Spending and debt responses to increased savings

Last registered on June 07, 2018

View Trial History

Pre-Trial

Trial Information

General Information

Title

Spending and debt responses to increased savings

RCT ID

AEARCTR-0003047

Initial registration date

June 06, 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

June 07, 2018, 1:16 PM EDT

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

Locations

Country

Mexico

Region

Mexico City

Primary Investigator

Name

Michaela Pagel

Affiliation

Columbia Business School

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Paolina Medina

PI Affiliation

Texas A&M University

Contact Investigator

Additional Trial Information

Status

Completed

Start date

2017-11-01

End date

2018-01-31

Keywords

Finance & Microfinance

Additional Keywords

JEL code(s)

Secondary IDs

Abstract

This project will study the changes in transactional volume of credit and debit cards when individuals increase their saving balances in a checking account. Specifically, we plan to study what are the spending categories that experience the larger decreases or increases in transactional volume, after increases in saving balances in checking accounts. We will also test whether individuals borrow more or not because they increase the balances in their checking accounts.

External Link(s)

Registration Citation

Citation

Medina, Paolina and Michaela Pagel. 2018. "Spending and debt responses to increased savings." AEA RCT Registry. June 07. https://doi.org/10.1257/rct.3047-1.0

Former Citation

Medina, Paolina and Michaela Pagel. 2018. "Spending and debt responses to increased savings." AEA RCT Registry. June 07. https://www.socialscienceregistry.org/trials/3047/history/30438

Sponsors & Partners

Experimental Details

Interventions

Intervention(s)

The period of analysis will cover a time during which the partner company performed a program evaluation of different methods to encourage their clients to increase their saving balances in checking accounts. The program evaluation consisted of sending text messages encouraging savings to a randomly selected set of clients. The messages were sent either via cell phone or on ATM screens. This type of program evaluation is part of the usual operations of the company, as they routinely communicate marketing campaigns to their users through their cell phones, or via ATM screens, and they routinely test the effectiveness of different versions of these messages with randomly selected groups of clients.
The analysis performed by the partner institution concluded that the messages increased the average checking account balance of the group of clients that received the messages. This evaluation took place in the end of 2017, before researchers and the partner company had any interaction. The researchers were not involved in any form in the planning, implementation, or analysis of the evaluation.

Intervention Start Date

2017-11-01

Intervention End Date

2018-01-31

Primary Outcomes

Primary Outcomes (end points)

Credit card and debit card transactions and balances

Primary Outcomes (explanation)

The data used in the analysis will contain all credit card and debit card transactions performed by the clients of the partner company during the period of analysis. Each transaction will contain a transaction category, a string description of the transaction, the date of the transaction, as well as the value of the transaction. The dataset will also contain balances on different types of credits, as well as binary variables indicating if a given user received the message encouraging savings or not.
• For each transaction: String description of the transaction, date, and value of the transaction, masked account identifies, and masked client identifier.
• For each account: Balances at the end of each month or billing cycle, interest paid or received at the end of each month or billing cycle, product characteristics (interest rate and maturity of the loan), masked account identifier, and masked client identifier.
• For each client: Latest credit score, indicator variable indicating if user received message, gender, age, estimated income, age of credit history, masked client identifier.

Secondary Outcomes

Secondary Outcomes (end points)

Secondary Outcomes (explanation)

Experimental Design

Experimental Design Details

Randomization Method

By computer.

Randomization Unit

Client id

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

One treatment with control on 129,949 clients

Sample size: planned number of observations

The control group is 19,984 clients and the treatment group is 153,280 clients. Control group could be increased using existing observational data of other non-treated clients.

Sample size (or number of clusters) by treatment arms

One treatment 173,264 clients in total, 153,280 treated. Control group could be increased using existing observational data of other non-treated clients.

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

We consider the deviation from normalized monthly spending, savings, and balances, which have (upper bound) an across-clients standard deviation of 70%, then the sample size of 153,280 treated clients, will enable us to detect a 0.5 percent deviation, with 79.9% statistical power (rule-of-thumb value is 80%) and a 5% alpha error.

Supporting Documents and Materials

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

Request Information

IRB