2024-25067. Student Debt Relief Based on Hardship for the William D. Ford Federal Direct Loan Program (Direct Loans), the Federal Family Education Loan (FFEL) Program, the Federal Perkins Loan (Perkins) Program, and the Health Education Assistance ...

Table 1—Proposed Model Inputs (“Predictors”)
Past and Present Repayment Statuses.
Total amount of debt outstanding.
Past and present types of loans held, and amounts borrowed.
Year of loan disbursement.
Ratio of current loan balance to balances from 4 months prior.
Repayment plans in which borrower currently participates.
Payments made on student loans.
Scheduled payments on student loans.
Interest rate on loans.
Years in repayment.
Pell Grant receipt.
Adjusted Gross Income from the borrowers' first FAFSA.
Expected Family Contribution calculated from inputs on the FAFSA.
Parent education level reported on the FAFSA.
Dependent/independent status.
Borrower age.
Highest academic level reported for the borrower's loans.
Highest degree the borrower ever reported pursuing.
Graduation indicator.
Year of graduation, for those graduated.
Predominant degree of the school the student last attended or from which they last graduated.
Ownership type of the school the student last attended or from which they last graduated.
Cohort default rates of the school the student last attended or from which they last graduated.
Earnings and debt information from College Scorecard of the school the student last attended or from which they last graduated.
Note: The Department proposes to use loan repayment statuses that reflect the benefits provided by On Ramp and Fresh Start policies.

The proposed process for designing and refining the statistical model to determine which borrowers meet the hardship standard in proposed § 30.91(a) based on 80 percent likelihood of being in default (as described in proposed § 30.91(c)) within two years would be as follows. First, the Department would develop and validate the model using multiple two-year random samples of data on Department-held loans with data ranging from 2017 to February 2020. The Department proposes to use samples from this time period because it contains the most recent period of at least two years of uninterrupted repayment before the COVID-19 payment pause, and therefore should provide the most up to date predictions about the relationship between the predictors described above and observed default over a two-year period. We would use these data to estimate the extent to which the previously described explanatory predictors (displayed in Table 1) would predict whether a borrower was likely to be in default on a student loan in any quarter within two years and therefore would meet the hardship standard in proposed § 30.91(a).

The Department would then evaluate a variety of methods to include the “predictors” in the proposed model to create the most accurate predictions of likelihood of being in default. There are generally two forms that predictors can take in the source data. The first form is continuous, which means that the predictor can be any value within a range. For example, the amount of outstanding debt that a borrower has could take on dollar values from greater than 0 to the maximum amount of outstanding debt in our data. The second form is categorical, which are predictors that have a finite number of distinct groups ( e.g., type of higher education institution). We propose to scale continuous predictors by their means and standard deviations, but would also consider those same predictors without scaling, and as categorical variables defined with different types of cutoff values to create those categories. The Department would also consider additional statistical model specifications such as those that include interactions among individual predictors, the use of higher order polynomials, and those that generate estimates using different subgroups of the model. Among these approaches to including variables in the model, the Department would estimate the model using logistic regression as well as machine learning approaches, such as gradient boosted trees.^[78]

Next, to select the proposed model from among various potential specifications and options, we would evaluate the performance of the model using a distinct random hold out test sample of Department-held loans from the same time period as the training sample. In this step, to evaluate the performance of the model, we would calculate commonly used metrics, including measures of model fit, confusion matrices with a variety of threshold levels, standard metrics derived from the confusion matrices, and other performance metrics. ⁷⁹ ( print page 87147) Generally, these measures provide different ways of comparing observed outcomes to outcomes predicted by the model, and a model would be considered to perform better if it more accurately classified borrowers into those who will be in default and those who will not be in default.

The proposed assessment based on this model would produce a score for each borrower that accumulates the prediction related to the predictors included in the model for likelihood of being in default within two years. The scores would range from 0 percent to 100 percent. This score could be interpreted as an estimate of the probability that a borrower is in default within the next two years. We would use the score from the model to assist with identifying borrowers who were at least 80 percent likely to be in default on a student loan in any quarter within two years of the proposed regulations' publication date.

Once the regulations are finalized and implemented, this model would be used to conduct an individualized determination of whether each borrower fits within the hardship standard in proposed § 30.91(a) and therefore qualifies for a waiver under proposed § 30.91(c).

For purposes of this NPRM, we estimated which borrowers would have an 80 percent likelihood of being in default within the applicable two-year period using a 5 percent sample of Department-held loans as of April 2024. At the time of the publication of the NPRM, however, the Department will have access to additional data that could be used to refine the model. For example, in the data used for modeling in this NPRM, the Department has recent borrower repayment history only for about five months since the end of the payment pause. At the time of the publication of the NPRM, however, the Department will be able to observe recent repayment and engagement experiences over a longer time horizon through the date of the NPRM.

The Department proposes to measure this two-year window as of the publication date of the NPRM to preclude strategic behavior to increase the likelihood of receiving hardship relief by defaulting on loans. The reason for measuring the two-year window as of the publication date of the NPRM is because we are intent on providing relief as soon as possible once the NPRM is finalized, and because we are concerned that a longer period between finalizing the regulations and measuring the two-year window could create incentives for borrowers to attempt to strategically adjust their repayment behavior to be more likely to obtain a waiver.

The committee reached consensus on this regulatory provision.

§ 30.91(d) Process for Additional Relief

Statute: Section 432(a) of the HEA (20 U.S.C. 1082(a)) provides that in the performance of, and with respect to, the functions, powers, and duties, vested in him by this part, the Secretary may enforce, pay, compromise, waive, or release any right, title, claim, lien, or demand, however acquired, including any equity or any right of redemption. Section 468(2) of the HEA endows the Secretary with similarly broad and flexible powers with respect to loans arising under the Perkins program.^[80]

Current Regulations: None.

Proposed Regulations: Proposed § 30.91(d) provides that the Secretary may rely on data in the Secretary's possession that may have been acquired through an application or any other means to provide relief, including automated relief, based on criteria demonstrating that the borrower has experienced or is experiencing hardship.

Reasons: Proposed § 30.91(d) would clarify the procedures the Department could use to provide relief if the Secretary were to exercise the discretion under this section to issue waivers.

The pathway for discretionary relief under proposed § 30.91(d) is for the Secretary to assess a borrower's circumstances in a holistic manner, which may be based in part on an application submitted by the borrower, to determine if the borrower is experiencing or has experienced hardship. Proposed § 30.91(d) operates fully independently and separately from proposed § 30.91(c) and would therefore be fully severable.

The Department intends the “hardship” necessary to trigger relief under proposed § 30.91(d) to be a substantial harm. The Department interprets the hardship required for relief under proposed § 30.91(d) as: the borrower must be highly likely to be in default or experience similarly severe negative and persistent circumstances, and other options for payment relief would not sufficiently address the borrower's persistent hardship. The requirement that other payment relief options would not sufficiently address a borrower's persistent hardship would apply both to borrowers who meet the standard because the Department finds they are highly likely to be in default and to borrowers who meet the standard because the Department finds they are highly likely to experience similarly severe negative and persistent circumstances.

Default is one of the strongest indications that a borrower has not been able to use options available to avoid hardship in repaying their student loans, so the Department would use a standard related to default as one part of the hardship test for individual applicants under proposed § 30.91(d).

In addition, the Department would also have to determine that other options for payment relief under the HEA, including IDR plans and other forgiveness opportunities, are not sufficient for the borrower to avoid a high likelihood of being in default or similarly severe and persistent negative circumstances.^[81] To determine whether the borrower faces a persistent hardship, the Department would consider the factors described in proposed § 30.91(b).

The Department makes student loans to students with the expectation that they will be repaid according to the terms provided under the HEA and laid out in the Master Promissory note. The Department understands that many borrowers experience difficulty repaying their loans at some point in their repayment experience that necessitates relief from monthly payments calculated under the standard 10-year repayment plan. As discussed above, there are many options under the HEA available to borrowers who may experience difficulty repaying their loans. Relief options include the short-term use of deferment or forbearance options. These proposed regulations are not designed to supplant any of the options available to borrowers. Rather, these proposed regulations are designed as a safety valve for those borrowers who cannot receive sufficient relief to avoid hardship via payment relief options already in existence under the ( print page 87148) HEA. For the purposes of proposed § 30.91(d), the Department would consider the availability of the following payment relief options ^[82] to determine whether such options could sufficiently address the borrower's hardship: deferment or forbearance; forgiveness opportunities such as borrower defense discharge and TPD discharge, and income-driven repayment (IDR) plans.

A payment relief option would not be sufficient if it would not prevent the borrower from still experiencing a hardship related to the loan that makes them highly likely to be in default or experience similarly severe negative and persistent circumstances that substantially impairs their ability to fully repay the loan. For example, a borrower that is on an IDR plan with a $0 monthly payment might still be eligible for a waiver if the borrower would still be highly likely to experience similarly severe negative and persistent circumstances because they have a persistent hardship and lack the disposable income needed to fully repay the loan without jeopardizing their basic financial security over an extended period of time. In other words, the Department could determine that a payment relief option was not sufficient if it only temporarily delayed—but did not eliminate—the need to discharge some or all of the borrower's loans to sufficiently address the hardship.

The Department seeks to provide relief for individuals who are experiencing hardship without creating incentives for borrowers to strategically choose to cease making payments in order to qualify for relief. Proposed § 30.91(c) would prevent this strategic behavior by specifying the Secretary's discretion to provide one-time immediate relief based on a predictive assessment that would use the publication date of this NPRM as the beginning of the two-year period. There would be no future opportunity to change behavior and to obtain relief under proposed § 30.91(c).

