The Missing Constitutional Law of Executive Conditions

Abstract

This Article contributes to the development of the constitutional law of executive conditions. The appropriations presidentialism of the second Trump Administration—the shift from bilateral executive/legislative cooperation and compromise in federal spending to executive unilateralism in defiance of congressional control—has forced courts into the administration of federal spending like never before. As a result, judges have been forced to confront legal questions that remained mostly dormant during the era of executive/legislative cooperation, including questions about the constitutionality of efforts to encourage through spending conditions what the federal government cannot mandate (or has not mandated) directly, like speech or silence on particular subjects. Early decisions resolving these questions have failed to appreciate an important distinction. Statutory conditions articulated by Congress or an executive actor formally exercising delegated legislative authority (statutory conditions) and executive conditions articulated by executive branch actors exercising non-enforcement, agenda-setting, or other executive powers (executive conditions) are different in constitutionally salient ways that can and should inform the development of workable doctrine. Yet courts, advocates, and scholars have only begun to map these differences and to draw the lines that will help to inform doctrinal development.

Work is needed to develop differences between statutory and executive conditions from the various perspectives courts consider in resolving constitutional questions—text, history, tradition, and function. After distinguishing statutory and executive conditions, the Article focuses on functional differences. It develops differences between statutory and executive conditions from the standpoint of the constitutional values that courts consider in resolving hard interpretive and precedential questions. Separation of powers, federalism, and anti-subordination values all favor closer constitutional scrutiny of executive conditions than courts usually apply to statutory conditions. At the same time, operational considerations create a reason for courts to hesitate to scrutinize executive conditions that may not be present with regard to statutory conditions—namely, that review of executive conditions could interfere with day-to-day program administration and so diminish federal capacity.

Together, the functional differences between statutory and executive conditions indicate that, rather than evaluating executive conditions by reflexively applying (or refusing to apply) precedents developed in review of statutory conditions, courts should develop distinctive constitutional doctrines for executive conditions tailored to target and address abuses without unduly interfering with day-to-day program administration. Doing so not only would lead to better outcomes from the standpoint of constitutional values but would also facilitate doctrinal development by supporting clear, workable lines in an area in which such lines are notoriously hard to draw. The Article illustrates with two specific prescriptions. In the First Amendment context, courts have struggled to discern when a spending condition that requires (or forbids) expression is a permissible aspect of program definition or is impermissible leveraging of a program to regulate speech more broadly. By giving closer scrutiny to executive conditions articulated by novel executive branch actors or compounded by the executive branch across distinct statutory programs, courts can draw better lines and check abuses without interfering with ordinary program administration. And in the Spending Clause context, courts have struggled to identify when pressure turns to compulsion. Here, we suggest executive conditions that pull resources to achieve executive goals in excess of amounts appropriated or purposes delineated by Congress require more searching judicial scrutiny to safeguard Congress’s power of the purse.

Introduction

Conditions that a person or entity must satisfy to receive something of value from the government are a familiar regulatory tool.^[3] At the national level, from civil rights to health care to defense, spending conditions help the federal government fulfill its purposes.^[4] For example, conditions on Medicare and Medicaid dollars helped to desegregate southern hospitals and fight COVID-19,^[5] and conditions on Medicaid and Affordable Care Act funding enable the federal government to operate those programs alongside—rather than in lieu of—states, combining federal fiscal capacity with state sovereignty and managerial expertise.^[6]

Like other tools that the federal government uses to achieve its purposes, federal conditions on spending are subject to constitutional constraints. In cases challenging the constitutionality of conditions Congress imposes through statutes, federal courts have developed doctrines to limit the leveraging of federal dollars to regulate speech protected under the First Amendment;^[7] to prevent disregarding federalism values through conditions on federal spending statutes;^[8] and to ensure that conditions do not circumvent constitutional limits on targeting individuals for punishment through bills of attainder.^[9]

The second Trump Administration has made unilateral, extensive, and often pioneering use of conditions to change behavior without new lawmaking by Congress,^[10] prompting scores of lawsuits in an area in which litigation had been relatively unusual.^[11] These lawsuits include challenges to an executive order imposing requirements that each recipient of federal funds “certify that it does not operate any programs promoting DEI that violate any applicable Federal anti-discrimination laws”;^[12] challenges to executive orders sanctioning law firms by revoking or precluding government contracts, making threats of such sanctions, and requiring deals between law firms and executive branch actors to agree to such conditions to avoid sanctions;^[13] challenges to executive actions penalizing universities by terminating grants, making conditional threats of losing federal funding for research, and negotiating deals to avoid sanctions;^[14] and challenges to executive actions imposing new conditions on grant awards, or non-termination, based on agency “priorities.”^[15] Of course, the Trump Administration is not the first presidency to have its use of conditions be questioned,^[16] but its unprecedented willingness to defy congressional control has wrought an extraordinary number of spending controversies.^[17] In this new era of appropriations presidentialism, what was once rare has become routine.^[18]

All this novel constitutional litigation is forcing judges to confront legal questions that lay dormant during the prior era of bilateral executive/legislative compromise around the administration of federal spending. The doctrines courts now refine and develop will come to shape the way the federal government will work in the future.

Unconstitutional conditions doctrines have been front and center in the new wave of disputes about federal spending conditions, in part because federal district courts cannot address statutory and procedural issues that would otherwise moot overlapping constitutional questions.^[19] But the Supreme Court’s leading unconstitutional conditions cases arose in the context of challenges to conditions imposed through federal legislation,^[20] creating a significant, largely unanswered question about whether and how such doctrines apply in the context of challenges to conditions imposed by executive branch actors. The early cases arising out of Trump’s appropriations presidentialism are inconsistent on this issue. Unsurprisingly, the Department of Justice has argued that doctrines developed in the context of constitutional challenges to conditions imposed by Congress through legislation do not apply to conditions imposed by the executive branch through executive orders and actions.^[21] Some courts have agreed,^[22] while others have applied doctrines developed in the context of statutory conditions in reviewing conditions imposed by executive action without comment.^[23] Questions have also emerged about the applicability of doctrine in the other direction—whether doctrines developed in the context of constitutional challenges to executive action unrelated to spending (such as conditions on criminal investigation and prosecution) apply to funding conditions articulated by executive actors.^[24] Questions have also emerged about the extent to which Spending Clause and separation of powers limits entail distinct checks on executive conditions in light of the risk that such conditions might encroach on congressional control over spending and undermine the Appropriations Clause’s goal of ensuring that Congress controls the funding necessary to accomplish resource-dependent goals.^[25]

This set of questions is ripe for synthesis, but prior scholarship has only begun to consider how the fact that a condition is articulated by the executive branch through some form of executive action, rather than by Congress through a statute, might impact the constitutional questions that conditions raise.^[26] This symposium contribution advances the conversation by developing a distinction between two varieties of conditions and arguing that this distinction can and should matter for the development of unconstitutional conditions doctrines. The Article differentiates statutory conditions imposed directly by legislation or by final agency action exercising delegated legislative authority (as through a regulation) from executive conditions imposed (or threatened) by an executive branch actor through its exercise of enforcement discretion, agenda-setting choices, or decisions whether to behave illegally.^[27] The Article argues that courts confronted with such hard constitutional questions about executive conditions should not simply apply doctrines developed for statutory conditions to answer those questions because executive conditions are different in constitutionally salient ways. Rather, courts should develop—and it is both possible and helpful to develop—particularized executive conditions doctrines tailored to the distinctive challenges such conditions present.

The distinction we draw may be intuitive to administrative law experts familiar with the test for whether an agency pronouncement is a “legislative rule” (insofar as it also focuses on authoritativeness) and constitutional lawyers familiar with Justice Jackson’s Youngstown framework (insofar as it also focuses on whether the executive branch is acting with the support of Congress, in the face of congressional silence, or in opposition to Congress).^[28] But the distinction is somewhat subtle and addresses the wonky mechanics of the administrative state, where some agency actions exercise authority delegated to the executive branch by a statute and others do not. As we explain, a “statutory condition” includes both conditions imposed explicitly in legislation (like the Hyde Amendment) and conditions imposed by an executive branch actor formally exercising legislative authority delegated by a statute, such as rulemaking or adjudication (using Administrative Procedure Act terms).^[29]

On the other hand, we describe an “executive condition” as any condition imposed by an executive branch actor exercising executive, rather than delegated legislative, power. This includes conditions on the exercise of non-enforcement discretion (like the requirement federal grantees certify elimination of DEI programs to avoid investigation), conditions on an agency’s choice to take some discretionary action (like those in the “compact” proposed to universities in exchange for favoritism in grant awards), or conditions on an agency’s choice whether to take a discretionary or even illegal (or arguably illegal) action (like the threat to cancel security clearances absent a promise of free legal services made to Paul Weiss, Jenner & Block, and other law firms).

To fully develop the missing constitutional law of executive conditions, it will be necessary to explore differences between statutory and executive conditions from all of the major perspectives courts consider in resolving constitutional questions—text, history, tradition, and function. The Article largely focuses on functional differences. Courts developing unconstitutional conditions doctrines have considered functional questions about how particular practices impact constitutionally salient values like political accountability, government capacity, state sovereignty, and individual liberty.^[30] The Article argues that statutory conditions, whether articulated explicitly in a statute or by an executive branch actor pursuant to a statutory delegation, are functionally similar, so they should be subject to the same doctrinal checks as courts have developed in challenges to conditions articulated directly in statutes. Indeed, as we discuss below, plaintiffs could defeat any effort to treat such conditions differently from purely legislative conditions by raising “as-applied” challenges to underlying statutes conferring the executive’s authority.

By contrast, executive conditions differ from statutory conditions in important and cross-cutting ways. Executive conditions are less susceptible to the checks of political processes and judicial scrutiny that improve accountability and protect liberty because they do not require debate or participation and can be targeted, ephemeral, and secret.^[31] Yet, judicial intervention to check allegedly unconstitutional executive conditions could pose a greater risk of interference with the day-to-day operation of government that is carried out by administrative agencies, potentially threatening both the federal government’s ability to achieve its purposes and Congress’s role in the administration of fiscal programs through legislation, appropriations, and oversight.^[32] And while both statutory conditions and executive conditions have the potential to threaten state sovereignty and individual liberty, executive conditions trade concerns about legislative overreach for a risk of executive incursion on Congress’s power of the purse.

Reconciling these contrasting considerations yields two overarching prescriptions. First, executive conditions are different from statutory conditions in constitutionally meaningful ways, so courts should not apply precedents developed in the statutory condition context to executive conditions, except in limited circumstances. Rather than treat such prior precedents as wholly applicable or inapplicable, which has largely been the pattern so far, courts should only apply statutory conditions doctrines mutatis mutandis—that is, with necessary changes made for the distinct context. Second, courts, advocates, and scholars should develop doctrinal checks on executive conditions that avoid unduly inserting courts into the day-to-day operations of federal programs while simultaneously ensuring executive conditions do not evade the accountability that usually occurs through judicial review and political processes.

While full elaboration of implications for the doctrinal questions surrounding executive conditions is not possible here, we briefly illustrate the workability of our intervention by explaining how it can inform constructive doctrinal line-drawing—tests that root out serious abuses and promote accountability without unduly interfering with everyday administration—on two unsettled questions. First, we propose that courts tasked with translating the Supreme Court’s 2013 anti-leveraging decision in United States Agency for International Development (USAID) v. Alliance for Open Society International, Inc. from the statutory to the executive context should focus on whether an executive condition is articulated by the executive branch actor ordinarily engaged in program administration or by a novel actor who is not. Second, when it comes to whether and how Spending Clause doctrine might check executive conditions, we suggest that courts should not simply import the Dole test but must try to isolate whether an executive condition threatens congressional prerogatives (similar to the state sovereignty evaluation typical in evaluating conditional spending for cooperative federalism programs). Courts should ask whether a particular executive condition would permit the executive to achieve identifiable non-statutory, resource-dependent goals despite Congress’s attempt to control resources to achieve those goals through appropriations.^[33]

The stakes are currently obvious and urgent. In particular, the doctrinal approaches that come to be endorsed by courts of appeals and the U.S. Supreme Court when they confront the first wave of substantive constitutional issues created by appropriations presidentialism will define the limits of unilateral presidential spending authority, shaping the structure of American government and reality of Americans’ liberty for as long as executive unilateralism is ascendant. But more subtle implications will materialize whether or not the norm of bilateralism in the administration of federal spending makes a comeback. Careless or aggressively rigorous doctrines for reviewing executive conditions could hobble executive capacity, limiting the ability of the federal government to “establish Justice, insure domestic Tranquility, provide for the common defense, promote the general Welfare, and secure the Blessings of Liberty.”^[34] Similarly, unless courts carefully distinguish executive from statutory conditions, new precedents responding to executive aggrandizement may inadvertently diminish legislative power, compounding and cementing the resulting diminution of Congress.

Part I briefly describes the history of spending conditions as a critical tool of governance and summarizes recent questions in litigation about the applicability of precedents developed in constitutional challenges to statutory conditions in the context of executive conditions (and vice versa). Part II develops a functional approach for separately analyzing statutory and executive conditions, concluding that executive conditions on non-enforcement, discretionary action, and illegal action are different from conditions articulated through statute, rulemaking, or binding adjudication in cross-cutting, constitutionally salient ways. Part III is prescriptive, illustrating how our approach facilitates line drawing in a notoriously murky area.

