SAVE in the balance: The future of income-driven repayment for federal student loans

Income-driven repayment (IDR) policy is at the center of a legal dispute impacting millions of borrowers and billions of dollars in loan balances. IDR is a critical but complicated element of the student loan program. Under IDR plans, borrowers make payments based on their income rather than a fixed amount every month for the life of the loan. Legal challenges to a Biden administration rule introducing a new IDR plan known as Saving on a Valuable Education (SAVE) have upended the system.

President Trump issued an executive order last month stating the secretary “shall to the maximum extent appropriate and permitted by law, take all necessary steps to facilitate the closure of the Department of Education.” However, the president does not have the authority to shutter the department or even to implement a proposed shift of the student loan program to the Small Business Administration (SBA). And even if Congress took action to eliminate the department or make changes to the student loan program for new borrowers, both of which are unlikely, there is $1.6 trillion in outstanding student loan debt to be collected or discharged. Right now, the system for student loan repayment is in chaos, and policymakers—in Congress, the Trump administration, or both—will need to take some action to address repayment in the coming months.

In this report, the first in a series to inform the future of IDR policy, we explain the purpose and history of IDR, what IDR plans exist and how they work, and how the litigation is affecting borrowers and taxpayers.

Why is action on repayment policy necessary?

Until recently, the federal student loan system operated four major IDR plans—Income-Contingent Repayment (ICR), Income-Based Repayment (IBR), Pay As You Earn (PAYE), and Revised Pay As You Earn (REPAYE)—in addition to the standard repayment option, graduated repayment plan, and extended repayment plan. Student loan repayment in general and IDR options in particular are confusing for borrowers, and this problem has likely worsened due to recent turmoil in the loan program. These programs have also suffered from administrative and implementation problems, leaving some borrowers unable to navigate the system effectively or to access affordable payment options.

The Biden administration introduced a new plan, Saving on a Valuable Education (SAVE), which replaced REPAYE and was more generous to borrowers, especially those with undergraduate debt. The SAVE plan opened to enrollment in August 2023, and nearly 8 million borrowers had signed up for (or were transferred automatically to) SAVE before it was closed to new enrollment in July of 2024 because of the litigation. Since then, SAVE borrowers have been in interest-free forbearance at a significant cost to taxpayers. We estimate that cost is about $2 billion per month, though this estimate is subject to considerable uncertainty because it depends on the alternative repayment regime borrowers would face if the litigation had not happened, among other assumptions.¹

Because borrowers are already enrolled in SAVE and because other provisions, such as the combined online application for IDR, have taken effect, the court cannot simply strike down the SAVE plan and return to a pre-SAVE status quo. An explicit plan to transition to a new set of rules will be necessary.²

Note that the terminology can be confusing: In the United States, income-driven repayment (IDR) is the catch-all term covering all of the repayment plans where payments depend on income. Some court documents use “ICR” to refer to the three plans that are based on the 1993 law authorizing the development of an income-contingent repayment plan: ICR, PAYE, and REPAYE/SAVE. We follow the convention of using “IDR” to refer to the class of repayment plans where the payment depends on income and only use “ICR” to refer to the specific ICR plan.

What is the purpose of IDR plans?

Income driven repayment plans exist because standard student loan repayment plans often fail both borrowers and lenders—borrowers can’t insure against disappointing earnings outcomes and fixed payment schedules often don’t align with graduates’ earnings trajectories, leading to poor repayment outcomes. When transitory periods of low income cause defaults, lenders are worse off than in a contract that would allow for eventual collection over a longer time horizon aligned with income trajectories. IDR plans can address these problems with the standard, fixed payment plan for student lending.

In practice, IDR plans are also a way to subsidize post-secondary education for some borrowers, though this is a departure from the original rationale for IDR and may be inefficient relative to more direct approaches such as financial aid. 

Income-driven repayment gives borrowers more time to repay

IDR aligns repayment with the trajectory of earnings. Once borrowers enter repayment after completing their education, they pay a constant monthly payment for a pre-specified amount of time, 10 years, on the standard repayment plan. This payment structure doesn’t always line up well with the financial returns to the post-secondary investment. Wages tend to rise over a worker’s career, especially for those with bachelor’s or graduate degrees, but under the ten-year repayment plan, borrowers need to pay a large share of their loans back early in their career when earnings are lower. That is, they have to repay their loans before they are receiving the full financial benefit of their educational investment.³

“Graduated” and “extended” payment plans can help by extending the term of the loan, but they still may not match the profile of earnings. Well-designed IDR plans can effectively create repayment plans that are customized to match the path of loan payments to the path of earnings. 