For proposed § 30.91(d), the Department would address the risk of strategic behavior with a two-fold requirement that the borrower must be highly likely to be in default, or experience similarly severe negative and persistent circumstances, and that other options for payment relief would not sufficiently address the borrower's persistent hardship, including IDR plans, for those eligible. As a result, a borrower who is experiencing a high likelihood of being in default that they could avoid by enrolling in an IDR plan but has chosen not to enroll as an attempt at strategic behavior, would be extremely unlikely to receive relief under proposed § 30.91(d). In cases where a borrower who could find sufficient relief from hardship through an IDR plan applies for relief under proposed § 30.91(d), the Department would encourage that borrower to enroll in IDR, and that borrower would be unlikely to be eligible for a waiver under proposed § 30.91(d). Nor would a borrower who faces default simply because they have chosen not to make payments, without any evidence of experiencing hardship, receive relief under proposed § 30.91(d). These requirements would advance the goal of the proposed regulations and apply the standard of proposed § 30.91(a), providing relief in cases of genuine hardship.

Moreover, should the Secretary choose to exercise authority under these regulations, proposed § 30.91(c) would provide relief to the millions of borrowers who are experiencing hardship already, and in many cases who have lacked access to the full range of repayment options that will now be fully available going forward. Relief would only be available to individuals under proposed § 30.91(d) who experience hardship that is not sufficiently addressed by other options for payment relief and have a high likelihood of being in default or experiencing similarly severe negative and persistent circumstances.

One type of borrower eligible for relief under proposed § 30.91(d) would be a borrower who is already enrolled in an IDR plan but who is highly likely to default or experience similarly severe negative and persistent circumstances even with an IDR plan's payment protections. Although IDR plans take into account income and household size, there are borrowers who would still experience hardship related to their loans that could not be remedied through other means. Consistent with the factors described in § 30.91(b), the Secretary could consider, for example, whether an individual has unusually high expenses (such as nondiscretionary medical or housing expenses) such that they are highly likely to be in default, or to experience similarly severe negative and persistent circumstances.

In general, to determine whether an individual has such high expenses, the Department would look to established benchmarks, such as the Department of Housing and Urban Development measures of a “rent burdened” or “severely rent burdened” household that pays rent m 30 or 50 percent of household income, respectively, or the Internal Revenue Code standard allowing for deduction of health expenses in excess of 7.5 percent of Adjusted Gross Income. The Department would consider these expenses in the context of the borrower's overall financial resources, including income, assets, and debt.

As an example, consider an individual who is earning $80,000 a year, has $35,000 in loans, few assets, three dependents, and a monthly payment obligation of approximately $277 a month under an income-based repayment plan. That obligation would not ordinarily lead to hardship. However, in this example, the individual lives in a high-rent area and pays the typical rent of $2,300 for a one-bedroom apartment (more than 30 percent of their income or “rent burdened” under the HUD standard) and has a dependent that requires medication and treatment for a chronic health condition that costs $1,600 per month (well in excess of 7.5 percent of AGI). In order to pay for these expenses in addition to other essentials, like food and transportation, the borrower is in default or is on the verge of being in default after missing seven months of payments. If this borrower demonstrated that they did not have the assets to avoid being in default, and that their circumstances were unlikely to improve within a period of time, then they could potentially receive relief under this provision.

There may also be cases where an individual can demonstrate hardship even in the absence of a payment burden (such as when a borrower has a $0 IDR payment). For example, a borrower may be able to show that they meet the standard for hardship described above if they can show that, even with a $0 IDR payment, the existence of the debt itself causes the required hardship. As stated above, a borrower on an IDR plan with a $0 monthly payment may also be able to show that they are still highly likely to experience similarly severe negative and persistent circumstances because they have a persistent hardship and lack the disposable income needed to fully repay the loan without jeopardizing their basic financial security over an extended period of time. The Department has also included a directed question regarding the circumstances in which this might occur.

Relief under proposed § 30.91(d), whether based on data “acquired through an application or by any other means” would be assessed on a holistic basis to determine whether the standard described above for proposed § 30.91(d) ( print page 87149) is met.^[83] The Department interprets the word “automated” as used in proposed § 30.91(d) to mean relief that the Secretary may grant based on information already in the Department's possession rather than acquired through an application. The Department anticipates that the number of borrowers for whom the Department would possess sufficient information to conduct the holistic review without data acquired from an application would be small. The Secretary would not be able to use a default risk model such as a model similar to the one described in § 30.91(c) in order to provide relief under proposed § 30.91(d). A borrower could only receive a waiver without an application under proposed § 30.91(d) if the Department's holistic review of the borrower's data satisfied the same stringent standard that the Department would apply for application-based relief. Such cases would be considered rare since the data that the Department possesses would have to sufficiently establish eligibility including that other options for payment relief would not sufficiently address the borrower's persistent hardship and would also need to sufficiently distinguish such borrowers from otherwise similar borrowers who would not be deemed to qualify for relief.

The Department recognizes that to meet this stringent standard, the Department would need data that would allow the Secretary to determine whether a borrower meets proposed § 30.91(d)'s standard. The Department notes that the Secretary would need to expand or refine data elements in the future to provide relief to borrowers under proposed § 30.91(d) without an application because, at the time of preparing this NPRM, the Department does not currently have sufficient data available to determine whether a borrower meets the eligibility standard. We seek feedback from the public about the type of data that could be used for relief without an application under proposed § 30.91(d), and how those data could be obtained.

As discussed throughout this NPRM including in the Regulatory Impact Analysis, the proposed process under § 30.91(d) would likely involve detailed reviews of applications submitted by borrowers or other data already in the Department's possession. We anticipate that the processes under § 30.91(d) would take time to implement following the publication of a final rule, including developing an application, producing clarifying guidance, and hiring and training staff. Given the administrative costs associated with this process, we also anticipate that the volume of applications the Department would be able to process would be low at first and would be dependent on the amount of funding received by FSA through the annual appropriations process. Therefore, depending on the number of applications, it would take time for the Department to make waiver determinations on a borrower's individual application, and the Department would not be in position to guarantee a response within a specific period. As a result, borrowers should anticipate continuing to make payments while their application is pending.

The committee reached consensus on this section.

Regulatory Impact Analysis

Executive Orders 12866 (as Modified by 14094) and 13563

Under Executive Order 12866, the Office of Management and Budget (OMB) must determine whether this regulatory action is “significant” and, therefore, subject to the requirements of the Executive Order and subject to review by OMB. Section 3(f) of Executive Order 12866, as amended by Executive Order 14094, defines a “significant regulatory action” as an action likely to result in a rule that may—

(1) Have an annual effect on the economy of $200 million or more (adjusted every 3 years by the Administrator of OIRA for changes in gross domestic product), or adversely affect in a material way the economy, a sector of the economy, productivity, competition, jobs, the environment, public health or safety, or State, local, territorial, or Tribal governments or communities;

(2) Create a serious inconsistency or otherwise interfere with an action taken or planned by another agency;

(3) Materially alter the budgetary impacts of entitlements, grants, user fees, or loan programs or the rights and obligations of recipients thereof; or

(4) Raise legal or policy issues for which centralized review would meaningfully further the President's priorities, or the principles stated in the Executive Order, as specifically authorized in a timely manner by the Administrator of OIRA in each case.

This proposed regulatory action would have an annual effect on the economy of $200 million or more. Table 4.1 in this regulatory impact analysis (RIA) provides an estimate of the net budget effects of each provision of these proposed regulations. We also provide estimates of the administrative costs for these provisions. Because the net budget effect is larger than $200 million a year, this proposed regulatory action is subject to review by OMB under section 3(f) of Executive Order 12866 (as amended by Executive Order 14094). Notwithstanding this determination, we have assessed the potential costs and benefits, both quantitative and qualitative, of this proposed regulatory action and have determined that the benefits would justify the costs.

We have also reviewed these regulations under Executive Order 13563, which supplements and explicitly reaffirms the principles, structures, and definitions governing regulatory review established in Executive Order 12866. To the extent permitted by law, Executive Order 13563 requires that an agency—

(1) Propose or adopt regulations only on a reasoned determination that their benefits justify their costs (recognizing that some benefits and costs are difficult to quantify);

(2) Tailor its regulations to impose the least burden on society, consistent with obtaining regulatory objectives and considering—among other things and to the extent practicable—the costs of cumulative regulations;

(3) In choosing among alternative regulatory approaches, select those approaches that maximize net benefits (including potential economic, environmental, public health and safety, and other advantages; distributive impacts; and equity);

(4) To the extent feasible, specify performance objectives, rather than the behavior or manner of compliance a regulated entity must adopt; and

(5) Identify and assess available alternatives to direct regulation, including economic incentives—such as user fees or marketable permits—to ( print page 87150) encourage the desired behavior, or provide information that enables the public to make choices.

Executive Order 13563 also requires an agency “to use the best available techniques to quantify anticipated present and future benefits and costs as accurately as possible.” The Office of Information and Regulatory Affairs of OMB has emphasized that these techniques may include “identifying changing future compliance costs that might result from technological innovation or anticipated behavioral changes.”

We are issuing these proposed regulations upon on a reasoned determination that their benefits would justify their costs. In choosing among alternative regulatory approaches, we selected those approaches that, in the Department's estimation, best balance the size of the estimated transfer and qualitative benefits and costs. Based on the analysis that follows, the Department believes that these proposed regulations are consistent with the principles in Executive Order 13563.

We have also determined that this regulatory action would not unduly interfere with State, local, territorial, and Tribal governments in the exercise of their governmental functions.

As described in OMB Circular A-4, we compare the proposed regulations to the current regulations. In this regulatory impact analysis, we discuss the need for regulatory action, the summary of key proposed provisions, potential costs and benefits, net budget impacts, and the regulatory alternatives we considered.

Elsewhere in this section under the Paperwork Reduction Act of 1995, we identify and explain burdens specifically associated with information collection requirements.

1. Congressional Review Act Designation

Pursuant to the Congressional Review Act ( 5 U.S.C. 801 et seq.), the Office of Information and Regulatory Affairs designated that these regulations are covered under 5 U.S.C. 804(2) and (3).

