Commonwealth of KENTUCKY, et al., Plaintiffs-Appellees, v. Joseph R. BIDEN, in his official capacity as President of the United States of America, et al., Defendants-Appellants.
In 1949, Congress passed a statute called the Federal Property and Administrative Services Act (“Property Act”) to facilitate the “economical and efficient” purchase of goods and services on behalf of the federal government. See 40 U.S.C. § 101. The Property Act serves an uncontroversial purpose; who doesn't want the government to be more “economical and efficient”? Yet that laudable legislative-branch prescription, in place for the last seventy years, has recently been re-envisioned by the executive. In November 2021, the Safer Federal Workforce Task Force, under the supposed auspices of the Act, issued a “Guidance” mandating that the employees of federal contractors in “covered contract[s]” with the federal government become fully vaccinated against COVID-19. 1 That directive sweeps in at least one-fifth of our nation's workforce, possibly more. And so an act establishing an efficient “system of property management,” S. Rep. 1413 at 1 (1948), was transformed into a novel font of federal authority to regulate the private health decisions of millions of Americans.
In response, three states (Ohio, Kentucky, and Tennessee) and two Ohio sheriffs’ offices filed suit. They collectively alleged that nothing in the Property Act authorizes the contractor mandate, that the contractor mandate violates various other federal statutes, and that its intrusion upon traditional state prerogatives raises serious constitutional concerns under federalism principles and the Tenth Amendment. The district court agreed. It enjoined enforcement of the contractor mandate throughout Ohio, Kentucky, and Tennessee. It also denied the subsequent motion of the federal-government defendants 2 to stay the injunction pending appeal. The government now comes to us with the same request. But because the government has established none of the showings required to obtain a stay, we DENY such relief.
On September 9, 2021, President Biden delivered an address in which he announced that his “patience” with “unvaccinated Americans ․ is wearing thin.” Amended Complaint at 1 n.1, R. 22 (citing Joseph R. Biden, Remarks by President Biden on Fighting the COVID-19 Pandemic, The White House (Sept. 9, 2021), https://perma.cc/GQG5-YBXK). Reflecting that fact, the President earlier that day had signed Executive Order 14042 (E.O. 14042), titled “Ensuring Adequate COVID Safety Protocols for Federal Contractors.” 86 Fed. Reg. 50,985 (Sept. 14, 2021). Citing the Property Act and 3 U.S.C. § 301 3 as the relevant statutory authorities, the Order directs federal contractors to “provide adequate COVID-19 safeguards to their workers performing on or in connection with a Federal Government contract[.]” Id. More specifically, it directs them to comply “with all guidance for contractor or subcontractor workplace locations published by the Safer Federal Workforce Task Force”—itself created by President Biden in January 2021—“provided that the Director of the Office of Management and Budget” (“OMB”) determines that such guidance “will promote economy and efficiency in Federal contracting.” Id.
The Safer Federal Workforce Task Force promulgated its Guidance a few weeks later. See Safer Federal Workforce Task Force, COVID-19 Workplace Safety: Guidance for Federal Contractors and Subcontractors (Sept. 24, 2021), https://perma.cc/2R27-9J4U. The Guidance requires “COVID-19 vaccination of covered contractor employees, except in limited circumstances where an employee is legally entitled to an accommodation,” id. at 1, such as for a disability or religious objection. See id. at 5 (“Covered contractors must ensure that all covered contractor employees are fully vaccinated for COVID-19, unless the employee is legally entitled to an accommodation.”) (emphasis added). Even fully vaccinated employees must also continue to wear masks if they work “[i]n areas of high or substantial community transmission.” Id. at 6. Covered contractors “may,” but apparently are not required to, relax the masking requirement when the employee is eating or “alone in an office with floor to ceiling walls and a closed door.” Id. at 7. Further, the Guidance requires vaccination even of employees “who are not themselves working on or in connection with a covered contract,” id. at 3–4 (emphasis added), at least if they are “likely to be present” where a covered contract is being performed. Id. at 8.
Four days after the Guidance issued, the OMB Director issued a perfunctory “determination” pursuant to E.O. 14042 that the Guidance would promote “economy and efficiency” under the Property Act. 86 Fed. Reg. 53,691, 53,691–92 (Sept. 28, 2021). 4 But in response to ensuing lawsuits (or so the plaintiffs allege), the OMB bolstered the Director's initial explanation with another notice of determination on November 16, 2021. 5 86 Fed. Reg. 63,418 (Nov. 16, 2021); see also Amended Complaint ¶ 120, R. 22. Though longer than its predecessor, the second notice of determination mostly recapitulates the relevant features of the contractor mandate: that it requires vaccination at least and, for some employees, both vaccination and masking; that it includes even those employees not themselves performing a covered contract; and that it admits of only limited exceptions. Id. at 63,419–21. The determination also includes some information in its final pages about how vaccination reduces COVID's net costs—for instance, by reducing absenteeism. Id. at 63,422–23. And thus it pronounces “OMB's expert opinion that the Guidance will promote economy and efficiency in Federal Government procurement.” Id. at 63,423.
Before turning to the procedural history underlying this litigation, we pause to make a few observations on the Guidance's breadth. First, according to the Department of Labor, “workers employed by federal contractors” constitute “approximately one-fifth of the entire U.S. labor force.” Amended Complaint ¶ 127, R. 22 (citing Dep't of Labor, History of Executive Order 11246, https://perma.cc/6ZXJ-WGR8). As the plaintiffs point out, contractors thus constitute “large portions of the labor force[s]” in Ohio, Kentucky, and Tennessee. Id. And second, we emphasize just how expansively the Guidance defines which share of those contractors are “covered.” The Guidance does not cover merely those employees performing a covered contract. Rather, it also sweeps in employees merely working “in connection with” such contracts, and even “employees of covered contractors who are not themselves working on or in connection with a covered contract.” Guidance, supra, at 3–4 (emphasis added); cf. Mont v. United States, ––– U.S. ––––, 139 S. Ct. 1826, 1832, 204 L.Ed.2d 94 (2019) (“This Court has ․ recognized that ‘ “in connection with” is essentially indeterminate because connections, like relations, stop nowhere.’ ” (quoting Maracich v. Spears, 570 U.S. 48, 59, 133 S.Ct. 2191, 186 L.Ed.2d 275 (2013))).
True, as we mentioned, the Guidance stipulates that such employees are “covered” only if they are “likely” to interact with employees performing a covered contract. Guidance, supra, at 4. But it then explains that unless the “covered contractor can affirmatively determine that none of its employees” working on matters other than the covered contract “will come into contact with a covered contractor employee during the period of performance,” then those employees must also become vaccinated. Id. at 10, 11 (emphasis added). Thus, as the Guidance explains, it includes personnel working in such areas as “human resources, billing, and legal review.” Id. at 13. Likewise, “covered” workplaces include “common areas,” parking garages, and even workplaces outdoors. Id. at 10, 11. Not only that, an employee working from home “is a covered contractor employee.” Id. at 11. So employees confined to their residences still “must comply with the vaccination requirement for covered contractor employees, even if the employee never works at either a covered contractor workplace or Federal workplace during the performance of the contract.” Id. (emphasis added). Given that expansive scope of the Guidance, the interpretive trouble is not figuring out who's “covered”; the difficult issue is understanding who, based on the Guidance's definition of “covered,” could possibly not be covered.
We turn now to the underlying suit. The plaintiffs filed their complaint against the relevant federal officials on November 4, 2021, Complaint, R. 1, followed by an amended, operative complaint on November 15, 2021, Amended Complaint, R. 22. Two aspects of the complaint are particularly relevant to our stay determination: the plaintiffs’ theory of injury, on the one hand, and their concomitant theories of standing to sue, on the other. As for injury, the complaint alleges that the contractor mandate will encounter substantial resistance from the covered workforces both in the Ohio sheriffs’ offices and throughout Ohio, Kentucky, and Tennessee. Id. ¶¶ 52e–f. Many covered employees, the plaintiffs allege, would rather quit their jobs than comply with the contractor mandate. Id. The mandate thus poses serious obstacles to the sheriffs’ offices’ and states’ continued contracting with the federal government. Id. The sheriffs’ offices may be impeded in their continued ability to carry out basic public-safety functions if they lose their contracts with the Department of Homeland Security (“DHS”), id. ¶¶ 28–34, while the states may lose millions in valuable contracts between the federal government and their state agencies and universities, id. ¶¶ 12, 22, 48.
From those allegations, the complaint then propounds three theories of standing, one of which is shared by the states and sheriffs’ offices and two of which are unique to the states. First, both the states and sheriffs’ offices allege that they are suing in their own proprietary capacities to vindicate their own pecuniary interests in continued contracting with the federal government. Id. ¶¶ 1, 13, 34, 41. Second, the states also allege that they are suing as “separate sovereigns” to redress the injuries they suffer from federal intrusions into areas of traditional state concern, such as the regulation of public health. See id.; see also Response at 8. So in this sense as well, the states sue to vindicate their own “sovereign” and “quasi-sovereign” interests against federal overreach. Id. And third, the states allege that they are suing in a parens patriae capacity to redress the injuries of third parties—namely, of their respective citizens allegedly injured by the contractor mandate. Id.; see also Response at 8–9. The parens patriae theory is thus distinct from the former two theories of injury in that the states purport to litigate on their citizens’ behalf, rather than on behalf of the states’ own proprietary, sovereign, or quasi-sovereign interests. See id.
The district court apparently rejected all but the parens patriae theory of standing. It claimed that neither the sheriffs’ offices nor the states had “provid[ed] an example of a new contract that is subject to the mandate in their briefing.” Kentucky v. Biden, No. 3:21-cv-00055-GFVT, ––– F. Supp. 3d ––––, ––––, 2021 WL 5587446, at *3 (E.D. Ky. Nov. 30, 2021). It thus seemed to reason that without some specific future contract identified, the plaintiffs could not be suing to redress their own injuries. But the district court then reasoned that the state plaintiffs could sue under at least the third theory—parens patriae—to redress their citizens’ alleged injuries from the contractor mandate. Id. It noted that “federal contracts bring in billions of dollars to the [plaintiff states] annually” and that “there is every indication that federal contractors and subcontractors throughout Kentucky, Ohio, and Tennessee will continue bidding for new contracting opportunities.” Id. at ––––, 2021 WL 5587446 at *4. Thus, the district court held that the states could litigate on behalf of those contractors within the states’ jurisdictions that would be injured by the contractor mandate. Id.
The district court then proceeded to the merits of the dispute. It rejected the plaintiffs’ contention that the Guidance is arbitrary and capricious, labeling OMB's second “economy-and-efficiency analysis” “robust.” Id. at ––––, 2021 WL 5587446 at *12. But it proved more receptive to the plaintiffs’ other arguments. It first concluded as a matter of statutory interpretation that the Property Act likely does not empower the President to “promulgat[e] a public health measure such as mandatory vaccination.” Id. at ––––, 2021 WL 5587446 at *6. Second, it reasoned that the contractor mandate seemed inconsistent with the Competition in Contracting Act (“CCA”). Id. at ––––, 2021 WL 5587446 at *8. The CCA requires “full and open competition through the use of competitive procedures,” and yet the contractor mandate seems to exclude from such competition otherwise-capable contractors unwilling to comply with the contractor mandate. Id. (quoting 41 U.S.C. § 3301(a)(1)). Third, the district court appeared to suggest, but did not explicitly rely on the point, that the Property Act engenders non-delegation concerns. Id. at –––– – ––––, 2021 WL 5587446 at *8–9. And last, it expressed “a serious concern that Defendants have stepped into an area traditionally reserved to the States,” in apparent contravention of federalism principles and the Tenth Amendment. Id. at ––––, 2021 WL 5587446 at *10. It thus preliminarily enjoined enforcement of the contractor mandate throughout Ohio, Kentucky, and Tennessee.
