Accountability in Higher Education and Access Through Demand- Driven Workforce Pell: Student Tuition and Transparency System (STATS) and Earnings Accountability

Executive Orders 12866 and 13563

Under Executive Orders (E.O.) 12866, the Office of Management and Budget (OMB) must determine whether a regulatory action is “significant” and, therefore, subject to the requirements of the E.O. and subject to review by OMB. Section 3(f) of E.O. 12866 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 $100 million or more 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, 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 novel legal or policy issues arising out of legal mandates, the President's priorities, or the principles stated in the E.O.

The Department estimates the net budgetary impacts to be $1,517 million from changes in transfers between the Federal Government and student loan borrowers and transfers of $8,782 million between the Federal Government and Pell Grant recipients resulting from replacing the current regulations with the accountability framework. Annualized, these transfers are estimated at $871 million and $862 million for Pell Grants and $149 million and $147 million at 3 percent and 7 percent discounting, respectively. Quantified benefits include a net reduction in costs of compliance with paperwork requirements ($113.4/$103.7 million) while quantified costs include administrative updates to Government systems ($2.4/$2.8 million), implementation staffing and contract costs ($1.4/$1.6 million), long-term staffing costs ($0.9/$0.9 million), and ongoing contract costs ($1.9/$1.8 million) at 3 percent and 7 percent discounting, respectively. Therefore, based on our estimates, the Office of Information and Regulatory Affairs (OIRA) has determined that this proposed rule is “economically significant” under section 3(f)(1) of E.O. 12866 and subject to OMB review.

We have also reviewed these regulations under E.O. 13563, which supplements and explicitly reaffirms the principles, structures, and definitions governing regulatory review established in E.O. 12866. To the extent permitted by law, E.O. 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 encourage the desired behavior, or provide information that enables the public to make choices.

The E.O. 13563 also requires an agency “to use the best available techniques to quantify anticipated present and future benefits and costs as accurately as possible.” OIRA has emphasized that these techniques may include “identifying changing future compliance costs that might result from technological innovation or anticipated behavioral changes.”

This final rule is considered an E.O. 14192 deregulatory action. We estimate that this rule generates $111.3 million in annualized cost savings at a 7% discount rate, discounted relative to

year 2024, over a perpetual time horizon. E.O. 14192 directs agencies of the executive branch to be prudent and financially responsible in the expenditure of funds, from both public and private sources, and to alleviate unnecessary regulatory burdens placed on the American people.

Consistent with OMB Circular A-4, we compare the final regulations to the current regulations. In this regulatory impact analysis, we discuss the need for regulatory action, potential costs and benefits, net budget impacts, and the regulatory alternatives we considered.

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

In this RIA, we discuss the need for regulatory action, the summary of comments and changes from the NPRM, the impact of the final regulation on institutions and programs, the costs and benefits of the final regulations, the net budget impacts, and the regulatory alternatives we considered in cases where the Department had discretion. Unless otherwise noted, throughout this RIA we compare the effects of the final regulation relative to a pre-statutory baseline where the WFTCA has not been enacted. This baseline includes the current Financial Value Transparency and Gainful Employment regulation (enacted October 10, 2023). Defining Key Terms

Key terms used throughout this RIA are defined as follows:

“Current Regulations”--refers to the current Financial Value Transparency and Gainful Employment regulations that were enacted on October 10, 2023 (88 FR 70004); \25\

\25\ Financial Value Transparency and Gainful Employment, 34 CFR parts 600 and 668 Docket ID ED-2023-OPE-0089. www.federalregister.gov/documents/2023/10/10/2023-20385/financial- value-transparency-and-gainful-employment.

“Accountability framework”--refers collectively to the debt to earnings (D/E) and earnings premium (EP) tests in the context of the current regulation, or to the revised EP test in the context of the final regulation; \26\

\26\ The EP and D/E metrics are defined in the “Methodology for Current Regulation Calculations” and “Methodology for Final Regulation Calculations” subsections below. In the context of the final regulations, “accountability framework” also includes the final revisions to the standards of administrative capabilities, discussed in the “Department Authority (Including GE and Quality Assurance Authority)” section above.

“GE programs”--refers to programs that are subject to the gainful employment rule and the accountability framework under the current regulations, which includes all non-degree programs and all types of programs offered at proprietary institutions;

“Non-GE programs”--refers to programs that are not subject to the gainful employment rule and accountability framework under the current regulations, which includes degree programs offered at public and non-profit institutions.

These final regulations are needed to implement certain provisions of the WFTCA that affect students and program participants in the Federal student loan programs authorized under title IV of the HEA. The WFTCA amended the HEA to create new eligibility criteria for programs of study at institutions to receive title IV loans. These changes establish an accountability framework for all undergraduate degree programs and all types of graduate programs that participate in the Direct Loan program. The final regulations are also needed to align existing accountability framework under the current FVT/GE rule (88 FR 70004) with those in the WFTCA.

The Department has limited discretion in implementing many of the provisions contained in the WFTCA. Many of the changes included in these final regulations simply modify the Department's regulations to reflect statutory changes made by the WFTCA. In some cases, the Secretary has exercised her limited discretion to implement certain provisions of the WFTCA. Areas of limited discretion include:

General definitions (Sec. 668.2), including how earnings would be measured and defined;

The student tuition and transparency system framework (Sec. 668.402), including the specific reporting requirements for institutions;

The method for calculating the earnings premium (Sec. 668.403), and whether the Department should adjust or exempt certain programs for various reasons;

The appeals process (Sec. 668.603), including the usage of alternative earnings data from State data systems; and

The scope and purpose of the earnings accountability framework (Sec. 668.601), including whether certain undergraduate certificate programs should be exempted, whether accountability framework should be delayed for certain programs, and whether the sanction for failing programs should be the removal from participation in all title IV, HEA programs.

These areas of limited discretion are discussed in the “Alternatives Considered” section below. In general, where the Secretary had discretion, she sought to align the accountability framework in Section 84001 of the WFTCA with the accountability framework under the current regulations such that all postsecondary programs are covered by the same accountability framework. In addition to the reasons stated earlier in the “Department Authority (Including GE and Quality Assurance Authority)” section, this alignment reduces complexity, burden, confusion, and compliance costs for both institutions and the Department. Additionally, the Secretary sought to reduce reporting burden under the STATS framework while maintaining the disclosure of relevant information on college costs and outcomes to students and families.

Table 3.1 provides a summary of changes from the NPRM to the final rule.

Table 3.1--Summary of Key Changes in the Final Regulations

Description of final

Provision Regulatory section provision

STATS and Earnings Accountability

Updating application information................ Sec. 600.21(a)(11)............. The final regulation is

updated to clarify that

the requirement for

institutions to report to

the Department certain

changes to eligible

programs now applies both

to GE and eligible non-GE

programs. General definitions............................. Sec. 668.2(b).................. The final regulation

revises the definition of

Cohort period to

streamline the cohort

expansion procedures.

It also revises the

definition of Earnings

threshold to establish a

minimum benchmark of one

dollar in cases where the

Census Bureau data

necessary to perform the

calculations is

unavailable. Program participation agreement................. Sec. 668.14(h)(3) and (4)...... The final regulation adds

two exceptions to the loss

of title IV, HEA

eligibility for all of an

institution's low-earning

outcome programs for

failure to comply with the

administrative capability

requirement at Sec.

668.16(t) in two out of

three years.

First, an institution's low-

earning outcome programs

are not subject to an

automatic loss of

eligibility for title IV,

HEA funds if the

institution is not

participating in the

Direct Loan program and

has not participated in

that program for the five

most recently completed

award years.

Second, the program is not

subject to the loss of

title IV, HEA funds if the

institution agrees in an

amendment to its PPA to

use its authority under

Sec. 685.203(m)(2) to

prevent students from

borrowing Direct Loans in

the program for at least

five years. This exception

will continue to apply for

as long as the institution

continues to prevent

Direct Loan borrowing in

the program. Initial and final decisions..................... Sec. 668.91(a)(3)(vi).......... The final regulation

clarifies that the

agency's action against a

low-earning outcome

program could be either a

limitation or termination

action. Student tuition and transparency system Sec. 668.402(b)................ The final regulation, for a

framework. GE program designed to

prepare students for

employment in a recognized

occupation that qualifies

for a deduction of tip

income under IRS “No Tax

on Tips” regulations and

50 percent or more of

individuals in the

occupation receive income

from tips, will not

consider the program to

have passed or failed the

earnings premium measure

for any award year in

which the graduate cohort

earnings data includes

earnings from 2025 or

prior. Earnings accountability scope and purpose....... Sec. 668.601................... The final regulation

exempts from the

accountability framework

programs at institutions

that enroll only

individuals with a

documented Specific

Learning Disability or

Autism, as defined under

34 CFR 300.8. Low-earning outcome programs.................... Sec. 668.603................... The final regulations

provide that an

institution may appeal

within 30 days the

Secretary's determination

that a program is a low-

earning outcome program

through a process

described by the Secretary

separate from the

limitation and termination

proceedings under Part 668

Subpart G. Student warnings................................ Sec. 668.605(c)(1)(iii)........ The final regulation adds

warning content to notify

students about a program's

potential loss of overall

title IV, HEA eligibility

under Sec. 668.14(h).

The Department received hundreds of comments related to the Regulatory Impact Analysis in the NPRM. This section responds to these comments. Many commenters submitted alternatives that were substantially similar to remarks submitted by other commenters; in these cases, we grouped those comments and responded to them collectively. Impact on the Economy

Comments: Some commenters expressed concern that the regulation will reduce economic activity and economic growth because the earnings test will lead to fewer educational opportunities as programs and colleges will be forced to close. Commenters noted this could reduce the supply of workers in critical fields, prevent individuals from starting new businesses, harm rural and local economies where college close, and lead to a less educated workforce, harming the nation's competitiveness internationally.

Discussion: The Department notes that the overall impact of the rule is estimated to increase the number of eligible educational programs (weighted by enrollment) and increase title IV, HEA disbursements relative to the current rule (Tables 5.12 and 5.15). While the final rule applies an earnings test to all degree programs for the first time, it also replaces the current Gainful Employment rule (implemented in 2023) with a new earnings test under which fewer programs fail. The net effect, as is shown throughout the Regulatory Impact Analysis, is an increase in program eligibility and a net increase in title IV, HEA volume for programs and students between 2027 and 2036 relative to current regulations (Tables 5.15,7.1A, 7.1B).

The Net Impact Budget Impact reflects an increase in outlays of $1,517 million in Direct Loan cohorts 2027 to 2036 and $8,782 million in Pell Grants in FYs 2027 to 2036. On an annual basis, the Department estimates that an additional $1.2 billion in title IV, HEA loans and Pell Grants will be disbursed to students relative to the baseline (Table 5.15).

