Breaking the Courts
by Jed Stiglitz
Last week the Supreme Court rejected the Trump Administration’s effort to rescind DACA, the policy started under President Obama to provide deportation relief to immigrants who arrived as children. In a 5-4 decision with Chief Justice Roberts writing for the majority, the Court held that the rescission was arbitrary and capricious and therefore invalid under the APA. The case is a major victory for the over 700,000 DACA recipients who rely on the program. The case also creates a number of wrinkles in administrative law and presents important puzzles, one of which I want to focus on—why did we have this case at all? Observers including my host have suggested a number of responses to this puzzle, but they all fit comfortably within normal politics. I wish to suggest one outside of normal politics.
A few details of the case to lay bare the puzzle. Good summaries of the case exist elsewhere, so I won’t belabor one here. But the key elements of the background follow. In 2012, the DHS announced the DACA policy via a memorandum. Two years later, the DHS expanded DACA and created the DAPA policy, which granted deportation relief to parents of children who were U.S. citizens or permanent residents. This 2014 expansion was challenged before it went into effect, and a District Court in Texas granted a preliminary injunction against the policy. The Fifth Circuit and later the Supreme Court affirmed the injunction, though the Supreme Court did so on the basis of a 4-4 vote and without opinion. The Fifth Circuit granted the injunction on the basis of procedural flaws (the policy needed notice-and-comment) and substantive flaws (the policy was contrary to a statute, the INA). Soon after taking office, President Trump issued an Executive Order calling for tightened immigration policy, and the DHS rescinded DAPA.
Notice that this earlier litigation did not affect DACA, as announced in the 2012 DHS memo. Yet soon after the DHS rescinded DAPA, Attorney General Sessions wrote to the Acting Secretary of DHS, Elaine Duke, to urge her to terminate DACA under the view that “the DACA policy has the same legal and constitutional defects that the courts recognized as to DAPA, [and] it is likely that potentially imminent litigation would yield similar results with respect to DACA.” A day later, Duke issued a memo rescinding the DACA policy—the “Duke memo.” After summarizing the DAPA litigation and the letter from Sessions, she concluded, “Taking into consideration the Supreme Court’s and the Fifth Circuit’s rulings in the ongoing litigation, and the September 4, 2017 letter from the Attorney General, it is clear that the June 15, 2012 DACA program should be terminated.”
Various parties immediately challenged Duke’s rescission as arbitrary and capricious under the APA (as well as a violation of equal protection). Viewing the Duke memo as inadequate under the APA, the D.C. District Court vacated the rescission, but stayed the order for a spell so that the DHS could re-issue the rescission with fuller explanation. This led to a memo by the then-Secretary of DHS, Kirstjen Nielsen—the “Nielsen memo.” Critically, her memo “declined to disturb” the Duke memo’s decision. Instead, standing on the Duke memo, Nielsen’s memo explained her “understanding of the Duke memorandum and why the decision to rescind the DACA policy was, and remains, sound.” She then pointed to the claim that DACA was contrary to law, to the idea that at least serious doubts about DACA’s legality existed, and to principles of enforcement policy. In closing, she acknowledged the reliance interests of the DACA recipients—an issue in the litigation—but argued those considerations were outweighed by other factors. The Nielsen memo, therefore, tried to spin out the justification offered in the Duke memo, but again critically did not issue a new decision rescinding DACA. The District Court rejected the Nielsen memo, and the Supreme Court eventually consolidated this D.C. case with two others and granted cert.
The majority held that the rescission was arbitrary and capricious. Along the way, the Court created a few more wrinkles in administrative law. For instance, the trick of hiding an important substantive policy change behind an enforcement guideline, sprinkling the guideline with a little individual-level discretion, and saying that the policy is unreviewable by courts because of prosecutorial discretion seems increasingly untenable. The majority also focused on reliance interests in its analysis. That is not a new consideration in administrative law, of course, but if fortified could have wide-ranging implications for the ability of agencies to fulfill statutory objectives and to adjust to changing circumstances. The majority did not confront some interesting questions about reasons based on legal conclusions, but the decision should still be recognized as a win for reason-giving and for the principle that agencies need to provide contemporaneous explanations for their actions.