The government took an interlocutory appeal of that preliminary injunction to this Court. See 28 U.S.C. § 1292(a)(1). It also moved the district court to stay its injunction pending appeal. The district court denied the government's motion and stood by its initial analysis. In response, the government likewise moved this Court for a stay of the injunction. We turn now to that request.
As our Court recently explained, “[a] stay is an intrusion into the ordinary processes of administration and judicial review.” In re MCP No. 165, OSHA Interim Final Rule: COVID-19 Vaccination & Testing, No. 21-7000, 21 F.4th 357, 365–66, 367–69, (6th Cir. Dec. 17, 2021) (cleaned up) (quoting Nken v. Holder, 556 U.S. 418, 427, 129 S.Ct. 1749, 173 L.Ed.2d 550 (2009)). Thus, a stay “is not a matter of right.” Id. (quoting Nken, 556 U.S. at 427, 129 S.Ct. 1749). Rather, “ ‘the heavy burden for making out a case for such extraordinary relief’ rests on ‘the moving part[y]’ ”—here, the federal government. Id. (quoting Winston-Salem/Forsyth Cnty. Bd. of Educ. v. Scott, 404 U.S. 1221, 1231, 92 S.Ct. 1236, 31 L.Ed.2d 441 (1971)).
A four-factor inquiry governs whether imposition of a stay is appropriate: “(1) whether the stay applicant has made a strong showing that he is likely to succeed on the merits; (2) whether the applicant will be irreparably injured absent a stay; (3) whether issuance of the stay will substantially injure the other parties interested in the proceeding; and (4) where the public interest lies.” Nken, 556 U.S. at 426, 129 S.Ct. 1749 (quoting Hilton v. Braunskill, 481 U.S. 770, 776, 107 S.Ct. 2113, 95 L.Ed.2d 724 (1987)). In other words, the government bears the “heavy burden” of showing that it is likely to succeed on the merits of its appeal, that it will be irreparably injured without a stay, that imposition of a stay would not “substantially injure” the plaintiffs, and that the public interest favors a stay. Id. at 427, 426, 129 S.Ct. 1749. But we remember that “[t]hese factors are not prerequisites that must be met”; they are instead “interrelated considerations that must be balanced together.” SawariMedia, LLC v. Whitmer, 963 F.3d 595, 596 (6th Cir. 2020) (quoting Serv. Emp. Int'l Union Loc. 1 v. Husted, 698 F.3d 341, 343 (6th Cir. 2012) (per curiam)). And “w[e] consider the motion de novo[,] because ‘we are not reviewing any district court decision or order.’ ” Priorities USA v. Nessel, 978 F.3d 976, 982 (6th Cir. 2020) (quoting A. Philip Randolph Inst. v. Husted, 907 F.3d 913, 917 (6th Cir. 2018)).
We will analyze the four stay factors in order, beginning with the government's likelihood of success on appeal in its arguments that (1) the plaintiffs lack standing to sue and (2) even if the plaintiffs have standing, the Property Act authorizes the contractor mandate. For the reasons explained below, we conclude that the government is not likely to succeed on appeal on either argument.
Factor 1: Likelihood of Success on the Merits
We begin with the government's contentions on standing. We explain first why the government is unlikely to succeed in its contention that the plaintiff states and sheriffs’ offices lack standing to sue in their own proprietary capacities. Next, we explain why the government is likely to succeed in its contention that the plaintiff states have no parens patriae standing to litigate on behalf of their allegedly injured citizens. But we conclude by explaining why the government is not likely to succeed in its contention that the plaintiff states cannot sue to vindicate their own interests in their sovereign and quasi-sovereign capacities. 6
Standing Based on the States’ and Sheriffs’ Offices’ Own Proprietary Capacities
Both the plaintiff states and sheriffs’ offices likely have standing to sue in their own proprietary capacities as contractors with the federal government. As the plaintiffs point out, the contractor mandate is already affecting the Seneca County Sheriff's Office's contract with DHS. Response at 10. After the mandate was announced, a “contract specialist” from DHS emailed Seneca County Sheriff Frederick W. Stevens instructing him to “please provide signature” on a contractual modification to incorporate the contractor mandate. See Hadden E-Mail, R. 27-2; Modification, R. 12-2. Declining to incorporate the modification clause could thus jeopardize the sheriffs’ offices’ contracts with DHS.
Moreover, the sheriffs’ offices have a demonstrated history of contracting with federal agencies like DHS and Immigration and Customs Enforcement (“ICE”). See Stevens Dec., R. 12-2; Hildenbrand Dec., R. 12-3. Requiring the contractor mandate in the sheriffs’ offices’ other, future contracts with DHS or ICE could also adversely affect the sheriffs’ offices’ own economic interests. They either will be unable to comply with the mandate, given anticipated resistance to it, and will lose the contracts, or they will comply with the contractor mandate but suffer serious hits to their workforces as employees resign in protest. The sheriffs’ offices likely have standing in their own proprietary capacities as contractors to contest these negative ramifications of the contractor mandate.
Likewise, whatever the complexities of parens patriae standing, no one claims that a prudential bar blocks the states from litigating in their own proprietary capacities to vindicate their own proprietary interests threatened by the contractor mandate. See Massachusetts v. EPA, 549 U.S. 497, 522, 127 S.Ct. 1438, 167 L.Ed.2d 248 (2007). As the states have shown, they and their state agencies are themselves federal contractors that will become subject to the contractor mandate but for the district court's injunction. 7 For instance, state universities, state departments of health, and jails reliant on the states’ coffers all contract extensively with the federal government. Relevant federal agencies with which the states have contracts include the United States Department of Justice, the United States Marshals Service, the Food and Drug Administration, the National Institutes of Health, the Bureau of Labor Statistics, the National Aeronautics and Space Administration, the Consumer Product Safety Commission, and the Centers for Disease Control and Prevention (“CDC”). See Amended Complaint ¶¶ 4, 16–17, R. 22.
As a result, each state so contracting is threatened with the imposition of the contractor mandate in two distinct ways. First, the federal government may enforce the contractor mandate any time the parties need to modify an existing contract with the state plaintiffs, given that the Guidance defines “contract” so expansively as to include modifications to an existing contract. See Guidance, supra, at 3 (“In addition to bilateral instruments, contracts include, but are not limited to, awards and notices of awards; job orders or task letters issued under basic ordering agreements; letter contracts; orders, such as purchase orders, under which the contract becomes effective by written acceptance or performance; exercised contract options; and bilateral contract modifications.”). Second, the state plaintiffs are also imminently threatened in their proprietary capacities should they renew those existing contracts (thus triggering the mandate as well) or should they choose to bid on new contracts to which the mandate applies. And if they chose not to renew such contracts given the contractor mandate, they could lose millions of dollars in funding from the federal government for critical state programs. See Amended Complaint ¶¶ 12, 22, 48, R. 22.
Given these realities, we are not persuaded by the government's claims that the states lack standing in their proprietary capacities because the contractor mandate applies “only in new or renewed contracts.” Gov't Mot. for Stay at 8. The events triggering imposition of the mandate are far broader than merely the signing of a new contract. See Guidance, supra, at 3. But the government also inexplicably discounts the virtual certainty that states will either bid on new federal contracts or renew existing ones. By engaging in such prolific federal contracting, the federal government has engendered substantial state reliance interests in securing future contracts. It is unreasonable, given those reliance interests, to expect states or their agencies to disavow their prior history of contracting and to decline to seek future such opportunities. And that point only underscores the states’ injury. The federal government of course knows that these reliance interests exist, which is why it seeks to purchase states’ submission by leveraging those interests to force their acquiescence to the contractor mandate. See Philip Hamburger, Purchasing Submission: Conditions, Power, and Freedom 18 (2021) (criticizing the federal government's use of the spending power “to sidestep congressional lawmaking, adjudication by the courts, the enumerated federal powers, federalism, and a host of constitutional rights”). We conclude, therefore, that the plaintiffs have plausibly alleged a theory of standing in their proprietary capacities.
The States’ Parens Patriae Standing
In addition to their proprietary interests, the states assert broader theories of standing based on parens patriae and their sovereign and quasi-sovereign interests. We turn first to parens patriae, which we conclude is not a viable means for standing, and then contrast it with standing in a sovereign or quasi-sovereign capacity, which, as we explain, the states have validly established.
A brief clarification of terms is useful at the outset. “Parens patriae,” at least in the context of standing, really encompasses two distinct concepts. First is the original parens patriae doctrine, a form of third-party standing that existed at common law. See Chapman v. Tristar Prod., Inc., 940 F.3d 299, 305 (6th Cir. 2019). Under this conception of parens patriae—a term literally meaning “parent of the country”—the King could litigate on behalf of those incapable of properly representing their own interests, such as the mentally disabled. Id. (quoting Alfred L. Snapp & Son, Inc. v. Puerto Rico ex rel. Barez, 458 U.S. 592, 600, 102 S.Ct. 3260, 73 L.Ed.2d 995 (1982)). The King, in other words, did not sue to redress his own injuries but the injuries of those who were, in effect, his wards. Alfred L. Snapp & Son, Inc., 458 U.S. at 600, 102 S.Ct. 3260. Today, states may not invoke this third-party-standing conception of parens patriae to sue the United States on behalf of state citizens allegedly harmed by the federal government. See Massachusetts v. Mellon, 262 U.S. 447, 485–86, 43 S.Ct. 597, 67 L.Ed. 1078 (1923). We will further explain the reasons for that change below, but we note at the outset that to the extent the plaintiff states here purport to sue purely on behalf of their own citizens’ interests, such a theory of standing is forbidden.
That brings us to the second, more modern conception of parens patriae, which, unlike its ancestor, generally is permissible. Under this more modern conception, states sometimes purport to sue in a “parens patriae” capacity, yet what they are really doing is asserting some injury to their own interests separate and apart from their citizens’ interests. Chapman, 940 F.3d at 305 (citing Alfred L. Snapp & Son, Inc., 458 U.S. at 601–02, 102 S.Ct. 3260). The classic cases involve public nuisances, in which a state sues to prevent pollution that not only injures its citizens but also invades the state's prerogative to superintend the public health. See, e.g., Georgia v. Tenn. Copper Co., 206 U.S. 230, 237, 27 S.Ct. 618, 51 L.Ed. 1038 (1907). The Supreme Court has said that in such instances, “the State has an interest independent of and behind the titles of its citizens” to safeguard “its domain,” id. at 237, 27 S.Ct. 618, and its “health, comfort and welfare,” Pennsylvania v. West Virginia, 262 U.S. 553, 592, 43 S.Ct. 658, 67 L.Ed. 1117 (1923), and thus that its suit may proceed.
The distinction between the two theories becomes most acute when a state sues the United States and its officers. While a state may so sue when it seeks to vindicate its own sovereign and quasi-sovereign interests against the United States, see infra pages 598–601, it cannot sue when it claims to represent its citizens in a purely third-party parens patriae capacity. The case most associated with this distinction is Massachusetts v. Mellon. There, Massachusetts sought to sue on behalf of its own citizens to vindicate their putative interests against being governed by an allegedly unconstitutional federal statute. Mellon, 262 U.S. at 485–86, 43 S.Ct. 597. After explicitly noting that the dispute did not involve “quasi sovereign rights actually invaded or threatened,” the Court explained:
We come next to consider whether the suit may be maintained by the state as the representative of its citizens. To this the answer is not doubtful. ․ It cannot be conceded that a state, as parens patriae, may institute judicial proceedings to protect citizens of the United States from the operation of the statutes thereof. While the state, under some circumstances, may sue for the protection of its citizens, it is no part of its power or duty to enforce their rights in respect of their relations with the federal government. In that field it is the United States, and not the state, which represents them as parens patriae, when such representation becomes appropriate; and to the former, and not to the latter, they must look for such protective measures as flow from that status.