Medical school graduates are a particularly stark example of this phenomenon. They may accumulate substantial student loan balances—often $200,000 or more—and enter repayment while they are in residency, where salaries average around $70,000. Physician incomes in mid-career and beyond are much higher, compared to during residency and compared to their student loan balances (although lifetime incomes of physicians vary considerably depending on their specialty and type of practice). Such borrowers could benefit from more time to repay their loans, to match the timing of repayment with the timing of income over their careers. IDR plans accomplish this.

Income driven repayment acts as insurance

IDR plans provide insurance where private markets won’t. Insurance can be purchased for a car or home, but private companies are reluctant to offer insurance protection against disappointing career outcomes. IDR fills this gap by reducing payments when an educational investment yields low returns.

The returns to higher education are uncertain, so borrowing to finance that education exposes students to risks. For example, those who graduate in a bad economy or have a worse employment outcome—and lower earnings—than they expected when they borrowed may be unable to repay their loans. In such cases, loan payments compete with consumption for basic needs, while loan defaults may generate knock-on financial consequences due to poor credit. Concerns about the down-side risk of borrowing to finance an education that doesn’t pay off as much as expected can lead to debt aversion and under-investment in higher education.

While well-designed income-driven repayment plans can act as insurance in the event of unexpectedly bad earnings, not all weak (or strong) outcomes are unpredictable. Insurance weakens the incentives to avoid the bad outcomes. When borrowers don’t expect they will need to repay all of their loans if their earnings are low, they may choose to borrow more than they expect to repay. For example, students may not avoid institutions, programs, and majors that data show do not reliably increase earnings (perhaps because they lead to occupations with low earnings). That is, the presence of insurance that softens the blow of an outcome where wages are low relative to borrowing may make such outcomes more likely to occur. Educational institutions might also respond by increasing tuition, with the expectation that students can borrow and have their loans forgiven. These responses not only increase the cost of the student loan program—a cost borne by taxpayers—but also can lead students to pursue post-secondary education even when the returns are likely to be low. 

Critically, the fact that insurance, in general, can cause some negative side effects does not mean there should be no insurance, and IDR is no exception. Rather, policymakers need to balance the insurance and other benefits of IDR against these potential unintended consequences.

Generous IDR is a subsidy for post-secondary education

Aside from spreading payments out over a longer time horizon and providing insurance, IDR can also be a way to subsidize post-secondary education for borrowers who have high debt relative to their income. When the terms of IDR are sufficiently generous to borrowers such that many borrowers do not expect to repay their loans in full, IDR has the effect of converting loans into grants. Whether students or institutions capture the benefit of this subsidy is complicated. For example, if institutions of higher education increase tuition in response to the IDR subsidy, they will capture some of the benefit of the subsidy even though on paper it is the student’s loan that is cancelled.

Relative to more direct subsidies, such as the Pell Grant program, subsidizing education through the loan forgiveness component of IDR encourages borrowing since students can only access the subsidy if they borrow. It is also less transparent to students since they face uncertainty about how much of their loan will be forgiven. The distributional implications of IDR subsidies also merit note in that the subsidy is larger for borrowers with larger balances.⁴ The fact that generous IDR functions in part as a grant is considered a positive aspect of the program by some and an unwelcome departure from the original purpose by others. 

How do IDR plans work?

Although IDR plans differ on a range of important details, all of the existing and proposed plans in the U.S. follow the same basic structure. We describe here the parameters policymakers have to choose when designing an IDR plan.

