Using Behavioral Economics to Help Individuals Reduce Debt in the United States

Fieldwork by:
Tulsa, Oklahoma, United States
465 Individuals
2010 - 2010
AEA RCT registration number:

Researchers designed and piloted a program called Borrow Less Tomorrow (BoLT) that took a behavioral approach to debt reduction, combining an accelerated loan repayment schedule with peer support and reminders. Results from a sample of free tax-preparation clients in Tulsa, United States suggest a strong demand for debt reduction: 41 percent of those offered BoLT used it to make a plan to accelerate debt repayment. The results also offer suggestive evidence that the BoLT package reduced credit card debt.

Policy issue

A mounting body of evidence suggests that behavioral factors, such as lack of self-control and an inability to remain focused on achieving a financial goal, impede individuals’ ability to accumulate wealth. Most financial products and policy instruments developed to overcome these behavioral issues focus on asset accumulation, such as retirement planning. For many households, however, debt reduction offers a more efficient path than asset accumulation to achieving greater wealth. Nevertheless, the availability of behaviorally-oriented financial interventions to reduce debt is far more limited, and additional research is needed to understand how such debt reduction programs should be structured and how they affect individuals’ financial health. This study is the first known evaluation to apply behavioral economics to debt reduction services.

Context of the evaluation

This study took place in 2010 in Tulsa, Oklahoma, a city located in the southern United States. Researchers partnered with the Community Action Project of Tulsa (CAP), an anti-poverty agency that provides a range of social services—including early childhood education, first-time homebuyer’s assistance, and free tax preparation—to low- and moderate-income individuals. The 465 participants of this study comprise individuals who sought tax preparation services from CAP under its Volunteer Income Tax Assistance program.

The majority of participants in the sample were low-income, with 75 percent reporting a total annual household income of less than US$30,000, which is equivalent to the bottom 31 percent of the national income distribution.1 The average individual credit card and auto loan debt of the sample was US$2,447 and US$5,546, respectively, which was low relative to U.S. averages. The mean age of the sample was 44 years and 74 percent was female.

Two people look at calculator and invoice

Details of the intervention

In 2010, researchers developed and piloted a program called Borrow Less Tomorrow (BoLT) to help CAP clients reduce their debt. During tax season (January-April), researchers and CAP staff asked tax preparation clients if they would be willing to complete a financial and behavioral survey in exchange for a US$5 gift card to a local gas station. Among the group that completed the survey, a total of 465 individuals were eligible to participate in the study because they had a positive balance on auto or credit card debt and had expressed interest in reducing their debt. All participants also granted permission for researchers to pull their credit reports on a regular basis to monitor debt payments and financial status.

Researchers randomly assigned 238 individuals to be offered BoLT (the intervention group), and 227 individuals to not be offered BoLT (the comparison group). For those offered BoLT, the research team explained the program components to the participant and worked to identify a single, suitable debt on which to focus effort (e.g. a debt with a substantial balance and a high interest rate).

BoLT comprised three separate interventions:

  1. Planning/Goal Setting: The surveyor used a simple repayment schedule calculator to show the participant how small increases in monthly payments could help dramatically reduce the time and cost to pay off their debt. The participant and surveyor would then establish a realistic repayment plan. In addition to an overall acceleration in repayment, participants were also offered the option to escalate payments every month. For example, a participant could commit to paying US$25 in month one, US$35 in month two, and so on to pay off debt at an even lower cost and faster pace.
  2. Peer Support: For those participants who agreed on an accelerated repayment plan, surveyors offered the participant the option of selecting one or more peers to be notified if she fell off-track with her repayment commitment. The peer could then offer encouragement (but not financial support) to help the participant regain momentum and reach her repayment goal.
  3. Reminder Notices: As a tool to focus participants’ attention on their debt reduction goals, those who agreed on an accelerated repayment plan were also offered the option of receiving a monthly reminder by email or phone to stay on track with their commitments.

Results and policy lessons

Demand for debt reduction support: Overall, researchers found strong demand for behaviorally-motivated debt reduction support. Among those randomly assigned to receive the offer to participate in BoLT, 41 percent signed up for an accelerated repayment plan. Of those who signed up for the plan, 41 percent signed up to escalate payments every month. Conditional on take-up of BoLT, 27 percent accepted the offer to receive peer support and 81 percent accepted the offer to receive monthly reminder notices. Households living in extreme poverty (i.e. incomes less than US$10,000 per year) were less likely to sign up for BoLT, but there is no evidence that take up differed by whether a participant demonstrated more or less self-control or attention paid to her finances.

BoLT Performance: By monitoring credit reports, researchers found that 51 percent of BoLT participants were on-schedule with their repayment plan after twelve months in the program. The study demonstrated weak evidence that those who were offered the opportunity to participate in BoLT achieved a lower overall level of debt after one year than the comparison group, which did not receive the BoLT offer. However, many of the estimated differences in debt reduction between the intervention and comparison groups were not statistically significant. The researchers found no evidence of a difference in credit scores, payment delinquencies, or credit line use between the intervention and comparison groups.

While noting that these pilot results should be considered with caution due to limited sample size and the use of just a single program design, the researchers found strong demand for debt reduction products and services, but only suggestive evidence that this product led to improved financial well-being for participants. They posit that debt reduction products and services could be used by businesses and financial advisors to enable employees and clients to achieve their financial goals.

Karlan, Dean, and Jonathan Zinman. "Borrow Less Tomorrow: Behavioral Approaches to Debt Reduction." Working Paper, May 2012.

U.S. Census Bureau. Accessed 1 July 2013.