sudarshan, Anant. 2022. "Liquidity constraints: Allocating clean water to the rural poor." AEA RCT Registry. AEARCTR-0010545. https://doi.org/10.1257/rct.10545-2.0

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

Fiona Burlig* , Amir Jina , Anant Sudarshan

* Primary investigator

Registry ID
AEARCTR-0010545
DOI
https://doi.org/10.1257/rct.10545-2.0
First registered
Last updated
Status
On going
Trial dates
2022-05-20 – 2025-05-01
Countries
Private
JEL codes
D45, O13, Q25, Q53
Registry record
http://www.socialscienceregistry.org/trials/10545

Abstract

How should scarce resources be allocated in developing countries? Weitzman (1977) highlights a trade-off between prices, which generate allocative efficiency, and quotas, which might have desirable distributional consequences. In partnership with a private company supplying clean water to rural Odisha, India, we plan to run an experiment to measure the relative effectiveness of different allocation mechanisms. We will measure the price elasticity of demand for clean water, health effects from consuming clean water, and the extent to which liquidity constraints and intra-household inefficiencies reduce consumption. To do so, we implement a cluster-randomized trial, where 160 villages are randomized into a pure control group and multiple treatment arms: (i) discounts; (ii) a monthly quota; and (iii) an exchangeable quota, where unused allocation can be exchanged for cash. We subsequently randomize which households within each treatment village will receive treatment. We plan to measure effects of treatment on water consumption and health outcomes using a combination of survey and administrative data.

Intervention

We implement a cluster-randomized trial in which villages are randomized into one of four conditions: pure control, discount, pure quota, and exchangeable quota. In each treatment (i.e. not pure control) village, households are subsequently randomized to receive clean water through one of the corresponding allocation mechanisms. The discount and exchangeable quotas have sub-arms that vary the level of the marginal incentive.

Experimental design

We conduct a cluster-randomized trial that assigns each of 160 villages to one of four groups, and then allocates approximately 40 households per treatment village to treatment and the remainder to control. We also sample approximately 15 households to survey in each village.

Design details

Randomization method
Randomization done in office by a computer
Randomization unit
First level: village. Second level: household.
Clusters
160 villages
Observations
160 villages x approximately 500 households per village: approx 80,000 households. Approx 1,800 households in the survey.
Arms
40 villages pure control; 40 villages discount; 40 villages pure quota; 40 villages exchangeable quota. Approximately 40 treatment households per treated village, and all remaining households are control.
Primary outcomes
Quantity of water purchased; health outcomes.
Secondary outcomes
Please see our PDF PAP for more detailed information on outcomes.

Cited by

No citation data available for this registration yet.

Keywords