SCOTUSfiles

SUPREME COURT OF THE UNITED STATES

_________________

No. 22–58

_________________

UNITED STATES, et al., PETITIONERS v. TEXAS, et al.

on writ of certiorari to the united states court of appeals for the fifth circuit

 Justice Alito, dissenting.

 The Court holds Texas lacks standing to challenge a federal policy that inflicts substantial harm on the State and its residents by releasing illegal aliens with criminal convictions for serious crimes. In order to reach this conclusion, the Court brushes aside a major precedent that directly controls the standing question, refuses to apply our established test for standing, disregards factual findings made by the District Court after a trial, and holds that the only limit on the power of a President to disobey a law like the important provision at issue is Congress’s power to employ the weapons of inter-branch warfare—withholding funds, impeachment and removal, etc. I would not blaze this unfortunate trail. I would simply apply settled law, which leads ineluctably to the conclusion that Texas has standing.

 This Court has long applied a three-part test to determine whether a plaintiff has standing to sue. Under that test, a plaintiff must plead and ultimately prove that it has been subjected to or imminently faces an injury that is: (1) “concrete and particularized,” (2) “fairly traceable to the challenged action,” and (3) “likely” to be “redressed by a favorable decision.” Lujan v. Defenders of Wildlife, 504 U. S. 555, 560–561 (1992) (internal quotation marks and alterations omitted). Under that familiar test, Texas clearly has  standing to bring this suit.1

 Nevertheless, the United States (the defendant in this case) has urged us to put this framework aside and adopt a striking new rule. At argument, the Solicitor General was asked whether it is the position of the United States that the Constitution does not allow any party to challenge a President’s decision not to enforce laws he does not like. What would happen, the Solicitor General was asked, if a President chose not to enforce the environmental laws or the labor laws? Would the Constitution bar an injured party from bringing suit? She responded:

 “That’s correct under this Court’s precedent, but the framers intended political checks in that circumstance. You know, if—if an administration did something that extreme and said we’re just not going to enforce the law at all, then the President would be held to account by the voters, and Congress has tools at its disposal as well.” Tr. of Oral Arg. 50 (emphasis added).

 Thus, according to the United States, even if a party clearly meets our three-part test for Article III standing, the Constitution bars that party from challenging a President’s decision not to enforce the law. Congress may wield what the Solicitor General described as “political . . . tools”—which presumably means such things as withholding funds, refusing to confirm Presidential nominees, and impeachment and removal—but otherwise Congress and the American people must simply wait until the President’s term in office expires.

 The Court—at least for now—does not fully embrace this  radical theory and instead holds only that, with some small and equivocal limitations that I will discuss, no party may challenge the Executive’s “arrest and prosecution policies.” Ante, at 12, n. 5. But the Court provides no principled explanation for drawing the line at this point, and that raises the concern that the Court’s only reason for framing its rule as it does is that no more is needed to dispose of this case. In future cases, Presidential power may be extended even further. That disturbing possibility is bolstered by the Court’s refusal to reject the Government’s broader argument.

 As I will explain, nothing in our precedents even remotely supports this grossly inflated conception of “executive Power,” U. S. Const., Art. II, §1, which seriously infringes the “legislative Powers” that the Constitution grants to Congress, Art. I, §1. At issue here is Congress’s authority to control immigration, and “[t]his Court has repeatedly emphasized that ‘over no conceivable subject is the legislative power of Congress more complete than it is over’ the admission of aliens.” Fiallo v. Bell, 430 U. S. 787, 792 (1977). In the exercise of that power, Congress passed and President Clinton signed a law that commands the detention and removal of aliens who have been convicted of certain particularly dangerous crimes. The Secretary of Homeland Security, however, has instructed his agents to disobey this legislative command and instead follow a different policy that is more to his liking. And the Court now says that no party injured by this policy is allowed to challenge it in court.

 That holding not only violates the Constitution’s allocation of authority among the three branches of the Federal Government; it also undermines federalism. This Court has held that the Federal Government’s authority in the field of immigration severely restricts the ability of States to enact laws or follow practices that address harms resulting from illegal immigration. See Arizona v. United States,  567 U. S. 387, 401 (2012). If States are also barred from bringing suit even when they satisfy our established test for Article III standing, they are powerless to defend their vital interests. If a President fails or refuses to enforce the immigration laws, the States must simply bear the consequences. That interpretation of executive authority and Article III’s case or controversy requirement is deeply and dangerously flawed.

I

 The Court’s opinion omits much that is necessary to understand the significance of its decision, and I therefore begin by summarizing the relevant statutory provisions, the challenged Department of Homeland Security (DHS) action, and the District Court’s findings of fact regarding the injury faced by the State of Texas as the result of what DHS has done.

