Skip to main content
Free

Reading Regents and the Political Significance of Law

When the Supreme Court handed down its decision in Department of Homeland Security v. Regents of the University of California, in June 2020, advocates celebrated. DACA—an acronym that no longer requires definition—lived to see another day.1 Newspaper headlines marked the decision as a decisive rebuff of the Trump administration’s efforts to end the Obama-era program that shielded so-called Dreamers from deportation while authorizing them to work in the United States.2 Initiated in 2012, the Deferred Action for Childhood Arrivals program had survived almost four years of a presidential administration overtly hostile to immigrants and immigration—a government bent on unraveling as much of the administrative and political legacy of its immediate predecessors as possible.3 The Supreme Court largely affirmed the Ninth Circuit’s holding that efforts by the Department of Homeland Security (DHS) to rescind DACA were arbitrary and capricious and therefore invalid, sending DHS back to the drawing board to accomplish its objectives.4 With the 2020 presidential election less than five months away and the very real possibility of regime change in the air, the decision seemed decisive. The Supreme Court had saved DACA, at least for the time being.

On the other side of the presidential election, we can now say that the Dreamers and their lawyers succeeded in using the courts to run out the clock on one of the more high-profile efforts of the Trump presidency. This success calls for an explanation. The original legal theory of DACA was predicated on its discretionary and therefore defeasible character. The government justified DACA as a series of individual acts of prosecutorial discretion, defined as the inherent discretion law enforcement officials possess to forbear from enforcement, at their convenience, in order to prioritize enforcement resources. DACA’s founding document—a memorandum issued by the Secretary of Homeland Security—included the disclaimer standard in Executive orders and agency guidance documents: “this memorandum confers no substantive right.”5 DACA’s promise, then, lasted as long as the Executive wanted it to. The promise was durable as long as President Obama remained in office but unenforceable should the Executive branch fall into the hands of officials hostile to the program.

Given the apparently weak anchor DACA provided, why was it so difficult for a new administration, whose enforcement priorities did not include categorical forbearance for Dreamers, to reorient the enforcement system in its preferred direction? A conventional answer, repeated as a description of many of the Trump administration’s stumbles across regulatory arenas, was that officials were incompetent, sloppy, and disingenuous. The myriad court opinions in the DACA rescission litigation of the Trump years, from across the country and up and down the judicial hierarchy, reflected a version of this thesis. No court concluded that DACA was required by law. All parties, including the Supreme Court, seemed to agree that an administration could end the program.6 But despite efforts to respond to the demands and criticisms of the lower courts, the Trump administration could not find its way to its desired conclusion.

But if the Court has implicitly acknowledged that DACA is not legally required and expressly stated that the government has the authority to wind it down, in what sense was Regents a major victory? In this essay, I argue that Regents is not a triumph in immigration law or even a decision of immigration law; far from it, the opinion contains a roadmap to DACA’s demise. The decision’s salutary outcome for immigrants also distracts us from a more ominous turn in the Roberts Court toward a reading of the immigration laws that empowers both Congress and the President to do as they please—a reading exemplified by one of the Term’s other decisions, Department of Homeland Security v. Thuraissigiam, in which the Court rejects a Suspension Clause challenge to expedited removal proceedings.7 Regents does reflect a kind of political triumph, however, not just because DACA was saved but because the Court calls attention to the profound interests its recipients have in remaining in the United States and thus to their new social status, separate and apart from their legal status. And yet, within the Regents decision itself, as well as in the legal claims made against the Trump administration, are the very tools with which courts might again stymie political change designed to advance immigrants’ rights, relying on the exacting procedural regularity championed in Regents by Chief Justice Roberts.

I.  The Trump Administration’s Failed Rescission

Chief Justice Roberts opened his opinion in 2012, at the moment of DACA’s inception. But to understand what DACA sought to achieve as a matter of administration, it is important to understand what it replaced. As most every court to have heard a DACA-related dispute has recognized, implicit in the operation of an immigration enforcement regime is the authority of Executive officials to set priorities for law enforcement agents. Those priorities can encourage agents to forbear from arresting or deporting otherwise removable non-citizens as part of a larger systemic interest in channeling resources toward removals in the government’s highest interests. Beginning in 2010, Obama-era DHS officials articulated a set of priorities in guidance documents (known as the Morton Memos) in an effort to encourage line-level officials to consider non-enforcement against certain types of individuals, including those who met the criteria that would eventually define DACA – the hundreds of thousands of non-citizens lacking immigration status who had been brought to the United States as youth.8 After two years of trying to steer the enforcement system with these exhortations, DHS officials determined that few obvious or publicly visible changes to enforcement practices had occurred. The Department’s political leadership, in conjunction with the White House, thus devised DACA to protect Dreamers from deportation. The program, adopted by what came to be known as the Napolitano memorandum, invited applications for forbearance from those who satisfied carefully drawn eligibility criteria, virtually ensuring, though not guaranteeing, protection and work authorization for Dreamers.9

By the time President Donald Trump took office, more than 750,000 Dreamers had been granted DACA status,10 which provided them actual and psychological relief from removal and enabled them to enter the workforce and live as if their immigration status were immaterial. As a candidate, Donald Trump vowed to rescind DACA immediately, but in his initial months in office, President Trump himself expressed ambivalence and even reservations.11 In September 2017, however, Attorney General Jefferson Sessions sent a one-page, four-paragraph letter to Acting DHS Secretary Elaine Duke to “advise” that DHS should rescind the Napolitano memorandum initiating DACA on the ground that DACA was “an open-ended circumvention of immigration laws” and “an unconstitutional exercise of the authority of the Executive Branch.”12 The next day, in what had to have been a coordinated decision within the administration, Secretary Duke released her own memorandum terminating DACA,13 citing the Attorney General’s letter and the litigation that had called into question the legal authority for a second but now moribund Obama-era deferred action policy (Deferred Action for Parents of Americans and Lawful Permanent Residents (DAPA)).14 The timing and content of each of these two administrative documents became central to the Supreme Court’s resolution of the legal question before it—whether the Trump administration’s efforts to rescind DACA had been lawful.

The court case began in three different circuits, where an array of plaintiffs raised numerous substantive claims, two of which ended up before the Supreme Court: that the rescission of DACA was arbitrary and capricious in violation of the Administrative Procedure Act and that the rescission violated the Equal Protection Clause of the Constitution. In its culminating opinion, the Supreme Court first concluded that DACA did not fall into the class of non-enforcement decisions long held to be unreviewable by courts on the authority of Heckler v. Chaney.15 DACA amounted, instead, to a full-blown program for granting immigration relief and attendant benefits, justifying judicial review—review that jurisdiction-stripping provisions of the Immigration and Nationality Act (INA) also did not preclude. The Court then proceeded to hold that the rescission of DACA in its entirety was indeed inconsistent with the requirements of the APA but that none of the plaintiffs’ allegations established a “plausible” claim of racial animus under the Equal Protection Clause.

But DACA was a discretionary program that the administration should have been able to undo easily, not a program that should have survived more than three years of a concerted rescission effort (assuming Trump officials’ hearts were in it). Why did a clearly permissible outcome evade the Trump administration? Though incompetence has been charged repeatedly in public commentary, the explanations offered by Chief Justice Roberts underscore that the federal courts’ conceptualization of the administration’s fault changed throughout the litigation.

In his letter to Secretary Duke, Attorney General Sessions justified the rescission as legally required, in part citing the litigation risk that maintaining the program posed, given that the Fifth Circuit had invalidated President Obama’s similarly structured DAPA initiative.16 At the time, Adam Cox and I argued that the administration was hiding behind flimsy legal arguments to duck political responsibility and accountability for ending a widely popular and successful program.17 The lower courts quickly put a stop to this evasion by demanding that the Trump administration provide reasons for the rescission beyond what the courts viewed to be erroneous legal claims.18 Judge John Bates in the District for the District of Columbia actually gave the administration an opportunity to remedy the APA violation by providing the court with a more extended rationale for the rescission.19 The administration obliged with a memorandum from a new DHS Secretary, Kirstjen Nielsen, in which she purported not to disturb the Duke memorandum and its legal conclusions but added multiple policy reasons to justify the rescission.20 By elaborating on its legal reasoning and offering a policy rationale for ending DACA that could justify the rescission if the court continued to find the legal reasoning wanting—just the sort of rationale courts typically do not second-guess—the path to rescission seemed to have been cleared.

And yet, at the Supreme Court, the sufficiency of the Nielsen memo went untested. In its opinion, the Court dismisses the memo’s relevance because it had been framed as an elaboration of the Duke memo and yet bore “little relationship” to the original purported basis for the agency action.21 Under hornbook administrative law, then, the Nielsen memo constituted an irrelevant post hoc rationalization. In support of this conclusion, the Court lists the familiar case law, for which the standard citation is SEC v. Chenery Corp.22, and the reasons for rejecting an elaborated justification and insisting that the agency start a new policy process or issue a new decision to invoke new reasons: that such requirements promote accountability, ensure the parties and public can respond to the agency’s authority, preserve the orderly process of review, and constrain the agency from making its reasons and therefore its policy a moving target.23

Thus focused on the Duke memo, the Supreme Court offers a two-part reason for finding the rescission procedurally flawed, each part of which I consider in more detail in Part II. The first is a legal rationale not yet hit upon by the federal courts but offered by the respondents from the District of Columbia: according to the Court, the Sessions letter had concluded that DACA was unlawful because it contained the same legal defects the Fifth Circuit had found in DAPA. Because the Fifth Circuit focused its analysis on the benefits DAPA conferred (primarily eligibility for work authorization), the Secretary failed to appreciate that the Sessions letter left her with discretion to decouple the two parts of DACA and consider whether its forbearance policy standing alone, without benefits, passed legal muster.24 The second of the Court’s rationales also sounds in basic administrative law—that when it changes a policy, an agency must consider the reliance interests engendered by that policy, not because those interests are necessarily legally dispositive, but because they are always substantively relevant.25 Leaving aside the puzzlement expressed by Justice Thomas in his partial dissent—why should these reliance interests matter if some or all of the program itself was without legal foundation?—the Court gave the administration two clear assignments on remand if it hoped to continue the rescission effort.

