Look Up Your Institution: Where Three Federal Policies Overlap

Three federal policies now bear on the same programs at the same time: the new earnings test, the elimination of Grad PLUS, and the new borrowing caps. The tool below combines them. Search your institution to see which of your programs are flagged, and by what.

Open the full tool in a new tab →

How to Read Your Result

The result has six parts, read top to bottom.

1. Start with the verdict. The banner counts the policies that flag you and says why in one line each: programs failing the earnings test, programs and borrowers over the new Grad PLUS cap, Parent PLUS dollars above the $20,000 cap. Below that, how much annual revenue sits in the path of those flags, and how many programs the earnings test can see.

2. Then the total. Revenue Exposure is the aggregate: the annual dollars sitting in the path of all three policies, with each piece listed. Tuition tied to programs that fail the earnings test, Parent PLUS dollars above the new $20,000 cap, and Grad PLUS dollars in programs whose borrowers exceed the new limits, discounted for grandfathering. For most institutions the two borrowing pieces dominate and the earnings test is the smallest. Where a piece cannot be computed the card says n/a, and the total is a floor.

3. Then the borrowing picture. PLUS Loan Exposure shows what families and graduate students borrowed in AY 2024-25, how many of those borrowers sit in programs whose average borrowing exceeds the new caps, and the five programs with the most dollars above them. This is the part institutions tend to underweight. The caps arrive July 1, 2026 regardless of any earnings result, and grandfathering only delays them for students already enrolled.

4. Then the earnings test. The Earnings Test card gives failing out of testable and lists three programs: those failing by the most, then those passing by the least. If nothing is failing, that is the honest answer for most institutions. Roughly 55% have no failing program and none within $2,000 of the benchmark. Read the testable count anyway. A small number means most of your portfolio is invisible to the test, not that it is safe.

5. Then the flagged programs. Programs Needing Attention gathers every program carrying any flag: red for failing, amber for at risk, plum for over the borrowing cap. Weigh two things the cards show directly. Enrollment tells you the scale of what is exposed. Margin, or the per-borrower gap above the cap, tells you how far from the line it sits. A program failing by $800 is a different problem from one failing by $12,000, and a program passing by $800 may be more fragile than one already failing.

6. Look for programs carrying more than one flag. In the list and the table, a program that fails the earnings test, carries a large per-borrower gap under the new caps, and enrolls hundreds of students is a categorically different risk from a small program that narrowly misses on earnings alone. That overlap is the subject of the rest of this piece.

Why Compound Risk Matters

Institutions do not experience these policies in isolation. They experience them simultaneously, and the programs most vulnerable to one are often the same programs most exposed to another. We have written about each on its own: the earnings test, Grad PLUS elimination, and the final rule timeline.

Consider a graduate nursing program at a mid-size private university. Under the earnings test alone, it passes. Its graduates earn above the benchmark, though only by a narrow margin. Under the Grad PLUS elimination alone, it has a large funding gap: students have been borrowing $45k per year, and the new cap is $20.5k, or $50k, depending on which week’s court ruling you are reading. Neither policy, viewed independently, is existential.

Together, they create a feedback loop. If the funding gap causes enrollment to decline, the program may fall below the reporting threshold and lose the ability to demonstrate passing earnings. If a future cohort’s earnings dip, because fewer students completed, or because the students who could not secure private loans to fill the gap were the ones who dropped out, the program could fail the earnings test. A first failure triggers mandatory disclosure to prospective students, which could depress enrollment further.

This is the compound effect: individually manageable risks that, stacked together, become structurally destabilizing.

Three Patterns That Emerge

Failure clusters, rather than isolated outliers

At most institutions the earnings test is a manageable issue. Among institutions with any failing programs, the median number failing is one. But at the tail the concentration is severe. Two institutions have 15 failing programs each, and several more have eight or nine, often with a large share of their enrollment flowing through them. For those institutions the earnings test is not a program-level problem. It is an institutional viability question, and the institution-level trigger in the final rule, provisional certification if low-earning programs account for more than half of Title IV recipients or dollars, makes it an explicit one.

