Community Lending with Outside Capital

How to cite

Brooks, Wyatt; Nakitende, Marie; Oluoch-Aridi, Jackline; Brown, Danice. 2020. "Community Lending with Outside Capital." AEA RCT Registry. AEARCTR-0005498. https://doi.org/10.1257/rct.5498-1.0

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

Wyatt Brooks* , Marie Nakitende , Jackline Oluoch-Aridi , Danice Brown

* Primary investigator

Registry ID
AEARCTR-0005498
DOI
https://doi.org/10.1257/rct.5498-1.0
First registered
Last updated
Status
On going
Trial dates
2019-09-19 – 2021-12-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/5498

Abstract

We provide temporary funding to community savings groups in rural Ugandan villages. These savings groups operate on an annual cycle where all members make small weekly contributes, take out loans from the accumulated savings, and distribute the value of the pool of funds at the end of the cycle. When a new cycle begins there with no funds it takes several months before the accumulated savings of group members is sufficient to meet loan demand. We treat a random subset of groups by providing a significant amount of funding (equal to approximately 25% of average annual group savings) at the beginning of group cycles. Unlike traditional debt financing, our repayment schedule and amount is the same as other group members. That is, the funding is repaid at the end of the cycle. The return on the funding is the same as that of the group members, which is lower if members default on group loans or higher if many loans are made at the group's established interest rate. Identification of the effects of outside financing comes by comparison to the untreated groups. We study two sets of outcomes. First, we evaluate how the outside financing affects within-group lending dynamics, such as the volume of loans within the groups, default rates, and saving rates. We observe these outcomes directly by digitizing the record books of the groups. Second, we evaluate how households are affected by this change in the availability of lending using household-level surveys of group members. We focus on consumption, investment and entrepreneurship.

Intervention

We provide funding to a random subset of community savings groups for the period of their annual cycle. We are repaid with the same timing and return as the members of the group.

Experimental design

The timing of payments coincides with the beginning of the annual cycle of each group. Our representatives visit each community at the time of the first meeting and explain the intervention, and undertake the consent process. If consent is given, then funding is contributed to the group's pool of funds, which are held in a community lockbox (with several locks controlled by different group members). There is a detailed accounting of group funds held in two identical ledgers held within the lockbox, which tracks contributions by members and loans outstanding. We have no further participation in the group until the end of the cycle when we return to take our share, which is calculated based on the same formula used to compute member payouts.

Design details

Randomization method
Randomization by computer
Randomization unit
Savings group
Clusters
121 groups
Observations
We expect an average of approximately 33 members per group for a total of approximately 4000 total members.
Arms
60 treatment groups and 61 control groups.
Primary outcomes
Lending dynamics: default rates, loan volume, saving rates Household outcomes: consumption of non-durables, consumption of durables, farm investment, farm outcomes, likelihood of running a business, business investment

Cited by

No citation data available for this registration yet.

Keywords