Can pre-commitment increase savings deposits? Evidence from a tax-time field experiment

For many low- and moderate-income (LMI) households in the U.S., the receipt of the tax refund marks one of the few times during the year when they can build savings. We present findings from a large-scale field experiment designed to encourage LMI tax filers to save their tax refund. The experiment tested combinations of behavioral strategies to promote savings including (1) asking filers at the start of tax preparation to pre-commit to saving their refund, and (2) choice architecture manipulations that emphasized directly depositing their refund into savings accounts or savings bond purchases. The four intervention conditions had a positive impact on the rate of refund deposits into savings accounts, increasing the deposit rate between 42% and 64%. However, the intervention conditions had minimal effects on the rate of bond purchases. The primary driver of increased savings deposits was the manipulation of the choice architecture. By comparison, savings pre-commitments had a more modest effect on savings deposits, on average. The impact of a pre-commitment on savings deposits varies over the course of the tax season, however, with early tax filers being much more responsive to pre-commitment than were later tax filers.