evaluation

Applying behavioural economics to improve microsavings outcomes

Evidence type: Evaluation i

Description of the programme
The study
What are the outcomes?
Key findings
Points to consider
Full report

Description of the programme

Many programmes, policies, and products designed to increase formal savings among poor households have had only limited large-scale impact. This study is an in-depth investigation of savings behaviour among clients at CARD Bank in the Philippines, where only 26 per cent of adults use formal financial services and almost 80 per cent do not have a deposit savings account. Almost 40 per cent of households report not having any cash on hand for emergencies and unexpected expenses.

The work was completed as part of Grameen Foundation’s Microsavings Initiative (GMFI)—a multi-year project to enhance access to safe, flexible, and convenient deposit accounts for the poor.

The study

This is an evaluation of a microsavings pilot project, conducted by ideas42 and Grameen Foundation in partnership with CARD Bank in the Philippines, from November 2012 to October 2013. The study used a combination of qualitative and quantitative research methods to uncover the causes of low savings balances and infrequent transactions, including a randomised control trial (RCT). It also sought to understand the psychological tendencies influencing low savings behaviour as well as specific contextual details that trigger or exacerbate the psychologies.

What are the outcomes?

Financial behaviour: opening deposit amounts, transaction frequency, deposit and withdrawal amounts, and final account balances.

Key findings

The study found four behaviour insights to explain low usage of savings accounts:

The minimum weekly deposit and opening deposit amounts anchored clients to deposit low amounts.
Clients opened new accounts with no plans or intentions of how they would use them.
Clients were not made sufficiently aware of the regular saving collection service at the time of account opening.
Saving goals are not salient.

Based on these barriers to improving savings outcomes CARD Bank’s account opening process was redesigned. This included: clients stating their saving intentions and goals on the application form, a separate savings plan for clients to complete, and a savings calendar for clients to use.

The RCT (with about 260 clients in each of the treatment and control group) tested the effects of the project and found statistically significant effects on savings behaviour:

Clients who received the treatment when opening a savings account made initial deposits 15 per cent higher than the control group, were 73 per cent more likely to initiate a transaction in the new account and made smaller and more frequent ongoing deposits as well as smaller withdrawals.
The treatment appears to have had the effect of increasing balances by 37 per cent compared to the control group.

The pilot project also generated lessons about microsavings and behavioural design that could be useful in future product and programme innovation:

Behavioural principles embedded into product design can trigger desired behaviours.
An attempt to close the intention-gap by helping clients take specific actions towards savings goals can influence behaviours among clients who already understand the importance of saving.
Rigorous data analysis is an important component of developing deep behavioural insights and institutions need to support this capability.
RCTs provide rigorous evidence that can be used to substantiate impact and support business decisions.

Points to consider

Methodological strengths: The study uses a RCT to measure the impact of the intervention. The report provides detail on the key stages that led to the design and implementation of the pilot intervention and the statistical methodology used for the analysis.
Applicability: The report includes general learning points about microsavings and about behavioural design. The authors suggest that the lessons from this pilot project are broadly applicable within the financial inclusion sector.
Generalisability: While this study is based in the Philippines, the behavioural insights will be of general interest to those interested in raising saving rates among low-income households.