A

 The relevant statutory provisions have figured in several prior decisions, and in those cases we have recounted how they came to be enacted and have clearly described what they require. These provisions were part of the Illegal Immigration Reform and Immigration Responsibility Act of 1996 (IIRIRA), which was adopted “against a backdrop of wholesale failure by the [Immigration and Naturalization Service] to deal with increasing rates of criminal activity by aliens.” Demore v. Kim, 538 U. S. 510, 518 (2003).2 Congress concluded that a central cause of that failure was the Attorney General’s “broad discretion to conduct individualized bond hearings and to release criminal aliens from custody during their removal proceedings.” Id., at 519. To remedy this problem, Congress “subtract[ed] some of that  discretion when it comes to the arrest and release of criminal aliens.” Nielsen v. Preap, 586 U. S. ___, ___ (2019) (slip op., at 15) (emphasis in original).

 Two such limits are important here. First, 8 U. S. C. §1226(c) directs the Government to “take into custody any alien” inadmissible or deportable on certain criminal or terrorist grounds “when the alien is released” from criminal custody, including when such an alien is released on “parole, supervised release, or probation.” Second, §1231(a) imposes a categorical detention mandate. Section 1231(a)(2) provides that the Government “shall detain [an] alien” “[d]uring the removal period,” which often begins either when an “order of removal becomes administratively final” or when an “alien is released from detention or confinement” not arising from immigration process, §1231(a)(1)(B). This requirement is reinforced by the direction that “[u]nder no circumstance during the removal period shall the [Government] release an alien” found inadmissible or deportable under almost any of the grounds relevant under §1226(c). §1231(a)(2). And §1231(a)(1)(A) commands that the Government “shall remove the alien” within the removal period.

 All of our recent decisions interpreting these provisions confirm that, for covered aliens, shall means shall; it does not mean “may.” See Johnson v. Guzman Chavez, 594 U. S. ___, ___–___, and n. 2 (2021) (slip op., at 2–3, and n. 2); Nielsen, 586 U. S., at ___–___ (slip op., at 16–17). Until quite recently, that was the Government’s understanding as well. See Biden v. Texas, 597 U. S. ___, ___–___ (2022) (slip op., at 8–9) (Alito, J., dissenting).

 Actions taken by Congress when IIRIRA was enacted underscore this conclusion. Because the provisions described above left the Executive with no discretion to refrain from arresting and detaining covered aliens, even during the time immediately after IIRIRA’s enactment when the Executive was still “expand[ing] its capacities” to enforce the  new law, Congress passed “transition rules [that] delayed the onset of the Secretary’s obligation to begin making arrests as soon as covered aliens were released from criminal custody.” Nielsen, 586 U. S., at ___ (slip op., at 21). If the Executive had possessed the discretion to decline to enforce the new mandates in light of “resource constraints,” see ante, at 8, those transition rules would have been entirely “superfluous.” Nielsen, 586 U. S., at ___ (slip op., at 21).

 Despite this clear text and background, the majority now claims that the President’s “enforcement discretion” survived these mandates, ante, at 7, but there is no basis for that conclusion. Certainly it is not supported by the cases it cites. They either underscore the general rule that the Executive possesses enforcement discretion, see Reno v. American-Arab Anti-Discrimination Comm., 525 U. S. 471, 490–491 (1999), or pair that general rule with the observation that the States cannot limit the Government’s discretion in pursuing removal, see Arizona, 567 U. S., at 396, 409. Nothing in those decisions is inconsistent with Congress’s power to displace executive discretion, and the fact that “five Presidential administrations” sometimes neglected the mandates is likewise irrelevant. See ante, at 8. As I have stressed before, the Executive cannot “acquire authority forbidden by law through a process akin to adverse possession,” Biden v. Texas, 597 U. S., at ___ (dissenting opinion) (slip op., at 15), and that is true even if the adverse possession is bipartisan.

B

 The events that gave rise to this case began on January 20, 2021, when the Acting Secretary of DHS issued a memorandum with “enforcement priorities” for the detention and removal of aliens found to be in this country illegally. This memorandum prioritized: (1) aliens “whose apprehension” implicated “national security,” (2) aliens not present “before November 1, 2020,” and (3) aliens due to be released  from criminal confinement who had both been “convicted of an ‘aggravated felony’ ” and were “determined to pose a threat to public safety.” 606 F. Supp. 3d 437, 454 (SD Tex. 2022) (internal quotation marks omitted); see §1101(a)(43) (defining “aggravated felony”). This prioritization was inconsistent with the §1226(c) arrest mandate, which extends to all aliens convicted of any crime within a long list of statutory categories. 606 F. Supp. 3d, at 454–455.

 In February, Immigrations and Customs Enforcement (ICE), an arm of DHS, issued a second memorandum that slightly modified the earlier priorities and stated that “ ‘preapproval’ ” would generally be required “for enforcement actions” against persons outside these priority groups. Id., at 455–456. This memorandum was also inconsistent with the relevant statutes.

 After some litigation regarding these two memoranda, a new DHS Secretary issued a Final Memorandum instructing that even aliens in priority groups need not necessarily be apprehended and removed. App. 113–115. Rather, the Final Memorandum directed DHS personnel to consider non-statutory “aggravating and mitigating factors” in deciding whether to detain an alien. Id., at 114–115. It further stated that DHS “personnel should not rely on the fact of [a qualifying] conviction” when exercising “prosecutorial discretion.” Id., at 115. Thus, the Final Memorandum did not simply permit deviations from the statutory mandates; it flatly contradicted those mandates by stating that qualifying convictions were insufficient grounds for initiating arrest, detention, and removal.