Overall, the Department's analysis shows that approximately 85,500 additional students will gain access to title IV aid under the proposed regulation (Table 5.12) because fewer programs fail the earnings test. Under the current regulations, the programs that these students attend would lose eligibility for all title IV, HEA program assistance, negatively impacting their ability to afford college. Under the final rule, failing programs usually only lose eligibility for title IV Federal student loans (unless the failing program is also at an institution that fails the Standards of Administrative Capability requirements).

The Department's estimates also show that the earnings test could result in higher earnings for graduates of associates degree programs (Table 6.1) because the regulations reduce the number of low-earning associates degree programs that may receive Title IV aid (many of which were previously exempt from the accountability framework). Put another way, the associate degree programs that are available to students will, on average, lead to higher earnings among students earning these degrees. The earnings for associate degree programs that pass the earnings test in the proposed rule but fail in the current rule are approximately $7,000 higher, on average, than programs that pass the earnings test in current regulation and fail under the proposed regulation. This is because the WFTCA applies an accountability framework to all degree programs (including associates degrees) at all types of institutions whereas the prior rule did not apply to associates degree programs offered at public and private non-profit institutions. In short, more associates degrees are subject to an accountability framework under the final rule, reducing the number of programs that lead to low earnings.

Ultimately, the Department's analysis shows that the assertions made about the harmful effects this regulation would have on the economy are misguided. In fact, when compared against the baseline Gainful Employment regulation, this final rule will result in fewer students being negatively impacted by program closures due to failing the earnings test, and it will therefore have a smaller

effect on worker shortages and the local economies who depend on such workers.\27\

\27\ Consequently, some students who attend programs that failed under the current regulation but now pass under this regulation may be harmed by the ability to attend passing programs that have relatively lower-earning programs. However, it is difficult for the Department to estimate this possibility because we lack counterfactual earnings data for what the students' earnings outcomes may have been had they not attended college at all.

Changes: None.

Comments: Some commenters expressed that the regulation would harm the economy because it would negatively impact state budgets. Commenters expressed that the regulation would result in the closure of many types of programs, including cosmetology programs. Commenters stated that this will result in fewer students taking state licensure exams, ultimately reducing state revenues from licensure examination fees.

Discussion: The Department evaluated the impact of the final rule on programs, students, and title IV, HEA student financial assistance. The Department's analysis indicates that fewer students will be impacted by the earnings test in the final rule relative to the share of students impacted under the current policy (Table 5.12). To the extent that the commenters' concerns about state budget are accurate, the Department's analysis implies that the final regulation would result in an increase in state revenue because more students will be anticipated to take state licensure examinations relative to the baseline policy.

Changes: None. Impact on Small Businesses

Comments: The Department received many comments about the effects the regulation would have on small businesses, particularly with respect to cosmetology businesses. Commenters explained that because the earnings test would cause cosmetology programs to lose access to title IV, HEA program assistance, there will be fewer cosmetologists, a group that tends to own and operate their own small businesses. Others noted that barber shops, salons and spas are small businesses that rely on trained and licensed cosmetologists as their employees. Cosmetology schools themselves are often small businesses, as some commenters noted.

Discussion: The Department disagrees with the commenters' assertions. We assessed the effects of the rule on cosmetology programs, single-program institutions (which are often cosmetology schools), and small institutions of higher education in the Regulatory Flexibility Act analysis. In each of these analyses, we find that when compared with the current Gainful Employment regulation, fewer programs are expected to fail the earnings test under this final rule (Tables 5.17, 5.18, 5.19, and 5.20). Relative to the baseline policy, the Department estimates the earnings test will reduce the negative effects on small businesses commenters have raised.

Although a smaller share of cosmetology programs are expected to fail under the final rule relative to the baseline policy, the Department does acknowledge that a high share of cosmetology certificate programs are likely to fail under this final rule (approximately 93%), and that the smallest institutions of higher education are more likely to have a high share of failing programs. The Department is, however, concerned that these fields and credentials do not produce adequate earnings to support student debt. The Department believes that institutions of higher education, employers, and state and local policymakers will have stronger incentives as a result of the earning premium measure to create or modify programs so that they lead to higher earnings, or reform employee pay policies, or credentialing requirements.

Lastly, the Department notes that many cosmetology programs do not participate in title IV, HEA programs. One study found that just 14 percent of barber and cosmetology programs in Texas participate in federal student loan and grant programs. Moreover, many of these non- federally funded programs charge lower tuition prices and have similar outcomes than cosmetology programs subsidized by taxpayers.\28\ These non-federally funded cosmetology programs will have incentives to increase their enrollment if fewer programs are eligible for title IV, HEA student aid, helping to supply the cosmetology workforce. As a result, it is possible that students benefit from this rule if they switch from more-expensive, Federally funded cosmetology programs to less expensive, non-Federally funded programs.

\28\ Cellini, S.R., & Onwukwe, B., (2022). Cosmetology Schools Everywhere: Most Cosmetology Schools Exist Outside of the Federal Student Aid System. Washington, DC: PEER Center. www.american.edu/ spa/peer/upload/peer_cosmetology_b.pdf.

Changes: None Impact on Consumers

Comments: Some commenters stated that the regulation will impact consumers by making the price of haircuts more expensive. Commenters stated that the rule would result in the closure of many cosmetology programs, creating a shortage of barbers and therefore resulting in an increase in the price of haircuts.

Discussion: The Department specifically examined the estimated impact of the regulation on cosmetology programs (Table 5.17, 5.18, 5.19, 5.20, 5.27, and 5.28). Relative to the current Gainful Employment regulation, the final rule is estimated to result in fewer failing cosmetology programs. As a result, this means there will likely be fewer cosmetology program closures as a result of this regulation, helping to blunt the possible shortage of barbers that would otherwise occur under the baseline policy. The Department also notes that there are hundreds of cosmetology programs participating in the title IV, HEA programs that will not be covered by the accountability framework in this regulation, further mitigating the extent to which the regulations contribute to a worker shortage (Table 5.27).

Changes: None.

Comment: Commenters expressed that the regulation would harm consumers in the particular context of cosmetology. Some commenters expressed that the rule would result in the closure of many cosmetology programs, which would therefore create negative effects on health, safety, and sanitary conditions because more services would be provided in homes and in unlicensed or uninspected facilities

Discussion: The Department notes that cosmetologist licensure and facility inspection are areas regulated and enforced at the State and local levels, not at the Federal level. The Department trusts the appropriate State and local entities to maintain appropriate standards for health and safety within their jurisdiction. Furthermore, as shown in the Department's analysis (Tables 5.17, 5.18, 5.19, 5.20, 5.27, and 5.28), fewer cosmetology programs are expected to fail the earnings test under the final rule relative to the current gainful employment regulation. Therefore, the Department believes this final rule reduces the health, safety, and sanitary concerns expressed by the commenters because it is likely that more cosmetology programs will remain open relative to the baseline policy.

Changes: None Impact on College Enrollment

Comments: Commenters expressed that the regulation would result in a decline in college enrollment, arguing that many students will no longer be able to afford to attend higher education

without access to title IV, HEA student aid programs. Commenters further noted that the decline in college enrollment will negatively impact the nation's economy, competitiveness, and entrepreneurialism.

Discussion: The Department is concerned about the low-earnings outcomes found in certain programs, and believes that students will be economically harmed by attending such programs. Students will have access to other title IV-eligible programs and can continue receiving grants and loans to attend non-failing programs. Furthermore, the Department notes that fewer students will be impacted by the final rule relative to the current policy (Table 5.12). Therefore, to the extent that the commenters' concerns are accurate, we note that this rule will enhance the ability for students to remain enrolled in college, thereby benefiting the nation's economy, competitiveness, and entrepreneurialism.

Changes: None. Impact on Specific Student Groups

Comments: Numerous commenters expressed concern about the impact the proposed rule would have on under-represented and disadvantaged student populations, including those from racial/ethnic minority groups. Some commenters stated opposition to the regulation because they believed it would disproportionately harm these groups, and requested the Department conduct an analysis to examine this issue. Some commenters expressed that the regulation would disproportionately harm certain types of institutions, such as Historically Black Colleges and Universities (HBCUs), Tribally Controlled Colleges and Universities (TCCUs), and other minority-serving institutions (MSIs).

Discussion: The Department specifically examined how the rule may impact individuals from different racial/ethnic backgrounds, shown in Table 5.21. For all racial/ethnic groups, the Department estimates that smaller shares of students will attend failing programs under the proposed regulation relative to the current regulation. We therefore disagree with these commenters, as this analysis shows that students from racial/ethnic minority groups will be less impacted overall by the proposed regulation relative to the way these students would be impacted under the current Gainful Employment regulations.

Furthermore, the Department estimated how the regulation would impact HBCUs, TCCUs, and other MSIs in Table 5.10. Relative to the current regulations, we estimate the final rule would result in a slight to moderate increase in the share of students and programs at HBCUs and MSIs that fail the earnings test. This is largely because of the statutory requirement to hold all programs accountable for their earnings outcomes, whereas the existing regulation only applied to non- degree programs and for-profit institutions.

Although a marginally higher share of programs and students at HBCUs and MSIs will be impacted by the final rule, the Department remains concerned that these programs regularly provided students with very low earnings after graduation, leaving them unable to afford their student debt burden and other financial expenses. The Department believes this final regulation will therefore benefit these students who otherwise would have attended these low-earning programs. Because of the final rule, some of these students may now consider attending other higher-earning programs that are not impacted by these final regulations, possibly resulting in higher earnings for students.

Changes: None.

Comments: Many commenters expressed concern that the regulation will negatively impact women. Commenters stated that many of the programs that will fail the earnings test--including education programs, social work programs, arts programs, cosmetology programs, and childcare programs--serve larger shares of women, resulting in a disproportionate impact on this student population.

Commenters also stated that programs that serve larger shares of women will be disproportionately impacted by the regulation because women are more likely to have family obligations and work part-time, downwardly skewing the median earnings value of the programs they attend. Many commenters expressed that this issue was particularly common among women in the cosmetology sector, where it is very common for women to work part-time as they manage other family obligations.

Discussion: The Department's analysis shows that fewer women will attend failing programs relative to the current baseline (Table 5.21). Under the final regulation, the Department estimates that only 5.5 percent of women who receive title IV, HEA funds would attend programs that are expected to fail the earnings test. Under the current Gainful Employment regulation, 6.4 percent of women who receive title IV, HEA funds attend programs that are estimated to fail the accountability framework. Some women may therefore benefit from this regulation due to the greater educational choices afforded to them, along with the greater amount of title IV, HEA funds they are eligible to receive.

The Department does acknowledge that certain women may not benefit, if they experience low earning outcomes after attending programs that remain open under this regulation but would have failed (and therefore likely closed) under the prior regulation. On net, the Department's analysis suggests there will be a greater number of students who attend programs with relatively lower-earning outcomes relative to the prior rule (Table 5.12). However, it is difficult for the Department to estimate if students benefit or are harmed by this provision because we do not have the ability to determine students' earning outcomes had they not attended such programs.

