Is Changing Your Student Loan Repayment Terms a Breach of Contract?

No. When Congress changes federal student loan repayment terms, it isn’t breaching your promissory note — because the note you signed says amendments to the law can change it. But there are situations where you have a real contract claim against a servicer or a guaranty agency, and those are the ones worth knowing about.

Your Master Promissory Note was never a negotiated contract. It’s a standardized federal form drafted by the U.S. Department of Education, not a document anyone bargained over — there is no version of it you could have redlined. In the older bank-based loan program, the Sixth Circuit took judicial notice in Mirandette v. Nelnet that the note’s language came directly out of the Department’s own regulations.

That form tells you what happens when the law changes, in a paragraph labeled NOTE on the Direct Subsidized/Unsubsidized version:

Amendments to the Act may change the terms of this MPN. Any amendment to the Act that changes the terms of this MPN will be applied to your loans in accordance with the effective date of the amendment. Depending on the effective date of the amendment, amendments to the Act may modify or remove a benefit that existed at the time that you signed this MPN.

The same note also says the terms “are determined in accordance with the Higher Education Act of 1965, as amended (the HEA), our regulations, and other federal laws and regulations.” As amended is doing a lot of work. Congress amending the Act doesn’t break the contract from outside. It changes the contract from the inside, which is exactly what the contract said would happen.

Then there’s what the note actually promised about repayment. The language is generic: “You have a choice of several repayment plans, including plans that base your required monthly payment amount on your income.” No plan name. No percentage. No forgiveness term. Nothing about 10% of discretionary income, and nothing about twenty years. That promise is still being kept as long as some income-based option exists — and IBR and the Repayment Assistance Plan both exist.

A stronger version of the theory is circulating alongside this one: that because the terms changed, the loans themselves are now void, unenforceable, or already forgiven. No court has accepted it. The one borrower who came closest — Daniel Skidmore, whose 2026 case against the Department of Education argued in almost these words that his loan agreements were void and unconscionable — never got a ruling on the question. His case was thrown out for being in the wrong courthouse, which is its own lesson and one we’ll come back to.

The note also anticipates the argument directly. Three sentences before the amendment paragraph, it says: “If any term of this MPN is determined to be unenforceable, the remaining terms remain in force.” Even if you knocked out a single term, your own paperwork says the rest survives. The amendment clause isn’t a defect that unravels the agreement — it’s a term of the agreement, and courts read it as one.

This is the same shape as the other debt-cancellation theories that circulate on social platforms, which generally identify a real grievance and then attach a remedy the law doesn’t provide.

Related: Can a data breach cancel your student loans?

You don’t have to take anyone’s word for any of this. Download your own promissory note from studentaid.gov and read the amendment paragraph.

A transfer of your loan to a different servicer or a different federal agency is not a material alteration that voids it. This version of the theory resurfaces every time the program moves. It came up in 2025 around a proposal — which did not proceed — to shift federal student loan administration to the Small Business Administration, and again in March 2026, when the Department of Education signed an interagency agreement with the U.S. Treasury Department beginning a phased handoff of the federal student loan portfolio, starting with defaulted loans. The argument each time is the same: the note names the Department of Education, so moving the program somewhere else supposedly breaks the agreement.

Two things answer it. The note broadly authorizes transfer and assignment — it contemplates the loan changing hands, and says so. And a federal student loan is a debt owed to the United States rather than to one particular agency, so moving administration between federal agencies doesn’t change who you owe. Even if a transfer were found to exceed the Department’s statutory authority, the natural remedy is that the work goes back where it belongs — not that the debt disappears.

The full version of this argument, including the chain-of-title claims that come with it, is covered in how the chain of title works with student loans.

Relying on the old terms for a decade doesn’t create a legal claim against the government, even when the loss behind it is real. And it is real. Someone who spent a decade making payments on a set of assumptions about what the endpoint would look like, and who would have paid down principal aggressively if they’d known otherwise, lost something concrete.

It’s worth being precise about what was lost, though, because it’s usually narrower than it feels. For most borrowers this is a loss of options, not of terms. Income-based repayment at 15% over 25 years has been the rule since 2007. The more generous 10%-over-20-years version was closed to borrowers who took loans before July 1, 2014 — and that happened back in 2014, not recently. What the 2026 changes took away were the workarounds that had grown up around those plans: the SAVE plan, and PAYE and ICR after June 30, 2028. The plan most people were on is still there.