C

 Texas and Louisiana challenged this Final Memorandum in federal court under the Administrative Procedure Act (APA). After a 2-day bench trial, the District Court found in favor of the States and made detailed findings of fact that bear on the issue of standing.

  Much of the District Court’s analysis of that issue focused on the Final Memorandum’s effect on the “detainer” system, 606 F. Supp. 3d, at 459–463, and it is therefore important to understand how that system works in relation to the relevant statutory provisions. When an alien in state custody for a criminal offense is identified as falling within a category of aliens whose apprehension and detention is required by §§1226(a) and (c), the Government should lodge a “detainer” with the State so that the Government can take the alien into custody when he or she is released by the State. Then, when an alien is about to be released, a cooperative State will notify DHS so that it can be ready to assume its obligation under §§1226(a) and (c) to take the alien into federal custody. When that occurs, the State is spared the burdens it would have to bear if the alien, after release, had been placed under state law on probation, parole, or supervised release. But if DHS rescinds a detainer before such an alien is released (or never lodges a detainer in the first place), those burdens fall on the State.

 After reviewing the parties’ evidence, the District Court found that in the first month after the substantive policy change brought about by the January 2021 DHS memorandum, ICE had rescinded 141 detainers in Texas.3 Ninety-five of the criminal aliens whose detainers were rescinded were then released on a form of state supervision. Seventeen of them went on to violate their terms of supervision, and four committed new crimes. Id., at 459.

 The court then examined what had taken place during just the time “since the Final Memorandum became effective” and found that “because of the Final Memorandum,” “ICE ha[d] continued to rescind detainers placed on criminal aliens in [Texas’s] custody,” and the court identified 15  specific cases in which this had occurred. Id., at 460. Rejecting the Government’s claim that these dropped detainers were necessary in light of “limited resources,” the court found that “the Government . . . persistently underutilized existing detention facilities” during the relevant time and that the average daily detained population in April 2022 was less than 40% of the 3-year high in August 2019. Id., at 453, 481, 488.

 Based on these findings of fact and historical data, the District Court identified four categories of costs that Texas had suffered and would continue to bear as a result of the relevant DHS actions. First, the court calculated the dollars-and-cents cost that Texas had to bear in order to supervise criminal aliens who were released in violation of §§1226(a), (c). Id., at 463. Second, it noted the costs associated with criminal recidivism. Id., at 464. Third, it found that some juvenile offenders who “are not detained by ICE because of the Final Memorandum” will attend Texas public schools (and at least one juvenile due to be released will do so). Ibid. Fourth, it concluded that the hundreds of millions of dollars that Texas annually spends on healthcare for illegal aliens would increase when some criminal aliens not detained “because of the Final Memorandum” make use of those services. Id., at 465.

 Concluding that these costs established Texas’s injury for standing purposes, the District Court went on to hold that the Final Memorandum was contrary to law and that Texas had therefore established a violation of the APA.4 As I will explain, it is a common practice for courts in APA cases to set aside an improper final agency action, and that is what  the District Court did here. It vacated the Final Memorandum pending further action by DHS, id., at 499, but it declined to issue injunctive or declaratory relief, id., at 501–502.

 The Government asked the Court of Appeals to stay the District Court’s order vacating the Final Memorandum, but that court refused to do so and observed that the Government had not “come close” to showing “ ‘clear error’ ” in the District Court’s factual findings on the injuries that Texas had already incurred and would continue to incur because of the Final Memorandum. 40 F. 4th 205, 216–217 (CA5 2022).

II

 Before I address the Court’s inexplicable break from our ordinary standing analysis, I will first explain why Texas easily met its burden to show a concrete, particularized injury that is traceable to the Final Memorandum and redressable by the courts. Lujan, 504 U. S., at 560–561.

A

 Injury in fact. The District Court’s factual findings, which must be accepted unless clearly erroneous, quantified the cost of criminal supervision of aliens who should have been held in DHS custody and also identified other burdens that Texas had borne and would continue to bear going forward. These findings sufficed to establish a concrete injury that was specific to Texas. TransUnion LLC v. Ramirez, 594 U. S. ___, ___ (2021) (slip op., at 9); see ante, at 4 (conceding that such costs are “of course an injury”).

 Traceability. The District Court found that each category of cost would increase “because of the Final Memorandum,” rather than decisions that DHS personnel would make irrespective of the directions that memorandum contains. 606 F. Supp. 3d, at 460, 464, 465 (emphasis added).

 The majority does not hold—and in my judgment, could  not plausibly hold—that these findings are clearly erroneous. Instead, it observes only that a “State’s claim for standing can become more attenuated” when based on the “indirect effects” of federal policies “on state revenues or state spending.” Ante, at 9, n. 3. But while it is certainly true that indirect injuries may be harder to prove, an indirect financial injury that is proved at trial supports standing. And that is what happened here. As Justice Gorsuch notes, just a few years ago, we found in a very important case that a State had standing based in part on indirect financial injury. Ante, at 3 (opinion concurring in judgment) (citing Department of Commerce v. New York, 588 U. S. ___, ___–___ (2019) (slip op., at 9–10)). There is no justification for a conflicting holding here.

