The Courts Restore the 'Remain in Mexico' Program: An End to Judicial Deference?

The Supreme Court on Aug. 24 declined to stay an injunction by Judge Matthew J. Kacsmaryk of the U.S. District Court for the Northern District of Texas against the Biden administration’s termination of the “Remain in Mexico” program, formally known as the Migrant Protection Protocols (MPP). The U.S. Court of Appeals for the Fifth Circuit had earlier also declined to stay the injunction. The courts’ decisions failed to accord appropriate deference to executive decisions about foreign affairs and resource allocation in immigration enforcement. However, given the Fifth Circuit’s broad reading of the Supreme Court’s decision in Department of Homeland Security v. Regents of the University of California, faulting the Trump administration’s stated reasoning for rescinding Deferred Action for Childhood Arrivals (DACA), the administration’s wisest approach may be a fresh, more comprehensive explanation of its reasons for ending MPP.

The MPP’s Operation and the Trump Administration’s Assessment

The MPP, a signature initiative of the Trump administration, relied on a little used Immigration and Nationality Act provision—8 U.S.C. §1225(b)(2)(C)—that allows immigration officials to “return” certain new entrants to a country that is “contiguous” (bordering on) the United States while those foreign nationals await a full hearing before an immigration judge. From January 2019 to January 2021, officials implementing MPP removed almost 70,000 asylum-seekers to Mexico. But as the U.S. Court of Appeals for the Ninth Circuit noted in 2020 in upholding an injunction against MPP, the program exposed asylum-seekers to substantial risks. For example, while waiting for months for removal hearings, asylum-seekers struggled to obtain housing and evade criminal gangs. Journeys to the border for transportation to U.S. temporary hearing facilities were fraught with peril, as Stephanie Leutert wrote here. In addition, access to counsel was far more difficult from Mexico than it is in the United States, where accessing counsel is difficult enough. According to Human Rights First, less than 10 percent of MPP enrollees had legal representation. Moreover, because of safety issues, many subjects did not show up for their hearings and received in absentia removal orders. In some cases, according to Human Rights First, MPP enrollees were kidnapped, making attendance at hearings impossible unless their captors had a soft spot for immigration adjudication.

The Trump administration in October 2019 provided its own analysis of MPP’s results. It cited two descriptively true facts, although their causation and normative implications are less clear: First, entries to the United States from Mexico by Central American families had dropped sharply, and second, in absentia removal orders were way up. According to Department of Homeland Security officials, both results occurred because MPP had reduced the number of noncitizens asserting baseless asylum claims.

Grading the Trump Administration’s MPP Analysis

Upon closer examination, the October 2019 memo’s conclusions rate a mixed grade. Statistics from U.S. Customs and Border Protection (CBP) demonstrate a significant reduction in apprehensions of Central American families at the border during operation of MPP, although these reductions had just started at the time of the October 2019 memo. Compare CBP statistics for fiscal 2020 to statistics for fiscal 2019. Numbers for October 2019 to February 2020 (that is, before the coronavirus had an impact) reflect a median drop of approximately 60 percent during this period.

The October 2019 memo is less persuasive on the causes of in absentia removals. The memo implies, but does not state expressly, that the principal cause of a noncitizen’s failure to appear is her claim’s lack of merit. On page 3, the memo refers to the substantial percentage of people who returned to their home country rather than continue in MPP. According to the memo, this data point suggested that “a significant proportion” of MPP enrollees had “chosen to abandon their claims.” The memo implies that enrollees with well-founded claims would stay the course. But the memo does not explore other reasons that at least some enrollees would leave the program.

Those other reasons would include both the safety issues mentioned earlier and concerns about whether MPP would in fact provide a meritorious claimant with a fair hearing. The October 2019 memo’s analysis of the latter issue is oblique and insufficiently concrete. On page 3, the memo states that U.S. immigration judges—part of the Department of Justice’s Executive Office for Immigration Review (EOIR)—had granted a “small subset” of MPP subjects’ asylum claims. But the memo does not provide more specific information, although presumably the government had access to data on case outcomes. According to Human Rights First, the grant rate in MPP was 4.1 percent. That would suggest, at least in comparison with non-MPP immigration judge hearings, that immigration judges in MPP were granting asylum claims at lower rates than usual, based on statistics about claims of applicants within the United States. According to EOIR, in groups of asylum-seekers that asylum officers determine have a credible fear of persecution—about 50 percent of the total—the grant rate is 15 percent. (See the account in Justice Samuel Alito’s opinion for the Supreme Court on page 8 in Department of Homeland Security v. Thuraissigiam.) In an unscreened group such as MPP enrollees, the grant rate should be half that: 7.5 percent. The Human Rights First’s estimate of the grant rate—4.1 percent—is below that threshold. Of course, samples vary. But the Trump administration should have at least provided a specific figure. Without that specificity, it is difficult to evaluate the administration’s claim.