That distinction matters legally as well as emotionally, because a claim that “you changed my deal” runs aground on the fact that, for most borrowers, the deal is the one it always was.

Even where something genuinely did change, reliance arguments against the federal government run into a decades-old wall. You generally cannot estop the government into paying money the statute doesn’t authorize, even when a government employee gave you bad advice and you relied on it — the Supreme Court held exactly that in Office of Personnel Management v. Richmond, a 1990 case in which a federal employee lost benefits after following incorrect guidance from the agency itself. And there’s no vested property right in a statutory benefit program: Flemming v. Nestor, decided in 1960, established that Congress can change the terms of a benefits program without compensating the people who were counting on the old ones.

The Department of Education already ran this exact analysis on the record. When it wrote the 2026 rules implementing the new repayment structure, commenters argued that the regulations couldn’t be materially altered after borrowers signed, and asked for borrowers already on track for forgiveness to be grandfathered in. The Department’s response, published in the Federal Register at 91 FR 23768, walked through what a reliance claim requires — a promise, reliance on it, a change in position for the worse, and reasonableness — and then concluded: “The MPN disclaims the idea that the terms and conditions of a Federal student loan are unalterable … meaning that any reliance interest is not reasonable.”

That’s the agency answering the question before anyone asked a court. It’s not a favorable answer, but it is the one a court would start from.

There’s a second layer worth knowing. Several of the changes people are angriest about aren’t discretionary choices the Department made — they’re statutory limits Congress wrote. When commenters asked the Department to let certain borrowers into a plan they’d been shut out of, the response was that the condition is statutory and the Department has no authority to amend it. Where an agency has no discretion, there’s nothing to challenge as an abuse of it.

The Contracts Clause restrains states, not Congress. It’s the single most common error in the forum threads, and the citation people reach for first.

Article I, Section 10 of the Constitution says no State shall pass any law impairing the obligation of contracts. It limits state legislatures, and Congress isn’t covered by it. Federal legislation that changes the terms of a federal program doesn’t implicate the Contracts Clause at all, which means the constitutional hook most borrowers think they have isn’t there.

A related doctrine does apply to the federal government, and it works against the argument rather than for it. Under what courts call the unmistakability doctrine — the Supreme Court’s 1996 decision in United States v. Winstar Corp. is the leading case — a claim that the United States gave up its power to change the law succeeds only if that surrender is unmistakable, spelled out in the agreement in terms that can’t be read any other way. A promissory note that expressly warns amendments may remove existing benefits is the opposite of unmistakable language.

The Higher Education Act’s sue-and-be-sued provision waives immunity for claims arising under the Act — which does not include a claim that your promissory note was breached. The statute says the Secretary may sue and be sued in any district court, and that sentence gets quoted constantly in forum threads as proof that a borrower can bring a contract case. A federal court in Illinois worked through it in 2024, in Witz v. Great Lakes Educational Loan Services, and reached the opposite conclusion for two reasons.

The waiver covers claims arising under the Act. A breach-of-promissory-note claim is a contract claim. It doesn’t arise under the Act, so it falls outside what Congress consented to.

The same sentence carves out the relief borrowers actually want. It expressly excludes “attachment, injunction, garnishment, or other similar process” against the Secretary. A borrower asking a court to adjust a balance, credit payments, or stop collection is asking for exactly that.

That decision also closed the workaround. Borrowers sometimes try to sue the servicer instead, on the theory that the servicer stepped into the Department’s shoes as a partial assignee of the note. But a partial assignee can generally be sued only if the assignor could be — so the servicer inherits the government’s immunity along with the assignment. Those claims were dismissed with prejudice.

Three courts across three circuits have now landed in the same place: Scoles v. Spellings in Oklahoma in 2008, Witz in Illinois in 2024, and Skidmore v. U.S. Department of Education in Michigan in March 2026. One district court in Pennsylvania went the other way in Ballard v. Navient in 2021 and let a partial-assignment theory proceed, so this isn’t unanimous. But the weight and the recency both run one direction.

Contract claims against the United States for more than $10,000 belong in the Court of Federal Claims, not in your local district court. It’s why a well-pleaded case can lose without a judge ever reaching the merits.