With this move, Chief Justice Roberts found a political sweet spot for someone hoping not to take sides on the merits. He did not allow the rescission to proceed, but he also avoided concluding that DACA was lawful. He thus did not close the door to an eventual gutting of DACA through elimination of the path to work authorization that made it so valuable. But whether Regents amounts to a “win” depends both on whose perspective we take and the timeframe we adopt. In the months after the decision, DACA recipients had clearly triumphed. The story’s denouement unfolded in a courtroom in the Eastern District of New York. After Regents, the government did indeed return to the drawing board. Attorney General William Barr rescinded all DOJ authorities relevant to the case, including a 2014 memorandum from the Office of Legal Counsel elaborating why the much larger DAPA program was consistent with the INA and within the Secretary’s authority.26 At DHS, Acting Secretary Chad Wolf rescinded the Nielsen and Duke memoranda and styled his own memo as beginning the process of considering DACA anew. He pledged that, while DHS conducted its fresh process, the agency would honor existing DACA grants but would no longer adjudicate new or pending applications. In his memorandum explaining these steps, Acting Secretary Wolf acknowledged the reliance interests of existing DACA holders by repeating back the Supreme Court’s articulation of those interests. But he then offered: “[w]hatever the merits of these asserted reliance interests on the maintenance of the DACA policy, they are significantly lessened, if not entirely lacking” for those who had never received deferred action in the first place.27

Litigants immediately challenged this new quasi-rescission. But rather than determine if DHS had properly adhered to the procedural path cleared by the Supreme Court, Judge Nicholas Garaufis found Wolf’s appointment to have been unlawful, thus invalidating his actions as lacking authorization—a conclusion reached by numerous courts reviewing various DHS actions in the waning days of the Trump administration.28 Whether DHS could correct this structural defect and try yet again to rescind DACA became moot with the election of Joseph R. Biden to the presidency in November 2020. Biden pledged during the campaign to shore up DACA. Not long after his inauguration, he followed through by declaring his intention to “fortify” the program.29

But even though the election brought the rescission saga to an end, a deeper, more speculative strain of Regents is now in play. Despite being a procedural decision on its surface and in its holdings, Chief Justice Roberts’s novel reasoning forecasts still more legal wrangling over both DACA’s validity and, more generally, the capacity of a new president to chart a different path on immigration policy.

II.  Immigration and the Roberts Court

Regents ensured that DACA would survive into a new administration determined to preserve it. But the decision itself is neither a victory for immigrants’ rights in a jurisprudential sense nor a particularly probative data point in a more holistic account of immigration law in the Roberts era. Most immediately, the opinion provides a roadmap to DACA’s demise by inviting litigants and judges to separate its two pillars—categorical forbearance from removal on the one hand and eligibility for work authorization and benefits tied to deferred action status on the other—and to invalidate the latter. Beyond DACA, the prospects for a jurisprudence that restrains the coercive power of the government against non-citizens grew even dimmer this Term, despite Regents. In the unrelated Thuraissigiam decision, the Roberts Court rejected yet another rights-based challenge to the assertion of a sweeping enforcement and removal power expressly authorized by Congress, continuing what increasingly appears to be the Court’s steady departure from its prior practice of infusing interpretations of the INA with a concern for basic due process principles. This opinion attracted much less interest from the media and general public than the fate of DACA, but it is of far greater importance to the future of immigration law qua law.

The first step to dismantling DACA through the courts is establishing its justiciability, which the Regents Court appears to do definitively. As originally conceived, DACA was an exercise of the Secretary’s enforcement discretion—a form of decision-making typically insulated from judicial review.30 The so-called benefits features of DACA merely flowed from the decision to forbear from removal in light of regulations and administrative policies, dating back decades, that linked those benefits to a grant of deferred action.31 Indeed, it seems quite plausible that the Obama administration chose deferred action as the vehicle through which to provide Dreamers some relief because of these legal parameters—unreviewable discretion connected to an already established regulatory structure, complete with a ready-made and routine process by which recipients of forbearance could apply for employment authorization documents (EADs).

But in various instantiations of the litigation over both DACA and DAPA, the Supreme Court and the lower courts have rebuffed the government’s argument (common across administrations) that the policies are unreviewable exercises of enforcement discretion.32 In Regents, the Court declines to apply Heckler v. Chaney and its holding that the decision to decline to enforce the law is not subject to judicial review because it is committed to agency discretion, emphasizing that “DACA is not simply a non-enforcement policy.”33 By erecting an application process to identify individuals who met enumerated criteria, the administration created a “program for conferring affirmative immigration relief,” not a “passive non-enforcement policy.”34 By reframing DACA as a program with component parts, each subject to legal review, the Court thus raises the stakes for the government by redefining the nature of what the administration seeks to accomplish in a way that imposes more procedural obligations and heightens the threat of judicial surveillance.35

But despite finding DACA to be reviewable, the Court does not purport to evaluate the legality of its component parts on the merits. Instead, the Court finds fault with the Secretary’s own failure to decouple DACA’s two pillars—forbearance and benefits—and then evaluate the legality and viability of a forbearance-only policy.36 In finding this avenue legally available to the Secretary, Roberts is arguably under-reading the Sessions letter to Secretary Duke, which, as the Chief Justice himself emphasizes, binds the Secretary’s discretion. That letter—one page in length—does not clearly draw the distinction Roberts identifies and arguably casts legal doubt on DACA as a whole.37 But Roberts seizes on the letter’s statement that “the DACA policy has the same legal … defects that the courts recognized as to DAPA” to chart his course through the fraught case. In the litigation over DAPA, the Fifth Circuit, in Texas v. United States,38 had determined the relevant legal question to be whether the Secretary had authority to make DAPA recipients eligible for benefits, not whether he had authority to forbear from removing the class of people who fit into the program’s criteria. In other words, even under Sessions’s letter and the specter of litigation risk raised by the Fifth Circuit, the Secretary could have considered a forbearance-only version of DACA. And under bedrock administrative law—Motor Vehicle Manufacturer’s Association v. State Farm39—an agency seeking to rescind a policy must consider in its “reasoned analysis” whether alternatives “within the ambit of the existing policy” might be viable. Because forbearance was at the very core of DACA, DHS ought to have considered a policy of “forbearance without benefits,”40 and the Sessions letter simply did not foreclose that possibility.

The Regents opinion thus gave the Trump administration a roadmap, albeit a lengthy one, to rescinding DACA once and for all. The agency could return to the drawing board, find the benefits prongs to be legally unfounded while maintaining a forbearance-only policy, which it then could have phased out if it had articulated policy reasons for abandoning this particular form of enforcement prioritization—reasons that adequately took into account reliance interests (more on which soon). And as noted above, in what turned out to be the waning months of the Trump administration, Chad Wolf formally started the rescission process anew, appearing to adopt a posture that threaded the Court’s needle before Judge Garaufis found his authority wanting.

Today, the Court’s roadmap is no longer of use to the administration itself. But it does steer litigants (the state of Texas, for example) seeking to challenge DACA’s very legality down a clear path that the Court already understands. Despite not addressing DACA’s legality squarely, the construction and reasoning of the Court’s opinion are both highly suggestive: whereas forbearance seems safe, the future of work authorization and other benefits is in doubt. This prediction flows in part from the extensive use, bordering on adoption, by the Chief Justice of the Fifth Circuit’s reasoning in Texas v. United States, which was, after all, a lower court decision concerning an immigration relief program not actually at issue in Regents and that did not culminate in a Supreme Court decision on the merits. To be sure, Attorney General Sessions put the Fifth Circuit opinion at issue by seeming to rely on it in his DACA letter to the Secretary. But not only does Chief Justice Roberts fold Texas v. United States and the fate of DAPA into his discussion of the history of the DACA rescission, he engages in an extended exposition of the Fifth Circuit’s analysis when evaluating Secretary Duke’s determination that she had no legal discretion to continue DACA. In his explication of the decision, Chief Justice Roberts ends up making a persuasive case for the bifurcation of the program. And by suggesting that the Secretary consider this alternative, he is at the very least implying that it may well be legally available, if not legally advisable.

Or put slightly differently, if the Chief Justice thought it likely that a DACA-style forbearance policy contravened the Executive’s legal duties, it would have been curious for him to send the matter back to the agency for consideration of an option doomed to fail. When the litigation over DAPA began, its challengers, not to mention critical commentators, raised doubts about the categorical forbearance it embodied—the same legal concerns Justice Thomas highlights in his Regents opinion dissenting from the APA holding, in which he emphasizes Congress has not authorized categorical exemptions to the INA’s removal requirements.41 But by the time the DAPA case had reached the Fifth Circuit, that court seemed to have accepted forbearance only as a manifestation of the Executive’s authority to prioritize removal resources.42 Indeed, for a court to reject the forbearance component of DACA truly would be a rejection of a very basic enforcement practice.43 As Adam Cox and I have argued, DACA is no less enforcement discretion for having shifted the locus of discretion to the Secretary level and away from individual agents, even as individual agents continue to make decisions.44

Now that we have a presidential administration that intends to fortify rather than wind down DACA, the central note of doubt in the Regents opinion comes into play: will the courts permit the Biden administration to continue extending eligibility for work authorization (and other benefits) to recipients of deferred action under the DACA program?45 DACA’s value to its beneficiaries turns on the answer to this question. Though a promise of forbearance diminishes the uncertainty and psychological anxiety associated with the threat of deportation, eligibility for work authorization is what has made DACA truly transformative for hundreds of thousands of non-citizens without legal status who are fundamentally American. Again, Chief Justice Roberts does not address on the merits whether DHS has the authority to extend eligibility for work authorization. But he also does not take the approach of several of the lower courts, which outright rejected the Trump administration’s conclusion that DACA was unlawful. He offers as an alternative for the agency a position that jettisons work authorization for being illegal (as Sessions had concluded) but continues on with forbearance. Perhaps if the administration had gone through with such an approach and it had reached the Supreme Court a second time, Roberts and his fellow justices, upon closer inspection, would have concluded, in fact, that extension of work authorization was perfectly legal. After all, the regulations making deferred action recipients eligible to apply for EADs date back to the Reagan era.46 The administration would then have had to rescind or modify those regulations (possibly after having gone through notice and comment) and explain why it was doing so – something that, as even Justice Thomas’s dissent recognized, would be no easy task.47 But would the Chief Justice really have sent the Trump administration back for more likely fruitless memoranda drafting? Perhaps he expected them to once and for all come up with sufficiently articulated policy reasons for rescinding all of DACA, obviating the need for the Court to address the legality of DACA on the merits.

