United States v. Texas, Regents, and the Roberts Legacy on DACA

Earlier this month, Judge Andrew Hanen—the Bush-appointed judge in Texas who, back in 2015, Republican state elected officials handpicked to give legal effect to their political attacks on the Obama administration’s immigration policies—issued his latest ruling invalidating DACA, the Deferred Action for Childhood Arrivals initiative. Compared to earlier episodes in the long-running litigation over the Obama-era deferred action initiatives, the responses to Hanen’s latest ruling have seemed somewhat muted—not least, perhaps, because the outcome was entirely expected. News organizations and commentators have devoted relatively limited attention to Hanen’s ruling, and statements criticizing the decision have seemed to carry a certain amount of resigned fatigue. DACA’s most outspoken political opponents have also seemed mostly preoccupied by other matters. Even Hanen himself seemed to lack in this decision the kind of big xenophobic energy that he often exhibits in his immigration-related opinions.

Hanen’s most recent decision is his fourth significant pronouncement declaring Obama-era deferred action initiatives unlawful, and the decision rests on more or less the same flawed legal premises as each of his previous rulings. Back in 2015, as I discussed and analyzed at length both here at Dorf on Law and subsequently in an article in the UCLA Law Review Discourse, Republican state elected officials successfully steered a lawsuit to Hanen—the only active judge hearing cases in his south Texas judicial division, and one with an extensive, well-documented record of anti-immigrant hostility—to challenge the legality of the Obama administration’s 2014 efforts to expand DACA and create a new initiative, DAPA, which would have made deferred action available to additional categories of undocumented immigrants. Hanen quickly enjoined the 2014 initiatives, and that ruling was validated twice by divided Fifth Circuit panels. In each instance, the Fifth Circuit opinions were written by Judge Jerry Smith: first, when the government sought a stay pending appeal, and subsequently, when the government appealed the merits of Hanen’s preliminary injunction decision. (I discussed and criticized the Fifth Circuit’s rulings here and here.) After the Supreme Court granted certiorari in 2016, it divided 4-4, thereby establishing no precedent but leaving the Fifth Circuit’s decision in place. As a result, the Obama administration never implemented DAPA, and the Trump administration formally terminated the initiative after assuming power in 2017. The original 2012 DACA initiative, however, was never subject to serious legal challenge and remained in place.

Hanen’s second encounter with the Obama-era deferred action initiatives came in the midst of the Trump administration’s efforts to terminate DACA. As lawsuits proceeded in other jurisdictions seeking to block DACA’s termination, Republican state elected officials returned to Hanen seeking essentially the opposite relief: a ruling declaring that DACA is unlawful and an injunction requiring the initiative to be immediately terminated. Hanen delivered on the first request, writing a lengthy opinion concluding that DACA violated the Administrative Procedure Act both procedurally, for not having been created through notice-and-comment rulemaking, and substantively, because it was purportedly not authorized by the immigration statutes. The opinion relied upon essentially the same reasons as those upon which the Fifth Circuit had relied when invalidating DAPA. However, Hanen declined to enjoin the initiative. Ultimately, of course, the Supreme Court vacated the Trump administration’s rescission of DACA in 2020 when—by a 5-4 margin, with Chief Justice John Roberts writing for the majority—it decided Department of Homeland Security v. Regents of the University of California.

Immediately upon assuming office in 2021, President Biden issued an executive order directing officials to take steps to “preserve and fortify” DACA. While those efforts were in progress, Republican officials rushed back to Hanen yet again, and in July 2021 he ruled yet again that in its original form, DACA violated the APA both procedurally and substantively. A Fifth Circuit panel consisting of Judges Priscilla Richman, James Ho, and Kurt Engelhardt affirmed Hanen’s ruling. However, while the appeal was pending, the Biden administration completed a notice-and-comment rulemaking process to implement DACA through regulations scheduled to take effect in October 2022. It was this more recent, rulemaking-based incarnation of DACA that Hanen deemed unlawful in his fourth, latest deferred action ruling earlier this month, relying again on many of the same legal arguments that have been leveled against the initiative for many years at this point. While the use of notice-and-comment rulemaking took the procedural APA arguments against DACA off the table, Hanen nevertheless concluded that the new incarnation of DACA “is flawed for the same substantive reasons as the 2012 DACA Memorandum.” As in 2021, however, Hanen nevertheless stayed his order as to existing DACA beneficiaries, thereby permitting the Biden administration to accept, process, and grant their renewal applications.