A borrower named Daniel Skidmore found this out in March 2026, in Skidmore v. U.S. Department of Education. He had paid on his loans for 27 years and sued the Department of Education directly, representing himself. His theories were almost verbatim the ones circulating online: that his loan agreements were “unenforceable, void, indefinite, and utterly unconscionable” because his balance and interest had been “consistently inflated” by bad record-keeping, and that the Department breached the implied duty of good faith and fair dealing by letting his servicer operate below industry standards.

He lost on jurisdiction, not on the merits. The Michigan federal court explained that sovereign immunity removes a court’s power to hear a case unless Congress has consented in unmistakable statutory text — and that for contract claims, Congress routed that consent through the Tucker Act:

• Over $10,000, the Court of Federal Claims has exclusive jurisdiction. That’s a specialized court in Washington, D.C., not the federal courthouse in your city.

• At $10,000 or less, district courts share jurisdiction. Most student loan claims clear that threshold without effort, which puts them out of district court.

• His contract counts were dismissed without prejudice so he could refile in the right court. The case survived only on Privacy Act and public-records claims.

That he was representing himself is part of the point rather than a footnote. A lawyer who handles these would have filed in the Court of Federal Claims from the start. The jurisdictional trap is precisely what catches people proceeding on advice they found online, and it costs them a year before anyone looks at what they’re actually complaining about.

Filing in the right court is no guarantee either. In 2016, a borrower did exactly that. In Nesselrode v. United States, he sued the government in the Court of Federal Claims for breaching his Master Promissory Note, arguing it was obligated to consolidate his loans. The court went to the text and found nothing there: there is nothing in the note, it held, that imposes a duty on the United States to consolidate a borrower’s loans. The statute he pointed to simply listed information schools have to make available, and other provisions made consolidation optional. No duty in the note, no breach.

That’s the pattern underneath all of this. The note is the document that decides these cases, and it does not promise what borrowers believe it promises.

Years enrolled is not the test. Qualifying payments are. It’s the most common factual misunderstanding among people who arrive at this question angry, and it’s separate from the legal argument entirely.

The same 2026 Michigan case makes the point cleanly. That borrower had been paying for 27 years, which sounds like far more than enough. He had made 276 of the 300 qualifying payments his forgiveness track required. What matters, the court said, is the number of qualifying payments made — not the number of years a borrower has been enrolled in an income-driven plan.

Which number applies to you depends on your plan. Some borrowers are on a 240-payment track, others on 300, and the Repayment Assistance Plan runs 360. Check which one you’re actually on before counting.

Two related points from the same decision:

• Academic deferment doesn’t fill the gap. He argued his years of graduate-school deferment should count as economic hardship. The statute lists economic hardship deferment specifically, and academic deferment isn’t on that list.

• A $0 payment does count. If your income calculation sets your payment at zero under an income-driven plan, that month counts toward your plan’s forgiveness clock.

If your count looks wrong, that’s a fixable administrative problem rather than a lawsuit, and there’s a process for it — see what to do when PSLF payments aren’t counting and which payments count toward IBR forgiveness.

The clause that kills the terms-change argument is the same clause that gives you a real claim against a servicer. Because your note incorporates the federal regulations as its terms, a company that violates one of those regulations may be breaching your contract — and that’s a state-law contract claim, which survives even though the Higher Education Act itself gives borrowers no private right of action.

Four theories come up. Two work reasonably well. Two are genuinely contested — courts disagree, and no appeals court has settled either one.

Direct breach of the promissory note. This is the theory with the clearest appellate support. In Bible v. United Student Aid Funds, decided in 2015, the Seventh Circuit held that the Master Promissory Note is a valid contract, that it incorporates the Act and its regulations, and that a guaranty agency which charged collection costs the regulations forbid had breached that contract. The Supreme Court declined to review it the following year.

On why federal law doesn’t wipe out the claim, the court held that the contract “simply incorporates applicable federal regulations as the standard for compliance,” so “the duty imposed by the state law is precisely congruent with the federal requirements,” and a state law claim that doesn’t seek to vary federal requirements doesn’t conflict with federal law. It added that the absence of a private right of action in a federal statute is no reason to dismiss a state law claim that incorporates part of that statute — otherwise you’d have to assume that where Congress provides no remedy, states may not provide one either.

