Income-based student loan repayment plans, which adjust payments according to a borrower’s earnings, can help reduce the risk of default for individuals with lower incomes. However, new research from Stanford Graduate School of Business suggests that while these plans may encourage some borrowers to work less or choose lower-paying jobs, the overall benefits surpass the associated costs.
Student loan debt in the United States has reached $1.8 trillion, making it the second-largest liability for American households after home mortgages. Tim de Silva, assistant professor of finance at Stanford Graduate School of Business, noted: “Over the past 10 to 15 years, there’s been massive growth in the amount of student debt. Part of that is because college has become more expensive.”
De Silva explained that income-contingent repayment offers a form of risk protection similar to equity-based financing used by corporations. “College is a large but risky investment that may or may not pay off,” he said. “Early-stage businesses often finance their risky projects using equity, which provides a form of insurance. It makes sense that we may want the same thing for education, where repayment is based on your income: If college turns out to pay off for you and you earn a high income, you pay more, but if it doesn’t pan out, you have a lower income and hence pay less.”
Despite its advantages over fixed-payment systems—where all borrowers must make set payments regardless of post-graduation income—income-based repayment introduces what economists call “moral hazard.” This means some borrowers might reduce their labor supply or opt for jobs with fewer hours to minimize repayments. Such behavior can result in decreased government revenue from both loan repayments and taxes.
To assess these trade-offs, de Silva analyzed Australia’s government-sponsored student loan program—the first country to introduce income-contingent loans in the late 1980s—as part of his doctoral dissertation at MIT. Using data from over four million Australian borrowers, he compared outcomes under income-based and fixed-payment systems.
His findings were published in The Quarterly Journal of Economics and indicate that although some borrowers work less under an income-based system—often those with greater debt burdens or flexible schedules—the insurance value provided to low-income individuals outweighs these drawbacks. On average, affected borrowers worked one to two weeks less per year to avoid repayments.
However, de Silva found that switching from fixed payment policies (as used in the U.S.) to an income-based approach could increase households’ average lifetime consumption by nearly 1.5%. He stated: “An increase of almost 1.5% is pretty sizable.” For comparison purposes, full forgiveness of all student loans would yield less than double this effect on consumption levels. “The research shows that a properly designed income-contingent loan gets you over half of the way to full forgiveness while still ensuring that borrowers repay the same amount on average.”
“In short,” de Silva concluded,“They do distort household labor supply in a non-trivial way, but the value of the insurance benefit they provide is much larger.”
Applying lessons from Australia requires caution due to differences between countries.“Any time you extract from one country to another,you need to be very careful,”de Silva said.“But there are lessons that apply to the U.S., where there has been a growing push to introduce income-contingent repayment plans.”
He also pointed out challenges within current U.S.income-driven programs,saying,“It’s a mess.It’s very unclear what’s going to happen under the current administration.A first-order issue is finding ways to improve knowledge and reduce regulatory burden of income-contingent loans.”
Recent policy changes have included significant debt cancellation efforts.The Biden administration wrote off $184 billion in student loans,but de Silva argued against such measures:“My research shows that forgiving student debt is not a very efficient policy.Full forgiveness leads to a very large cost to taxpayers while also targeting forgiveness in an ineffective way.”
Similarly,de Silva criticized designs where balances are forgiven after set periods:“Adding forgiveness…creates an incentive for peopleto lower their incomes when they’re close tothe point wherethey wouldn’t have topaytheloan regardless.Income-contingent loans do alotof good things without needing second-best tools like forgiveness.”