If the Court was seeking to avoid a decision on the merits, the 2020 election may have foiled its plans. Again, the Biden administration has announced its intention to fortify DACA through notice and comment rulemaking. This move seems intended to add more procedural armor to the program in the hopes of preserving it through what may be a lengthy legal battle already begun by the state of Texas and some of its allies in the Southern District of Texas, where the district judge who invalidated DAPA now sits on the case.48 If and when DACA returns to the Court under this new guise, no one should be surprised if only forbearance stands at the end, leaving it to Congress to provide a meaningful anchor of belonging for the Dreamers.49 The way the Chief Justice in Regents presents the work authorization question as potentially expendable for legal reasons, and the incredulity expressed by some of the Justices at oral argument in the DAPA case back in 2015, justify characterizing the litigation risk associated with DACA as high.50

If the Court does indeed come to rest on a forbearance-only version of DACA, its decision will stand alongside Regents as an example of the Court’s skepticism of administrative innovation and perhaps an insistence that expansive Executive policymaking be supported by clear and even express statutory authority. In so doing, the Court would hamstring the ability of the Executive branch to introduce humanity and stability into its management of a massive deportation regime that Congress thus far has been unable or unwilling to reform. This form of disabling the Executive in the name of accountability and the separation of powers would in turn exacerbate another feature of today’s immigration law that the Supreme Court slowly but surely has been re-enforcing in recent years, including during the October 2019 Term.

In a string of decisions over the last several years, the Court has taken the inverse approach of its opinion in Regents, finding that DHS has robust power to choose how to enforce the immigration laws in light of expansive statutory delegations. In at least two startling cases, the Court outright rejects the application of meaningful and direct constitutional limits on the political branches themselves, in ways that depart from past precedent.51 And in other cases, the Court increasingly reads statutory provisions with strenuous resistance to interpretive possibilities that would be protective of immigrants’ interests, including in ways that sound in basic fairness. In other words, the real story of the Roberts Court’s immigration jurisprudence has been to empower the political branches working in tandem, with waning interest in scrutinizing how the Executive wields the power Congress has delegated to it.

Three data points are especially worth noting as part of an initial sketch of judicial review and immigration in the Roberts era: Trump v. Hawaii, DHS v. Thuraissigiam, and Jennings v. Rodriguez. These other jurisprudential developments help to put Regents in a more holistic perspective. The contrast underscores that Regents’ rigorous review of Executive action that advances the interests of immigrants cannot be generalized. To the contrary, the government’s interest in the enforcement of the immigration laws, not the interests and rights of non-citizens, remains at the heart of modern immigration law and policy.

The clearest example of a Roberts-era decision that reads a statutory mandate to authorize sweeping Executive power with minimal to no constitutional constraint is Trump v. Hawaii,52 in which the Court effectively upholds President Trump’s proclamation barring the entry of nationals from several Muslim-majority countries on the ostensible ground that their governments could not provide adequate security information.53 Two of that opinion’s features bear mention as signals of the direction of the Roberts Court on immigration law.54 First, the Court, in an opinion by Chief Justice Roberts, offers an uncomplicated, neatly textualist reading of the specific statutory provision at issue, which is indeed straightforward and clearly grants the president the power to deny entry to “any aliens or class of aliens” whose entry would be “detrimental to the United States.”55 More important, the Court rejects the complex arguments grounded in the INA’s structure offered by challengers, who attempted to read subsequent additions to the INA focused on national security screening as limiting the suspension power first adopted in 1952.56

Second, and still more important, even as Chief Justice Roberts got the statutory arguments correct, he offered a startling vision of the (lack) of constitutional scrutiny of the president’s actions pursuant to his clear authority. Not only does the opinion apply a highly deferential standard of review that requires the government to provide only a “facially legitimate and bona fide reason” for its actions, it purports to apply a heightened version of rational basis review for the sake of argument and concludes that even an “established discriminatory motive would not have warranted invalidation of the government’s actions, as long as another, facially legitimate reason for the exclusion existed.”57 Instead of attempting to contextualize, diminish, or otherwise cabin the anti-Muslim statements of the president, which Justice Sotomayor powerfully recounts in dissent, Chief Justice Roberts concludes that those statements, and the motivations they arguably reflect, are beside the point. For the first time since the notorious cases of the Chinese exclusion era, the Court upholds an action by the political branches that could reasonably be characterized as a violation of ordinary domestic constitutional law, in this case government action predicated on grounds ordinarily considered illicit.58 In other words, the president not only has power (delegated by Congress), the Court will impose almost no constraint on its exercise (under the Constitution).

But at least Trump v. Hawaii, when it was decided, could be cabined in various ways. Its constitutional analysis addressed a specific sort of intent-based antidiscrimination claim, and its national security-infused deference involved non-citizens on the precipice of entry, outside the custody and control of the United States.59 But this past Term, the Court issued a decision that signaled that these implicit limits on the government’s power might be hard to maintain. Thuraissigiam v. Department of Homeland Security60 rejects a Suspension Clause challenge to a provision of the system of expedited removal—a statutory scheme that authorizes the government to remove in summary fashion, without a hearing, certain categories of non-citizens seeking admission, subject to a staged screening process for migrants who express the intention to seek asylum.61 The Court overturns a Ninth Circuit decision that had invalidated the provision at issue. That provision authorized a non-citizen to obtain review via a writ of habeas corpus in only three discrete circumstances, none of which applied to Mr. Thuraissigiam, an ethnic Tamil who crossed the southern border unlawfully and indicated to an immigration officer his fear of persecution if returned to Sri Lanka.62 In his habeas petition, Thuraissigiam challenged the officer’s determination that he lacked a credible fear of persecution, contending that the officer had not given him a meaningful opportunity to establish his claim. Had he succeeded in establishing credible fear, he would have been entitled to a full asylum hearing before an immigration judge followed by judicial review, rather than to summary removal.

Justice Samuel Alito’s majority opinion prompted a lengthy concurrence and dissent and involves technical details of immigration and habeas law, the ins and outs of which deserve a treatment all their own.63 For present purposes, the opinion’s marked departure in tone and substance from some of the Court’s major precedents concerning not only habeas, but also the Due Process Clause, make it noteworthy. The decision represents another installment—perhaps the most important yet—in the Roberts Court’s resistance to reading immigration statutes to advance rights-based constitutional goals. It also shows the Court in a tightening embrace of a regulatory scheme created by Congress that dramatically empowers the Executive to enforce the immigration laws without meaningful judicial scrutiny.

Whereas Roberts’s voice in Regents channels the aspirations of DACA recipients and acknowledges their connection to American communities, Justice Alito’s voice in Thuraissigiam is suspicious of non-citizens and friendly to the government’s interests in efficiency.64 In the exposition of the case, for example, he laments the amount of time removal cases take to be adjudicated, noting the considerable expense to the government of detaining the non-citizen or the “attendant risk” that someone released might not ever be found again.65 He refers to the “burdens” asylum screenings pose by “overwhelming our immigration system.”66 This sort of framing may not be outlandish (the backlog of immigration cases is real), or even tendentious, but it does reflect a valuation of the efficiency that the system of expedited removal prioritizes as opposed to a robust consideration of potential asylum claims that the system arguably underplays. And Alito’s rhetoric borders on sarcasm as he dismisses Thuraissigiam’s habeas claim by saying that the government “is happy to release him—provided the release occurs in the cabin of a plane bound for Sri Lanka.”67

The core holding of the case was that the Suspension Clause as understood in 1789 does not reach the sort of claim Thuraissigiam sought to raise – a challenge to his credible fear screening and the chance to make his asylum claim anew. That holding may not obviously strain credulity. As the Court put it, “the historic role of habeas is to secure release from custody,” but Thuraissigiam sought “the opportunity to remain lawfully in the United States” through an application for asylum and other relief, the rejection of which he claimed was flawed.68 Justices Breyer and Ginsburg concur that the section of the INA at issue was not unconstitutional as applied to the respondent, agreeing that Congress has the power to foreclose habeas review of the sort of claim Thuraissigiam raised – one that sought to challenge the immigration officer’s factual findings.69 But Breyer and Ginsburg depart from the majority’s opinion because it calls into question the availability of habeas to challenge removal orders at all. As Gerald Neuman noted powerfully in the aftermath of the decision, Justice Alito’s opinion “eviscerates the Suspension Clause” and departs from the Court’s seminal habeas decisions in Boumediene v. Bush and INS v. St. Cyr.70 And as Justice Sotomayor emphasizes, in a dissent joined Justice Kagan, the implication of the majority opinion is that “expedited removal proceedings shall be functionally unreviewable through the writ of habeas corpus, no matter whether the denial is arbitrary or irrational or contrary to governing law”—an implication that runs counter to a century of the Court’s precedents.71

The fact that Justice Alito reached still further to address a due process claim not clearly raised by respondent—a reach wholly unnecessary to resolve the case—expands the opinion’s remarkable breadth and further announces the Court’s intention to retreat from constitutional review of congressional schemes to enforce the immigration laws. Alito disclaims that that the Due Process Clause applies to Thuraissigiam simply because he set foot into US territory; he remained “on the threshold,” which meant that he had only those rights Congress saw fit to give him. Though Alito does not explain whether and how far this reasoning extends beyond an individual like Thuraissgiam, who has just entered the country, the stunning implications of his reasoning become apparent when he underscores that being stopped at the border is in fact a term of art and that someone “paroled elsewhere in the country for years pending removal” would also appear to be at the threshold.72 Not only does this reference suggest that Congress’s authorization of expedited removal for any non-citizen who has not been admitted and has been in the United States for fewer than two years is perfectly constitutional73—an authorization no Executive until the Trump administration had sought to utilize—it also suggests Congress may go still further.