2. Need for Regulatory Action

These proposed regulations describe circumstances in which the Secretary might exercise the longstanding discretionary waiver authority under sections 432(a)(6) and 468(2) of the HEA to waive all or part of a Federal student loan held by the Department to provide relief to a borrower who has experienced or is experiencing hardship.

Addressing the issue of hardship is critical in building a strong student loan program. While the Department currently offers a number of options for payment relief, including IDR plans and forgiveness opportunities, the complexity of borrowers' lives may lead to hardships that are not sufficiently addressed by these existing options such that these hardships are likely to impair their ability to repay a Federal loan in full or cause the anticipated costs of collecting the loan to exceed the likely benefits of continued collection of the entire debt. The Department frequently hears from borrowers about a range of these situations, such as borrowers facing significant unexpected expenses caring for loved ones with chronic illnesses, living with disabilities that limit but do not eliminate work opportunities, dealing with financially burdensome medical bills, or fearing that they will struggle to repay loans as they prepare to exit the workforce by retiring.

Sections 432(a)(6) and 468(2) of the HEA provide the Secretary with discretion to address these situations. Issuing a clear regulatory framework to address hardship would better inform the public about how the Secretary might exercise this waiver authority by considering a set of factors and standards that would allow for the consistent treatment of similarly situated borrowers, while also recognizing the inherent variability of each borrower's particular situation.

As further explained in the preamble, these proposed regulations specify two different pathways by which the Secretary may exercise discretion to grant relief to borrowers experiencing hardship: a pathway for immediate relief using a “predictive assessment” (proposed § 30.91(c)) and a separate pathway for additional relief based on a “holistic assessment” of information submitted by the borrower through an application or acquired by any other means (proposed § 30.91(d)).

For the immediate relief described in proposed § 30.91(c), the Secretary proposes to assess whether the borrower meets the hardship standard by determining whether a borrower has at least an 80 percent chance of being in default within two years, using a predictive assessment based on data already in the Department's possession to analyze the hardship factors in proposed § 30.91(b). The use of a predictive assessment would allow the Department to recognize situations in which similarly situated borrowers face comparable challenges likely to impair their ability to fully repay the loan, or that would cause the costs of collecting the loan to outweigh the benefits.

Further, this predictive assessment under proposed § 30.91(c) would be based on data in the Department's possession and therefore would promote efficiency and reduce administrative costs. For example, the predictive assessment would promote efficiency because it would eliminate the need for individual borrowers to complete applications and for the Department to process those applications. Furthermore, using this predictive assessment would also avoid the risk that many borrowers in need of relief would miss out on the opportunity for relief because they are unaware of the need to apply or be unable to overcome the administrative challenges of applying.

While the predictive assessment described in proposed § 30.91(c) would reduce administrative burden for both borrowers and the Department and could be implemented quickly because it would rely on data the Department already has, it would not be able to capture all borrowers who are experiencing hardship that satisfies the proposed standard for waiver. One reason is that the Department currently does not have data on several of the factors described in proposed § 30.91(b), such as information on debts not owed to the Department or a borrower's expenses for caretaking, housing, and other factors, which could be burdensome for some borrowers.

Therefore, the Department would also need the discretionary option of receiving additional information from borrowers through an application process so that the Department could conduct a holistic assessment of the borrower's circumstances to determine whether the borrower meets the applicable standard for hardship under proposed § 30.91(d). As described in the preamble, under this process, the Department would determine whether: (i) the borrower is highly likely to be in default or experience similarly severe and persistent negative circumstances, and (ii) other options for payment relief would not sufficiently address the borrower's persistent hardship. The Department could also make this determination based on information already in the Department's possession, or a combination of information already in the Department's possession or received through an application. This application-based pathway would be important to give borrowers the opportunity to provide additional information and data that might not be captured in existing data systems to which the Department has access, or to describe any additional relevant circumstances. ( print page 87151)

Overall, the relief contemplated in these proposed regulations would provide important support in situations where a borrower's investment in postsecondary education fails to yield the potential benefits from completing such an education. Generally, postsecondary education provides significant individual and societal benefits. Earning a postsecondary credential typically provides individuals with a range of personal benefits in the labor market, including higher income and lower unemployment risk.^[84] In addition to individual benefits related to earnings and employment, additional education can provide a host of individual benefits, including greater access to health insurance, increased job satisfaction, and overall happiness.^[85] Increasing levels of postsecondary attainment also have spillover benefits for communities and society, including benefits to those who never attended or completed postsecondary education. For example, researchers have documented that wages of non-college graduates rise when the supply of college graduates increases.^[86] Increases in education are also linked to higher civic participation, reduced crime, and improved health of future generations.^[87]

For some borrowers, financing an education does not lead to individual net benefits. Loans taken out for postsecondary education commonly take a decade or more to repay, and borrowers may never reach sustained periods of income security necessary to afford and manage their loans. This could be because they never complete their program and therefore never receive a meaningful earnings return, or they may lose the income security attendant to an education when they face unexpected and significant life events outside their control, such as the need to care for sick dependents, expensive medical problems, or the onset of disabilities that limit work opportunities.

These proposed regulations would establish a framework for the Secretary to exercise the discretionary waiver authority in a consistent and transparent manner. This framework would fill existing gaps in relief that are otherwise available from the Department to assist borrowers with managing repayment of their loans. The Department's existing avenues for payment relief, for example, may be insufficient to assist older borrowers with high student loan debt burdens at increased risk of default and resulting financial insecurity, or those with significant obligations expenses for child or dependent care. Therefore, these proposed regulations would specify the Secretary's authority to grant relief where the Secretary determines the borrower's hardship impairs the borrower's ability to fully repay loans or makes collecting the loans unjustifiably costly.

Summary of Proposed Key Provisions

Table 2.1 below summarizes the proposed provisions in the NPRM.

Table 2.1—Summary of Proposed Provisions
Provision	Regulatory section	Description of proposed provision
Standard for waiver due to likely impairment of borrower ability to fully repay or undue costs of collection	§ 30.91(a)	Provides that the Secretary may waive up to the outstanding balance of a Federal student loan held by the Department if the Secretary determines that the borrower has experienced or is experiencing hardship related to such a loan such that the hardship is likely to impair the borrower's ability to fully repay the Federal government or the costs of enforcing the full amount of the debt are not justified by the expected benefits of continued collection of the entire debt.
Factors that substantiate hardship	§ 30.91(b)	Provides a non-exclusive list of factors the Secretary could consider in determining whether a borrower meets the standard for waiver based on hardship.
Immediate relief for borrowers likely to default	§ 30.91(c)	Provides that the Secretary may consider the borrower's factors indicating hardship described in proposed § 30.91(b) to exercise discretion to waive all or some of outstanding loans held by borrowers who the Secretary determines have experienced or are experiencing hardship such that their loans are at least 80 percent likely to be in default in the two years after the publication of the proposed regulations.
Process for additional relief	§ 30.91(d)	Provides that the Secretary may rely on data obtained from an application or by any other means, or potentially a combination or both, to provide relief for borrowers who are highly likely to be in default or to experience similarly severe and persistent negative circumstances, and other payment relief options do not sufficiently address the borrower's persistent hardship.

3. Discussion of Costs, Benefits and Transfers

Overall, waivers that the Secretary grants under the proposed regulations would result in costs in the form of transfers from the Federal Government to student loan borrowers. The size of these transfers would vary based upon the number of borrowers who the Secretary determines are at least 80 percent likely to be in default and, therefore, eligible for waiver under proposed § 30.91(c). It would also depend on the number of borrowers who are approved for waivers under proposed § 30.91(d). The Department believes that these transfers would ( print page 87152) provide significant benefits to borrowers in the form of waiving their obligation to repay some or all of their Federal student loan debt. The Department would also see benefits from waivers granted on loans that are unlikely to be repaid in a reasonable period, which would prevent or reduce costly collection efforts.

The transfers to borrowers in the form of waivers could result in costs to the Federal Government and in turn taxpayers, to the extent that borrowers receiving waivers might otherwise have repaid the loan in part or whole, or the financial costs of collecting the loan might have proved less than the benefits of collection. The proposed rules would also result in administrative expenses for the Department to implement these provisions. When considering all these factors, the Department believes that the benefits from these proposed regulations would outweigh the costs.

What follows is a description of the data used to create estimates in this RIA, followed by a discussion of the costs, benefits, and transfers for each of the distinct regulatory provisions.

Data Used in This RIA

This section describes the data used in the RIA. To generate information about the expected number of borrowers who would be eligible to receive relief under these proposed regulations, the Department relied upon non-public records contained in the administrative data the Department uses to administer the title IV, HEA programs.

The primary data used in the RIA to estimate the number of borrowers who could potentially qualify for a waiver under proposed § 30.91(c) is a 5 percent random sample of the Federal Department-held student loan portfolio with at least one open title IV, HEA student loan as of April 30, 2024. We are using a random sample including over 2 million borrowers, but we present all estimates in the analyses below in terms of the full Department-held student loan portfolio. The data we use for modeling in the RIA are stored in the National Student Loan Data System (NSLDS), maintained by the Department's Office of Federal Student Aid. The Department determined that a sample of this size was appropriate to provide reasonable estimates of the impact of the proposed regulation. A sample of this size is similar to what the Department uses in other modeling, such as for the annual President's budget and for the net budget impact modeling in this RIA.

Analysis of Costs, Benefits, and Transfers for Each Proposed Regulatory Section

The sections that follow contain a discussion of the costs, benefits, and transfers for the different proposed regulatory provisions if the Secretary chooses to grant waivers under such provisions. We separately discuss the relief potentially provided under proposed § 30.91(c)'s pathway for “immediate relief” and proposed § 30.91(d)'s pathway for “additional relief” based on an application or information already in the Secretary's possession, or both, because those provisions would represent different pathways for the Secretary to exercise discretion to grant a waiver for a borrower. Implementation of each of these provisions would include administrative costs for the Department. Because these administrative costs generally would represent baseline implementation expenses, we provide a separate discussion of administrative costs at the end of this part of the RIA.