Id. (citation omitted). So, in other words, when sovereign and quasi-sovereign interests are not on the line, a state cannot litigate in a third-party capacity as parens patriae against the United States. A solitary state is not the “parent of the country”; that distinction belongs to the United States. See also Georgia v. Penn. R.R. Co., 324 U.S. 439, 446, 65 S.Ct. 716, 89 L.Ed. 1051 (1945) (“Commonwealth of Massachusetts v. Mellon ․ make[s] plain that the United States not the State represents the citizens as parens patriae in their relations to the federal government.”); South Carolina v. Katzenbach, 383 U.S. 301, 324, 86 S.Ct. 803, 15 L.Ed.2d 769 (1966) (“Nor does a State have standing as the parent of its citizens to invoke these constitutional provisions against the Federal Government, the ultimate parens patriae of every American citizen.”). For better or worse, later cases label this prudential constraint the “Mellon bar.” See, e.g., Gov't of Manitoba v. Bernhardt, 923 F.3d 173, 179 (D.C. Cir. 2019).
In the government's view, the Mellon bar precludes the states’ entire suit because, also in the government's view, the complaint seeks to vindicate purely the third-party interests of covered contractors that happen to reside within the plaintiff states. See Gov't Mot. for Stay at 9–10. Yet reading the complaint in even the worst possible light cannot produce the government's desired result. The complaint refers to “sovereign, quasi-sovereign, proprietary, and parens patriae interests,” which implies that the states view themselves as asserting both a (permissible) sovereign-and-quasi-sovereign theory and an (impermissible) third-party parens patriae theory, rather than wholly the latter theory. Amended Complaint ¶¶ 1, 13, 41, R. 22.
To the extent that the complaint asserts purely third-party interests of covered contractors that happen to reside within the states, we agree with the government that this third-party theory is impermissible under the Mellon bar. The United States, not individual states, is the modern-day “parent of the country” for purposes of third-party parens patriae standing. But we disagree that the Mellon bar, ipso facto, precludes all but the proprietary-capacity theory. For as their complaint makes clear, the plaintiff states also seek to assert their own “sovereign” and “quasi-sovereign” interests against the federal government. Id. Mellon explicitly does not speak to this situation, since it disavowed that “quasi sovereign rights” were there at stake—a point the Supreme Court later confirmed in another Massachusetts dispute, Massachusetts v. EPA. See Massachusetts, 549 U.S. at 520 n.17, 127 S.Ct. 1438 (“Mellon itself disavowed any such broad reading when it noted that the Court had been ‘called upon to adjudicate, not rights of person or property, not rights of dominion over physical domain, [and] not quasi-sovereign rights actually invaded or threatened.” (quoting Mellon, 262 U.S. at 484–85, 43 S.Ct. 597)). There is thus no Mellon bar against the plaintiff states’ suit in their sovereign and quasi-sovereign capacities.
The government's strongest case in support of its contrary view is apparently the D.C. Circuit's decision in Bernhardt. True, Bernhardt appears to reject the distinction we draw here—between (1) permissible quasi-sovereign-interest suits against the federal government, sometimes nominally labeled “parens patriae” suits, and (2) classical parens patriae suits in which states impermissibly claim to represent merely the interests of third parties against the federal government. See Bernhardt, 923 F.3d at 182 (“The distinction is not, as Missouri suggests, between two types of parens patriae lawsuits, one permissible and one not. It is between a parens patriae lawsuit (what Mellon prohibits) and a State suing based on ‘its rights under federal law’ (not a parens patriae lawsuit at all).”). We reject the D.C. Circuit's analysis for two reasons. First, it mistakenly conflates quasi-sovereign-interest suits with third-party parens patriae suits to suggest that Mellon categorically bars both. Id. at 182. Yet as we have shown, Mellon does not. That case invalidates the traditional third-party-standing conception of parens patriae, but it does not invalidate (or even address) the quasi-sovereign-interest theory. See Massachusetts, 549 U.S. at 520 n.17, 127 S.Ct. 1438. And second, the D.C. Circuit's putative “Bernhardt bar” conflicts with Supreme Court precedent. As the Court recognized in Massachusetts v. EPA, post-Mellon precedent endorses the view that a state has “standing to bring a cross-claim against the United States to vindicate its ‘quasi-sovereign’ interests which are ‘independent of and behind the titles of its citizens.’ ” Id. (quoting Nebraska v. Wyoming, 515 U.S. 1, 20, 115 S.Ct. 1933, 132 L.Ed.2d 1 (1995)).
We therefore conclude that Mellon likely bars the state plaintiffs’ claims to the extent that they seek to litigate in a purely third-party parens patriae capacity against the United States and its agents. But we also conclude that Mellon likely does not bar the state plaintiffs’ claims to the extent they assert sovereign and quasi-sovereign interests against those defendants. Having established that the latter theory is permissible, we turn now to whether the plaintiffs have sufficiently shown the sovereign- and quasi-sovereign-interests theory of standing.
The States’ Sovereign and Quasi-Sovereign Interests
As our Court, other circuits, and the Supreme Court have all recognized, states have a variety of sovereign and quasi-sovereign interests that they validly may seek to vindicate in litigation. States have a sovereign interest to sue the United States when a federal regulation purports to preempt state law. See Ohio ex rel. Celebrezze v. U.S. Dep't of Transp., 766 F.2d 228, 232–33 (6th Cir. 1985) (holding that Ohio could sue to contest purported federal preemption of a state law); see also Alaska v. U.S. Dep't of Transp., 868 F.2d 441, 442–43 & n.1 (D.C. Cir. 1989) (same); Texas v. United States, 809 F.3d 134, 153 (5th Cir. 2015) (“[S]tates may have standing based on ․ federal preemption of state law[.]”). States also have sovereign interests to sue when they believe that the federal government has intruded upon areas traditionally within states’ control. Texas, 809 F.3d at 153 (explaining that states also “may have standing based on ․ federal assertions of authority to regulate matters they believe they control” (citing Tex. Off. of Pub. Util. Counsel v. FCC, 183 F.3d 393, 449 (5th Cir. 1999))); see also Alfred L. Snapp & Son, Inc., 458 U.S. at 601, 102 S.Ct. 3260 (noting states’ “sovereign interests” in both “the exercise of sovereign power over individuals and entities within the relevant jurisdiction” and “the demand for recognition from other sovereigns”). And states also have a recognized quasi-sovereign interest in the health and “economic well-being” of their populaces. Alfred L. Snapp & Son, Inc., 458 U.S. at 605, 102 S.Ct. 3260; see also Missouri v. Illinois, 180 U.S. 208, 241, 21 S.Ct. 331, 45 L.Ed. 497 (1901) (noting states’ interest in “the health and comfort of the[ir] inhabitants”); Pennsylvania, 262 U.S. at 592, 43 S.Ct. 658 (concluding that a state could litigate to defend the “health, comfort, and welfare” of its citizens); Chapman, 940 F.3d at 305 (“[A] state has a quasi-sovereign interest in the ‘health and well-being—both physical and economic—of its residents in general.’ ”).
As we noted above and now emphasize here, none of these sovereign-and-quasi-sovereign-interest theories relies on impermissible notions of third-party standing in which a state asserts in a purely vicarious manner the interests of its citizens. Rather, as the Supreme Court has recognized, these theories involve “interest[s] apart from the interests of particular private parties.” Alfred L. Snapp & Son, Inc., 458 U.S. at 607, 102 S.Ct. 3260. In these cases, in other words, the state is not merely a “nominal party.” Id. The state instead asserts an injury that, while possibly overlapping with individual citizens’ injuries, is really an additional injury to the state itself. Id.
On that understanding, then, we conclude that the state plaintiffs have plausibly shown standing in the states’ sovereign and quasi-sovereign capacities. They have done so in two ways. First, they have shown that each of the states follows its own, contrary vaccination policy, and that the contractor mandate threatens to override those policies. See Amended Complaint ¶¶ 11, 21, 47, R. 22. They have also plausibly alleged that the federal government has intruded upon an area traditionally left to the states—the regulation of the public health of state citizens in general and the decision whether to mandate vaccination in particular. See infra pages 608–10 (discussing the federalism implications of the contractor mandate). The contractor mandate thus likely implicates states’ power to make and enforce policies and regulations, as well as states’ traditional prerogative to superintend their citizens’ health and safety. See Alfred L. Snapp & Son, Inc., 458 U.S. at 601, 603–04, 102 S.Ct. 3260.
And second, the plaintiff states have plausibly shown that the contractor mandate threatens to damage each of the states’ economies. 8 See id. at 605, 102 S.Ct. 3260. For instance, the states note that “almost 70% of unvaccinated Americans would quit their jobs if a vaccine mandate were required and their exemption were denied.” Amended Complaint ¶ 52e, R. 22 (citing Jordan Burrows, Employees Not Given Exemption Prefer to Quit Job Than Get COVID Vaccine, Poll Shows, Salt Lake City ABC4.com (Sept. 15, 2021), https://perma.cc/6A95-CJXD). Likewise, “9 in 10 large employers fear reductions in their workforces if they have to implement vaccine mandates.” Id. (citing Karl Evers-Hillstrom, 9 in 10 Employers Say They Fear They'll Lose Unvaccinated Workers Over Mandate: Survey, The Hill (Oct. 18, 2021), https://perma.cc/V5ZJ-7XUQ). The states thus plausibly allege that resistance to the contractor mandate will result in layoffs, further supply-chain issues, and rising prices, all to the detriment of their state economies. The states likely have a quasi-sovereign interest in defending their economies from the alleged negative ramifications of the contractor mandate. And, because the contractor mandate implicates these sovereign and quasi-sovereign interests, the states likely have standing to contest it.
Constitutional Standing Under Article III
Thus far we have concluded that likely no prudential bar prevents the states from suing the United States to vindicate their proprietary, sovereign, and quasi-sovereign interests, and that similarly no such bar likely prevents the sheriffs’ offices from suing in their own proprietary capacities. Having defined the interests at stake—proprietary, sovereign, and quasi-sovereign—we now examine whether the plaintiffs have also shown “the irreducible constitutional minim[a]” to establish Article III standing. Lujan v. Defs. of Wildlife, 504 U.S. 555, 560, 112 S.Ct. 2130, 119 L.Ed.2d 351 (1992). The Supreme Court's modern standing test has coalesced around three elements. First, “the plaintiff must have suffered an ‘injury in fact’—an invasion of a legally protected interest which is (a) concrete and particularized ․ and (b) actual or imminent, not conjectural or hypothetical.” Id. (cleaned up). “Second, there must be a causal connection between the injury and the conduct complained of—the injury has to be fairly traceable to the challenged action of the defendant, and not the result of the independent action of some third party not before the court.” Id. (cleaned up). And “[t]hird, it must be likely, as opposed to merely speculative, that the injury will be redressed by a favorable decision.” Id. at 561, 112 S.Ct. 2130 (cleaned up).
