Can Incentives Create a Habit of Saving?

Last registered on January 24, 2014
View Trial History

Pre-Trial

Trial Information

General Information

Title
Can Incentives Create a Habit of Saving?
RCT ID
AEARCTR-0000068
First published
January 24, 2014, 3:47 AM EST

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

Locations

Country
Ghana
Region
Northern Region

Primary Investigator

Name
Gharad Bryan
Affiliation
London School of Economics
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Margaret McConnell
PI Affiliation
Harvard
Contact Investigator

Additional Trial Information

Status
In development
Start date
2013-11-01
End date
2015-01-01
Keywords
Finance & Microfinance
Additional Keywords
Habit Formation, Microsavings, Incentives
JEL code(s)
Secondary IDs
Abstract
This research is designed to understand whether monetary incentives can be used to create a habit of savings. The study population consists of 600 market women in Tamale, Northern Ghana. Several recent studies suggest that women similar to those in the study population are willing and able to save, but that they face challenges in reaching their savings goals. Studies have also shown surprisingly large impacts from programs that facilitate savings. The psychological literature suggests that habits guide the behaviour of many people, and that there is potential for habit formation to help people to reach goals that they find difficult, for example, because of self control problems. While there are many ways to think about forming habits, most are information and time intensive. Providing incentives for the repetitive performance of particular behaviours is among the cheapest and least information intensive options. This work will explore three main issues: first, can an experiment be designed that can discriminate between the two key possibility -- that incentives are sufficient to create a habit, or that incentives alone cannot create a habit; second, are incentive complementary to other potential avenues for promoting habits; third, what is the role of consistency and repetition in habit formation; and fourth, what population characteristics determine whether or not it is possible to form a habit? The study will also consider the welfare impacts of habit formation, if indeed it is possible to form a habit in the study population.
External Link(s)

Registration Citation

Citation
Bryan, Gharad and Margaret McConnell. 2014. "Can Incentives Create a Habit of Saving? ." AEA RCT Registry. January 24. https://doi.org/10.1257/rct.68-1.0
Former Citation
Bryan, Gharad and Margaret McConnell. 2014. "Can Incentives Create a Habit of Saving? ." AEA RCT Registry. January 24. https://www.socialscienceregistry.org/trials/68/history/956
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)
Subjects in the treatment group will be paid 1 Ghana Cedi for every 1 Ghana Cedi increase in their bank balance relative to the previous week. This is to the maximum of 2 Cedis. The deposit must be placed into a Tigo mobile money account. Some subjects will also be given advice on how to go about forming a habit.
Intervention Start Date
2013-11-14
Intervention End Date
2014-05-01

Primary Outcomes

Primary Outcomes (end points)
Savings levels, stress levels, response to stress stimulus, self reported wellbeing, consumption, health outcomes.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
The primary arm of the experiment splits the sample (600 market women from Tamale) into two equally sized groups. Half the sample will be the treatment group, members of which will be provided with a savings incentive; the size of which will vary on a randomized basis. The length of incentive provision will be determined by an endogenous measure of habit formation, which has been shown in past research to reach a maximum. Incentives will be ceased when a pre-specified portion of the group has reached this maximum. The remaining half will be control, and will receive no incentive.

There will also be additional cross-randomizations. First, because we are concerned that the habit measurement amounts to a treatment, 75% of the control and 25% of the treatment will not receive this measurement. Second, to validate the habit measurement, 33% of the treatment group will receive incentives for an additional 4 weeks, with treatment ceasing at the same time as teh main treatment group. Theory suggests that these group should experience no additional propensity to form a habit. Third, to assess the role of consistency in creating a habit, 33% of the treatment group will be subject to four weeks in which incentives will not be paid (although the total time period for which incentives are paid will remain the same). These four weeks will be randomly allocated through the term of the incentives. Finally, one quarter of all subjects receiving incentives will also be given some advice on how to effectively meet the savings requirements to receive the incentive. This advice will be based on a psychologically informed model of habit formation, which highlights that a savings habit requires the constant repetition of several sub-behaviors.

Data will be collected weekly during the habit formation phase. Baseline data will be collected at the time of the cessation of incentive payment, as well as 6 months post treatment, and at additional dates if effects are seen at 6 months. Savings activity in the Tigo account will be monitored on an ongoing basis.
Experimental Design Details
Randomization Method
In office and done by a computer.
Randomization Unit
The individual.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
600 market women.
Sample size: planned number of observations
600
Sample size (or number of clusters) by treatment arms
300 control, 300 treatment.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
Institutional Review Board of the Harvard School of Public Health
IRB Approval Date
2013-12-12
IRB Approval Number
IRB13-0014
Analysis Plan

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