 In any event, many of the costs in this case are not indirect. When the Federal Government refuses or fails to comply with §§1226(a) and (c) as to criminal aliens, the direct result in many cases is that the State must continue its supervision. As noted, the District Court made specific findings about the financial cost that Texas incurred as a result of DHS’s failure to assume custody of aliens covered by §§1226(a) and (c). And the costs that a State must bear when it is required to assume the supervision of criminal aliens who should be kept in federal custody are not only financial. Criminal aliens whom DHS unlawfully refuses to detain may be placed on state probation, parole, or supervised release, and some will commit new crimes and end up in a state jail or prison. Probation, parole, and corrections officers are engaged in dangerous work that can put their lives on the line.

 Redressability. A court order that forecloses reliance on the memorandum would likely redress the States’ injuries. If, as the District Court found, DHS personnel rescind detainers “because of ” the Final Memorandum, then vacating that memorandum would likely lead to those detainers’ remaining in place.

 B

 While the majority does not contest redressability, Justice Gorsuch’s concurrence does, citing two reasons. But the first is contrary to precedent, and the second should not be addressed in this case.

 The first asserted reason is based on the inability of the lower courts to issue a broad injunction forbidding enforcement of the Final Memorandum. See §1252(f )(1).5 In this case, the District Court did not issue injunctive relief. Instead, it vacated the Final Memorandum, and Justice Gorsuch argues that this relief did not redress Texas’s injuries because it does not “require federal officials to change how they exercise [their prosecutorial] discretion in the [Final Memorandum’s] Guidelines’ absence.” Ante, at 6. There are two serious problems with this argument.

 First, §1252(f )(1) bars injunctive relief by courts “other than the Supreme Court.” (Emphasis added.) As a result, redress in the form of an injunction can be awarded by this Court. According to the Court’s decision last Term in Biden v. Texas, our authority to grant such relief “le[ft] no doubt” as to our jurisdiction even if §1252(f )(1) precluded the lower courts from setting aside an administrative action under the APA. 597 U. S., at ___ (slip op., at 10). We have not been asked to revisit this holding, see id., at ___–___ (Barrett, J., dissenting) (slip op., at 3–4), and I would not do so here.

 Second, even if Biden v. Texas could be distinguished and  no injunctive relief can be awarded by any court, setting aside the Final Memorandum satisfies the redressability requirement. Our decision in Franklin v. Massachusetts, 505 U. S. 788 (1992), settles that question. There, the Court held that a declaratory judgment regarding the lawfulness of Executive Branch action satisfied redressability because “it [was] substantially likely that the President and other executive . . . officials would abide by an authoritative interpretation” of the law “even though they would not be directly bound by such a determination.” Id., at 803 (opinion of O’Connor, J.).6 Here, we need not speculate about how DHS officers would respond to vacatur of the Final Memorandum because the District Court found that the DHS personnel responsible for detainers were rescinding them “because of ” the Final Memorandum. 606 F. Supp. 3d, at 460. This point was effectively conceded by the Government’s application for an emergency stay pending our decision in this case. The Government argued that the Final Memorandum was needed to guide prosecutorial discretion, Application 38–39, and if the District Court’s order were ineffectual, that would not be true. For these reasons, the harm resulting from the Final Memorandum is redressed by setting aside the Final Memorandum.

 As to the concurrence’s second argument—that the APA’s “set aside” language may not permit vacatur—the concurrence acknowledges that this would be a sea change in administrative law as currently practiced in the lower courts. Ante, at 16 (opinion of Gorsuch, J.); see, e.g., Data Marketing Partnership, LP v. United States Dept. of Labor, 45   

 F. 4th 846, 859 (CA5 2022) (“The default rule is that vacatur is the appropriate remedy” under the APA); United Steel v. Mine Safety and Health Admin., 925 F. 3d 1279, 1287 (CADC 2019) (“The ordinary practice is to vacate unlawful agency action”).7 We did not grant review on this very consequential question, and I would not reach out to decide it in a case in which Biden v. Texas resolves the issue of redressability.

 To be clear, I would be less troubled than I am today if Justice Gorsuch’s concurrence had commanded a majority. At least then, Congress would be free to amend §1252(f ). But the majority reaches out and redefines our understanding of the constitutional limits on otherwise-available lawsuits. It is to this misunderstanding that I now turn.

III

 The majority adopts the remarkable rule that injuries from an executive decision not to arrest or prosecute, even in a civil case, are generally not “cognizable.” Ante, at 4 (internal quotation marks omitted). Its reasoning has three failings. First, it fails to engage with contrary precedent that is squarely on point. Second, it lacks support in the cases on which it relies. Third, the exceptions (or possible exceptions) that it notes do nothing to allay concern about the majority’s break from our established test for Article III standing. I address each of these problems in turn.

