The Department of Government Efficiency’s effect on executive agencies last spring was swift.  DOGE teams hopped from agency to agency, terminating thousands of grants worth billions of dollars.  From medical labs to art museums, grantees of all stripes sprinted to the courthouse in a bid to salvage their funding.  Judges are now facing an onslaught of grants-termination lawsuits.  Now that the Supreme Court has issued its much-anticipated emergency docket ruling in NIH v. APHA (in which the Court stayed a lower court’s vacatur of various grant terminations), we arrive at an opportune moment to assess the state of play.

Can grantees find relief? And in what tribunal (or tribunals) might they seek it? APHA provides a single sentence about the scope of jurisdiction and relief: a district court cannot “adjudicate claims based on the research-related grants or [ ] order relief designed to enforce any obligation to pay money pursuant to those grants” (quotations omitted). But what does it mean to bring a claim “based on” a grant? And what does it mean to “enforce any obligation to pay money”? This post will take up each of these questions in turn. But first, a word on nomenclature. Although there is a colorable argument that grants should be considered legally distinct from contracts, APHA treats grants as a type of contract, so I use these two terms as the high court does—synonymously.

***

Imagine that a government agency terminates a grant, and the grantee claims that the termination was unlawful. The grantee could allege that the termination violated the terms and conditions of the grant.  Alternatively, the grantee could allege that the termination violated sources of federal law outside the grant agreement (such as a regulatory, statutory, or constitutional provision).  The former claim would be “based on” on the grant; the latter would be based on federal law. The distinction between the two types of claims has significant ramifications for determining both the proper forum and the available remedy.

Claims based on the grant’s terms are breach-of-contract actions. As such, they are nonjusticiable in federal district court because the Tucker Act vests jurisdiction over contract claims in the Claims Court (officially, “the Court of Federal Claims”), an Article I tribunal with special expertise in contract law. Alternatively, claims based on violations of federal law can be heard in district court under the Administrative Procedure Act (APA).[1] The Claims Court can provide only money damages to compensate grantees for past-due sums; reinstatement of grants is not on the table in that forum.[2] But in district court, vacatur of an unlawful grant termination would have the effect of snapping the grant back into legal effect.

Plaintiffs seeking reinstatement thus have strong incentive to allege violations of federal law so that they can stay in district court and get their terminations vacated. Article III courts accordingly are on the lookout for plaintiffs who camouflage breach-of-contract claims as violation-of-federal-law claims.  It can sometimes be tricky for courts to tell the difference between a disguised contract action and one that is legitimately grounded in federal law. One appeals court judge recently framed the analysis as follows in a dissent that was quickly endorsed by the en banc D.C. Circuit Court of Appeals:  “What matters is what the court must examine to resolve the case: If a plaintiff’s claim depends on interpretations of statutes and regulations rather than the terms of an agreement negotiated by the parties, the claim is not in essence contractual.”

Courts sometimes note that a grantee would not be able to bring any claims for grant terminations but for their grants. From that premise, more than one court has concluded that the plaintiff’s rights stem from the grant itself and thus their claims sound in contract. But Crowley Gov’t Servs. v. GSA (D.C. Cir. 2022) expressly rejected any such “‘but-for’ test for identifying the source of the right.” The operative question is not whether there is a grant but rather whether the rights turn on the grant or on other sources of federal law.  A helpful analogy might be found in the U.S. Constitution’s right to due process for property owners.  You wouldn’t have a claim to a due process violation but for the deed to your house; nevertheless, your due process rights plainly come from the Constitution, not from the deed.  The key here is to avoid the erroneous conflation of the source of a plaintiff’s relationship with the government (i.e., their grant agreement) for the source of the plaintiff’s rights (i.e., federal law).

What if the termination violates a provision of the grant’s terms and independently violates a provision of federal law? There are two different tests that the D.C. Circuit has put forward to address claims that turn on both contracts and federal law.  First, in Spectrum the court employed a gravamen test: if the “gravamen” of a claim turns on the contract, then it’s a breach-of-contract case. But then in Transohio the court suggested that the analysis hinges on whether “plaintiffs’ claims are founded only on a contract, or whether they stem from a statute or the Constitution.” (emphasis added) In other words, if the court has to interpret federal law at all, the claim doesn’t sound in contract. So it’s an open question where exactly the courts will draw the line in any given case.  Arguably the Transohio approach is better aligned with the APA’s “strong presumption in favor of reviewability of agency action.”

