Borrowers have more choices than ever for paying down their student loans. Although many borrowers still automatically enroll in standard 10-year, fixed-payment plans, some opt to enroll in programs that allow smaller monthly payments, such as income-driven repayment plans, or plans with longer repayment periods. I present a new analysis of borrower-level credit bureau data, showing that borrowers who start repayment with smaller, or modified, payments have better short-term loan outcomes but do not fully avoid delinquencies and defaults.

My three primary findings:

• Borrowers who started student loan repayment with smaller, modified monthly payments between August 2011 and August 2012 are roughly 11 percentage points less likely to default before August 2014 than borrowers with standard payments (8 percent of modified-payment borrowers default, compared with 19 percent of standard-payment borrowers).
• Borrowers who use smaller, modified monthly payments are more likely to be older, have higher student loan principals, and have higher credit scores. However, they are also more likely to have at least one dollar of deferred, or postponed, student loan debt in their first year of repayment, and they are more likely to have had a severe delinquency on a student loan in the previous three years.
• After accounting for borrower characteristics, I determine that the use of modified payments plays a much smaller role in predicting the likelihood of default, reducing the likelihood by about 3 percentage points compared with standard repayments.

Methodology

Federal loans offer several modified repayment plans that borrowers can opt in to, including plans that cap repayment at a certain percentage of a borrower’s annual discretionary income (REPAYE, PAYE, IBR, and ICR), as well as plans that lengthen the repayment period and/or allow graduated payments over time. Private loans have fewer repayment options, though some student lenders permit lengthening the repayment period above the standard 10 years or using graduated payments that start low and increase over time.

To examine the impact of making smaller payments on credit outcomes, I use a dataset containing credit records over a six-year period on a random sample of more than 5 million US consumers. This dataset provides an annual snapshot of consumers’ credit records each August from 2010 to 2015. The data available include the original student loan amount (or principal) in repayment, the amount of loans deferred, the expected monthly payment, and the delinquency or collections status on student loans.

I analyze a sample of borrowers who entered repayment on their student loans between August 2011 and August 2012 (i.e., had no expected monthly payment listed for the previous two years). To identify borrowers who are paying less than what would be expected under a standard 10-year repayment plan, I calculate an expected monthly payment based on the standard repayment of their loan with a 5 percent interest rate and no fees. If the borrower has a monthly payment that is less than 90 percent of this expected payment, I identify her as having a smaller, or modified, payment plan. For example, a borrower with a $10,000 student loan principal would be expected to have a monthly payment of $106 under a 10-year plan with a 5 percent interest rate. Thus, borrowers in my sample with a $10,000 open student loan principal and an initial monthly payment of less than $95 are counted as having a modified payment plan.

Results

In my sample, 27 percent of student loan borrowers who started repayment in 2011–12 had a monthly payment that was lower than would be expected given the open principal amount on their loan. These borrowers had a median monthly payment that was about 63 percent of what would be expected under a standard 10-year plan.

According to my results, lower modified payments are associated with lower student loan defaults. About 8 percent of people who entered repayment using a lower-than-standard monthly payment were in default after two years, compared with 19 percent of people who had standard payments (figure 1).¹ Borrowers with modified payment plans were also more likely to stay current and less likely to have delinquencies (late payments of up to 120 days).

These results come with several caveats. Most important, the borrowers in my modified payment group differ substantially from those in the standard payment group. Modified-payment borrowers tend to be older (average age of 36 at the start of repayment, compared with 32). Those making modified payments also had better credit scores (average credit score of 647, compared with 598, slightly less than half a standard deviation difference). This difference in credit score is consistent across age categories of borrowers.

Despite having better credit overall, borrowers with modified monthly payments appear more likely to be consolidating or rehabilitating their loan. Nine percent of modified-payment borrowers had severely delinquent (more than 120 days late) or defaulted student debt within the previous three years, compared with 4 percent of those with standard payments. Those with modified monthly payments were also more likely to still have some deferred student debt in 2012 (23 percent compared with 17 percent of standard-payment borrowers).

A recent White House report shows that the use of income-driven repayment plans is concentrated among borrowers who tend to have large student loan balances (Executive Office of the President 2016). Although I can’t specifically identify whether borrowers are on income-driven plans, my findings echo these results. Individuals in the modified-payments group have more than double the student loan principal of those using standard payments (approximately $33,300, compared with $15,700). Just 14 percent of borrowers with a loan principal of less than $10,000 appear to have a modified-payment plan, while 60 percent of borrowers with more than $50,000 in debt start with a modified payment.

Regardless of monthly payment size, borrowers with relatively low loan amounts are the most likely to default in my dataset (figure 2). However, in every bracket of debt, borrowers who start with modified payments are always less likely to default than borrowers with similar loan amounts and standard payments.

Previous research on similar credit bureau datasets has shown that student loan delinquency is distributed unequally across the United States: delinquency rates appear to be highest in the South and West.² I find that borrowers with modified payments in 2012 are more likely to live in the Northeast and less likely to live in the Midwest and South. Even after I control for borrower characteristics such as age, previous delinquency, student loan principal, presence of deferred loans, and credit score, borrowers in the Northeast are still slightly more likely to use modified payments than borrowers in other regions.

My descriptive analysis indicates that those entering repayment in 2011–12 with modified monthly payments reduce their likelihood of default by 2014 by roughly 11 percentage points (8 percent of modified-payment borrowers default, compared with 19 percent of standard-payment borrowers). However, after I account for differences in age, previous delinquency, loan principal, presence of deferred loans, credit score, and US region, the difference in default rates drops to just 3 percentage points. It appears that borrowers who choose to use modified repayment have many characteristics that already make them less likely to default. In particular, these borrowers have better credit scores and higher student loan balances, traits that may indicate college completion or graduate education (factors not observed in the credit bureau data).