 A

 Prior to today’s decision, it was established law that plaintiffs who suffer a traditional injury resulting from an agency “decision not to proceed” with an enforcement action have Article III standing. Federal Election Comm’n v. Akins, 524 U. S. 11, 19 (1998). The obvious parallel to the case before us is Massachusetts v. EPA, 549 U. S. 497 (2007), which has been called “the most important environmental law case ever decided by the Court.” R. Lazarus, The Rule of Five: Making Climate History at the Supreme Court 1 (2020). In that prior case, Massachusetts challenged the Environmental Protection Agency’s failure to use its civil enforcement powers to regulate greenhouse gas emissions that allegedly injured the Commonwealth. Massachusetts argued that it was harmed because the accumulation of greenhouse gases would lead to higher temperatures; higher temperatures would cause the oceans to rise; and rising sea levels would cause the Commonwealth to lose some of its dry land. The Court noted that Massachusetts had a “quasi-sovereign interes[t]” in avoiding the loss of territory and that our federalist system had stripped the Commonwealth of “certain sovereign prerogatives” that it could have otherwise employed to defend its interests. Massachusetts, 549 U. S., at 519–520. Proclaiming that Massachusetts’ standing claim was entitled to “special solicitude,” the Court held that the Commonwealth had standing. Id., at 520.

 The reasoning in that case applies with at least equal force in the case at hand. In Massachusetts v. EPA, the Court suggested that allowing Massachusetts to protect its sovereign interests through litigation compensated for its inability to protect those interests by the means that would have been available had it not entered the Union. In the present case, Texas’s entry into the Union stripped it of the power that it undoubtedly enjoyed as a sovereign nation to  police its borders and regulate the entry of aliens. The Constitution and federal immigration laws have taken away most of that power, but the statutory provisions at issue in this case afford the State at least some protection—in particular by preventing the State and its residents from bearing the costs, financial and non-financial, inflicted by the release of certain dangerous criminal aliens. Our law on standing should not deprive the State of even that modest protection. We should not treat Texas less favorably than Massachusetts. And even if we do not view Texas’s standing argument with any “special solicitude,” we should at least refrain from treating it with special hostility by failing to apply our standard test for Article III standing.

 Despite the clear parallel with this case and the States’ heavy reliance on Massachusetts throughout their briefing, the majority can only spare a passing footnote for that important precedent. Ante, at 13, n. 6; see Brief for Respondents 11, 12, 14, 16–18, 23; see also Brief for Arizona and 17 Other States as Amici Curiae 7–12. It first declines to say Massachusetts was correctly decided and references the “disagreements that some may have” with that decision. Ante, at 13, n. 6. But it then concludes that Massachusetts “does not control” since the decision itself refers to “ ‘key differences between a denial of a petition for rulemaking and an agency’s decision not to initiate an enforcement action,’ ” with the latter “ ‘not ordinarily subject to judicial review.’ ” Ante, at 13, n. 6 (quoting 549 U. S., at 527) (emphasis added).

 The problem with this argument is that the portion of Massachusetts to which the footnote refers deals not with its key Article III holding, but with the scope of review that is “ordinarily” available under the statutory scheme. Importantly, Massachusetts frames its statement about declining enforcement as restating the rule of Heckler v. Chaney, 470 U. S. 821 (1985). See 549 U. S., at 527. And  as the Court acknowledges when it invokes Heckler directly, that decision is not about standing; it is about the interpretation of the statutory exception to APA review for actions “committed to agency discretion by law.” 5 U. S. C. §701(a)(2); see 470 U. S., at 823; ante, at 11. And even in that context, Heckler expressly contemplates that any “presumption” of discretion to withhold enforcement can be rebutted by an express statutory limitation of discretion—which is exactly what we have here. 470 U. S., at 832–833.

 So rather than answering questions about this case, the majority’s footnote on Massachusetts raises more questions about Massachusetts itself—most importantly, has this monumental decision been quietly interred? Cf. ante, at 3 (Gorsuch, J., concurring in judgment).

 Massachusetts v. EPA is not the only relevant precedent that the Court brushes aside. “[I]t is well established that [this Court] has an independent obligation to assure that standing exists, regardless of whether it is challenged by any of the parties.” Summers v. Earth Island Institute, 555 U. S. 488, 499 (2009). Yet in case after case, with that obligation in mind, we have not questioned the standing of States that brought suit under the APA to compel civil enforcement.

 In Little Sisters of the Poor Saints Peter and Paul Home v. Pennsylvania, 591 U. S. ___ (2020), two States sued under the APA and sought to compel the Department of Health and Human Services to cease exercising regulatory enforcement discretion that exempted certain religious employers from compliance with a contraceptive-coverage mandate. Id., at ___–___ (slip op., at 11–12). The issue of the States’ standing was discussed at length in the decision below, see Pennsylvania v. President United States, 930 F. 3d 543, 561–565 (CA3 2019), and in this Court, no Justice suggested that the Constitution foreclosed standing simply because the States were complaining of “the Execu tive Branch’s . . . enforcement choices” regarding third parties. Ante, at 7.

