Gambling for retirement

Last registered on July 23, 2026
View Trial History

Pre-Trial

Trial Information

General Information

Title
Gambling for retirement
RCT ID
AEARCTR-0019175
Initial registration date
July 17, 2026

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
July 23, 2026, 8:03 AM EDT

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

Locations

Country
Colombia
Region

Primary Investigator

Name
Egon Tripodi
Affiliation
Hertie School
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Matthias Rodemeier
PI Affiliation
Bocconi University
Contact Investigator
PI Name
Jared Gars
PI Affiliation
University of Florida
Contact Investigator
PI Name
Justin Holz
PI Affiliation
University of Michigan
Contact Investigator
PI Name
Miguel Villa
PI Affiliation
Inter-American Development Bank
Contact Investigator

Additional Trial Information

Status
Completed
Start date
2019-04-11
End date
2019-12-31
Keywords
Behavior, Finance & Microfinance, Governance, Labor
Additional Keywords
Retirement Savings, Prize-Linked Savings, Inattention, Informality, Household Finance
JEL code(s)
D14, D91, H55, C93, G52
Secondary IDs
Prior work
This trial is based on or builds upon one or more prior RCTs.
Abstract
This study evaluates the effectiveness of savings lotteries in increasing retirement savings among informal workers in Colombia. Participants were randomly assigned to receive lottery offers tied to savings deposits, including small cash prizes, large housing prizes, both types of prizes, or no lottery offers. Using administrative data from the public pension system, the study examines the effects of these lottery incentives on savings behavior, the timing of deposits, eligibility for complementary program benefits such as life and disability insurance, and the persistence of any behavioral changes after the intervention.
External Link(s)

Registration Citation

Citation
Gars, Jared et al. 2026. "Gambling for retirement." AEA RCT Registry. July 23. https://doi.org/10.1257/rct.19175-1.0
Sponsors & Partners

Partner

Name
Colpensiones
Type
government
URL
https://www.colpensiones.gov.co
Experimental Details

Interventions

Intervention(s)
The intervention used text messages to inform account holders in Colombia’s voluntary public retirement-savings program, Beneficios Económicos Periódicos (BEPS), about savings lotteries. Depending on treatment assignment, account holders were offered opportunities to win small cash prizes, a housing prize, or both, conditional on reaching specified savings thresholds. The control group received standard savings reminders without information about lottery prizes.
Intervention (Hidden)
The intervention was implemented by Colpensiones beginning in April 2019. All experimental groups received savings-reminder text messages on the same biweekly schedule. The messages differed only in whether they advertised lottery incentives.

Account holders were assigned to one of four arms:

1. Control: General reminders to save in BEPS, without mention of a lottery.
2. Cash lottery: Eighteen cash-prize drawings were held approximately every two weeks between April and December 2019. Each drawing awarded 200 prizes of COP 100,000. Account holders earned one lottery ticket for every COP 55,000 deposited during the relevant two-week qualification period, with no limit on the number of tickets.
3. House lottery: Two housing-prize drawings were held, one for each six-month qualification period. Each drawing awarded a housing voucher worth up to COP 102,000,000. Account holders earned one ticket for every COP 500,000 accumulated during the relevant qualification period, with no limit on the number of tickets.
4. House and cash lotteries: Account holders received offers for both the cash and housing lotteries under the same qualification rules.

Colpensiones randomized the text-message information rather than formal eligibility for the drawings. In principle, all account holders who met a lottery’s savings requirements could enter the drawing. However, the randomized text messages were the only channel through which Colpensiones informed account holders about the lotteries during the experiment. A random subset of accounts in the Cash and House-and-Cash groups received messages only through June 2019, while the remaining accounts received the full intervention through the end of 2019.
Intervention Start Date
2019-04-11
Intervention End Date
2019-12-31

Primary Outcomes

Primary Outcomes (end points)
1. Savings-exactly-at and savings-at-and-above the lottery qualification thresholds during the intervention.
2. Biweekly retirement-account savings during the intervention period.
3. Eligibility for the BEPS life and disability insurance benefit.
Primary Outcomes (explanation)

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
The study is a nationwide natural field experiment conducted among account holders in Colombia’s BEPS voluntary retirement-savings program. Individual accounts were randomly assigned to receive standard savings reminders or reminders advertising one of three lottery-incentive arrangements: cash prizes, housing prizes, or both cash and housing prizes.

Lottery eligibility was tied to reaching specified savings thresholds over either two-week or six-month qualification periods. Administrative account records are used to measure deposits, total savings, the timing and frequency of transactions, and eligibility for associated social-insurance benefits.
Experimental Design Details
The experimental sample consisted of 387,449 BEPS account holders registered by the end of 2018. Accounts were assigned in approximately equal proportions to Control, House, House-and-Cash, and Cash arms.

Beginning on April 11, 2019, all four arms received text-message reminders on the same biweekly schedule. Control messages encouraged retirement saving without mentioning prizes. Treatment messages described the savings thresholds and prizes corresponding to the assigned lottery arm.

The Cash intervention comprised 18 two-week qualification periods. Account holders received one ticket for every COP 55,000 deposited within a qualification period. Each drawing offered 200 prizes of COP 100,000.

The House intervention comprised two six-month qualification periods. Account holders received one ticket for every COP 500,000 accumulated during the relevant period. Each drawing offered one housing voucher worth up to COP 102,000,000.

The House-and-Cash arm combined both incentive structures. There was no cap on the number of tickets an account holder could earn.

A randomly selected subset of the Cash and House-and-Cash arms received messages only during the first three months of the intervention, while the remaining accounts received messages through the end of 2019. The main analysis of the separate lottery treatments focuses on the full-dose groups where relevant.
Randomization Method
Randomization conducted by computer using Colpensiones administrative account records.
Randomization Unit
Individual BEPS retirement-savings account. Each account corresponds to one account holder.
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
387,449 individual accounts.

Because treatment was assigned individually, each account is its own randomization unit; there are no multi-account treatment clusters.
Sample size: planned number of observations
387,449 BEPS account holders.
Sample size (or number of clusters) by treatment arms
* Control: 96,862 accounts
* House lottery: 96,862 accounts
* House-and-Cash lotteries: 96,863 accounts
* Cash lottery: 96,862 accounts
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
No formal ex ante power calculation is available. The experiment included the near-universe of 387,449 eligible BEPS accounts and was therefore determined by the size of the administrative study population rather than by a target minimum detectable effect.
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
University of Essex
IRB Approval Date
2022-05-03
IRB Approval Number
ETH2122-1112
IRB Name
University of Michigan IRB
IRB Approval Date
2023-12-04
IRB Approval Number
HUM00246681
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