The certificate and PLUS overlap

Undergraduate certificate programs account for 58% of all earnings test failures. Many of these same programs carry significant Parent PLUS borrowing: parents taking on debt for programs whose graduates earn less than the regional high school median. The tool makes this overlap visible program by program, so you can see which failing programs also carry substantial PLUS volumes.

The at-risk band deserves more attention than the failing count

The 934 programs passing within $2,000 of the benchmark are, in some ways, more strategically important than the 2,880 that currently fail. Failing programs have clarity. They know where they stand and can plan accordingly. At-risk programs face uncertainty: a small shift in the next cohort’s earnings, a benchmark adjustment, or a change in the completer mix could push them over the line. For institutions with clusters of at-risk programs, the question is not “do we fail today?” but “how fragile is our passing status?”

What You Cannot See

The tool has real limitations.

76% of programs have no earnings data. The IRS suppresses earnings when a program has 15 or fewer graduates in the measurement cohort. Only 49,860 of 209,321 programs can be tested at all, and 835 institutions have no testable programs whatsoever. Small programs, which may be disproportionately vulnerable, are invisible here. The 2,880 failures are almost certainly a lower bound.

The data is backward-looking. The PPD:2026 earnings reflect students who completed programs in 2017-18 and 2018-19, with earnings measured in 2022 and 2023. These graduates entered the workforce before the pandemic, before generative AI reshaped entry-level labor markets, and before anyone knew this test was coming. The first operational test will use a different cohort, primarily completers from the 2021 award year with earnings measured in calendar year 2025, at a finer CIP6 granularity. That first round is informational; the rule takes effect July 1, 2027 and the first award year carrying sanctions begins July 2028.

Revenue exposure is a proxy with uneven coverage. The earnings-test piece needs IPEDS finance data and is computable for only 938 of the 5,096 institutions in the file. The borrowing pieces come from FSA actuals and reach further: 1,507 institutions have Grad PLUS volume and 580 have Parent PLUS borrowing above the cap. In all, 2,127 institutions get a figure. It is an order of magnitude, not a financial model.

The two PLUS sources do not perfectly agree. The tool shows FSA institution-level disbursement totals alongside PPD program-level per-student averages. These come from different cohort windows and different suppression rules, so dividing the FSA total by FSA recipients will not reproduce the PPD per-student figure. At the median the PPD-derived figure runs about 16% below what FSA actuals imply, and a third of institutions diverge by more than 25%. The likely cause is that PPD excludes privacy-suppressed small programs that FSA totals still include. FSA is the right source for institutional volume and PPD for program-level comparison; they should not be read against each other.

PLUS exposure reflects the outgoing regime. The Grad PLUS and Parent PLUS volumes shown are from AY 2024-25, the last year under uncapped borrowing. They represent what students were borrowing, not what they will be able to borrow under the new caps. The gap between the two is the problem.

Methodology

The tool combines three federal files.

• Program Performance Data (PPD:2026). 209,321 programs at 5,096 institutions, released for the AHEAD negotiated rulemaking. Earnings are IRS medians four years after exit for the pooled 2017-18 and 2018-19 completer cohorts, in 2024 dollars. Pass or fail comes from the master OBBBA flag (mstr_obbb_fail_cip2_wageb) in the corrected January 2, 2026 release. Program-level enrollment and loan counts come from the same release.
• FSA Direct Loan Dashboard. AY 2024-25, Q4. Institution-level Grad PLUS and Parent PLUS disbursements and recipient counts. Actual figures, not subject to privacy suppression.
• IPEDS Finance. FY 2022-23 net tuition revenue per FTE.

Revenue exposure

• Earnings test. The student count comes from federal aid records, so it includes everyone receiving Pell or a Direct Loan and excludes students who pay with no federal aid. Computable for the 938 institutions with both failing programs and IPEDS finance data.
• Parent PLUS over cap. $20,000 is the new annual cap per dependent student under OBBBA for loans made on or after July 1, 2026. The figure works from the institution’s average, so it nets families above the cap against families below it. It is not discounted for grandfathering.
• Grad PLUS at risk. The share is the fraction of the institution’s program-level Grad PLUS dollars sitting in programs whose average borrower exceeds the cap: 100 percent at 582 institutions, partial at 403, zero at 65. Another 457 institutions have FSA volume but no program-level data and are treated as fully exposed, about $60 million of the $14 billion total. The 50 percent stands in for grandfathering, since students already borrowing before July 1, 2026 keep Grad PLUS for up to three more years.