I. Background

A. Conditions and Government Capacity

Conditions on spending are a common and usually unobjectionable tool that the federal government uses to achieve national policies. It is hard to imagine a spending or tax program that does not come with some kind of condition—a subsidy of $1,000 for every citizen conditions payment on citizenship; a gas tax rebate of $500 per vehicle owner conditions the rebate on vehicle ownership; a procurement contract for the purchase of ten gunboats requires bidders to build and deliver the gunboats; and so on. Moreover, conditions on federal spending and taxing are a way to influence behavior without immediately threatening sanction, arrest, or imprisonment, though these can follow for violation of such conditions. This can make conditions preferable to mandates from the perspective of liberty, autonomy, and state sovereignty. A government subsidy to encourage replacement of aging cars^[35] may seem a slight to those unwilling to part with gas-guzzlers, but even the staunchest opponents presumably would not prefer a government mandate forbidding aging vehicles from public roadways altogether.

Conditions can achieve results that mandates alone could not. For example, after a federal court held in Simkins v. Moses H. Cone Memorial Hospital that the Equal Protection Clause prohibited racial segregation of federally funded hospitals,^[36] most hospital segregation was unconstitutional.^[37] However, such segregation remained commonplace throughout the South until the Civil Rights Act of 1964 and the Medicare Act combined to make compliance with Title VI, which prohibited racial and ethnic discrimination in programs that used federal funds, a condition of hospital eligibility for Medicare payments for senior citizens in 1965. The intersection of funding conditions facilitated desegregation of most hospitals within a span of months.^[38]

As this illustrates, Congress often creates conditions explicitly in legislation. Indeed, the very first Congress put conditions on funds for military procurement to ensure the dollars it appropriated purchased the sorts of equipment it thought necessary—conditions on spending enabled it to fund the purchase of gunships but not muskets.^[39] However, statutory conditions may also be clarified and elaborated by administrative agencies in the regular course of program administration. In Biden v. Missouri, for example, the Supreme Court upheld a binding regulation (promulgated by the Secretary of Health and Human Services (HHS) through rulemaking) that required Medicare providers to certify their staff were either vaccinated against COVID-19 or regularly tested for the virus.^[40] The Court found that Congress had authorized HHS to create such a condition in implementing the statute’s requirement that hospitals and other health care providers protect patient safety.^[41]

Moreover, conditions can emerge in the give-and-take of cooperative federalism or privatized programs as a federal agency works with a state, contractor, grant recipient, or other funding recipient to identify and address compliance and steer program execution consistent with agency priorities. Medicaid and Affordable Care Act waivers offer a rich, well-studied example of this give-and-take.^[42]

B. Constitutional Checks on Statutory Spending Conditions

The Spending Clause has long been understood to permit Congress to spend for the general welfare and to place conditions on federal spending; indeed, as discussed above, any effort to define a program can be framed as creating conditions on the spending.^[43] Under the Supremacy Clause, conditions imposed through federal law preempt contradictory state laws.^[44] And Congress may delegate authority to impose conditions on spending so long as it provides intelligible principles to guide agency implementation of the conditions.^[45]

That said, conditions can raise important constitutional issues for the freedom of speech, due process, prohibition on bills of attainder, and federalism limits on the spending power, among others.^[46] We do not have space here to comprehensively survey all these doctrines, but we briefly highlight two of the most important: First Amendment anti-leveraging and federalism-based limits on conditional spending under South Dakota v. Dole.

1. First Amendment Protection from Leveraging

In USAID, a statutory provision limited eligibility for funding to combat HIV/AIDS to non-governmental organizations that both promised not to use federal funds in efforts to support legalization of prostitution and had a policy opposing prostitution and sex-trafficking. USAID implemented these conditions through a certification requirement in grant applications.^[47] The Court held the second condition (requiring an explicit policy opposing prostitution and sex trafficking) violated the First Amendment by not merely defining the spending program but seeking to leverage spending to influence advocacy outside of the program.^[48] The Court stated: “[T]he relevant distinction that has emerged from our cases is between conditions that define the limits of the government spending program—those that specify the activities Congress wants to subsidize—and conditions that seek to leverage funding to regulate speech outside the contours of the program itself.”^[49] Courts and scholars have understood USAID as the latest in a line of “anti-leveraging” decisions that address the federal government’s use of conditions, especially conditions on funding, to compel or suppress speech.^[50]

2. Conditions on Spending

In South Dakota v. Dole, the Court announced a four-part test for analyzing the constitutionality of conditions on funding Congress offers to states or private parties^[51] to participate in national policies.^[52] The spending must promote the general welfare, conditions must be unambiguous, conditions must be germane to the federal interest, and the conditions must not violate other constitutional provisions (like the First Amendment); in dicta, the Court observed that a point might exist where financial inducement impermissibly coerces states.^[53] Several decisions built out aspects of this test, including Pennhurst, which pre-dated Dole (and became the second element of the Dole test),^[54] wherein the Court held that certain language in the Developmentally Disabled Assistance and Bill of Rights Act of 1965 could not be a condition on funding if states could not have “ascertained” the conditions.^[55] More recently, National Federation of Independent Business v. Sebelius turned the Dole coercion dicta into doctrine by holding the ACA’s mandatory expansion of Medicaid eligibility was an unconstitutionally coercive condition on states receiving Medicaid funding.^[56] The coercion doctrine is nebulous, because the Court’s analysis seemed to rely at least in part on the germaneness of the Medicaid expansion to the rest of the Medicaid program, as well as the possibility that a state could lose 100 percent of its funding for noncompliance (which was always possible under the Medicaid Act). Whether the constitutional problem is the amount of federal funding, the percentage that could be lost, or the link between new funding conditions and older spending programs, as well as how to analyze these issues, remains undecided.^[57]

C. Appropriations Presidentialism Centers Questions About Constitutional Checks on “Executive” Conditions

These and other constitutional checks on federal spending conditions (such as vagueness doctrine under National Endowment for the Arts v. Finley^[58] and the ambiguity of conditions under Pennhurst) present difficult questions, like how to separate program definition from leveraging outside of the program in First Amendment cases^[59] and the point at which encouragement becomes coercion in Spending Clause cases.^[60] These questions have moved to the fore amidst the appropriations presidentialism of the second Trump Administration.

The Trump Administration’s assertion of broad, unilateral power to condition and terminate federal spending programs^[61] prompted scores of lawsuits challenging new conditions on federal funds.^[62] The lawsuits arising out of appropriations presidentialism have forced courts repeatedly to address questions about the limits of unconstitutional conditions doctrines that rarely came up in the past for at least two reasons. First is simply the volume of executive actions imposing unconstitutional conditions during 2025 and the frequency with which these actions have occurred without clear or even arguably legal basis.^[63] Second, and more subtly, the Court’s “shadow docket”^[64] ruling in Department of Education v. California created pressure for plaintiffs to forgo overlapping regulatory or statutory claims in favor of constitutional claims to avoid being channeled out of federal district court and into the Court of Federal Claims, which lacks the power to award timely emergency relief.^[65] Cases that would have been litigated as APA cases prior to California are now being litigated as constitutional cases.

The details of this expanding body of constitutional spending doctrine matter. Careless development of doctrine today may yield “false negatives,” failing to award relief to plaintiffs that deserve it and facilitating executive branch aggrandizement that threatens liberty directly (by exposing Americans to new forms of unchecked presidential control) and indirectly (by degrading the separation of powers). At the same time, careless development of doctrine even in “no brainer” cases may yield future “false positives,” setting up decades of undesirable judicial interference in program administration that hobbles both administrative and legislative capacity.

A recurring question in these cases has been whether and how constitutional doctrines developed in the context of challenges to statutory funding conditions (such as the conditions in Pennhurst, NFIB, and USAID) apply to conditions imposed by the executive branch in the course of program administration. In some cases, advocates have argued and courts have applied doctrine developed in challenges to statutory conditions to executive conditions without comment.^[66] In other cases, however, the potential difference between statutory and executive conditions has been an important point of contention. Consider four recent controversies.

First, in Wilmer-Hale’s challenge to an executive order and associated agency actions revoking security clearances, scrutinizing Wilmer-Hale clients’ government contracts, and otherwise targeting the firm for adverse discretionary executive actions,^[67] the firm raised a Spending Clause challenge arguing the order “impose[d] unconstitutional conditions on federal contracts.”^[68] The Department of Justice responded that Spending Clause doctrine was inapplicable because the conditions were imposed by executive order rather than legislation. The district court agreed,^[69] and its approach has been adopted in at least one subsequent case.^[70]

Second, in Harvard’s challenge to an executive order and associated agency actions terminating $2.2 billion worth of federal grants, awarded by more than a dozen federal agencies, as punishment for alleged failure to address antisemitism,^[71] the university argued the termination entailed an effort to condition continued receipt of funds on the university’s expressive choices in violation of the First Amendment’s anti-leveraging doctrine, citing USAID.^[72] The DOJ argued in response that the First Amendment’s anti-leveraging doctrine does not apply because the condition was a proposed executive settlement of a potential legal action for underlying violations.^[73]

Third, in Perkins Coie’s challenge to an Executive Order analogous to the Wilmer-Hale order, amici argued the order is an unconstitutional bill of attainder.^[74] Again, DOJ’s response was that the prohibition on bills of attainder does not apply to executive actions in the first place.^[75] In an opinion avoiding the question as not properly presented, the district court noted that the application of the prohibition on bills of attainder “to unilateral presidential action not otherwise authorized by Congress, is worth further study and consideration.”^[76]

Fourth, a number of grant termination and grant condition cases hinge on interpretation of a standard clause that agencies include in grant agreements made with awardees pursuant to 2 C.F.R. § 200.340(a)(4) (the “convenience clause”) or raise questions about conditions imposed through this clause. The convenience clause permits termination associated with changes in “program goals or agency priorities.”^[77] The Trump Administration asserted this clause gives it broad authority to cancel awarded grants for failure to comply with newly announced policy priorities. State grantees have argued that reading the clause to have such an effect would violate Dole’s requirement that conditions on funds be clearly stated.^[78] This argument is in some tension with the fact that, historically, courts have ignored Pennhurst and instead applied ordinary contract interpretation principles in disputes over the interpretation of grant agreements entered between a state and an agency.^[79]

Related questions loom. Courts have also been asked to consider how doctrines developed to check executive enforcement discretion in other contexts might apply to grant administration. Specifically, district courts have divided as to the relevance of the Court’s recent unanimous decision in National Rifle Association of America v. Vullo in the spending context. In Vullo, the Court found a threat to investigate and prosecute banks that did business with the NRA violated the First Amendment’s free speech protection,^[80] holding: “To state a claim that the government violated the First Amendment through coercion of a third party, a plaintiff must plausibly allege conduct that, viewed in context, could be reasonably understood to convey a threat of adverse government action in order to punish or suppress the plaintiff’s speech.”^[81] Two district courts have found Vullo is inapplicable to executive conditions on spending-related enforcement actions,^[82] while one court found Vullo’s principles—in particular the functional focus on the likely effect of a condition on speech—to be informative in adapting existing doctrines.^[83]

Little prior precedent or scholarship aids courts in addressing these questions.^[84] Moreover, while adversarial litigation can tease out precedential or interpretive arguments vis-à-vis particular doctrines, it is poorly positioned to explore overarching implications of the manner or source of a federal condition’s creation. The next Part turns to these implications.

II. Purely Executive Conditions Are Different

A. Statutory Versus Executive Conditions

In any given case, fully exploring the way the source or manner of a condition’s creation might impact its constitutionality from the standpoint of a particular constitutional provision (freedom of speech, spending power, etc.) could depend, among other things, on historical, textual, or precedential considerations specific to that constitutional provision.^[85] A comprehensive analysis of all such factors is beyond the scope of this Article, though we hope to inform future work.

That said, as a starting point we see a clear dividing line between two types of conditions with important, cross-cutting implications for the values underlying constitutional conditions doctrines. This is the line between statutory and executive conditions previewed in the introduction. The line is not simply between conditions communicated by the legislature, on the one hand, and conditions communicated by the executive, on the other. Conditions communicated by the executive often merely interpret or implement conditions first articulated by Congress, using power to do so delegated by statute. Rather, the important distinction depends on the nature of the authority on which a condition communicated by an executive actor is based. The important distinction is between conditions articulated by executive actors exercising delegated legislative authority, on the one hand, and conditions articulated by executive actors wielding executive powers, on the other.^[86]

In this Article, we thus use the term “statutory conditions” to include both conditions imposed explicitly or specifically by a statute and conditions articulated by executive branch actors exercising delegated legislative power, either to clarify or specify a statutorily articulated condition or to articulate new conditions. And we use the term “executive conditions” to refer to conditions articulated by executive branch actors not wielding delegated legislative power.

Agencies exercising delegated legislative authority to create legally binding rules, which can include conditions, typically do so through rulemaking.^[87] Agencies may also issue binding conditions exercising legislative grants of authority through adjudication.^[88] On the other hand, agencies also articulate conditions that, while influential, are not themselves created through the exercise of delegated legislative power. These conditions do not have the force of law but might rely upon executive non-enforcement discretion, agenda-setting discretion, or an ultra vires threat of illegal action.^[89] They may come in the form of “guidance,” individualized letters, email messages, settlement offers, social media commentary, or other ad hoc methods.

