Does offering overdraft protection encourage people to set up automatic transfers?

Interventions
Overdraft protection
Experiment Type
Field Experiment
Goals
Increase program enrollment
Outcomes
Increase short-term savings
Focus Areas
Marketing & messaging
Behavioral Concepts
Default bias
Partner
SaverLife
Partner Type
Non-profit

What Happened

It didn't work. There were null results. Offering overdraft protection in the body of the email did not affect savings behavior or weather people enrolled in automatic transfers.

Lessons Learned

The way overdraft protection was used in this email intervention provided was unsucessful and provided null results. The authors suggest that mentioning "overdraft" within the email may have scared people away if overdrafts were not already top-of-mind.

Background

Having a small cushion of savings to fall back on is critical to someone’s financial well-being and security, yet far too many people struggle to build sufficient savings. There are many reasons why we find saving difficult, but one important reason is that saving feels painful. Building savings requires sustaining motivation over time and regularly making contributions to savings, effectively giving up some amount of spending in the short-term.

To explore how we might make saving a little easier, we partnered with a non-profit organization called SaverLife (formerly EARN). SaverLife adopts the philosophy that small savings can make a big difference and has a variety of interventions such as matched savings programs, prize-linked savings games, and other fun challenges aimed at making saving easier.

While the financial incentive increases savings behavior, some evidence suggests that making incentives contingent upon ongoing performance – such as paying people for successfully saving – can “crowd out” people’s willingness to continue the behavior after the incentive is taken away. In this instance, we worked with SaverLife members engaging with a 6-month matched savings program. We wanted to both increase the number of users who reached their monthly savings goals but also the number who continued to do so after the program ended.

Key Insights

One of the most effective ways to help people save regularly is to have them set up an automatic transfer to their savings. Automatic transfers work because setting one up is a decision someone only has to make once, but doing so locks them into a behavior that is good for them in the long-term. While automatic transfers make sense in theory, they are practically difficult for several reasons:

  • Setting up automatic transfers is more difficult if someone has irregular or variable income.

  • When people set up automated payments, they tend to set them up for smaller amounts. This is problematic because people feel a sense of achievement: they are making consistent progress. In reality, they are slower to achieve their longer-term savings goals.

  • Low-income individuals are at greater risk of overdraft fees. Since the consequences can be high, many fear automated transfers for savings.

Experiment

We tested our hypothesis that reducing the consequences of overdraft will increase uptake of automated savings transfers by sending SaverLife users two different emails:

Results

We found that users who were offered overdraft protection messaging were not more interested in setting up an automatic transfer, as measured by email click rate, and were not more likely to set up an automatic transfer. In the

longer term, both conditions saved about the same amount. The analysis also suggests that simply mentioning the possibility of “overdraft” in an email scared people away if overdrafts were not already top-of-mind.

However, regardless of condition assignment, users who opened the email were more than twice as likely to set up an automatic transfer. Users who clicked in the email, were three times more likely to set up an auto transfer. This can be attributed to self-selection bias (there is something different about people who opened an email vs. those who didn’t open an email), but regardless, this indicates that there is latent demand for automatic transfers and that simply reminding people can make a difference.