With its due process analysis, Thuraissigiam fits together with the Court’s 2017 detention decision in Jennings v. Rodriguez, which also revealed the Roberts Court’s skepticism that the Due Process Clause meaningfully restrains Congress from authorizing extraordinary enforcement measures. In that case, the Court declined to read several of the INA’s mandatory detention provisions as containing either an implicit time limit or the requirement of a bond hearing out of respect for underlying due process rights of those detained.74 Also written by Justice Alito, Jennings signaled the Court’s departure from an important jurisprudential predisposition of the early 2000s—the application of principles of constitutional avoidance to temper the reach of enforcement statutes, particularly those authorizing detention and the deprivation of the core liberty interest to be free from it. The lodestar in this turn-of-the-century jurisprudence has been the Court’s decision in Zadvydas v. United States, which energetically reenforced the principle that the Due Process Clause applies to non-citizens and then proceeded to read a detention statute that could have given rise to indefinite detention as containing an implicit six-month limit on detention, subject to the government establishing anew that detention still served its purposes and was justified.75 Zadvydas has had an enormous influence in the way lower courts have read detention statutes as well as on government detention practices. The decision has come under fair criticism for relying on the avoidance canon to rewrite Congress’s statute rather than directly confronting the constitutional question raised by potentially indefinite detention. But it has also come to stand for the Court’s willingness to cautiously superintend the massively coercive powers of the immigration system. It may be possible to reconcile cases like Trump v. Hawaii and Thuraissigiam with the Zadvydas holding. But the Court’s inclinations have very clearly changed with respect to both detention itself and challenges by non-citizens more generally to the enforcement apparatus authorized by Congress and mobilized by DHS.76 As Zadvydas fades further back in time, its hold on the Court’s approach to constitutional review appears to be weakening.77

III.  Regents and the Politics of Judicial Review

Even if DHS had not run afoul of State Farm and Chenery in its attempts to rescind DACA, the Court identifies another flaw in the Duke memorandum. The claim emerged early in the litigation and reflected the moral heart of the DACA recipients’ case, translated into the language of administrative law. The Duke memorandum was arbitrary and capricious for failure to take into account the reliance interests that DACA had generated.78 As a legal matter, the Court’s recognition of these has limited significance; the Court makes clear that reliance interests are not dispositive, only that they must be taken into account and explained away as less significant that the administration’s policy goals. But the Court presents the interests not only as serious and weighty, but also as “radiating outward” from the recipients themselves to the economic and social institutions with which they have become intertwined.79 Also important, the Court rejects the government’s argument, echoed by the dissent, that because DACA never promised an entitlement, no reliance interests could be said to have arisen.80 The Court here privileges a social fact over a legal construct—what we all know to be true about DACA’s implicit promise to forbear and include, even as the express trappings of the initiative disclaimed any sense of obligation. Though the reliance passage is brief, it underscores that the agency cannot elide the way DACA has changed the social world, immediately raising the political costs to the agency and the administration of abandoning DACA altogether.

This remarkable passage connects, in a way, to the unsuccessful equal protection claims made in the case. Only Justice Sotomayor credits those claims formally, concluding (in an opinion concurring in part and dissenting in part) that President Trump’s statements denigrating Mexican immigrants, both during the campaign and once in office, created the “strong impression” that the attempted rescission of DACA was “contaminated by impermissible animus.”81 The Court dismisses these claims in two pages, concluding that none of the plaintiffs’ arguments, “either singly or in concert, establishes a plausible equal protection claim.”82 This debate echoes the exchange between these two justices in Trump v. Hawaii, but here the majority opinion at least communicates that the rescission would have been unfair or disruptive, even if legally unavoidable or within the government’s discretion.

Regents thus offers us an example of a phenomenon that abounds in immigration law and other domains where substantive constitutional claims have become elusive due to doctrinal development—the use by litigants and courts of structural doctrines as de facto stand-ins for rights claims.83 The typical lament heard about this turn to structure over substance is that it precludes the development of the substance or otherwise occludes the malign motives of the government, which in turn compounds societal failure to grapple with anti-Muslim, anti-immigrant, and anti-Latino sentiment. But at least in the case of DACA, the evidence cited of anti-Mexican or anti-immigrant sentiment seemed far off of what would have been required to establish an equal protection claim, even as it does not seem controversial to cite racial resentment and nativism as factors relevant to the whole of the Trump administration’s immigration policies.

More to the point, despite being focused on structural questions, both DACA itself and the litigation surrounding its rescission have been political projects to entrench the Dreamers’ claims in public perception and in our governing institutions.84 In retrospect, we may see both DACA and the rescission litigation as bridges to legalization. Both have bought time for DACA recipients, and both have highlighted the simultaneous desert and precarity of the Dreamers and others like them, underscoring the urgency of legalization.85 Chief Justice Roberts’s recitation of the individual and social reliance interests generated by DACA certainly highlights how non-citizens, advocates, and determined officials, through legal forms and legal contestation, have changed the social meaning of unlawful status.86 This sociological development arguably brings us closer to a political transformation of the immigration laws—a transformation that eludes the increasingly unsuccessful constitutional litigation discussed in Part II. Perhaps one day we will be able to characterize the strong assertion of Executive authority in DACA as having disrupted the political economy of immigration lawmaking that for the last generation has given rise mostly to the sorts of punitive, anti-due process enforcement schemes to which the justices increasingly defer.

But even if Regents embodies this kind of triumph for immigrant advocacy, we might still cast a skeptical eye on the way the Court reaches its conclusion precisely because of the way it constrains the government. Had the 2020 election turned out differently, it seems probable that DHS would have jumped through the additional procedural hoops required by the Court and pursued the policy goals it had all along. The question would have become what value those extra procedural steps had, other than inhibiting policy change that everyone in the litigation agreed was legally permissible because DACA was not legally required. Justices Thomas and Kavanaugh, in their separate dissents, make the point that insistence on exacting procedural regularity is at odds with the discretionary nature of the DACA program and what courts ought to expect of agencies seeking to make changes in policy.

In his partial concurrence and partial dissent, Justice Kavanaugh provides an alternate account of Chenery that may not be true to what the decision requires but that provides a more sensible approach to evaluating government action. He would have evaluated the government’s actions based on the Nielsen memorandum, which expressly addressed the reliance interests generated by DACA, provided policy reasons for the rescission alongside the legal ones, and was in fact invited by a district court judge in the litigation. Kavanaugh criticizes what he describes as the Court’s distortion of Chenery, which he characterizes as a constraint on the government’s ability to justify its actions in litigation, not on the agency’s authority to supplement or develop its reasons for a policy through other forms of final agency action.87 As he points out, the Court’s decision seems to permit the agency on remand to “relabel and reiterate” the contents of the Nielsen memo, so what function other than delay (for the agency and the Court’s own reckoning with DACA’s legality) does the decision perform?88 Chief Justice Roberts rejects Justice Kavanaugh’s view, offered by the government, too, that requiring a formally new policy process would be “an idle and useless formality,” arguing that procedural requirements can serve “important values of administrative law,” including giving the parties and the public the opportunity to “respond fully and in a timely manner to an agency’s exercise of authority.”89

In our consideration of whether and how courts might discipline the Executive’s enforcement powers, Adam Cox and I have emphasized the value of importing reason-giving requirements from the mainstream of administrative law but also warned against the imposition of overly wooden procedural requirements that “stand in the way of an incoming regime’s ability to better tailor policy to its own political views.”90 Some scholars have begun to interpret Regents as standing on the right side of this line—as promoting accountability in government.91 Others have suggested that the decision distorts the policy process, making what should otherwise be swift and decisive policy change subject to a slog in the courts.92 In a contemporary context in which the administrative state itself and the deference doctrines that sustain it are under concerted challenge, from both opponents of regulation and proponents of certain formal conceptions of the separation of powers, we have every reason to be skeptical of procedural doctrines that make change within government cumbersome or difficult.93 Indeed, because just about every agency action generates reliance interests of some kind, the way the Court deploys them in Regents could undermine the very act of policy change. And in this particular case, the Court’s accountability rhetoric comes up somewhat empty. Neither DACA itself nor any new policy process adopted to rescind it would have entailed opportunity for public comment or participation. If Secretary Nielsen’s primary mistake was in treating her memo as an outgrowth of the Duke memo, forcing DHS back to the drawing board to start anew is arguably overkill, given that the Nielsen memo from the Secretary more fully elaborated a policy position that in the first instance should have been sufficient to pass court review. Any suggestion to the contrary that the reasons given in her memo if presented in the proper procedural posture would not have been sufficient to justify the government’s policy choice would be ratcheting up the power of the courts to inhibit policy change.

Thus far in the lower courts, Regents has taken its place alongside canonical administrative law cases such as State Farm and Overton Park as a citation for the proposition that an agency must take account of the alternatives available to it.94 But one key litigant—the state of Texas—is behaving predictably, seizing the Regents rationale to attempt to stymie every effort of the new Biden administration to change course from its predecessor. One court in Texas, for example, greeted the new administration with a preliminary injunction of its efforts to shift immigration enforcement policy by pausing removals in order to take stock of what the Trump administration had done to the system.95 Partisan litigants may well find partisan courts to use arbitrary and capricious review to demand ever clearer and more elaborate explanations for their positions – in turn pushing the government into ever more elaborate policy processes – when it arguably ought to be enough for an administration to offer that it was elected to implement a new set of policy views.96 Whether this exacting proceduralism contributes to good and accountable government ought to be an ongoing debate and a point of true reflection for the courts called upon to referee our political disputes.

Conclusion

Despite surviving the concerted efforts of the Trump administration to rescind it, the future of DACA remains uncertain—a state that arguably befits a status that is itself temporary and provides no path to permanent membership, even as it mitigates the unfairness and stress of unlawful status. The optimistic version of the future culminates in the achievement of legalization for the Dreamers and perhaps still others, a political outcome that DACA and the activist energy that produced it will have helped make possible. If these events come to pass, they will in one sense prove DACA’s detractors correct: that with its adoption, President Obama and Secretary Napolitano transformed immigration law by grounding the Dreamers’ claims to membership in official policy. By prompting Executive action, the Dreamers expanded the limits of the law in a way even the most determined of opponents could not undo, even once they controlled the enforcement power. The pessimistic end of the story is of endless conflict and permanent instability, not just for the Dreamers, but across the regulatory state, as the structural tools honed in litigation against the Trump administration are used to stymie the attempts at change by all of its successors, through courts that have become players in the political scrum. It is too early to say that the courts’ vaunted insistence on reasoned decision-making has thrown sand into the government’s gears. But neither Regents nor the Roberts Court’s immigration jurisprudence will fix our immigration system to make it more humane, and we can only hope that the Court will not stand in the way of the political actors seeking to do just that.

