More than 46 million Americans owe more than
.6 trillion in federal student loan debt, making it the second-largest category of household debt behind mortgages. It takes a decade or more to pay off these loans, so it’s little surprise that millions of borrowers default each year.
Behavioral science experts at Wharton have found a way to help struggling borrowers get back on track — and it costs the federal government practically nothing. In a large-scale study with 13 million student loan borrowers, they determined that sending emails to remind borrowers about alternative repayment options after they missed a payment can significantly reduce continued delinquency.
“Nagging really works,” said Wharton operations, information and decisions professor Katy Milkman, who studies how nudges can encourage positive behaviors such as exercising and saving. (Listen to the podcast.)
Robert Kuan, a doctoral candidate in the OID department who led the study, said, “Nudges work because they rely on psychological insights to improve communications, like the insight that people are forgetful and reminders can help them follow through.”
“Nudges work because they rely on psychological insights to improve communications.”— Robert Kuan
Reducing Student Loan Delinquency Rates Across the U.S.
The study, “Behavioral Nudges Prevent Loan Delinquencies at Scale,” was published in January in the journal PNAS. Kuan, Milkman, and their five co-authors partnered with the U.S. Department of Education to conduct a field experiment beginning in the fall of 2023. The timing is significant because it marked the end of a three-and-a-half year pause on repayments that was granted during the COVID-19 pandemic. After such a long period of forbearance, it can be challenging for borrowers to return to the habit of monthly payments, the scholars noted in the paper.
The researchers designed different email messages that were sent to borrowers who had missed a recent loan payment. The experiment randomized whether or not borrowers received standard missed payment emails or redesigned emails informed by behavioral science principles, which also encouraged them to apply for an income-driven repayment plan (IDR) and/or sign up for auto-pay. Among those who received redesigned emails, the experiment also randomized whether borrowers received a follow-up reminder message; whether their predicted savings with an IDR plan were described in dollars or percentages; and whether two separate emails encouraged both IDR and auto-pay enrollment or whether each email encouraged just one action.
Overall, email nudges reduced delinquencies by 0.42%. The researchers determined that if they had sent their best-performing nudges to all 13 million borrowers, it could have averted nearly 80,000 delinquencies.
“The key insights gleaned here can be relevant to anyone hoping to promote positive behavior change, whether a bank is hoping to nudge on-time credit card payments or a doctor is hoping to encourage follow through on healthy behaviors,” Kuan said.
“What’s exciting is it shows the power of behavioral science and testing at scale to make an impact on lives.”— Katy Milkman
Surprising Findings
Kuan and Milkman said there were two notable takeaways from the experiment. First, they expected that simplified messages, each encouraging a single action, would be more effective than equally frequent messages that repeatedly encourage two actions. Previous research on nudges has shown that simpler messages are more likely to incite action, such as applying for government benefits. But even though messages recommending two actions meant more information for participants to digest, these nudges proved better than messages suggesting one action at a time. This finding challenges the assumption that simpler communications are more effective because they are easier to understand.
“We were able to isolate, at least in this case, that repetition beats simplification,” Milkman said.
Second, they suspected that participants would respond more to seeing their potential savings in dollar amounts rather than percentages, which can be confusing to people who have a hard time interpreting numbers. But percentages prevailed in encouraging more participants to apply for income-driven repayment plans. Kuan thinks it has to do with how people perceive the size of the savings.
“We hypothesize that this might be because people typically evaluate percentages on a 0 to 100 scale, so numbers close to 100 feel larger and more impactful, even for individuals low in numeracy,” he said.
The scholars said the study builds on many others that prove nudges can be a cost-effective method to promote positive behaviors.
“I think what’s exciting is it shows the power of behavioral science and testing at scale to make an impact on lives,” Milkman said. “These are essentially free adjustments to a messaging campaign that have the potential for large impact.”