The distinction between conditions imposed by executive branch actors exercising delegated legislative power, on the one hand, and conditions on the exercise of executive power, on the other, is familiar in administrative and constitutional law. It tracks the distinction between “legislative rules” that must go through notice and comment rulemaking under the APA and “guidance” that, because “nonbinding,” need not,^[90] as well as the distinction under United States v. Mead Corp. between agency actions with the “force of law” and those lacking such effect.^[91] It also tracks the distinctions that Eloise Pasachoff has drawn among federal grants—“policy,” “pork,” and “punishment.”^[92] Finally and most fundamentally, this distinction also tracks Justice Robert Jackson’s Youngstown reasoning and framework for assessing the limits of presidential power. A condition imposed exercising delegated legislative power reflects congressional support and fits in the realm of maximum authority to act with the “express or implied authorization of Congress,” while a condition imposed without exercising delegated legislative power would either contradict congressional power or fall within the “zone of twilight” in which the President and Congress may have “concurrent authority.”^[93] (We return to Youngstown’s utility for building a framework in Part III.)

Enforcement discretion—the power to decide whether or not to pursue sanctions or other means of enforcement for alleged violations of law—is perhaps the best-understood way executive actors can impose conditions without exercising delegated legislative authority.^[94] Using enforcement discretion, executive branch actors can impose conditions on settlement, non-enforcement, or abandonment of/restraint from an investigation. While not legally binding when offered (a target need not accept a settlement offer), such conditions can (like other executive conditions) become legally binding if incorporated in an enforceable settlement agreement or contract. Conditions on the exercise of enforcement discretion are routine in the operation of cooperative federalism programs. For example, in assessing whether to pursue sanctions for a state’s alleged non-compliance with statutory conditions under the federal law of the Medicaid program, HHS might forgo enforcement proceedings if a state takes specified remedial steps.^[95]

In addition to enforcement discretion, executive branch actors may also impose de facto influential conditions by threatening to act^[96] in ways that, though illegal or potentially illegal, would cause irrevocable harm to the target.^[97] The executive’s role in distributing funds appropriated by Congress and collecting taxes makes it distinctively capable of action or will^[98]—seizing funds, refusing to pay grants, and so on—whether or not those actions are legally authorized. Moreover, the executive can move quickly and first, whereas other branches move slowly and, in the case of courts, may only react. As a result, threats of unlawful action may be used as leverage to force desired conduct or terms on threatened actors, even if the threatened action does not occur. The effectiveness of an executive threat to take illegal action presumably depends on many things, including the likelihood that subsequent judicial review or other remedial action would make the threatened entity whole and the threatened party’s strength and capacity to bear risk. But regardless of the precise usefulness of threats of illegality as a tool of influence, there is wide space in existing doctrine for such threats to be effective, created by the inherently limited capacity of courts, threshold (non-merits) barriers to judicial review, the cost of legal representation to litigate (and lack of free services), and limits on courts’ remedial authority.^[99]

Furthermore, executive branch actors may impose de facto binding conditions by offering to exercise discretion to take (or not take) a legally authorized action only if certain conditions are met. In short, they may create a condition via the exercise of agenda-setting power. Consider, for example, HHS guidelines governing whether to grant state requests to depart from federal Medicaid rules pursuant to a “demonstration project waiver.”^[100] The statute giving this power to the HHS Secretary is worded broadly, and the implementing guidelines have filled the gaps in a variety of ways, including, for example, a requirement that state waiver proposals be “budget neutral” for the federal government. Until very recently this was not a condition drawn from the statute, but a policy HHS created for determining whether to exercise its discretion to grant a state’s waiver application.^[101] Though this condition entailed (to some experts) funny math, states honored it in developing demonstration waiver applications.^[102]

To illustrate this distinction, Table 1 lists several well-known conditions that we would describe as “statutory conditions” because they were created through the exercise of direct or delegated legislative authority. Table 2 then offers well-known conditions that we would describe as “executive conditions” because they are implemented as a condition on enforcement discretion, on the exercise of a discretionary power, or as part of a threat to take illegal executive action.

Table 1: Illustrative Statutory Conditions

Condition	Statutory Source	Manner of Expression
Medicaid requirement that states reimburse providers sufficiently to support access	42 U.S.C. § 1396a(a)(30)(A)	Directly articulated in statute
Medicare requirement that providers periodically test unvaccinated employees for COVID	42 U.S.C. § 1302(a); 86 Fed. Reg. 61555 (Nov. 5, 2021)	Agency promulgated condition through notice and comment rulemaking in exercise of statutory authority to promulgate rules for efficient administration of Medicare and Medicaid.
Requirement that recipients of certain federal grants adopt a policy against human trafficking (at issue in USAID)	22 U.S.C. § 7631(f)	Condition in statute; agency promulgated regulation with more specific interpretation of statute pursuant to statutory authority to promulgate implementing regulations.
Requirement that Medicaid waivers be “budget neutral” (after 2025)	H.R. 1 § 71118	Congress amended the Medicaid statute to make this de facto condition on the executive branch’s approval of waivers an explicit statutory requirement.

Table 2: Illustrative Executive Conditions

Condition	Underlying Executive Power	Manner of Expression
Requirement in Cole memo that marijuana dispensers comply with state laws to be eligible for federal enforcement discretion	Enforcement discretion	Articulated by Attorney General in open letter[103]
Requirement that Medicaid waivers be “budget neutral” (prior to 2025)	For state waiver applications, executive discretion to deny based on this interpretation of a broadly worded statute. For approved waivers, a deal between HHS and state that was reflected in waiver “terms and conditions”	For waiver applications, a condition articulated by HHS in guidance to states regarding its priorities in considering waiver requests For approved waivers, inclusion in waiver terms and conditions
Requirement that DACA applicant be under 18	Enforcement discretion	Included in executive memorandum announcing DACA
Requirement that Paul Weiss donate pro bono time to government causes	Enforcement discretion/threat of illegality	Included in communication to firm threatening to cut off client contracts and security clearances

B. Statutory Conditions Are Functionally Alike Regardless of Which Branch Communicates Them

Conditions articulated directly by statute and by executive branch actors exercising delegated legislative power are not different in ways that necessarily require distinct doctrinal treatment. By contrast, major functional differences justifying distinct doctrinal treatment do exist between statutory conditions and executive conditions.

To start, three considerations counsel against differentiating between conditions articulated directly in legislation, on the one hand, and binding conditions articulated by an agency through rulemaking or adjudication pursuant to delegated legislative authority, on the other.^[104] First, regulation is functionally analogous to legislation—it is transparent (rules and binding adjudications must be published), generally applicable, precedential, and susceptible to objective evaluation. While agency adjudication can take many forms, it is also ordinarily transparent and may be precedential. Moreover, the degree of interference with administrative capacity threatened by judicial review of rulemaking and completed adjudications is analogous to the degree of interference with legislative capacity threatened by judicial review of legislation. In either case, judicial review of a completed action hinders development of a particular rule but does not impede day-to-day implementation of that or other legal requirements.

Second, treating conditions imposed by the executive branch using delegated legislative authority differently from conditions imposed directly by legislation would create an easily exploited loophole.^[105] Where barred by First Amendment, Spending Clause, or other doctrines from directly legislating toward a particular end, Congress could simply enact open-ended legislation giving an executive official or the President power to achieve the same end. This would frustrate values such as liberty, political accountability, state sovereignty, anti-subordination, and other ends served by constitutional limits on statutory conditions. However executive in nature regulatory or adjudicatory actions exercising delegated lawmaking power might be, they are also “legislative.”

Third, a procedural obstacle exists for any effort by courts to treat conditions articulated by statute differently from binding conditions articulated by an executive branch actor exercising delegated legislative power. Plaintiffs could simply frame their challenge in any case as an as-applied challenge to the constitutionality of the underlying statute articulating broad conditions for an agency to further specify or empowering the agency to issue new conditions with the force of law of its own, rather than as a direct challenge to the regulation or adjudication promulgating a specific condition pursuant to that delegated authority.^[106] Indeed, that procedural approach may have benefits from the perspective of the value of political accountability because framing actions as “as-applied” challenges to particular statutes rather than only as challenges to executive actions would remind voters that such assertions of executive power are ultimately traceable to Congress’s choices in enacting or failing to amend broad legislative authority for executive branch actors.^[107]

In light of these points of overlap, we do not see a need from a functional perspective for development of distinct doctrines governing legislative conditions imposed by statute, on the one hand, and legislative conditions imposed by executive actors exercising delegated legislative power, on the other. Harder questions are presented by conditions that do not exercise delegated legislative authority.

C. Executive Conditions Function Very Differently

A skeptic might argue the distinction we are drawing is overly technical or trivial. We would call an HHS rule requiring hospitals to test for COVID pursuant to a statutory provision authorizing the agency to impose conditions of participation a “statutory condition” but HHS guidance telling states the agency would refuse to exercise its discretion to approve a Medicaid waiver unless it is budget neutral an “executive condition.” Yet, from the perspective of regulated parties the two seem equally binding, and from the perspective of those not steeped in administrative law, both may look like agency actions. Are “statutory conditions” and “executive conditions” really so different? Yes.

Despite superficial similarities, major functional differences exist between statutory conditions and executive conditions and profoundly impact values including liberty, accountability, and capacity that underlie constitutional checks. These differences, summarized in Table 3 and elaborated upon below, are that executive conditions (1) are an inevitable and desirable source of flexibility in program administration; (2) can be secret, ephemeral, and threatened; (3) are controlled by the executive branch; (4) are not subject to political process safeguards of federalism and liberty including bicameralism and presentment; and (5) can be imposed by an executive branch actor other than the one selected by Congress in related delegations of legislative power. These differences are cross-cutting; they counsel neither more stringent constitutional checks for executive conditions nor less stringent such checks; rather, they counsel carefully tailored checks on executive conditions.

First, a functional consideration counsels care in the development of constitutional checks on executive conditions. Such conditions can be inevitable and desirable sources of flexibility in the day-to-day administration of programs that aid regulators and regulated parties alike. They thereby increase the capacity of the government. From the bonds required of brokers to host livestock auctions^[108] to the staffing required of nursing homes to be eligible for Medicare reimbursement,^[109] binding legal requirements can be difficult to interpret, and the penalties for their violation can be stark. Discretion can bring on-the-ground expertise and essential, ongoing relationships of collaboration and trust into day-to-day governance.^[110] It can enhance executive capacity while minimizing regulatory burdens.^[111] Or it can serve as a warning to those who are violating the law to cease while encouraging those who are not to continue on their way.^[112]

Judicial review poses a potential threat to the flexibility that executive conditions may bring. Executive conditions can be developed informally—a few words, even a look from an investigator, or a hasty report and recommendation by an audit team^[113]—without the formal legal process entailed in the issuance of binding pronouncements through rulemaking or adjudication. Judicial review of such conditions could impose costly and time-consuming processes, and the mere threat of such review may prompt prophylactic formalization. Administrators may choose rigidity and secrecy to reduce litigation risk.

Second, to state the obvious, executive conditions empower the executive rather than the legislative branch. This can cut two ways for the development and application of constitutional checks. On one hand, it means that concerns do not apply about non-review creating a loophole by which the legislative branch can achieve indirectly what it cannot achieve directly. So, to the extent that courts are concerned with limiting legislative power qua legislative power, rather than limiting governmental power more generally, that concern does not apply to executive conditions.

At the same time, executive conditions bring a danger of executive encroachment on powers the Constitution gave Congress—namely, the Appropriations Clause and the spending power. Conditions can be used to generate resources for the government; for example, consider law firms’ agreement to provide pro bono services consistent with President Trump’s priorities to avoid threatened sanctions.^[114] As the Supreme Court recently emphasized, the Constitution’s reservation of control over the attainment and utilization of resources by the federal government reflects a founding-era consensus about the importance of separating power over the purse and the sword: “By the time of the Constitutional Convention, the principle of legislative supremacy over fiscal matters engendered little debate and created no disagreement.”^[115] The use of executive conditions to coerce behavior arguably violates the Appropriations Clause and the Spending Clause, both by undermining their design and by usurping congressional authority.^[116] Where conditions entail the exercise of delegated legislative power, Congress can effectively prevent an agency from imposing conditions by either repealing the delegation or including language that prevents such use. But Congress’s ability to check conditions on the exercise of enforcement discretion or the taking of discretionary actions is severely limited.^[117]

This concern that executive conditions can encroach on legislative power in ways that Congress cannot check led the Ninth Circuit to find that a limit imposed by the DOJ on grant funding for “sanctuary cities” that was not authorized by statute was not only ultra vires but also a violation of the Spending Clause; other courts have reached similar conclusions.^[118] On this theory, to which we return in Section III.B, the Appropriations Clause and the Spending Clause are an outer bound on executive powers much the way that the Court in NFIB saw federalism concerns forming an outer bound on the commerce power and the spending power.^[119]

Third is a difference that impacts the likelihood courts will intervene to check unconstitutional executive conditions, whatever doctrines apply.^[120] Executive conditions can be not only targeted but made in secret (never shared with the public), threatened (not merely imposed, but leveraged to achieve concessions), and ephemeral (issued not as a command but as a mere suggestion or in a way that is ambiguous about whether it is a command or a suggestion, and temporary). All of these things increase the likelihood that unconstitutional executive conditions will go unchecked by courts, no matter how stringent the doctrines courts apply.^[121] The possibility of secrecy and threats also brings rule of law concerns, increasing the likelihood that individuals’ treatment may be due to personal factors and that like situations will not be treated alike—a fundamental feature of the rule of law.^[122]

Consider Scott v. Department of Health and Human Services, in which Florida alleged HHS unconstitutionally threatened not to renew the state’s existing Medicaid waiver unless the state agreed to expand Medicaid eligibility under the ACA.^[123] The grant of a demonstration project waiver itself is an exercise of delegated legislative power under Section 1115 of the Social Security Act and must be done publicly, but threats regarding the conditions under which the Secretary might approve, decline, or renew a waiver may be informal and could occur behind closed doors.^[124] In the litigation, federal officials denied they ever made the threat that state officials claimed.^[125] Amidst the factual dispute, Florida agreed to dismiss the case. But how do we know whether or not HHS made similar threats to other states, which then chose to expand rather than risk losing millions of federal dollars?

The potential for secret, ephemeral, and threatened conditions cuts in two directions. Plainly, it cuts in favor of tailoring doctrine to facilitate review and/or increase the costs to the executive of unconstitutional threats that come to light in order to deter those threats. On the other hand, however, it cuts against doctrines that would permit review only of publicly announced executive conditions for fear of discouraging the executive branch from publicizing conditions and potentially driving them into the shadows.

Fourth, executive conditions may be created unilaterally, meaning they do not need to reflect agreement of elected representatives in Congress or satisfy the constitutional checks of bicameralism and presentment, limiting political accountability. These limits on congressional action—the need for agreement among a large number of diverse representatives and the requirement of passage through both the House and the Senate—were intended by the Framers as a check on the risk of faction, or subordination, meaning that those who hold power will use it to punish or extract rents from those who are not in power.^[126] As James Madison put it, given that members of Congress would inevitably represent a “great variety of interests, parties, and sects . . . a coalition of a majority of the whole society could seldom take place on any other principles than those of justice and the general good,” leaving “less danger to a minor from the will of a major party.”^[127] So, too, the political-process limits built into the legislative process are understood to be an important safeguard of federalism and state sovereignty.^[128] Where conditions are imposed by the executive, such checks are not present. Thus, as Justice Kennedy explained, “liberty is threatened” when “the decision to spend [is] determined by the Executive alone.”^[129]

Fifth, and relatedly, executive conditions need not be articulated by the agency or actor designated by Congress in creating a program and need not satisfy the procedural checks on agency actions that exercise delegated legislative power set out in the APA. Because delegated legislative power to create binding rules can be exercised only by the agency actor(s) designated by Congress, the role of agency expertise is increased in condition-creating decisions. This may mitigate the magnitude of conditions by designating the agency (or agencies) responsible for a given rule and increase congressional influence insofar as particular agencies tend to be in close touch with and somewhat responsive to congressional committees with jurisdiction over their authorities and appropriations.^[130] Likewise, the procedural checks on binding agency action set out in the APA replicate (to an extent) process checks on legislative action.^[131] By contrast, executive conditions can be articulated by the President, the Attorney General, or other individual executive branch actors not designated by Congress and not subject to usual agency processes. This facilitates tying together the leverage made possible by multiple spending statutes^[132] while diminishing the effectiveness of committee-agency relationships that bring congressional influence into conditions, as well as circumventing procedural and workforce checks on agency action.

* * *

Executive conditions clearly can function differently from statutory conditions in ways that significantly impact the nature of judicial review and the effectiveness, benefits, and risks of constitutional checks. Table 3 summarizes these differences and implications.

Table 3: Summary of functional differences between statutory and executive conditions

Difference	Implication	Implication	Upshot
Source of program administration flexibility	Judicial review could interfere, reducing government capacity	Too much deference could leave executive discretion unchecked	Judicial review potentially costly, deference desirable in some situations
Executive, not legislature, controls	Lower risk of legislative self- aggrandizement	Higher risk executive uses conditions to circumvent Congress’s power of the purse	More stringent judicial check needed when condition is used for resource-dependent tasks[133]
Can be secret, targeted, threatened	Reduced checks such as political accountability, judicial review	Judicial checks only on public conditions may backfire and increase secrecy.	Checks should ensure programmatic success without sacrificing possible upside to closed-door negotiations but also should protect a threatened party.
No constitutional checks (or political safeguards) like bicameralism and presentment	Increased risk of executive power aggrandizement without other branches’ input and balance	Increased risk of faction, subordination, threats to state sovereignty	Stronger checks, especially when used to target parties outside political and judicial processes
Conditions can be imposed by an executive actor other than a legislatively designated agency/official.	Circumvent iron triangles	Pool inducements across agencies	Diminished political safeguards; contradict enabling legislation

From a functional standpoint, what should courts do? At a minimum, we suggest the possibility that these distinctions matter, perhaps a lot. Courts should not assume that either constitutional limits or particular doctrines operationalizing those limits that developed within challenges to statutory conditions would have the same effects if applied without alteration to executive conditions. Instead, they should consider ways that particular doctrines might exacerbate, address, or fail to address these functional differences and adjust the scrutiny and doctrinal tests they apply accordingly.

Importantly, the functional concerns weighing for or against judicial review do not apply equally to all executive conditions. Concerns that judicial review will interfere with desirable flexibility in implementation and enforcement of a statutory program do not apply if the executive uses conditions—such as threats of illegal action—not to advance the statutory program in good faith but rather to achieve ends outside of or even impermissible under the program. Similarly, concerns that the possibility of secrecy, targeting, and threatening may reduce the likelihood of judicial review apply only to secret, targeted, and threatened conditions. And the concern that executive conditions may reach across agencies, circumventing congressional influence, applies only when conditions are bundled or compounded. In other words, though valid, the concerns we anticipate do not outweigh the greater value in carefully drawing doctrinal lines that facilitate judicial review for executive conditions.

III. Line-Drawing Possibilities

Courts, advocates, and scholars should develop executive conditions doctrines that can thread the needle between stifling desirable flexibility in program implementation and ensuring meaningful checks to protect liberty, separation of powers, and state sovereignty. Courts and scholars developing the constitutional law of executive conditions could productively focus on doctrinal approaches that would account for the functional differences between executive and statutory conditions in terms of the kind of judicial review necessary (that is, how much deference or scrutiny) and the substance of the analysis (which doctrine is applicable).

We recognize that this point may seem abstract and conceptual and that courts are tasked not only with conceptualizing unconstitutional conditions but with identifying them in practice. Indeed, line-drawing in practice has proven a stubborn impediment to the development of unconstitutional conditions doctrines.^[134] The distinction between executive and statutory conditions can be helpful in such real-world line drawing. Consider two specific areas of ongoing controversy: Section A addresses limits on the use of executive conditions to influence expression protected by the First Amendment. Section B addresses limits on the use of executive conditions to circumvent congressional control through the Appropriations and Spending Clauses.

A. Novelty and Compounding as Considerations in Anti-Leveraging Cases

As we have discussed, executive branch actors often use conditions to influence the expression of institutions and individuals, implicating First Amendment limits on such leveraging. For example, the Trump Administration threatened to terminate security clearances and contracts for certain law firms unless they agreed to advocate in court on behalf of particular positions pro bono.^[135]

In assessing whether such efforts violate freedom of speech, courts have looked to USAID. As noted above, the Court explained in that case “the relevant distinction . . . is between conditions that define the limits of the government spending program—those that specify the activities Congress wants to subsidize—and conditions that seek to leverage funding to regulate speech outside the contours of the program itself.”^[136] The trouble with the USAID test—as the Court acknowledged in articulating it—is that it can be difficult to discern whether a given condition “define[s] the limits of [a] Government spending program” or “seek[s] to leverage funding to regulate speech outside the contours of the federal program.”^[137]

Focusing on what makes executive conditions distinctive yields two discrete, workable considerations that courts might use in adapting USAID from the statutory condition context to determine whether a particular threatened executive condition “seek[s] to leverage funding to regulate speech outside the contours of [a] program.”^[138] First, courts could focus on the executive branch actor who articulates a condition. If articulated by the agency staff or leadership ordinarily engaged in administering a program—for approving awards, assessing compliance, and the like—then that could be a reason to believe the condition “define[s] the limits” of the program. On the other hand, if a condition is articulated by an executive actor not ordinarily engaged in administering a program, such as the President, Attorney General, or an unusual political appointee not routinely involved in the program, then this could be treated as a reason to more closely scrutinize the condition on the concern that the condition seeks to “leverage funding” to regulate “outside the contours of the program.”

This line has much to commend it. As a matter of simple logic, conditions articulated by unusual actors are more likely to reflect an effort to use conditions to achieve goals “outside” a program because internal program goals can be advanced by the actors ordinarily engaged in administration, while external or novel goals may require enlisting unusual actors.^[139] This approach addresses the concern that executive conditions circumvent congressional controls by taking decisions out of the hands of agency actors who were given congressionally created authority. At the same time, focusing on conditions imposed by novel executive actors leaves ordinary program administration—in which conditions can be inevitable and desirable—to the usual workings of administrative law principles.^[140]

How would “usual” actors be discerned from “unusual” ones? Program participants and current or former administrators could, for example, file declarations describing statutory delegations, past practices, or typical signatories in analogous agency correspondence. Courts assessing materiality for False Claims Act purposes, for example, routinely make an analogous inquiry into what factors an agency considers in administering a particular spending program.^[141] This consideration would employ the same sort of evidence but as to a simpler question: not what the agency considers but who communicates the decision. And in some cases, the inquiry would be quite simple. To return to the example of litigation around Trump Administration executive orders addressing specific universities and law firms: While presidents routinely instruct agencies on priorities or agendas by executive order, it is not ordinary for the President to direct the administration of particular grants.

Second, courts could discern that executive conditions cutting across or compounding programs—such as threats to terminate all federal grants across an administering agency or even multiple agencies or to provide preference in all such grants for a particular actor—also seek to leverage funding to regulate speech “outside the contours of a program” rather than define the program.^[142] Indeed, while this Article was in press, a district court employed such an approach in adjudicating various groups’ challenges to new grant conditions imposed by HUD and HHS. The executive conditions prohibited funding for programs that advance “gender ideology; diversity, equity, and inclusion; elective abortions; and antidiscrimination,” raising issues under USAID.^[143] Judge Dubose held:

The landscape established by [USAID and related cases] considers challenges to funding conditions that were baked into the development of specific government programs by Congress, but here, the Challenged Conditions are being enacted “as extrinsic, extra-statutory conditions” on top of already existing and well-established programs. The categorical and expansive nature of the Challenged Conditions telegraph that the Defendants will deny federal funding to a whole class of programs based on viewpoint alone . . . . [T]hey likely violate the First Amendment.^[144]

Judge Dubose had it right. Simple logic counsels that compounding conditions across programs (which Judge Dubose described as “class” treatment^[145] and might be called “pooling”^[146]) seeks to leverage multiple programs to achieve a goal beyond any one component program. Moreover, compounding to maximize influence on a particular target warrants increased judicial scrutiny because it would be consistent with the constitutional prohibitions we have explored.^[147] It is also consistent with the Court’s concern in NFIB v. Sebelius regarding a threat to withhold funds allocated by previously enacted statutes (echoing Dole’s third prong, germaneness).^[148] While the NFIB plurality did not interpret the Medicaid Act accurately,^[149] the Court’s overarching concern is still salient: Just as the Court considered whether the possibility of withholding federal funding allocations across time and health insurance programs could increase pressure on states, so too pooled threats that combine federal funding allocations across programs may heighten pressure (which appears to be the intent). And again, scrutinizing conditions running across multiple programs leaves the flexibility deemed desirable in ordinary program administration undisturbed while maintaining checks on extrinsic executive conditions.

The two factors we suggest, actor novelty and compounding programs, also complement the Court’s recent decision in Vullo, which furthered anti-leveraging principles by stating an “official cannot do indirectly what she is barred from doing directly: A government official cannot coerce a private party to punish or suppress disfavored speech on her behalf.”^[150] The Court continued: “To state a claim that the government violated the First Amendment through coercion of a third party, a plaintiff must plausibly allege conduct that, viewed in context, could be reasonably understood to convey a threat of adverse government action in order to punish or suppress the plaintiff’s speech.”^[151] Vullo is especially useful for puzzling through the anti-leveraging analysis when combined with USAID, which might be combined as follows:

To state a claim that the government/agency/actor violated the First Amendment through coercion, a plaintiff must plausibly allege conduct that, viewed in context, does not define the limits of the government program and could be reasonably understood to convey a threat of adverse government/agency/official action in order to punish or suppress the plaintiff’s speech outside the contours of the program itself.

Considering novel actors and compounding programs would not render a “clear” line between program definition and impermissible leveraging. But, acknowledging the distinct circumstances of executive conditions, and focusing on these factors while drawing together USAID and Vullo, could bring clarity to a challenging analysis while tracking the distinctive constitutional and cross-cutting functional considerations presented by executive conditions.

B. Appropriations Clause and Spending Clause Cases

Litigants have invited courts to apply some or all of the Dole test for analyzing the constitutionality of conditional spending to executive conditions. Courts are divided so far, ranging from some applying the Dole test with little concern for the distinctiveness of executive conditions, to others rejecting it as wholly inapplicable. Relatedly, litigants have contended the executive broadly violates the Spending Clause and/or Appropriations Clause by using executive conditions to exercise power over funding.^[152] We suggest a middle ground: The Dole test provides guidance but is not straightforwardly adaptable to the executive-condition environment. Dole does not wholly account for the ways in which executive conditions are different from legislative conditions, including that its analysis relies on federalism values in conditional spending programs but not on the separation of powers concerns that executive conditions raise;^[153] that political safeguards of federalism applicable to the federal legislative process are a limited protection (at best) against purely executive action; and that judicial checks on executive conditions threaten the effective administration of cooperative federalism and other programs.

Therefore, we suggest the analysis may be guided by constitutional allocation of power—because Congress has the power of the purse—while recognizing how executive conditions may either further or thwart congressional choices to allocate or limit resources dedicated for particular programs. Executive actors have threatened state and private funding recipients with conditions using enforcement discretion; threats/promises of illegality; and threats/promises of discretionary action to obtain services, donations, and even funding for military action.^[154] These conditions implicate not only Congress’s enumerated powers and separation of powers but also federalism principles, because executive unilateralism that circumvents or thwarts the existing terms of cooperative federalism programs shifts the power balance between the federal government and states. This power balance is often navigated by federal and state actors in dynamic negotiations that preserve state authority and policy preferences within federal programs (consider the example of Medicaid demonstration waivers above), and executive conditions engage in a unilateralism that not only avoids the checks inherent in legislative and rulemaking processes but also threatens the vertical balance of power.^[155]

When this occurs—that is, the executive actor circumvents congressional funding decisions for an unsanctioned function through executive conditions—Youngstown’s framework for determining when the President exercises lawful implied authority may be instructive.^[156] To be clear, we are not wading into debates over the nature of presidential power; rather, the Youngstown opinions created a taxonomy that offers a useful guide for deciphering when executive conditions should be reviewed more searchingly.

We propose analyzing the lawfulness of executive conditions by assessing when such conditions usurp, contradict, or otherwise interfere with Congress’s power, which reflects three well-known concurring opinions in Youngstown. The Court held the President could not issue executive orders taking over steel mill operations to support military action in Korea because he did not have express power to seize businesses (Congress actually debated and denied that power to the President).^[157] The Frankfurter, Jackson, and Douglas concurrences explored how to determine whether the President may act if the Constitution or Congress has spoken (or not) to the issue. Justice Douglas focused on separation of powers, specifically other branches’ powers being usurped or hindered in performing their constitutional duties.^[158] Justice Jackson’s concurrence included the usurpation notion but offered a three-part framework on which courts have continued to rely:

When the President acts pursuant to an express or implied authorization of Congress, his authority is at its maximum, for it includes all that he possesses in his own right plus all that Congress can delegate. . . . When the President acts in absence of either a congressional grant or denial of authority, he can only rely upon his own independent powers, but there is a zone of twilight in which he and Congress may have concurrent authority, or in which its distribution is uncertain. . . . When the President takes measures incompatible with the expressed or implied will of Congress, his power is at its lowest ebb, for then he can rely only upon his own constitutional powers minus any constitutional powers of Congress over the matter. . . . [W]hat is at stake is the equilibrium established by our constitutional system.^[159]

This presidential power framework is useful here because of the particular, distinctive risks executive conditions pose, whether imposed by the President or another executive branch actor, when they are used to thwart or obtain resources allocated by Congress.

Only Congress has the power of the purse, and it sometimes enacts spending programs with broad contours that give executive actors power to create conditions to execute a program. This legislative allocation of power is consistent with the “maximum” articulated by Jackson’s concurrence. If Congress has not addressed the matter for which executive conditions are imposed, or if the executive creates conditions that are resource-dependent but not allocated by Congress, then the “zone of twilight” concept indicates courts may want to exercise some caution. The zone of twilight could result in either deferential or more searching judicial review, depending on the totality of a given case’s circumstances.

We agree with Price’s contribution to this volume^[160] that the balance is tipped by considering whether the condition induces states or private parties to provide resources to the executive in excess or contravention of those provided by Congress, which resounds in the spending and appropriations powers. If the executive conditions in question provide resources to achieve resource-dependent purposes that might have been appropriated, then the executive may be acting with power at its “lowest ebb”—that is, the executive is usurping congressional authority over governmental resources—and the executive condition should receive more searching scrutiny.

In statutory interpretation terminology, legislation appropriating funds for various executive purposes might be best understood as appropriating only the funds designated in legislation only for the purposes so designated. Reading appropriations legislation to impliedly prohibit the executive’s generation and use of resources through executive conditions would place such conditions within the “lowest ebb” category of congressional opposition, rather than the “zone of twilight” in which executive conditions that do not generate resources might fall. And such an expressio unius est exclusio alterius reading can be justified for appropriations legislation by the Appropriations Clause, the Miscellaneous Receipts Act, and appropriations principles like the “pick and stick” rule that requires the executive branch to limit funding for a purpose similar to that provided by the first appropriation tapped for that purpose, even if others might have been available.^[161]

This approach may address underlying concerns about (1) respecting rather than displacing congressional control over spending and (2) preventing leveraging from allowing the executive to accomplish resource-dependent goals without congressional approval. For example, C. Boyden Gray worried that bank settlement agreements during the Obama Administration posed a threat to congressional control;^[162] this resource-based consideration would address precisely that concern. Resource-dependence is just one example—but a key one—for considering how states or private parties might demonstrate that executive conditions should not receive judicial deference and may be impermissible. It also offers a clear line-drawing example because Congress exercises the power of the purse, which also can be a check on constitutionally allocated executive powers.

Finally, returning to the question of whether Dole can be imported to executive conditions, we think the answer depends on whether a court is tasked with reviewing executive conditions that act upon states specifically. The Dole Court’s concern for preserving federalism values is a meaningful principle for evaluating executive conditions that involve resources and act upon states, because such values may underscore the need for courts to be skeptical of the asserted executive power. Through Dole as amended by NFIB and other key spending power decisions, the Court effectively warned Congress and agencies tasked with implementing conditional spending programs that statutory conditions will be deemed binding for states when they are clear and unambiguous (under Pennhurst and Arlington), the conditions are germane, and states cannot be coerced into participation by the threat of too much money being taken away (NFIB’s coercion doctrine). The Court also warned Congress’s spending power influence can be deemed unconstitutional when leveraging occurs, though it is more likely to occur when private parties are subject to conditions on spending (Dole’s fourth prong and its application in Rust and USAID). So, we suggest the Dole test may be relevant for executive conditions when they are resource-dependent and act upon states. This offers a possible analytical approach that draws on existing doctrine, tracks with the Jackson concurrence’s three-prong analysis of executive power, and provides sensible line-drawing.

Conclusion

Statutory conditions and executive conditions are distinct. While executive conditions may provide flexibility in program administration, they may also evade political accountability, judicial review, and political safeguards of federalism while threatening separation of powers and individual liberty. Developing distinct doctrines for executive conditions that target abuses while avoiding undue interference in program administration is both theoretically desirable and practically feasible. Courts confronting First Amendment anti-leveraging challenges to executive conditions should give closer scrutiny to conditions articulated by executive branch actors not typically involved in program administration, as well as conditions bundled across discrete statutory programs. And courts confronting Spending Clause challenges to executive conditions should consider the presence of a condition that generates resources for the executive in excess of limits set by Congress as a factor favoring a finding of unconstitutionality.

1. * Edward R. Utley Professor of Health Law, Boston University School of Law and School of Public Health. ↑

2. ** Professor of Law, Emory University School of Law. Many thanks for invaluable feedback to participants in the Washington University Law Review symposium “Taxing, Spending, and the Constitution”; the Health Law, Policy, Bioethics, and Biotechnology Workshop at Harvard Law School; and the faculty workshops at Emory Law, University of Texas Law School, and Boston University School of Law. ↑

3. . See, e.g., Kousisis v. United States, 605 U.S. 114, 140 n.1 (2025) (Thomas, J., concurring) (“The Government often tries to use monetary incentives to advance political objectives.”). ↑

4. . Our focus is conditions on spending, though our analysis may apply to other kinds of regulatory conditions. We leave this question to future work. ↑

5. . E.g., Vanessa Burrows & Barbara Berney, Creating Equal Health Opportunity: How the Medical Civil Rights Movement and the Johnson Administration Desegregated U.S. Hospitals, 105 J. Am. Hist. 885 (2019) [https://doi.org/10.1093/jahist/jaz004]; Biden v. Missouri, 595 U.S. 87 (2022) (upholding executive condition that health care entities test or vaccinate their workers for coronavirus to be eligible for Medicare or Medicaid funding). ↑

6. . Cf. David A. Super, Rethinking Fiscal Federalism, 118 Harv. L. Rev. 2544 (2005). ↑

7. . Agency for Int’l Dev. v. All. for Open Soc’y Int’l, Inc., 570 U.S. 205, 206 (2013) (addressing constitutionality of condition in United States Leadership Against HIV/AIDS, Tuberculosis, and Malaria Act of 2003). ↑

8. . South Dakota v. Dole, 483 U.S. 203, 205 (1987) (assessing constitutionality of “23 U.S.C. § 158, which directs the Secretary of Transportation to withhold a percentage of federal highway funds otherwise allocable from States” that allow purchase of alcohol by persons under twenty-one (citation omitted)). ↑

9. . Selective Serv. Sys. v. Minn. Pub. Int. Rsch. Grp., 468 U.S. 841, 843 (1984) (rejecting claim that “Section 3 of the Military Selective Service Act” was an unconstitutional bill of attainder). ↑

10. . See generally Matthew B. Lawrence, Eloise Pasachoff & Zachary S. Price, Appropriations Presidentialism, 114 Geo. L.J. Online 1 (2025). ↑

11. . See Matthew B. Lawrence, Second-Class Administrative Law: Lincoln v. Vigil’s Puzzling Presumption of Unreviewability, 101 Wash. U. L. Rev. 1029 (2024) (discussing relative rarity of litigation about implementation of appropriations). ↑

12. . Exec. Order No. 14173, 90 Fed. Reg. 8633 (Jan. 21, 2025). ↑

13. . E.g., Wilmer Cutler Pickering Hale & Dorr LLP v. Exec. Off. of the President, 784 F. Supp. 3d 127 (D.D.C. 2025), modified, No. CV 25-917, 2025 WL 2105262 (D.D.C. June 26, 2025), appeal filed and consolidated sub nom. Perkins Coie LLP v. U.S. Dep’t of Just., No. 25-5241 (D.C. Cir. argued May 14, 2026); Jenner & Block LLP v. U.S. Dep’t of Just., 784 F. Supp. 3d 76, 97 (D.D.C. 2025), appeal filed and consolidated sub nom. Perkins Coie, No. 25-5241. ↑

14. . Conor Clarke, The Conservative Case for Leaving Harvard Alone, The Atlantic (Apr. 18, 2025), https://www.theatlantic.com/ideas/archive/2025/04/harvard-trump-irs-nonprofit/682507/ [https://
    perma.cc/J5QU-V6MY]. ↑

15. . See Governing for Impact, Challenging Federal Award Terminations (2025), https://governingforimpact.org/wp-content/uploads/2025/03/Challenging-Federal-Award-Terminations
    -2.pdf [https://perma.cc/4C6L-UEXD]. ↑

16. . Cf. Matthew B. Lawrence, Fiscal Waivers and State “Innovation” in Health Care, 62 Wm. & Mary L. Rev. 1477, 1530–41 (2021) (discussing questions about conditions on Medicaid waivers dating back decades). ↑

17. . Lawrence et al., supra note 8. ↑

18. . For example, in the decade plus after it was decided, the Supreme Court’s 2013 decision in USAID was cited about a dozen times each year, but the case was cited in forty-six published decisions in 2025 alone. Shepard’s Citing Decisions Analysis by Date, LEXIS, advance.lexis.com (search “Agency for Int’l Dev. v. All. for Open Soc’y Int’l, Inc., 570 U.S. 205 (2013)” in search bar; then click “Citing Decisions” and next to “Shepard’s Citing Decision Analysis” select “Date”). ↑

19. . See infra notes 62–63 and accompanying text (explaining that challenges to executive conditions are getting funneled to the Court of Federal Claims, which has limited jurisdiction). ↑

20. . See, e.g., South Dakota v. Dole, 483 U.S. 203 (1987); Nat’l Fed’n of Indep. Bus. v. Sebelius, 567 U.S. 519 (2012); All. for Open Soc’y Int’l, 570 U.S. 205; Rust v. Sullivan, 500 U.S. 173 (1991). ↑

21. . See infra Section I.C. ↑

22. . See infra Section I.C. ↑

23. . See infra Section I.C. ↑

24. . See infra Section I.C. (discussing Nat’l Rifle Ass’n of Am. v. Vullo, 602 U.S. 175 (2024)). ↑

25. . See infra Section I.C.; see generally Kate Stith, Congress’ Power of the Purse, 97 Yale L.J. 1343 (1988) (describing principles of appropriations control); Zachary S. Price, Funding Restrictions and Separation of Powers, 71 Vand. L. Rev. 357 (2018) (contrasting resource-dependent and resource-independent executive functions); Consumer Fin. Prot. Bureau v. Cmty. Fin. Servs. Ass’n of Am., Ltd., 601 U.S. 416, 427 (2024) (describing origins of the Appropriations Clause). ↑