Notes

Cristina M. Rodríguez is Leighton Homer Surbeck Professor of Law, Yale Law School.

Author’s note: I am enormously grateful for the outstanding research assistance of Callie Bruzzone, Kayla Crowell, Daniel Esses, and Becca Steele.

1 For the sake of precision, DACA is the acronym for an Obama-era initiative called Deferred Action for Childhood Arrivals, which invited non-citizens lacking lawful immigration status who met certain criteria to apply for relief, which entailed extending the deferral of their removal for two years and rendered them eligible to apply for work authorization and other discrete benefits tied to “lawful presence” as recognized by the Attorney General. See Memorandum from Janet Napolitano, Sec’y, U.S. Dep’t of Homeland Sec., to David V. Aguilar, Acting Comm’r, U.S. Citizenship & Immigr. Servs., et al. (June 15, 2012), https://www.dhs.gov/xlibrary/assets/s1-exercising-prosecutorial-discretion-individuals-who-came-to-us-as-children.pdf (hereinafter Napolitano memorandum).

2 See, e.g., Robert Barnes, Supreme Court Blocks Trump’s Bid to End DACA, a Win for Undocumented ‘Dreamers,’ Wash. Post (June 18, 2020), https://www.washingtonpost.com/politics/courts_law/supreme-court-rules-against-trump-administration-attempt-to-end-daca-a-win-for-undocumented-immigrants-brought-to-us-as-children/2020/06/18/4f0b6c74-b163-11ea-8758-bfd1d045525a_story.html; Alan Gomez, ‘We Won’: DACA Recipients Overwhelmed by Surprise Supreme Court Victory Over Trump, USA Today (June 18, 2020), https://www.usatoday.com/story/news/nation/2020/06/18/dreamers-daca-recipients-celebrate-rare-supreme-court-win-over-trump/3213617001. The appellation “Dreamer” long predates DACA and has been self-consciously adopted by immigrant youth mobilized to claim a legal status that matches their felt sense of belonging—a status that would be made available by the DREAM Act (an acronym for Development, Relief, and Education for Alien Minors), which was first introduced in Congress in 2001 and has remained a mainstay of immigration reform proposals that have attracted support from both major parties but have yet to be enacted.

3 No court has yet declared DACA to be illegal. During the Obama years, both ICE agents and other enforcement-minded officials at the state level challenged DACA’s legality, but the lawsuits were all dismissed. See Crane v. Napolitano, No. 3:12-CV-03247-0, 2013 WL 8211660, at *2 (N.D. Tex. July 31, 2013) (concluding that the Civil Service Reform Act provides “comprehensive and exclusive procedures for settling work-related controversies between federal civil -service employees and the federal government”), aff’d sub nom. Crane v. Johnson, 783 F.3d 244 (5th Cir. 2015); Arpaio v. Obama, 797 F.3d 11, 14 (D.C. Cir. 2015) (holding that Maricopa County Sheriff Joe Arpaio lacked standing to challenge DACA). One lawsuit challenging the legality of DACA, brought by the state of Texas, remains pending in the Southern District of Texas. See infra notes 48 and 84 and accompanying text.

4 Dep’t of Homeland Sec. v. Regents of the Univ. of Cal., 140 S. Ct. 1891 (2020).

5 Napolitano memorandum, supra note 1. This framing arguably reflected the administration’s desire to proceed quickly, not its intention for DACA status to be unstable or unreliable for its participants. By casting it as an exercise of discretion, DHS could avoid the notice and comment requirements applicable to the promulgation of a binding legislative rule under the Administrative Procedure Act.

6 See Regents, 140 S. Ct. at 1905 (“The dispute before the Court is not whether DHS may rescind DACA. All parties agree that it may. The dispute is instead primarily about the procedure the agency followed in doing so.”).

7 140 S. Ct. 1959 (2020).

8 Adam Cox and I have recounted this story in detail, explaining how the Morton Memos departed in important respects from previous administrations’ prioritization memos, and situating DACA in a long history of presidential control over immigration law and developing practices of political officials supervising the bureaucracy. See Adam B. Cox & Cristina M. Rodríguez, The President and Immigration Law 162–90 (2020).

9 The eligibility criteria for DACA applied to non-citizens without immigration status who were under age 31 in 2012; had continuously resided in the United States since 2007; were current students, high school graduates, or honorably discharged veterans; had not been convicted of serious crimes; and did not present national security or public safety threats. See Napolitano memorandum, supra note 1, at 1. In November 2014, DHS attempted to expand these criteria as part of the Deferred Action for Parents of Americans and Lawful Permanent Residents (DAPA)—an expanded deferred action policy quickly enjoined by the courts and later abandoned by the Trump administration. See Memorandum from the Office of Legal Counsel to the Sec’y of Homeland Sec. and the Counsel to the President 10 (Nov. 19, 2014), http://www.justice.gov/sites/default/files/olc/opinions/attachments/2014/11/20/2014-11-19-auth-prioritize-removal.pdf [http://perma.cc/85Y5-N94M] [hereinafter OLC Memorandum Op.].

10 See Number of Form I-821D, Consideration for Deferred Action for Childhood Arrivals, Fiscal Year 2012-2017, U.S. Citizenship & Immigration Servs. (Sept. 30, 2017), https://www.uscis.gov/sites/default/files/document/data/daca_performancedata_fy2017_qtr4.pdf.

11 See, e.g., Gregory Krieg, Trump’s Many Shifting Positions on DACA, from the Campaign to Right Now, CNN (Jan. 25, 2018), https://www.cnn.com/2018/01/25/politics/donald-trump-positions-daca.

12 Letter from Jefferson B. Sessions, Att’y Gen., U.S. Dep’t of Just., to Elaine Duke, Acting Sec’y, Dep’t of Homeland Sec. (Sept. 4, 2017), https://www.dhs.gov/sites/default/files/publications/17_0904_DOJ_AG-letter-DACA.pdf [hereinafter Sessions Letter].

13 Memorandum from Elaine Duke, Acting Sec’y, U.S. Dep’t of Homeland Sec., to James W. McCament, Acting Dir., U.S. Citizenship & Immigr. Servs., et al. (Sept. 5, 2017), https://www.dhs.gov/news/2017/09/05/memorandum-rescission-daca.

14 Texas v. United States, 809 F.3d 134, 166, 179-81 (2015) (affirming a lower court preliminary injunction, concluding that DAPA was “much more than nonenforcement: It would affirmatively confer ‘lawful presence’ and associated benefits on a class of unlawfully present aliens” and that plaintiffs were likely to succeed on their claim that DAPA should have gone through notice and comment rulemaking and that the APA required DAPA’s invalidation because it was “manifestly contrary” to the INA).

15 Regents, 140 S. Ct. 1905–07 (citing Heckler v. Chaney, 470 U.S. 821 (1985)).

16 See Texas, 809 F.3d at 181–82 (concluding that DAPA was “manifestly contrary to the INA” because it “would make 4.3 million otherwise removable” non-citizens eligible to apply for work authorization and receive other benefits).

17 Adam B. Cox & Cristina M. Rodríguez, Don’t Let Trump Hide Behind the Constitution in Ending DACA, Just Security (Sept. 6, 2017), https://www.justsecurity.org/44735/dont-trump-hide-constitution-daca. In the wake of the Regents decision, Benjamin Eidelson has characterized the litigation and the Court’s decision as accountability forcing. See Benjamin Eidelson, Reasoned Explanation and Political Accountability in the Roberts Court, 130 Yale L.J. 1748 (2021).

18 See, e.g., Batalla Vidal v. Nielsen, 279 F. Supp. 3d 401, 437–38 (E.D.N.Y 2018) (granting plaintiffs’ motion for preliminary injunction); but see Casa De Md. v. Dep’t of Homeland Sec., 284 F. Supp. 3d 758 (D. Md. 2018) (concluding that it was reasonable for DHS to have concluded that DACA was unlawful and should be wound down in orderly manner).

19 NAACP v. Trump, 298 F. Supp.3d 209 (D.D.C. 2018).

20 The policy reasons included that DHS should exercise prosecutorial discretion on a case-by-case basis, leaving categorical relief to Congress, and that it was important for DHS to “project a message” that the immigration laws were to be enforced against all categories of violators. Memorandum from Kirstjen M. Nielsen, Sec’y, U.S. Dep’t of Homeland Sec. (June 22, 2018), https://www.dhs.gov/sites/default/files/publications/18_0622_S1_Memorandum_DACA.pdf.

21 Dep’t of Homeland Sec. v. Regents of the Univ. of Cal., 140 S. Ct. 1891, 1908 (2020).

22 332 U.S. 194 (1947).

23 140 S. Ct. at 1908–09 (citing Chenery, 332 U.S. at 201 (1947), and D.C. Circuit precedent establishing that on remand, an agency can either amplify its original reasons or take new agency action that complies with the procedural requirements for new action).

24 Id. at 1912 (“The Attorney General neither addressed the forbearance policy at the heart of DACA nor compelled DHS to abandon that policy. … Duke’s memo offers no reason for terminating forbearance. She instead treated the Attorney General’s conclusion regarding the illegality of benefits as sufficient to rescind both benefits and forbearance, without explanation.”). What the Court characterized as a “unique statutory provision,” id. at 1910, made the Attorney General’s legal conclusions binding on the Secretary. See 8 U.S.C. § 1103(a)(1) (2018).

25 Regents, 140 S. Ct. at 1913–15 (citing Encino Motorcars, LLC v. Navarro, 136 S. Ct. 2117, 2126 (2016), and FCC v. Fox Television Stations, Inc., 556 U.S. 502, 515 (2009)).

26 Letter from William P. Barr, Att’y Gen., to Chad F. Wolf, Acting Sec’y, U.S. Dep’t of Homeland Sec. (June 30, 2020), https://www.dhs.gov/sites/default/files/publications/20_0630_doj_aj-barr-letter-as-wolf-daca.pdf.