 Just last Term in Biden v. Texas, two States argued that their spending on the issuance of driver’s licenses and the provision of healthcare for illegal immigrants sufficed to establish Article III standing and thus enabled them to sue to compel enforcement of a detain-or-return mandate. See Texas v. Biden, 20 F. 4th 928, 970–971 (CA5 2021). The Court of Appeals held that the States had standing, ibid., and the majority in this Court, despite extended engagement with other jurisdictional questions, never hinted that Article III precluded the States’ suit. 597 U. S., at ___–___ (slip op., at 8–12).

 If the new rule adopted by the Court in this case is sound, these decisions and others like them were all just wasted ink. I understand that what we have called “ ‘drive-by jurisdictional rulings’ ” are not precedents, see Arbaugh v. Y & H Corp., 546 U. S. 500, 511 (2006), but the Court should not use a practice of selective silence to accept or reject prominently presented standing arguments on inconsistent grounds.

B

 Examination of the precedents the majority invokes only underscores the deficiencies in its analysis.8 The majority says that the “leading precedent” supporting its holding is Linda R. S. v. Richard D., 410 U. S. 614 (1973). Ante, at 5. But as Justice Barrett notes, this Court has already definitively explained that the suit to compel prosecution in Linda R. S. was rejected “because of the unlikelihood that  the relief requested would redress appellant’s claimed injury.” Duke Power Co. v. Carolina Environmental Study Group, Inc., 438 U. S. 59, 79, n. 24 (1978); see ante, at 2 (opinion concurring in judgment).

 The Court notes in a quick parenthetical that the “Linda R. S. principle” was once “cit[ed] . . . in [the] immigration context” in Sure-Tan, Inc. v. NLRB, 467 U. S. 883, 897 (1984), ante, at 5. But Sure-Tan’s single “[c]f.” cite to Linda R. S. provides the Court no help. 467 U. S., at 897. Sure-Tan only rejected (quite reasonably) any standalone “cognizable interest in procuring enforcement of the immigration laws” by a party who lacked any “personal interest.” Ibid. (emphasis added). And it did so, not as part of a standing analysis, but as part of its explanation for rejecting two employers’ attempt to assert that seeking to have employees deported as retaliation for union activity was “an aspect of their First Amendment right ‘to petition the Government for a redress of grievances.’ ” Id., at 896.

 After these two inapposite precedents, the majority’s authority gets even weaker. I agree with Justice Barrett that neither Heckler, nor Castle Rock v. Gonzales, 545 U. S. 748 (2005), has real relevance here. Ante, at 4–5. Castle Rock considered the “deep-rooted nature of law-enforcement discretion” as a tool for interpreting a statute, not as a constitutional standing rule. 545 U. S., at 761. And as explained above, Heckler is not about standing and only states a presumptive rule. The Court’s remaining authorities are likewise consistent with the understanding that prosecution decisions are “generally committed to an agency’s absolute discretion” unless the relevant law rebuts the “presumption.” Heckler, 470 U. S., at 831 (emphasis added). For example, TransUnion states that it is only when “unharmed plaintiffs” are before the Court that Article III forecloses interference with the “discretion of the Executive Branch.” 594 U. S., at ___ (slip op., at 13) (emphasis deleted).

  In sum, all of these authorities point, not to the majority’s new rule, but to the same ordinary questions we ask in every case—whether the plaintiff has a concrete, traceable, and redressable injury.

C

 Despite the majority’s capacious understanding of executive discretion, today’s opinion assures the reader that the decision “do[es] not suggest that federal courts may never entertain cases involving the Executive Branch’s alleged failure to make more arrests or bring more prosecutions,” despite its otherwise broad language covering the “exercise of enforcement discretion over whether to arrest or prosecute.” Ante, at 5, 9. The majority lists five categories of cases in which a court would—or at least might—have Article III jurisdiction to entertain a challenge to arrest or prosecution policies, but this list does nothing to allay concern about the Court’s new path. The Court does not identify any characteristics that are shared by all these categories and that distinguish them from cases in which it would not find standing. In addition, the Court is unwilling to say that cases in four of these five categories are actually exempted from its general rule, and the one remaining category is exceedingly small. I will discuss these categories one by one.

 First, the majority distinguishes “selective-prosecution” suits by a plaintiff “to prevent his or her own prosecution,” ante, at 9. But such claims are ordinarily brought as defenses in ongoing prosecutions, as in the cases the Court cites, and are rarely brought in standalone actions where a plaintiff must prove standing. This category is therefore little more than a footnote to the Court’s general rule.

 Second, the majority grants that “the standing analysis might differ when Congress elevates de facto injuries to the status of legally cognizable injuries,” and it hypothesizes a situation in which Congress “(i) specifically authorize[s]  suits against the Executive Branch by a defined set of plaintiffs who have suffered concrete harms from executive under-enforcement and (ii) specifically authorize[s] the Judiciary to enter appropriate orders requiring additional arrests or prosecutions by the Executive Branch.” Ante, at 10 (emphasis added). It is puzzling why the presence or absence of such a statute should control the question of standing under the Constitution. We have said that the enactment of a statute may help us to determine in marginal cases whether an injury is sufficiently concrete and particularized to satisfy the first prong of our three-part standing test. Spokeo, Inc. v. Robins, 578 U. S. 330, 341 (2016). But once it is posited that a plaintiff has personally suffered a “de facto” injury, i.e., an injury in fact, it is hard to see why the presence or absence of a statute authorizing suit has a bearing on the question whether the court has Article III jurisdiction as opposed to the question whether the plaintiff has a cause of action. In the end, however, none of this may matter because the majority suggests that such a statute might be unconstitutional. Ante, at 10, and n. 4.