A certain amount of Andrew Hanen fatigue, therefore, seems entirely understandable. Reading Hanen’s lengthy, often strident opinions on the deferred action initiatives has long had a Groundhog’s Day quality, since in many respects the core substantive issues in the litigation have not changed very much over the years. As a result, there often has seemed very little to be surprised about or new to say about Hanen’s decision-making. However, coming after the Supreme Court’s decision in Regents, Hanen’s most recent two rulings do provide occasion to say something about the politicized immigration decision-making of John Roberts, for they draw attention to some of the ways in which Roberts’s opinion—celebrated by many DACA supporters at the time for preserving the initiative—may in fact now be contributing to DACA’s demise.

That should not be altogether surprising. As Professors Adam Cox and Cristina Rodríguez wrote soon after the case was decided, Roberts’s opinion was “as ominous as it [was] encouraging,” insofar as it appeared to provide a “roadmap” for DACA to be successfully terminated and to create barriers to “a future Biden administration’s ability to pursue a humanitarian path going forward.” (In a subsequent law review article, Professor Rodríguez suggested that the hurdles created by Regents might also “undermine the very act of policy change” for future incoming administrations. The Regents-based claims made in the flood of lawsuits filed by Republican state officials seeking to entrench and insulate Trump’s immigration agenda from his successor’s administration suggests that Rodríguez’s concern has been borne out to some extent, although those politicians almost certainly would have charged forward with their lawsuits even if Regents had never been decided.)

Three years later, conservative lower court judges like Hanen and Richman have indeed read Roberts’ opinion to bolster their anti-DACA positions—perhaps not in precisely the manner that Professors Cox and Rodríguez predicted in every last particular, but certainly to similar effect. Professors Cox and Rodríguez anticipated a scenario in which future administrations, litigants, or courts would draw upon Regents to argue that an initiative like DACA could permissibly carve out and confer deferred action itself—a longstanding means of exercising prosecutorial discretion, which provides tentative, discretionary forbearance from enforcement action—or something like it, but could not permissibly provide recipients of deferred action with other collateral “benefits,” such as the ability to apply for employment authorization, Social Security, or Medicare.

At least so far, anti-immigration litigants and judges such as Hanen have not taken up Roberts’ invitation to declare victory by accepting a hollowed-out version of DACA along these lines. Instead, they have invoked Roberts’ opinion more opportunistically and selectively to reinforce their existing, full-throated arguments calling DACA (and by extension, similar initiatives) more fundamentally into question altogether. That, too, should not be particularly surprising.

I.

Start with the basis on which Roberts concluded that the Trump administration’s rescission of DACA was subject to arbitrary-and-capricious review under the APA in the first place. The Trump administration’s effort to terminate DACA involved two steps. First, Attorney General Jeff Sessions sent a vague, one-page letter to Acting DHS Secretary Elaine Duke informing her that DHS “should” rescind DACA because, Sessions asserted, the initiative had been “effectuated . . .  without proper statutory authority” and in a manner that “circumvent[ed the] immigration laws.” Sessions further opined that “[b]ecause” DACA (in his view) had the same alleged “legal and constitutional defects” that Hanen and the Fifth Circuit had “recognized as to DAPA,” it was likely that “potentially imminent litigation would yield similar results with respect to DACA.” Second, upon receipt of that letter, Duke issued a guidance memo announcing—based solely on the legal position in Sessions’s letter, together with the Fifth Circuit’s ruling in the DAPA litigation—that she had decided to terminate DACA “in the exercise of [her] authority.” Neither document, however, made at all clear or even really attempted to explain precisely what DACA’s supposed legal or constitutional infirmities ostensibly were.

When faced with litigation challenging this decision, the Trump administration maintained that the decision to terminate DACA fell within the APA’s narrow exception precluding review of agency actions that are “committed to agency discretion by law,” for essentially the same reasons as the nonenforcement decision at issue in the Supreme Court’s landmark decision in Heckler v. Chaney. Like the FDA decision not to take enforcement action that was at issue in Chaney, the Trump administration argued, DACA similarly involved a “discretionary policy of nonenforcement” that was “unsuitabl[e] for judicial review,” since it similarly “involve[d] a complicated balancing of a number of factors which are peculiarly within [the agency’s] expertise,” including “whether agency resources are best spent on this violation or another” and whether enforcement “best fits the agency’s overall policies.”