27 Memorandum from Chad F. Wolf, Acting Sec’y, U.S. Dep’t of Homeland Sec., to Mark Morgan, Senior Off. Performing the Duties of Comm’r, U.S. Customs & Border Prot. (July 28, 2020), https://www.dhs.gov/sites/default/files/publications/20_0728_s1_daca-reconsideration-memo.pdf.

28 See, e.g., Pangea Legal Servs. v. U.S. Dep’t of Homeland Sec., No. 20-CV-09253-JD, 2021 WL 75756, at *4-5 (N.D. Cal. Jan. 8, 2021) (enjoining a rule altering procedures for asylum and withholding of removal); Batalla Vidal v. Wolf, No. 16CV4756NGGVMS, 2020 WL 6695076, at *9 (E.D.N.Y. Nov. 14, 2020) (granting summary judgment for plaintiffs in challenge to memorandum effectively suspending DACA program pending DHS review); Nw. Immigrant Rts. Project v. U.S. Citizenship & Immigration Servs., No. CV 19-3283 (RDM), 2020 WL 5995206, at *24 (D.D.C. Oct. 8, 2020) (enjoining a rule requiring first-time asylum applicants to pay a fee and reducing the availability of fee waivers), appeal dismissed, No. 20-5369, 2021 WL 161666 (D.C. Cir. Jan. 12, 2021); Immigrant Legal Res. Ctr. v. Wolf, 491 F. Supp. 3d 520, 533 (N.D. Cal. 2020) (enjoining rule implementing fee changes for immigrant benefit requests); Casa de Md., Inc. v. Wolf, 486 F. Supp. 3d 928, 960 (D. Md. 2020) (enjoining rules overhauling an employment authorization scheme for asylum seekers). This improper installation of Wolf and the subsequent invalidation of numerous administration policy initiatives may be the best candidate for the award of administrative incompetence during the Trump years. At the same time, it may have been that the virtue of avoiding the scrutiny of confirmation and the ease of installing preferred agents as acting officials outweighed whatever discrete policy losses the administration suffered. In the case of DACA, unwinding it may have mattered less to the officials most focused on immigration policy than simply signaling hostility to even the most publicly sympathetic immigrants.

29 See Memorandum of January 20, 2021, 86 Fed. Reg. 7053 (Jan. 25, 2021); Statement by Homeland Security Secretary Mayorkas on DACA, Dep’t of Homeland Sec. (Mar. 26, 2021), https://www.dhs.gov/news/2021/03/26/statement-homeland-security-secretary-mayorkas-daca.

30 As the Court in Regents notes, the basic presumption of judicial review embodied in the Administrative Procedure Act “can be rebutted by a showing that the relevant statute ‘preclude[s]’ review, § 701(a)(1), or that the ‘agency action is committed to agency discretion by law,’ § 702(a)(1).” 140 S. Ct. at 1905. Under the Court’s precedent in Heckler v. Chaney, 470 U.S. 821, 831–32 (1985), a decision not to bring an otherwise authorized enforcement action is ordinarily not subject to judicial review because the decision requires balancing a number of factors “peculiarly within [the agency’s] expertise,” mirrors the decision of a prosecutor not to indict, and “as a practical matter” provides no action to focus judicial review. Regents, 140 S. Ct. at 1906 (citing Chaney, 470 U.S. at 831–32).

31 See 8 C.F.R. § 274a.12(c)(14) (2020) (establishing that deferred action recipients are eligible to apply for employment authorization); 8 C.F.R. §1.3(a)(4)(vi) (establishing that recipients of deferred action are “lawfully present” for the purposes of receiving certain Social Security benefits). In addition to these federal regulations, in many states, recipients of deferred action or holders of EADs are eligible under state law to acquire driver’s licenses. Again, this eligibility flows from the deferral of removal and is not part of a holistic regulatory scheme to confer a status on certain non-citizens. See, e.g., Wis. Stat. § 343.14 (2020); Ind. Code § 9-24-9-2.5 (2020); Kan. Stat. Ann. § 8-240 (2019); Tex. Transp. Code § 521.142 (2019); Ariz. Rev. Stat. Ann. § 28-3158 (2020); S.D. Codified Laws §32-12-3.1 (2021); Fla. Stat. § 322.08 (2020); 75 Pa. Cons. Stat. § 1506 (2019).

32 For the government’s position on reviewability in the litigation surrounding DAPA, see Brief for Petitioner at 36, United States v. Texas, 136 S. Ct. 2271 (2016) (No. 15-674) (arguing that DAPA was “an unreviewable exercise of enforcement discretion with unreviewable consequences” under the APA). The Fifth Circuit viewed DAPA as “much more than nonenforcement,” saying that the program would “affirmatively confer ‘lawful presence’ and associated benefits on a class of” unlawfully present noncitizens. Texas v. United States, 809 F.3d 134, 166 (5th Cir. 2015), as revised (Nov. 25, 2015), aff’d by an equally divided court, 136 S. Ct. 2271 (2016). The Fifth Circuit stated that “to be reviewable agency action, DAPA need not directly confer public benefits—removing a categorical bar on receipt of those benefits and thereby making a class of persons newly eligible for them ‘provides a focus for judicial review.’” Id. at 167 (citing Chaney, 470 U.S. at 832). It referred to “DAPA’s issuance of lawful presence and employment authorization” as “affirmative agency action” and concluded that the INA provided enough of a framework that judicial review to determine whether the agency had exceeded its statutory powers was appropriate. Id. at 168.

33 140 S. Ct. at 1906.

34 Id.

35 Even though this conceptualization of DACA as non-enforcement “plus” seems to constrain the Executive, it also in some sense raises the Executive’s stature by treating enforcement judgments as a form of policymaking. This move is of a piece with the way Justice Kennedy treated enforcement priorities as a form of federal policymaking that could preempt state-level enforcement laws that might conflict or interfere with the scope of federal law as defined by Executive enforcement judgments in Arizona v. United States, 567 U.S. 387 (2012). It also accords with the theory of policymaking through enforcement that Adam Cox and I have developed, through which we emphasize how history, legal structure, and administrative practice have transformed supposedly one-off discretionary judgments into a form of systematic policymaking. See Cox & Rodríguez, supra note 8, at 103-32, 191–214.

36 See supra note 24.

37 If we read the Sessions letter as impugning the legality of DACA as a whole, that is, as foreclosing the bifurcation of its parts to save forbearance from a conclusion of legal invalidity, then the Secretary truly would have been required to end DACA altogether in light of that letter, given the Attorney General’s authority to determine the scope of the immigration laws—an authority the Court recognizes. The Court’s options for disposition would then have changed and perhaps become politically less comfortable for the Chief Justice, because it would have been impossible to avoid taking a position on DACA’s legality. As Justice Thomas notes in his partial dissent and concurrence in the judgment, if DACA was not lawful, then reliance interests, no matter how strong, could not have justified keeping it in place. And thus, to send DHS back to the drawing board to better justify its decision to rescind DACA would have required Roberts implicitly to accept DACA’s legality as a whole and to demand a better policy rationale for the rescission, including an explanation for disrupting the reliance interests. By instead bifurcating DACA, Roberts is able to remain agnostic on DACA’s legality, leaving open the possibility of narrowing the Executive’s authority in the future, while rebuffing the Trump administration’s effort to disrupt the lives of the enormously sympathetic Dreamers.

38 809 F.3d 134 (5th Cir. 2015), as revised (Nov. 25, 2015), aff’d by an equally divided court, 136 S. Ct. 2271 (2016).

39 463 U.S. 29 (1983).

40 Dep’t of Homeland Sec. v. Regents of the Univ. of Cal., 140 S. Ct. 1891, 1913 (2020).

41 Id. at 1921 (Thomas, J., concurring in the judgment in part and dissenting in part). Citing both the reasoning of the elephants-in-mouseholes canon and the major-questions doctrine, Justice Thomas, with whom Justices Alito and Gorsuch join, concludes “the detailed statutory provisions governing temporary and lawful permanent resident status, relief from removal, and class-wide deferred-action programs lead ineluctably to the conclusion that DACA is ‘inconsistent with the design and structure of the statute as a whole.’” Id.

42 According to the Fifth Circuit, Texas and the other litigating states “do not challenge the Secretary’s decision to ‘decline to institute proceedings, terminate proceedings, or decline to execute a final order of deportation,’” and nothing in its decision “requires the Secretary to remove any alien or to alter” the agency’s class-based “enforcement priorities.” Texas, 809 F.3d at 168, 166, 169.

43 In his partial dissent, Justice Thomas takes precisely this position. He would have called into question not just the legal authority for DACA but also for other categorical grants of deferred action that defenders of DACA have cited as evidence of past and probative administrative practice. Regents, 140 S. Ct. at 1924, n.6 (Thomas, J., dissenting) (“In the DAPA litigation, DHS noted that some deferred action programs have been implemented by the Executive Branch without explicit legislation. But ‘past practice does not, by itself, create [Executive] power… . Moreover, if DHS has the authority to create new categories of new aliens eligible for deferred action, then all of Congress’ deferred-action legislation was but a superfluous exercise.” (citations omitted)). The district court in the DAPA litigation took a similar position. “Instead of merely refusing to enforce the INA’s removal laws against an individual, the DHS has enacted a wide-reaching program that awards legal presence … to individuals Congress has deemed deportable or removable.” Texas v. United States, 86 F. Supp. 3d 591, 654 (S.D. Tex. 2015), aff’d, 809 F.3d 134 (5th Cir. 2015), as revised (Nov. 25, 2015), aff’d by an equally divided court, 136 S. Ct. 2271 (2016). Some commentators have maintained this position, too. See, e.g., Zachary Price, Enforcement Discretion and Executive Duty, 67 Vand. L. Rev. 671, 688–96 (2014).

44 Cox & Rodríguez, supra note 8, at 178–79.

45 Adam B. Cox & Cristina M. Rodríguez, The Supreme Court’s Ominous DACA Decision: Perils for Dreamers in What Comes Next, Just Security (June 22, 2020), https://www.justsecurity.org/70956/the-supreme-courts-ominous-daca-decision-perils-for-dreamers-in-what-comes-next.