26. . See, e.g., Douglas M. Spencer, Sanctuary Cities and the Power of the Purse: An Executive Dole Test, 106 Iowa L. Rev. 1209, 1232–35 (2021) (addressing how Dole test might be different for executive conditions). Solomon and Zaring focus on administrative procedure questions posed by what they call “transactional administration,” not constitutional questions, but their examples and emphasis on transparency are helpful. Steven Davidoff Solomon & David Zaring, Transactional Administration, 106 Geo. L.J. 1097 (2018); see also Steven M. Davidoff & David Zaring, Regulation by Deal: The Government’s Response to the Financial Crisis, 61 Admin. L. Rev. 463, 468 (2009). Finally, one of us previously explored the greater susceptibility to subordination of executive conditions. See Matthew B. Lawrence, Subordination and Separation of Powers, 131 Yale L.J. 78, 90 (2021). Two additional articles address the interaction of the Pennhurst doctrine (regarding interpretation of conditions imposed on states) and the Chevron doctrine; although Chevron has been invalidated, those articles’ discussions of the relationship between conditions doctrine and executive power remain invaluable. See David Freeman Engstrom, Drawing Lines Between Chevron and Pennhurst: A Functional Analysis of the Spending Power, Federalism, and the Administrative State, 82 Tex. L. Rev. 1197, 1216–22 (2004); Peter J. Smith, Essay, Pennhurst, Chevron, and the Spending Power, 110 Yale L.J. 1187, 1233–36 (2001). Many fine articles address doctrines we discuss but, while invaluable, do not specifically study distinctions between conditions imposed by statute (directly or via delegation) and conditions imposed through executive action. ↑

27. . “Illegal behavior” should not be a source of power for the executive, but the executive has the first mover advantage, and its illegal actions can cause irreparable harm regardless of whether courts might later intervene. See Terry M. Moe & William G. Howell, Unilateral Action and Presidential Power: A Theory, 29 Presidential Stud. Q. 850, 867 (1999) (discussing “the president’s powers of unilateral action”). In practice, this means that the executive may threaten (or promise) illegal action as a way to coerce a person or institution to agree to terms that might themselves later come to be practically or legally binding. The terms of any deal resulting from such a threat are “executive conditions,” as we develop herein. ↑

28. . See infra notes 88–91 and accompanying text (discussing existing doctrinal tests that overlap with distinction between statutory conditions and executive conditions drawn here). ↑

29. . For an example of a statutory condition, consider an HHS regulation now going through notice and comment rulemaking that would amend the Code of Federal Regulations to require state Medicaid plans to exclude coverage for certain treatments for gender dysphoria. See Medicaid Program; Prohibition on Federal Medicaid and Children’s Health Insurance Program Funding for Sex-Rejecting Procedures Furnished to Children, 90 Fed. Reg. 59441 (proposed Dec. 19, 2025) (to be codified at 42 C.F.R. pts. 441, 457). HHS purports to be exercising power to impose binding conditions delegated by statute in the proposed regulation, so if it is finalized, we would consider this condition to be “statutory” even though it was originally articulated by an executive branch actor. ↑

30. . See Printz v. United States, 521 U.S. 898, 922–23 (1997) (pointing to state sovereignty, federal effectiveness, accountability, and liberty as underlying values). ↑

31. . See Lawrence, supra note 9. ↑

32. . See Lawrence et al., supra note 8, at 22–29 (arguing that judicial interference in spending cases risks displacing Congress’s oversight role). ↑

33. . See Price, supra note 23, at 365 n.12; infra Section III.B. (identifying concurring opinions in Youngstown Sheet & Tube Co. v. Sawyer as this framework’s inspiration). ↑

34. . U.S. Const. pmbl. ↑

35. . See, e.g., Cash for Clunkers, https://www.cashforclunkers.org/ [https://perma.cc/H96Z-XASG]. ↑

36. . Simkins v. Moses H. Cone Mem’l Hosp., 323 F.2d 959 (4th Cir. 1963), cert. denied, 376 U.S. 938 (1964). ↑

37. . Joshua Herb, Health Policy in Action – A Brief History of Hospital Desegregation in the United States, Ass’n for Acad. Surgery (Feb. 27, 2025), https://www.aasurg.org/health-policy-in-action-a-brief-history-of-hospital-desegregation-in-the-united-states/ [https://perma.cc/P3K5-SGVC]. ↑

38. . Burrows & Berney, supra note 3. ↑

39. . See generally Price, supra note 23. ↑

40. . 595 U.S. 87, 93 (2022). ↑

41. . Id. ↑

42. . See Abbe R. Gluck & Nicole Huberfeld, What Is Federalism in Health Care For?, 70 Stan. L. Rev. 1689 (2018); Lawrence, supra note 14; Samuel R. Bagenstos, Federalism by Waiver After the Health Care Case, in The Health Care Case: The Supreme Court’s Decision and Its Implications 227 (Nathaniel Persily, Gillian E. Metzger & Trevor W. Morrison eds., 2013); David J. Barron & Todd D. Rakoff, In Defense of Big Waiver, 113 Colum. L. Rev. 265 (2013); 42 U.S.C. § 1315 (Social Security Act waiver authority (Medicaid)); Patient Protection and Affordable Care Act, Pub. L. No. 111-148, § 1332, 124 Stat. 120, 203–08 (2010) (state innovation waivers for commercial health insurance in the ACA). ↑

43. . Fullilove v. Klutznick, 448 U.S. 448, 474 (1980) (“Congress [may] . . . further broad policy objectives by conditioning receipt of federal moneys upon compliance by the recipient with federal statutory and administrative directives.”). ↑

44. . Nicole Huberfeld, Spending Programs and the New Roberts Court, 24 Yale J. Health Pol’y, L. & Ethics 1, 40 (2026). ↑

45. . FCC v. Consumers’ Rsch., 606 U.S. 656, 673 (2025). ↑

46. . E.g., ACORN v. United States, 618 F.3d 125 (2d Cir. 2010) (bills of attainder). ↑

47. . Agency for Int’l Dev. v. All. for Open Soc’y Int’l, Inc., 570 U.S. 205 (2013). ↑

48. . Id. ↑

49. . Id. at 214–15 (emphasis added). ↑

50. . See, e.g., Sam Bagenstos, Employing the Anti-Leveraging Test to Effectively Protect Grantee Speech, Knight First Amend. Inst.: Fed. Funding & the First Amend. (May 21, 2025), https://
    knightcolumbia.org/blog/employing-the-anti-leveraging-test-to-effectively-protect-grantee-speech [htt
    ps://perma.cc/LB67-NDHR] (discussing USAID case); Michael C. Dorf, Conditional Funding Can Raise Difficult Legal Questions. Trump’s Freezes Don’t, Knight First Amend. Inst.: Fed. Funding & the First Amend. (May 13, 2025), https://knightcolumbia.org/blog/conditional-funding-can-raise
    -difficult-legal-questions-trumps-freezes-dont [https://perma.cc/ZY8Z-NGBS]. ↑

51. . Daniel S. Cohen, A Gun to Whose Head? Federalism, Localism, and the Spending Clause, 123 Dick. L. Rev. 421, 436–37 (2019). ↑

52. . South Dakota v. Dole, 483 U.S. 203 (1987). ↑

53. . Id. at 207–11. In Arlington Central School District Board of Education v. Murphy, the Court amended the second Dole prong slightly to clear and unambiguous conditions, further promoting state-protective federalism values. 548 U.S. 291, 296 (2006). ↑

54. . Nicole Huberfeld, Clear Notice for Conditions on Spending, Unclear Implications for States in Federal Healthcare Programs, 86 N.C. L. Rev. 441 (2008) (describing Pennhurst and explaining the Arlington modification to Dole). ↑

55. . Pennhurst State Sch. & Hosp. v. Halderman, 451 U.S. 1, 17–18 (1981); see also Samuel R. Bagenstos, Spending Clause Litigation in the Roberts Court, 58 Duke L.J. 345 (2008) (discussing implementation of constitutional checks on spending power through interpretive doctrines). ↑

56. . 567 U.S. 519, 588 (2012) (plurality opinion). ↑

57. . See Huberfeld, supra note 42 (mapping the high volume of state-led litigation attempting to discern contours and outer limits of the coercion doctrine). ↑

58. . Nat’l Endowment for the Arts v. Finley, 524 U.S. 569 (1998). ↑

59. . Agency for Int’l Dev. v. All. for Open Soc’y Int’l, Inc., 570 U.S. 205, 215 (2013) (“The line is hardly clear.”). ↑

60. . NFIB, 567 U.S. at 585 (plurality opinion) (“It is enough for today that wherever that line may be, this statute is surely beyond it. Congress may not simply ‘conscript state [agencies] into the national bureaucratic army . . . .’”). ↑

61. . Lawrence et al., supra note 8. ↑

62. . At the authors’ last count there were already more than 338 cases in which a lower court had reached a decision on a request for injunctive relief, with lower courts granting such relief in 226 of those cases and denying them in 112. See Litigation Tracker: Legal Challenges to Trump Administration Actions, Just Sec. (Mar. 12, 2026), https://www.justsecurity.org/107087/tracker-litigation-legal-challenges-trump-administration/ [https://perma.cc/WB5M-R2FM]. ↑

63. . E.g., AIDS Vaccine Advoc. Coal. v. U.S. Dep’t of State, 803 F. Supp. 3d 164, 184 (D.D.C. 2025) (“Defendants’ reasons for not developing an argument here over the numerous months and opportunities given may be many, including that, even having changed their position, there is not a plausible interpretation of the statutes that would justify the billions of dollars they plan to withhold.”). ↑

64. . Cf. Maureen E. Brady & Richard Re, The Docket That Cannot Be Named, SCOTUSblog: Controlling Opinions (May 4, 2026), https://www.scotusblog.com/2026/05/the-docket-that-cannot-be-named/ [https://perma.cc/7D3W-ZBAM] (describing naming confusion surrounding processes other than merits docket by which Supreme Court issues orders). ↑

65. . The APA’s legal, procedural, and substantive checks on agency action are at least as protective as underlying constitutional safeguards, see Vt. Yankee Nuclear Power Corp. v. Nat. Res. Def. Council, 435 U.S. 519 (1978) (APA requirements ordinarily satisfy the Constitution), so ordinary disputes about agency action center around questions of statutory authority, procedure, or rationality (arbitrary and capricious review under the Administrative Procedure Act). But as interpreted by the Supreme Court in Department of Education v. California, 604 U.S. 650 (2025), the Tucker Act channels many disputes about federal spending out of APA actions in district court and into contract disputes in the Court of Federal Claims. The Court of Federal Claims cannot grant preliminary relief, lacks many APA protections, and has very limited ability to consider constitutional questions. The Court’s opinion in California focused on the connection between a lawsuit and underlying grant or contract provisions, so plaintiffs have understandably shied away from claims focused on such provisions and toward claims focused on larger constitutional violations in bringing cases in district court. For a prominent issue brief highlighting this interaction in encouraging constitutional claims, see Governing for Impact, Challenging Funding Conditions Imposed by the Executive Branch 17 (2025), https://governingforimpact.org/wp-content/uploads/2025/09/Funding-Conditions-Issue-Brief-final.pdf [https://perma.cc/CN97-MATX%5D. ↑

66. . Jenner & Block LLP v. U.S. Dep’t of Just., 784 F. Supp. 3d 76, 109–10 (D.D.C. 2025) (applying legislative condition caselaw to executive condition, referring to case addressing constitutionality of statutory condition as more broadly applying when “the government” seeks to leverage its power to regulate speech), appeal filed and consolidated sub nom. Perkins Coie LLP v. U.S. Dep’t of Just., No. 25-5241 (D.C. Cir. argued May 14, 2026); City of Fresno v. Turner, No. 25-cv-7070, 2025 WL 2469330, at *5–6 (N.D. Cal. Aug. 27, 2025) (applying Pennhurst and NFIB), appeal filed, No. 25-7378 (9th Cir. Nov. 24, 2025); Am. Ass’n of Univ. Professors v. Trump, 815 F. Supp. 3d 907 (N.D. Cal. 2025) (invalidating threat to cut off U.C. system’s grant funding unless system agreed to proffered conditions). ↑

67. . Exec. Order No. 14250, 90 Fed. Reg. 14549 (Mar. 27, 2025). ↑

68. . Complaint for Declaratory and Injunctive Relief at 54–55, Wilmer Cutler Pickering Hale & Dorr LLP v. Exec. Off. of the President, 784 F. Supp. 3d 127 (D.D.C. 2025) (No. 25-917), modified, No. CV 25-917, 2025 WL 2105262 (D.D.C. June 26, 2025), appeal filed and consolidated sub nom. Perkins Coie, No. 25-5241. ↑

69. . “The Court finds that the firm has failed to state a Spending Clause claim.” Wilmer-Hale, 784 F. Supp. 3d at 162 (rejecting Spending Clause challenge to contractual condition imposed by executive order because “[t]he Spending Clause is implicated when Congress imposes a spending or funding condition” but “[t]he President issued the Order”). ↑

70. . Bd. of Educ. for the Silver Consol. Schs. v. McMahon, 791 F. Supp. 3d 1272, 1288 (D.N.M. 2025) (“[W]ithout alleging congressional action, Plaintiff ‘cannot state a claim under the Spending Clause.’” (quoting Wilmer-Hale, 784 F. Supp. 3d at 162)). ↑

71. . President & Fellows of Harvard Coll. v. U.S. Dep’t of Health & Hum. Servs., 798 F. Supp. 3d 77 (D. Mass. 2025), appeal filed, No. 25-2231 (1st Cir. Dec. 31, 2025). ↑

