Preparedness for old age and the demand for contributory pensions in Uganda

Last registered on October 16, 2026
View Trial History

Pre-Trial

Trial Information

General Information

Title
Preparedness for old age and the demand for contributory pensions in Uganda
RCT ID
AEARCTR-0019899
Initial registration date
October 08, 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
October 16, 2026, 2:08 PM EDT

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

Locations

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

Request Information

Primary Investigator

Name
Alison Andrew
Affiliation
Contact Primary Investigator

Other Primary Investigator(s)

PI Name
Selim Gulesci
PI Affiliation
Trinity College Dublin
Contact Investigator

Additional Trial Information

Status
On going
Start date
2026-09-15
End date
2026-11-30
Keywords
Finance & Microfinance, Welfare
Additional Keywords
Pensions; Social Protection
JEL code(s)
Secondary IDs
Prior work
This trial does not extend or rely on any prior RCTs.
Abstract
We conduct a survey of 560 working-age individuals in Uganda to assess how individuals prepare for old age. Alongside collecting rich descriptive data on individuals' preparation behavior, attitudes and expectations, we use a series of hypothetical experiments to gauge stated demand for, and contributions to, a new contributory pension, similar to the scheme the government is planning to introduce. We vary the co-contribution rate and the degree of liquidity to assess how these design choices might affect uptake. We embed an information experiment on the share of elderly individuals living alone or only with dependents, to assess the impact of information on the current situation of older people in Uganda on demand. Finally, we examine which demographic characteristics (e.g. not having living children), economic characteristics (e.g. having some disposable income), expectations of old age and attitudes towards old age predict demand.
External Link(s)

Registration Citation

Citation
Andrew, Alison and Selim Gulesci. 2026. "Preparedness for old age and the demand for contributory pensions in Uganda." AEA RCT Registry. October 16. https://doi.org/10.1257/rct.19899-1.0
Sponsors & Partners
Experimental Details

Interventions

Intervention(s)
Most of the study consists of collecting novel survey data on how working-age individuals in Uganda prepare for old age. The experimental component has two parts. First, we vary, within respondent, the characteristics of a hypothetical contributory pension (the co-contribution rate and the liquidity of savings) and measure stated uptake. Second, we give respondents accurate information from the 2024 Ugandan census on the share of elderly individuals living alone or only with dependents, to assess how this information affects demand.
Intervention Start Date
2026-09-15
Intervention End Date
2026-11-30

Primary Outcomes

Primary Outcomes (end points)
Whether the respondent would open a pension account (binary) and the monthly amount they would contribute (UGX), in each of four pension design scenarios (no government match and no early withdrawal; a 25% government match and no early withdrawal; no government match and early withdrawal allowed; a 25% government match and early withdrawal allowed).
Primary Outcomes (explanation)
Demand is measured in a tablet-based app that presents four scenarios for a new voluntary contributory pension, crossing a government matching contribution of 0% or 25% with whether early withdrawal before age 55 is allowed. In each scenario, the respondent sets a monthly contribution (in UGX, or zero) using a slider. The binary measure equals one if the chosen contribution is positive. The continuous measure is the chosen monthly contribution in UGX.

Secondary Outcomes

Secondary Outcomes (end points)
Potential secondary outcomes that may be affected by the information treatment are: (a) attitudes to old age; (b) subjective expectations of old age; (c) psychological well-being. If we have sufficient variation, we will also assess heterogeneity of treatment effects with respect to respondent’s gender, age, marital status, number and gender composition of children, socio-economic status, ethnic background (in particular presence of age-set practices).
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
Inclusion criteria: 25-50 years old, member of a Village Savings and Loan Association.

The study involves two randomizations. First, each respondent sees all four versions of the pension design (a 0% or 25% government matching contribution, crossed with whether early withdrawal before age 55 is allowed), and we compare demand across versions within respondent. The order in which the four versions are shown is randomized. Second, for a random half of the sample, we reveal the actual share of older men vs. women in Uganda who live alone or only with dependents, taken from the 2024 Ugandan census, before the pension demand exercise. The other half receive this information after the exercise.
Experimental Design Details
Not available
Randomization Method
Random number generated internally by SurveyCTO.
Randomization Unit
Individual, for both experiments. In the pension design experiment, every respondent sees all four scenarios, and the order in which they are shown is randomized at the individual level. In the information experiment, respondents are randomized individually to receive the census information either before the pension demand exercise (half the sample) or after it (the other half).
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
560 individuals
Sample size: planned number of observations
560 individuals
Sample size (or number of clusters) by treatment arms
Experiment 1: 560 individuals (within-person comparison across the four scenarios). Experiment 2: 280 individuals in each arm (information before the pension demand exercise; information after).
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Experiment 1 (within-person comparison of scenarios, N = 560): the MDE depends on the within-person correlation in demand across scenarios (rho), on which we have no strong prior. With 80% power and 5% significance, the MDE is 0.17 SD if rho = 0, 0.12 SD if rho = 0.5 and 0.09 SD if rho = 0.7. Experiment 2 (280 individuals per arm): the MDE is 0.23 SD with 80% power and 5% significance.
Supporting Documents and Materials
IRB

Institutional Review Boards (IRBs)

IRB Name
Economics (Econ) DREC
IRB Approval Date
2026-01-12
IRB Approval Number
DREC - 2661470
IRB Name
UNCST
IRB Approval Date
2026-09-14
IRB Approval Number
SS5800ES
Analysis Plan

Analysis Plan Documents