x

We are happy to announce that all trial registrations will now be issued DOIs (digital object identifiers). For more information, see here.
Estimating the Effects of Digital Credit in Malawi
Last registered on August 05, 2019

Pre-Trial

Trial Information
General Information
Title
Estimating the Effects of Digital Credit in Malawi
RCT ID
AEARCTR-0004139
Initial registration date
July 28, 2019
Last updated
August 05, 2019 3:05 PM EDT
Location(s)

This section is unavailable to the public. Use the button below to request access to this information.

Request Information
Primary Investigator
Affiliation
University of California, Santa Cruz
Other Primary Investigator(s)
PI Affiliation
ID Insight
PI Affiliation
Stanford University
Additional Trial Information
Status
In development
Start date
2019-07-29
End date
2020-11-30
Secondary IDs
Abstract
In the past few years, Sub-Saharan African countries have seen the emergence of digital credit – loans made through cell phones. These types of loans provide a potential solution to common problems associated with microcredit, which include high transaction costs such as the cost of traveling to the nearest bank branch or time taken by lenders in approving and processing loans. In Malawi, digital credit is on offer through a mobile network operator (thereafter telco). In terms of loan terms and conditions, digital loans look similar to payday loans in the developed world; just like with payday loans, there is an active debate over whether consumers are informed about loan terms, and whether they accurately gauge the costs of these loans.

In this project, we seek to answer: what is the impact of consumer knowledge of digital credit terms and conditions on take-up of the digital loan? We will evaluate this question using a randomized controlled trial where we test the effect of an information treatment on take-up of the digital loan. The information treatment consists of an interactive-voice-response (IVR) quiz that is designed to teach participants about the terms and conditions of the loan. We will use surveys, as well as telco-generated administrative data, to compare outcomes of those sampled for the information treatment, hose sampled for a lighter information treatment (consisting only of SMSs which state loan conditions), tohse sampled for a salience treatment (they receive an IVR call like those in the IVR information group, but the quiz is just a minute long and does not include information) and a control group. By applying this research design, we hope to identify the effect of consumer awareness of loan conditions on demand for digital credit.
External Link(s)
Registration Citation
Citation
Brailovskaya, Valentina, Pascaline Dupas and Jonathan Robinson. 2019. "Estimating the Effects of Digital Credit in Malawi." AEA RCT Registry. August 05. https://doi.org/10.1257/rct.4139-2.0.
Former Citation
Brailovskaya, Valentina, Pascaline Dupas and Jonathan Robinson. 2019. "Estimating the Effects of Digital Credit in Malawi." AEA RCT Registry. August 05. https://www.socialscienceregistry.org/trials/4139/history/51253.
Sponsors & Partners

There are documents in this trial unavailable to the public. Use the button below to request access to this information.

Request Information
Experimental Details
Interventions
Intervention(s)
We randomly assign our sample to four treatment groups:
(1) the first group (the financial literacy group) will receive an automated call which leads them through an interactive- voice-response (IVR) quiz. This IVR quiz is designed to educate consumers about the costs, terms and conditions of Airtel’s digital loans. This group will also receive an SMS inviting them to call in to the IVR quiz, as well as follow-up SMSs which state loan terms and conditions.
(2) The second group (the placebo /salience group) will also receive an automated call that leads them through an IVR quiz; this quiz, however, merely mentions Airtel loans and services, without educating consumers about loan fees or conditions. As with the financial literacy group, the placebo group will also receive an SMS inviting them to call in to the IVR quiz.
(3) The third group (the SMS group) will receive the same SMSs as the financial literacy group, i.e. SMSs which state loan terms and conditions. This group will not be invited to participate in an IVR quiz.
(4) The fourth group is the control and will not receive an intervention.
Intervention Start Date
2019-07-29
Intervention End Date
2019-12-30
Primary Outcomes
Primary Outcomes (end points)
- demand for Kutchova loans
- timeliness of repayment and late fees incurred
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
- knowledge of digital loan terms
- purpose of loan
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
We randomly assign all phone numbers eligible for the digital loan to one of four groups: a pure control group, an info-SMS group, a placebo/salience group, and a treatment group.

The pure control group receives no intervention.
The info-SMS group will receive a limited intervention comprising three text messages
which state the terms and conditions of digital loans.
The placebo group will receive an SMS invitation to take the IVR quiz and an automated call leading directly into the placebo quiz. The placebo IVR quiz is about telco-related products and services, but it does not state any terms or conditions of digital loans.
The treatment group will receive three text messages stating the terms and conditions of digital loans (these are the same SMSs as those received by the info-SMS group), as well as an SMS invitation to take the IVR quiz, and an automated call which leads directly into the financial literacy IVR quiz. (The financial literacy quiz walks listeners through a fictional scenario in which the listener makes decisions and responds to questions about the fees, terms, and conditions associated with digital loans.) The financial literacy quiz is designed to inform users about the terms and conditions of digital loans.

The randomized assignment was stratified by: (1) credit limit; (2) gender; (3) whether sim was registered in a metro area or not; (4) whether sim is registered to a unique ID or not; (5) whether the user wanted to enable mobile money when they filled out KYC.
To measure the impact of each treatment, we will observe loan demand, loan repayment times, and late fees incurred by analyzing telco-generated administrative data for the full sample. We will also conduct a survey on a subsample which will provide information on consumer knowledge and purpose of the loans taken.
Experimental Design Details
Not available
Randomization Method
Randomization by computer, in Stata
Randomization Unit
Individual phone number
Was the treatment clustered?
No
Experiment Characteristics
Sample size: planned number of clusters
NA
Sample size: planned number of observations
A minimum of 50,810 subscribers total (assigned equally to 4 treatment groups); 6000 for the survey (In the Phase 1 Pilot Launch of digital loans by our partner telco, the digital loan will be available to 50,810 subscribing phone numbers. After Phase 1 trial launch, the product is due to scale up to many more. If all goes to plan, then this RCT will also scale up accordingly.)
Sample size (or number of clusters) by treatment arms
A minimum of 12,775 subscribers in each of the treatment arms
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
Malawi National Commission for Science and Technology
IRB Approval Date
2018-06-05
IRB Approval Number
P.02/18/250
IRB Name
University of California, Santa Cruz
IRB Approval Date
2018-01-30
IRB Approval Number
2959