SBS Credit Score

Last registered on February 27, 2018
View Trial History

Pre-Trial

Trial Information

General Information

Title
SBS Credit Score
RCT ID
AEARCTR-0002354
Initial registration date
February 26, 2018

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
February 27, 2018, 11:03 AM EST

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Country
Peru
Region
Latin America

Primary Investigator

Name
Jonathan Zinman
Affiliation
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Xavi Gine
PI Affiliation
World Bank
Contact Investigator
PI Name
Kareem Haggag
PI Affiliation
Carnegie Mellon University
Contact Investigator
PI Name
Dean Karlan
PI Affiliation
Northwestern University
Contact Investigator

Additional Trial Information

Status
On going
Start date
2017-06-07
End date
2018-12-31
Keywords
Finance & Microfinance
Additional Keywords
credit score, Peru
JEL code(s)
Secondary IDs
Abstract
Credit users are often not conscious of the status of their credit history, this could lead them to take sub-optimal decisions (e.g. not being able to access a credit when needed). In this sense, to ameliorate the financial decisions of peruvians, the Superintendencia de Banca y Seguros del Perú (SBS) has created an online portal and a mobile app that lets consumers obtain their credit history for the last 5 years (including a credit score for each debt). The main objective of this portal is to provide consumers with information about their debts so they could take action and improve their credit situation. However, although 1.3 million Peruvians have already logged in into the portal, only 21% of them have logged in more than one time. This might suggest that a vast majority of credit users are not utilizing credit reports to ameliorate their credit situation.The main goal of this study is to evaluate an SMS Campaign that will aim at increasing usage of the portal and ameliorating the credit situation of users. The content of the text messages will be based on the behavioral economics literature.
External Link(s)

Registration Citation

Citation
Gine, Xavi et al. 2018. "SBS Credit Score." AEA RCT Registry. February 27. https://doi.org/10.1257/rct.2354-1.0
Former Citation
Gine, Xavi et al. 2018. "SBS Credit Score." AEA RCT Registry. February 27. https://www.socialscienceregistry.org/trials/2354/history/26131
Sponsors & Partners

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

Request Information
Experimental Details

Interventions

Intervention(s)
We will send debt users a message to remind them to revise their credit information (total debt, credit score, among others). This message will be sent via SMS.
Intervention Start Date
2017-09-22
Intervention End Date
2017-12-01

Primary Outcomes

Primary Outcomes (end points)
A binary indicator for whether the individual logged in to the portal between September 29, 2017 and December 31, 2017. [Primary outcome measure]
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
2. The total number of times a user logs into the portal over that time period (September 29, 2017 and December 31, 2017)
3. The number of days between September 29 and the first date the user checks the portal.
4. Credit Score (0 to 4)
5. Total amount of debt
6. Total available credit
7. Number of lines of credit
Secondary Outcomes (explanation)
For point 5 and 6, a functional form will be used.
To decide any functional form (e.g. log vs. level) adjustments to our dependent variables, we will use the following approach: We construct a fake (binary) random treatment variable and run the analysis using both the unadjusted and adjusted dependent variables regressed on this fake treatment. Whichever approach delivers the highest precision (i.e. the standard error of the treatment effect divided by the control group standard deviation)
will be taken as our primary outcome variable for the analysis with the true treatment variable. For outliers, will examine the data without the true treatment variable, and make any trimming or winsorizing decisions based a subjective assessment of whether outliers are consistently high across multiple similar measures.

Experimental Design

Experimental Design
We have nearly 52,000 peruvian phone numbers. We will randomly assign these phone numbers to one of six conditions: Control, T1, T2, T3, T4 or T5. In the five treatment conditions, subjects will receive a text message with the following contents:

T1: "Get your free SBS credit score here [link] "
T2: “Get your free SBS credit score here [link]. Did u know that [n] out of 10 people have the lowest score?"
T3: “Get your free SBS credit score here [link]. Did u know that [n] out of 10 people have the highest score?”
T4: "Get your free SBS credit score here. Improve it today by paying off any late debts."
T5: "Get your free SBS credit score here. Do you know your score?

Independent from the randomization of T1-T5, half of the treated sample will receive an additional message stating the following: "We have not checked your score". We will send five text messages in total during a period of three months. The first out of the five messages of this campaign will inform each of the subjects about the treatments. Specifically, all of the individuals in the treatment conditions will receive a first message stating the following:

"The Superintendency of Banking, Insurance and AFP (SBS) starts its campaign to promote the access of its credit report portal. Wait for our messages with the signature @SBS_PERU."

The following four messages will utilize the contents from T1-T5. At the end of this period, we will assess if there were any significant differences between the control and treatment groups in the outcome measures.
Experimental Design Details
Randomization Method
Randomization will be done at our offices using STATA. We will make sure that the randomization process is replicable.
Randomization Unit
The unit randomization is the individual.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
0
Sample size: planned number of observations
51,947 individuals
Sample size (or number of clusters) by treatment arms
Control: 25,839
T1: 5,153
T2: 5,246
T3: 5,238
T4: 5,189
T5: 5,282
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
Innovations for Poverty Action IRB
IRB Approval Date
2017-09-05
IRB Approval Number
14392
Analysis Plan

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

Request Information

Post-Trial

Post Trial Information

Study Withdrawal

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

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials