Last week, the Supreme Court made waves by declining to stay a lower-court ruling that blocked the Biden Administration’s rescission of the Trump Administration’s “Migrant Protection Protocol” or “Remain in Mexico” policy. Though an unfortunate result, this outcome follows logically, if not inevitably, from the Supreme Court’s equally unfortunate decision in June 2020 preventing the Trump Administration from rescinding the “Deferred Action for Childhood Arrivals” or “DACA” policy. Some cheered by the DACA case may now regret that decision’s implications.

In Department of Homeland Security v. Regents of the University of California, the 2020 decision on DACA’s repeal, the Supreme Court held that the Trump Administration violated the Administrative Procedure Act by rescinding DACA without adequate explanation. Adopted by the Obama Administration in 2012, the DACA program invited hundreds of thousands of unauthorized immigrants who arrived in the United States as children to apply for “deferred action,” a renewable two-year promise of non-deportation that carried with it certain other benefits including potential work authorization. Though in practice this policy altered immigration law for its beneficiaries, the administration justified DACA as a mere exercise of prosecutorial discretion to prioritize some unauthorized immigrants over others for enforcement.

As I argued at the time, the proffered legal basis for DACA should have meant it could be revoked with little or no explanation. Enforcement priorities can normally change without notice; they are revocable because legal obligations ultimately come from substantive laws, not executive officials’ temporary choice of focus. But the Supreme Court held otherwise. In UC Regents, a five-justice majority deemed the DACA repeal arbitrary and capricious because the agency’s explanation inadequately addressed whether to continue enforcement forbearance without associated benefits. In addition, even though the government had informed DACA beneficiaries that the program was revocable, the Court faulted the administration for failing to “assess whether there were reliance interests, determine whether they were significant, and weigh any such interests against competing policy concerns.” DACA thus survived the Trump presidency, and President Biden has ordered the Secretary of Homeland Security to “preserve and fortify DACA,” though in July a district court held the policy unlawful and ordered the suspension of new applications.

For his part, after taking office in January, President Biden sought to rescind the Trump Administration’s Remain-in-Mexico policy. Under this policy, DHS broadly exercised a discretionary statutory authority to “return” migrants with pending asylum claims to “a foreign territory contiguous to the United States” if the migrant “arriv[ed] on land” from that territory. Instead of detaining migrants with pending asylum claims or paroling them into the United States, the Remain-in-Mexico policy often required returning those apprehended on the southern border to Mexico.

Like DACA, the Remain-in-Mexico policy should have been revocable with little or no explanation. Not only is the authority to return migrants discretionary, not mandatory, but returning them in large numbers to a neighboring country could raise delicate diplomatic questions that require a variable policy over time. Absent a more specific statutory mandate, courts have no business freezing executive policies like this in place.

In the waning days of the Trump Administration, however, Texas entered an agreement with the federal government guaranteeing advance notice prior to any rescission of Remain-in-Mexico. So, when the Biden Administration nevertheless rescinded the policy, Texas and Missouri sued, and a district court enjoined the administration from terminating the policy. Invoking UC Regents, the Fifth Circuit affirmed. It reasoned that, as with the Trump Administration’s explanation for the DACA repeal, the government’s explanation of the policy change showed inadequate consideration for Texas’s reliance interests and other potential negative effects of the policy change. In a terse order, the Supreme Court denied the government’s request for a stay pending appeal. Citing UC Regents, the Court declared simply that “[t]he applicants have failed to show a likelihood of success on the claim that the memorandum rescinding the Migrant Protection Protocols was not arbitrary and capricious.”

As should be clear, both UC Regents and the Fifth Circuit’s Remain-in-Mexico decision strike me as erroneous. But they are flawed in parallel ways, meaning that, as Jonathan Adler has observed, the two cases should properly stand or fall together.

In UC Regents, the court effectively applied an intensive form of arbitrariness review, faulting the government for failing to consider certain aspects of the policy change in sufficient depth. As I argued at the time, the opinion appeared designed to have limited precedential effect; it went out of its way to be narrow and fact-specific. As I also noted, however, the Court’s reasoning, if taken seriously, all but invited efforts to lock in Trump Administration policies. “To the extent such policies invite reliance,” I wrote, “a future administration will be foreclosed from reversing them unless its self-explanation meets the Regents court’s meticulous standards.” Though I did not have Remain-in-Mexico in mind, this prediction has now come to pass: courts have frozen in place a discretionary Trump Administration policy based on purported deficiencies in the repealing agency’s self-explanation.

I am unpersuaded by the efforts I have seen to distinguish the two cases. Some have suggested that in UC Regents was unique because the Trump Administration initially justified DACA’s repeal by asserting the policy was unlawful. On this account, the administration’s original rationale was specious and served only to deflect political responsibility for an unpopular policy choice.

In fact, the rationale was not specious: DACA was unlawful, and in any event the administration’s later statements took responsibility for the change in policy. But even if the Trump Administration’s legal conclusions were wrong, the error would provide no basis for distinguishing UC Regents, because UC Regents expressly declined to rule on DACA’s legality. As Chief Justice Roberts’s majority opinion explained, “[w]hether DACA is illegal is . . . a legal determination, and therefore a question for the Attorney General.” Because the Attorney General had deemed DACA unlawful and the Secretary of Homeland Security was bound by that determination, the Court expressly declined to “evaluate the claims challenging the explanation and correctness of the illegality conclusion.” The Court’s holding in UC Regents thus turned on purported deficiencies in the agency’s policy discussion—precisely the problem that doomed the Remain-in-Mexico rescission.

More generally, political-process concerns provide no clear basis for distinguishing the two cases. Although DACA is popular and repealing it was not, some argue that political accountability nonetheless justified requiring the Trump Administration to say the quiet part out loud and articulate its indifference to the costs of repealing DACA. To the extent that rationale explains UC Regents, however, much the same logic could justify the Fifth Circuit’s holding. The government’s explanation for rescinding Remain-in-Mexico certainly seems to telegraph sympathy for immigrants while saying almost nothing about real or potential costs of allowing asylum seekers into the United States.

Another line of argument posits that, in UC Regents, cancelling legal benefits while continuing enforcement forbearance was so obvious an alternative to ending DACA that the agency’s failure to consider it was arbitrary. As I have discussed elsewhere, this feature of the UC Regents reasoning was borderline disingenuous: despite declining to address DACA’s legality, the Court faulted the Secretary for failing to parse options supposedly afforded by prior court opinions invalidating a different deferred action program. In any event, the Fifth Circuit found the same deficiency in the Remain-in-Mexico record that UC Regents found in the DACA repeal: it too failed to show serious consideration of ways in which the policy might have been modified instead of abolished outright. Furthermore, as a general matter, UC Regents invited close scrutiny of agencies’ reasons for repealing past policies. That inquiry is inevitably slippery and subjective, but it now applies equally to both Democratic and Republican administrations.

Swinging wildly between opposite policies in multiple domains is probably not the best way to run a superpower. But this feature of contemporary administration reflects the divided and polarized character of our current politics. Courts can properly limit such inconstancy by invalidating executive actions that go beyond what statutes allow. They should not have taken the further step of flyspecking agencies’ reasons for repealing past discretionary policies. Having gone down this path, however, it is unsurprising, and in some ways welcome, to see the Court applying this approach in symmetrical fashion to Biden Administration policy changes. In the law, a win in one case often requires paying the piper down the road.

Zachary Price is a professor of law at UC Hastings Law in San Francisco.