We do not include a discussion of proposed § 30.91(a) or (b), which would establish the standard for hardship and the indicators to be considered in determining if a borrower is facing hardship, because these provisions do not describe discretionary pathways for relief that may result in costs, benefits, and transfers.

§ 30.91(c) Immediate Relief for Borrowers Likely To Be in Default

Should the Secretary choose to grant waivers under proposed § 30.91(c), the proposed regulations would result in costs in the form of transfers from the Department to borrowers through waiver of outstanding debt to the Department. Waiving these amounts would eliminate future payments by these borrowers to the Department, which is a cost to the Federal Government and, by extension, to taxpayers. The extent of transfers and their associated cost would vary depending on the eligible borrower's amount of outstanding debt, loan type(s), age of the loan, likelihood of repayment, and other factors. The proposed regulations would also result in administrative expenses for the Department to implement these provisions. When considering all these factors, the Department believes that the benefits from these proposed regulations would outweigh the costs.

Borrowers who are in default are likely to have repeated instances of default or be in default for a protracted time. Department data show that almost all of those who were likely to be in default in the next two years had struggles with loan repayment in the past, as evidenced by instances of current or prior default, or of payment delinquency. Acknowledging past hardship recognizes that previous periods of hardship may have current and future consequences for a borrower. For example, a borrower who struggled to repay their loans may have seen their balance increase in size such that full repayment of that greater amount is no longer feasible. The likelihood of prior or persistent repayment struggles observed in Department data is similar to that found in other data. A Federal Reserve Bank of Philadelphia survey of borrowers demonstrated that most of the individuals who anticipated difficulties making loan payments after the payment pause ended also reported making no or partial payments prior to the pandemic forbearance.^[88] These data also suggest that there is greater prevalence of longer-term or repeated defaults among communities with greater shares of Black and Hispanic residents, and that student loan default commonly co-occurs with delinquency and collections on other types of debt, such as medical debts and utilities.^[89] These distributional effects reflect underlying differences in income, completion status, and other factors that correlate with student loan struggles.^[90]

In addition, many of the borrowers who might receive a waiver under proposed § 30.91(c) have been in repayment for an extended time. For instance, based on analysis of Department data, in 2022, more than 1 million borrowers held loans that had been in default for at least 20 years. These borrowers could have been subject to negative credit reporting, wage garnishment, tax refund offset, and even litigation. If these loans are still outstanding after all this time, notwithstanding the availability of those powerful collection tools, the odds that they would be fully repaid in a reasonable period are unlikely.

Older loans are also likely to be held by older borrowers. Analysis of Department data indicates that almost a ( print page 87153) quarter of borrowers who would receive a waiver are over 55 years old. The older the borrower, the greater the likelihood that they will stop working prior to successful repayment. Forty-one percent of non-Parent PLUS borrowers 62 years of age and older with an open loan have held their student loans for more than 20 years, and 30 percent of borrowers 62 years of age and older with an open loan have held their student loans for more than 25 years.^[91] Waiving such loans would not create significant costs for the Government in the form of transfers because the Department is unlikely to receive significant additional payments from a retired borrower.

About two-thirds of borrowers who may receive a waiver received a Pell Grant in our data, but this number is likely an underestimate because Pell Grant status is unavailable for most borrowers who entered repayment on their last loan before 1999.

Borrowers would receive significant benefits from no longer having to repay loans, and the Federal Government would also see benefits from conserved administrative costs through discontinued servicing or collecting on loans that the Department does not expect to be repaid in full.

As noted above, these transfers would create some costs for the Federal government and, by extension, taxpayers. However, as discussed above, these waivers would generally affect loans with lower expected repayment rates (therefore have a low likelihood of generating funds for the Federal government), and any limited lost revenue from waiving some of the Department's worst-performing loans would likely be outweighed by significant individual and social economic benefits to the borrower. Specifically, the waivers proposed here would provide borrowers facing hardship with a greater ability to avoid financial distress, and potentially lower delinquency rates on other types of debt, promote consumption (which can benefit the economic wellbeing of their communities), improve access to credit, and may reduce reliance on other forms of the Federal safety net.^[92]

To estimate the number of borrowers we would expect to be eligible for relief under proposed § 30.91(c), we followed the process described for implementing proposed § 30.91(c) above, where we used predictors that correspond to the 17 factors described in proposed § 30.91(b) to predict whether borrowers were at least 80 percent likely to be in default on a student loan in any quarter for the subsequent two years after the NPRM's publication. For the purposes of the NPRM, we used a 5 percent sample of Department-held loans as of April 2024.

Should the Secretary choose to exercise authority under these regulations, we estimate that approximately 6.0 million borrowers would be eligible to receive relief under proposed § 30.91(c). This estimate is based on output of the proposed model developed to estimate the likelihood that a borrower would have been in default within two years.

The estimate of borrowers who may receive waivers under this provision uses data as of April 30, 2024, and calculates the two-year measurement window as of April 30, 2024. The Department chose April 30, 2024, because it was the most recent comprehensive dataset available to the Department at the time that the Department was developing the proposed model for this NPRM. The borrower portfolio may change between April 2024 and the publication of the NPRM, both in terms of its composition ( i.e., which borrowers are in the portfolio) and in the borrowers' circumstances ( e.g., the loans held by borrowers and outstanding debt amount may change between April 30, 2024 and the publication of the NPRM). It is not clear what the substantive effects of such changes would be, as they could drive the model's outcomes in different directions. Estimates presented in the NPRM also do not include potential overlap with relief that would be provided by any proposed rules that are not yet final as of the publication of this NPRM, or of waivers through other provisions that were not yet implemented as of April 30, 2024.

§ 30.91(d) Process for Additional Relief

Borrowers would benefit from any waivers granted by the Secretary under proposed § 30.91(d)'s pathway for additional relief based upon a holistic assessment of information already in the Secretary's possession or submitted by the borrower through an application process, or both in conjunction, to determine whether: (i) the borrower is highly likely to be in default or to experience similarly severe and persistent negative circumstances, and (ii) other payment relief options would not sufficiently address the borrower's persistent hardship. As further described in this NPRM's preamble, such waivers would address challenges that these borrowers face while trying to repay their loans. While this approach would provide overall financial benefits, the specific benefits for borrowers who receive a waiver would vary depending on the nature of their qualifying hardship. Waivers granted under proposed § 30.91(d) would also create administrative costs for the Department to implement, which are discussed at the end of this subsection of the RIA.

Consider several examples of borrowers who may receive waivers based on a holistic assessment of the factors in proposed § 30.91(b). For example, a borrower whose qualifying hardship is a result of advanced age, having an old loan, and no longer working would benefit from no longer having to manage a loan payment in their final years of life. If they were in default, they could also potentially see an increase in the total amount of Social Security benefits they could retain since they would not be at risk of having amounts offset. By comparison, a borrower who is facing hardship due to having extensive expenses caring for an elderly relative could also accrue benefits, but in a different form, such as being able to better afford necessary care for that individual, including potentially paying for better services for that relative. Since the precise facts that support waiver under proposed § 30.91(d) would vary across individual borrowers' circumstances based on a holistic assessment of their factors in proposed § 30.91(b), the specific benefits of waiver would vary. But in general, to the extent that the hardship results in the borrower being overburdened by necessary expenses, the waiver would help a borrower better afford those expenses while maintaining basic financial security and also greatly reduce or eliminate their risk of experiencing the substantial harms of default, or other similarly severe negative and persistent circumstances.

Waivers granted under proposed § 30.91(d) would create costs to the government in the form of transfers to student loan borrowers. These costs would also accrue to taxpayers. However, we believe the benefits would exceed these costs. As discussed, the Secretary may provide a waiver under ( print page 87154) proposed § 30.91(d) only after determining: (i) the borrower has experienced, or is experiencing, hardship such that the borrower is highly likely to be in default or to experience similarly severe and persistent negative circumstances, and (ii) other payment relief options do not sufficiently address the borrower's persistent hardship. Therefore, borrowers who may receive waivers are those with lower-than-expected repayments who are highly likely to struggle with repaying their loans. By contrast, as described above, the benefits to borrowers could be significant. And such waivers could provide benefits to the government as well. The Department would no longer pay to collect on or service loans that are highly unlikely to be repaid. And to the extent borrowers are facing hardship while receiving other Federal benefits, such as Social Security, no longer having those amounts at risk of being offset would allow broader Federal benefits to better achieve their intended purposes, such as keeping senior citizens out of poverty.

Estimating the number of borrowers who could, at the Secretary's discretion, be approved for relief under proposed § 30.91(d) depends on assumptions about: (1) the number of borrowers who have characteristics that are likely to make them eligible for relief based on the Department's holistic assessment of their circumstances, and (2) the share of borrowers who are potentially eligible who would actually apply. At the end of this section, to inform the estimates of administrative costs, we discuss further assumptions about the number of borrowers who would apply, but who we would expect would not be approved for discretionary relief. In the sections below, we describe the information the Department considered to reach estimates used in this NPRM. Recognizing data limitations and that there are no perfect historical analogs that could inform estimates with perfect precision, we included a directed question that solicits feedback and input from the public about other data or information that could be used to improve and refine estimates.

First, we consulted available Department data on borrowers and national survey data related to student debt holders to inform the number of borrowers who have characteristics that are likely to make them eligible for relief based on the Department's holistic assessment of their circumstances. These borrowers would need to have indicators that show they are highly likely to be in default or experience similarly severe negative and persistent circumstances, that would not be sufficiently addressed by other options for payment relief. In addition, the Department anticipates that proposed § 30.91(c) would, at the Secretary's discretion, be implemented first and that such relief would likely be sufficient to address the hardship of a borrower who receives such relief. Therefore, because we do not want to double count borrowers, the estimate for proposed § 30.91(d) discussed below does not include borrowers who would be expected to receive full relief under proposed § 30.91(c).