The lower courts, however, declined to apply Chaney’s presumption of nonreviewability to the decision to rescind DACA, noting that Chaney had expressly reserved the question of whether judicial review would be available for nonenforcement decisions based on the agency’s view that it lacked jurisdiction or power to act. In the District of Columbia, U.S. District Judge John Bates applied D.C. Circuit precedent addressing that reserved question and holding that Chaney’s presumption does not apply to “a legal interpretation phrased as a general enforcement policy, even if that interpretation concerns the scope of the agency’s lawful enforcement authority.” In the Northern District of California, Judge William Alsup reached a similar conclusion, and the Ninth Circuit affirmed this conclusion on appeal, expressly holding in an opinion by Judge Kim Wardlaw that “an agency’s nonenforcement decision is outside the scope of the Chaney presumption—and is therefore presumptively reviewable—if it is based solely on a belief that the agency lacked the lawful authority to do otherwise.”

In fact, the government evidently conceded in the California litigation that “if the agency’s interpretation of a statute is embedded in a non-reviewable enforcement policy, the former may be reviewable as such.” Ultimately, however, the Trump administration argued—unsuccessfully in all of the lower courts—that the decision to rescind DACA should not be characterized or understood as a legal interpretation phrased as a general enforcement policy, in part because Sessions’s cursory legal position did not refer to any specific statutory provisions. The lower courts rejected this argument, which essentially (and rather shamelessly) sought to capitalize and make a virtue of the vagueness and insufficiency of Sessions’s legal explanation as a basis for insulating that position from judicial review. As Bates noted in his opinion, “[t]he Court fails to perceive any meaningful difference between an agency’s conclusion that it lacks statutory authority and its interpretation of a specific statutory provision.” Writing for the Ninth Circuit, Wardlaw pointedly stated that the court “take[s] Attorney General Sessions literally at his word when he wrote to Acting Secretary Duke that ‘DACA was effectuated . . . without proper statutory authority,’ and that DACA ‘was an unconstitutional exercise of authority by the Executive Branch.’”

In Regents, however, Roberts approached the issue of reviewability rather differently. While he agreed that the rescission decision was reviewable under the APA, Roberts chose not to follow the approaches taken by any of the lower courts to reach that conclusion. Instead, Roberts maintained that DACA fell outside of Chaney’s presumption of nonreviewability for nonenforcement decisions because it was “not simply a non-enforcement policy” at all. Rather, Roberts insisted, DACA was a “a program for conferring affirmative immigration relief,” creating an application submission and review process that yields a set of “proceedings” that are “effectively ‘adjudicat[ions].’” Moreover, Roberts continued, the fact that DACA’s beneficiaries become eligible to apply for employment authorization and receive benefits under Social Security and Medicare “provide[d] further confirmation” that “DACA is more than simply a non-enforcement policy” and therefore is subject to review under the APA.

This line of reasoning—evidently discussed by none of the lower courts, and urged by none of the parties in their briefs—hardly was necessary to find the decision to rescind DACA reviewable. In fact, it was not even the most straightforward way to do so. After all, as the plaintiffs in the D.C. litigation argued to the Supreme Court, since Sessions’s memo made a legal determination that DACA was unlawful, Duke’s decision to terminate the initiative expressly on the basis of that legal determination did not involve any exercise of discretion at all. Since a provision in the immigration statute, 8 U.S.C. § 1103(a)(1), makes “determination[s] and ruling[s] by the Attorney General with respect to all questions of law . . . controlling” on other executive officials and agencies, Duke did not have any discretion to “maintain a policy that she was bound to treat as illegal.” It would have been “nonsensical,” they argued, “to say that an agency’s disavowal of its own discretion is ‘committed to agency discretion’” under the APA. Justice Ruth Bader Ginsburg picked up on this point at oral argument, noting a “strange element” in the Trump administration’s argument:

[Y]ou're arguing this is a discretionary matter; it's not reviewable because it's committed to agency discretion. But, on the other hand, you say the agency had no discretion because this program was illegal. In other words, the law requires you to drop DACA. So how can it be committed to your discretion when you're saying we have no discretion; this is an illegal program?