The government does not contest the latter two elements of causation and redressability. The contractor mandate may be “fairly trace[d]” to the defendants, as they are the agents of the United States responsible for devising, promulgating, and potentially enforcing the Guidance. Id. at 560, 112 S.Ct. 2130. Likewise, the government does not seriously contest redressability. If the government's enforcement of the contractor mandate is enjoined, then the harms that will allegedly flow from the contractor mandate will have been prevented, and thus the plaintiffs’ claimed injuries redressed.
The government instead asserts that the plaintiffs have failed to establish imminent injury. Gov't Mot. for Stay at 8. But the government is unlikely to succeed on this objection on appeal. As for constitutional standing in the states’ and sheriffs’ offices’ proprietary capacities, the plaintiffs have shown that their existing contracts are threatened with modification under the contractor mandate, that they have a history of bidding on other federal contracts, that federal contracts provide critical funding for their functions, that they are thus likely to continue bidding on federal contracts, that such contracts are likely subject to the contractor mandate, and that resistance to the contractor mandate will likely lead either to the loss of contracts or difficulty executing such contracts. As for constitutional standing in the states’ sovereign and quasi-sovereign capacities, the states have made a similarly adequate showing. They allege that many thousands of citizens throughout their states are employed by federal contractors, that a substantial portion will likely resist the contractor mandate, that resistance could lead to the loss of federal contracts or difficulty performing such contracts, and that these facts will harm the states’ economies. Likewise, they have shown that they have sovereign interests and traditional prerogatives in regulating public health and compulsory vaccination and that the contractor mandate invades these prerogatives.
All of this easily suffices to establish imminence. In addition to showing negative effects on existing contracts, as the plaintiffs have done, they may also establish imminence with “an adequate showing that sometime in the relatively near future [they] will bid on another Government contract” that allegedly violates the law. Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 211, 115 S.Ct. 2097, 132 L.Ed.2d 158 (1995). In Adarand itself, the Supreme Court concluded that Adarand had satisfied the imminence requirement by claiming that it would bid on contracts in the future and by showing that it had a demonstrated history of such bidding. Id. at 212, 115 S.Ct. 2097; see also Sherbrooke Turf., Inc. v. Minn. Dep't of Transp., 345 F.3d 964, 967 (8th Cir. 2003) (explaining that the plaintiffs had Article III standing because they “ha[d] bid on federally assisted highway projects in the past, will continue to bid in the future, and suffer competitive harm” from a new federal policy making it harder to win future contracts). That is precisely the showing the plaintiffs have made here. Both the plaintiff states and sheriffs’ offices, then, likely have shown that their claims satisfy Article III's standing requirements.
The Existence of a Cause of Action
Just because the plaintiffs have standing to sue, of course, does not mean that they have a cause of action with which they can vindicate their purported interests. See Texas, 809 F.3d at 161 (“[A] state that has standing still must have a cause of action.”). Yet the plaintiffs are likely to succeed on appeal in their contention that they have a cause of action under the Administrative Procedure Act (“APA”), and specifically under 5 U.S.C. § 702. 9 See Amended Complaint ¶ 79, R. 22. That provision explains that “[a] person suffering legal wrong because of agency action, or adversely affected by agency action within the meaning of a relevant statute, is entitled to judicial review thereof.” 5 U.S.C. § 702. It also waives the federal government's sovereign immunity from injunctive relief. Id. For better or worse, the Supreme Court has read this language as creating a cause of action, rather than merely providing that plaintiffs with preexisting rights in law or equity may sue agencies to vindicate those rights. See Japan Whaling Ass'n v. Am. Cetacean Soc'y, 478 U.S. 221, 230 n.4, 106 S.Ct. 2860, 92 L.Ed.2d 166 (1986); see also Caleb Nelson, “Standing” and Remedial Rights in Administrative Law, 105 Va. L. Rev. 703, 708–09 (2019). And courts have also read § 702 to extend to states as well. 10 See Bernhardt, 923 F.3d at 181 (“There is little doubt that a State qualifies as a ‘person’ under the APA.” (citing Md. Dep't of Hum. Res. v. Dep't of Health & Hum. Servs., 763 F.2d 1441, 1445 n.1 (D.C. Cir. 1985))). The plaintiffs are thus likely to succeed on appeal in their contentions not only that they have prudential and constitutional standing, but that they also have a concomitant cause of action.
Whether the Property Act Authorizes the Contractor Mandate
At last we turn to the Property Act—the claimed source of authority for the contractor mandate. Whether the Property Act authorizes the President to impose such a measure (and thus whether he validly may delegate such authority to an agency) is a question of statutory interpretation. “The controlling principle,” then, is that we “must give effect to the clear meaning of statutes as written.” Star Athletica, L.L.C. v. Varsity Brands, Inc., ––– U.S. ––––, 137 S. Ct. 1002, 1010, 197 L.Ed.2d 354 (2017) (quoting Est. of Cowart v. Nicklos Drilling Co., 505 U.S. 469, 476, 112 S.Ct. 2589, 120 L.Ed.2d 379 (1992)). To do so, we assign each word of the statute “its ‘ordinary, contemporary, common meaning,’ ” id. (quoting Walters v. Metro. Educ. Enters., Inc., 519 U.S. 202, 207, 117 S.Ct. 660, 136 L.Ed.2d 644 (1997)), while keeping in mind that “[s]tatutory language has meaning only in context.” Graham Cnty. Soil & Water Conserv. Dist. v. United States ex rel. Wilson, 545 U.S. 409, 415, 125 S.Ct. 2444, 162 L.Ed.2d 390 (2005); see also Univ. of Tex. Sw. Med. Ctr. v. Nassar, 570 U.S. 338, 356, 133 S.Ct. 2517, 186 L.Ed.2d 503 (2013) (“Text may not be divorced from context.”).
The government points to two portions of the Property Act that it claims authorize the contractor mandate: 40 U.S.C. §§ 101 & 121. We reproduce that language in full below because it is, ironically, likely the best evidence against the government's position. First comes the Act's statement of purpose:
The purpose of this subtitle is to provide the Federal Government with an economical and efficient system for the following activities:
(1) Procuring and supplying property and nonpersonal services, and performing related functions including contracting, inspection, storage, issue, setting specifications, identification and classification, transportation and traffic management, establishment of pools or systems for transportation of Government personnel and property by motor vehicle within specific areas, management of public utility services, repairing and converting, establishment of inventory levels, establishment of forms and procedures, and representation before federal and state regulatory bodies.
(2) Using available property.
(3) Disposing of surplus property.
(4) Records management.
40 U.S.C. § 101. Section 121(a) then states, “The President may prescribe policies and directives that the President considers necessary to carry out this subtitle. The policies must be consistent with this subtitle.” 40 U.S.C. § 121(a).
The government contends that these two statutory provisions “plainly authorize[ ] the President” to order the contractor mandate—the imposition of an irreversible medical procedure without precedent in the history of the Property Act's application. Gov't Mot. for Stay at 13 (emphasis added). But the government's argument is unlikely to succeed for two central reasons: the relevant text, in fact, unambiguously precludes the government's theory, and, even if there were some ambiguity, the relevant canons of interpretation would foreclose construing the ambiguity in the government's favor. We analyze these points in turn.
By its plain text, the Property Act does not authorize the contractor mandate. The government itself offers virtually no textual analysis, which is unsurprising given that the text undermines its position. The government apparently supposes that the statute's statement-of-purpose section authorizes the President to procure “economical and efficient” “nonpersonal services.” 11 It then claims that the contractor mandate fulfills that goal, since it allegedly makes federal contractors more “economical and efficient” by reducing absenteeism. Gov't Mot. for Stay at 12. The first issue with the government's approach is its heavy reliance on the statement of purpose in § 101. Statements of purpose may be useful in construing enumerated powers later found in a statute's operative provisions. Sturgeon v. Frost, ––– U.S. ––––, 139 S. Ct. 1066, 1086, 203 L.Ed.2d 453 (2019). But statements of purpose are not themselves those operative provisions, so they cannot confer freestanding powers upon the President unbacked by operative language elsewhere in the statute. Id.; see also Gundy v. United States, ––– U.S. ––––, 139 S. Ct. 2116, 2127, 204 L.Ed.2d 522 (2019) (plurality) (describing a statement of purpose as simply an “appropriate guide to the meaning of the statute's operative provisions”) (quoting Antonin Scalia & Bryan A. Garner, Reading Law: The Interpretation of Legal Texts 220 (2012)) (cleaned up). Despite that fact, the government relies almost exclusively on § 101, leaving unexplained the link between its statement-of-purpose language and some other operative provision of the Property Act.
But even if we construed the statement of purpose in § 101 as an operative grant of power, its text does not, in fact, authorize the President to take “necessary measures” to procure “economical and efficient” “nonpersonal services.” It permits him to employ an “economical and efficient system” to “procur[e]” those nonpersonal services. 40 U.S.C. § 101 (emphases added). Webster's defines those critical terms as follows: “System,” in context, refers to “[a] formal scheme or method of governing organization, arrangement.” System, Webster's New International Dictionary 2562 (2d ed. 1959). And “procure” means “[t]o bring into possession; to acquire; gain; get; to obtain by any means, as by purchase or loan.” Procure, id. at 1974. The President thus has statutory authority to implement an “economical and efficient” method of contracting—a “system,” in other words—to obtain nonpersonal services. But there is no textual warrant to suggest that after the President or his agents have “economical[ly] and efficient[ly]” acquired those services that they then may impose whatever medical procedure deemed “necessary” on the relevant services personnel to make them more “economical and efficient.”
Likewise, we note for ourselves (given the dearth of textual analysis from the government) that the “performing related functions including contracting” language also cannot sustain the contractor mandate. The government's argument seems to implicitly assume that all the employees subject to the contractor mandate are continuously “contracting” under § 101, 12 and thus that the President can impose upon them those measures necessary to make the discharge of their “contracting” duties more “economical and efficient.” Yet this view has two major problems. First, “contracting” within § 101 refers to the government's initial entry into a contractual agreement to procure nonpersonal services—not all the subsequent tasks performed in connection with the contract. See Contracting, 2 The Oxford English Dictionary 914 (1933) (defining “contracting” as “[e]ntering into a contract or mutual agreement”). If the latter were true, then why specify all the other “related functions” that § 101 covers, like inspection, storage, transportation, repairs, regulation of inventory, and so on? Those “functions” will naturally occur pursuant to or in connection with a contract's performance, so it seems pure surplusage to enumerate other such functions if “contracting” really means not only making the agreement, but all the other activities done pursuant to it. See Direct Mktg. Ass'n v. Brohl, 575 U.S. 1, 13, 135 S.Ct. 1124, 191 L.Ed.2d 97 (2015) (explaining that courts should avoid “broad construction[s]” that render statutory terms “mere surplusage”). By contrast, of course, if “contracting” is properly confined to its contextual meaning—the government's making of the agreement, rather than all subsequent performance of it—the surplusage issues disappear. Second and relatedly, § 101 refers to “contracting” as a “function[ ]” “perform[ed]” by “the Federal Government”—not by the employees of federal contractors. 40 U.S.C. § 101 (emphasis added). That makes perfect sense if “contracting” in § 101 refers to the government's initial entry into a procurement contract, since the government there performs the act of contracting—“entering into a contract”—when it signs an agreement to procure nonpersonal services. But given that statutory text, it makes little sense to construe “contracting” as likewise covering the subsequent performance of the contract. Those actions are “perform[ed]” by the private employees of the contractors whom the government procured—not by the government itself. Id. Section 101 thus authorizes the President to implement systems making the government's entry into contracts less duplicative and inefficient, but it does not authorize him to impose a medical mandate directly upon contractor employees themselves because he thinks it would enhance their personal productivity.