Furthermore, many of the programs listed by the commenters-- including cosmetology programs, social work programs, and education programs--have lower fail rates under the accountability framework in this final regulation relative to the fail rates under the current regulation (Table 5.18 and 5.19). Again, this means certain women may benefit if they value the greater amount of educational choices and opportunities that will be available to them under this rule relative to the prior regulation. In summary, the Department disagrees with the commenters' assertions because our estimates show that relative to the current baseline, more female students and the programs they attend are likely to have access to title IV funding under the proposed rule, which may benefit certain female students.

Changes: None.

Comments: Some commenters expressed concern that the proposed regulation will negatively impact low-income students. The commenters stated that fewer low-income students will enroll in college as a result of the regulation because they will have less access to the title IV, HEA student aid programs they depend on. Commenters believed that low-income students would therefore be disproportionately harmed because they will have no other way to afford postsecondary education, forcing them to drop out of college or skip higher education entirely.

Discussion: Relative to the current Gainful Employment regulation, the Department estimates that fewer students will attend failing programs (Table 5.12). Additionally, students will continue to receive Federal Pell Grants, even if they attend failing programs. This is because the failing programs in

the final rule will usually only lose access to Federal student loans, which differs from the current Gainful Employment regulation, where students at failing programs would lose access to all types of title IV, HEA aid. This means low-income students who attend failing programs under the final regulation will usually have access to higher amounts of Federal student financial assistance relative to the amount they would have access to under the current policy. To the extent that low- income students pay less tuition as a result (due to the Federal title IV, HEA aid they can receive under this regulation), they may benefit.

Changes: None.

Comments: Some commenters stated that the regulation would harm formerly incarcerated individuals, veterans, and the family members of veterans because these individuals rely on cosmetology programs to re- enter the labor force. The commenter argued that the rule would result in the closure of many cosmetology programs, creating a disproportionate impact on these populations.

Discussion: The Department does not have data to evaluate how the rule would impact veterans, the family members of veterans, and formerly incarcerated individuals. However, the Department specifically examined the estimated impact of the regulation on cosmetology programs (Tables 5.17, 5.18, 5.19, 5.20, 5.27, and 5.28). Relative to the current Gainful Employment regulation, the final rule is estimated to result in fewer failing cosmetology programs. As a result, there will likely be fewer cosmetology program closures as a result of this rule. This may benefit formerly incarcerated individuals and veterans who may desire to enroll in these programs because they will enjoy greater educational choice as a result of this regulation. Therefore, while the Department is unable to evaluate the specific concerns raised by commenters, we believe that the rule will enhance the ability for students to attend cosmetology programs due to the overall reduction in the rule's impact on cosmetology programs.

Changes: None.

Comments: Commenters expressed concern that the proposed regulation would disproportionately impact first-generation college students, asserting that the earnings test will result in the closure of programs that serve these students.

Discussion: The Department notes that assessing the degree to which first-generation students are impacted by the rule is difficult and that commenters provided no data or analysis to support their concerns. While the Department does not have comprehensive data to directly estimate the impact of the regulation on first-generation students, we estimate the impact on less-than-two-year institutions and two-year institutions, which prior research has found could be more likely to enroll first-generation students.\29\ The results are mixed. ED's analysis (Table 5.7) shows that fewer programs at less-than-two-year institutions will fail the earnings test under the proposed regulation when compared against the current regulations but programs at two-year institutions will fail at slightly higher rates under the proposed rule. Additionally, some first-generation students may benefit from the rule in the form of higher earnings, if those students instead enroll in programs that have higher earnings outcomes as a result of this regulation.

\29\ PNPI (2025). First Generation Students in Higher Education. Washington, DC: Postsecondary National Policy Institute. https:// pnpi.org/wp-content/uploads/2025/05/ FirstGenStudents_FactSheet_Apr25.pdf.

Changes: None. Impact on Rural Communities

Comments: Many commenters expressed concern about how the regulation will impact programs in rural areas and communities. Commenters suggested the rule will negatively impact programs in these areas for factors that are outside of their control, including lower costs of living and localized labor market conditions.

Discussion: The Department addressed these comments above in the “Earnings Threshold--Geographic Scope of Data” section. To briefly reiterate, the Department specifically examined the impact of the Regulation on rural colleges, and notes that the regulation will have a slightly larger impact on rural programs and the students who attend them relative to the current regulation (Table 5.10). The Department clarifies that this increased impact on rural communities is driven by the statutory requirement to hold degree programs offered at public colleges accountable for earnings outcomes.

Changes: None. Comments About RIA Methodology

Comments: Some commenters stated that the Regulatory Impact Analysis did not explain how the earnings test would impact programs and colleges and suggested that terms were not well defined. Other commenters expressed that the Department's analysis in the RIA is confusing, inaccurate, and misleading.

Discussion: The Department disagrees with the commenters' assertions. The commenters did not provide any specific details on which parts of the Department's analysis they found confusing, inaccurate, or misleading. The Department notes that it clearly explained the data and methods used to produce its estimates, which are shown in Tables 5.7 to 5.21 in the Regulatory Impact Analysis.

However, to further enhance transparency in the Department's analysis, we included two new tables in the RIA, Tables 5.1 and 5.2. These tables explain how the Department used the PPD:2026 data to create an analytic sample of programs for which its estimates are based on. Furthermore, these tables list which types and number of programs that are anticipated to be covered by the accountability framework in this regulation. All subsequent tables were re-numbered to account for the inclusion of these new tables.

Changes: None.

Comments: Some commenters stated that the income definition used in the Department's analysis in the RIA was unclear. Commenters stated that the definition used to calculate the earnings benchmarks--personal income from wages and salary and personal income from self-employment and farm income--is an incomplete measure of individual income.

Discussion: The Department disagrees with the commenters' assertions. When calculating the earnings benchmarks, the Department includes personal income from wages and salary and personal income from self-employment and farm income. This is the most comprehensive measure of personal income available in the ACS, the dataset the Department will use to calculate the earnings benchmarks. Commenters who believed this measure was incomplete offered no alternative dataset or data that the Department could use as a better measure for earnings. In the absence of any other data, the Department will use the available data from the ACS.

Changes: None. Comments About PPD:2026 Data

Comments: Some commenters expressed that PPD:2026, the dataset used for the Department's analysis, contained errors and inaccuracies. Commenters alleged that the dataset included programs at colleges that did not exist, and suggested that this data would be inappropriate to use for the accountability framework.

Discussion: The Department stated in the Regulatory Impact Analysis that

there are several important differences between the PPD:2026 data and the dataset that will ultimately be used to administer the final rule (Table 5.3). The Department used PPD:2026 for the analysis in this final rule because we will not have the data to implement the final rule until 2027. The Department reiterates that the estimates presented in the Regulatory Impact Analysis are the best available estimates based on currently-available data on programs and earnings.

In response to commenters who expressed the PPD:2026 dataset contained programs for which their college never offered, the Department notes that the Technical Data Appendix clearly explains the process used to include programs in PPD:2026. The data utilizes program-level information on enrollments and title IV, HEA disbursements from COD in a particular program during a particular period of time. Programs where there was at least one title IV enrollee in the program during the 2024 or 2025 award years were maintained in the analytic sample.

The Department reiterates that, to mitigate this concern when the final rule is implemented, all colleges will have an opportunity to review their program completer lists for each program they offer. During this process, colleges will be able to amend completer lists for each program to ensure that the college is held accountable for the correct set of program completers.

Changes: None.

Comments: One commenter representing a college expressed concern about the PPD:2026 data used to produce estimates in the Regulatory Impact Analysis. The commenter was concerned with the way federal loan disbursements were associated with programs, noting that they believed the data inaccurately apportioned federal loan disbursements to a program at their institution. The commenter argued that borrowers who graduated with loans from one program who then re-enrolled in another program were having their loans counted towards the second program. The commenter also expressed confusion regarding the pass/fail indicators in the PPD:2026 dataset.

Discussion: The Department disagrees with the concerns expressed by the commenter. The Technical Data Appendix clearly articulates how loans are tied to programs. Specifically, FSA data include information on which program students were enrolled in when the loan was disbursed and which program and college received the loan. This means loans are not double counted for individuals who enroll in a new program and have loans from a prior program.

Furthermore, the Variable Codebook articulates which programs would pass or fail the proposed regulation; the variable “mstr_obbb_fail_cip2_wageb” indicates whether the program is estimated to pass or fail the earnings test in the proposed regulation. The Department further notes that institutions can reach out to us directly if they have specific questions about how to use the data we produce.

Changes: None.

Comments: One commenter indicated that of the 16 graduate Acupuncture and Herbal Medicine (AHM) programs in the Department's own College Scorecard, 12 have earnings data suppressed due to privacy thresholds. Because of this, the commenter suggests that existing federal data infrastructure cannot produce reliable earnings estimates for the majority of AHM programs, and recommends using alternative data sources.

Discussion: The Department notes that prior data from the College Scorecard is not the dataset that will be used to determine program earnings. Rather, in conjunction with a Federal Agency with earnings data, the Department will assemble new program-level dataset on program earnings, and this newly-assembled data will have a more-robust cohort aggregation process to account for small programs. This will greatly reduce the concern raised by the commenter who believes that AHM programs will not be covered by the earnings test. We decline the commenter's request to use alternative datasets because the earnings data must be pasted on program completers, not broad data based on the earnings of individuals with certain degrees.

Changes: None.

Comments: One commenter requested that the Department impute the earnings of programs with missing earnings data in PPD:2026. The commenter argued that this would provide a better comparison to estimate the potential impact of eliminating the D/E metric on programs.

Discussion: The Department declines the commenter's request. The commenter did not offer a way in which the Department could feasibly impute the earnings outcomes for programs with missing data. In addition, the Department is concerned that the imputation of missing data could produce confusion, as some earnings values would based on statistical estimates rather than the program's actual earnings outcomes from a Federal agency with earnings data. The Department also notes that the specific method used to impute missing earnings data could be highly sensitive to model specifications, further contributing to potential confusion and the possibility of misleading results.

Changes: None.

This section presents the Department's analysis of the anticipated impact of the final regulations. For this analysis, the Department estimated which programs would fail the accountability framework in the final regulations relative to the current regulations, which is the baseline for the analysis. The Department also analyzed the characteristics of these failing programs and the characteristics of students who attend them. The following subsections describe the data and methodology the Department used and the estimated impact of the final regulation on students, programs, and institutions. Data Description

Throughout this RIA, we use data from a modified version of the 2026 Program Participation Data (PPD:2026) that the Department compiled for the negotiated rulemaking sessions. This data was made publicly available on the Department's website prior to the January 2026 negotiated rulemaking sessions, along with additional details. You may find the data at www.ed.gov/laws-and-policy/higher-education-laws-and- policy/higher-education-policy/negotiated-rulemaking-for-higher- education-2025-2026.\30\

\30\ U.S. Department of Education AHEAD Session 2 Program Performance Data Fact Sheet, https://www.ed.gov/media/document/ ahead-session-2-program-performance-data-fact-sheet-112902.pdf; AHEAD Session 2 Program Performance Data Variable Codebook, https:// www.ed.gov/media/document/ahead-session-2-program-performance-data- variable-codebook-112904.pdf; AHEAD Session 2 Program Performance Data Technical Appendix https://www.ed.gov/media/document/ahead- session-2-program-performance-data-technical-appendix-112901.pdf.

