Relaxing Borrowing Constraints in Savings Groups: Evidence from Uganda
Last registered on May 01, 2019


Trial Information
General Information
Relaxing Borrowing Constraints in Savings Groups: Evidence from Uganda
Initial registration date
December 12, 2018
Last updated
May 01, 2019 10:05 PM EDT

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

Request Information
Primary Investigator
University Of Oregon
Other Primary Investigator(s)
PI Affiliation
University of Maryland
Additional Trial Information
On going
Start date
End date
Secondary IDs
The study randomly assigns VSLAs to two treatments: one providing a group-level savings account from a local bank, and the second providing the same savings account plus the possibility of getting a group level loan from the bank. The study is intended to provide groups with access to a safe place to store money, in addition to relaxing possible constraints that limit the ability of groups to lend to members. At the time of writing of this PAP, researchers had implemented the intervention and carried out one round of data collection (midline). An endline will be concluded in 2019. The data have yet to be analyzed.
External Link(s)
Registration Citation
Burlando, Alfredo and Jessica Goldberg. 2019. "Relaxing Borrowing Constraints in Savings Groups: Evidence from Uganda." AEA RCT Registry. May 01.
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
Savings groups (SGs), including VSLAs, generally self-fund all their activities, including the provision of loans to one another through the accumulation of group savings. Successful groups are capable of meeting all internal capital needs, allowing members to satisfy the credit needs of their families and their income generating activities. In practice, however, many VSLAs are formed by rural poor with limited access to capital, and they often face significant financial barriers, with members unable to accumulate enough savings to meet the financial needs of their members (Burlando et al, 2018). In addition, groups also face the possibility of theft from their savings, which itself could create distortionary behavior; for instance, it could induce people to reduce savings, or to over-lend at the end of the cycle to reduce the amount of cash in their savings box. In this context, linkage programs with external financial providers that provide loans to VSLAs could alleviate these financial barriers.

In this study, we propose to carry out an evaluation of a randomized control trial implemented in 2015-2016 to measure the impact of introducing formal banking products to members of savings groups in Uganda. The main question we are interested in addressing is whether savings groups participants benefit from an enhanced access to external (bank) credit. The study partnered a research team with a financial services provider (FSP), two local implementing partners with ties with savings group, and the Stromme Foundation. OBUL developed two financial products specifically targeted to savings groups: a group saving account, and a group loan. In our study, we randomly assigned savings groups to three interventions: a loan intervention, in which groups receive access to loans and group savings; a savings intervention, in which groups receive access to group savings only; and a control. This PAP covers the analysis of midline data; this analysis will inform our plans for an endline, and we plan to create a new PAP that covers the endline.

Access to external credit also affects incentives to join and remain in savings groups. The groups may become more attractive to households seeking larger loans, or less attractive to more risk averse members. We will monitor the composition of groups across cycles.

The inclusion of a savings intervention serves two purposes. First, it allows us to understand the importance of providing savings groups with a safe place to store excess funds. If concerns over the safety of savings leads to over-lending at the end of each cycle (which is something some group members have hinted at to us in the past), or if money saved in the bank is not easily accessible to members, we expect bank lending to be lower in this treatment. Second, the presence of the savings intervention allows us to cleanly separate the effect of credit from the effect of bank savings. That’s because OBUL loan products require groups to maintain a savings account, which makes it impossible to evaluate a lending program without a savings component.

The group saving account allows savings groups to keep part or all of their balance deposited in a mobile-money linked OBUL saving account. The loan product involves a financial transfer to the group saving account. Loans carry a 2.5\% monthly interest rate, and repayments occur monthly according to a payment schedule. Repayment is initially set at 3 months. The bank also charges a search fee‚ and application fee to each loan; these fees are rolled into the payment schedule. Finally, a one-time ‚ financial card application fee‚ is applied to groups for the purpose of obtaining Bank of Uganda permission to receive credit.
Intervention Start Date
Intervention End Date
Primary Outcomes
Primary Outcomes (end points)
Total savings
Total borrowing
Investments in agriculture and microenterprise
Household income
Primary Outcomes (explanation)
See Pre-analysis plan for details.
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
The study will have three treatment arms:
• An intervention (treatment) group, which is offered (1) a bank loan (2) access to a group savings account provided by the bank, and (3) training on the use and management of bank loans and savings accounts;
• A savings only group, which is offered (1) access to a group savings account provided by the bank and (2) training on the use of the savings account;
• A “monitoring only” or “comparison” group, which is not given any contact to the bank.

The financial products
The financial product is described in an appendix. This description has been approved by the relevant Ugandan financial authorities.

We study 2 products. The first is a group savings account. The bank will open an account for the treatment and savings only groups (the bank will have no interactions with the no-intervention group). The group can use the account to deposit their savings (a good way to protect against theft) The account belongs to the entire group, and it is administered in the following way: There are 3 account administrators, who are elected by the group. Each administrator is given a PIN number. To make any transaction in the account, all three PINs must be used.

The second product is the loan product. The loan product is a 3 month loan given to the entire group. The size of the loan is decided by the group, within the limits imposed by the bank. The loan is then deposited in the savings account. This is then divided up among members who want to borrow. A bank official visits the groups on a monthly basis to ensure partial repayment of the loan. (This is very consistent with standard practice within the group, where internal loans are repaid once a month until the balance is paid in full after 3 months). The loan carries a monthly interest of 2.75%, which is less than the 3-5% monthly interest rate charged by internal group loans. However, there are also a couple of financial fees charged to the group.

Conditional on the repayment of the loan, the bank will be able to extend a second, 3-month loan to the group. This will be repaid before the end of the cycle. In future years, the groups will be able to negotiate loans with longer duration (4 to 6 months), which will cut down on the service costs for these groups.
Experimental Design Details
Not available
Randomization Method
Randomization done in the office by a computer.
Randomization Unit
The level of treatment is at the savings group level; the randomization was done at the village level.
Was the treatment clustered?
Experiment Characteristics
Sample size: planned number of clusters
There are 156 savings groups that were entered in the study.
Sample size: planned number of observations
The IRB has approved a protocol for up to 4,620 participants of savings groups.
Sample size (or number of clusters) by treatment arms
61 Loan + Savings groups
28 Savings only groups
65 Control
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB Name
IRB Approval Date
IRB Approval Number
Analysis Plan
Analysis Plan Documents

MD5: c0865ccebbcca8645b83d0e568a0ab02

SHA1: ea5da60d7c298a3d9e4b6a45d960db62f66487ac

Uploaded At: November 27, 2018

Revised PAP_Uganda_Burlando_Goldberg_April2019

MD5: 1805c6817bc54592cae351a64c1f5945

SHA1: 66370357b57cb1ffa32cfbe45ddbd99dd4bc69f5

Uploaded At: May 01, 2019