46 The rationale for these longstanding notice-and-comment rules was to ensure that recipients of deferred action, whose presence the government had chosen to tolerate for some period of time, would have the means to sustain themselves while the government tolerated their presence in the country despite their lack of lawful status. An animating concern of the courts that have rejected categorical work authorization is that DACA seemed to go beyond what was intended by the work authorization originally—it’s one thing to say small numbers can be authorized to work, quite another to do so for hundreds of thousands when Congress has clearly prohibited employers from hiring people not authorized to work.

47 See Regents, 140 S. Ct. at 1929 n.14 (Thomas, J., concurring in the judgment in part and dissenting in part).

48 When DACA was first announced in 2012, legal challenges to it quickly emerged, but none of them gained traction. One lawsuit by ICE officials who claimed that the policy required them to violate their oaths to protect and defend the Constitution was dismissed for lack of subject-matter jurisdiction on the ground that this personnel complaint had to proceed through an alternate administrative system. And in a second lawsuit brought by Sheriff Joseph Arpaio of Maricopa County, Arizona, the D.C. Circuit also rejected the case on justiciability grounds, finding that Arpaio lacked standing to challenge the federal policy. Perhaps because the litigation over DAPA consumed the attention of litigants hostile to immigration relief (and to the Obama administration) for a period of years, Texas and the other usual suspects at the state level did not organize to challenge DACA until the Trump administration showed early ambivalence about whether to rescind it. See supra note 3.

49 As was predicted and hoped, DACA has had substantial stabilizing benefits for its recipients. See Tom K. Wong et al., DACA Recipients’ Livelihoods, Families, and Sense of Security Are at Stake This November, Ctr. for Am. Progress (Sept. 19, 2019), https://www.americanprogress.org/issues/immigration/news/2019/09/19/474636/daca-recipients-livelihoods-families-sense-security-stake-november (reporting on results of a survey of over 1,000 DACA recipients that documents their improved lives, including that 59 percent of respondents moved to a job with better pay, 48 percent to a job with better working conditions, 53 percent to a job that better fits their education and training, and 53 percent to a job with health insurance or other benefits). And yet it remains a contingent and precarious status, as the rescission effort and ongoing lawsuits challenging its legality underscore. See Cox & Rodríguez, supra note 8, at 241–42 (citing studies highlighting the precarity of discretionary statuses).

50 As Justice Kennedy put it at oral argument, DAPA turns the Constitution “upside down,” because the Executive has claimed authority that belongs only to Congress. Transcript of Oral Argument at 24, United States v. Texas, 136 S. Ct. 2271 (2016) (No. 15-674). One feature of the DAPA litigation that did not arise in Regents but may complicate the government’s defense of DACA moving forward is that the extension of Social Security benefits, in particular, arises by virtue of an Attorney General decision in the 1990s to treat recipients of deferred action as having “lawful presence” for the purposes of certain benefits under a 1996 companion law to the public benefits reforms of that era known as the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA). Unlike the 2012 DACA memorandum issued by Secretary Napolitano, the memorandum from Secretary Jeh Johnson announcing DAPA adverted to this term of art and as a result sowed massive confusion and consternation in the lower courts, as well as at oral argument at the Supreme Court. How could the government maintain that deferred action recipients had lawful presence for one purpose but were not given lawful status more generally, something only Congress has the authority to do? In fact, the issue is a red herring: the term “lawful presence” as used in DAPA and DACA means eligibility for benefits, not lawful immigration status.

51 In the case of DACA, invalidating it on legal grounds could call into question longstanding Executive branch practice through which administrations have developed case-management and programmatic techniques to manage the massive amount of enforcement discretion the law effectively delegates to the Executive by virtue of rendering so many millions of non-citizens deportable. See Cox & Rodríguez, supra note 8, at 198–201.

52 138 S. Ct. 2392 (2018).

53 Presidential Proclamation No. 9645, 82 Fed. Reg. 45,161 (Sept. 24, 2017).

54 I have written about this case at length elsewhere. See Cristina M. Rodríguez, Trump v. Hawaii and the Future of Presidential Power over Immigration, Am. Const. Soc’y Sup. Ct. Rev., 2017–2018, at 161 (2018), https://www.acslaw.org/wp-content/uploads/2019/01/ACS-Supreme-Court-Review-2018-Revised.pdf.

55 8 U.S.C. § 1182(f) (2018).

56 Among the problems with these sorts of arguments is that they embody a logic that would invalidate a raft of Executive branch practices of the last several decades undertaken to enforce the INA, including, arguably, DACA itself. See Rodríguez, supra note 54, at 170–73.

57 Id. at 179.

58 For development of this argument, see Adam B. Cox, Ryan Goodman & Cristina Rodríguez, The Radical Supreme Court Travel Ban Opinion—But Why It Might Not Apply to Other Immigrants’ Rights Cases, Just Security (June 27, 2018), https://www.justsecurity.org/58510/radical-supreme-court-travel-ban-opinion-but-apply-immigrants-rights-cases.

59 Rodríguez, supra note 54, at 166.

60 140 S. Ct. 1959 (2020).

61 Under a statutory provision enacted in 1996, an applicant for admission is subject to expedited removal if the applicant is inadmissible for lacking a valid entry document, has not been physically present in the United States for more than two years, or has been designated by the Secretary for expedited removal. 8 U.S.C. § 1225(b)(1)(A)(i), (iii)(I)-(II) (2018). If an immigration officer determines that the person is inadmissible, the officer must order the non-citizen removed without “further hearing or review.” Id. § 1225(b)(1)(A)(i). An applicant who expresses an intention to apply for asylum is referred to a screening to determine if he has a “credible fear of persecution.” Id. § 1225(b)(1)(B)(v). If the screening officer finds no credible fear, the non-citizen is subject to expedited removal. A non-citizen who shows a credible fear is referred for full consideration of the asylum claim. 8 C.F.R. § 208.30(f) (2020).

62 8 U.S.C. § 1252(e)(2) (limiting habeas review to “whether the petitioner is an alien,” was ordered removed, or already was granted entry as a lawful permanent resident).

63 For three excellent treatments of the case, see Gerald Neuman, The Supreme Court’s Attack on Habeas Corpus in DHS v. Thuraissigiam, Just Security (Aug. 25, 2020), https://www.justsecurity.org/72104/the-supreme-courts-attack-on-habeas-corpus-in-dhs-v-thuraissigiam (noting, in particular, how the Court departs from its seminal habeas decision in Boumediene v. Bush); Ahilan Arulanantham & Adam Cox, Immigration Maximalism at the Supreme Court, Just Security (Aug. 11, 2020), https://www.justsecurity.org/71939/immigration-maximalism-at-the-supreme-court (describing Thuraissigiam as demonstrating the Court’s immigration maximalism, in contrast to Regents, and showing how the case reflects disregard for the interests of non-citizens in ways that deviate from past precedent); and Amanda L. Tyler, Thuraissigiam and the Future of the Suspension Clause, Lawfare (July 2, 2020), https://www.lawfareblog.com/thuraissigiam-and-future-suspension-clause (noting how the opinion calls into question aspects of Boumediene v. Bush and expresses a more narrow view of the Suspension Clause than embodied in that case).

64 Justice Alito increasingly has become the voice of the conservative majority of a sharply divided Court on immigration, having authored five significant opinions in the last three years. In addition to Thuraissigiam, most notable are Nielsen v. Preap, 139 S. Ct. 954 (2019), in which the Court holds that the INA’s mandatory-detention provision applies to all non-citizens convicted of predicate crimes and not just to those immediately taken into custody from the state by DHS, and Jennings v. Rodriguez, 138 S. Ct. 830 (2018), in which the Court declines to read the INA’s mandatory-detention provisions as time-limited or necessitating bond hearings. As a judge on the Third Circuit, Alito took a “harder line” on criminal and immigration cases than even other Republican appointees on the courts of appeals. See Amy Goldstein & Sarah Cohen, Alito, In and Out of the Mainstream: Nominee’s Record Defies Stereotyping, Wash. Post (Jan. 1, 2006), https://www.washingtonpost.com/archive/politics/2006/01/01/alito-in-and-out-of-the-mainstream-span-classbankheadnominees-record-defies-stereotypingspan/2ba042c7-773e-48da-9056-14300cea054e; Cong. Rsch. Serv., RL33218, Immigration: Selected Opinions of Judge Samuel Alito (2006) (offering an extensive survey of then-Judge Alito’s majority and dissenting opinions in immigration cases heard by the Third Circuit).

65 140 S. Ct. at 1964–65.

66 Justice Alito notes that there has been an almost 2,000 percent increase in credible-fear claims and that the majority of them have “proven to be meritless.” He also characterizes expedited removal as “protecting” the Executive’s discretion from “undue” interference by the courts. Id. at 1966.

67 Id. at 1970.

68 Id. at 1970–1971.

69 Id. at 1990 (Breyer, J., concurring) (“Respondent, to be sure, casts the brunt of his challenge to this adverse credible fear determination as two claims of legal error. But it is the factual findings underlying that determination that respondent, armed with strong new factual evidence, now disputes.”).

70 See Neuman, supra note 63 (“The expectation that the court would give DHS a second chance to decide petitioner’s case lawfully, rather than ordering his immediate release, follows longstanding practice in immigration law and other fields.”). Alito addresses Boumediene simply by noting that it “is not about immigration at all,” and that nothing in the Court’s decision suggests those held at Guantanamo could have used a petition in the way Thuraissigiam sought to—in order to gain entry into the United States. Thuraissigiam, 140 S. Ct. at 1981. He similarly dismisses the relevance of St. Cyr in three paragraphs as not addressing the sort of claim respondent sought to raise. Id. For an outstanding treatment of St. Cyr in the larger context of habeas jurisprudence, see Richard H. Fallon, Jr. & Daniel J. Meltzer, Habeas Corpus Jurisdiction, Substantive Rights, and the War on Terror, 120 Harv. L. Rev. 2029, 2099 (2007) (“In St. Cyr, the Court came close to holding that the petitioner, an alien seized and detained within the United States, was entitled to review not merely of constitutional but also statutory questions underlying his claim of unlawful detention… . St. Cyr’s understanding that the core function of habeas review is to ensure judicial determination of fundamental legal issues—statutory as well as constitutional—is not unique.”). See also Gerald H. Neuman, Habeas Corpus, Executive Detention, and the Removal of Aliens, 98 Colum. L. Rev. 961, 967-1021, 1044–45 (1988) (arguing that historical evidence supports the interpretation of the Suspension Clause as extending to proceedings for the removal of aliens from the United States by immigration officials and that “habeas corpus does aim at detentions ancillary to other deprivations [removal orders], and not only at pure detention orders”).

