NEW UPDATE: Completed trials may now upload and register supplementary documents (e.g. null results reports, populated pre-analysis plans, or post-trial results reports) in the Post Trial section under Reports, Papers, & Other Materials.
Private vs. Public Mental Accounts: Experimental Evidence from Savings Groups in Colombia
Last registered on September 25, 2016


Trial Information
General Information
Private vs. Public Mental Accounts: Experimental Evidence from Savings Groups in Colombia
Initial registration date
January 08, 2014
Last updated
September 25, 2016 12:43 AM EDT
Primary Investigator
City University of New York
Other Primary Investigator(s)
Additional Trial Information
Start date
End date
Secondary IDs
I designed and implemented a Randomized Controlled Trial to study whether simple modifications to the framing and labeling of a commitment savings product affected savings accumulations and other outcomes of low-income individuals in newly-formed Village Savings and Loan Associations in Colombia. Motivated by insights from behavioral economics, the study tests if behavioral responses vary depending on whether subjects are led to label and create ‘mental savings accounts’ in private or public ways. Individuals in the private-labeling treatment groups were led to label their savings as earmarked for a particular purpose and to state savings accumulation targets, information which was shared only privately with a member of the research team. Individuals in the public-labeling treatment groups received the same intervention but were then asked to reveal publicly and announce their chosen goals to other members of their savings group. The average treatment effects of the public-labeling intervention are very strong and significant. Savings accumulations increased by an average of 35% and savings goals were 8.5% more likely to be reached in comparison to the untreated group. Further results strongly suggest evidence of differentiated behavioral responses of individuals in the private-labeling treatment group: private commitment to a savings goal is more effective for individuals who, after random assignment but prior to the intervention, had been measured to be less constrained by economic circumstances and institutional barriers.
External Link(s)
Registration Citation
Salas, Luz. 2016. "Private vs. Public Mental Accounts: Experimental Evidence from Savings Groups in Colombia." AEA RCT Registry. September 25. https://doi.org/10.1257/rct.210-3.0.
Former Citation
Salas, Luz. 2016. "Private vs. Public Mental Accounts: Experimental Evidence from Savings Groups in Colombia." AEA RCT Registry. September 25. http://www.socialscienceregistry.org/trials/210/history/10753.
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
The RCT randomly assigned 137 newly formed Village Savings and Loan Associations (VSLA), mainly in rural areas from nine municipalities of Colombia into two treatments and a control group. Individuals in the control group were exposed to the standard VSLA model. In the private-labeling treatment members received an additional module with a short organized guided conversation aimed at discussing and highlighting the difficulties of committing to a savings path and the potential role and use of mental accounts in strengthening those commitments. I asked participants to voluntarily state in writing a savings purpose and weekly savings goals. This was intended to help guide individuals to form and label their mental ‘savings’ account to privately commit to achieving that savings purpose. Goals were not stated publicly. The public-labeling treatment was similar except that the group discussion encouraged members to make commitments to themselves as well as to others in their group in order to explore the possibility that this might lead to different outcomes. As in the private-labeling treatment members were asked to voluntarily state a savings purpose and weekly savings goals in writing, but in this intervention, those commitments were shared with all members of the group. By doing this, individuals explicitly label their ‘mental savings account’ and publicly commit to their own decision. Individuals were not informed about the (public) nature of the treatment prior to the public announcement. This module included a trust building game and a guided conversation aimed at highlighting the achievements of group commitments.
Intervention Start Date
Intervention End Date
Primary Outcomes
Primary Outcomes (end points)
Savings accumulations (number of shares purchased and savings balance in first savings cycle) and likelihood of goals achieved in first savings cycle
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
I randomly assigned newly formed VSLA to two treatments and one control group in 9 municipalities of Colombia. The assignment of the VSLA was carried out using a simple lottery and the method is called spot-randomization in which the assignment to treatment was random at the time when the VSLA was formed. For example, if 5 new VSLA were formed in a week in Cartagena, I draw the type of intervention that the VSLA would be assigned to (private, public or control). This status is maintained throughout all the experimental period. The unit of randomization is the savings group (or VSLA) and the unit of analysis is at the individual level.
Experimental Design Details
Randomization Method
Simple lottery, stratification at the municipality level, spot randomization done in office by a computer.
Randomization Unit
Unit of randomization: savings group or VSLA.
Was the treatment clustered?
Experiment Characteristics
Sample size: planned number of clusters
137 savings groups or VSLA with an average of 13 individuals by group
Sample size: planned number of observations
1700 individuals
Sample size (or number of clusters) by treatment arms
45 VSLAs control, 45 VSLAs to private-labeling treatment and 45 VSLAs to public-labeling treatment
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
MDE of 0.2 s.d. for private-labeling treatment and of 0.3 s.d. for public-labeling treatment
IRB Name
Graduate Center, City University of New York
IRB Approval Date
IRB Approval Number
Post Trial Information
Study Withdrawal
Is the intervention completed?
Intervention Completion Date
November 30, 2012, 12:00 AM +00:00
Is data collection complete?
Data Collection Completion Date
December 15, 2012, 12:00 AM +00:00
Final Sample Size: Number of Clusters (Unit of Randomization)
137 VSLAs
Was attrition correlated with treatment status?
Final Sample Size: Total Number of Observations
1,663 individuals
Final Sample Size (or Number of Clusters) by Treatment Arms
45 VSLAs in control, 45 VSLAs in private-labeling treatment, 47 VSLAs in public-labeling treatment
Data Publication
Data Publication
Is public data available?

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

Request Information
Program Files
Program Files
Reports, Papers & Other Materials
Relevant Paper(s)