PPD:2026 was assembled by combining data from a variety of public and private sources, including the Integrated Postsecondary Education Data System (IPEDS), the College Scorecard, the IRS, FSA, and the ACS. It includes information on enrollments, earnings, and title IV, HEA disbursements, among other variables.

The unit of analysis in PPD:2026 is the unique combination of institutional ID (opeid6), credential level (credlev), and four-digit classification of instructional program (CIP) code (cip4).31 32 When necessary, OPEIDs are

linked to UNITIDs using the UNITID of the main campus, identified using the College Scorecard crosswalk files.\33\ The universe of programs in PPD:2026 includes all programs eligible for title IV, HEA funds that had at least one title IV enrollee reported to the National Student Loan Data System (NSLDS) during the 2023-24 or 2024-25 award years. In total, PPD:2026 includes information for 209,321 unique programs offered at 5,096 unique higher education institutions.

\31\ Programs are defined using the unique combination of institutional ID (OPEID6), credential level, and 4-digit CIP codes. In almost all cases, 4-digit CIP code titles align with the 2010 CIP code taxonomy. In some cases, 4-digit CIP codes appear only in the 2020 CIP code taxonomy. In these cases, the 2020 CIP code taxonomy is used to title the program. Programs with CIP codes that do not appear in the 2010 or 2020 CIP code taxonomies are dropped. This removes fewer than 20 individual programs, representing less than 0.006% of all higher education programs in the final data set.

\32\ Credential levels are defined by the following eight categories: (1) Undergraduate certificate programs, (2) Associate degree programs, (3) Bachelor's degree programs, (4) Post- Baccalaureate degree programs, (5) Master's degree programs, (6) Doctoral programs, (7) First-Professional Programs, and (8) graduate certificate programs.

\33\ UNITIDs are the unique institution identifiers used in IPEDS data.

There are two key differences between the data used in this RIA and the public dataset available on the Department's website. First, the data used in this RIA contains information on the specific counts of enrollees regardless of program size. This differs from the publicly released version of PPD:2026, where student counts fewer than 20 are privacy-suppressed or perturbed. Estimates in this RIA may therefore differ slightly from those using the publicly released version of PPD:2026. Additional information is available in the technical documentation for PPD:2026 on the Department's website.\34\ Second, the data used in this RIA contains additional variables from publicly available sources (such as IPEDS and the College Scorecard) that were not originally included in the PPD:2026 data released by the Department in January. The Department needed to include these additional variables in its analysis of the final rule to estimate the impact of certain provisions, such as the provision allowing failing programs to voluntarily remove themselves from the Federal student loan program for a period of five years (discussed in the “Department Authority (Including GE and Quality Assurance Authority)” section).

\34\ For more information, see: www.ed.gov/media/document/ahead- session-2-program-performance-data-technical-appendix-112901.pdf.

The analysis in this RIA uses data from PPD:2026. These data include information on 209,321 individual programs. We then limit these data to exclude programs that the Department estimates are unlikely to be subject to the accountability framework in this rule. This includes:

Programs located in U.S. territories with unavailable ACS data; \35\

\35\ This includes programs at institutions located in U.S. territories that primarily enroll in-territory students for which ACS data are unavailable.

Programs with no Federal loan participation from 2021- 2025; \36\

\36\ This restriction removes programs that received $0 in Federal student loan disbursements for each award year between 2021 and 2025. This method slightly differs from the process outlined in this regulation, where programs at institutions that did not participate in the Federal student loan program for the prior five award years would not have an outcome calculated. This analysis uses program-level participation since we observe loan disbursement data at the program level.

Graduate programs in states where ACS data will likely be unavailable or unreliable; \37\ and

\37\ This includes graduate programs at colleges that primarily enroll in-state students located in states and fields where the ACS had fewer than 30 respondents used in the same-state, same-field earnings benchmark (see Table 5.5 for further explanation).

Programs that are estimated to not meet the minimum size requirement to generate a program earnings measure.\38\

\38\ To determine if a program would likely meet the minimum size requirement, we used PPD:2026 and summed the number of title IV completers from the 2021-22 to 2024-25 award years. Following the aggregation process described in the “Cohort Expansion” section, we summed together completers from the same college and credential level who shared the same four-digit or two-digit CIP code and completed within during the 2021-22, 2022-23, 2023-24 and 2024-25 award years. If this value did not exceed 30 unique title IV completers, we assume the program (defined at the OPEID6 x CREDLEV x CIP4 level) would not reach the minimum size requirement needed to be included under the final rule and therefore removed it from the sample. Additionally, approximately 160 cosmetology programs and massage therapy programs will receive two or more years of delay because they are small and additional cohorts of students with earnings under the “No Tax on Tips” policy will be needed for the cohort aggregation process (described in the “Earnings of Program Completers--Use of IRS Data” section). These small programs are excluded from the final sample of programs; however, we note that these programs will eventually be subject to the earnings test when their cohort sizes become large enough in future years, which would occur between 2030 and 2032.

Table 5.1 reports the number of programs removed by each of these restrictions as well as the final set of programs included in the analysis. Approximately 30 percent of all programs are expected to be covered by the accountability framework in this final rule, and these programs enroll approximately 79 percent of all title IV, HEA students and receive 84 percent of all annual title IV, HEA disbursements (Table 5.2). The fail rates presented in the “Impact of the Final Regulations on Institutions” and “Impact of the Final Regulations on Programs” sections are therefore based on the roughly 61,900 programs in PPD:2026 that the Department anticipates will be subject to the accountability framework in this final rule.\39\ The other programs in PPD:2026, which, based on the Department's analysis, are unlikely to be covered by the accountability provisions in this regulation, are not included in the numerator or denominator for fail rates displayed in these sections.

\39\ In other words, programs that are not expected to be covered by the accountability framework are not included in the pass rates or fail rates in the analysis presented in the “Impact of the Final Regulations on Institutions” section.

BILLING CODE 4000-01-P

[GRAPHIC] [TIFF OMITTED] TR01JY26.040

[GRAPHIC] [TIFF OMITTED] TR01JY26.041

Data Limitations & Assumptions

The data used in this RIA differ slightly from what the Department would use to evaluate programs under the current regulations and these final regulations. These differences are summarized in Table 5.3. We make several assumptions in our analysis to account for these differences and note that the estimates in this RIA may slightly differ from the actual rates.

[GRAPHIC] [TIFF OMITTED] TR01JY26.042

BILLING CODE 4000-011-C

First, our estimates may slightly overcount the share of programs that ultimately fail the accountability framework in the current regulations

and final regulations because programs in PPD:2026 are identified by a unique combination of 6-digit OPEID, credential level, and 4-digit CIP code. However, in the current and final regulations, programs are identified by a unique combination of 6-digit OPEID, credential level, and 6-digit CIP code. We therefore assume that earnings outcomes of programs within the same 4-digit CIP code, credential level, and institution are equally distributed across (unobserved) 6-digit CIPs. Fail rates in this RIA may vary from the actual fail rates if different programs (defined at the 6-digit CIP level) nested within the same overarching 4-digit CIP have different earnings outcomes. Some research has shown that program earnings outcomes may vary across 6-digit CIP codes within the same 4-digit CIP code; however, the Department believes that any effect this has on our estimates should be small because approximately 83 percent of 4-digit CIP codes have only a single 6-digit CIP code nested within it.\40\

\40\ Blagg, K., (2026). Measuring Program-Level Outcomes in Higher Education. Washington, DC: The Urban Institute. www.urban.org/sites/default/files/2026-01/Measuring_Program- Level_Outcomes_in_Higher_Education.pdf; Soliz, A., & McCann, C., (2026). Earnings of Programs Can Vary Widely--Even Within the Same Field of Study. Washington, DC: PEER Center. www.peer-center.org/ research/earnings-vary-widely-within-field.

Second, student completer cohorts in PPD:2026 are constructed differently than under the current and final regulations. Specifically, the cohort for earnings in PPD:2026 includes title IV completers from two pooled award years; these data were drawn from the College Scorecard for expediency and therefore use the cohort construction from that source. Under the current regulation, cohorts would include title IV completers from two or four pooled award years, depending on program size. Under the final regulation, cohorts will generally include title IV completers from a single award year unless the program does not meet the minimum size threshold (discussed in the “Cohort Expansion” section above), in which case cohorts will be aggregated with similar programs for up to three prior award years, until a statistically reliable cohort size is achieved. We therefore must assume that the cohort aggregation processes for the current and final regulations would result in programs having similar earnings as the cohorts used in PPD:2026.

Third, our estimates use the earnings outcomes from the single pooled cohort of completers, but in the current and final regulations, programs face sanctions only if they fail the accountability framework in two out of three consecutive years.\41\ A single cohort is used for the analysis in this section because PPD:2026 does not include multiple, consecutive years of program-level earnings outcomes. This is another reason why our estimates may slightly overcount the share of programs that fail the accountability framework.\42\

\41\ In the current regulation, GE programs that fail the D/E test in two out of three consecutive years or GE programs that fail the EP test in two out of three consecutive years lose access to all title IV, HEA funds. Under the final regulations, programs that fail the EP test in two out of three consecutive years lose access to title IV Federal student loans.

\42\ We anticipate this issue will be very small given that program earnings outcomes are based on completers who exited a program approximately six calendar years prior to the date in which earnings are measured. For this reason, there is little, if anything, colleges could do to alter the earnings outcomes of their former students. In other words, we anticipate that failing the earnings test one year will be highly correlated with failing the earnings test in the subsequent year.

Fourth, our analysis in this section (“5. Impact of the Final Regulations”) assumes that students who attend failing programs will not switch to a different, non-failing program.\43\ Program-switching is not accounted for in this section because the Department is unable to precisely estimate which programs individual students will select when switching. However, we expect this to have a minimal impact on our estimates because this assumption is applied consistently to our estimates of both the current and final regulations.\44\

\43\ The Department's analysis in the “5. Impact of the Final Regulations” section does not account for the possibility of program switching. This is because the analysis in this section is presented at specific fields of study and credential levels, and the Department is unable to predict how students may switch across specific types of fields of study and credentials. This differs from the assumptions made in the “7. Net Budget Impact” section, which does account for the possibility of program switching when estimating the budgetary impact of the final rule. The “7. Net Budget Impact” section can account for program switching because the estimates are derived with assumptions using broad volume-based groups that students may switch to; the budget estimates are not disaggregated by specific fields of study or credential levels.

\44\ Our estimates could differ from the true impact of the final regulation if students would differentially switch programs under the final regulation relative to how they would switch programs under the current regulation. This could occur, for example, if under the current regulation students are more likely to drop out of college (due to losing access to both Pell Grant and Federal student loan eligibility) relative to the extent that students drop out of college (rather than switch programs) under the final rule, since under the final rule students would lose access to Federal student loans only.