The Texas-U.S. Agreement in the Trump Administration’s Final Days

The Trump administration took another step to reinforce MPP. Just before President Biden took office, the Department of Homeland Security signed an unprecedented agreement with Texas to consult with that state prior to any change in policy. But the validity of this agreement was in doubt from the start. First, the department’s signatory to the agreement, Ken Cuccinelli, had not been confirmed by the Senate and could not legally assume the position of deputy secretary of homeland security or even acting deputy secretary. That is why the memo identified Cuccinelli as the “Senior Official Performing the Duties of the Deputy Secretary.” It is not clear that under the law Cuccinelli even had the authority to act on the department’s behalf in executing the agreement. Moreover, Texas did not provide what contract law calls “consideration”—something of value—in exchange for the benefits that it received under the agreement. Under contract law, that lack of consideration would render a contract unenforceable. Texas agreed to honor U.S. immigration detainers—requests sent to local police and jails to detain an individual already in custody, when Immigration and Customs Enforcement (ICE) believes that the individual may not be lawfully present in the United States. However, Texas had been routinely honoring detainers for years. Rather than an agreement entered into at arm’s length and in good faith, the agreement seemed much more like an effort by Trump’s handpicked Homeland Security official and pro-immigration-enforcement Texas state officials to tie the Biden administration’s hands.

The Effect on MPP of the Coronavirus and the Transition to the Biden Administration

Other events changed the outlook for MPP. First, during the Trump administration, concerns about the coronavirus prompted the Centers for Disease Control and Prevention to issue an order under Title 42—which governs the federal government’s public health powers—to bar entry at the southern border. While that Title 42 order did not completely moot out MPP, it made the latter less significant in any immediate sense, at least while the coronavirus remained a threat and immigration courts were closed.

When President Biden assumed office in January 2021, he indicated that he would fulfill a campaign pledge to end MPP. In February, the Biden administration summarily terminated the Texas agreement and began winding down MPP, limiting new removals to Mexico and processing some enrollees’ claims in the United States. To complete the termination of MPP, the Biden administration would have to consider one further issue: the relevance of the Supreme Court’s 2020 Regents decision on the Trump administration’s DACA decision.

Regents and the Role of Reasoned Decisionmaking in Termination of Programs

In Department of Homeland Security v. Regents of the University of California, the Supreme Court held that the Trump administration had failed to engage in “reasoned decisionmaking” under the Administrative Procedure Act when the Trump administration sought to end President Obama’s DACA program. (See my analysis of Regents here.) According to Chief Justice John Roberts, who wrote the opinion of the court, then-Homeland Security Secretary Elaine Duke had failed to address basic questions, such as how ending DACA would affect the reliance interests of recipients, their families, and U.S. persons and entities with educational, service, and employment relationships with DACA recipients. In addition, Roberts noted that Duke had failed to address whether the Department of Homeland Security could bifurcate the DACA program, ending the portion of the program that provided eligibility for work authorization, but continuing the program’s reprieve from removal for its recipients.

Reasons for Rescinding MPP: The Mayorkas June 1 Memo

That discussion of failures in the Trump administration Department of Homeland Security’s reasoned decisionmaking regarding DACA tees up the decisions of the Biden Department of Homeland Security about MPP. Here, the June 1 memo by Biden Homeland Security Secretary Alejandro Mayorkas is key. Both the district court and the U.S. Court of Appeals for the Fifth Circuit found that Mayorkas’s memo failed the “reasoned decisionmaking” test that the Supreme Court outlined in Regents. Assessing that conclusion requires a closer look at the memo’s reasoning.

In his memo, Mayorkas observed that MPP consumed substantial resources and Homeland Security employees’ time and effort. For MPP hearing sites that were not adjacent to ports of entry, CBP officers had to coordinate travel of MPP enrollees from the border to the hearing location and back to the border. Other department employees staffed the tent facilities where hearings took place. Moreover, MPP enrollees did not necessarily stay in Mexico until their hearing dates. Many sought to enter the United States at undesignated points, along with other non-MPP foreign nationals. CBP bore the brunt of apprehending these individuals, reducing the time that CBP officers had for other enforcement tasks.

In addition, Mayorkas raised a significant foreign policy concern. According to the secretary, MPP had come to dominate the U.S.-Mexico relationship, occupying a disproportionate share of exchanges between diplomats and others in the “bilateral relationship.” Mayorkas asserted that ending MPP would allow both U.S. and Mexican officials to focus more readily on matters such as “the root causes of migration from Central America; improving regional migration management; and expanding cooperative efforts to combat smuggling and trafficking networks.”