The most immediate puzzle about the case, though, is why it exists. None argue that the DHS lacked the substantive authority to rescind the DACA policy. The only question is whether the DHS offered a sufficient explanation. If the administration had “turn[ed] square corners,” nine justices would likely have gone along with the rescission. Moreover, it would have been easy to offer a more fulsome explanation.
This might have been accomplished at almost any point along the way. But the most obvious time at which the administration could have avoided this issue is during the short stay offered by the D.C. District Court precisely for this purpose. Nielson could at that point have easily issued a new decision rescinding the DACA policy with a more complete explanation. The D.C. District Court envisioned her doing just this. But instead of issuing a new rescission, she left Duke’s rescission undisturbed. It is not even entirely clear that the modestly expanded explanation provided in the Nielsen memo, which glanced for instance at reliance interests, was itself inadequate if her memo constituted a new rescission. But looking to explanations contemporaneous with the action at issue, the majority credited only the Duke memo in its analysis, so her justification did not count. Hence the puzzle. Why didn’t they just issue a new rescission with a better explanation and avoid this case?
Observers have proposed a number of responses to this puzzle.
This is perhaps the most common explanation in circulation. The administration has certainly revealed itself to be incompetent many times in the last three years, so incompetence cannot be discounted as a possibility. There is also Hanlon’s razor, which counsels in favor of this explanation. All the same, the incompetence in this case is extraordinary. The District Court noted problems with the explanation, including the failure to consider reliance interests, and invited Nielsen to respond, allowing in its stay that “the Secretary of Homeland Security or her delegate may reissue a memorandum rescinding DACA, this time providing a fuller explanation for the determination that the program lacks statutory and constitutional authority.” It is true that, so far as I can tell, the concern about separating DACA’s deportation forbearance from collateral benefits eligibility was not yet squarely on the table, and that argument carried part of the day for the majority. But even if that issue was not addressed, had Nielsen simply re-issued the rescission, with her justification as it was including its gesture to reliance interests, it is quite plausible that it would have passed muster for a majority of justices. That may be part of why the majority invested itself in discounting the Nielsen memo and focused on the Duke memo—which memo counts, indeed, was the main fight between the majority and Kavanaugh in dissent.
2. Deep State Sabotage
Perhaps DHS sought to undermine the viability of the rescission by providing an inadequate explanation. Supporting this view, Duke herself apparently disfavored DACA rescission, and the New York Times reports that she “did not want her name on what she saw as anti-immigrant policy rationales put forth by Mr. Sessions.” The challenge to this theory is that the critical moment was Nielsen’s failure to re-issue the rescission, and so far as I can tell Nielsen had few moral qualms about the policy change. The thesis of bureaucratic resistance is a strong one, but I’m not sure it fits this case.
3. Judicial Bailout
Another leading theory is that the administration sought to stoke the base by rescinding the policy, but was actually unwilling to bear the political consequences of rescission. One of the pro-DACA lawyers, Luis Cortes Romero, for instance expressed this idea. DACA is a popular policy, and rescinding the policy in reality would generate thousands upon thousands of testimonials from child-immigrant Americans who were torn from their lives and deported to countries they did not know. So the best scenario for the administration was for it to appear to make a good-faith effort at rescinding the policy, only to be frustrated by the courts. This would allow it to gain a measure of credit from the base without suffering the consequences of a rescission in reality. This theory is not so far from one advanced by Justin Fox and Matthew Stephenson, which posits that “judicial review may rescue elected officials from the consequences of ill-advised policies” that were adopted to posture to voters, thus “bailing out” politicians from their own bad decisions.
By this logic, the big winner last week was the administration. But if that is right, the vote lineup is a little odd. The four liberals voted for the position best for the administration, and four conservatives voted against the position best for the administration. True, perhaps the liberals were unwilling to risk living with the reality of a DACA rescission. But it is harder to understand why the conservatives would vote against the administration’s favored outcome. Maybe Roberts was the only one to figure out the game. Or maybe this theory doesn’t explain the case.
Another oddity is that the government’s litigation strategy did not maximize the delays available to it. The administration filed petitions for certiorari while appeals from the three district court decisions were pending. The administration had injunctions at that point, which under this theory is what they sought, so why rush to the Supreme Court? That could only expedite the possibility that their bailout would be upset. More generally, the bailout strategy is risky as the Court might end up siding with the administration’s stated rather than preferred position. Indeed, the Court nearly accepted the administration’s stated position.