Another caveat is that earnings are missing for many programs in PPD:2026, usually due to the IRS's privacy protocols.\45\ Earnings data will, however, be collected for many of these programs because cohorts will be aggregated to include more students under the final regulation. As discussed in the prior section, the Department anticipates that roughly 61,900 programs will be subject to the accountability framework in this final rule (Table 5.1). Approximately one third of these programs (32 percent) have missing earnings data in PPD:2026, but the Department expects these programs will have earnings data available when the final rule is implemented due to the expanded cohort aggregation process (Table 5.4). Rates of missingness are higher (approximately 59 percent) for GE programs covered by the accountability framework in the final rule (Table 5.4).

\45\ The IRS (which was the agency that provided the Department with earnings data used in PPD:2026) is usually unable to provide the Department with median earnings estimates for programs where there is a relatively small number of working title IV recipients with available tax records. In some cases, the IRS may be unable to provide the Department with median earnings estimates depending on the distribution of earnings within programs. This results in many small programs having unobserved (missing) program earnings.

To estimate the pass and fail rates for programs in PPD:2026 where earnings data are missing, we assume that these programs fail the accountability framework at equivalent rates as similar programs with reported earnings data in PPD:2026. Specifically, using the subset of programs where earnings data are available, we calculate the fail rates within each sector, broad field of study, credential level, and institutional level. Using those rates, we then assume that programs with missing earnings data will fail the accountability frameworks under the current and final regulations at the same rate as programs with reported earnings data from the corresponding sector, broad field of study, credential level, and institutional

level.\46\ This method for estimating pass and fail rates slightly differs from the preliminary analysis presented by the Department during the negotiated rulemaking sessions in January 2026 (available at: www.ed.gov/media/document/2025-ahead-results-of-earnings-test-and- ge-changes-112932.pdf) because that preliminary analysis excluded programs with missing earnings data.\47\

\46\ This means our estimates may undercount the true share of programs that fail the earnings test in the final regulation if the earnings of small programs are systematically lower than the earnings of larger programs. Conversely, our estimates may overcount the true share of programs that fail the earnings test if the earnings of small programs are systematically higher than the earnings of large programs.

\47\ In most cases, the Department's preliminary analysis presented during the negotiated rulemaking sessions in January 2026 are within a fraction of a percentage point from the estimates presented in this RIA. The marginal difference in estimates is because the preliminary analysis presented at the negotiated rulemaking sessions excluded programs with missing earnings, whereas the analysis in this RIA includes these programs and assumes they pass and fail the accountability framework at equivalent rates as similar programs (programs in the corresponding sector, broad field of study, credential level, and institutional level) with observed earnings data.

BILLING CODE 4000-01-P

[GRAPHIC] [TIFF OMITTED] TR01JY26.101

[GRAPHIC] [TIFF OMITTED] TR01JY26.102

BILLING CODE 4000-01-C

Lastly, the Department's analysis may slightly overestimate the share of programs and students that fail the accountability framework under the final rule because the Department is unable to accurately incorporate two policies into its analysis, both of which are described in the “Low-earning outcome programs (Sec. 668.603)” section above. First, the final rule includes a teach-out provision that allows institutions to continue receiving title IV, HEA funds if they agree to an orderly program closure. Failing programs that exercise this teach- out option can continue to receive title IV, HEA funds for the lesser of three years or the full-time normal duration of the program, meaning students currently enrolled in these failing programs would not be immediately impacted.

Second, the final rule includes an appeals process that allows institutions to appeal the Department's determination for failing programs. If institutions successfully appeal, those programs initially identified as failing would not lose eligibility for Federal student loans.\48\ As estimated in the PRA, the Department anticipates that approximately 1,000 programs may appeal the determination annually. However, it is not possible for the Department to estimate how many of these appeals may be successful. Therefore, we are unable to account for the possibility that the appeals process reduces the program fail rates in this RIA.

\48\ The inability to account for these factors may ultimately result in a slight overestimate in the share of programs, students, and title IV, HEA student aid disbursements impacted under the final rule.

Throughout the RIA, the Department estimates the share of programs that fail the accountability framework under the current and final regulations to determine the net-effects of the final regulations relative to the baseline.\49\ We first estimate which programs would fail the earnings premium measure and D/E metrics under the current regulations. Consistent with the current GE regulation, we count a GE program as failing if it failed either the earnings premium measure or the D/E metric according to our estimates of these measures using PPD:2026.

\49\ While the baseline assumes the current regulations are in effect, we note that no program has actually failed the current regulations at the time this regulation is published because the EP and D/E metrics (as defined under the current regulations) have not yet been computed by the Department. However, in the absence of this final rule, these rates would be calculated, which is why we use the impact of the current regulation as the baseline to judge the impact of the final rule.

GE programs are counted as failing the earnings premium measure under the current regulation if the median earnings of program graduates is below the earnings threshold, which is defined as the median annual earnings of working individuals aged 25 to 34 whose highest level of educational attainment is a high school diploma (or equivalent) in the relevant geographic area.\50\ For this calculation, we used the three-year median earnings of title IV program graduates in the labor market who completed during the 2014-15 and 2015-16 pooled award years, obtained from the IRS.\51\ To calculate the earnings threshold, we used the median annual earnings of working high school graduates using the 2023 ACS 5-Year Estimates, obtained from IPUMS.\52\ All monetary values were adjusted to constant 2024 dollars using the CPI-U.

\50\ Under the current regulations, for programs that enrolled more than 50 percent of their students from the state where the institution is located, the relevant geographic area is the state in which the college is located. For programs that enroll 50 percent or less of their students from the state where the institution is located, the relevant geographic area is the entire United States. In our analysis, we do not observe the share of enrollees in a program that are from out of State. We proxy for program-level in- State enrollment shares using institution-level data on the share of students across the entire institution who are from in-State. To determine whether an institution enrolls more than half its students from out-of-state, we use each enrolled student's address reported in the student's most recently reported FAFSA relative to the award year being evaluated.

\51\ These data were collected by the Department for the original version of the PPD during development of the current GE/FVT regulation. It is the only cohort of students for which readily available earnings data match the requirements of the current GE/FVT regulation. The Department does not have more recent data that match these requirements. Note that one of the measurement years for earnings was during the COVID-19 pandemic. Specifically, title IV completers from the 2014-15 award year had their earnings measured during the 2019 calendar year, and title IV completers from the 2015-16 award year had their earnings measured during the 2020 calendar year. Program graduates who were enrolled in postsecondary education at the time earnings were measured are excluded from this calculation. The median earnings value includes statistical noised added by the IRS to protect student privacy.

\52\ Earnings were defined as the combined sum of personal income from wages and salary (incwage) and personal income from self-employment and farm income (incbus00). The median earnings value is taken using individuals who live in the relevant geographic area (e.g., the corresponding state, or nationally), who have the relevant educational attainment level (e.g., those who only have a high school diploma or equivalent with no postsecondary education), who are between 25-34 years old (inclusive), and who have a positive, non-zero income. Appropriate survey weights were utilized to ensure estimates were representative at the national and state levels.

Next, we counted GE programs as failing the D/E metric if they failed either the Annual Earnings Rate measure or the Discretionary Earnings Rate measure under current regulation.\53\ For this calculation the Department used program-level data on cumulative student debt from the College Scorecard for individuals who completed during the pooled 2017-18 and 2018-19 award years.\54\ To

calculate the annual loan payment amount (which is used in both the Earnings Rate Measure and Discretionary Earnings Rate measure), we assumed a 4.45 percent interest rate on loans for undergraduate programs and a 6.23 percent interest rate for graduate programs.\55\ To calculate the denominator for the Annual Earnings Rate measure and the Discretionary Earnings Rate measures, we used the same program-level earnings measure described above. When relevant, we used 150 percent of the Federal Poverty Guidelines for a single person in 2024, which was $22,590.\56\

\53\ The Annual Earnings Rate measure is defined as Annual Earnings Rate = (Annual Loan Payment)/(Annual Earnings); the Discretionary Earnings Rate measure is defined as Discretionary Earnings Rate = (Annual Loan Payment)/(Discretionary Earnings). Under current regulation, programs are counted as failing the D/E metric if the Annual Earnings Rate measure exceeds 8% or if the Discretionary Earnings Rate measure exceeds 20% in two out of three consecutive years.

\54\ To calculate the Annual Loan Payment amount, the Department used the following amortization periods: undergraduate certificate, associate degree, post-baccalaureate certificate programs, and graduate certificate programs are amortized over 10 years; bachelor's and master's degree programs are amortized over 15 years; and doctoral and first professional degree programs are amortized over 20 years. These differing amortization periods account for the typical outcome that borrowers who enroll in higher-credentialed programs (e.g., bachelor's and graduate degree programs) are likely to have more loan debt than borrowers who enroll in lower- credentialed programs and, as a result, are more likely to take longer to repay their loans. The amortization rates mirror those used in the 2014 and 2023 prior rules.

\55\ These interest rates were determined by taking a weighted average of the interest rates on Undergraduate Unsubsidized Stafford Loans, Graduate Stafford Loans, and Grad PLUS Loans between 2016 and 2019, which were the available interest rates on these loans around the time that borrowers completed their programs.

\56\ The Federal Poverty Guideline is used to calculate the denominator of the Discretionary Earnings Rate measure.

The Department estimated which programs would fail the revised earnings premium measure under the final regulation using PPD:2026. Programs are counted as failing if the median earnings of working program graduates are below the relevant earnings threshold. For this calculation, we used the four-year median earnings (from the College Scorecard) of title IV program graduates who completed during the 2017- 18 and 2018-19 pooled award years and were working at the time earnings was measured.\57\

\57\ Specifically, title IV, HEA completers from the 2017-18 award year had their earnings measured during the 2022 calendar year, and title IV, HEA completers from the 2018-19 award year had their earnings measured during the 2023 calendar year. Earnings were inflation adjusted to 2024 using CPI-U. Program graduates who were enrolled in postsecondary education at the time earnings were measured are excluded from this calculation. Individuals are determined to be “working” if they had positive income reported to the IRS from wages, salary, or self-employment during the calendar year earnings were measured. The median earnings value includes statistical noise added by the IRS to protect student privacy.

The relevant earnings thresholds for each program are listed in Table 5.5. There are six different earnings thresholds in which a program could be judged under the final regulation, including:

In-State High School (HS). The median earnings of individuals aged 25-34 in the state where the college is located, who are working,\58\ and have only a high school diploma or its recognized equivalent.

\58\ For all six ETs in the final regulation, we determine an individual was working if the individual reported positive, non-zero personal income from wages, salary, or self-employment income (including farm income) during the year. Individuals who reported that they were currently unemployed (at the time they completed the survey) but had otherwise worked during other parts of the year (meaning they had positive personal income from wages, salary, or self-employment) are still counted as working.