What about when a grant agreement incorporates a regulation? In that scenario, an agency’s failure to comply with that regulation is simultaneously a violation of federal law and a breach of contract.  Would a plaintiff’s claim sound in contract? It depends.  Generally speaking—no, the government cannot evade district court jurisdiction simply by incorporating federal law into a contract and then arguing that any violations of federal law give rise exclusively to breach-of-contract claims. In the influential Megapulse case, the D.C. Circuit expressly rejected this line of reasoning, pointing out that agencies could insulate themselves from judicial review for violations of federal law simply by incorporating federal law into a contract.  And the court was right to do so.  Imagine that an agency incorporates the First Amendment into a grant and then terminates the grant on the grounds that the grantee authored an editorial espousing a political viewpoint disfavored by the current presidential administration. The grantee files suits on First Amendment grounds. In turn, the government argues that the claim violates the grant agreement and thus belongs in the Claims Court. It would surely be absurd for a district court to punt a fundamental question of constitutional interpretation to a Claims Court that has a narrow specialty in contract law. 

But that is a general principle, not a universal axiom.  Ingersoll-Rand marks an exception. In that case, the government allegedly violated regulations that were incorporated into a contract, but the district court nonetheless found it was deprived of jurisdiction.  Importantly, these were not just any regulations—they came from the Federal Acquisition Regulation and specifically concerned contracts.  In other words, the regulations at issue fell within the unique expertise of the Claims Court. No court has explicitly reconciled the apparent tension between Megapulse and Ingersoll-Rand, but the relationship between the cases is best understood as follows: Megapulse’s insistence that the government can’t incorporate its way out of judicial review for violations of federal law is the default rule; Ingersoll-Rand creates a narrow carve-out for incorporated regulations that directly implicate contract law. 

Admittedly, the notion that Ingersoll-Rand is an exception to Megapulse’s general rule serves to tip the scales in favor of district court jurisdiction.  Such an approach is consistent with the Supreme Court’s opinion in Maine Community Health Options where it held, “The Tucker Act yields . . . when the Administrative Procedure Act [ ] provides an avenue for relief.”  Despite a strong case for treating Ingersoll-Rand as a limited exception, in our current climate district courts may be hesitant to exercise jurisdiction over any regulation incorporated into a grant agreement. Plaintiffs accordingly may be inclined toward claims based on federal law lying exclusively beyond their grants’ terms.

The foregoing analysis aside, there is one statutory claim that appears to be unavailable to grantees in district court: the arbitrary-and-capricious provision in the APA. Why? Because most any breach-of-contract claim could be artfully plead as an arbitrary-and-capricious claim. Both in APHA and Department of Education v. California (the other major grants case on the emergency docket, from April), the Supreme Court stayed grant reinstatements based on a lower court ruling that the terminations were arbitrary and capricious.  In neither APHA nor California did the Court’s slender opinion say as much explicitly, but many district court judges will now be inclined to presumptively treat arbitrary-and-capricious claims as disguised contract claims and thus not properly before them.  Undoubtedly, some grantees have colorable arbitrary-and-capricious claims that turn on their termination notices and do not require recourse to the grant’s terms; those plaintiffs may try to press those claims in district court. But APHA and California create an atmosphere in which grantees will seek to vindicate rights based on other sources of federal law in order to assure district courts of their jurisdiction.

***

Once a district court finds that a claim is indeed based on federal law rather than the grant itself, the court must then determine what remedies can be properly provided to the plaintiff. APHA bars district courts from providing relief “designed to enforce any obligation to pay money”—but what does that mean?