4. Breaking the Judiciary with Legal Absurdity
Let me float another explanation that is outside of normal politics and fits with a broader pattern in this administration. Suppose you were trying to degrade the judiciary. To undermine values core to the institution, that is, and moreover to publicly demonstrate that the judiciary is spent and subordinate. What might you do?
One strategy might be to advance poor cases. In fact, the poorer the better. If you convince the judiciary to go along with a strong case, that helps you little. But if you can convince the judiciary to submit to a weak case, that is very helpful. Submission to a weak case is helpful in the first instance because it directly undermines the integrity of the institution’s internal logic. The institution must contort itself to accommodate the weak case and find ways to ignore its own history. It is helpful in the second instance because the weaker the case, the more of a public statement the judiciary’s submission is. Submission to absurdity shows all that the administration is dominant and without judicial constraint.
Here, the fact that Nielsen very easily could have fixed the explanation is part of the point. She would not take the smallest steps to accommodate the court or its internal logic. It was helpful, moreover, that the case confronted the adequacy of reason-giving, a value core to the judiciary (and in ordinary times, the administrative state). Submission to these facts would be very useful to breaking the judiciary. This theory also explains the rush to the Supreme Court, where the administration thought it might receive a more readily submissive audience than in the relevant Circuits. The Supreme Court, of course, is also more high profile, all the better for public display. Further, the theory more closely explains the vote by the Supreme Court. It was the institutionalist, Roberts, who voted against the administration despite his conservative ideology. The other four conservative justices were willing to go along with the administration.
This may be part of a larger pattern in the administration’s judicial strategy. Before suggesting that pattern, though, consider other domains where the administration has adopted a similar strategy. We had our first taste of official absurdity when the White House Press Secretary insisted that Trump’s inauguration crowd was the “largest audience to ever witness an inauguration – period” despite photographs obviously showing crowd sizes smaller than those from President Obama’s inaugurations. The Press Secretary’s submission to absurdity set expectations for government employees and tested the media. The administration regularly asks Republicans in Congress to submit to absurdity, and that they almost all always do makes it hard for them to recover an independent moral center and publicly demonstrates the subordination of the Republican-controlled parts of the legislative branch. More recently, the administration induced the Chairman of the Joint Chiefs of Staff to appear in a photo op with combat fatigues after using gas to clear protesters from outside the White House. The backlash to this last effort was swift and widespread, leading to an apology from General Milley and the suggestion that he was misled into appearing in the photo op. But again, an effort to undermine and publicly subordinate through submission to acts contrary to the values of the institution in question.
Why should we think the legal domain is immune to this approach? The DACA case is, indeed, but one example. The Census case over the citizenship question, where the administration asked the courts to accept manifestly contrived reasons, represents another effort to degrade the judiciary. The Court rejected that bid, again with Roberts as the pivotal player. “If judicial review is to be more than an empty ritual,” Roberts wrote, “it must demand something better than the explanation offered for the action taken in this case.” For their parts, Thomas, Gorsuch, and Kavanaugh were fine with judicial review as empty ritual, and Alito preferred to avoid review altogether. The administration’s deregulation program more generally regularly asks judges to accept poorly reasoned administrative actions. So far, courts have largely rejected these efforts. Other times the administration asks the judiciary to suspend its senses and ignore statements by the president that demonstrate the animus underlying various actions. Sometimes courts suspend their senses; other times they refuse to.
We probably have life tenure and a culture of reason-giving to thank for the relative success of the judicial resistance to institutional subordination. Justices and judges feel the pulls of partisanship and ideology, but they do not fear for their jobs, and they have reputations they wish to protect based on their ability to publicly reason. Still, the record is not uniform, and even the victories often appear to be close-run affairs, as in the DACA case.
Maybe this theory is off. Maybe the administration’s record reflects legal incompetence. Maybe the inaugural photo mess was incompetence. Maybe the near-daily humiliation of Republican Members of Congress is incompetence. Maybe inducing General Milley to appear in the Lafayette Square photo op with combat fatigues was incompetence. Maybe. But maybe not.