National HS. The median earnings of individuals aged 25-34 in the entire United States, who are working, and have only a high school diploma or its recognized equivalent.

Same-State, Same-Field Bachelor's degree (BA). The median earnings of individuals aged 25-34 in the state where the college is located, who are working, and have a bachelor's degree in the same field of study.

Same-State BA. The median earnings of individuals aged 25- 34 in the state where the college is located, who are working, and have a bachelor's degree.

National Same-Field BA. The median earnings of individuals aged 25-34 in the entire United States who are working and have a bachelor's degree in the same field of study.

National BA. The median earnings of individuals aged 25-34 in the entire United States who are working and have a bachelor's degree. BILLING CODE 4000-01-P

[GRAPHIC] [TIFF OMITTED] TR01JY26.044

To calculate each of the six earnings thresholds, we used data from the 2023 ACS 5-Year Estimates, obtained from IPUMS. We defined earnings using the same approach described in the “Methodology for Current Regulation Calculations” section.\59\ We computed each specific earnings threshold by using the corresponding group of individuals aged 25-34 who live in the relevant geographic area (e.g., the corresponding state, or nationally), who have the relevant educational attainment level (e.g., high school diploma/recognized equivalent or bachelor's degree in the relevant field of study), and who are not currently enrolled in college.\60\

\59\ Specifically, earnings are defined as the combined sum of personal income from wages and salary (incwage) and personal income from self-employment and farm income (incbus00), and are adjusted to 2024 dollars using CPI-U. Individuals with $0 or non-positive earnings are omitted from the medians.

\60\ Appropriate survey weights were utilized to ensure estimates were representative at the National and state levels. For institutions that enrolled fewer than 50 percent of their students from in-state, the relevant geographic area is the State in which the college is located. For institutions that enrolled 50 percent or more of their students from out-of-state, the relevant geographic area is the entire United States. To determine if students are from in-state or out-of-state, we use each enrolled Title IV student's address reported in the student's most recently reported FAFSA relative to the award year being evaluated.

A majority of programs are compared against the in-state earnings thresholds, reflective of the fact that most postsecondary students are in-state residents of the college they attend. Summary statistics on the share of programs that compared against the in-state thresholds are shown in Table 5.6. This reveals variation in the rate at which certain types of programs are judged against the in-state thresholds. For example, undergraduate Culinary & Personal Services programs are most likely to be compared to the in-state earnings threshold (96 percent), whereas graduate-level Religious Studies programs are least likely to be compared to the in-state earnings threshold (55 percent).

[GRAPHIC] [TIFF OMITTED] TR01JY26.106

[GRAPHIC] [TIFF OMITTED] TR01JY26.107

BILLING CODE 4000-01-C

For the “Same-State, Same-Field BA” and “National Same-Field BA” earnings thresholds, we determined programs to have the same field of study if the graduate program shared the same 2-digit CIP as the bachelor's degree program.\61\ Monetary values were adjusted to constant 2024 dollars using the CPI-U. In a small number of cases we could not reliably calculate the “Same-State, Same-Field BA” earnings threshold due to the small number of individuals sampled in the ACS in the correct age range who had a bachelor's degree in a specific field and were located in the relevant state.\62\ In these cases, we removed these graduate programs from the analysis because they are expected to automatically pass the earnings test, consistent with the process described in the “Earnings threshold--Data Source” section.\63\ The Department estimates that this removes approximately 2,650 graduate programs from the analysis (Table 5.1).

\61\ We used the variable DEGFIELD from the ACS to crosswalk fields of study to two-digit CIP codes. To do so, we subtracted 10 from the value of DEGFIELD to match the corresponding two-digit CIP code. One exception was DEGFIELD=38 (Military Technologies), where we had to subtract 11 (rather than 10) to achieve the corresponding two-digit CIP code. For more information, see https://usa.ipums.org/ usa-action/variables/DEGFIELD#codes_section.

\62\ If an earnings threshold would be based on fewer than 30 individuals in the ACS, we did not calculate the earnings threshold because we believed it would not be statistically reliable or representative. As described earlier in this RIA, these programs are removed from the analysis.

\63\ These programs are expected to automatically pass the earnings test because the earnings threshold they would be compared against is $1.

Programs are counted as failing the revised earnings premium test under the final regulations if the median earnings of program graduates were below the relevant earnings threshold in Table 5.5. Accounting for Additional Provisions in the Final Regulation

The Department's impact analysis also accounts for four provisions that are included in this final rule. First, the final regulations modify the standards of administrative capability as described in the “Administrative Capability and Consequences” section. Under this provision, all programs (GE- and non-GE programs alike) that fail the revised earnings premium test will lose access to Pell Grants (in addition to losing access to Federal student loans) if:

More than half of title IV, HEA funds disbursed to an institution are to students attending programs that fail the revised EP test under the final regulations; or

More than half of title IV students enrolled at an institution are in programs that fail the revised EP test under the final regulations.

To account for this provision, the Department used PPD:2026 to calculate the total amount of title IV, HEA funds (Pell Grants and Federal student loans) disbursed to failing programs during the 2024-25 award year. Then, we calculated the share of title IV students and title IV, HEA funds in failing programs at each institution. If either of those shares exceeded 50 percent, the institution is considered as failing the standards of administrative capability requirements, and its failing programs are counted as losing eligibility for both Federal student loans and Pell Grants.\64\

\64\ Our estimates may slightly overcount the share of programs that lose access to Pell Grants under the final regulation's revisions to the standards of administrative capabilities. This is because the revised standards in the final regulation apply to institutions after three years of failing the revised earnings premium metric. However, as described above, we do not observe multiple, consecutive years of program-level earnings data in PPD:2026. Thus, our estimates may slightly overcount the share of programs that are impacted by the final regulation's revisions to the standards of administrative capabilities since we assume they fail this standard after failing the accountability framework in a single year.

Second, as described above in the “Department Authority (Including GE and Quality Assurance Authority)” section, the accountability framework in this rule will not apply to programs at institutions that have not participated in the Direct Loan program for the prior five award years. To account for this provision, the Department removed programs with $0 in federal student loan disbursements between 2020-21 and 2024-25 from this analysis.

Third, as described in the “Department Authority (Including GE and Quality Assurance Authority)” section, the final rule allows colleges to voluntarily remove programs from the Federal student loan program for a period of five years after the initial year a program fails the revised earnings premium test. Voluntarily removing a program from the Federal student loan program will mean the programs will not be at risk of failing the standards of administrative capability requirements. As a consequence, these failing programs can preserve their Pell Grant eligibility by pre-emptively opting out of the Federal student loan program after the first year they fail the revised earnings premium test.

The Department used PPD:2026 and supplemental data from IPEDS to estimate which failing programs would likely choose to opt out of the Federal loan program under this provision. For this estimate, we began by calculating which programs and institutions would fail the revised earnings premium test under the final rule and the standards of administrative capability requirements, respectively. Then, we used Pell Grant disbursement data from PPD:2026 to calculate the average Pell Grant award disbursed to Pell-recipients in each program during the 2024-25 award year (or from the 2023-24 award year, if data from 2024-25 is missing). Lastly, using institutional tuition data from IPEDS for the 2024-25 award year (or the 2023-24 award year, if data for 2024-25 were missing) we calculated the weighted average of the in- state and out-of-state average net tuition and fees.\65\

\65\ We calculate an institution's average tuition as the weighted average of the in-state and out-of-state average net tuition price, where the average tuition and fees are weighted by the share of in-state and out-of-state students at the institution.

For the subset of failing programs at institutions that also fail the standards of administrative capability requirements, we assume that colleges would voluntarily remove these programs from the Federal student loan program if their Pell Grant recipients received an average Pell Grant disbursement within 120 percent of the institution's weighted average tuition and fees. This is because these programs are likely able to continue operating on Pell Grant funding alone, given the relative closeness between the institution's tuition and fees and the average Pell Grant awards received by students in the failing programs. Ultimately, for this subset of failing programs, our analysis assumes those programs will only lose eligibility for Federal student loans because it is likely

those programs will choose to opt-out of the Federal student loan program and therefore avoid the penalties associated with failing the standards of administrative capability requirements.

Fourth, as described in the “Earnings of Program Completers--Use of IRS Data” section, the final rule delays the implementation of the accountability framework for a subset of programs associated with predominantly-tipped occupations. The programs that qualify for the delay are listed in Table 5.22.\66\ The Department accounts for this provision in the following manner. First, for our main analysis presented in the “Impact of the Final Regulations on Institutions”, “Impact of the Final Regulations on Programs”, and “Impact of the Final Regulations on Students” sections, these programs are included, meaning they are not counted as either passing or failing the earnings test. This is because the analysis in those sections assumes this rule is fully implemented (i.e., it assumes the delay period has ended).\67\ However, in the “Impact of the Delayed Implementation for Certain Programs” section, we conduct a secondary analysis where we examine how “first-year” fail rates (i.e., the fail rates of programs in 2028/29) are likely to differ from the fail rates when the delay provision has expired. In that analysis, we remove from the sample programs that qualify for the delay provision, thereby showing the rule's impact on programs after those programs are excluded from the earnings test.

\66\ See section “Earnings of Program Completers--Use of IRS Data” for an explanation of how this list of programs was assembled.

\67\ In other words, that analysis assumes the rule has been in effect for at least four years, such that the delay period is over and most programs covered by the rule, including those listed in Table 5.22, have been subject to the accountability provisions in this regulation. One exception is 160 small cosmetology and massage therapy programs, which continue to be excluded from the main analysis because these programs will not be subject to the earnings test until a later year between 2030 and 2032.

Although the current and final regulations establish an accountability framework for individual programs, we considered their effects on institutions with two approaches: by estimating how many programs fail within each level of institution type, and how many institutions will see high and low rates of program failure. For this analysis, higher education institutions are categorized into levels by their predominant degree offered (Less-than 2-year; 2-year; 4-year; Exclusively Graduate-degree Granting).

Examining the rate of program failure within levels, we estimate that 30.2 percent of programs at the less-than 2-year level would fail under the current regulation, whereas only 23.7 percent are estimated to fail under the final regulation (Table 5.7). At the two-year, four- year, and graduate levels, we estimate an increase in the share of programs that fail the accountability framework under the final regulation. Overall, the Department estimates that slightly more programs will fail under the final regulation relative to the current regulation (5.2 percent vs. 4.6 percent), but these programs enroll fewer students than those that fail under the final regulation (4.2 percent vs. 4.7 percent), which contributes to a smaller loss in title IV disbursements (4.1 percent vs. 5.2 percent). This is consistent with the estimated net cost from the final regulation in the net budget impact section, as shown in Tables 7.1A and 7.1B. BILLING CODE 4000-01-P

[GRAPHIC] [TIFF OMITTED] TR01JY26.046

Next, we estimated how many programs within each institution would fail the accountability framework under the current and final regulations and then assigned institutions to one of five groups based on the degree to which students attend programs that fail (Tables 5.8 and 5.9). For this analysis we disaggregate institutions by level and sector (Public; Private Non-Profit; Proprietary).\68\

\68\ Estimates for the current regulation are based on enrollments in GE programs with non-missing earnings data, while estimates for the final regulation are based on enrollments in all types of programs with non-missing earnings data.

