Roughly 60 million U.S. adults are unbanked or underbanked. These disproportionately low-income and minority Americans lack access to a safe financial account – an important vehicle for saving and a precursor to life-changing asset building. For many low- and moderate-income people, their federal tax refund is the largest lump sum of income they will receive during the year. The government sends out tax refunds via paper check to an estimated 5 million unbanked Americans and over 10 million low-income tax filers receive refunds via high-cost refund anticipation loans or checks. For these reasons, tax time provides an important opportunity to get people out of the world of high fees and into the financial mainstream. 

The MyAccountCard pilot, launched in January 2011, offers low-income taxpayers a safe, low-cost financial account and provides a safe and efficient mechanism for the electronic delivery of their federal tax returns. This pilot includes roughly 800,000 low-income individuals who were randomly assigned to one of eight treatment groups. These individuals received an offer to enroll in a prepaid debit card account into which their tax refund can be direct-deposited. The eight treatment groups differ along three dimensions: (1) no monthly fee versus low monthly fee ($4.95), (2) linked savings account versus no linked savings account, and (3) offer letter messaging –convenience vs. safety.

The evaluation, which is based on data through two tax filing seasons (January 2011-May 2012), is estimating the effects of the different prepaid card offers on card take-up and usage over time. The differential product features (i.e., card pricing and the availability of a linked savings account) and differential messaging approaches are being tested and evaluated as part of the pilot. Specific research questions include:

1. Are prepaid cards an effective way to keep people banked over time? Do people use the prepaid card as a plastic check or as a bank account?
2. Is there a difference in card use between those who receive a card with no monthly fee versus a low monthly fee? Is there a difference between those who receive a card with a linked savings account versus no linked savings account?
3. Do people offered the linked savings account have higher average card balances (transaction plus savings) than those not offered a savings account?

The impact analysis makes comparisons between different treatment groups. Because sampling variation can lead to differences in the average characteristics of members of each group, we use a regression-based method to control for measured characteristics (e.g., individual-level and zip code-level characteristics).

Overall, the evaluation findings will provide a comprehensive understanding of card use, banked status, and how card features affect card use and savings over time. The study will provide important information on whether prepaid cards at tax time are an effective way to keep people banked over time and whether providing low-income people with a linked savings account promotes savings.