Mayorkas also addressed procedural concerns with the MPP removal proceedings, including the high volume—44 percent—of in absentia removal orders. Here, Mayorkas trod carefully, although his measured tone conveyed the unmistakable message that “conditions faced by some MPP enrollees in Mexico” such as the prevalence of gangs and kidnapping may have affected the “safety” of enrollees. Safety threats in the route from Mexico to a U.S. port of entry severely reduced attendance at hearings, Mayorkas suspected. Based on the information that the Department of Homeland Security had gathered, Mayorkas also believed that it was reasonable to infer that such conditions could have caused the “abandonment of potentially meritorious protection claims.” In other words, MPP was a blunt instrument that discouraged unfounded and well-founded claims alike, lacking the tailoring to distinguish between the two. Finally, Mayorkas said that in his judgment, any modification of the program would entail a “total redesign,” with considerable opportunity costs in staff time and funding. Nevertheless, while Mayorkas believed a programmatic commitment to MPP was unwise, he acknowledged that on a spot basis the department would continue to use the contiguous-territory authority in the Immigration and Nationality Act.

Although Mayorkas’s memo was more detailed than Duke’s memo rescinding DACA, it still left gaps that the district court and U.S. Court of Appeals for the Fifth Circuit later highlighted in their decisions. Most importantly, Mayorkas was insufficiently precise in discussing MPP’s effect on the all-important migration of Central American families to the United States. Mayorkas’s analysis was bland, relying on the banal observation that “border encounters [between CBP and would-be entrants during MPP] increased during certain periods and decreased during others.” If Mayorkas wished to assert that immigration ebbed and flowed during the relevant period, he should have provided facts and figures to support this claim, just as the Trump administration should have been more specific about the percentage of asylum grants in MPP.

But banality was not the biggest problem with Mayorkas’s discussion of border encounters during MPP. His biggest problem was the failure to specifically acknowledge the sharp reduction during MPP of family apprehensions—the biggest single driver of unlawful immigration during the past decade. Putting this figure into perspective should have been a key function of the memo, since a specific account of border apprehensions should be top of mind for the official in charge of immigration enforcement.

This gap went hand in hand with Mayorkas’s failure to address the Trump administration’s October 2019 Homeland Security memo. That omission was harmless on the question of in absentia removals, since Mayorkas addressed the underlying issue. However, the failure to mention the Trump Homeland Security memo was more problematic on the issue of border arrests.

In addition, Mayorkas did not mention the agreement with Texas on consultation regarding immigration policy. Because all states have a stake in immigration policy and Texas as a border state has a larger stake than many, mentioning the agreement would have been appropriate in depicting the backdrop for policymaking. Here, Mayorkas could have flagged the issue of whether unconfirmed acting Homeland Security officials serving in violation of legislation had authority to enter into a contract and whether Texas’s promise to honor immigration detainers constituted adequate consideration. On a more affirmative level, Mayorkas could have assured Texas and other concerned stakeholders at the state level that the department would work with them on all of their concerns, including coordinating with other federal agencies to provide Texas with resources in law enforcement, public health and education as needed. Mayorkas could also have noted scholars’ views that any short-term effect on state budgets or services would be outweighed by immigrants’ long-term positive economic impact.

The Courts’ Adverse Response to the MPP Halt Breaks With Judicial Deference

Both the district court and the Fifth Circuit analogized these gaps in Mayorkas’s memo to the lack of reasoned decisionmaking of Trump’s Department of Homeland Security regarding DACA, which the Supreme Court called out in Regents. The Fifth Circuit viewed Regents as a template for the MPP decision, taking Mayorkas to task for failing to respond at all or respond adequately. In particular, the district court, backed up by the Fifth Circuit, faulted the Mayorkas memo for failing to fully address the Trump administration’s claim in its October 2019 memo that MPP had substantially reduced CBP family unit apprehensions at the border. In addition, the courts found the Mayorkas memo’s analysis of in absentia removal orders to be incomplete, and its discussion of resource allocation and foreign policy concerns to be unpersuasive. Both the district court and the Fifth Circuit took the Trump administration’s October 2019 claims at face value, while exhibiting skepticism regarding Mayorkas’s June 2020 memo. The Fifth Circuit also observed that Mayorkas should have considered in greater detail the option of keeping part of MPP in effect, just as the Supreme Court in Regents had flagged the Trump administration’s failure to consider keeping in place DACA’s reprieve from removal. Based on these findings, the Fifth Circuit declined to stay the district court’s injunction, although the Fifth Circuit did hold that the Biden administration merely needed to make a “good-faith effort” to restore MPP, contingent on Mexico’s cooperation.