Perhaps Roberts simply wanted to avoid addressing the issues reserved in Chaney, although those issues were addressed and resolved in the lower courts, squarely presented to the Supreme Court, and expressly discussed at oral argument—including in response to Ginsburg’s question. Nor does Roberts’s opinion in Regents at any point say that the Court was, in fact, seeking to avoid addressing those reserved issues.

What Roberts’s approach did allow him to do was to nonchalantly—and inaccurately—recharacterize DACA as being not a mechanism by which agency officials exercise prosecutorial discretion in the form of deferred action, but rather a more comprehensive, aggregated package of intertwined “benefits” in a manner that approximates lawful immigration status. Roberts’s characterization of deferred action under DACA closely tracks the characterizations made by DACA’s political opponents, and draws directly from the Fifth Circuit’s opinion enjoining DAPA, in which Smith similarly insisted—using almost the exact same words—that deferred action “is more than nonenforcement. It is the affirmative act of conferring ‘lawful presence’ on a class of unlawfully present aliens.” However, as I previously explained in the midst of the DAPA litigation, it is not correct, legally and precisely speaking, to say that the guidance memos creating either DACA or DAPA, by their own terms, operate to confer benefits upon anyone in the manner that Smith intimates. While it is true that the recipients of deferred action under those initiatives can become eligible for various benefits, those legal consequences are disaggregated, piecemeal, and collateral to the exercise of enforcement discretion—the result of other legal authority long predating DACA and DAPA in most cases, not the result of the criteria and frameworks for exercising discretion under the DACA and DAPA memos themselves.

At a minimum, Roberts’s characterization of DACA is highly contested—indeed, his assertion goes to factual and legal disputes that have been central to the litigation over the deferred action initiatives from the very beginning. His opinion, however, breezily proceeds as if it were straightforwardly factual and uncontroversial that DACA is properly understood as “more than nonenforcement,” and offers a citable soundbite for lower courts and others to use in support of that proposition.

II.

Having found the decision to terminate DACA to be reviewable, Roberts nevertheless declined to evaluate on the merits either the sufficiency of the Trump administration’s legal explanation for that decision or whether its legal position was correct. This choice, too, was a curious one, with similarly curious consequences. The lower federal courts—having found the decision reviewable because it was based entirely on a legal conclusion that the initiative was unlawful and unconstitutional—proceeded, quite understandably, to actually engage and assess that legal position when determining whether the rescission decision might be arbitrary and capricious. In the District of Columbia, Bates concluded that the legal reasoning and explanations given by both Duke and Sessions to support their legal position were “scant,” “opaque,” and “barebones.” On that basis, Bates concluded that the Trump administration had failed to satisfy its obligation to give a sufficiently reasoned explanation in support of its departure from the government’s previous position, under the Obama administration, that DACA was lawful. Given the shoddy nature of the reasoning and explanation provided by Duke and Sessions, Bates concluded, it was unnecessary to go further to analyze whether the legal position was also substantively wrong. Although Duke’s successor, Kirstjen Nielsen, responded to Bates’s order by issuing a new rescission memo, Bates concluded that Nielsen’s position, too, was insufficient because it provide[d] “almost no meaningful elaboration” on Duke’s assertion that DACA was unlawful. Instead, Nielsen once again relied primarily on Sessions’s letter—whose “conclusory legal assertions,” Bates reiterated, the court had already found were “themselves inadequately explained.”

In the California and New York cases, the lower courts went further to analyze and evaluate the substance of the Trump administration’s legal position—and concluded that legal position was incorrect. In the Eastern District of New York, Judge Nicholas Garaufis concluded that it was “legally erroneous” for Sessions to determine that DACA was unconstitutional based on the “conclusory” assertions in his one-page letter, and that Sessions’s assertion that DACA had “the same legal and constitutional defects that the courts recognized as to DAPA” not only rested on an “obvious factual mistake,” but was legally incorrect. Garaufis also concluded that, to the extent that Sessions meant to assert that DACA was inconsistent with the immigration statutes, that assertion also was legally incorrect—to the contrary, Garaufis expressly determined, “DACA is lawful.”