A Nebraska federal court put the mechanism even more plainly in 2021, in Johansson v. Nelnet, rejecting a servicer’s argument that a breach claim can’t rest on a general governing-law provision: the Department’s regulations are specifically designated as part of the note’s terms and conditions, so breaching a regulation is breaching the note.

State-law misrepresentation. This is the theory that has survived even in the decisions most hostile to borrowers on everything else. The Seventh Circuit drew the practical dividing line in Nelson v. Great Lakes Educational Loan Services in 2019, and the Eleventh Circuit agreed the following year in Lawson-Ross v. Great Lakes Higher Education Corp. It comes down to two sentences that sound similar and aren’t:

• “They didn’t tell me” is preempted. Federal law exempts these loans from state disclosure requirements, so a claim that a servicer should have volunteered information generally fails.

• “They told me something false” is not preempted. When a servicer voluntarily chooses to make an affirmative statement and that statement is false, a state-law claim based on it can proceed.

The reason this line is durable is which courts have drawn it. The two decisions most hostile to borrowers on every contract theory — Hyland v. Navient in New York in 2019 and Witz in Illinois in 2024 — both let the misrepresentation claims through. It’s also the theory with money attached: in Hyland, every contract theory was dismissed and only the New York consumer-protection claim survived — and that case went on to class certification and settled.

The catch is that the state statute decides it, and states differ sharply. In Hyland, claims under California, Maryland, and Florida consumer statutes mostly failed the heightened pleading standard for fraud while the New York claim survived. Nebraska has knocked these claims out on two separate grounds: its consumer protection act exempts conduct in heavily regulated industries, and its deceptive trade practices act reaches only the purchase of goods or services — a limit a Nebraska federal court applied again in Montgomery v. U.S. Bank in September 2025, noting that no case had extended it to loan servicing. Where you live changes the answer.

Third-party beneficiary of the servicing contract. This one is genuinely split. The theory is that the Department’s contract with your servicer was meant to benefit you, so you can enforce it. Three district courts — Johansson and Olsen v. Nelnet in Nebraska, and Ballard v. Navient in Pennsylvania — have let versions of it proceed past a motion to dismiss. Two, Hyland and Witz, have rejected it. No appeals court has decided the question.

The side against borrowers has the stronger starting position, because federal common law treats individual members of the public as incidental beneficiaries of government contracts unless a different intention is manifested. The Hyland court added a point that’s hard to answer: the servicing contracts weren’t even available to borrowers, and it would be surprising for parties who intended a contract to be enforceable by third parties to withhold its terms from those same third parties. Witz dismissed the theory with prejudice in 2024, reasoning that letting borrowers enforce the servicing contract would render the absence of a private right of action meaningless, and Skidmore followed it in 2026. One of the borrower-side decisions, Johansson, has never been cited by any other court. It’s an open question, not a settled route.

Suing the servicer as assignee of the note. This one is mostly blocked, for the sovereign-immunity reason described above. The servicer inherits the government’s immunity along with the partial assignment, and the sue-and-be-sued clause doesn’t reach contract claims. One district court has gone the other way, but it’s the minority position.

Two caveats. The borrower-favorable decisions above are almost all rulings that a claim survived a motion to dismiss — a court found the allegations sufficient to proceed, not that the borrower won. And Mirandette is unpublished, which limits its use as authority; it illustrates the theory rather than establishing a rule.

There’s one more thing worth keeping in view. The note and its accompanying rights statement say that under applicable state law, unless federal law preempts it, you may have borrower rights, remedies, and defenses in addition to the ones stated in the note. The document contemplates state-law claims on its face.

The empirical picture, as of July 2026. A search of the case law for breach-of-contract claims tied to the promissory note and income-driven repayment turns up 23 decisions. Six of them aren’t student loan cases at all — “master promissory note” is ordinary commercial lending language, and the search picks up farm credit lines, commercial real estate notes, and insurance premium financing. Of the seventeen that are student loan cases, thirteen are about something a servicer or guaranty agency did: a forbearance applied in a way the note didn’t authorize, payments credited late, collection costs that violated the regulations. The remaining four were brought against the government — over a refusal to consolidate, over what a borrower was told about the risks of borrowing, and two dismissed for pleading failures.