This interpretation is, moreover, the only one plausible considering the historical concerns that motivated the passage of the Property Act. The fear shortly after World War II was not that personnel executing duties under nonpersonal-services contracts were themselves performing in an uneconomical and inefficient manner, but instead that the manner in which federal agencies were entering into contracts to procure goods and services was not economical and efficient. See S. Rep. 1413 at 2–3. 13 Specifically, given the lack of centralized coordination of procurement efforts, many agencies entered duplicative contracts supplying the same items and creating a massive post-war surplus. Id.; see also James F. Nagle, A History of Government Contracting 411 (2d ed. 1999) (describing the wartime “procurement free-for-all” among federal agencies). The Property Act thus aimed to “integrate[ ] and centraliz[e]” procurement responsibility to prevent agencies from “unnecessary buying” of “the same articles in the same markets.” S. Rep. 1413 at 3. Yet no one seems to have envisioned the Property Act as a latent well of authority to order the medical enhancement of contractor employees to make them more “economical and efficient.”
So what in the statute does the government claim supports its position? The statutory language the government relies on the most comes from § 121(a), which explains that “[t]he President may prescribe policies and directives that the President considers necessary to carry out this subtitle.” 40 U.S.C. § 121(a). In the government's view, this broad language permits the President to direct certain medical decisions for the employees of federal contractors under the rationale of promoting economical and efficient performance of their duties. But the government reads too much into § 121(a). The President cannot “carry out this subtitle,” see § 121(a), by exerting a power the subtitle never actually confers. So while he may enjoy a modest valence of necessary and proper powers surrounding those powers enumerated in § 101, he cannot wield a supposedly necessary and proper power without showing how it clearly stems from a power enumerated. Cf., e.g., McCulloch v. Maryland, 17 U.S. 316, 423, 4 Wheat. 316, 4 L.Ed. 579 (1819). And that he has not done with the contractor mandate.
Even if the government had managed to show textual ambiguity, which it likely has not, related canons of interpretation still would likely foreclose construing such ambiguity in the government's favor. Two considerations are of particular concern—the contractor mandate's potentially vast economic significance and its potential implications for “the balance between federal and state power.” Ala. Ass'n of Realtors v. Dep't of Health & Hum. Servs., ––– U.S. ––––, 141 S. Ct. 2485, 2489, 210 L.Ed.2d 856 (2021). As we have detailed above, the contractor mandate sweeps in at least one-fifth of the American workforce. The true proportion may be even larger, given that the contractor mandate defines so capaciously who qualifies as a covered contractor. And the plaintiffs have shown that the resistance the contractor mandate is sure to encounter will engender economic disruption throughout the plaintiff states. If an agency really had the power to promulgate a so-called “Guidance” with such “vast economic and political significance,” we would need a clear statement from Congress delegating such authority to the executive branch. Util. Air. Regul. Grp. v. EPA, 573 U.S. 302, 324, 134 S.Ct. 2427, 189 L.Ed.2d 372 (2014) (“We expect Congress to speak clearly if it wishes to assign to an agency decisions of vast ‘economic and political significance.’ ” (quoting FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 160, 120 S.Ct. 1291, 146 L.Ed.2d 121 (2000))); see also Tiger Lily, LLC v. HUD, 5 F.4th 666, 671 (6th Cir. 2021) (“There is no clear expression of congressional intent in § 264 to convey such an expansive grant of agency power, and we will not infer one.”). Yet that is just what we lack—a clear statement from Congress that it intended the President to use a property-and-services procurement act, for a purpose never-before recognized, to effect major changes in the administration of public health.
We note that our application of the major-questions canon conforms with our Court's recent discussion of the same issue regarding OSHA's emergency temporary standard (“ETS”). Our Court there found the canon inapposite for two reasons. First, we concluded that “the statutory language” in the OSHA statute “unambiguously grant[ed] OSHA authority” for its emergency standard. In re MCP No. 165, 21 F.4th at 373–74, (emphasis added). Here, by contrast, the Property Act unambiguously excludes the purported power. 14 And second, in the OSHA case we posited that “OSHA's issuance of the ETS is not an enormous expansion of its regulatory authority,” given OSHA's long history of regulating workplace safety, which has included “vaccination and medical examinations” to “control[ ] the spread of disease.” Id. at 372–73, . Here, by contrast, the government has propounded no relevant history showing that it has ever wielded the Property Act to mandate “vaccination and medical examinations” or to “control[ ] the spread of disease.” Id. The only examples the government offered were instances in which the federal government said federal contractors (1) could not discriminate, (2) had to abide by wage and price controls, (3) had to hang posters advising employees that they could not be forced to join a union, and (4) had to confirm employees’ immigration status. Gov't Mot. for Stay at 11 (citing AFL-CIO v. Kahn, 618 F.2d 784, 790, 785 (D.C. Cir. 1979) (en banc); UAW-Labor Emp. & Training Corp. v. Chao, 325 F.3d 360, 366–67 (D.C. Cir. 2003); Chamber of Com. v. Napolitano, 648 F. Supp. 2d 726, 729 (D. Md. 2009)). Each of those requirements has a “close nexus” to the ordinary hiring, firing, and management of labor. Kahn, 618 F.2d at 792. But none of those comes even close to the deployment of the Property Act to mandate a medical procedure for one-fifth (or more) of our workforce. Indeed, the government's preferred authority, Kahn, repeatedly stresses the narrowness of its decision to uphold the anti-discrimination order. See id. at 793 (“[O]ur decision today does not write a blank check for the President to fill in at his will.”); id. at 797 (Tamm, J., concurring) (“Lest we later be construed as having broadly interpreted the [Property] Act, I write separately only to emphasize my belief that the opinion we issue today is a narrow one. It does not allow the President to exercise powers that reach beyond the Act's express provisions.”).
It is telling that none of the history from 1949 to present supplied by the government involves the imposition of a medical procedure upon the federal-contractor workforce under the rationale of “reducing absenteeism.” The dearth of analogous historical examples is strong evidence that § 101 does not contain such a power. See In re MCP No. 165, OSHA Interim Final Rule: COVID-19 Vaccination & Testing, 20 F.4th 264, 284 (6th Cir. 2021) (Sutton, C.J., dissenting from denial of initial hearing en banc) (“A ‘lack of historical precedent’ tends to be the most ‘telling indication’ that no authority exists.”) (quoting Free Enter. Fund. v. Pub. Co. Acct. Oversight Bd., 561 U.S. 477, 505, 130 S.Ct. 3138, 177 L.Ed.2d 706 (2010))). After all, the threat of absenteeism is hardly unique to COVID-19. Many diseases, like influenza and the common cold, have inflicted absenteeism on federal contractors for the last seventy-two years. See, e.g., Matthew R. Groenewold et al., Health-Related Workplace Absenteeism Among Full-Time Workers—United States, 2017–2018 Influenza Season, CDC (July 5, 2019), cdc.gov/mmwr/volumes/68/wr/mm6826a1.htm (last visited Dec. 23, 2021) (noting that “absenteeism increased sharply” during the “2017–18 influenza season”). But the government has cited for us no instance in which the President invoked the Property Act to mandate that nearly all federal-contractor employees get a flu shot.
Moreover, if the President can order medical interventions in the name of reducing absenteeism, what is the logical stopping point of that power? Even vaccinated employees may contract the flu (or COVID-19) at family gatherings, concerts, sporting events, and so on. May the President, in the name of the Property Act, mandate that covered employees also wear masks in perpetuity at each of those events to reduce the chances of contracting an airborne communicable disease and later spreading it to coworkers, thus creating absenteeism? Such off-the-job conduct very well may threaten to cause on-the-job absenteeism. So why, if the government's interpretation is correct, does the Property Act not confer a de facto police power upon the President to dictate the terms and conditions of one-fifth of our workforce's lives? The government has never reckoned with the implications of its position or proposed any limiting principle to allay our concerns. And those points underscore just how inapposite are the government's historical examples—wage and price controls, union posters, confirmation of immigration status, and anti-discrimination in hiring. Each is a modest, “work-anchored” measure with an inbuilt limiting principle. In re MCP No. 165, 20 F.4th at 275–76 (Sutton, C.J., dissenting from denial of initial hearing en banc). The contractor mandate, by contrast, requires vaccination everywhere and all the time. It is not “anchored” to the statutory text, nor is it even “anchored” to the work of federal contractors.
Other tools of construction likewise undercut the government's view. Consider the “federalism canon”—the notion that Congress must use “exceedingly clear language if it wishes to significantly alter the balance between federal and state power.” Ala. Ass'n of Realtors, 141 S. Ct. at 2489. We obviously lack “exceedingly clear language” that the Property Act supports the contractor mandate, so we focus on why the contractor mandate would also “significantly alter the balance between federal and state power.” Id. Since the Framing, the power to regulate the public health has been “part and parcel” of states’ “traditional police power.” In re MCP No. 165, 20 F.4th at 275–76 (Bush, J., dissenting from denial of initial hearing en banc). Indeed, “the States, not the Federal Government, are the traditional source of authority over safety, health, and public welfare.” Id. at 273 (Sutton, C.J., dissenting from denial of initial hearing en banc).
The Supreme Court has recognized this principle time after time. See Jacobson v. Massachusetts, 197 U.S. 11, 25, 25 S.Ct. 358, 49 L.Ed. 643 (1905) (calling it a “settled principle[ ]” that states enjoy a police power to promulgate “legislative enactment[s to] protect the public health and the public safety”); Chicago, B. & Q. Ry. Co. v. Illinois, 200 U.S. 561, 592, 26 S.Ct. 341, 50 L.Ed. 596 (1906) (“[T]he police power of a State embraces ․ regulations designed to promote the public health.”); Berman v. Parker, 348 U.S. 26, 32, 75 S.Ct. 98, 99 L.Ed. 27 (1954) (describing regulation of “public health” as a “traditional application of the police power”). The Court has also reiterated this point twice “in the specific context of compulsory vaccination.” In re MCP No. 165, 20 F.4th at 287 (Bush, J., dissenting from denial of initial hearing en banc) (citing Jacobson, 197 U.S. at 24–25, 25 S.Ct. 358; Zucht v. King, 260 U.S. 174, 176, 43 S.Ct. 24, 67 L.Ed. 194 (1922)); see also Zucht, 260 U.S. at 176, 43 S.Ct. 24 (describing it as “within the police power of a State to provide for compulsory vaccination” (citing Jacobson, 197 U.S. at 24–25, 25 S.Ct. 358)).
What the contractor mandate seeks to do, in effect, is to transfer this traditional prerogative from the states to the federal government under the guise of a measure to make federal contracting more “economical and efficient.” But see Tiger Lily, LLC, 5 F.4th at 671 (declining to permit an agency “to interpret a statute to push the limit of congressional authority ․ when ‘the administrative interpretation [would] alter[ ] the federal-state framework by permitting federal encroachment upon a traditional state power’ ”) (quoting Solid Waste Agency of N. Cook Cnty. v. U.S. Army Corps of Eng'rs, 531 U.S. 159, 172–73, 121 S.Ct. 675, 148 L.Ed.2d 576 (2001))). The government protests that “federal contracts are not an area traditionally reserved to the states,” and thus, apparently, that states may not complain when the federal government uses contracting as a naked pretext 15 to invade traditional state prerogatives. 16 See Gov't Mot. for Stay at 18. In making such an argument, the government frames the issue at the wrong level of generality. States may have no power to dictate what and how much of something the federal government may buy. See, e.g., Perkins v. Lukens Steel Co., 310 U.S. 113, 127, 60 S.Ct. 869, 84 L.Ed. 1108 (1940). But they certainly have a traditional interest in regulating public health and, specifically, in determining whether to impose compulsory vaccination on the public at large. See In re MCP No. 165, 20 F.4th at 273 (Sutton, C.J., dissenting from denial of initial hearing en banc) (“[T]he only Supreme Court cases that permitted a government to impose a vaccination mandate on individuals arose from the States, not the National Government.”) (citing Jacobson, 197 U.S. at 11, 25 S.Ct. 358; Zucht, 260 U.S. at 174, 43 S.Ct. 24)). And, by extension, they may validly complain when the federal government seeks to usurp those roles by doing something that it has no traditional prerogative to do—deploy the Property Act to mandate an irreversible medical procedure. 17
We thus conclude that the federal government is unlikely to prevail on its argument that the Property Act authorizes imposition of the contractor mandate. 18 We turn now to the remaining factors relevant to the stay—irreparable injury, harms to the non-movants, and the public interest.