As a starting point the Department consulted economic studies of individuals experiencing poverty. We believe estimates of persistent poverty provide an important perspective on borrowers who may have enduring negative economic conditions, even if it is not the perfect comparison. Poverty by itself may not lead to relief under proposed § 30.91(d), but people in poverty often face challenges such as not being able to afford necessary expenses. In addition, other available payment relief options might address episodic spells of poverty. As described earlier, the Department expects that a borrower would need to have indicators showing a high likelihood of persistent hardship rather than a short-term hardship to receive relief under this provision. On the other hand, it is also possible that borrowers could be facing persistent hardship and receive relief, even if they are not considered in poverty. Even acknowledging these limitations, we believe estimates of persistent poverty are a reasonable consideration for estimates under proposed § 30.91(d).

Studies suggest that a meaningful, but small, share of the population experience persistent poverty, defined in many studies as having an income below the Federal Poverty Level. For example, longitudinal studies of families experiencing poverty, using the Panel Study of Income Dynamics suggest that between about 3 to 5 percent of adults are exposed to instances of poverty that last five years or more across their adult lifetimes.^[93] Therefore, if we applied the persistent poverty rate of 3 percent to 5 percent to the number of ED-managed borrowers in the current portfolio, we might expect somewhere between 1 and 2 million borrowers in the current portfolio to experience persistent economic hardship at some point in their adulthood that would meet the eligibility requirements under § 30.91(d).

This range of 1 to 2 million borrowers from the current portfolio is corroborated through other data sources. In Department data, as of December 2023, there were about 9 million borrowers who were recorded as past due in their payments.^[94] Solely being behind on student loan payments would not lead to eligibility for a waiver under § 30.91(d). Therefore, the Department also reviewed information that was reported in the Federal Reserve Board's Survey of Household Economics and Decisionmaking (SHED).^[95] The SHED provides information about borrowers and their personal finances, and indicates whether borrowers who were behind on student loan payments also reported some other condition that could indicate hardship, such as being unemployed or underemployed due to a health or medical limitation or a disability or living with parents or adult children to provide help with child or medical care.^[96] Those data indicate that about 13 to 25 percent of borrowers who are behind on student loan payments also reported one of these other indicators. Applying those percentages to the current portfolio of borrowers implies that between 1 to 2 million borrowers may be experiencing some ( print page 87155) substantial economic hardship and report being behind on payments. We would expect that borrowers reporting a greater number of economic challenges in survey data might be more indicative of the type of hardship that would qualify for a waiver under § 30.91(d). In the SHED data, about 40 percent of the student loan borrowers who are behind on payments experience both conditions described above, which would imply an estimate of the number of borrowers in the current portfolio in the range of 0.4 to 0.8 million borrowers. These ranges are based on a single point in time. Under § 30.91(d), borrowers would need to face hardship consistently in order qualify for a waiver, so estimates based on an isolated observation likely overestimates the number of borrowers who are facing persistent challenges on these factors. On the other hand, a point in time observation would not count borrowers who may experience struggles in the future, even if not showing such markers at the time of the survey.

For our base assumptions, we take the high end of the range suggested by available evidence and assume that about 2 million borrowers from the current portfolio would potentially have characteristics that might make them highly likely to be in default or experience similarly severe negative and persistent circumstances at some point in their remaining repayment. As we discuss later, other payment relief options might sufficiently address the hardship for some of these borrowers, not all borrowers who might be eligible will apply for relief, and most borrowers will not have sufficient data already in the Department's possession that would be necessary for any non-application-based relief.

We assume a take up rate of 75 percent among the 2 million borrowers from the current portfolio that we estimate would be potentially eligible for relief. This take up rate is blended across borrowers who might meet the standard for relief under proposed § 30.91(d), described above, based on the Department conducting holistic assessments that may either rely on data already in the Secretary's possession, data submitted by the borrower through an application, or a combination of both. This borrower estimate assumes that, as noted above, the Department would need to expand or refine data elements in the future to provide relief to borrowers under proposed § 30.91(d) without an application because, at the time of preparing this NPRM, the Department does not have sufficient data available to determine whether a borrower meets the eligibility standard. Borrowers for whom the Department possesses sufficient information to conduct the holistic review without data acquired from an application would not need to apply (implying a take up rate of 100 percent), but the Department anticipates that the number of such borrowers based on future data matches or data collections would be small. Ample evidence suggests that borrowers do not always apply for benefits for which they are eligible for many reasons, including because of the burden associated with application and lack of knowledge about the benefits. Evidence from other settings—none being perfect analogies—including from SNAP, the Earned Income Tax Credit (EITC), and Unemployment Insurance, suggest a range of take-up rates that differ across benefit amounts, salience, and eligible populations.^[97] Many of the programs with high take-up rates, such as SNAP and EITC, which have take-up rates at or above 80 percent, are well known and have been around for a long time, and some programs have infrastructures that help beneficiaries apply. Other researchers have reported take up of Unemployment Insurance of 42 percent to 55 percent.^[98] There are reasons to believe that the take-up rate for forgiveness proposed under § 30.91(d) could be lower than those for SNAP or EITC, since this would be a new program, benefits would be uncertain, and many borrowers do not engage with student loan programs that can be beneficial to them. In sensitivity analyses below, we assume a lower take-up rate.

We also assume that about one-third of these remaining borrowers would benefit from other payment relief options that could sufficiently address their hardship. Some estimates suggest that payment relief options available from the Department can benefit large swaths of borrowers; ^[99] however, not all borrowers who benefit from existing payment relief options will have their hardship sufficiently addressed.

This leads to an estimate of 1 million borrowers, or about 2.5 percent of the current portfolio, who we expect could, at the Secretary's discretion, be approved for relief under proposed § 30.91(d) in the period after this regulation is implemented and throughout the remainder of the borrowers' repayment periods. We also estimate that an additional 1 million borrowers who could, at the Secretary's discretion, be approved among the next 10 cohorts of borrowers. To arrive at estimates of borrowers who would be affected in the next ten future cohorts, we assume that 5 percent of each cohort could, at the Secretary's discretion, be approved for a waiver under proposed § 30.91(d) at some point in their repayment. This is double the share of borrowers estimated in the current portfolio because borrowers in the current portfolio would receive waivers on a one-time basis under proposed § 30.91(c), whereas borrowers in future cohorts would not. Using an assumption of roughly 2 million new borrowers each year for the next ten years, this leads to an estimate of roughly 100,000 borrowers per cohort, and 1 million borrowers over these cohorts.

In addition to our primary estimate, we include low and high estimates to bound the range of reasonable possible waivers under proposed § 30.91(d). Our low estimate assumes that the take-up of application-based relief is 50 percent (instead of 75 percent), but still assumes that one-third of eligible borrowers benefit from other payment relief options. This results in a total of 1.33 million borrowers who could, at the Secretary's discretion, be approved for relief, 0.67 million among the current portfolio, and 0.67 million from future borrower cohorts. In our high estimate, we assume a larger share of borrowers would qualify for relief. Specifically, we assume that 10 percent of borrowers in the current portfolio could be approved for a waiver under proposed § 30.91(d). This larger share aligns with estimates from research suggesting that 10 percent of adults experience spells of poverty that last at least three years, whereas the 5 percent in our base estimate was based on five-year spells.^[100] Similar to our ( print page 87156) base case, we assume a take-up rate of 75 percent and that 1/3 of borrowers get relief through another option. In this high scenario, we estimate that 2 million borrowers in the current portfolio, and 2 million borrowers in future cohorts would qualify. We note that overall estimates could be reduced once they account for other, anticipated regulatory actions that provide relief to borrowers with education debt.

We also consider the possibility that the Department could potentially use data aligned with the factors listed in § 30.91(b) that was not obtained through an application ( e.g. from additional or refined data to which the Department has access in the future). In such potential cases, a borrower could receive a waiver if the Department's holistic review of the borrower's data satisfied the same stringent standard that the Department would apply for application-based relief under proposed § 30.91(d) ( e.g., the borrower must be highly likely to be in default or experience similarly severe and persistent negative circumstances, and other payment relief options would not sufficiently address the borrower's persistent hardship). Such cases would be considered rare since the data that the Department may possess would have to sufficiently establish eligibility, sufficiently show that other options for payment relief did not address the borrower's hardship, and sufficiently distinguish such borrowers from otherwise similar borrowers who would not be deemed to qualify for relief.

We interpret the potential for such waivers to occur on the margin of the take-up rate that we have built into our overall estimates. Changes to the assumptions about the total number of borrowers who could be approved because of the potential for non-application waivers would not be due to differences in the applicable eligibility standard, but rather assumptions about the precision with which various data sources could identify borrowers who were experiencing hardship that could qualify under the standard for relief in proposed § 30.91(d). Our base case assumption includes a 75 percent take-up rate, and we believe this already generously incorporates a high implied take-up rate for a small share of borrowers who might receive waivers under proposed § 30.91(d) without an application. However, we consider the possibility that there could be a higher take-up rate, for example, 80 percent. We also consider that a greater number of borrowers could potentially be approved. Assuming that 5 percent more borrowers could be approved, and a take-up rate of 80 percent, our primary estimates of who would receive relief under proposed § 30.91(d) would increase from 1 million borrowers in the current portfolio to about 1.1 million. We do not formally run a new budget scenario below with these different assumptions, as we believe those estimates would be below the high scenario already discussed above.

The Department also considered how to estimate how many applications it would receive, and the rate at which an application for waiver would be likely to be approved at the Secretary's discretion. As with the discussion above, there is no perfect comparison on which to rely. However, considering that some borrowers who are ineligible will apply, and that for other borrowers, other options for payment relief would sufficiently address the borrower's hardship, we assume that for every borrower who is approved at the Secretary's discretion, there would be one that is rejected, i.e., we assume an approval rate of 50 percent.