Key Parameters of Income-Driven Repayment

• Eligibility of borrowers and loans: Not all borrowers and loans are eligible to participate in all IDR plans. Eligibility depends on the type of loans, when they were originated, and whether the loans have ever been consolidated. In addition, some plans (IBR and PAYE) have a partial financial hardship (PFH) requirement which requires borrowers to have incomes sufficiently low (and standard payments sufficiently high) that their payment in the IDR plan would be less than the standard payment.⁵
• Income protection limit: When calculating payments under IDR, some income is excluded; this income exemption threshold is specified as a percentage of the federal poverty level.⁶ Payments are calculated based on income above this threshold (sometimes referred to as discretionary income), and borrowers with income below the threshold will have a “zero payment,” meaning they do not pay anything but typically get credit for time toward forgiveness.
• Payment rate: IDR plans in the United States calculate payments as a percentage of discretionary income, 5% to 20%, depending on the specific plan.
• Maximum payment: Some IDR plans cap payments at the standard 10-year payment amount while other plans allow payments to rise above this amount as income rises.
• Time to forgiveness: IDR plans also set a number of qualifying payments a borrower needs to make before the remaining balance is forgiven. In the U.S., this ranges from 10 to 25 years.⁷
• Treatment of “accrued-but-unpaid” or “uncovered” interest: Sometimes a borrower’s IDR payment based on income is not enough to cover the interest on the loan, in which case, the loan balance will grow. The extent to which uncovered interest is subsidized differs across programs, with important implications for how much different types of borrowers benefit from IDR.
• Treatment of married couples: For married couples, IDR plans need to specify whether just the individual borrower’s income or the household income will be counted when calculating the IDR payments. Plans have different rules about the treatment of a spouse’s income when calculating payments.

The choice of the key parameters described above will determine how much the IDR plan costs taxpayers, which will be reflected in the federal subsidy rate—the share of the loan plus interest that taxpayers, rather than the borrower, pay back—for the affected loans. If the payment rate is high and/or the time to forgiveness is long, the subsidy could be quite low or even zero. However, when a combination of higher income protection thresholds, lower payment rates, and shorter times to forgiveness make IDR more generous to borrowers, the federal subsidy rate and cost of IDR increase, sometimes substantially.

Aside from these key parameters, how IDR plans are administered has important effects on how well the plans work in practice. If borrowers do not know what plans they qualify for, cannot find and complete the correct IDR application, or cannot file paperwork necessary to maintain eligibility, IDR will not be effective in meeting its goals.

The administration of IDR programs has long been plagued by problems.⁸ Leading up to the development of the SAVE Rule, administrative barriers continued to prevent many borrowers from taking advantage of an IDR plan for which they qualified. The SAVE Rule introduced several changes to simplify the application and recertification processes for all IDR plans, including the new SAVE program.⁹

How do IDR plans interact with PSLF?

Many borrowers who enroll in IDR plans also participate in the Public Service Loan Forgiveness (PSLF) program, which reduces the time to loan forgiveness. PSLF is not itself an IDR plan, but it interacts with IDR. Borrowers who work full-time in the government or non-profit sectors can enroll in PSLF, and their loans will be forgiven after making payments for 10 years, instead of the 20 to 25 years for borrowers who qualify for IDR but not PSLF. Importantly, PSLF and IDR serve different functions. IDR is primarily meant to give borrowers more time to pay and provide insurance against poor earnings outcomes as described above, while PSLF is specifically meant to subsidize public sector and nonprofit employment by forgiving student loans only for workers in those sectors.¹⁰

How do IDR plans differ from each other?

Although we expect the SAVE plan will either be eliminated or altered through the reconciliation process or as a result of the ongoing litigation, it is still instructive to understand its provisions to inform the options that Congress and/or the Department of Education have in developing an alternative. We discuss not only how the SAVE plan compares to earlier IDR plans but also the distinctions among non-SAVE plans. 

Before the introduction of SAVE, there were four IDR plans.

• The 1993 Omnibus Reconciliation Act (OBRA93) authorized the Department of Education to establish the first IDR program, known as Income Contingent Repayment (ICR). ICR opened for enrollment in 1994. This program is available to all borrowers, including parents who borrowed Parent PLUS loans.¹¹ ICR is less generous than later programs and enrollment is low relative to the other plans. 
• The 2007 College Cost Reduction and Access Act specified a second, more generous IDR option known as Income Based Repayment (IBR). The Health Care and Education Reconciliation Act (HCERA) of 2010 specified a more generous version of IBR for new borrowers after July 2014. For most borrowers, IBR is a better deal than ICR, but Parent PLUS loans are not eligible for IBR.
• The Pay As You Earn (PAYE) program was created in 2012. PAYE offered the more generous terms of “new” IBR to more borrowers, but participation in this program was limited to new borrowers. 
• The Revised Pay as You Earn (REPAYE) program, created in 2015, expanded access to IDR for more borrowers and made several other changes relative to PAYE.
• REPAYE was overwritten and rebranded as Saving on a Valuable Education (SAVE). SAVE was opened for enrollment in 2023, and borrowers in REPAYE were transitioned automatically to SAVE at that time.