This analysis shows that approximately 94 percent and 98 percent of public and private non-profit institutions, respectively, have 0 percent of their enrollment in failing GE programs under the baseline (Table 5.8). At the other end of the distribution, we find that less than 1 percent of public and private non-profit institutions have all (100 percent) of their enrollment in failing GE programs under the baseline.

These rates noticeably differ from shares estimated for the final regulation (Table 5.9). Sixty-nine percent and 81 percent of public and private non-profit institutions, respectively, are unaffected by the final regulation (these institutions have 0 percent of their enrollment in failing programs). On the other end of the distribution, about 3 percent of private non-profit institutions have 100 percent of their enrollment in failing programs, which is four times the rate as the current regulation.

[GRAPHIC] [TIFF OMITTED] TR01JY26.047

[GRAPHIC] [TIFF OMITTED] TR01JY26.048

The Department also estimated the impact of the final regulation on special types of institutions and those with unique missions, including Historically Black Colleges and Universities (HBCUs), Tribally Controlled Colleges and Universities (TCCUs), Minority Serving Institutions (MSIs), religiously affiliated institutions, rural institutions, and institutions located in U.S. territories, and single- program institutions (Table 5.10). We find that a larger share of programs at HBCUs, MSIs, religiously affiliated colleges, rural institutions, and institutions in U.S. territories will be more negatively impacted by the final regulation relative to the current regulation. Single-program institutions are less impacted under the final regulation because fewer programs at these institutions are estimated to fail the accountability framework relative to the baseline.

[GRAPHIC] [TIFF OMITTED] TR01JY26.049

The Department also estimated the share of institutions that fail the standards of administrative capability requirements under the current and final regulations (Table 5.11). Overall, similar shares of institutions will fail the standard (15.4 percent and 14.5 percent under the final and current regulations, respectively). Although these are similar shares, the penalty for failing the standard is less under the final regulation. Under the final rule, only the subset of failing programs at institutions that fail the standards of administrative capability requirements lose access to all title IV, HEA aid. This differs from the current regulations, where all programs at an institution that fails to meet the standard would lose access to all title IV, HEA funds.

Proprietary institutions are expected to fail the standards of administrative capabilities at the highest rates. Under the final rule, the Department estimates that approximately 43 percent of proprietary institutions are estimated to fail. One reason the proprietary sector has higher fail rates is because these institutions offer fewer programs, making it more likely that failing programs will account for a majority of enrollment or title IV HEA funds at the institution. Less than two-year proprietary institutions offer an average of just 5 programs (Panel C, column 4). If just one or two programs at one of these institutions fails the accountability framework under the final regulations, the institution has a higher probability of failing the standards of administrative capabilities than other institutions that offer a broad range of programs where enrollments and title IV, HEA funds are more disbursed.

[GRAPHIC] [TIFF OMITTED] TR01JY26.050

The Department estimated the share of programs by credential level that will be impacted by the final regulation (Table 5.12). We estimate the final regulation will result in a slight increase in the total share of programs that would fail the accountability framework, growing from 4.6 percent of all programs under current regulation to 5.2 percent under the final regulation (Panel A). This increase is driven by associate's, bachelor's, master's, and professional degree programs. Many of these programs are offered at institutions that are exempt from the accountability framework under the current regulation but are now subject to it under the final regulation. In contrast, undergraduate and graduate certificate programs are expected to fail the accountability framework at lower rates under the final regulations. These reductions in fail rates, however, are not enough to offset the large increase in expected fail rates among associate's, bachelor's, master's, and professional degree programs, resulting in a net increase in the share of programs that fail.

In terms of students (Panel B), we estimate the final regulation will result in a slight reduction in the total share of students enrolled at programs that would fail the accountability framework, dropping from 4.8 percent of students under the current regulation to 4.2 percent under the final regulation. Even though a larger share of programs would fail under the final regulation relative to the baseline, those programs are smaller and enroll fewer students than programs that fail under the current regulation on average.

[GRAPHIC] [TIFF OMITTED] TR01JY26.051

When looking at the effect of the final regulation on sectors (Table 5.13), we estimate that the public and non-profit sectors will experience a net-increase in the share of programs that fail. This is because many programs offered in these sectors are exempt from the accountability framework under the current regulation. Conversely, under the final regulations, we estimate that the proprietary sector will experience a reduction in the share of programs expected to fail-- primarily driven by the reduction in failing undergraduate certificate programs. This is largely because under the current regulations, undergraduate programs at proprietary institutions are subject to a more-punitive earnings premium measure relative to the final regulations.\69\

\69\ By “more-punitive”, we mean the earnings premium metric under the current regulations uses median program earnings measured 3-years after completion and includes the earnings of non-working individuals. Conversely, the revised earnings premium metric in the final regulations uses median program earnings measured 4-years after completion and only includes the earnings of working individuals.

We find this pattern holds when weighting by title IV enrollment (Table 5.14). Under the final regulations, we estimate that 1.8 percent of students at programs in the public sector attend programs expected to fail the accountability framework. This is modestly higher than the estimated share who attended failing programs under current regulations (1.1 percent). The increase is driven by the large number of students who attended associate's, bachelor's, and master's degree programs at public institutions that would be subject to the

accountability framework but are currently exempt. This increase more than offsets the reduction in students attending undergraduate certificate programs at public institutions that are no longer expected to fail the accountability framework. A similar pattern exists for students who attend private non-profit programs (Panel B). While there is a reduction in the share of students who attended failing certificate programs at those institutions, there are increases in the shares of students who attend failing programs in all other credentials, resulting in an overall increase in the share of students who attend failing programs.

In the proprietary sector (Panel C), we estimate that fewer students attend programs that fail the accountability framework in the final regulations relative to the current regulations (18.0 percent vs. 29.9 percent), mainly because the test was made slightly less punitive under the final regulation (program earnings are measured after 4-years and include only working individuals).

[GRAPHIC] [TIFF OMITTED] TR01JY26.052

[GRAPHIC] [TIFF OMITTED] TR01JY26.053

The Department also estimated the share of title IV, HEA funds disbursed to students who attend programs expected to fail the accountability framework, disaggregated by credential level (Table 5.15).70 71 Overall, the Department estimates that failing programs under the final regulations would lose a smaller amount of title IV, HEA funds than failing programs under the current regulations (4.0 percent vs. 5.2 percent). Undergraduate certificate programs will experience the largest change. Under the current regulations, more than half (50.2 percent) of all title IV, HEA funds disbursed to undergraduate certificate programs are projected to be lost due to the accountability framework. Under the final regulations, only 29.4 percent is projected to be lost. This reduction is driven by the fact that programs lose eligibility for only Federal student loans under the final regulation (unless the failing program is offered at an institution that also fails the standards of administrative capabilities, and the program is also estimated to not opt out of the federal loan program), and because undergraduate certificate programs face a less-punitive accountability framework.

\70\ Under the current regulations, lost title IV HEA funds includes both Pell Grants and Federal student loans. Under the final regulations, lost title IV, HEA funds include only Federal student loans, unless the program is at an institution that is estimated to also fail the administrative capability standards--in which case, lost title IV HEA funds also includes Pell Grants for that program.

\71\ Our estimates on title IV, HEA student aid disbursements assume no program switching for students who switch from a failing program to a passing program and therefore continue to receive title IV, HEA funds. [GRAPHIC] [TIFF OMITTED] TR01JY26.054

When disaggregated by sector (Table 5.16), we estimate no net change in the overall amount of title IV, HEA funds disbursed to failing programs at public institutions. While undergraduate and graduate certificate programs in the public sector are expected to lose less Title IV, HEA aid under the final regulations (Panel A), most other credential levels at those institutions are expected to lose more title IV, HEA aid, resulting in no net change in the title IV, HEA disbursements.

The Department's estimates suggest that there could be a sizeable reduction in title IV, HEA funds disbursed to the non-profit sector (Panel B) because all programs in this sector are now subject to an accountability framework. Under the current regulations, just 0.8 percent of title IV, HEA funds disbursed to non-profit programs were to failing programs. Under the final regulations, that figure is estimated to increase to 3.9 percent, driven by failing associate's, bachelor's, master's, and professional degree programs. The proprietary sector (Panel C) demonstrates the opposite pattern. Under the final regulations, these programs will see an increase in

title IV, HEA funds relative to the baseline. This is because the accountability framework is (generally) easier for these programs to pass, and the final regulation allows failing programs to continue receiving Pell Grants (as long as the institution does not fail the standards of administrative capability requirements). [GRAPHIC] [TIFF OMITTED] TR01JY26.055

BILLING CODE 4000-011-C

Next, we estimated how the final regulations would impact specific fields of study.\72\ The Department estimates that some undergraduate programs, such as Education programs (CIP=13), Humanities/Liberal Arts programs (CIPs=05, 16, 23, 24, and 50), and Religious Studies programs (CIPs=38 and 39) will fail at a higher rate relative to the baseline (Table 5.17). Other programs, such as Health-Related undergraduate programs (CIPs=51, 60, 34, and 61), Business/Management undergraduate programs (CIP=52), and Vocational/Technical undergraduate programs (CIPs=15, 41, 46, 47, 48, and 49), are estimated to fail at lower rates. Culinary & Personal Services undergraduate programs (CIP=12) will fail the accountability framework at the highest rates among all broad field of study categories, though the share that they fail under the final regulation (77 percent) is slightly lower than the share under the current regulation (83 percent). This implies that certain types of institutions offering these types of programs will likely benefit because fewer of their programs will be at risk of failing the accountability framework under this final rule, relative to the current baseline.

\72\ The Department grouped programs into broad field of study categories using a slightly modified version of the field of study categories defined in the variable “MAJORS12” from NPSAS:2020. The following adjustments were made to the field of study categories for conformability: Multi/Interdisciplinary Studies was combined with “Other Technical Professional”; Math was combined with “Engineering”; new categories were created for “Culinary and Personal Services” (CIP2=12) and “Religious Studies” (CIP2=38 or 39). For more information, see https://nces.ed.gov/datalab/ codebooks/by-subject/157-national-postsecondary-student-aid-study- 2020-undergraduate-students.

A somewhat different pattern is observed when estimates are weighted by enrollment (Table 5.18). For example, approximately half as many students attend failing Religious Studies undergraduate programs under the final rule than under the current baseline (3.9 percent vs. 7.8 percent). One reason for this difference is because the final regulation does not apply the accountability framework to programs that received no Federal student loans during the five award years prior to the enactment of the Working Families Tax Cut Act--many of which are religious programs that would have otherwise failed the revised earnings premium test.