Factor 2: Whether the Federal Government Will Suffer Irreparable Injury Absent a Stay
The government protests that absent a stay, it will incur the irreparable injuries of absenteeism and productivity loss. Gov't Mot. for Stay at 19–20. We are not persuaded by the government's claims of “irreparable injury” for several reasons. First, COVID-19 vaccines have been generally available in the United States—and the present administration has been in office—for nearly a year. Yet the contractor mandate did not emerge until September 2021. See Guidance, supra, at 1. The government then delayed the compliance deadline from December 8, 2021, to January 18, 2022. See 86 Fed. Reg. at 63,423. The government's actions undercut its representations of great urgency in implementation of the contractor mandate.
Second, we note the additional tension between the government's conduct regarding the contractor mandate versus OSHA's vaccine-or-mask standard. The government claims that implementing the contractor mandate's vaccine requirement is crucial to avoiding irreparable injury. Yet it has delayed formal enforcement of OSHA's similar standard until February 9, 2022, despite our Court recently dissolving the Fifth Circuit's stay of its enforcement. See Dep't of Labor, Statement from the U.S. Department of Labor on the 6th Circuit Court of Appeals Dissolving the Stay of OSHA Emergency Temporary Standard on Vaccination and Testing (Dec. 18, 2021), https://www.dol.gov/newsroom/releases/osha/osha20211218 (last visited Dec. 22, 2021) (explaining that “[t]o account for any uncertainty created by the stay, OSHA is exercising enforcement discretion” to decline to “issue citations for noncompliance with the standard's testing requirements before February 9” so long as employers take reasonable steps to comply on their own). The government has failed to explain why it must immediately implement the contractor mandate to avoid irreparable injury yet considers it permissible to voluntarily delay enforcement of OSHA's laxer vaccines-or-mask requirement, which regulates basically the same conduct, for a month and a half.
Last, we note from a practical perspective that the contractor mandate is already subject to a nationwide injunction out of the Southern District of Georgia that the Eleventh Circuit recently declined to stay. See Georgia v. Biden, No. 1:21-cv-163, ––– F. Supp. 3d ––––, 2021 WL 5779939 (S.D.Ga.), motion to stay preliminary injunction denied, No. 21-14269 (11th Cir. Dec. 17, 2021) (holding that the government failed to establish that it would be “irreparably injured absent a stay”). The government is correct that the Eleventh Circuit's decision does not affect the reviewability of the underlying decree in this case, but it does make our decision here somewhat academic. See Gov't Reply at 11. For even if we thought the district court's injunction an abuse of discretion, our dissolution of it could not revive the contractor mandate and prevent the government's allegedly irreparable injuries. We thus conclude that the government has not made a strong showing on the second stay factor.
Factor 3: Whether a Stay Would Harm the States and Sheriffs’ Offices
A stay of the injunction, by contrast, would harm the plaintiff states and sheriffs’ offices. Two discernible theories of injury are relevant. First, the contractor mandate will deter the plaintiffs from bidding on or renewing covered contracts that they otherwise would have bid on or renewed but for the contractor mandate. A stay of the injunction pending appeal would thus deter plaintiffs from entering into economically valuable federal contracts. 19 And second, if the contractor mandate were to become enforceable during the appeal, covered employees who chose to comply with the mandate rather than lose their jobs would incur the irrecoverable compliance cost of a coerced vaccination that could not be reversed if the contractor mandate were later held invalid. See BST Holdings, L.L.C. v. OSHA, 17 F.4th 604, 618 (5th Cir. 2021) (quoting Texas v. EPA, 829 F.3d 405, 433 (5th Cir. 2016)).
Factor 4: Where the Public Interest Lies
The last stay factor, the public interest, is equivocal at best. The federal government claims that the public interest lies in increasing vaccinations, reducing absenteeism, and decreasing hospital visits from COVID-19. Those are valuable goals no doubt. But the states raise countervailing concerns that suggest denial of a stay is in the public interest. For instance, despite the government's asserted interest in stable supply-chains, the contractor mandate itself may engender serious resistance and thus serious economic disruption. Employees of contractors who choose to comply rather than resist may be compelled to submit to a potentially illegal mandate and suffer irrecoverable compliance costs. And as we have explained before, the public's true interest lies in the correct application of the law. See Coal. to Def. Affirmative Action v. Granholm, 473 F.3d 237, 252 (6th Cir. 2006) (“ ‘[T]he public interest lies in the correct application’ of the federal constitutional and statutory provisions upon which the claimants have brought this claim[.]” (citation omitted)). In any event, the public-interest factor does not sufficiently outweigh all the other defects with the government's case to warrant a stay.
The government has not made the “strong showing” required to justify the grant of a stay. Nken, 556 U.S. at 426, 129 S.Ct. 1749. Most concerning, the Property Act likely confers no authority upon the President to order the imposition of the contractor mandate. For that reason and the others explained above, we DENY the government's requested stay. 20
CONCURRING IN PART AND DISSENTING IN PART
I disagree with the majority's conclusion that both the states and the sheriffs’ offices have standing. I also disagree with the conclusion that the President “re-envisioned” the Federal Property and Administrative Services Act (“Property Act”) to take the actions contemplated by Executive Order No. 14042. Maj. Op. 589. I recognize that the Eleventh Circuit recently declined to stay the national injunction imposed by Georgia v. Biden, ––– F. Supp. 3d ––––, No. 1:21-CV-163, 2021 WL 5779939, (S.D. Ga. Dec. 7, 2021). See Georgia v. Biden, No. 21-14269, slip op. at 1 (11th Cir. Dec. 17, 2021). Even still, I find that the government has made a “strong showing” in this case that it will prevail on the merits and has established that it will suffer irreparable harm without a stay. See Nken v. Holder, 556 U.S. 418, 426, 129 S.Ct. 1749, 173 L.Ed.2d 550 (2009). For these reasons, I dissent.
As to standing, because “the Plaintiffs did not provide an example of a new contract that is subject to the mandate,” they lack standing. Kentucky v. Biden, No. 3:21-CV-00055-GFVT, ––– F.Supp.3d ––––, ––––, 2021 WL 5587446, at *3 (E.D. Ky. Nov. 30, 2021) (“Kentucky”). While the majority notes that the government has requested modification of at least one contract, there is no harm in asking that a contract be modified. For a modification to go into effect, the agreement must be bilateral—that is, both the government and the contracting party must agree to it. For contracts that are subject to renewal, the contract modification language could be different by the renewal date, like any other condition to the contract. Any purported reliance interest is not sufficient to constitute a harm here, any more than it is when Congress leverages the power of the purse to encourage states to comply with federal law. See South Dakota v. Dole, 483 U.S. 203, 207–08, 107 S.Ct. 2793, 97 L.Ed.2d 171 (1987). As to future contracts, at no point do the states provide any legal support for their contention that failure to bid on a contract or failure to receive federal funds is a cognizable harm. It is a potential contracting party's choice not to contract with the federal government—the government's mandate does not prevent states from bidding or otherwise contracting. See United States v. Wunderlich, 342 U.S. 98, 156, 72 S.Ct. 154, 96 L.Ed. 113 (1951) (noting that federal contractors are “not compelled or coerced into” contracting with the government).
Nor do the states have standing in a sovereign or quasi-sovereign capacity. The majority contends that the states have “plausibly shown that the contractor mandates threatens to damage ․ [their] economies” due to individual employees’ refusal to comply with the mandate. Maj. Op. 599–600. Any contention that the parties will be harmed by failing to comply with the mandate, however, is wholly speculative. According to the Centers for Disease Control and Prevention (“CDC”), 85.5% of individuals over the age of 18 have received at least one dose of the vaccine, while 72.8% are fully vaccinated. See CDC, COVID-19 Vaccination in the United States, COVID Data Tracker, (Jan. 3, 2022, 6:00 AM), https://covid.cdc.gov/covid-data-tracker/#vaccinations_vacc-total-admin-rate-total. If federal contractors constitute one fifth of the American workforce, Kentucky, ––– F.Supp.3d at ––––, 2021 WL 5587446, at *1, the executive order affects, at most, five percent of the workforce—a much smaller scale than the majority and the states implicate. The states have also failed to provide evidence of what percentage of their workforce are federal contractors, so the number is likely exaggerated further. In addition, while 70% of unvaccinated workers say they would leave their job to avoid a vaccine requirement, in practice only five percent of workers have done so. KFF, KFF Covid-19 Vaccine Monitor: October 2021, (Oct. 28, 2021), https://www.kff.org/coronavirus-covid-19/poll-finding/kff-covid-19-vaccine-monitor-october-2021/. Therefore, the evidence indicates that most contractors are either already vaccinated or would choose to get vaccinated rather than quit their jobs. Thus, given that there is no harm, neither the states nor the sheriffs’ offices have standing in their proprietary capacities.
Further, nothing in the contractor mandate threatens to override state policies, nor is the federal government intruding upon an area traditionally left up to the states. This guidance solely applies to future or bilaterally modified federal contracts—an area of governance that has never been, nor could be, left to the states. To the extent that states seek to vindicate the interests of the alleged persons who would leave their job rather than be vaccinated, this is—as the majority notes—impermissible litigation on behalf of third parties. See Massachusetts v. Mellon, 262 U.S. 447, 485–86, 43 S.Ct. 597, 67 L.Ed. 1078 (1923). Without standing, the lawsuit cannot proceed, and the injunction should not have been issued.
Even if the plaintiffs had standing, a stay would be appropriate. As a reminder, we must consider: “(1) whether the stay applicant has made a strong showing that he is likely to succeed on the merits; (2) whether the applicant will be irreparably injured absent a stay; (3) whether issuance of the stay will substantially injure the other parties interested in the proceeding; and (4) where the public interest lies.” Nken, 556 U.S. at 426, 129 S.Ct. 1749 (quoting Hilton v. Braunskill, 481 U.S. 770, 776, 107 S.Ct. 2113, 95 L.Ed.2d 724 (1987)).
The government has demonstrated that it is likely to succeed on the merits. The Property Act is a procurement statute. It was enacted “to provide the Federal Government with an economical and efficient system for the following activities: (1) [p]rocuring and supplying property and nonpersonal services, and performing related functions ․ [,] (2) [u]sing available property[,] (3) [d]isposing of surplus property[, and] (4) [r]ecords management.” 40 U.S.C. § 101. The Property Act further provides that “[t]he President may prescribe policies and directives that the President considers necessary to carry out this subtitle[,]” so long as the policies are “consistent” with it. 40 U.S.C. § 121(a). “ ‘Economy’ and ‘efficiency’ are not narrow terms; they encompass those factors like price, quality, suitability, and availability of goods or services that are involved in all acquisition decisions.” Am. Fed'n of Lab. & Cong. of Indus. Orgs. v. Kahn, 618 F.2d 784, 789 (D.C. Cir. 1979). Courts have interpreted the act to encompass policies and directives that have a “sufficiently close nexus to the values of providing the government an economical and efficient system for ․ procurement and supply.” UAW-Lab. Emp. & Training Corp. v. Chao, 325 F.3d 360, 366 (D.C. Cir. 2003) (citation and quotations omitted).