For estimating the potential application rates, the Department considered situations that might be closely analogous to the application-based approach contemplated by proposed § 30.91(d), whereby the Secretary would conduct a holistic assessment of the borrower's factors indicating hardship to determine if the borrower met the standard for waiver described in proposed § 30.91(d). We identified few comparative situations. Applications similar to the ones for IDR, Public Service Loan Forgiveness, or the prior pandemic-related student debt relief plan under the HEROES Act were not closely relevant for this estimation, because they generally only involve completing straightforward background questions and checking certain boxes, and there is no meaningful open response required from the borrower. By contrast, the Department expects that the application for relief under proposed § 30.91(d) would solicit a range of qualitative and quantitative information from the borrower to inform the Department's determination of whether the borrower satisfies the hardship standard.

We also considered using the rate of approvals when borrowers submit applications to use a different payment amount when seeking to rehabilitate their loans, but that information is not readily tracked by the Department's contractor. Even if an approval rate were available, that form may still not be an appropriate comparison, since it only affects borrowers in default and those borrowers have particular characteristics in terms of postsecondary completion, type of institution, and debt balance that might be different than the broader population of borrowers. Another approval rate we considered was borrower defense to repayment. Borrower defense also is not a perfect comparison because it tends to have disproportionate numbers of applications from borrowers who attended private for-profit colleges than might be expected to occur here, and there are significant differences between the factors that justify borrower defense to repayment and the waivers proposed here.

Unless specified otherwise in the above discussion, estimates of borrowers who would be eligible for relief under proposed § 30.91(d) do not account for potential overlap with relief that may be provided by any other proposed regulations that are not yet final as of the publication of the NPRM, or of waivers through other provisions that were not yet implemented as of April 30, 2024.

We invite feedback from the public about how to refine these estimates.

Administrative Costs

The proposed regulations could result in significant administrative costs for the Department. These costs would be relatively small for immediate relief granted under proposed § 30.91(c). For that type of relief, the Department would expend one-time resources on developing the predictive assessment contemplated in proposed § 30.91(c) that would predict the likelihood that a borrower will be in default within two years after the publication of these proposed regulations. But the Department would not need to expend significant further resources to apply the predictive assessment to a borrower's information and would not need to expend any resources developing an application, disseminating the application, and reviewing and processing the application.

For relief granted under the application-based pathway described in proposed § 30.91(d), however, the Department would incur significant costs to create, disseminate, and process applications to complete a holistic assessment of the information submitted by the borrower related to hardship. The Department would need to either repurpose or hire additional staff for this purpose. This would create expenses for systems to accept and track the status of applications as well as call- ( print page 87157) center staffing costs to address inquiries related to the application process. The degree of these costs would vary based upon the number of applications the Department has the capacity to process in a year. Increasing the Department's capacity to holistically assess applications would require hiring more staff, either directly or through subcontractors. Greater initial costs for staff could result in lower long-term costs, however, as we anticipate that most borrowers who are initially interested in the application-based process would do so soon after such process is available. We anticipate that future applications would come from newer borrowers or those with a significant change in their circumstances such that they have now decided to seek a hardship waiver.

The Department has developed estimates of the administrative costs for the application-based pathway specified in proposed § 30.91(d) by considering existing analogous administrative processes, particularly the costs related to reviewing applications for borrower defense to repayment. Those processes share some similarities, particularly that borrowers submit applications that may reveal information and evidence that is not otherwise available to the Department and must be reviewed. There are also some key differences. First, borrower defense requires conducting fact-finding related to an institution. That can be a significant upfront investment of time, but any findings from that work can then be applied to multiple applications. Second, the review of borrower defense applications is generally carried out by attorneys. This reflects the legal standards used for borrower defense approvals, which often include making determinations about the nature of misrepresentations and meeting certain evidentiary bars that are grounded in concepts similar to those used by States in their unfair and deceptive acts and practices (UDAP) laws.

The proposed application-based approach for hardship waivers under proposed § 30.91(d) would be different. First, we do not anticipate that the fact-finding related to institutional conduct would apply to the Department's review of applications under proposed § 30.91(d). Second, we do not anticipate that the individuals reviewing hardship applications would need to be attorneys. That means the typical staffing cost could be lower than it is for borrower defense.

Based upon these considerations, the Department modeled the possible administrative costs of the application-based pathway described in proposed § 30.91(d) in the following manner. First, we assumed that the cost per hour to review was $50. This is based on the current hourly rate used by subcontractors for the Office of Federal Student Aid in the Department of Education, which is roughly half the hourly rate were Department staff hired for the same process. We then assumed that it would take each reviewer on average 30 minutes to review an application and render a recommended decision to the Secretary. The 30-minute estimate is similar to how long Department contractors typically take to review a form where borrowers submit detailed income and expense information when seeking to rehabilitate a defaulted loan (information collection 1845-0120). We expect that some applications would be faster while others that include significant additional information might take longer. The overall administrative cost would then depend on how many applications the Department anticipates receiving annually as well as how many we anticipate being able to review in a year. The number would also be higher or lower depending on the number of applications processed each year, the total number of anticipated applications, and the average time to review the application. If we expect a total of 4 million applications, at the current hourly rate and expected review time, the personnel costs for application review is estimated at about $100 million. We will continue to refine these estimates based upon comments received from the public.

In addition to staffing costs, the Department also anticipates incurring some administrative costs for updating and maintaining data systems to process the intake of applications from borrowers seeking hardship waivers, staffing call centers for questions, and costs to train system users. We estimate this amount to be approximately $9 million in the first year, and an additional $1.7 million each year thereafter.

4. Net Budget Impact

Table 4.1 provides an estimate of the net Federal budget impact of these proposed regulations that are summarized in Table 2.1 of this RIA. This includes both costs of a modification to existing loan cohorts and costs for loan cohorts from 2025 to 2034. A cohort reflects all loans originated in a given fiscal year. Consistent with the requirements of the Federal Credit Reform Act of 1990 (FCRA), budget cost estimates for the student loan programs reflect the estimated net present value of all future non-administrative Federal costs associated with a cohort of loans. The baseline for estimating the cost of these final regulations is the President's Budget for 2025 (PB2025) as modified for changes to debt management policies. The baseline ^[101] does not include any changes related to the student debt relief provisions described in the NPRM published April 17, 2024.^[102] Should such debt relief provisions be finalized as proposed in the April NPRM and should the Department provide waivers under such provisions, the Department expects that the estimated costs of these regulations would decrease.

( print page 87158)

Table 4.1—Estimated Net Budget Impact of the NPRM for Direct Loans and ED-Held Loans

[$ in millions]
Section	Description	Modification score (1994-2024)	Outyear score (2025-2034)	Total (1994-2034)
§ 30.91(c)	Immediate relief for borrowers likely to be in default	70,200		70,200
§ 30.91(d)	Application-based relief for borrowers experiencing hardship	29,600	12,100	41,700

It is possible that a borrower who is eligible for immediate relief under proposed § 30.91(c) may also be inclined to apply for relief under proposed § 30.91(d). For budgeting purposes, however, we assume that all relief would be full relief, and that if a borrower qualifies for and receives a waiver under proposed § 30.91(c) then they would not also receive a waiver under proposed § 30.91(d). Accordingly, the primary budget estimate stacks the scores in the order shown on the assumption that immediate relief under proposed § 30.91(c) would be provided to eligible borrowers prior to any additional relief under proposed § 30.91(d) to different borrowers. The Department believes this stacked estimation is appropriate for the primary estimates of the proposed regulations.

Methodology for Budget Impact

The Department estimated the budget impact of the proposed provisions in this NPRM through changes to the Department's Death, Disability, and Bankruptcy (DDB) assumption that handles a broad range of loan discharges or adjustments, the collections assumption to reflect balance changes on loans that ever defaulted, and the IDR assumption for effects on borrowers in those repayment plans. The DDB assumption is used in the Student Loan Model (SLM) to determine the rate and timing of loan discharges due to the death, disability, bankruptcy, or other discharge of the borrowers (this model is not the same as the predictive assessment that is described for determining whether a borrower may be eligible for a waiver under proposed § 30.91(c)). The SLM is designed to calculate cash flow estimates for the Department's Federal postsecondary student loan programs in compliance with the FCRA and all relevant Federal guidance. The SLM calculates student loan net cost estimates for loan cohorts where a cohort consists of the loans originated in a given budget (fiscal) year. The model operates with input data obtained from historical experience and other relevant data sources. The SLM cash flow components range from origination fees through scheduled principal and interest payments, defaults, collections, recoveries, and fees. The cash flow time period begins with the fiscal year of first disbursement and ends with the fiscal year of the events at the end of the life of the loan: repayment, discharge, or forgiveness.

For each loan cohort, the SLM contains separate DDB rates by loan program, population (Non-Consolidated, Consolidated Not From Default, and Consolidated From Default), loan type, and budget risk group (Two-Year Proprietary, Two-Year Public and Not-for-Profit, Four-Year Freshman and Sophomore of all institution types, Four-Year Junior and Senior of all institution types, and Graduate Student of all institution types). The DDB rate is the sum of several component rates that reflect underlying claims data and assumptions about the effect of policy changes and updated data on future claims activity. In general, DDB claims are aggregated as the numerator by fiscal year of origination and population, program, loan type, risk group, and years from origination until the DDB claims. Zeros are used for any missing categories in the numerator. Net loan amounts are aggregated as the denominator by fiscal year of origination and population, program loan type, and risk group. The DDB rate is simply the ratio of the numerator to the denominator. Because the SLM only allows for DDB rates to be specified up to 30 years from origination, DDB claims occurring more than 30 years after origination are included in the year 30 rate. DDB rates for future cohorts are forecasted using weighted averages of prior year rates and have a number of additions and adjustment factors built into them to capture policies or anticipated discharges that are not reflected in the processed discharge data yet, including adjustments for anticipated increased borrower defense and closed school activity.