 Third, the majority tells us that the standing outcome “might change” if the Federal Government “wholly abandoned its statutory responsibilities,” but that statement is both equivocal and vague. Ante, at 11 (emphasis added). Under what circumstances might the Court say that the Federal Government has “wholly abandoned” its enforcement duties? Suppose the Federal Government announced that it would obey 80% of the immigration laws or 70% of the environmental laws. Would the Court say that it had “wholly abandoned” enforcement of these bodies of law? What would happen if the Final Memorandum in this case had directed DHS agents not to arrest anyone convicted of any covered crime other than murder? DHS would still be enforcing the arrest mandate as to one of the many covered crimes. Would this only-murder policy qualify as complete abandonment? And why should the ability of a particular  party to seek legal redress for an injury turn on the number of others harmed by the challenged enforcement policy? Standing is assessed plaintiff by plaintiff. The majority has no answers, and in the end, it cannot even bring itself to commit to this complete-abandonment exception. It says only that “the standing calculus might” or “arguably could” change. Ibid. (emphasis added).

 Fourth, the Court says that a plaintiff might have standing to challenge an “Executive Branch’s arrest or prosecution priorities and the Executive Branch’s provision of legal benefits or legal status . . . because the challenged policy might implicate more than simply the Executive’s traditional enforcement discretion.” Ibid. Exactly what this means is not easy to ascertain. One possibility is that the majority is talking about a complaint that asserts separate claims based on the grant or denial of benefits, the grant or denial of legal status, and harms resulting from non-enforcement of a statutory mandate. In that event, standing with respect to each claim would have to be analyzed separately. Another possibility is that the majority is referring to a claim asserting that non-enforcement of a statutory requirement requiring the arrest or prosecution of third parties resulted in the plaintiff ’s loss of benefits or legal status. Such a situation is not easy to imagine, and the majority cites no case that falls within this category. But if such a case were to arise, there is no reason why it should not be analyzed under our standard three-pronged test.

 Fifth, and finally, the majority states that “policies governing the continued detention of noncitizens who have already been arrested arguably might raise a different standing question than arrest or prosecution policies.” Ante, at 12 (emphasis added). The majority provides no explanation for this (noncommittal) distinction, and in any event, as the majority acknowledges, the States in this case challenged noncompliance with the §1231(a)(2) detention mandate in addition to the §1226(c) arrest requirement. Ante, at 2, 13.  The Court points to what it sees as a “represent[ation]” by the Solicitor General that the Final Memorandum does not affect “continued detention of noncitizens already in federal custody.” Ante, at 12, n. 5. But as Justice Barrett notes, the Government argued that when it chooses not to remove someone under the Final Memorandum’s guidance, its mandatory detention obligation ends—meaning it is asserting discretion over continued detention. Ante, at 3 (opinion concurring in judgment).

 In any event, arrest policy cannot be divided from detention policy in this case. When a person is arrested, he or she is detained for at least some period of time, and under the detainer system involved here, “arrest” often simply means transferring an immigrant from state custody to federal custody. As best I can tell, the majority’s distinction between arrest and detention is made solely to avoid the obvious inference that our decision last Term in Biden v. Texas should have dismissed the case for lack of standing, without analyzing “the Government’s detention obligations.” 597 U. S., at ___ (slip op., at 14).

 In sum, with the exception of cases in the first (very small) category (civil cases involving selective-prosecution claims), the majority does not identify any category of cases that it would definitely except from its general rule. In addition, category two conflates the question of constitutional standing with the question whether the plaintiff has a cause of action; category three is hopelessly vague; category four is incomprehensible; and category five actually encompasses the case before us.

IV

 The Court declares that its decision upholds “[o]ur constitutional system of separation of powers,” ante, at 9, but as I said at the outset, the decision actually damages that system by improperly inflating the power of the Executive and cutting back the power of Congress and the authority of the  Judiciary. And it renders States already laboring under the effects of massive illegal immigration even more helpless.

 Our Constitution gives the President important powers, and the precise extent of some of them has long been the subject of contention, but it has been widely accepted that “the President’s power reaches ‘its lowest ebb’ when he contravenes the express will of Congress, ‘for what is at stake is the equilibrium established by our constitutional system.’ ” Zivotofsky v. Kerry, 576 U. S. 1, 61 (2015) (Roberts, C. J., dissenting) (quoting Youngstown Sheet & Tube Co. v. Sawyer, 343 U. S. 579, 637–638 (1952) (Jackson, J., concurring)).