The California litigation proceeded similarly. In the district court, after analyzing the legality and constitutionality of DACA in great detail, Alsup concluded that the rescission was “based on a flawed legal premise,” since the more sound legal conclusion was that “DACA was and remains a lawful exercise of authority by DHS.” On appeal, the Ninth Circuit agreed after conducting its own detailed legal and constitutional analysis. The Ninth Circuit squarely concluded that DACA had been a legally permissible exercise of executive discretion, “notwithstanding the Fifth Circuit’s conclusion that the related DAPA program exceeded DHS’s statutory authority.” As such, the Trump administration’s decision to rescind DACA could not rest on Sessions’s “erroneous view of what the law required,” and was therefore arbitrary and capricious.

Once again, however, Roberts chose a different, more convoluted approach—and once again, it was an approach that none of the lower courts appear to have discussed and none of the parties appear to have urged or even argued. Rather than scrutinizing either the sufficiency of Sessions’s legal explanation or the correctness of the Trump administration’s legal position on the merits, as the lower courts had done, Roberts essentially changed the subject. The respondents’ arguments, he maintained, had “overlook[ed] an important constraint” on Duke’s decision-making authority—namely, that “she was bound by the Attorney General’s legal determination” under 8 U.S.C. § 1103(a)(1). Of course, the respondents had done no such thing: as noted above, they expressly invoked that same provision to emphasize that Duke’s decision could not have been understood as “committed to agency discretion by law” and therefore was reviewable. If anything, it was Roberts himself who “overlook[ed]” the significance of that provision for purposes of reviewability, only to opportunistically rediscover its potential use as a tool to avoid confronting the reasons actually given by the lower courts in support of the conclusion that DACA’s termination had been arbitrary and capricious.

The only reason Roberts offered to avoid directly engaging the sufficiency or soundness of Sessions’s legal position is a suggestion, almost in passing, that the lawsuits challenging DACA’s rescission might not be “proper vehicles for attacking the Attorney General’s legal conclusion” that DACA is unlawful and unconstitutional. But even as he took a swipe at the respondents for these supposed “gaps” in their briefing, Roberts himself did not discuss any reasons why the lawsuits would not provide an appropriate occasion to address whether Sessions’s position was adequately explained or legally correct. (For its part, the government does not appear to address this issue in its briefing either, but Roberts faults the respondents alone.) It is difficult to understand why those issues should not appropriately have been addressed in the context of that litigation. After all, Duke’s memo expressly referred to and incorporated the reasons vaguely advanced in Sessions’s letter as the sole basis for her decision to terminate DACA. It was precisely on that basis that the lower courts examined the Sessions letter when ascertaining whether the Trump administration’s decision had been inadequately explained or rested on an incorrect legal premise. As Garaufis’s opinion put it, “[Duke’s memo] offers no independent legal reasoning as to why Defendants believed the DACA program to be unlawful, so the court turns to the Sessions Letter.”

Nevertheless, having extracted 8 U.S.C. § 1103(a)(1) from the place in the litigation where it was more directly relevant, Roberts then proceeded to analyze whether DACA’s termination was arbitrary and capricious exclusively in terms of any possible discretionary choices that might have remained available to Duke within the four corners of Sessions’s letter—regardless of whether Sessions’s legal position (as adopted by Duke herself) had been adequately explained or was legally correct. As such, Roberts’s analysis necessarily proceeded as if the Trump administration’s legal position was, in fact, legally sound. To the extent Sessions’s analysis was vague and conclusory, as Bates twice concluded, Roberts generously filled in those details himself, explaining and elaborating upon Sessions’s position primarily by drawing approvingly from the Fifth Circuit’s opinion in the DAPA litigation and embracing it as his own. To the extent that the Fifth Circuit’s legal analysis about DAPA may have been unsound or inapplicable to DACA—as some of the lower courts had concluded—Roberts simply ignored those possibilities and their potential relevance altogether. He then proceeded to maintain, rather dubiously, that the Fifth Circuit had been “careful to distinguish” between DAPA’s “forbearance component” from its component purportedly granting “eligibility for benefits,” such as employment authorization, Social Security, and Medicare—and that the “legal defects” with which the Fifth Circuit had been concerned only involved the latter. As such, Roberts insisted, Duke’s main fault was her failure to consider whether DACA might be saved by turning it into a more barebones initiative that only provided some form of enforcement forbearance, not eligibility to apply for other collateral benefits.