Participation in IDR has increased substantially over the years: Nearly 60% of federally-held student loans ($714.4 billion) and 40% of borrowers in repayment at the end of 2024 were participating in IDR programs, up from 28% of loans and 15.5% of borrowers a decade ago.

Although they were introduced later, the regulations that created PAYE, REPAYE, and SAVE also relied on the provisions in OBRA93 that were the basis for ICR, which was introduced in 1994. Those provisions authorize the secretary of education to develop “an income contingent repayment plan, with varying annual repayment amounts based on the income of the borrower, paid over an extended period of time prescribed by the Secretary, not to exceed 25 years,” whereas the two versions of IBR are authorized under different laws. This distinction is important because the plaintiffs in the SAVE litigation are arguing that the 1993 law does not authorize loan cancellation at the end of the repayment term. Loan cancellation in SAVE/REPAYE, PAYE, and ICR is currently on pause for this reason.

Eligible borrowers and loans

Table 1 summarizes the rules about which loans and borrowers are eligible for each IDR plan. The student loan program is comprised of several different loan programs. There are three key distinctions that are important for purposes of IDR eligibility:

Was the loan made to a student or a parent? Most student loans are borrowed by undergraduate or graduate students to pay for their own education, but parents can also borrow to pay for their children’s education through the Parent PLUS program.¹² ICR is the only IDR plan for which loans made to parents are eligible. Parent borrowers can take advantage of the “double consolidation loophole” to access IBR or PAYE, but this is difficult to execute and the loophole was set to be closed by July of 2025 by the SAVE Rule. It is unclear how the litigation blocking the SAVE Rule affects the provision closing this loophole. While the loophole may be technically reopened because the SAVE Rule is enjoined, it has been difficult or impossible to access the loophole since last July because the IDR and loan consolidation applications have been frequently unavailable due to the litigation.

Is the loan a Direct Loan or not? Before around 2010, most student loans were made by private lenders and guaranteed by the federal government through Federal Family Education Loan (FFEL) and Perkins programs. The Direct Lending program, whereby the government holds the loans directly, was introduced as a pilot in 1992 and made more widely available by OBRA93, but was initially a small share of loans. The Direct Lending share of loans increased over time and in 2010 the government shifted completely to Direct Lending. Older FFEL or Perkins loans can be consolidated and converted to a Direct Loan. IBR is the only plan that borrowers with FFEL loans can access without consolidating to a Direct Loan. A borrower can “consolidate” a single FFEL loan to convert it to a Direct Loan to access other IDR plans. Still, borrowers might hesitate to consolidate FFEL loans if consolidation would result in a higher interest rate and because unpaid interest gets capitalized during consolidation. 

When did the borrower take the loans (or consolidate them)? ICR, REPAYE, and SAVE are available to all eligible borrowers and loans on the same terms. Eligibility for IBR does not depend on when the loans were originated, but those who first borrowed after July 1, 2014 can access more favorable terms. PAYE has the most complex timing requirements, summarized in Table 1.

Protected income and payment rate

Together, the protected income threshold and payment rate determine the monthly payment in an IDR plan. The payment owed is equal to discretionary income—that is, income above the protected income threshold—multiplied by the payment rate. Consider an IDR plan with a protected income threshold of 150% of the poverty line and a payment rate of 10% (as in PAYE and new IBR). Table 2 illustrates the calculation for two scenarios. In 2023, for example, the poverty line for a single individual was $14,580, so the protected income threshold was $21,870. Any borrower with income below that level would pay nothing. A single borrower with no dependents and $40,000 in income, for example, would owe $1,813 per year (10% of $40,000 minus $21,870) or $151 per month (Example 1 in Table 2).