Not one of the twenty-three challenges Congress changing repayment terms. The claim the internet is recommending has apparently never been brought. The adjacent claim — that a company broke a rule your note incorporated — gets brought constantly and regularly survives dismissal.

If you’re still years away from forgiveness, courts have said you don’t have a case yet — you have a fear. This came up in the largest borrower-side servicing case in the country, In re FedLoan Student Loan Servicing Litigation — a multidistrict litigation in Pennsylvania brought by 33 borrowers starting in 2018 against both the Department of Education and its servicer. It covered teacher grant conversions, PSLF payment counting, and income-driven repayment administration, and it brought contract claims among many others.

It ended in February 2025 without a ruling on any of it, dismissed for lack of subject-matter jurisdiction. Ten of the fifteen PSLF plaintiffs were between one and seven years away from 120 qualifying payments when the complaint was filed. The court held that a feared future denial isn’t an injury — it’s speculation — so those plaintiffs had no standing. Others had already been made whole through the Department’s own reconsideration process, which left them nothing to sue about. The appeal was voluntarily dismissed with prejudice in July 2025.

Seven years of litigation, 33 borrowers, real counsel, and no merits ruling. That’s the practical answer to “should I sue now”: if the harm you’re describing hasn’t happened yet, the case ends before anyone evaluates it.

As of July 2026, the active litigation isn’t about changed terms — it’s about credit borrowers already earned and procedure the government skipped. That’s the difference between the arguments that get dismissed and the arguments that get traction.

• Forcing forgiveness processing — and this one produced a result. American Federation of Teachers v. U.S. Department of Education, in the District of Columbia, pressed the Department to actually process income-driven repayment forgiveness for borrowers who had reached their payment threshold. It settled, and the settlement fixed the effective discharge date at the date the borrower hit the qualifying-payment milestone rather than the date the Department got around to the file.

• Procedure, not terms — pending. An amended complaint filed in June 2026 by four borrowers alleges the Department effectively repealed the REPAYE plan without going through required rulemaking, after a court vacated the SAVE plan. A preliminary injunction has been requested and the Department has moved to dismiss. Nothing has been decided.

• State challenges to the new rule — pending. Twenty-five states and the District of Columbia are challenging provisions of the 2026 repayment rule.

None of these is a breach-of-contract case. They’re administrative law challenges, and they’re built on “you didn’t give me what I already earned” or “you didn’t follow the process” rather than “you changed my deal.” That framing is the one courts have been willing to hear.

Related: How an active student loan lawsuit actually plays out

The mechanism that has actually delivered relief to borrowers at scale is administrative, not judicial — an observation from the litigation record, not an opinion. What ended most of the claims in that Pennsylvania multidistrict case wasn’t a defense win. It was legislation fixing teacher grant conversions in 2021, the limited PSLF waiver, and the one-time income-driven repayment account adjustment, all of which delivered the forgiveness, refunds, and corrected payment counts the plaintiffs had sued for. The court found the relief complete. Administrative fixes got there before the courts did.

What that leaves within your control is documentation:

• Pull your payment count and keep a copy. Download it rather than screenshotting a dashboard, and save a dated copy each time you check. Counts change when loans transfer, and the record you kept is often the only evidence that something moved.

• Keep every servicer communication in writing. Ask for written confirmation of anything you’re told on a phone call. A call log entry saying you spoke to someone is not the same as a document saying what they told you.

• Write down affirmative statements that turned out to be false. Note the date, the channel, and the exact wording as best you can reconstruct it. Under the preemption line described above, “they told me something specific and it was wrong” is the difference between a claim and no claim, and it’s the piece people almost never document at the time.

• Use the correction processes before the courts. Payment count disputes, reconsideration requests, and complaints to the Department’s ombudsman have resolved at scale what litigation has not.

• Read your own promissory note. Not a summary of it — the actual document, including the amendment paragraph. It settles most of the argument in about ninety seconds.

Every case and source referenced above, with full citations. Where a decision is unpublished or was later reviewed, that’s noted — it affects how much weight the decision carries.

Whether the promissory note can change

• United States v. Winstar Corp., 518 U.S. 839 (1996) — the unmistakability doctrine.