In describing the DAPA litigation and the Fifth Circuit’s opinion in this manner, Roberts served up large helpings of judicial truthiness and revisionist history. As I discussed at length in an essay for Dorf on Law at the time, neither the plaintiffs in the DAPA litigation nor the judges ruling their favor sharply distinguished between “forbearance” and “benefits” in the manner that Roberts suggests. To the contrary, the rhetoric they deployed against the deferred action initiatives depended heavily on blurring the lines between those two categories, in order to argue that DACA and DAPA did not simply provide guidance for the agency’s exercise of prosecutorial discretion, but instead usurped congressional authority by conferring something more closely approximating and resembling lawful immigration status. Indeed, the plaintiffs scrupulously avoided making any such distinction throughout most of the litigation, and never directly challenged the separate legal authority that actually governs such benefits.

For their part, the judges ruling in the plaintiffs’ favor fully embraced their framing of DAPA as involving an inextricably intertwined package that approximated lawful immigration status, albeit sometimes advancing that position in confused and jumbled fashion. While Roberts suggested that Smith’s opinion for the Fifth Circuit carefully left “the Secretary’s forbearance authority . . . unimpaired,” that assertion is simply false. Hanen’s injunction did not block DHS only from accepting and processing applications for “benefits” that Hanen believed to be impermissible, or from conferring those benefits. Rather, it enjoined federal officials “from implementing any and all aspects or phases” of the initiative, including grants of deferred action. On appeal, the Fifth Circuit affirmed the scope of that injunction in full. In Roberts’s Regents opinion, however, all of these basic facts about the DAPA litigation were conveniently memory holed.

III.

In their more recent opinions on the Biden administration’s efforts to preserve DACA, both Hanen and the Fifth Circuit’s Richman have taken Roberts’s misleading characterization of DACA as “more than nonenforcement” and run with it—not as in Regents for purposes of analyzing APA reviewability, but to support other aspects of their anti-DACA positions on the merits. In his 2021 ruling, for example, Hanen suggested that Roberts’s characterization of the initiative as “more than nonenforcement” provided conclusive, binding support for his own determination that the DACA was also not a “general statement of policy” and accordingly should have been instituted through notice-and-comment rulemaking. On appeal, Richman invoked Regents as support for the same proposition.

Notably, however, when presented with the opportunity to do so in the course of his most recent decision, Hanen angrily declined to take Roberts up on the suggestion that DACA might be whittled down to a much narrower initiative, in the manner that Cox and Rodríguez hypothesized might occur, by distinguishing between forbearance and benefits. In its new, rulemaking-based version of DACA, the Biden administration actually had left room for the possibility of falling back on a hollowed out version of DACA along these lines in the event of an adverse litigation outcome. The DACA rule included express severability clauses that, for example, would have allowed the provisions for forbearance to be preserved even if Hanen or another anti-immigration judge were to invalidate other provisions, such as those providing for work authorization or other benefits.

Nevertheless, in his latest decree Hanen refused to take what seemingly would have been the Roberts-esque approach of giving effect to the DACA rule’s administrative severability provisions. After grousing that he was “perplexed” at why he “should try to tailor the Final Rule when the agency made no attempt to do so,” he grudgingly went through the motions of determining whether the DACA rule should be treated as severable and—surprise, surprise—concluded that the DACA rule should not be severed and instead must be set aside in its entirety. (At the same time, as noted above, Hanen did also stay his order as to existing DACA beneficiaries.) Leaving to one side for now the question of whether this conclusion about administrative severability was legally sound, Hanen’s discussion is nevertheless striking in showing just how wrong Roberts was in trying to implausibly characterize the DAPA litigation as having yielded some sort of clear factual or legal distinction in the lower courts between “forbearance” and “benefits.”

While Hanen acknowledged Roberts’s suggestion that “forbearance” could be separated from “benefits,” he concluded that DHS “would not have adopted the Final Rule without the benefits provisions” because the two categories of provisions were necessarily and inextricably intertwined—which is more or less the same position he has taken since 2015. In the course of rejecting the notion that DACA could “function sensibly as a forbearance only policy,” Hanen also falsely insisted (as Roberts had similarly done in Regents) that a judicial decision setting DACA aside would have no effect on DHS’s exercise of prosecutorial discretion—thereby seeming not to understand basic facts about the very nature of deferred action, even after presiding over lawsuits about deferred action for eight-and-a-half years. On that basis, Hanen concluded that “forbearance with no benefits would be superfluous.” Were it “stripped of the benefits it directly or indirectly bestows,” Hanen maintained, DACA “would become a nullity.” In a laughably brazen touch, Hanen cited none other than John Roberts in support of his conclusion that DACA could not “function sensibly” as a forbearance-only initiative: “As previously noted by the Supreme Court in Regents,” Hanen emphasized, “DACA is primarily a benefits only rule.”