If APHA makes anything clear, it is this: district courts cannot directly order agencies to disburse funds. When a grantee is owed past due sums, they must recover those funds in the form of money damages. And the Administrative Procedure Act forecloses the possibility of money damages as a remedy in district court. Meanwhile, the Tucker Act invests the Claims Court with the power to provide that very form of relief. There does not appear, then, to be great controversy over retrospective monies—a grantee can get made whole in the Claims Court.

The real fight is over grant reinstatement in district court. Such courts can set aside, or vacate, agency action that violates federal law. Imagine the following chain reaction: (1) a district court finds that a grant termination violates federal law, (2) that court, accordingly, vacates the termination, (3) the default effect of vacatur is that the grant snaps back into legal effect, (4) the restoration of the grant means the resumption of the agency’s obligations to disburse funds moving forward under the grant. Importantly, the legal obligation to pay up comes not from the court’s order but from the grant’s terms.

Some might argue that because vacatur ultimately leads (albeit indirectly) to the disbursement of funds, district courts should not be allowed to order vacatur in grants cases. By that logic, vacatur is underneath it all “relief designed to enforce [an] obligation to pay money[.]” And, in fact, the Supreme Court’s order in APHA stayed a district court’s vacatur of various grant terminations—potentially leading to the impression that vacatur is per se impermissible relief in a grants-termination case.

But APHA does not appear to require such a result.  The lone sentence in APHA concerning the district court’s jurisdiction cites to Department of Education v. California.  And California made plain that “a district court’s jurisdiction ‘is not barred by the possibility’ that an order setting aside an agency’s action may result in the dis­bursement of funds,” here drawing language from the landmark case Bowen v. Massachusetts (1988). Bowen found that a district court can remedy violations of federal law even if the “byproduct” of that court’s ruling might result in an agency’s disbursement of money. So while some may read into APHA a judicial fiat that vacatur is categorically unavailable as a remedy to grantees, that interpretation is ultimately unpersuasive given that APHA expressly reaffirms California, which itself expressly reaffirms Bowen.  And there is yet another reason to think vacatur is not really “relief designed to enforce [an] obligation to pay money”: vacatur might not actually lead to the disbursement of funds. After all, an agency could reterminate a reinstated grant, provided that the agency does so lawfully.

If vacatur is still on the table, then why did the Supreme Court stay the district court ruling in APHA, which vacated the grant terminations at issue? Because vacatur is only available to remedy violations of federal law. The high court treated APHA as a contractual action—in other words, as a case whose claim was “based on” the grant rather than on sources of federal law beyond the grant’s terms.  Notably, the district court in APHA had vacated the terminations on the grounds that they were arbitrary and capricious.  And, as discussed above, a claim grounded in the arbitrary-and-capricious provision of the APA is the one breed of statutory claim that will be presumptively treated as a disguised breach-of-contract claim. Because most any breach-of-contract claim can be dressed up as an arbitrary-and-capricious claim, grantees must bring other kinds of claims if they hope to stay in district court and ultimately find relief in the form of vacatur.

If vacatur were per se impermissible in grants cases, it would lead to absurd results. Imagine an agency overtly terminates a grant because of the grantee’s religious identity. Then the grantee brings a constitutional claim via the APA. The district court would be powerless to remedy a brazen violation of that grantee’s constitutional right to the free exercise of their faith, merely because vacatur would indirectly lead to the disbursement of funds. It would be rather remarkable for courts to derive from a single sentence in APHA a newfangled doctrine that fundamental liberties are contingent on agency whim and beyond judicial safeguard.

Andrew Porwancher is a Professor of Constitutional History at Arizona State University.

---

[1] Some claims that an agency violated federal law can also be brought outside the APA, but for the purposes of this post, I am concerned here with APA cases.

[2] The Claims Court can provide precisely three other forms of relief “as an incident of and collateral to any [monetary] judgment,” forms that are unlikely to be relevant in grants cases: (1) “orders directing restoration to office or position,” (2) “placement in appropriate duty or retirement status,” and (3) “correction of applicable records[.]” 28 U.S. Code § 1491(a)(2). The Federal Circuit has recently and unambiguously rejected the idea that the Claims Court can provide any additional relief beyond the foregoing “explicit and limited list of remedies.”