 That is the situation here. To put the point simply, Congress enacted a law that requires the apprehension and detention of certain illegal aliens whose release, it thought, would endanger public safety. The Secretary of DHS does not agree with that categorical requirement. He prefers a more flexible policy. And the Court’s answer today is that the Executive’s policy choice prevails unless Congress, by withholding funds, refusing to confirm Presidential nominees, threatening impeachment and removal, etc., can win a test of strength. Relegating Congress to these disruptive measures radically alters the balance of power between Congress and the Executive, as well as the allocation of authority between the Congress that enacts a law and a later Congress that must go to war with the Executive if it wants that law to be enforced.9

  What the majority has done is to apply Oliver Wendell Holmes’s bad-man theory of the law to the separation of powers. Under Holmes’s theory, as popularly understood, the law consists of those things that a bad man cannot get away with.10 Similarly, the majority’s understanding of the “executive Power” seems to be that a President can disobey statutory commands unless Congress, by flexing its muscles, forces capitulation. That is not the Constitution’s conception of “the executive Power.” Art. II, §1. The Constitution, instead, requires a President to “take Care that the Laws be faithfully executed.” §3 (emphasis added).

 Neither the Solicitor General nor the majority has cited any support for the proposition that a President has the power to disobey statutes that require him to take enforcement actions, and there is strong historical evidence to the contrary.11 The majority’s conception of Presidential authority smacks of the powers that English monarchs claimed prior to the “Glorious Revolution” of 1688, namely, the power to suspend the operation of existing statutes, and to grant dispensations from compliance with statutes.12 After James II was deposed, that changed. The English Bill of Rights of 1689 emphatically rejected “the pretended Power of Suspending of Laws or the Execution of Laws by Rega[l] Authority without Consent of Parl[i]ament” and  “the pretended Power of Dispensing with Laws or the Execution of Laws by Rega[l] Authorit[y] as it ha[s] bee[n] assumed and exercised of late.”13

 By the time of the American Revolution, British monarchs had long abandoned the power to resist laws enacted by Parliament,14 but the Declaration of Independence charged George III with exercising those powers with respect to colonial enactments. One of the leading charges against him was that he had “forbidden his Governors to pass Laws of immediate and pressing importance, unless suspended in their operation till his Assent should be obtained; and when so suspended, . . . ha[d] utterly neglected to attend to them.”15

 By 1787, six State Constitutions contained provisions prohibiting the suspension of laws,16 and at the Constitutional Convention, a proposal to grant the President suspending authority was unanimously defeated.17 Many  scholars have concluded that the Take Care Clause was meant to repudiate that authority.18 See 1 Works of James Wilson 399, 440 (R. McCloskey ed. 1967) (describing Clause as providing that the President holds “authority, not to make, or alter, or dispense with the laws, but to execute and act the laws”).

 Early decisions are inconsistent with the understanding of Executive Power that appears to animate the majority. In 1806, Justice Patterson, while presiding over a criminal trial, rejected the argument that the President could authorize the defendant to violate the law. United States v. Smith, 27 F. Cas. 1192, 1201 (No. 16,342) (CC NY 1806). He concluded:

 “The president of the United States cannot control the statute, nor dispense with its execution, and still less can he authorize a person to do what the law forbids. If he could, it would render the execution of the laws dependent on his will and pleasure; which is a doctrine that has not been set up, and will not meet with any supporters in our government. In this particular, the law is paramount.” Id., at 1230.

 In Kendall v. United States ex rel. Stokes, 12 Pet. 524 (1838), the full Court rejected the President’s claim that he had the authority to disregard a statutory duty to pay certain sums to a government contractor: “To contend that the obligations imposed on the President to see the laws faithfully executed, implies a power to forbid their execution, is  a novel construction of the constitution, and is entirely inadmissible.” Id., at 613. This Court made the obvious connection to the separation of powers: “vesting in the President a dispensing power” would result in “clothing the President with a power entirely to control the legislation of congress, and paralyze the administration of justice.” Ibid.; see also Office of Personnel Management v. Richmond, 496 U. S. 414, 435 (1990) (White, J., concurring) (citing Kendall to explain that the “Executive Branch does not have the dispensing power on its own” and “should not be granted such a power by judicial authorization”).

 The original understanding of the scope of the Executive’s prosecutorial discretion was not briefed in this case, and I am reluctant to express a firm position on the question. But it is indisputable that we have been provided with no historical support for the position taken by the Solicitor General or the majority.

*  *  *

 This sweeping Executive Power endorsed by today’s decision may at first be warmly received by champions of a strong Presidential power, but if Presidents can expand their powers as far as they can manage in a test of strength with Congress, presumably Congress can cut executive power as much as it can manage by wielding the formidable weapons at its disposal. That is not what the Constitution envisions.

 I end with one final observation. The majority suggests that its decision rebuffs an effort to convince us to “ ‘usurp’ ” the authority of the other branches, but that is not true. Ante, at 3. We exercise the power conferred by Article III of the Constitution, and we must be vigilant not to exceed the limits of our constitutional role. But when we have jurisdiction, we have a “virtually unflagging obligation” to exercise that authority. Colorado River Water Conservation Dist. v. United States, 424 U. S. 800, 817 (1976). Because  the majority shuns that duty, I must respectfully dissent.