Limitations

My findings are constrained by the dataset, which contains anonymized credit records, similar to the Consumer Credit Panel used by the Federal Reserve Bank. As a result, I cannot distinguish whether an individual has a federal or private student loan (or a combination of the two), though an estimated 10 percent of the student loan volume in the past five years was nonfederal loans (College Board 2015). Also, I may be capturing a small portion of borrowers who are making small monthly interest payments on a private student loan, or on a Parent Direct PLUS loan, in my modified-payment group.

Another limitation to my analysis is that I cannot determine when, or if, a borrower completed higher education, or how long a borrower was in an economic hardship deferral or forbearance before starting payment. Since borrowers in my modified-payment group are more likely to have had delin-quent or defaulted student debt before entering repayment, part of the reduction in defaults may stem from modified-payment borrowers trying to avoid the consequences of a second student loan default.

Conclusion

Though lower monthly payments are associated with lower defaults and delinquencies, borrowers who start with modified repayments do not escape defaults entirely. There may be several explanations for this phenomenon. Borrowers on a graduated repayment plan, where payments start low and grow larger, may not be able to keep up with payments once they start to increase. Individuals enrolled in income-driven repayment plans must recertify their income and family size each year to stay in the program, so some individuals may drop out of the program after the first year. According to a 2015 Consumer Financial Protection Bureau report, over half of borrowers on federal income-driven repayment plans fail to recertify their income on time, and nearly one-third do not recertify within six months of the deadline (CFPB 2015). In my sample, 31 percent of borrowers who started with a modified monthly payment plan had a payment that resembled a standard payment by 2014. Unlike income-contingent programs in other countries, which automatically deduct a percentage of a worker’s take-home pay each pay period, US federal loan repayment plans are slow to respond to shocks in earnings, such as the sudden loss of a job or temporary disability.³ All these factors could contribute to delinquency and default.

Lower monthly payments—whether through longer repayment periods, graduated payments, or income-driven repayment plans—may help put a dent in short-term delinquencies and defaults. However, individuals who opt into these plans tend to differ significantly from borrowers who rely on standard plans. As adoptions of these programs increase, particularly for low-balance, low-income borrowers, policymakers should focus on making these programs easier to navigate and more responsive to unexpected borrower hardship.

Data Appendix

I use data from a random 2 percent sample of five years of depersonalized consumer data (2010–15) from a major credit bureau. Information was collected each August from 2010 through 2015. If a consumer drops out of the data (for example, because he or she passes away), a new consumer is added in a manner that retains the randomness in the sample.

Although I refer to the repayment group as a cohort, this metric is not equivalent to the two- and three-year cohort default rate on federal loans used to evaluate colleges and universities. My metric includes all individuals entering repayment on a student loan, not just individuals who withdrew or graduated from an institution in the prior year. In addition, the federal cohort default rate excludes loans that are rehabilitated from default during the measurement period from the numerator of the calculation. In contrast, I identify the worst status on any student loan in the period and do not allow for the possibility of student loan rehabilitation in the given period.

For this brief, I focus on the results from the cohort of borrowers in my sample who appear to have entered repayment between August 2011 and August 2012 (2012 cohort). In table A.1 below, I also present my results for borrowers who entered repayment between August 2012 and August 2013 (2013 cohort).

Notes

1. There are different definitions of default for federal and private student loans. I define default as having a severe delinquency (above the maximum 120 days late payment reported in my dataset) on a student loan and at least one dollar of student loans in collections.↩
2. Marshall Steinbaum and Kavya Vaghul, “An Introduction to the Geography of Student Debt,” Equitablog, Washington Center for Equitable Growth, December 1, 2015, http://equitablegrowth.org/research-analysis/an-introduction-to-the-geography-of-student-debt/.↩
3. Susan Dynarski, “America Can Fix Its Student Loan Crisis. Just Ask Australia,” The Upshot (blog), New York Times, July 9, 2016, http://www.nytimes.com/2016/07/10/upshot/america-can-fix-its-student-loan-crisis-just-ask-australia.html.↩

References

CFPB (Consumer Financial Protection Bureau). 2015. Student Loan Servicing: Analysis of Public Input and Recommendations for Reform. Washington, DC: CFPB.

College Board, The. 2015. Trends in Student Aid 2015. Washington, DC: The College Board.

Executive Office of the President of the United States. 2016. “Investing in Higher Education: Benefits, Challenges, and the State of Student Debt.” Washington, DC: Executive Office of the President of the United States.

About the Author

Kristin Blagg is a research associate in the Income and Benefits Policy Center at the Urban Institute, focusing on education policy. Before joining Urban, she spent four years as a math teacher in New Orleans and New York City. She has also been a human capital consultant for federal government clients. Concurrent with her work at Urban, she is pursuing a PhD in public policy and administration at George Washington University. Blagg holds a BA in government from Harvard University, a MSEd from Hunter College, and an MPP from Georgetown University.

Acknowledgments

This brief was funded by a grant from Urban Institute’s Fleishman Fund.

The views expressed are those of the author and should not be attributed to the Urban Institute, its trustees, or its funders. Funders do not determine research findings or the insights and recommendations of Urban experts. Further information on the Urban Institute’s funding principles is available at www.urban.org/support.

The author thanks Matthew Chingos for his guidance and helpful comments in producing this brief. The author also thanks Sandy Baum and Victoria Lee for their thoughtful feedback.