For estimates related to waivers granted to borrowers enrolled in IDR repayment plans, the Department has a borrower and loan type level submodel that generates representative cashflows for use in the SLM. This IDR submodel contains information about borrowers' time in repayment, the use of deferments and forbearances, estimated incomes and filing statuses, and annual balances. Therefore, we are able to identify or assign the borrowers in the IDR submodel who would be eligible for one of the waivers in proposed § 30.91 and incorporate that effect by ending the payment cycle for borrowers who would be eligible to receive a waiver under proposed § 30.91(c) or (d).

The estimated cost of waivers under proposed § 30.91(c) or (d) varies depending on whether the borrower is in IDR, as well as whether the waiver is a full or partial discharge of the loan. Partial waiver of balances for borrowers already modeled to be on an IDR plan could have three different effects depending upon whether the borrower was expected to get IDR forgiveness prior to these waivers, and whether the waiver changes that anticipated outcome. These potential effects would be:

1. Before and after the waiver is applied, borrowers are expected to receive some IDR forgiveness at the end of their repayment term. For these borrowers, the waivers would affect the amount ultimately forgiven, but because payments are based upon income and the amount of time borrowers would be expected to repay is unchanged, there would be no effect on the amount of anticipated future payments.

2. The borrower was expected to receive IDR forgiveness before the waiver's application, but afterward is now expected to pay off their balance before receiving IDR forgiveness. Because these borrowers would now be expected to repay in less time, there would be some reduction in the amount of anticipated future payments.

3. Before applying the waiver, the borrower was expected to retire their loan balance prior to receiving IDR forgiveness, but as a result of the policy would now be expected to retire their balance sooner. Because these borrowers would now be expected to repay in less time, there would be some reduction in the amount of anticipated future payments. ( print page 87159)

Generally, we project that most partial waivers for borrowers modeled to be on IDR would end up in the first group. Since these borrowers would not see a change in the amount they pay before receiving forgiveness, we do not assign a cost to the waivers for these borrowers. Any costs for waivers granted to borrowers who are modeled to be on IDR come from either full waivers or the minority of borrowers in the second and third groups, for whom the waivers would reduce the number of payments needed to fully repay their loan.

For estimates related to the effects of the proposed waiver provisions on borrowers with loans not in IDR plans, the Department's approach would be to: (1) estimate the potential waiver amounts borrowers would be eligible for and aggregate them by loan cohort, loan type, and budget risk group used in the SLM; (2) add the waiver amounts for non-defaulted, non-IDR borrowers to the Department's baseline DDB assumption in FY 2025; and (3) remove the amounts associated with the waiver provisions from defaulted, non-IDR borrowers from the baseline collections assumption. The revised IDR, DDB and collections groups are run in a SLM scenario for each provision to generate the estimates in Table 4.1. To produce the potential waiver amounts in Step 1 of this process, the Department developed a loan-level file based on the FY2023 sample of NSLDS information used for preparing budget estimates, with balance information supplemented by redrawing key loan information as of June 13, 2024, to account for the discharges and waivers that occurred in FY2024 and reduced borrower's balances to zero. Information from this file would allow the evaluation of times in repayment that would qualify for one of the provisions and anticipated future balances for use in calculating the amount that the Secretary might waive for borrowers who have experienced changes in balance.

These estimates are all based on the same random sample of borrowers that the Department uses for all other budget estimation activity related to Federal student loans. Currently, the most recent sample available is from the end of FY2023, which is the best currently available data that maintains the Department's consistent scoring practices.

The Department followed two different approaches for modeling the estimated cost of the provisions in proposed § 30.91(c) and (d). For proposed § 30.91(c), the Department considered the output of the model developed to project the likelihood that a borrower would be in default within two years. We used that model to estimate the number of borrowers by risk group as well as repayment status ( e.g., in default or in repayment) and estimated the cost of forgiving those loans. For the outyear cohorts, we randomly assigned borrowers to default based on default rates by cohort, risk group and loan type assumed in the President's Budget for FY2025 baseline. This approach reflects that under proposed § 30.91(c), the Secretary may identify borrowers eligible for relief based on a predictive assessment, without requiring any action by those borrowers.

The Department took a different approach for proposed § 30.91(d). That proposed provision describes an application-based pathway for relief whereby the Secretary may conduct a holistic assessment of the borrower's factors indicating hardship based on information obtained through an application process in addition to potentially supplementing that data with information already in the Department's possession. To model this approach, the Department generated assumptions of the number of borrowers who would be eligible for a waiver. Of this number of assumed eligible applicants, we then calculated the distribution of borrowers across risk group ( e.g., 2-year proprietary, graduate borrowers, and 4-year nonprofit or public institutions), and by cohort year. We describe this process in greater detail below.

First, as described earlier, the Department considered multiple potentially comparable situations it has dealt with in the past to estimate the number of applications from borrowers seeking proposed waivers related to hardship. We examined the volume associated with borrowers on IDR plans, those who applied for student debt relief under the HEROES Act, and those who applied for economic deferment or hardship from a period prior to the payment pause. With no perfect analog but based on the best available data, we use a base estimate that about 1 million borrowers in the current portfolio would be approved for relief. The estimate of borrowers who would be affected in future cohorts over the next ten years is 1 million.

To reduce the possibility in the net budget impact estimate that borrowers who might be otherwise captured under proposed § 30.91(c) and could potentially be double counted in both proposed § 30.91(c) and proposed § 30.91(d), we estimated the cost of the waivers proposed in proposed § 30.91(d) after accounting for the cost of proposed § 30.91(c) and only allowed borrowers to receive a waiver under one of the proposed provisions. The Department seeks feedback about the assumed number of borrowers who would be approved for a waiver under proposed § 30.91(d), and we will continue to refine this estimate.

Next, to estimate the distribution of approved applicants across risk group and cohort year, the Department consulted the closest past situations that might operate similarly to the proposed waivers. This included looking at the distribution of borrowers across risk group and cohort year who submitted applications for relief under the HEROES Act Plan that was announced in August 2022, borrowers who ever defaulted on loans based on Department data, and borrowers who had a qualifying economic hardship forbearance or deferment. While not exact corollaries, these data nonetheless provide useful information on which types of borrowers might choose to apply. We believe these are better comparisons for thinking about the distribution of approved borrowers than other types of existing information. For instance, we do not think closed school loan discharges would be a good comparison, because those borrowers only come from colleges that closed, which would largely exclude public institutions. We also chose not to use borrowers who documented income and expense information when seeking a loan rehabilitation, because these borrowers are disproportionately likely to have not finished their postsecondary programs, whereas the hardship applications could come from borrowers who graduated but who are struggling on their loans in ways beyond just being in default.

Specifically, to estimate the distribution of approved applicants across cohort years, the Department equally weighted the distributions observed under submitted applications for relief under the HEROES Act Plan that was announced in August 2022 and borrowers who ever defaulted on loans based on Department data. To calculate the distribution of borrowers across risk groups, the Department equally weighted the distributions observed under submitted applications for relief under the HEROES Act Plan that was announced in August 2022 and borrowers who had a qualifying hardship forbearance or deferment. If a cell reached 100 percent of sampling in current Department data, the excess approved applicants were distributed within the respective cohort range row by weight. The resulting estimated ( print page 87160) distribution of approved applicants is shown in Table 4.2.

Table 4.2—Estimated Percentage Distribution of Approved Applicants by Cohort and Risk Group
	Non- consolidated	Consolidated non-default	Consolidated default	Total
	Non- consolidated	Consolidated non-default	Consolidated default	2-Yr Proprietary	2-Yr Public	4-Yr PubPri FrSo	4-Yr PubPri JrSr	Grad
1994-2004	0.6	0.2	1.3	1.3	1.4	3.5	0.9	9.1
2005-2009	0.7	0.7	2.1	3.2	1.6	2.5	0.6	11.5
2010-2014	2.3	2.4	6.9	10.5	5.1	8.3	2.1	37.4
2015-2019	1.8	1.8	5.2	7.9	3.8	6.3	0.9	27.6
2020-2024	0.9	0.9	2.8	4.2	2.0	3.3	0.1	14.3
Total	6.3	6.0	18.2	27.1	13.8	23.9	4.6	100.0

Next, we randomly identified non-defaulted, non-IDR borrowers within each risk group and cohort year cell, based on the percentages shown in Table 4.2, to be in the approved applicant pool. We then waived the assumed future balances in the sample to generate the estimated increase in DDB claims or reduction in collections associated with the hardship application provisions. To provide a maximally conservative budget estimate, we assumed all of these approvals would result in full relief. Lesser amounts of relief would reduce the estimated cost.

Table 4.3 shows the outstanding loan balances by risk group for approved borrowers from existing cohorts that entered repayment by 2024.

Table 4.3—Outstanding Loan Balances by Risk Group

[$ Millions]
Risk group	Outstanding loan balances
2-Yr Proprietary	835
2-Yr Public	1,103
4-Yr PubPri FrSo	6,099
4-Yr PubPri JrSr	5,680
Grad	5,034
Consol	7,523
Total	26,275

The sampling process described above generated the estimated forgiveness for borrowers from existing cohorts for loans that entered repayment by 2024. For future cohorts and loans that enter repayment in 2025 and later from existing cohorts, we calculated the percent of net volume that was associated with borrowers that entered repayment by 2024 assigned to receive forgiveness by origination cohort, consolidation status, budget risk group, and time to receiving hardship forgiveness from entering repayment (offset). We then took the average forgiveness percentage of volume across origination cohorts by risk group, consolidation status, and offset to estimate the percent of volume that will enter repayment from 2025 and out that we estimate will receive hardship forgiveness. As we expect it will take borrowers some years in repayment to demonstrate persistent hardship, we have distributed forgiveness in the outyears evenly from years 5 to 15 in repayment. This estimated forgiveness is then summarized by origination cohort, consolidation status, budget risk group, loan type, and offset and added to the baseline estimate for the discharge assumption to generate the cost of § 30.91(d).