• Office of Personnel Management v. Richmond, 496 U.S. 414 (1990) — no estoppel against the government for money the statute doesn’t authorize.

• Flemming v. Nestor, 363 U.S. 603 (1960) — no vested property right in a statutory benefit program.

• Mirandette v. Nelnet, Inc., 720 F. App’x 288 (6th Cir. 2018) — unpublished; Sixth Circuit Rule 28 limits its citation. Reversed dismissal of a breach-of-note claim; affirmed dismissal of the Nebraska consumer-protection claims.

Jurisdiction and sovereign immunity

• Skidmore v. U.S. Department of Education, 2026 U.S. Dist. LEXIS 45606 (E.D. Mich. Mar. 5, 2026), appeal dismissed, 2026 U.S. App. LEXIS 17737 (6th Cir. June 17, 2026) — plaintiff appeared without counsel.

• Nesselrode v. United States, 127 Fed. Cl. 421 (Ct. Fed. Cl. 2016) — plaintiff appeared without counsel.

• Witz v. Great Lakes Educational Loan Services, Inc., 752 F. Supp. 3d 980 (N.D. Ill. 2024).

• Scoles v. Spellings, 2008 U.S. Dist. LEXIS 70647 (W.D. Okla. Sept. 18, 2008).

• Ballard v. Navient Corp., 2021 U.S. Dist. LEXIS 64255 (M.D. Pa. Mar. 31, 2021) — contrary on partial assignment.

When the note is actually breached

• Bible v. United Student Aid Funds, Inc., 799 F.3d 633 (7th Cir. 2015), cert. denied, 578 U.S. 989 (2016).

• Johansson v. Nelnet, Inc., 2021 U.S. Dist. LEXIS 58657 (D. Neb. Mar. 26, 2021).

• Nelson v. Great Lakes Educational Loan Services, Inc., 928 F.3d 639 (7th Cir. 2019).

• Lawson-Ross v. Great Lakes Higher Education Corp., 955 F.3d 908 (11th Cir. 2020).

• Hyland v. Navient Corp., 2019 U.S. Dist. LEXIS 113038 (S.D.N.Y. July 8, 2019); the case later reached class certification and settled, 2020 U.S. Dist. LEXIS 211676 (S.D.N.Y. Oct. 9, 2020).

• Olsen v. Nelnet, Inc., 392 F. Supp. 3d 1006 (D. Neb. 2019).

• Montgomery v. U.S. Bank N.A., 2025 U.S. Dist. LEXIS 191163 (D. Neb. Sept. 26, 2025).

• Travis v. Navient Corp., 284 F. Supp. 3d 335 (E.D.N.Y. 2018).

Standing, and the litigation that didn’t reach the merits

• In re FedLoan Student Loan Servicing Litigation, MDL No. 2833, 2025 U.S. Dist. LEXIS 27941 (E.D. Pa. Feb. 18, 2025), appeal dismissed, 2025 U.S. App. LEXIS 24669 (3d Cir. July 30, 2025).

• American Federation of Teachers v. U.S. Department of Education, No. 1:25-cv-802-RBW (D.D.C.).

Statutes and regulations

• 20 U.S.C. § 1082(a)(2) — the sue-and-be-sued provision, and its carve-out for injunctions and similar process.

• 20 U.S.C. § 1098g — exemption of Title IV loans from state disclosure requirements.

• 20 U.S.C. § 1098e(b)(7)(B) — qualifying payments toward income-driven forgiveness, including the economic-hardship deferment categories.

• 28 U.S.C. § 1491 (Tucker Act) and 28 U.S.C. § 1346(a)(2) (Little Tucker Act) — the $10,000 jurisdictional line.

• 34 C.F.R. § 685.209(k)(4)(i) — a $0 monthly payment earns a month of forgiveness credit.

• 34 C.F.R. § 682.209(g) — the bank-based program’s promissory note language, codified.

Primary documents

• Direct Subsidized/Unsubsidized Master Promissory Note and Borrower’s Rights and Responsibilities Statement — U.S. Department of Education. The amendment paragraph and the severability sentence quoted above are both in Section E. You can also download the copy you personally signed from studentaid.gov.

• Final rule implementing the 2026 repayment changes, 91 Fed. Reg. 23768 — the Department’s own reliance and promissory-estoppel analysis appears in the preamble responses to comments.