The total measures dollars in the path of the policies, not a forecast of lost revenue. Where a component cannot be computed the card shows n/a and the total is a floor.

A worked example

Shenandoah University, a private nonprofit in Winchester, Virginia, with 4,040 students and 93 programs, shows how the three cards are built. Two of its 17 testable programs fail the earnings test, five graduate programs have borrowers above the new Grad PLUS cap, and four undergraduate programs have average Parent PLUS loans above the new $20,000 cap.

Revenue Exposure. The card’s $9.1M is the sum in the figure above.

PLUS Loan Exposure. FSA reports 379 Grad PLUS borrowers and 364 Parent PLUS recipients. The PPD program files place 281 of the graduate borrowers in five programs whose average borrower exceeds the cap, led by the rehabilitation doctorate with 106 and the allied health master’s with 97. The pharmacy program is not among them: its 88 borrowers average $20,463 in Grad PLUS, but pharmacy is one of the professional fields with a $50,000 cap. Four undergraduate programs, nursing, music, drama, and dance, have average Parent PLUS loans above $20,000, 108 borrowers in all. The program ranking multiplies each program’s per-borrower gap by its borrowers: the rehabilitation doctorate’s 106 borrowers times a $25,687 gap puts $2.7M above the cap, the top line on the card.

Earnings Test. Both failing programs are in music. The bachelor’s earns $30,758 against a $35,969 benchmark, a shortfall of $5,211, and the master’s earns $42,787 against $47,421, a shortfall of $4,634. Drama (bachelor’s) is the closest passing program at $4,488 above its benchmark, outside the $2,000 at-risk band, which is why the card shows zero programs there.

PLUS loan exposure

• Institution totals and borrower counts come from the FSA file.
• Borrowers over the new caps and the top five programs use PPD program-level borrower counts. A program is counted as over the cap when its average borrower is, which misses over-cap borrowers in programs whose average sits below the line. The PPD omits privacy-suppressed small programs, so both figures are floors.
• The cap is $50k for the eleven professional fields in the RISE final rule of May 1, 2026 and $20.5k for every other graduate program. The professional list is the rule’s original eleven fields, not the 29-field expansion from the June 24 court stay.
• Unsubsidized borrowing is capped at the $20.5k statutory maximum before the gap is computed. About 1,365 programs report more than that, an artifact of mismatched cohort windows. Where a borrower has maxed unsubsidized loans, the gap equals the Grad PLUS amount; that is about 13 percent of programs.

Earnings status

Tipped occupations get a delayed timeline under Section 668.402(c)(3) of the final rule, which lists 20 six-digit CIP codes: cosmetology and personal grooming, bartending, casino services, and massage and bodywork. The PPD identifies programs only at the four-digit level, so the tool flags the families 12.04, 12.05, and 51.35. That over-includes culinary programs that are not tipped work, and it leaves out casino services, whose codes sit inside the much broader hospitality family 52.09.

Data sources: (1) U.S. Department of Education, Office of the Chief Economist and Office of Federal Student Aid. (2026). Program Performance Data (PPD:2026) for AHEAD Negotiated Rulemaking. (2) U.S. Department of Education, Federal Student Aid. (2026). Direct Loan Program Data: Award Year 2024-25, Quarter 4. (3) National Center for Education Statistics. (2024). IPEDS Finance Survey, FY 2022-23.

Read More

For the underlying framework, How the New Federal Earnings Test for Colleges Works explains what a failure actually triggers, and Every Program, One Test covers the final rule and its compliance timeline. To pull your own institution’s numbers directly from the federal files, see How to Download and Use the Program Performance Data.

Questions? Get in touch! We’re happy to help.

← All writing