The poverty threshold for a family of four was $30,000, so a married couple with two kids and $120,000 in income would pay $625 per month. Note that the IDR payment does not depend on the loan balance, but it would not be desirable to participate in IDR unless the payment would be lower than under the standard ten-year plan, which does depend on the loan balance.

Later IDR plans have income protection limits and payment rates that generate lower payments compared to their predecessors, as summarized in Table 3. The last two columns show what the monthly payment would be for the same two hypothetical borrowers in Table 2 under each plan.

The oldest plan, ICR, has a protected income threshold equal to the poverty line and a payment rate of 20%.

In IBR, PAYE, and REPAYE the protected income threshold is higher, at 150% of the poverty line. The payment rates in PAYE, REPAYE, and for more recent borrowers in IBR is 10%. IBR participants who borrowed before July 2014 face a payment rate of 15%, but those borrowers would generally qualify for—and benefit from enrolling in—PAYE and/or REPAYE. Indeed, PAYE was largely designed to offer the new IBR terms to older cohorts of borrowers, and REPAYE was meant to expand eligibility to those excluded from PAYE because they borrowed before October 2007 (as well as address several design issues with the PAYE plan).

In SAVE, the threshold for protected income is higher, 225% of the poverty line. And for undergraduate borrowers, the payment rate was reduced to 5% but was still 10% for graduate loans (or a weighted average of the undergraduate and graduate loan payment rates for borrowers with both types of loans).

The examples in Table 3 illustrate the increasing generosity of IDR plans. For the low-income single borrower in the first example, the payment would be $302 under ICR, $227 under old IBR, and $151 under (new) IBR, PAYE, or REPAYE. Under SAVE, the payment would be only $30 if the loans were for undergraduate studies and $60 if the loans were for graduate study (or a weighted average if the loans were a mix of undergraduate and graduate loans). The difference between SAVE and the earlier plans is smaller for graduate borrowers and those with somewhat higher incomes but still substantial. For the moderate-income married couple in the second example, their payment would fall from $625 under REPAYE to $438 under SAVE if they borrowed for graduate school. The two plans have the same payment rate for graduate borrowers, so this savings is due to the higher protected income threshold.

Another important distinction among plans is whether the payment is capped. In ICR, the payment is capped at what it would have been if the balance at the time the borrower entered ICR were amortized over 12 years (rather than 10 on the standard plan). Borrowers can only enroll in PAYE or IBR if the payment calculated based on income is less than what they would pay in the standard plan—this is referred to as the “partial financial hardship requirement.” Once enrolled, their payment is capped at what the payment would have been on the standard 10-year plan when they first enrolled. By contrast, in REPAYE or SAVE, the payment is uncapped, but the borrower may be able to change to a different plan if their income (and income-based payment) rises, though they may lose the opportunity to get credit towards PSLF or IDR forgiveness.

Time to forgiveness

The number of payments borrowers need to make before remaining balances are cancelled—the time to forgiveness—also varies across IDR plans, as summarized in Table 4. Under pre-SAVE IDR plans, borrowers have to make payments for either 20 or 25 years. As for the payment rate and income protection limits discussed above, the time to forgiveness generally became more generous in later IDR plans, though the REPAYE plan introduced a longer timeline for graduate borrowers, a feature that was retained in SAVE. Recall that PSLF shortens the time to forgiveness for those employed in the public or nonprofit sectors.

SAVE also shortened the timeline for forgiveness to 10 years for undergraduate borrowers who borrowed less than $12,000 in total. The time to forgiveness scales up proportionally for borrowers with original balances between $12,000 and $22,000, where the time to forgiveness for borrowers with only undergraduate loans is 20 years. That is, a borrower with an initial principal balance of $13,000 will see forgiveness after 11 years, a borrower with an initial principal balance of $14,000 will see forgiveness in 12 years, and so on.

Perhaps counterintuitively, student loan borrowers with the lowest balances tend to have the highest default rates, while those with large balances struggle less, on average. Borrowers with large balances typically completed a graduate or professional degree, investments that often pay off in the form of higher incomes which allow them to repay their loans.¹³ Many small-balance borrowers, on the other hand, attended some college but did not complete a degree. Without a degree, their incomes are often low, making it difficult to pay back even the relatively small amounts they borrowed.