Similarly, undergraduate programs in Religious Studies are also estimated to lose fewer title IV, HEA funds under the final regulation (Table 5.19). These programs will lose about half as much title IV, HEA funds under the final rule relative to the amount they would lose under the current regulations (2.9 percent vs. 8.1 percent). Again, this difference is driven by the provision in the final rule that exempts certain programs from the accountability framework if they are at institutions that received no Federal student loans during the prior five award years.

Other types of undergraduate programs, including Business/ Management programs, Culinary & Personal Services programs, Health- related programs, Humanities/Liberal Arts programs, Other Technical/ Professional programs, Social/Behavioral Science Programs, and Vocational/Technical programs also experience increases in title IV, HEA funding under the final rule relative to the current baseline (Table 5.19). Certain types of institutions offering these programs will likely benefit under the final regulation (relative to the baseline) because fewer of their programs will be at risk of failing the accountability framework. This is because the final regulation includes a less-punitive earnings test and because programs that fail the accountability framework lose access only to Federal student loan eligibility (unless the failing program is at an institution that fails the standards of administrative capability requirements. BILLING CODE 4000-011-P

[GRAPHIC] [TIFF OMITTED] TR01JY26.056

[GRAPHIC] [TIFF OMITTED] TR01JY26.057

[GRAPHIC] [TIFF OMITTED] TR01JY26.058

To conclude the program-level analysis, the Department examined the programs estimated to fail the final regulations at the highest rates (Table 5.20).\73\ Some of these--such as Cosmetology (CIP=12.04), Somatic Bodywork (CIP=51.35), and Dental Support Services (CIP=51.06)-- have substantially lower fail rates under the final regulation relative to the baseline. For example, the Department estimates that 99 percent of undergraduate certificate programs in Cosmetology would fail the accountability framework under the current regulations, but only 93 percent are expected to fail under the final regulation. So, while many Cosmetology certificate programs are estimated to fail under the final regulation, it is less punitive for these programs than the estimated impact of the current regulation.

\73\ The Department included the twelve programs (defined by the unique combination of cip4 and credlev) that had the highest share of programs estimated to fail under the final regulations. Programs where there were fewer than 100 observations with non-missing earnings data nationally were excluded from the ranking.

On the other hand, Mental and Social Health & Allied Professions master's degree programs (CIP=51.15), Teacher Education and Professional Development associate's degree programs (CIP=13.12), and Drama/Theater Arts bachelor's degree programs (CIPs=50.05, 50.07, and 50.09) are anticipated to be most negatively impacted by the final regulations relative to the current baseline. These programs are often between 10 and 20 times more likely to fail the accountability framework under the final regulation relative to their fail rates under the current regulations. These higher fail rates are driven by the fact that a large share of these programs are offered at public and non- profit institutions. Unlike the policy in the current regulations, these programs would no longer be exempt from the accountability framework under the final rule.

[GRAPHIC] [TIFF OMITTED] TR01JY26.059

BILLING CODE 4000-011-C Impact of the Final Regulations on Students

The Department also estimated the share of students in failing programs from each sex and race category using data from IPEDS for completers from the 2017-18 and 2018-19 pooled award years (Table 5.21). For each program, we multiplied the number of title IV enrollees from the 2024-25 award year by the ratio of completers from the given sex or race category.\74\ As reported above, fewer students attend programs that are estimated to fail under the final regulation relative to the baseline. Consistent with this finding, we estimate a reduction in the overall share of students from both sex categories and

all race categories who attend failing programs. The estimated reduction is largest for male students, Hispanic students, and students with another race category not listed in Table 5.21.

\74\ For example, we multiplied the number of title IV, HEA enrollees in each program by the ratio of completers from the program (using IPEDS data) who were male to calculate the share of male students in passing and failing programs. [GRAPHIC] [TIFF OMITTED] TR01JY26.060

Next, the Department estimated the impact of the provision that delays when the accountability framework first takes effect for qualifying programs. As described in the “Earnings of Program Completers--Use of IRS Data” section, programs listed in Table 5.22 will receive at least a one-year delay in when they are first counted as passing or failing the accountability provision.\75\ The delay applies for all programs sharing one of the 6-digit CIP codes listed in Table 5.22, regardless of credential level or institutional sector in which the program is offered.\76\

\75\ As explained above in the “Earnings of Program Completers--Use of IRS Data” section, this list was determined in the following manner. First, we started with the list of occupations listed in the final regulation “Occupations That Customarily and Regularly Received Tips; Definition of Qualified Tips” from the Internal Revenue Service and Treasury (91 FR 19026). This final regulation became effective June 12, 2026. For more information, see: www.federalregister.gov/documents/2026/04/13/2026-07104/ occupations-that-customarily-and-regularly-received-tips-definition- of-qualified-tips. We then narrowed the list to predominantly tipped occupations, meaning that 50 percent or more of workers in a given occupation report tipped income to the IRS. Then, we used the Department's CIP-SOC crosswalk to link occupations (defined using 6- digit SOC codes) to programs (defined using 6-digit CIP codes). Twenty unique programs (defined using 6-digit CIP codes) qualify for the delay.

\76\ Some of these programs may qualify for additional years if they have fewer than 30 title IV, HEA completers in cohorts that graduated after the 2021-22 award year. This is because the Department will not aggregate completers for programs listed in Table 5.22 using graduates prior to the 2021-22 award year. Ultimately, this means that some share of programs within those listed in Table 5.22 will not meet the minimum cohort size (30 title IV, HEA completers) until a future point between 2030 and 2032, giving them additional years in which they will not be subject to the earnings test.

The Department examined the number of programs that will likely qualify for a delay under this provision using data from FSA (not PPD:2026 \77\) for the 2024-25 award year. This data contains information on the number of title IV enrollees, completers, and title IV, HEA student aid disbursements, disaggregated by institution (OPEID6), credential-level, and program (6-digit CIP code).\78\ The Department finds that approximately 3 percent of all programs will qualify for at least a one-year delay. These programs collectively enroll approximately 1.4 percent of all title IV, HEA students and receive approximately 1.4 percent of all title IV, HEA disbursements (Table 5.23).

\77\ The Department could not use PPD:2026 for this analysis because programs in PPD:2026 are identified by 4-digit CIP codes, but the delay provisions applies to programs at the 6-digit CIP code level.

\78\ Using this data, the Department dropped programs with fewer than 8 title IV completers during 2024-25 because these programs would likely have fewer than 30 title IV completers after the cohort aggregation process, exempting them from the earnings test. The N sizes differ from prior those in prior table because this is a different dataset (programs are measured at the 6-digit CIP code level) and the same restrictions cannot be applied to it as PPD:2026.

BILLING CODE 4000-011-P

[GRAPHIC] [TIFF OMITTED] TR01JY26.061

[GRAPHIC] [TIFF OMITTED] TR01JY26.062

Two fields of study benefit most from the delay: Personal and Culinary Services (CIP=12) and Health Professions and Related Clinical Sciences (CIP=51). Specifically, the Department estimates that 77 percent of all Personal and Culinary Service programs and 1 percent of all Health Professions and Related Clinical Sciences programs will qualify for the delay (Table 5.24, Panel A). These programs collectively enroll 226,000 title IV, HEA students annually, which is roughly 1% of all title IV, HEA students (Table 5.24, Panel B).

[GRAPHIC] [TIFF OMITTED] TR01JY26.063

Next, the Department estimated how the delay provision will impact the number of students enrolled in failing programs during the first years of sanctions. For this analysis, the Department used PPD:2026 and removed all programs sharing a 4-digit CIP code with one of the 6-digit CIP codes listed in Table 5.22. With these delay-qualifying programs excluded, the Department estimates that 3.0 percent of all title IV, HEA students are enrolled in programs expected to fail the accountability framework during the first award year in which sanctions go into effect (i.e., the award year starting with July 2028) (Table 5.25, Panel A). After the delay period has expired, the Department estimates that 4.3 percent of students will have enrolled in failing programs (Table 5.25, Panel B), which corresponds to our main estimates shown in Table 5.12.

Lastly, the Department examined how the delay provision will impact the share of title IV, HEA student aid disbursements to failing programs in the first year of sanctions. We again use PPD:2026 and remove programs sharing a 4-digit CIP code with a 6-digit CIP code in Table 5.22. With these programs excluded, the Department estimates that 2.9 percent of all title IV, HEA student aid will be cut off from failing programs in the first award year that sanctions go into effect (Table 5.26, Panel A). When the delay period has expired and programs listed in Table 5.22 are subject to the accountability framework, the Department estimates that 4.0 percent of all title IV, HEA student aid will be cut off from failing programs (Table 5.26, Panel B), corresponding to our main estimates in Table 5.15.

[GRAPHIC] [TIFF OMITTED] TR01JY26.064

[GRAPHIC] [TIFF OMITTED] TR01JY26.065

The Department has previously discussed the impact of the regulation on the Cosmetology sector above (Tables 5.17, 5.18, 5.19, and 5.20), demonstrating that the regulation will likely result in a substantial reduction in the share of these programs that are at risk of failing the accountability framework relative to the current regulations. Institutions offering cosmetology programs will likely benefit from the final regulation because they will have fewer programs at risk of failing the accountability framework. This means fewer of their programs would be at risk of losing eligibility for certain types of title IV, HEA funds. Given the many comments the Department received about the rule's impact on cosmetology programs, the Department includes additional analysis on these programs to further clarify the rule's impact on this sector.

Overall, the Department estimates that 93 percent of cosmetology certificate programs are estimated to fail the accountability framework in this rule (Table 5.20). This is a reduction from the current regulation, under which 99 percent of cosmetology certificate

programs are expected to fail. These percentages are calculated by dividing the number of failing cosmetology programs by the total cosmetology programs that are subject to the earnings test. In other words, to determine the share of cosmetology certificate programs impacted by this final regulation, we divided the 840 cosmetology certificate programs that are expected to fail the earnings test by the 900 total cosmetology certificate programs that are subject to the earnings test (yielding roughly 93 percent).

However, the Department clarifies that there are many more cosmetology programs that receive Federal student aid that will not be impacted by the earnings test because these programs are too small to form a cohort, or because these programs do not receive Federal student loans. As shown in Table 5.27, there are a total of 1,450 cosmetology certificate programs nationally. Approximately 450 of these programs are entirely exempted from the earnings test (columns 2 and 3), and an additional 100 programs or will receive two or more years of delay before the earnings test applies to them (column 4). Thus, when these programs are factored into the analysis, the Department finds that just 58 percent of all cosmetology programs are expected to fail the earnings test, which we refer to as the “effective fail rate” for these programs (Table 5.28). Furthermore, the Department finds that just 80.6 percent of students in cosmetology programs are enrolled in programs that are expected to fail the earnings test under this regulation, a significant reduction compared with the current regulations under which 99 percent of students in cosmetology certificate programs are enrolled in programs expected to fail. [GRAPHIC] [TIFF OMITTED] TR01JY26.066

[GRAPHIC] [TIFF OMITTED] TR01JY26.067

BILLING CODE 4000-01-C