Courts have recognized that the Property Act gives the President “necessary flexibility and broad-ranging authority.” Id. (quotations omitted). Congress clearly intended to grant the President “direct and broad-ranging authority over those larger administrative and management issues that involve the Government as a whole.” Kahn, 618 F.2d at 789. Congress also intended for that authority to “be used in order to achieve a flexible management system capable of making sophisticated judgments in pursuit of economy and efficiency.” Id. Accordingly, the President and Congress have “frequently imposed on the procurement process social and economic programs somewhat removed from a strict view of efficiency and economy.” Id. at 789–90. For this reason, courts have found a sufficiently close nexus between presidential action and economical procurement systems even when the connection is attenuated, or where arguments that the action arguably impairs economic interests could be advanced. Chao, 325 F.3d at 366–67.
Given this history, Executive Order 14042 is consistent with the provisions of the Property Act. The express language of the Order promotes economy and efficiency in federal contracting by ensuring federal contractors implement adequate COVID-19 safeguards to protect their workers and reduce the spread of COVID-19, “which will decrease worker absence, reduce labor costs, and improve the efficiency of contractors and subcontractors at sites where they are performing work for the Federal Government.” Exec. Order No. 14,042, 86 Fed. Reg. 50,985 (Sept. 9, 2021). That the first goal—decreasing the spread of COVID-19—happens to also be a valid public health objective does not mean that it intrudes upon state liberties or exceeds the President's authority under the Property Act. See Kahn, 618 F.2d at 790 (executive order requiring federal contractors to meet antidiscrimination provisions helps establish an economical system for procurement). The health and safety of the government's workforce amid a global and worsening pandemic has direct and tangible effects on the economy, and, by extension, on the government's ability to procure and supply services. The fact that vaccination mandates—whether for COVID-19, influenza, or other, vaccine-preventable diseases—are commonplace and mandated in workplaces and schools around the county plainly demonstrates that such mandates have a “close nexus” to the ordinary hiring, firing, and management of labor. To find otherwise prioritizes a state's right to implement a hypothetical public health measure over the federal government's right to control the terms and conditions of its contracts. Put simply, it prioritizes a hypothetical individual's decision to work as an unvaccinated federal contractor over the federal government's right to control “administrative and management issues.” Kahn, 618 F.2d at 789.
Although the majority contends that the Order is merely pretext to increase vaccination rates, it relies not on the language of the Order, but rather on the words of the executive branch in doing so. Courts have been historically reluctant “to consider the President's motivation[s] in issuing [an] Executive Order.” Chamber of Com. of U.S. v. Reich, 74 F.3d 1322, 1335 (D.C. Cir. 1996); see also Trump v. Hawaii, ––– U.S. ––––, 138 S. Ct. 2392, 2417–19, 201 L.Ed.2d 775 (2018). And with good reason: allowing “political opponents of executive actions to generate controversy with accusations of pretext, deceit, and illicit motives ․ could lead judicial review of administrative proceedings to devolve into an endless morass of discovery and policy disputes[.]” Dep't of Com. v. New York, ––– U.S. ––––, 139 S. Ct. 2551, 2575, 204 L.Ed.2d 978 (2019) (Thomas, J., concurring in part and dissenting in part). As the Supreme Court has repeatedly stressed, outside of a narrow line of animus cases, see Romer v. Evans, 517 U.S. 620, 632, 635, 116 S.Ct. 1620, 134 L.Ed.2d 855 (1996), what matters is the lawful scope of a president's authority, not the statements they make. Trump, 138 S. Ct. at 2418; see also id. at 2424 (“[T]he statements ․ of Government officials are not subject to judicial scrutiny[.]”) (Kennedy, J., concurring). The challenged executive order easily falls within the scope of President Biden's authority.
As to the remaining factors, the federal government has clearly demonstrated irreparable harm in the form of significant productivity losses not only from leave and health care costs for workers who are sick, quarantined, and unable to perform due to COVID-19, but also scheduling delays and reduced performance quality—by its estimate approximately two billion dollars per month that the injunction is in place. In contrast, the states have not identified any irreparable harm. Ordinary compliance costs are not “irreparable”—they are “fully compensable by monetary damages.” Overstreet v. Lexington-Fayette Urb. Cnty. Gov't, 305 F.3d 566, 578 (6th Cir. 2002) (citing Basicomputer Corp. v. Scott, 973 F.2d 507, 511 (6th Cir. 1992)); see also Wilson ex rel. Est. of Wilson v. United States, 405 F.3d 1002, 1009 (Fed. Cir. 2005) (“A claim may be asserted under the Tucker Act [28 U.S.C. § 1491(a)(1)] ‘for recovery of monies that the government has required to be paid contrary to law.’ ” (quoting Aerolineas Argentinas v. United States, 77 F.3d 1564, 1572 (Fed. Cir. 1996))); 28 U.S.C. § 1491(b)(1)–(2) (permitting bid protests where an “interested party” objects to “any alleged violation of a statute or regulation in connection with a procurement or a proposed procurement” provided that monetary relief is “limited to bid preparation and proposal costs”). Any future federal contractors subject to the executive order will be aware of the vaccine requirement before bidding, thereby agreeing to be vaccinated—or provide a valid exemption—by soliciting and entering the contract.
Contrary to the states’ speculative contention, which the majority adopts, there is no evidence that contractors will leave their positions to avoid complying with the mandate. The Government does not “assume that all the employees subject to the contractor mandate are continuously ‘contracting’ under § 101.” Maj. Op. 605. The mandate explicitly only applies to new contracts or bilateral modifications. Therefore, the only way any current federal contractor would become subject to the mandate is if an employer agreed to the bilateral modification—and the states and the sheriffs’ offices are under no obligation to do. There is also no evidence that those who leave will disrupt workplace operations, because the actual scope of the mandate is smaller than it seems, as discussed previously.
Because “[t]he first two factors of the traditional standard are the most critical” when assessing whether to stay a court's order, Nken, 556 U.S. at 434, 129 S.Ct. 1749, I would grant the government's motion to stay the district court's injunction pending appeal.
Accordingly, I dissent.
1. We call this directive the “contractor mandate.”
2. We refer to the federal-government defendants collectively as “the government.”
3. This statute simply permits the President to delegate statutory authority that he already possesses to his agents within the executive branch. See 3 U.S.C. § 301. It is not an independent grant of authority to the President. Id.
4. The notice of determination contains a single sentence of analysis, which we reproduce here in full: “Based on my review of the Safer Federal Workforce Task Force's COVID-19 Workplace Safety: Guidance for Federal Contractors and Subcontractors, scheduled for issuance on September 24, 2021, and exercising the President's authority under the Federal Property and Administrative Services Act (see 3 U.S.C. 301I) [sic] delegated to me through Executive Order No. 14042, I have determined that compliance by Federal contractors and subcontractors with the COVID-19 workplace safety protocols detailed in that guidance will improve economy and efficiency by reducing absenteeism and decreasing labor costs for contractors and subcontractors working on or in connection with a Federal Government contract.” 86 Fed. Reg. at 53,692.
5. In the meantime, the Federal Acquisition Regulatory Council promulgated a model “deviation clause”—language to be inserted into federal contracts to make them consistent with the Safer Federal Workforce Task Force Guidance. See Amended Complaint ¶ 113, R. 22.
6. We reserve the constitutional standing analysis—injury-in-fact, causation, and redressability—until we have defined the plaintiffs’ relevant interests. Until we have defined those interests (that is, the sovereign, quasi-sovereign, and proprietary interests at stake), we cannot analyze whether those interests have suffered a redressable injury-in-fact caused by the defendants.
7. The government and the dissent complain that the state plaintiffs have not introduced specific contracts into the record that will become subject to the contractor mandate. Gov't Mot. for Stay at 8. This argument is unpersuasive. The complaint is rife with well-pleaded allegations that the state plaintiffs and their state agencies contract with multiple federal agencies, including the United States Department of Justice, the United States Marshals Service, the Food and Drug Administration, the National Institutes of Health, the Bureau of Labor Statistics, the National Aeronautics and Space Administration, the Consumer Product Safety Commission, and the Centers for Disease Control and Prevention. See Amended Complaint ¶¶ 4, 16–17, R. 22. The plaintiffs also bolstered those allegations with supporting evidence, as was required to obtain a preliminary injunction. See Niknejad Dec., R. 12-1; Maddox Dec., R. 12-4; Flowers Dec., R. 22-2; see also Leary v. Daeschner, 228 F.3d 729, 739 (6th Cir. 2000). And the Guidance is written so broadly that it would obviously apply to these contracts, whether via modifications, renewals, options, or if the states should pursue additional contracts with these agencies. See Guidance, supra, at 3–5.