Forgiving small balances sooner can benefit both borrowers and the government. Consider, for example, someone who borrowed $5,500 (the maximum for the first year of college) at an interest rate of 6.5%, but dropped out after one year and has earnings of $40,000 per year (like Example 1 in Table 2 above). Such a borrower would pay about $30 per month in the SAVE program, but this is only enough to cover the interest on the loan, so the loan balance will remain roughly constant until it is forgiven. The borrower is not making large payments, and the government faces a fixed cost of administering each loan, so forgiving such a loan in 10 instead of 20 years does not cost the government that much, benefits the borrower, and is well-targeted to low-income borrowers who need the protection most.

Treatment of uncovered accrued-but-unpaid interest

Sometimes a borrower’s payment under IDR is not enough to cover the interest on the loan in the same period. Without some type of interest subsidy, the loan balance grows because payments are applied to interest first, so the borrower is not paying down principal. This phenomenon is known as “negative amortization.” IDR plans differ in how they treat this “accrued-but-unpaid interest,” and this is another dimension along which the SAVE plan is significantly more generous than earlier plans. In fact, the SAVE plan eliminated the possibility of negative amortization, which can be confusing and demoralizing for borrowers who see their balances increase even though they are making payments. But subsidizing interest is also costly for the government.

To facilitate understanding of the treatment of uncovered interest in IDR plans, we first define a few key terms:

• “Capitalized balance” is the portion of the borrower’s loan balance that accrues interest. This is sometimes referred to as the “loan principal,” though it can include interest if that interest has capitalized (see below).
• “Accrued-but-unpaid interest” or “uncovered interest” is the excess of interest accrued in the current year above the payments made. For example, consider the borrower illustrated in Example 2 in Table 3 who enters repayment and has an income-based payment of $625. If their initial balance was $150,000 with an interest rate of 7%, the interest for the first year is $10,500, or about $875 per month. The payment is not enough to cover the interest accrued that year. The difference of $250 ($875-$625) is the “accrued-but-unpaid interest” for that period. Whether that interest is paid by the government (technically “not charged”), accumulates, or capitalizes depends on the plan and type of loan, as described below.
• “Accumulated interest” is interest that has accrued but is not part of the capitalized balance on which interest will be calculated in the next year. Depending on the type of loan and IDR plan, in the example above, some or all of the $250 of uncovered interest may accumulate. That means it is added to the borrower’s outstanding balance but will not be subject to interest in the next year (unless there is a capitalization event). That is, the interest in the next year will again be $10,500, even though the total loan balance is larger ($150,250). The capitalized balance (principal) and accumulated interest balance are tracked separately.
• “Capitalization” is when accumulated interest is added to the capitalized balance (principal) on which interest is calculated. A “capitalization event” is a condition that triggers capitalization. An earlier change to the student loan regulations, finalized by the Biden administration in November of 2022, eliminated capitalization events associated with IDR plans other than IBR. For example, when a borrower in IBR stops making payments based on income or leaves IBR, that will trigger a capitalization event. This was also true for borrowers in PAYE and REPAYE before the 2022 changes. Imagine the Example 2 borrower made payments of $625 for five years before getting a promotion that made their income too high to qualify to make payments based on income in IBR. In that case, they would have $1,250 in accumulated interest that would capitalize; going forward, interest would be calculated based on their new capitalized balance of $151,250. Capitalization can also occur if a borrower does not recertify their income, a potentially costly error.

Note that in this report we are discussing what happens after a borrower enters repayment, but the distinction between interest accumulation and capitalization is also important while borrowers are in school. Prior to July 2023 (the effective date for the changes to student loan regulations finalized in November 2022), interest on unsubsidized loans accumulated but did not capitalize while the borrower was in school; the accumulated interest capitalized when the borrower first entered repayment.¹⁴ The November 2022 regulation changes eliminated capitalization of interest on unsubsidized loans when a borrower first enters repayment, but interest still does capitalize when the borrower returns to repayment after a subsequent in-school deferment.

Table 4 summarizes the treatment of uncovered accrued interest in each of the IDR plans before and after the November 2022 regulations went into effect (the SAVE Rule retained the changes from the earlier rule). The legal status of these proposed changes is complex and some provisions are enjoined; here we explain what the treatment of uncovered accrued interest and capitalization rules were prior to any changes and what they would be if all of the Biden-era changes were implemented.

