Self-Timed Credit Contracts (STICC)

How to cite

Snowberg, Erik; Agness, Daniel; Chassang, Sylvain; Dupas, Pascaline. 2018. "Self-Timed Credit Contracts (STICC)." AEA RCT Registry. AEARCTR-0002949. https://doi.org/10.1257/rct.2949-1.0

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

Erik Snowberg* , Daniel Agness , Sylvain Chassang , Pascaline Dupas

* Primary investigator

Registry ID
AEARCTR-0002949
DOI
https://doi.org/10.1257/rct.2949-1.0
First registered
Last updated
Status
On going
Trial dates
2018-04-01 – 2019-12-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/2949

Abstract

In order to understand how loan contract design affects take up of credit, and eventual repayment rates, we test three different approaches of credit for a single product: an agricultural hip-pump. The first (“M500”) is a standard loan with monthly payments for a year. The second (“Freedom”) only requires that half of the payment be made by the middle of the year, and the full payment by the end of the year. A third contract (“Flex”) has basically the same payoff structure as the “Freedom” contract, but asks those who take up the product to make a repayment plan, and sends reminders when they pledged to pay a certain amount. This experiment uses the same sample as STRIP (https://www.socialscienceregistry.org/trials/183/history/799), and consists of people who are in clusters where 2 people have already received the same agricultural hip-pump under different allocation mechanisms. As such, we will also be able to determine how different mechanisms to identify those who will “experiment” on the behalf of the cluster affect the demand for credit. We will also examine how a "group nudge" encouraging group ownership of the hip-pump through the provision of information relevant to group ownership affects take up of the credit offer, as well as group ownership.

Intervention

Each household in our sample of 7,994 farming households from 390 clusters will be offered a randomly chosen experimental loan contract from a group of three. The first (“M500”) is a standard loan with monthly payments for a year. The second (“Freedom”) only requires that half of the payment be made by the middle of the year, and the full payment by the end of the year. A third contract (“Flex”) has basically the same payoff structure as the “Freedom” contract, but asks those who take up the product to make a repayment plan, and sends reminders when they pledged to pay a certain amount. A group nudge, randomized at the cluster level, will provide participants information about a common group ownership structure.

Experimental design

All participants in the STRIP trial (https://www.socialscienceregistry.org/trials/183/history/799) who did not receive a free pump through STRIP, which constitutes 7994 farming households across 390 clusters will be offered one of three credit contracts. Loan contract types are stratified by cluster, so there will be an approximately equal number of each contract type offered in each cluster. The group nudge is randomized at the cluster level.

Design details

Randomization method
Randomization done in office by a computer.
Randomization unit
Individual.
Clusters
390, with only group nudge randomized by cluster
Observations
7,994, equally split between each contract type. 954 used for piloting, 7,040 for main sample.
Arms
Roughly evenly split across all six arms (three contract types by receiving group nudge or not)
Primary outcomes
Interest in each credit contract, rates of group ownership, take-up of contract (defined as the down payment of 1,000 KSH, at which point the participant will get the pump), full repayment rates. The last three will be examined overall, and conditional on group ownership
Secondary outcomes
Patterns of repayment over time under each contract, difficulty and cost of repossessing pumps from those who are in default, assessments of loan structure, pump, and agricultural productivity

Cited by

No citation data available for this registration yet.

Keywords