8. The CDC statistic judicially noticed by the dissent, see Dissenting Op. at 613–14, does not contradict the plaintiff states’ plausible showing of economic harm. We may take judicial notice of generally known information or government websites, see Broom v. Shoop, 963 F.3d 500, 509 (6th Cir. 2020), but judicially noticed data should at least be particular to the issue at hand. On a national basis for the entire population, the CDC statistic reflects that “85.5% of individuals over the age of 18 have received at least one dose of the vaccine, while 72.8%”—about three-quarters—“are fully vaccinated.” Dissenting Op. at 613. Or, said differently, about 27.2%, or one-quarter, are not fully vaccinated. The dissent then assumes that (1) the vaccination rate of the overall population describes the vaccination rate of the national workforce, and (2) because federal contractors are about one-fifth of the national workforce, only 27.2% must not be fully vaccinated. That is how the dissent arrives at its “five percent” figure, since about one-quarter of twenty percent of the national workforce is also about five percent of the national workforce. But this string of questionable assumptions relies on a CDC statistic that does not even directly address the vaccination rate in the workforce of the plaintiff states. For one, the CDC's national figure includes states with highly vaccinated populations (for instance, New York) despite the vaccination rate in those states being much higher than in Ohio, Kentucky, and Tennessee. See, e.g., New York State COVID-19 Vaccine Tracker: Vaccination Progress to Date, N.Y. State Dep't of Health (Jan. 3, 2022), https://perma.cc/Z4L4-RE5X?type=image (explaining that 80.6% of New York's over-18 population is fully vaccinated). Additionally, the CDC statistic includes all Americans aged 65 years old and older, who, as a group, are very highly vaccinated but who are also much less likely to be in the workforce. To accurately discuss the potential effects of the contractor mandate in the plaintiff states, we need data tailored to both the plaintiff states and the working-age populations within them. It turns out that such data show that vaccination rates are far lower for those populations than the CDC's “72.8%” statistic would suggest. In Tennessee, for instance, only 42.6% of the 21–30 population, 51.5% of the 31–40 population, 57.6% of the 41–50 population, and 64.4% of the 51–60 population are fully vaccinated. See Vaccination Reporting, Tenn. Dep't of Health (Jan. 2, 2022), https://www.tn.gov/health/cedep/ncov/covid-19-vaccine.html (last visited Jan. 3, 2022). Likewise, in Kentucky, only 40% of the 18–24 population, 46% of the 25–39 population, 55% of the 40–49 population, and 65% of the 50–64 population are fully vaccinated. See Kentucky COVID-19 Vaccination Dashboard, Ky. Cabinet for Health & Family Servs. (Jan. 3, 2022), https://dashboard.chfs.ky.gov/views/KYPublicFacingDashboard_16191000580170/KentuckyCOVID-19Vaccination?%20% 3A%20iid=1&%20% 3A%20isGuestRedirectFromVizportal=y&%20% 3A%20embed=y (last visited Jan. 3, 2022). Last, in Ohio, only 48.32% of the 20–29 population, 55.41% of the 30–39 population, 60.96% of the 40–49 population, and 67.62% of the 50–59 population are fully vaccinated. COVID-19 Dashboard, Ohio Dep't of Health, https://perma.cc/3V99-2H5X?type=image. Concededly, 76.29% of Ohio's 60–64 population is fully vaccinated. Id. So, while that lone sub-population exceeds the CDC's national-average figure, every other working-age population within the plaintiff states falls below it—in some cases far below. We also note that the CDC, at least for now, still defines “fully vaccinated” as being two weeks post-second dose in a two-shot series or two weeks post dose in a single-dose vaccine. COVID-19 Vaccine Booster Shots, CDC, https://www.cdc.gov/coronavirus/2019-ncov/vaccines/booster-shot.html (last visited Jan. 3, 2022). But the definition of “fully vaccinated” is highly likely to change to include both the initial doses and a booster shot, as new variants of COVID-19 continue to spread. See, e.g., Nathaniel Weixel, Fauci: It's ‘when, not if’ definition of fully vaccinated will change, The Hill (Dec. 8, 2021), https://thehill.com/policy/healthcare/584943-fauci-when-not-if-definition-of-fully-vaccinated-will-change?rl=1 (last visited Jan. 3, 2022); see also Emily Anthes & Noah Weiland, As Omicron Spreads, Officials Ponder What It Means to Be ‘Fully Vaccinated,’ New York Times (Dec. 29, 2021), https://www.nytimes.com/2021/12/29/health/covid-vaccinations-boosters.html (last visited Jan. 3, 2022). E.O. 14042 and the deviation clauses implementing the contractor mandate implicitly anticipate that this definition could change, given that they simply incorporate whatever definition is supplied by the CDC, which is itself subject to change. And to the extent they are available, data from the plaintiff states suggest that very little of the working-age population has received a booster shot. In Kentucky, for instance, only 6% of the 18–24 population, 11% of the 25–39 population, 17% of the 40–49 population, and 27% of the 50–64 population have received boosters. See Kentucky COVID-19 Vaccination Dashboard, supra. So, contrary to what the dissent implies, implementation of the contractor mandate would be far from a non-event in the plaintiff states.
9. It is not clear from the government's briefing that it even disputes the existence of a cause of action, at least insofar as the plaintiffs are found to have standing.
10. Specifically, § 702 explains that “person” carries the definition provided in 5 U.S.C. § 551, i.e., “an individual, partnership, corporation, association, or public or private organization other than an agency.” This language likely includes a sheriff's office as well.
11. The term “nonpersonal services” refers to services that the federal government procures from individuals who are employees of a federal contractor with which the government has a contract, but who are not themselves employees of the federal government. Compare 48 C.F.R. § 37.104(a) (“A personal services contract is characterized by the employer-employee relationship it creates between the Government and the contractor's personnel.”), with 48 C.F.R. § 37.101 (“Nonpersonal services contract means a contract under which the personnel rendering the services are not subject, either by the contract's terms or by the manner of its administration, to the supervision and control usually prevailing in relationships between the Government and its employees.”). This distinction further undercuts the government's position, given that the reference in § 101 to “nonpersonal services” implies the federal government's lack of the heightened degree of “supervision and control” it might exercise over its own employees.
12. The dissent references this observation to claim that we misunderstand the scope of the Guidance, since the Guidance “only applies to new contracts or bilateral modifications” rather than to unmodified existing contracts. See Dissenting Op. at 616. Yet here we are interpreting § 101 of the Property Act—not the Guidance. The relevant statutory-interpretation question is not which contracts the Guidance affects, but why the President or his agents have any power under the statute to impose medical procedures on federal contractors. Whether they are working pursuant to existing contracts, bilaterally modified contracts, or wholly new contracts does not matter to that inquiry. Even in the context of wholly new contracts, contractors performing such contracts still are not “contracting” within the meaning of § 101. As explained above, that term refers to a function of the government—not of contractor employees—and specifically a governmental function of entering contracts—not contractors’ subsequent performance of them. The dissent never addresses that point other than by reciting the same handful of non-binding cases that we later distinguish at length. See, e.g., Dissenting Op. at 614–15 (citing Kahn, 618 F.2d at 789).
13. It is by now axiomatic that “legislative history is not the law” and that it cannot trump unambiguous statutory text. Azar v. Allina Health Servs., ––– U.S. ––––, 139 S. Ct. 1804, 1814, 204 L.Ed.2d 139 (2019) (citing Epic Sys. Corp. v. Lewis, ––– U.S. ––––, 138 S. Ct. 1612, 1631, 200 L.Ed.2d 889 (2018)). We include our brief discussion of the Senate Report in a way that Justice Scalia once advocated—to show that not only does the text preclude the government's fanciful construction but that, indeed, the text's progenitors likewise did not understand themselves to be encoding within it such a novel and sweeping power. See Green v. Bock Laundry Mach. Co., 490 U.S. 504, 527, 109 S.Ct. 1981, 104 L.Ed.2d 557 (1989) (Scalia, J., concurring in the judgment) (“I think it entirely appropriate to consult all public materials, including the background of Rule 609(a)(1) and the legislative history of its adoption, to verify that what seems to us an unthinkable disposition ․ was indeed unthought of[.]”).
14. We thus disagree with the district court that the Property Act likely presents non-delegation concerns. Those might arise if the Property Act had “merely announce[d] vague aspirations” and then gave “the executive carte blanche” to do whatever the President saw fit. Gundy, 139 S. Ct. at 2133, 2144 (Gorsuch, J., dissenting). The Property Act instead grants the President specific, enumerated powers to achieve specific, enumerated goals in administering the federal procurement system. That the district court raised non-delegation concerns, however, is understandable. If the government's interpretation were correct—that the President can do essentially whatever he wants so long as he determines it necessary to make federal contractors more “economical and efficient”—then that certainly would present non-delegation concerns.
15. The federal government's actions are, of course, simply a pretext to increase vaccination, as its own documents confirm. See, e.g., Off. Fed. Procurement Pol'y, Memorandum for Chief Acquisition Officers 3 (Sept. 30, 2021) (“To maximize the goal of getting more people vaccinated”—rather than to enhance the goal of efficient procurement—“the Task Force strongly encourages agencies to apply the requirements of its guidance broadly[.]”) (emphasis added); see Dep't of Com. v. New York, ––– U.S. ––––, 139 S. Ct. 2551, 2575, 204 L.Ed.2d 978 (2019) (“[W]e are ‘not required to exhibit a naiveté from which ordinary citizens are free.’ ” (quoting United States v. Stanchich, 550 F.2d 1294, 1300 (2d Cir. 1977))).
16. The dissent faults us for examining “the words of the executive branch” to discern that branch's motivations, relying primarily on a partial dissent from a Supreme Court opinion in which a majority actually endorsed such an inquiry. See Dissenting Op. at 615–16 (citing Dep't of Commerce, 139 S. Ct. at 2576 (Thomas, J., concurring in part and dissenting in part)). But see Dep't of Commerce, 139 S. Ct. at 2575–76; id. at 2584 (Breyer, J., concurring in part and dissenting in part, joined by Ginsburg, J., Sotomayor, J., and Kagan, J.). Dissents may state relevant points but, of course, do not bind us, especially when a clear majority of the Supreme Court has adopted a different rule of decision. In any event, the pretextual nature of the contractor mandate is not outcome-determinative here, and nowhere do we claim that it is. As the dissent acknowledges, what really matters “is the lawful scope of [the] president's authority.” Dissenting Op. at 615. And so, because the “lawful scope” of his authority under the Property Act likely does not include the contractor mandate, we deny the government's requested stay.
17. The dissent claims that the contractor mandate has no federalism implications because it does not “intrud[e] upon an area traditionally left to the states”—an area that it narrowly defines as federal contracts. Dissenting Op. at 613–14. The dissent can only frame the mandate so narrowly by simply ignoring its real-world effects, which would include a de facto authority to dictate public health measures for sizeable portions of the plaintiff states’ populations. Yet as the Supreme Court recently reminded us, agencies cannot skirt the federalism implications of their actions by pretending that “decades-old statute[s]” somehow “indirectly” grant them novel powers to intrude into “particular domain[s] of state law.” Ala. Ass'n of Realtors, 141 S. Ct. at 2486, 2488–89.
18. Having concluded that the President likely lacked statutory authority to promulgate the contractor mandate, we decline to consider whether the contractor mandate is also arbitrary and capricious or violates the Competition in Contracting Act. See 41 U.S.C. § 253.
19. The dissent claims that the federal government has “clearly demonstrated irreparable harm” with its assertion that, given sundry “productivity losses,” it will lose “approximately two billion dollars per month that the injunction is in place.” See Dissenting Op. at 616. But the dissent then says that the monetary losses the states will incur from compliance with the mandate are “not ‘irreparable’ ” because “they are ‘fully compensable by monetary damages.’ ” Id. (quoting Overstreet v. Lexington-Fayette Urb. Cnty. Gov't, 305 F.3d 566, 578 (6th Cir. 2002)). This line of reasoning is flawed on at least four levels. First, whatever one's view of the merits of the contractor mandate, it simply cannot be the case that monetary losses are reparable when suffered by the states, yet somehow become “irreparable” when suffered by the federal government. Second, who exactly will the states sue to obtain such “monetary damages”? Certainly not the federal government, at least to the extent the states continue to rely on the APA's cause of action, since the APA does not waive federal sovereign immunity from money-damages claims. See Haines v. Fed. Motor Carrier Safety Admin., 814 F.3d 417, 425 (6th Cir. 2016). The dissent appears to envision the states bringing an independent action under the Tucker Act, 28 U.S.C. § 1491, in which they attempt to quantify the economic harm the contractor mandate imposes and then seek to recover concomitant damages. But the Tucker Act itself creates no cause of action, so under what source of law would the states even sue? See Bowen v. Massachusetts, 487 U.S. 879, 905 n.42, 108 S.Ct. 2722, 101 L.Ed.2d 749 (1988). Third, even if the dissent had a solution to those problems, some of the intangible harms asserted here—invasions of state sovereignty and coerced compliance with irreversible vaccinations—likely cannot be economically quantified, and thus cannot be monetarily redressed. See id. (questioning whether a “naked money judgment against the United States” can be an “adequate substitute for prospective relief” given the “complex ongoing relationship” between states and the federal government). And fourth, the dissent subtly lessens the burden upon the federal government to obtain a stay by misstating the third stay factor. See Dissenting Op. at 615–16. The government is required to disprove that imposition of a stay would injure the states substantially—not irreparably. See Nken, 556 U.S. at 426, 129 S.Ct. 1749. That is a more difficult showing, since it requires disproving a broader class of harms. In other words, a harm to the states from the stay that is substantial, even if potentially compensable and thus technically reparable, still suffices to defeat the government's stay request.
20. The separate motion to file an amicus brief on behalf of the American Medical Association and fourteen other organizations is GRANTED.
JOHN K. BUSH, Circuit Judge.
BUSH, J., delivered the opinion of the court in which SUHRHEINRICH, J., joined, and COLE, J., joined in part. COLE, J. (pp. 613–16), delivered a separate dissenting opinion.