Mandatory Spending

Function 500 - Education, Training, Employment, and Social Services

Limit Forgiveness of Graduate Student Loans

CBO periodically issues a compendium of policy options (called Options for Reducing the Deficit) covering a broad range of issues, as well as separate reports that include options for changing federal tax and spending policies in particular areas. This option appears in one of those publications. The options are derived from many sources and reflect a range of possibilities. For each option, CBO presents an estimate of its effects on the budget but makes no recommendations. Inclusion or exclusion of any particular option does not imply an endorsement or rejection by CBO.

Billions of Dollars 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2019-
  Savings Estimated Using the Method Established in the Federal Credit Reform Act
Change in Outlays  
  Increase payments under IDR plans -0.3 -0.7 -1.0 -1.2 -1.6 -1.9 -2.3 -2.7 -3.1 -3.3 -4.7 -17.9
  Extend repayment period for IDR plans -0.2 -0.4 -0.6 -0.8 -1.0 -1.3 -1.5 -1.8 -2.0 -2.2 -3.1 -11.9
  Increase payments and extend repayment perioda -0.5 -1.2 -1.7 -2.2 -2.8 -3.4 -4.0 -4.7 -5.4 -5.9 -8.3 -31.7
  Savings Estimated Using the Fair-Value Method
Change in Outlays  
  Increase payments under IDR plans -0.2 -0.6 -0.9 -1.1 -1.4 -1.7 -2.1 -2.5 -2.8 -3.1 -4.3 -16.4
  Extend repayment period for IDR plans -0.1 -0.3 -0.5 -0.6 -0.8 -1.0 -1.2 -1.4 -1.6 -1.7 -2.4 -9.2
  Increase payments and extend repayment perioda -0.4 -1.0 -1.5 -1.9 -2.4 -3.0 -3.5 -4.2 -4.8 -5.2 -7.3 -27.9

This option would take effect in July 2019.
By law, the costs of federal student loan programs are measured in the budget according to the method established in the Federal Credit Reform Act. The fair-value method is an alternative and is included in this table for informational purposes.
IDR = income-driven repayment.
a. If both alternatives were adopted, the total savings would be greater than the sum of the savings if the alternatives were individually adopted because of interactions between the two alternatives.


Federal student loans can be forgiven under certain circumstances. The federal government offers several income-driven repayment (IDR) plans in which borrowers make monthly payments for a certain period of time based on their income, after which the outstanding balance of their loans is forgiven. IDR plans do not impose a limit on the amount that can be forgiven. The Congressional Budget Office expects that the biggest benefits of those plans currently go to people who borrow to attend graduate or professional school, because those people tend to borrow larger amounts than do people who borrow for undergraduate studies.


This option includes two alternatives that would reduce loan forgiveness for borrowers who took out federal student loans to pay for graduate school, starting with loans made to new borrowers in July 2019.

The first alternative would increase the percentage of income above 150 percent of the poverty guidelines that graduate borrowers in IDR plans pay on loans to 15 percent, up from the current 10 percent in most plans. (The amount those borrowers pay is capped by the amount that would be required under the Standard Repayment Plan with a 10-year repayment period, so borrowers with sufficiently high income would pay less than 15 percent of their income.)

The second alternative would extend the repayment period from 20 years to 25 years for several IDR plans used by borrowers who take out loans to finance graduate school. (The percentage of income required for monthly payments and the length of the repayment period for borrowers with only undergraduate loans would continue to be 10 percent and 20 years, respectively.)

Effects on the Budget

When estimating the budgetary effects of proposals to change federal loan programs, CBO is required by law to use the method established in the Federal Credit Reform Act (FCRA). That approach uses accrual accounting—which, unlike cash accounting, records the estimated present value of credit programs' expenses and related receipts when the legal obligation is first made rather than when subsequent cash transactions occur. (Present value is a single number that expresses a flow of current and future payments in terms of an equivalent lump sum paid today and that depends on the rate of interest, or discount rate, that is used to translate future cash flows into current dollars.) FCRA accounting, however, does not consider all the risks borne by the government. In particular, it does not consider market risk—which arises from shifts in macroeconomic conditions, such as productivity and employment, and from changes in expectations about future macroeconomic conditions. The government is exposed to market risk because, when the economy is weak, borrowers default on their debt obligations more frequently, and recoveries from borrowers are lower. Under an alternative method, the fair-value approach, estimates are based on market values—market prices when they are available, or approximations of market prices when they are not—which better account for the risk that the government takes on. As a result, the discount rates used to calculate the present value of higher loan repayments under this option are higher for fair-value estimates than for FCRA estimates, and the savings from those higher repayments are correspondingly lower.