71 Thuraissigiam, 140 S. Ct. at 1993 (Sotomayor, J., dissenting).

72 Id. at 1982 (majority opinion).

73 See 8 U.S.C. § 1225(b)(1)(A)(iii)(II) (2018).

74 138 S. Ct. 830 (2018). At issue were 8 U.S.C. § 1125(b) (applying to individuals arriving at the border who seek admission to the United States); 8 U.S.C. § 1226(c) (applying to non-citizens convicted of enumerated crimes); and 8 U.S.C. § 1231(a) (applying to non-citizens who have been ordered removed and “shall” be detained for up to ninety days while government effectuates their removal).

75 Zadvydas v. Davis, 533 U.S. 678, 692 (2001) (“The serious constitutional problem arising out of a statute that, in these circumstances, permits an indefinite, perhaps permanent, deprivation of human liberty without any such protection is obvious.”).

76 This is not to say that the Court’s opinions in recent years have blithely accepted every last government argument about the scope of its authority under the INA. In Sessions v. Dimaya, 138 S. Ct. 1204 (2018), for example, the Court found a provision of the INA unconstitutionally vague. In an opinion written by Justice Kagan, the Court rejected the government’s argument that a more lenient form of the void for vagueness doctrine should apply in this civil context. In his concurrence, Justice Gorsuch writes a kind of treatise on the importance of the vagueness doctrine and the provision of clear notice by the law to avoid arbitrary power, id. at 1223–34 (Gorsuch, J., concurring in part and concurring in the judgment), and in so doing he rejects a conclusion embraced by Justice Thomas, id. at 1245–48 (Thomas, J., dissenting), that non-citizens at the time of the Founding did not possess rights under the Due Process Clause.

For a reading of Jennings and other cases that suggest a possible “new era for immigration exceptionalism” that does not bode well for immigrant interests, see David Rubenstein, The Future of Immigration Exceptionalism, scotusblog (June 29, 2017, 2:29 PM), https://www.scotusblog.com/2017/06/immigration-symposium-future-immigration-exceptionalism. For a more extended discussion of these themes, see David S. Rubenstein & Pratheepan Gualsekaram, Immigration Exceptionalism, 111 Nw. U. L. Rev. 583 (2017), which synthesizes the courts’ treatment of immigration exceptionalism with respect to rights, federalism, and the separation of powers and argues that the Court sometimes but not always treats immigration as exceptional, and that litigants sometimes, but not always, want it that way, but that this approach is unstable and requires coming to terms with trade-offs.

77 I have explored the arc of the Court’s treatment of due process claims in immigration cases elsewhere. See Rodríguez, supra note 54, at 198–208.

78 Dep’t of Homeland Sec. v. Regents of the Univ. of Cal., 140 S. Ct. 1891, 1913 (2020).

79 Id. at 1914 (noting that the rescission would radiate outward to DACA recipients’ families, their U.S.-citizen children, the schools where they study and teach, to their employers and the labor force more generally, resulting in the loss of billions of dollars in economic activity and tax revenue).

80 Id. at 1913. The Court does note that this feature of DACA might go to the weight the agency gives to the reliance interests.

81 Regents, 140 S. Ct. at 1917 (Sotomayor, J., concurring in part, concurring in the judgment, dissenting in part).

82 Id. at 1915 (majority opinion).

83 See, e.g., Jennifer M. Chacón, The Inside-Out Constitution: Department of Commerce v. New York, 2019 Sup. Ct. Rev. 231, 236 (noting that the litigation over the 2020 census exemplifies how “substantive-rights claims are increasingly being vindicated, if they are vindicated at all, through procedural channels”); Jenny-Brooke Condon, The Preempting of Equal Protection for Immigrants?, 73 Wash. & Lee L. Rev. 77, 84 (2016) (arguing that federalism has transformed “equal protection doctrine involving state alienage classifications into a preemption-like inquiry”); Geoffrey Heeren, Persons Who Are Not the People: The Changing Rights of Immigrants in the United States, 44 Colum. Hum. Rts. L. Rev. 367, 375 (2013) (describing the shift away from jurisprudence assessing laws regulating migrants under a rights-based analysis to “one focused on the rights and responsibilities of the federal government”).

84 DACA’s structural move is to emphasize the proper supervision and the distribution of enforcement resources, and the litigation focused on how the government organizes its policy process. The long arc of the DACA and DAPA litigation may even have changed the perceptions of some of the judicial participants. When grappling post-Regents with the challenge to DACA’s legality brought by the state of Texas, Judge Andrew Hanen of the Southern District of Texas, who emphatically found DAPA unlawful, appeared reticent, at least for a time, to apply the same analysis to DACA. At one point in the litigation, he called for a status conference on the impact of the American Dream and Promise Act, a bill introduced in the House of Representatives that contained a path to legal status for DACA recipients, on the litigation. American Dream and Promise Act of 2021, H.R. 6, 117th Cong. (2021), https://www.congress.gov/bill/117th-congress/house-bill/6/text. This move seemed at the time like a delay tactic. At the conference Judge Hanen asked for further submissions with a deadline of April 9, 2021. As this essay went to press, Judge Hanen issued his decision, citing Regents and concluding that DACA violated the Administrative Procedure Act and exceeded DHS’s statutory authority. Texas v. United States, No. 18–cv–00068, 2021 WL 3025857, at *20 (S.D. Tex. July 16, 2021), appeal filed, No. 21–40680 (5th Cir. Sept. 16, 2021).

85 Even in dissent, Justice Kavanaugh adverts to the uncertainty Dreamers face and presumes that the solution would be action by Congress to address their uncertainty. Even though he leaves it to Congress to decide their fate “one way or another,” this passage evokes sympathy rather than suspicion. Regents, 140 S. Ct. at 1935 (Kavanaugh, J., concurring in the judgment in part and dissenting in part).

86 Id. at 1913–15 (majority opinion).

87 Id. at 1934 (Kavanaugh, J., concurring in the judgment in part and dissenting in part).

88 Id. at 1935.

89 Id. at 1909 (majority opinion).

90 Cox & Rodríguez, supra note 8, at 230; see also id. (“An expectation of reason giving could prompt courts to ask various questions of an enforcement policy. Does the policy reflect deliberation within the agency or the administration, such that the outcome reflects considered judgments …? Was an agency’s choice to shift its enforcement strategy based on values that can be defended as public-regarding and legitimate? … [T]his reason-giving approach should re-enforce the complexity of governance to the Executive, helping to sustain its culture of deliberation grounded in evidence and sound judgment.”).

91 Benjamin Eidelson, for example, argues that the Court “recast Chenery as an accountability-forcing tool” in response to the agency’s post-hoc rationalizations in the Nielsen memo. See Eidelson, supra note 17, at 1768. He posits that the Administration’s handling of the rescission, including by issuing the Nielsen memo as a follow-on to the Duke memo rather than a new agency action, id. at 1765–66, was motivated by a desire to avoid public consequences of an unpopular policy choice. He notes that although the policy statements in the Nielsen memo gave the Court reason to believe that the agency would not reach a different decision if it redid the action properly, the Court nevertheless chose to force the agency to face public scrutiny. Id. at 1769–70. Other commentators have suggested that the decision is democracy forcing. Regents, in particular, implicates an anti-regulatory strategy one scholar has called statutory abnegation, according to which agencies disclaim statutory authorities on which they had previously relied—a strategy courts typically resist in order to promote various forms of political accountability. See William W. Buzbee, Agency Statutory Abnegation in the Deregulatory Playbook, 68 Duke L.J. 1509 (2019); see also Eidelson, id. at 1777–81 (arguing that Regents does not let the Trump administration claim Congress has made the decision that the administration itself is making).

92 See Zachary Price, Symposium: DACA and the Need for Symmetric Legal Principles, scotusblog (June 19, 2020, 3:51 PM), https://www.scotusblog.com/2020/06/symposium-daca-and-the-need-for-symmetrical-legal-principles (arguing that Regents engages in ad hoc reasoning driven by a strategic judgment that departs from courts’ typical reluctance “to recognize any estoppel based on government assurances … and claims of reliance based on forbearance,” which also has the effect of freezing policies in place).

93 See Gillian Metzger, The Supreme Court, 2016 Term – Foreword: 1930s Redux: The Administrative State Under Siege, 131 Harv. L. Rev. 1 (2017).

94 See, e.g., Invenergy Renewables LLC v. United States, 476 F. Supp. 3d 1323, 1344–45 (Ct. Int’l Trade 2020) (relying on Regents as an outline of the established principles courts should focus on when they analyze agency explanations of their actions); MediNatura, Inc. v. FDA, No. CV 20-2066 (RDM), 2020 WL 6262121, at *27–28 (D.D.C. Oct. 23, 2020) (making ample use of Regents for analyzing the FDA’s change in position and its impact on homeopathic drug companies’ reliance interests).

95 Texas v. United States, No. 6:21-CV-00003, 2021 WL 723856 (S.D. Tex. Feb. 23, 2021). For critical commentary warning that this case is an opening salvo in an effort to use administrative law to stymie the Biden administration, see Noah Feldman, Biden Didn’t Deserve to Lose That Immigration Case, Bloomberg Opinion (Jan. 27, 2021, 12:30 PM EST), https://www.bloomberg.com/opinion/articles/2021-01-27/biden-s-immigration-order-lost-in-court-but-didn-t-deserve-to.

96 This development also underscores how the DAPA litigation changed the landscape by recognizing states’ standing to sue the federal government for its enforcement policies—a development that commentators at the time warned would open the government up to vexatious litigation and stymie policy change. For excellent examples of the growing literature on how state attorneys general have been using their litigating and enforcement powers to shape national policy, see Jessica Bulman-Pozen, Federalism All the Way Up: State Standing and “The New Process Federalism,” 105 Calif. L. Rev. 1739 (2017); and Mark L. Earley, “Special Solicitude”: The Growing Power of State Attorneys General, 52 U. Rich. L. Rev. 561, 564–65 (2018).