Both before and after the change, under IBR, PAYE, and REPAYE, the government pays uncovered interest on subsidized loans for up to three years from the time the borrower first enrolls in IDR (technically, the interest is “not charged”).¹⁵ For REPAYE, the government also pays half the uncovered interest for subsidized loans after the three year period and for unsubsidized loans as long as the borrower is enrolled in REPAYE.

Under what conditions does accumulated interest get added to principal—or capitalize? Before the changes in the November rule, accumulated interest capitalized if a borrower exited the IBR, PAYE, or REPAYE plans or had income high enough such that they stopped making payments based on income in IBR or PAYE (this latter provision is not relevant for REPAYE because payments are not capped in REPAYE). Capitalization could also occur if a borrower did not recertify income on time, since that was effectively equivalent to not making payments based on income in IBR and PAYE or leaving REPAYE. In PAYE and ICR, accumulated interest also capitalized annually until the balance was 110% of the original balance, after which uncovered interest accumulated but did not capitalize unless one of the other conditions just described applied (e.g., leaving the plan). 

The November 2022 changes made the capitalization rules more favorable for borrowers across all the plans except IBR, which did not change because its provisions are specified in statute. That change eliminated capitalization for not making payments based on income in PAYE and for leaving PAYE or REPAYE; the rule also eliminated capitalization events for failure to recertify. This means that there was, briefly, a version of REPAYE between original REPAYE and when it was replaced with SAVE by the SAVE Rule. In the SAVE plan, all uncovered interest is paid by the government, so there is never accumulated interest that could be capitalized.

Treatment of married couples

The SAVE Rule increased the number of married couples who could have lower payments by using the “married filing separately” tax filing status. When a married couple files a joint tax return, as most do, their combined income and combined student debt determine their combined payment under all IDR plans. If both spouses have student debt, the payment is credited to each spouse’s loan proportional to the balance. Under ICR, IBR, and PAYE plans, the spouse’s income is not considered when calculating a borrower’s payment if the couple files taxes separately; the payment is calculated separately for each spouse based only on their student loan debt and income.¹⁶ This can result in a lower combined payment, but taxes could be higher than if the couple filed separately. Couples need to consider the trade-off between lower loan payments and higher taxes. 

This potential benefit of filing separately was not available to married couples participating in the REPAYE plan before it was changed to SAVE. Married borrowers in REPAYE had to include their spouse’s income when calculating eligibility and payments even if they filed separately. Under the SAVE Rule, the treatment of married couples who file separately in SAVE (the new name for REPAYE) is the same as in the other IDR plans. That is, each spouse’s payment is calculated separately based only on their own income and loan balance. For borrowers in IBR or PAYE (where married couples could already ignore their spouse’s income if they filed separately) who switch to SAVE, the potential benefit of filing separately could go up or down. Married borrowers formerly in REPAYE—who have been transitioned to SAVE automatically—can now exclude their spouse’s income in calculating their loan payment if they file taxes separately, though the benefit of doing so depends on their situation.

IDR costs are difficult to predict and can be high

Federal student loan programs are different from private banking services; on average, student loans are at least somewhat subsidized by taxpayers. These subsidies can be substantial, vary across loan types and depend on what repayment plans are available. Historically, the subsidy rate was largest for subsidized Stafford loans, which are available only to lower-income undergraduate borrowers. For these loans, interest does not accrue while students are in school, and interest rates are lower. Other loan types such as Parent PLUS loans have actually generated net revenue for the government. The effect of introducing income-driven repayment plans on government costs for loan programs could, in theory, be positive or negative. On the one hand, some borrowers will be less likely to default and will repay more over a longer horizon if they have access to IDR. On the other hand, some borrowers will have part of their balance forgiven under IDR, and some of the plans subsidize some uncovered interest. The balance of these two factors determines whether IDR is cost-saving or cost-increasing.