IV.

None of us would still be wasting our time talking about Andrew Hanen at all had Roberts taken a more straightforward and candid approach to his decision in Regents. Had he concluded that the Trump administration was incorrect in its legal position that DACA was unlawful, as some of the lower courts had done, the Trump administration still could have terminated DACA, but it would have had to take responsibility for doing so in policy terms—which subsequent reporting makes clear it had been avoiding—rather than deflecting responsibility and blaming its decision on “the law,” “the Constitution,” and Barack Obama. Conversely, a future administration inclined to support the use of deferred action in a manner similar to DACA or DAPA—whether Biden or someone else in future years—would have had a green light and clear path to follow if they wished to do so.

Why, then, didn’t Roberts make this choice? Supreme Court Kremlinology can itself feel like a tiresome waste of time, but indulge me for a moment. The most obvious explanation of course seems to be that Roberts does not, in fact, believe that the Obama-era deferred action initiatives are lawful. His skepticism about DACA and DAPA was plain to see during oral argument in the DAPA litigation, and while we cannot know for sure, the Supreme Court’s divided 4-4 outcome in that case seems to indicate pretty clearly that he likely voted to affirm the Fifth Circuit’s ruling. Four years later, his warm embrace in Regents of the Fifth Circuit’s decision and its reasoning only reinforces that conclusion.

So given his almost certain belief that DACA is unlawful, why then did Roberts not simply rule in the Trump administration’s favor by concluding that its legal position was sound? As Professors Cox and Rodríguez note, his opinion seems designed to set DACA up for its eventual undoing in any event. Why wouldn't Roberts have just let that happen in 2020? Writing in dissent in Regents, Justice Clarence Thomas offers his own answer: Roberts’s decision “must be recognized for what it is: an effort to avoid a politically controversial but legally correct decision.”

It seems difficult to disagree with Thomas’s assessment (well, on the avoidance of political controversy part, at least). Instead of coming down clearly on one side or another on the question of DACA’s legality when that question was squarely presented—in a presidential election year, no less—Roberts appears to have opted instead, as he often seems to do, to play a politicized long game. The approach he took in Regents might have been convoluted and his own invention, but it allowed him to invalidate the Trump administration’s attempt to end DACA—a popular initiative that has long garnered strong public support—while simultaneously giving a subtle but unmistakable boost to a Fifth Circuit decision calling DACA’s longer-term viability into question. In the lower courts, anti-DACA judges like Hanen have used Regents as the simulacrum of more substantive precedent to give their own existing anti-DACA positions a little bit more weight. In a future Supreme Court case, one can easily imagine Roberts (or one of his fellow travelers) treating the invalidation of DACA, or some future DACA-like use of deferred action, as merely an incremental step that follows from what Roberts, quoting and summarizing the Fifth Circuit, has to say in Regents.

Observers including Professor Benjamin Eidelson have characterized Roberts’s opinion in Regents as fashioning a mechanism to encourage political accountability for the executive branch’s discretionary decisions. That way of understanding Regents, and the litigation challenging DACA's rescission generally, has much to commend itself. However, it seems difficult to explain Roberts's departures from the approaches taken by the lower courts—which themselves can be understood as accountability-forcing—entirely in accountability-forcing terms. Moreover, as Professor Rodríguez explained in her law review article on the decision, the outcome in Regents also serves to deflect attention away from the aggressive, anti-immigrant turn in the larger sweep of the Roberts Court’s decisions on immigration, some less prominent aspects of which have also been examined in a terrific law review article by Professor Nancy Morawetz. Three years after the decision, as the hopelessly politicized anti-DACA litigation persists in the lower federal courts in Texas, it seems hard to escape the conclusion that the majority’s opinion in Regents was written in a manner tailor-made to help the Roberts Court deflect accountability for its own politicized choices on immigration.