72. . Id. at 123. ↑

73. . Id. at 124 (referring to amicus brief of Professor Aaron Caplan); Memorandum in Support of Defendants’ Cross-Motion for Summary Judgment and Opposition to Plaintiff’s Motion for Summary Judgment at 38, President & Fellows of Harvard Coll., 798 F. Supp. 3d 77 (No. 1:25-cv-11048) (“It is an open question whether the unconstitutional conditions doctrine can even be properly applied to settlement negotiations at all.”). ↑

74. . Perkins Coie LLP v. U.S. Dep’t of Just., 783 F. Supp. 3d 105, 173 n.36 (D.D.C. 2025), appeal filed, No. 25-5241 (D.C. Cir. argued May 14, 2026). ↑

75. . Id. at 174 n.36. ↑

76. . Id. ↑

77. . See, e.g., Washington v. U.S. Dep’t of Com., 812 F. Supp. 3d 1169, 1177 (W.D. Wash. 2025) (describing provision, citing 2 C.F.R. § 200.340(a)(4)). ↑

78. . See Complaint for Declaratory and Injunctive Relief at 68–69, New Jersey v. U.S. Off. of Mgmt. & Budget, No. 1:25-cv-11816 (D. Mass. filed June 24, 2025) (“[N]o statute authorizes Defendants to terminate grant awards based on newly identified ‘program goals or agency priorities’ unknown to Plaintiffs at the time of the award. If there were such a statute, it would violate the Spending Clause because it would not set forth unambiguously the terms upon which grants may be terminated.”); cf. City of Fresno v. Turner, No. 25-cv-7070, 2025 WL 2721390, at *14 (N.D. Cal. Sept. 23, 2025) (reasoning, in voiding grant conditions imposed by agency on vagueness grounds, that the conditions “are not ‘merely add[itive]’” to those imposed by the underlying statute; “they digress from them” (alteration in original)), appeal filed, No. 25-7378 (9th Cir. Nov. 24, 2025). ↑

79. . Bridget A. Fahey, Federalism by Contract, 129 Yale L.J. 2326 (2020). ↑

80. . Nat’l Rifle Ass’n of Am. v. Vullo, 602 U.S. 175 (2024). ↑

81. . Id. at 191. ↑

82. . For cases declining to apply Vullo in the context of a challenge to termination of funding for failure to abide by executive conditions, see Nat’l Urb. League v. Trump, 783 F. Supp. 3d 61, 103 (D.D.C. 2025) (concluding in challenge to non-DEI certification requirement that “this is not a case of the government ‘relying on the threat of invoking legal sanctions’ to ‘achieve the suppression of disfavored speech’” because “the Certification Provision addresses only DEI programs that violate federal antidiscrimination law” (emphasis added) (quoting Vullo, 602 U.S. at 189)); Chi. Women in Trades v. Trump, 778 F. Supp. 3d 959, 986 (N.D. Ill. 2025) (“[E]ven if ‘coercion’ encompasses more than investigation and prosecution, it likely does not encompass the termination of government funding.”); see also Nat’l Urb. League, 783 F. Supp. 3d at 104 (following Chi. Women in Trades). ↑

83. . For a case incorporating aspects of Vullo into the resolution of a challenge to termination of funding for failure to abide by executive conditions, see Jenner & Block LLP v. U.S. Dep’t of Just., 784 F. Supp. 3d 76, 97 (D.D.C. 2025) (concluding in challenge to law firm order that “[i]n this context, retaliation amounts to something akin to the impermissible ‘scheme of informal censorship’ that arises when government actors use the ‘threat of invoking legal sanctions and other means of coercion to achieve the suppression of disfavored speech’” (quoting Vullo, 602 U.S. at 188–89)), appeal filed and consolidated sub nom. Perkins Coie LLP v. U.S. Dep’t of Just., No. 25-5241 (D.C. Cir. argued May 14, 2026). ↑

84. . See supra note 24 (collecting sources). ↑

85. . Cf. John F. Manning, Separation of Powers as Ordinary Interpretation, 124 Harv. L. Rev. 1939, 1945 (2011) (“[T]he idea of separation of powers, properly understood, reflects many particular decisions about how to allocate and condition the exercise of federal power.”). ↑

86. . Trump v. United States, 603 U.S. 593, 607 (2024) (“No matter the context, the President’s authority to act necessarily ‘stem[s] either from an act of Congress or from the Constitution itself.’” (quoting Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 585 (1952)). ↑

87. . For example, HHS’s promulgation of the requirement that health care providers test or vaccinate for COVID was such a rule: The interim final rule was published with a preamble and explanation in the Federal Register and published in the Code of Federal Regulations. Medicare and Medicaid Programs; Omnibus COVID–19 Health Care Staff Vaccination, 86 Fed. Reg. 61555 (Nov. 5, 2021) (to be codified at 42 C.F.R. pts. 416, 418, 441, 460, 482, 483, 484, 485, 486, 491, 494). This created a legally binding condition on Medicare and Medicaid funding through exercise of the power delegated by Congress to HHS to promulgate regulations elaborating statutory “health and safety” conditions. See, e.g., 42 U.S.C. § 1395x(e)(9); 42 U.S.C. § 1396d; see also Biden v. Missouri, 595 U.S. 87 (2022). ↑

88. . See SEC v. Chenery Corp., 332 U.S. 194 (1947). ↑

89. . On the threat of illegal action as a source of presidential power, see supra note 25. ↑

90. . See Cass R. Sunstein, Essay, “Practically Binding”: General Policy Statements and Notice-and-Comment Rulemaking, 68 Admin. L. Rev. 491, 495–99 (2016). ↑

91. . United States v. Mead Corp., 533 U.S. 218 (2001). ↑

92. . Eloise Pasachoff, Executive Branch Control of Federal Grants: Policy, Pork, and Punishment, 83 Ohio St. L.J. 1113 (2022). ↑

93. . Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 635–38 (1952) (Jackson, J., concurring). ↑

94. . See generally Zachary S. Price, Enforcement Discretion and Executive Duty, 67 Vand. L. Rev. 671 (2014). ↑

95. . Simon Brewer, Note, The Attorney General’s Settlement Authority, 130 Yale L.J. 174, 184 (2020) (“No statute explicitly authorizes DOJ to enter into settlements that make policy or otherwise constrain the future exercise of executive-branch discretion or policymaking.”). ↑

96. . Cf. Trump v. United States, 603 U.S. 593, 608 (2024) (“If the President claims authority to act but in fact exercises mere ‘individual will’ and ‘authority without law,’ the courts may say so.”). ↑

97. . As described by Professors Starr and Lakier, certain threats the Trump Administration has made to universities to cut off funding on alleged Title VI grounds are in fact threats to engage in action that courts would likely hold to be unlawful because they are based on unprecedented and incorrect interpretations of Title VI and the Equal Protection Clause. See Sonja A. Starr & Genevieve Lakier, The Constitution and the War on DEI 7 (Univ. Chi. L. Sch., Working Paper No. 25-35, 2025) (“Many of those targeted by the war on DEI have chosen to abandon DEI-related policies and expression, notwithstanding the weakness of the administration’s legal theories. That outcome—preemptive compliance—is obviously what the administration is attempting to achieve.” (footnote omitted)). Cf. David Pozen, Regulation by Deal Comes to Higher Ed, Balkinization (July 23, 2025), https://balkin.blogspot.com/2025/07/regulation-by-deal-comes-to-higher-ed.html [https://perma.cc/D7
    94-KELQ]. ↑

98. . Cf. Moe & Howell, supra note 25. ↑

99. . See, e.g., Lawrence et al., supra note 8 (describing barriers to judicial review of unilateral executive funding actions). ↑

100. . 42 U.S.C. § 1315. ↑

101. . See Lawrence, supra note 14, at 1494–96. The July 2025 budget reconciliation bill formalized budget neutrality as a Section 1115 Waiver statutory requirement. One Big Beautiful Bill Act of 2025, Pub. L. No. 119-21, § 71118, 139 Stat. 72, 306. See Nicole Huberfeld & Matthew Lawrence, Why It Matters: HR1’s Change to Medicaid Waiver Budget Neutrality Rules, Petrie-Flom Ctr.: Bill of Health (Feb. 19, 2026), https://petrieflom.law.harvard.edu/2026/02/19/why-it-matters-hr1s-change-to-medicaid-waiver-budget-neutrality-rules/ [https://perma.cc/E7KH-Z45H%5D. ↑

102. . The Government Accountability Office has issued several reports regarding the lack of standards for budget neutrality in Section 1115 Waiver applications and implementation. See, e.g., U.S. Gov’t Accountability Off., GAO-25-107445, Medicaid Demonstrations: Action Needed to Address New Cost Concerns (2025), https://www.gao.gov/products/gao-25-107445 [https://
     perma.cc/GUR7-R4TW]. ↑

103. . See Memorandum from Deputy Att’y Gen. James M. Cole, U.S. Dep’t of Just., Guidance Regarding Marijuana Enforcement (Aug. 29, 2013), https://www.justice.gov/iso/opa/resources/3052
     013829132756857467.pdf [https://perma.cc/DV2E-SFAM]. ↑

104. . See Manning, supra note 83; Lawrence, supra note 24. ↑

105. . Spencer, supra note 24, at 1215 (“Congressional delegation should not be a loophole in the Dole doctrine.”). Concern about this loophole is one reason that courts and scholars have cited for routine application of the First Amendment to executive action despite the fact that the Amendment’s text specifically refers only to lawmaking by Congress. See Daniel J. Hemel, Executive Action and the First Amendment’s First Word, 40 Pepp. L. Rev. 601 (2013). ↑

106. . Cf. Moody v. NetChoice, 603 U.S. 707 (2024) (recent Supreme Court case emphasizing utility of “as applied” challenges). ↑

107. . Cf. Lawrence et al., supra note 8 (arguing there is no substitute for congressional action in addressing appropriations presidentialism). ↑

108. . See Packers and Stockyards Act of 1921, 7 U.S.C. §§ 181–229(b). ↑

109. . Cf. Off. of Inspector Gen., U.S. Dep’t of Health & Hum. Servs., OEI-04-18-00450, Some Nursing Homes’ Reported Staffing Levels in 2018 Raise Concerns: Consumer Transparency Could Be Increased (2020) (describing CMS practice of downgrading nursing home star ratings for insufficient staffing). ↑

110. . Price, supra note 92, at 675 (finding enforcement discretion “create[s] a safety valve that protects citizens from overzealous enforcement of general prohibitions” (footnote omitted)). ↑

111. . Cf. David A. Super, Against Flexibility, 96 Cornell L. Rev. 1375 (2011) (describing benefits of flexibility while arguing that these benefits are overstated); Aaron L. Nielson, How Agencies Choose Whether to Enforce the Law: A Preliminary Investigation, 93 Notre Dame L. Rev. 1517 (2018). ↑

112. . See, e.g., Off. of Inspector Gen., U.S. Dep’t of Health & Hum. Servs., Special Fraud Alert: Speaker Programs (2020). ↑

113. . See John Blevins, Retaliation by Raised Eyebrow, 60 Ga. L. Rev. (forthcoming 2026), https:
     //papers.ssrn.com/sol3/papers.cfm?abstract_id=5503698 [https://perma.cc/LG95-HK5L] (manuscript at 7–19) (describing history of informal regulatory approaches). ↑

114. . See Melissa Quinn, Law Firm Skadden Cuts $100 Million Pro Bono Deal with Trump to Avoid Executive Order, CBS News (Mar. 28, 2025), https://www.cbsnews.com/news/law-firm-skadden-cuts-100-million-pro-bono-deal-trump-avoid-executive-order/ [https://perma.cc/72UC-KGC4%5D. ↑

115. . Consumer Fin. Prot. Bureau v. Cmty. Fin. Servs. Ass’n of Am., Ltd., 601 U.S. 416, 431 (2024); see also id. at 463 (Alito, J., dissenting) (“[T]he Appropriations Clause demands legislative control over the source and disposition of the money used to finance Government operations and projects.”); see also U.S. Dep’t of the Navy v. Fed. Lab. Rels. Auth., 665 F.3d 1339, 1347 (D.C. Cir. 2012) (explaining without congressional appropriations power, “the executive would possess an unbounded power over the public purse of the nation[] and might apply all its monied resources at his pleasure” (quoting 3 Joseph Story, Commentaries on the Constitution of the United States § 1342, at 213–14 (1833))). ↑

116. . For sources pointing toward such an argument, see infra notes 152–58. ↑

117. . See Cary Coglianese, Gabriel Scheffler & Daniel E. Walters, Unrules, 73 Stan. L. Rev. 885, 952–59 (2021) (discussing barriers to litigation and judicial review for challenges to agency carveouts and dispensations). ↑