Because loan repayments under IDR plans would be expected to increase under this option, the government would face less risk on loans in those plans; however, in estimating the budgetary effects of this option, CBO did not decrease the fair-value discount rates to account for the anticipated decline in risk.

Under current law, the student loan program will generate $18 billion for the government from 2019 to 2028, according to the FCRA method, CBO estimates. Under the first alternative, the government would save an additional $18 billion over the same period, according to FCRA accounting. According to the fair-value method, over the same period, federal costs would be reduced from $212 billion to $196 billion, for a savings of $16 billion. Under either method, the annual savings grow over time, because each year the number of borrowers and volume of loans are projected to increase as more borrowers enter the repayment plans. (The numbers for savings and costs account only for mandatory costs—both subsidy and administrative costs—for direct student loans.)

Under the second alternative, CBO estimates, federal spending from 2019 to 2028 would be reduced by $12 billion, according to the FCRA method. According to the fair-value method, spending would be reduced by $9 billion.

If both alternatives were implemented, the total savings would be slightly greater than the sum of the savings if the alternatives were individually adopted because of interactions between the two alternatives.

Both alternatives would encourage prospective borrowers who use an IDR plan to limit their borrowing because the cost of repaying the loan would increase. Under the first alternative, the cost of repaying the loan could be as much as 50 percent higher than under current law. The second alternative would increase by 25 percent the number of payments made by affected borrowers—and because income tends to increase with work experience, adding more years of payments would probably increase the sums that borrowers would have to repay by an even larger percentage.

Accordingly, under both alternatives CBO expects the volume of loans in IDR plans would be reduced. Under current law, CBO estimates that 45 percent of the volume of the loans made to all student borrowers and about 55 percent of those made to graduate student borrowers will enter an IDR plan. Under this option, CBO estimates that by 2028, the volume of loans originated to graduate student borrowers who entered an IDR would be reduced by about 20 percent (to about 44 percent of the loans originated to graduate student borrowers) in the first alternative and by 15 percent in the second alternative.

There are several sources of uncertainty in the estimates associated with this option. CBO must project future enrollment, the number of students who will take out a government loan, and the future earnings of those borrowers under current law and under each of the two alternatives. To estimate the effects of the option, CBO must then predict how those borrowers would respond to increases in the effective cost of borrowing that would occur under either or both alternatives.

It is difficult to determine how savings would be affected by variations in the option. For example, increasing the share of income borrowers pay on their loans from 10 percent to 20 percent (rather than from 10 percent to 15 percent, as specified in the first alternative) would not double the savings under the first alternative. That is because, if loan repayments had to be a higher portion of their income, more borrowers would completely pay off their loans or switch to other types of repayment plans. Similarly, if the repayment period was increased by 10 years (rather than by 5 years as specified in the second alternative), the savings would not double.

Other Effects

An argument in favor of this option is that reducing the amount of student debt that is forgiven—either by increasing the amount of the monthly payment or by extending the repayment period—would reduce students' incentive to borrow and would encourage them to enroll in graduate programs whose benefits, in terms of improved opportunities for employment, justified the costs of the additional schooling.

A second argument in favor of this options is that it focuses on people who have borrowed for graduate studies, who often have relatively high income and are therefore more likely to be able to eventually pay back their loans. Under both alternatives, affected borrowers would pay back more of their loans than they otherwise would, and more of those borrowers would completely pay off their debt before the end of the repayment period. (Under either alternative, IDR plans would continue to forgive any amount that was not repaid, so debt relief would be provided to borrowers who, despite making regular payments for 20 years or 25 years, could not pay off their debt.)

An argument against this option is that it would increase the risk that students would not be able to repay their loans. The increased risk might lead some students to choose less graduate education or to forgo it altogether. Both alternatives would disproportionately affect prospective graduate students with fewer financial resources, such as those who come from low-income families. Such students would be less likely to attend graduate school and consequently would have lower future earnings; and if they chose to take out loans to attend graduate school, they would be likelier to have heavy student debt later in life.