The Department also considered how to estimate how many applications it would receive, and the rate at which an application for waiver would be likely to be approved.

As described previously, we assume that for every two borrowers who are eligible, there is one that is rejected because their needs are met via other Department payment relief options, such as IDR plans. We also assume that there would be borrowers who apply but do not meet the standard. On net, we assume that for every eligible applicant, there is also one ineligible applicant, for an effective approval rate of 50 percent. The Department seeks feedback about these assumed approval rates, and we will continue to refine this estimate.

5. Accounting Statement

As discussed in OMB Circular A-4, we have prepared an accounting statement showing the classification of the expenditures associated with the provisions of these proposed regulations. Table 5.1 provides our best estimate of the changes in annual monetized transfers that may result from these proposed regulations.

Expenditures are classified as transfers from the Federal government to affected student loan borrowers.

Table 5.1—Accounting Statement: Classification of Estimated Expenditures

[In millions]
Reduction in loans that are unlikely to be repaid in full in a reasonable period
Increased ability for borrowers to repay loans on which they have or are experiencing hardship
Reduced administrative burden for Department due to reduced servicing, default, and collection costs
Category:
Paperwork Reduction Act burden on borrowers to complete applications	$11.14
Administrative costs to Federal government to update systems and contracts to implement the proposed regulations	$2.5
Administrative costs of staff reviews	$12.1
Reduced transfers from borrowers due to waivers:	2%
Based on high likelihood of being in default	$7,657
Based on applications	$4,432

( print page 87161)

6. Alternatives Considered

The Department considered the option of not proposing these regulations, as the Secretary has existing waiver authority under sections 432(a)(6) and 468(2) of the HEA. However, we believe these regulations are important to inform the public about how the Secretary would exercise this longstanding discretionary waiver authority in a consistent and transparent manner. The Department believes that foregoing these proposed regulations would reduce transparency about the Secretary's discretionary use of waiver. For all the reasons detailed above, hardship waivers would produce substantial, critical benefits for borrowers and the Department. Overall, as discussed above in the context of the relevant Executive Orders, the Department's analysis suggests that the benefits of the proposed regulations will outweigh their costs.

As part of the development of these proposed regulations, the Department engaged in a negotiated rulemaking process in which we received comments and proposals from non-Federal negotiators representing numerous impacted constituencies. These included higher education institutions, legal assistance organizations, consumer advocacy organizations, student loan borrowers, civil rights organizations, State officials, and State attorneys general. Non-Federal negotiators submitted a variety of proposals relating to the issues under discussion. Information about these proposals is available on our negotiated rulemaking website at https://www2.ed.gov/policy/highered/reg/hearulemaking/2023/index.html.

Because the negotiators reached consensus on the proposed regulations in this NPRM, the Department did not consider alternative regulations in the drafting of this NPRM. We did, however, consider some alternatives during the negotiated rulemaking process.

The Department considered including the issue of borrowers affected by servicer errors as a potential sign of hardship. However, we decided not to explicitly include that as a factor under proposed § 30.91(b) because the Department has existing procedures to address administrative errors without needing these specific regulations for them.

The Department also considered using an exclusive list of factors indicating hardship in proposed § 30.91(b) but concluded that a non-exhaustive list would provide necessary flexibility to consider unanticipated factors indicating hardship and to incorporate new types of data as they become available.

As noted above, the Committee reached consensus on the regulatory language proposed in this NPRM.

7. Regulatory Flexibility Act

The Secretary certifies, under the Regulatory Flexibility Act (5 U.S.C. 601 et seq.), that this final regulatory action would not have a significant economic impact on a substantial number of “small entities.”

These regulations will not have a significant impact on a substantial number of small entities because they are focused on arrangements between the borrower and the Department. They do not affect institutions of higher education in any way, and those entities are typically the focus on the Regulatory Flexibility Act analysis. As noted in the Paperwork Reduction Act section, burden related to the final regulations will be assessed in a separate information collection process and that burden is expected to involve individuals.

8. Paperwork Reduction Act

As part of its continuing effort to reduce paperwork and respondent burden, the Department provides the general public and Federal agencies with an opportunity to comment on proposed and continuing collections of information in accordance with the Paperwork Reduction Act of 1995 (PRA) (44 U.S.C. 3506(c)(2)(A)). This helps make certain that the public understands the Department's collection instructions, respondents can provide the requested data in the desired format, reporting burden (time and financial resources) is minimized, collection instruments are clearly understood, and the Department can properly assess the impact of collection requirements on respondents.

Proposed § 30.91 in this NPRM contains information collection requirements. Under the PRA, the Department would, at the required time, submit a copy of these sections and an Information Collections Request to the Office of Management and Budget (OMB) for its review.

A Federal agency may not conduct or sponsor a collection of information unless OMB approves the collection under the PRA and the corresponding information collection instrument displays a currently valid OMB control number. Notwithstanding any other provision of law, no person is required to comply with, or is subject to penalty for failure to comply with, a collection of information if the collection instrument does not display a currently valid OMB control number. In the final regulations, we would display the control numbers assigned by OMB to any information collection requirements proposed in this NPRM and adopted in the final regulations.

Section 30.91—Waiver due to likely impairment of borrower ability to repay or undue costs of collection.

Requirements: The NPRM proposes to add a new § 30.91 to 34 CFR part 30 in which the Secretary would consider granting a waiver for borrowers experiencing hardship. To implement proposed § 30.91(d), the Department would use an “additional relief” process using a holistic assessment approach, where the Department would consider information provided by a borrower through an application, based on a non-exclusive list of factors in proposed § 30.91(b), indicating that they are experiencing hardship. Information would include items such as a borrower's household income and assets, payments on debt relative to household income, and exceptional amounts of costs for caretaking.

While some of the information in proposed § 30.91(b) could be obtained from the Department's administrative data, other information must be obtained from the borrowers themselves through an application. The information collected on the application would be used to assess eligibility for a hardship determination. The Department expects that the application for relief under proposed § 30.91(d) would solicit a range of qualitative and quantitative information from the borrower to inform the Department's determination of whether the borrower satisfies the hardship standard.

Burden Calculations:

§ 30.91 Waiver due to likely impairment of borrower ability to repay or undue costs of collection.

The proposed regulatory changes would add burden to borrowers and would require a new information collection. As discussed in the net budget impact section, we estimate that between 2.67 million and 8 million borrowers would submit hardship applications. The costs are estimated using the median hourly wage of $23.11 reported by the Bureau of Labor Statistics for all occupations.^[103] We estimated the number of hours needed to complete the proposed application based upon discussions with Department staff that have worked on similar processes in the past. Through those conversations, we estimate that it ( print page 87162) would take a typical borrower 1 hour to complete the application form to indicate they want to pursue the application-based process. The Department's two closest analogous types of application forms are the one that borrowers submit when filing a borrower defense to repayment application and the one that borrowers fill out to document their income and expenses when seeking to rehabilitate a defaulted loan. For borrower defense forms, the Department estimates that it takes a borrower 30 minutes (0.5 hours) to complete, while the rehabilitation form takes an estimated 60 minutes (1 hour) per borrower. We anticipate that the application form for the proposed hardship waiver would likely take as long as the rehabilitation form to fill out. We came to this conclusion because borrowers who want to provide information about indicators of hardship from their finances or assets may need to provide supplemental financial information. Applicants may have to put together documentation related to high essential expenses, such as health care or dependent care costs. They may also need to provide information about how they are experiencing hardship as a result of the items identified and why it is likely to persist.

Because we do not want to double count borrowers who may qualify for and receive relief under proposed § 30.91(c), the estimate for proposed § 30.91(d) illustrated below does not include borrowers who would be expected to receive full relief under proposed § 30.91(c).

These figures and considerations are the basis for the following estimations.

§ 30.91 Hardship Application—OMB Control Number 1845-NEW

Affected entity	Applications	Burden hours	Cost $22.31 per hour
Individual low scenario	2,667,000	2,667,000	59,500,770
Individual medium scenario	4,000,000	4,000,000	89,240,000
Individual high scenario	8,000,000	8,000,000	178,480,000
Average Total	4,889,000	4,889,000	109,073,590

Consistent with the discussions above, the following chart describes the sections of the proposed regulations involving information collections, the information being collected and the collections that the Department would submit to OMB for approval and public comment under the PRA, and the estimated costs associated with the information collections. The monetized net cost of the increased burden for borrowers using wage data was developed using Bureau of Labor Statistics (BLS) data. For individuals, we have used the median hourly wage for all occupations, $23.11 per hour according to BLS.^[104]

Collection of Information
Regulatory section	Information collection	OMB control number and estimated burden	Estimated cost $23.11 per hour
§ 30.91	Would allow the Secretary to receive applications that provide information for the Secretary to conduct hardship determinations.	1845-NEW 4,889,000 average hours	$109,073,590
Total		1845-NEW 4,889,000	109,073,590

Visit the Official Version

Document Information

Published:: 10/31/2024
Department:: Education Department
Entry Type:: Proposed Rule
Action:: Notice of proposed rulemaking.
Document Number:: 2024-25067
Dates:: We must receive your comments on or before December 2, 2024.
Pages:: 87130-87163 (34 pages)
Docket Numbers:: Docket ID ED-2023-OPE-0123
RINs:: 1840-AD95: Student Debt Relief Hardship
RIN Links:: https://www.federalregister.gov/regulations/1840-AD95/student-debt-relief-hardship
Topics:: Claims, Income taxes
PDF File:: 2024-25067.pdf
Supporting Documents:: » Student Debt Relief Supplemental Report; » Forms and Instruments; » Supporting Statements
CFR: (1): 34 CFR 30