118. . City & Cnty. of San Francisco v. Trump, 897 F.3d 1225, 1232 (9th Cir. 2018) (“Aside from the power of veto, the President is without authority to thwart congressional will by canceling appropriations passed by Congress.”); Cnty. of Santa Clara v. Trump, 250 F. Supp. 3d 497, 530–31 (N.D. Cal. 2017); Nat’l Council of Nonprofits v. Off. of Mgmt. & Budget, 763 F. Supp. 3d 36, 56 (D.D.C. 2025) (rejecting executive “attempt[] to wrest the power of the purse away from the only branch of government entitled to wield it”), appeal filed, No. 25-5148 (D.C. Cir. argued Feb. 6, 2026); New York v. Trump, 764 F. Supp. 3d 46, 51 (D.R.I. 2025) (“The Executive Branch has a duty to align federal spending and action with the will of the people as expressed through congressional appropriations, not through ‘Presidential priorities.’” (emphasis omitted)), aff’d in part and vacated in part, 171 F.4th 1 (1st Cir. 2026); PFLAG, Inc. v. Trump, 766 F. Supp. 3d 535, 564–65 (D. Md. 2025); Washington v. Trump, 768 F. Supp. 3d 1239, 1262 (W.D. Wash. 2025) (concluding because executive did not identify any statute arguably granting authority to condition funding on institutions’ non-provision of gender-affirming care, “[t]he President’s power is . . . ‘at its lowest ebb’” (quoting Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 637 (1952) (Jackson, J., concurring))), appeal filed, No. 25-1922 (9th Cir. argued Mar. 5, 2026); see Washington, 768 F. Supp. 3d at 1262 (“President Trump’s Executive Orders purport to do something not even Congress may do: ‘surprise[] states with post acceptance . . . conditions’ on federal funds and ‘impose conditions on federal grants that are unrelated to the federal interest in particular national projects or programs.’” (alteration in original) (citing City of Los Angeles v. Barr, 929 F.3d 1163, 1175 (9th Cir. 2019))). ↑

119. . Nat’l Fed’n of Indep. Bus. v. Sebelius, 567 U.S. 519, 536 (2012); see also id. at 631 (Ginsburg, J., concurring in part, concurring in the judgment in part, and dissenting in part). ↑

120. . Cf. Daryl J. Levinson, Rights Essentialism and Remedial Equilibration, 99 Colum. L. Rev. 857 (1999). ↑

121. . A secret condition that is never disclosed cannot be challenged by third parties whom it might impact and will be disclosed by the target only if they overcome barriers, such as the fear of retaliation and the cost of litigation, to bringing suit. Cf. Nicholas R. Parrillo, Administrative Law as a Choice of Business Strategy: Comparing the Industries Who Have Routinely Sued Their Regulators with the Industries Who Rarely Have, 93 Geo. Wash. L. Rev. 1031 (2025). The fact that executive conditions can be mere suggestions (or in APA terms, guidance) reduces the likelihood that affected parties will have standing and a cause of action. ↑

122. . Richard A. Epstein, Government by Waiver, 7 Nat’l Affs. 39, 54 (2011) (“The fate of our rights and liberties is left to the wisdom and discretion of individuals; we are therefore governed by men, not by laws. It was this exact circumstance that our system of government was designed to avoid . . . .”); ACUS Recommendation 2017-7, Regulatory Waivers and Exemptions, 82 Fed. Reg. 61742 (Dec. 29, 2017) (notice) (“[W]hen an agency decides to waive legal requirements for some but not all regulated parties, the decision to grant a waiver or exemption may create the appearance—or perhaps even reality—of irregularity, bias, or unfairness.”). ↑

123. . Complaint for Injunctive and Declaratory Relief, Scott v. U.S. Dep’t of Health & Hum. Servs., No. 3:15-cv-195 (N.D. Fla. dismissed Apr. 28, 2015). ↑

124. . Lawrence, supra note 14, at 1513. ↑

125. . Id. at 1534–35; Defendants’ Memorandum in Opposition to Plaintiffs’ Motion for Preliminary Injunction, Scott, No. 3:15-cv-195. ↑

126. . Lawrence, supra note 24, at 96–97. By subordination in this separation of powers context, we mean Madison’s goal of structuring the Constitution “to guard one part of the society against the injustice of the other part.” The Federalist No. 51, at 323–24 (James Madison) (Clinton Rossiter ed., 1961) (“Different interests necessarily exist in different classes of citizens. If a majority be united by a common interest, the rights of the minority will be insecure . . . In a society under the forms of which the stronger faction can readily unite and oppress the weaker, anarchy may as truly be said to reign as in a state of nature . . . .”). For expositions of Madison’s view of “justice” as “the protection of ‘minority’ groups against systematic ‘oppression’ or ‘tyrannization’ by more powerful groups acting through the political process and the government,” see James S. Liebman & Brandon L. Garrett, Madisonian Equal Protection, 104 Colum. L. Rev. 837, 840 n.3, 862 (2004); see also Akhil Reed Amar, The Bill of Rights and the Fourteenth Amendment, 101 Yale L.J. 1193, 1215 (1992) (noting that insofar as Madison viewed “liberty” as “liberty against popular majorities” he “was a man ahead of his time”). ↑

127. . The Federalist No. 51, supra note 123, at 325 (James Madison). ↑

128. . See, e.g., Robert A. Mikos, Medical Marijuana and the Political Safeguards of Federalism, 89 Denv. U. L. Rev. 997 (2012) (describing the theory and applying it to medical marijuana legalization). ↑

129. . Clinton v. City of New York, 524 U.S. 417, 451 (1998) (Kennedy, J., concurring); see also Bond v. United States, 572 U.S. 844 (2014). ↑

130. . See Keith E. Hamm, Patterns of Influence Among Committees, Agencies, and Interest Groups, 8 Legis. Stud. Q. 379, 396 (1983) (summarizing literature). ↑

131. . See Jon D. Michaels, An Enduring, Evolving Separation of Powers, 115 Colum. L. Rev. 515, 548–49 (2015). ↑

132. . Cf. Daphna Renan, Pooling Powers, 115 Colum. L. Rev. 211 (2015) (describing how multiple agencies may pool power to achieve purposes not possible for any one agency). ↑

133. . Price, supra note 23. ↑

134. . See Agency for Int’l Dev. v. All. for Open Soc’y Int’l, Inc., 570 U.S. 205, 215 (2013) (“The line is hardly clear . . . .”); Nat’l Fed’n of Indep. Bus. v. Sebelius, 567 U.S. 519, 585 (2012) (plurality opinion) (“It is enough for today that wherever that line may be, this statute is surely beyond it.”). ↑

135. . Jenner & Block LLP v. U.S. Dep’t of Just., 784 F. Supp. 3d 76, 88–91 (D.D.C. 2025), appeal filed and consolidated sub nom. Perkins Coie LLP v. U.S. Dep’t of Just., No. 25-5241 (D.C. Cir. argued May 14, 2026). ↑

136. . USAID, 570 U.S. at 214–15 (emphasis added). ↑

137. . Id. ↑

138. . Id. ↑

139. . This reflects the inevitable specialization and relationship formation that occurs even among political leadership who are responsible for overseeing implementation of a particular program. Cf. Anya Bernstein & Cristina Rodríguez, The Accountable Bureaucrat, 132 Yale L.J. 1600, 1628 (2023) (“[P]olitical accountability comes not only from the presence and influence of the political appointee, but also from the relationships between politicals and careers . . . .”); Gonzales v. Oregon, 546 U.S. 243, 253–54 (2006) (expressing skepticism because interpretive rule regarding applicability of the Controlled Substances Act to assisted suicide drugs was articulated by Attorney General rather than the Administrator of the Drug Enforcement Administration, who ordinarily oversees CSA implementation). ↑

140. . While we hesitate to use the term “major conditions doctrine,” in this way the above approach is broadly analogous to the skepticism for novel agency actions—and, in the view of some commenters, focus on who drives agency actions in assessing their legality—of the major questions doctrine. See Jodi L. Short & Jed H. Shugerman, Major Questions About Presidentialism: Untangling the “Chain of Dependence” Across Administrative Law, 65 B.C. L. Rev. 511 (2024) (noting that major questions cases decided by Roberts Court have involved agency actions that were presidentially directed). ↑

141. . See Universal Health Servs., Inc. v. United States ex rel. Escobar, 579 U.S. 176 (2016). ↑

142. . Cf. Jenner & Block LLP v. U.S. Dep’t of Just., 784 F. Supp. 3d 76, 102 (D.D.C. 2025) (“Rather than ensuring that national secrets remain with those who will keep them, Section 2’s process ‘seek[s] to leverage’ the Executive’s control over security clearances as a way to change speech.” (emphasis added) (citing Agency for Int’l Dev. v. All. for Open Soc’y Int’l, Inc., 570 U.S. 205 (2013))), appeal filed and consolidated sub nom. Perkins Coie LLP v. U.S. Dep’t of Just., No. 25-5241 (D.C. Cir. argued May 14, 2026). ↑

143. . R.I. Coal. Against Domestic Violence v. Kennedy, 807 F. Supp. 3d 94, 95 (D.R.I.), modified, 812 F. Supp. 3d 180 (D.R.I. 2025). ↑

144. . Id. at 104–05. ↑

145. . Id. ↑

146. . See Renan, supra note 130. ↑

147. . See U.S. Const. art. I, § 9, cl. 3; art. I, § 10, cl. 1. ↑

148. . Nat’l Fed’n of Indep. Bus. v. Sebelius, 567 U.S. 519, 582 (2012) (plurality opinion). ↑

149. . Nicole Huberfeld, Elizabeth Weeks Leonard & Kevin Outterson, Plunging into Endless Difficulties: Medicaid and Coercion in National Federation of Independent Business v. Sebelius, 93 B.U. L. Rev. 1, 73 (2013). ↑

150. . Nat’l Rifle Ass’n of Am. v. Vullo, 602 U.S. 175, 189 (2024) (“[T]he First Amendment prohibits government officials from relying on the ‘threat of invoking legal sanctions and other means of coercion . . . to achieve the suppression’ of disfavored speech.” (quoting Bantam Books, Inc. v. Sullivan, 372 U.S. 58, 67 (1963))). ↑

151. . Id. at 191. Scholarship around “jawboning” focuses on use of executive power to coerce a person or entity to censor or regulate the speech of a third party. See Derek E. Bambauer, Against Jawboning, 100 Minn. L. Rev. 51, 57 (2015) (defining jawboning as “a specific type of informal pressure by a government actor on a private entity: one that operates at the limit of, or outside, that actor’s authority” and discussing attempts by “regulators to employ informal rather than formal means” of censorship); Will Duffield, Jawboning Against Speech: How Government Bullying Shapes the Rules of Social Media, Cato Inst. (Sept. 12, 2022), https://www.cato.org/policy-analysis/jawboning-against-speech [https://perma.cc/D29U-TQUR] (“Government officials can use informal pressure — bullying, threatening, and cajoling — to sway the decisions of private platforms and limit the publication of disfavored speech. The use of this informal pressure, known as jawboning, is growing.”); Genevieve Lakier, Enforcing the First Amendment in an Era of Jawboning, 93 U. Chi. L. Rev. (forthcoming 2026) (manuscript at 1), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5162523 [https://perma.cc/WR
     J8-W9CY] (explaining that jawboning entails creation of “informal system of censorship” (quoting Bantam Books, 372 U.S. at 67)). Vullo nominally involved such a pattern. One question facing courts is how much they should draw from Vullo in assessing claims that executive branch actors have conditioned spending to coerce people or entities to do things other than regulate the speech of third parties, such as remain silent about particular topics or engage in particular expression. ↑

152. . See supra notes 56–61 and accompanying text (discussing arguments about applicability of checks to executive conditions); supra note 116 and accompanying text (discussing Ninth Circuit precedent finding executive conditions violate Spending Clause). ↑

153. . Prof. Spencer suggested a stringent Dole test for executive conditions to address accountability concerns. Spencer, supra note 24, at 1238–40. ↑

154. . See Zachary S. Price, Public Functions and Private Resources, 103 Wash. U. L. Rev. 1813 (2026) (developing an analogous argument). ↑

155. . Gluck & Huberfeld, supra note 40, at 1733–46 (detailing the dynamic negotiation of Medicaid waivers to expand eligibility under the ACA). ↑

156. . See Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579 (1952). ↑

157. . Id. at 598–99, 602 (Frankfurter, J., concurring) (noting that legislative history showed Congress debated and withheld such authority); see also id. at 657–58 (Burton, J., concurring) (same). ↑

158. . Id. at 629–33 (Douglas, J., concurring). See also id. at 631–32 (“The President has no power to raise revenues. That power is in the Congress by Article I, Section 8 of the Constitution. . . . [U]ntil and unless Congress acted, no condemnation would be lawful. The branch of government that has the power to pay compensation for a seizure is the only one able to authorize a seizure or make lawful one that the President had effected [sic]. That seems to me to be the necessary result of the condemnation provision in the Fifth Amendment.”). ↑

159. . Id. at 635–38 (footnotes omitted) (Jackson, J., concurring). ↑

160. . See Price, supra note 152. ↑

161. . See 1 U.S. Gov’t Accountability Off., GAO-04-261SP, Principles of Federal Appropriations Law (3d ed. 2004), https://www.gao.gov/assets/gao-04-261sp.pdf [https://perma.cc/
     K6DQ-JW85]. ↑

162. . See Settling the Question: Did Bank Settlement Agreements Subvert Congressional Appropriations Powers?: Hearing Before the Subcomm. on Oversight & Investigations of the H. Comm. on Fin. Servs., 114th Cong. 47 (2016) (statement of Ambassador C. Boyden Gray, Partner, Boyden Gray & Assocs.) (describing bank settlement agreements that “obligate the banks to provide hundreds of millions of dollars to third party credit counseling services and lawyers’ trust funds” and asserting that “[t]hese provisions raise difficult questions under the Appropriations Clause and the Miscellaneous Receipts Act”); Paul J. Larkin, Jr., The Justice Department’s Third-Party Payment Practice, the Antideficiency Act, and Legal Ethics, Federalist Soc’y Rev., Oct. 2016, at 28, 29 (concluding that the Appropriations Clause, the Antideficiency Act, and the Miscellaneous Receipts Act prohibit the Justice Department from entering settlements requiring payments to third parties). ↑