The cost to the government of student loans has been consistently underestimated, largely because of changing economic circumstances and larger-than-anticipated responses from borrowers and institutions.¹⁷ Forecasting the cost of IDR programs has proven especially difficult. When the first IDR program was adopted in the 1990s it was expected to have minimal budget impacts, but costs have ballooned in recent decades. According to a 2022 GAO report, the Department of Education’s IDR cost estimates increased by approximately $70 billion since 2013 due to “updated assumptions about borrowers’ repayment plan selection” and another $68 billion due to “changes related to the underlying data and estimated income growth for borrowers in IDR plans.” What this means in practice is that when the Department of Education was able to use better data on incomes, expected payments dropped and more borrowers selected IDR plans than originally anticipated; both factors contributed to the substantial underestimate of costs.

The SAVE plan, if fully implemented, could increase the costs of the student loan program enormously. Cost estimates for the SAVE plan vary widely but are in the hundreds of billions over 10 years. The Department of Education’s July 2023 estimate suggested SAVE would cost $156 billion over 10 years, while the Congressional Budget Office (CBO) and Penn-Wharton expected much higher costs, $276 billion and $475 billion, respectively. These represent additional costs on top of the substantial cost of the already-existing IDR plans and are large relative to other federal spending on higher education. For example, the Pell Grant program, which provides financial aid for lower-income students, had an annual cost of $31.4 billion in 2023-24.

The high cost of the SAVE program is a direct result of the lower payments—and more loan forgiveness—offered to borrowers. According to analysis from the Urban Institute, only about 22% of bachelor’s degree recipients would be expected to repay their balances in full if SAVE is available, while about 59% of borrowers would be expected to repay their original balances under PAYE. 

While the exact cost of SAVE, if implemented, is uncertain, there is no question it would increase the level of subsidy required to sustain the student loan program substantially. 

What is the current status of IDR and what will happen next?

It’s complicated. Since July 2024, nearly 8 million borrowers enrolled in SAVE have been placed in interest-free forbearance. These borrowers cannot make payments that count toward loan forgiveness under IDR or PSLF programs, setting them back on their forgiveness timelines. Meanwhile, other borrowers struggling to access affordable repayment options face potential delinquency or default. Indeed, the number of borrowers behind on their payments or in delinquency surged in 2024.

The availability of IDR options and the online application has changed several times since last summer (and even when the application has been available it is not clear if they are being processed). After weeks of both paper and online IDR applications being completely unavailable in February, the federal IDR application is back online, but with a major catch—as of this writing, applications are being accepted but not processed. This latest development came in response to a lawsuit brought by the American Federation of Teachers (AFT), which argued that their members were being harmed by blocking access to the IBR plan—which is not contested in the ongoing litigation. The agreement to reopen the IDR application is for now temporary pending a fuller hearing.

In February, the Eighth Circuit sent the SAVE case back to District court with instructions to enjoin the entire SAVE Rule, suggesting the court believes the 1993 law (OBRA93) may not authorize loan forgiveness. Such a finding would call into question the legality not only of SAVE but also the PAYE and ICR programs, while acknowledging that IBR and PSLF were clearly authorized by Congress. Existing IDR plans vary significantly in terms of payments required and forgiveness timelines. Many borrowers who benefited from SAVE’s generous terms may find other IDR plans less advantageous or even unavailable depending on their specific circumstances.

Over 40% of federal student loan borrowers (approximately 12 million people) are enrolled in IDR plans, accounting for more than $700 billion in loan balances. Almost 8 million of those are enrolled in the SAVE plan and in interest-free forbearance—so not making payments—due to the litigation. For several years, borrowers have faced considerable uncertainty about what they will need to pay and when. Even without these disruptions, student loan repayment policy is extraordinarily complex, as demonstrated by the length of this explainer. This complexity undermines the ability of the government to administer IDR well and of borrowers to access critical relief. The current turmoil presents an important opportunity for Congress to simplify and strengthen IDR.

Whatever IDR system emerges once the SAVE litigation is resolved and Congress potentially enacts a new law governing IDR, transitioning to a new system will require massive coordination among the Department of Education, loan servicers, and borrowers. Communicating complex changes to millions of people, many of whom haven’t made payments since before the pandemic, will be a monumental task. A well-managed transition will be essential to rebuild trust in the federal student loan system and ensure borrowers can access the protections to which they’re entitled. This will be difficult to accomplish in the face of the Trump administration’s cuts to staffing at the already-understaffed Federal Student Aid office in the Department of Education.