The Fifth Circuit’s approach, along with the Supreme Court’s decision not to stay the district court’s injunction, conflict with the long-standing deference that courts accord the executive branch on issues of foreign policy. This deference took center stage in a host of decisions, including Holder v. Humanitarian Law Project—which upheld legislation prohibiting “material support” to terrorist groups—and Trump v. Hawaii—which ruled that Trump’s travel ban was within the president’s statutory authority and was also consistent with the Establishment Clause.

In Hawaii, for example, Chief Justice Roberts, writing for the Supreme Court, noted that judges “cannot substitute our own assessment for the Executive’s predictive judgments” on matters of foreign affairs, which are “delicate, complex, and involve large elements of prophecy.” Roberts, in a passage from Humanitarian Law Project that he cited in Hawaii, also acknowledged the need to give the executive branch’s “evaluation of the underlying facts … appropriate weight,” especially in cases involving “sensitive and weighty interests of national security and foreign affairs.” Indeed, this strand in U.S. law and governance goes back to the Framers: In Federalist No. 70, Alexander Hamilton wrote famously about the importance of virtues such as energy, speed and decisiveness in the executive branch.

Such deference is not unlimited, and members of the public and experts alike can disagree about a range of cases, including Hawaii, which I have argued was unduly deferential. But Mayorkas’s decision about MPP in his June 1 memo was squarely within the range of cases that should trigger deference. Unlike the DACA rescission, which involved matters that played out in the domestic arena and addressed the interests of individuals domiciled in the United States, MPP by definition operates outside the country, in Mexico. At its core, MPP involves Mexico’s agreement to host tens of thousands of foreign nationals, largely from countries in Central America, for protracted periods while those persons await adjudication in the United States. That is a sensitive matter for any country. Indeed, it is sensitive for the United States; MPP arose because the Trump administration believed that immigration enforcement would benefit from off-loading noncitizens to Mexico who would otherwise be present in the United States.

It is true that Mexico cooperated with MPP during the Trump administration, but from the standpoint of judicial review, that decision by Mexico is beside the point. Like any other sovereign state, Mexico has interests of its own. Moreover, as with officials of any country, Mexican officials have limited political capital and bandwidth. Cooperation with MPP comes at a price: A concession there means that Mexico will have neither the inclination nor the capacity to work with the United States on other matters, including broader regional matters of migration and security. That was the gravamen of Mayorkas’s discussion of cooperation with Mexico in the June 1 memo. That is exactly the kind of decision to which courts typically defer.

The same concern prevails for the June 1 memo’s discussion of in absentia removals. The executive branch is in a far better position than the courts to assess MPP enrollees’ personal safety in Mexico. The Trump administration’s October 2019 memo addressed this issue only obliquely, noting on page 4 that “many” of Mexico’s states had “low unemployment and crime.” However, the Trump administration did not discuss crime in Mexico’s border area, which the FBI has linked to a “tide of violence” by drug cartels. Giving the Trump administration’s fuzzy explanation a pass, the district court and the Fifth Circuit second-guessed Mayorkas’s own characterization of removal proceedings in which the Department of Homeland Security is the moving party. Dialing skepticism down for the Trump administration while ratcheting it up for Biden shows a disparity in standards of review that conflicts with the courts’ impartial role.

The courts’ avid armchair-quarterbacking seems particularly inapposite because a more comprehensive discussion by Mayorkas of the causes of in absentia removals under MPP would inevitably have involved more details about safety and security in Mexico—sensitive matters whose public discussion could prompt embarrassment in Mexico and lessen cooperation by Mexican officials on immigration and other issues. In light of those concerns, judicial demands for more details about the circumstances of in absentia removals entail trade-offs with other interests—trade-offs that courts are ill equipped to assess.

The standard for a judicial stay of an injunction tells the same tale. That standard, set out in Nken v. Holder—another decision in which Chief Justice Roberts wrote for the Supreme Court—asks courts considering a stay of a ruling below to consider multiple factors, including the likelihood of success on the merits and the public interest. We have already discussed the likelihood of success, in which the district court and the Fifth Circuit unduly discounted the rationale for judicial deference. On the public interest, Roberts in Nken made two relevant points. First, he acknowledged that the public interest was served by enforcement of the immigration laws—since Nken concerned a stay of the removal of a noncitizen pending appeal of an agency order removing that individual, Roberts’s reminder was apt.

Second, more broadly, Roberts implied that courts should take care to fully weigh the public interest in allowing the politically accountable executive branch to make decisions on behalf of the public without undue judicial interference. Again, as Roberts noted in his opinion for the court in yet another case, Natural Resources Defense Council v. Winter—involving an injunction that the court vacated against a Navy training exercise—that public interest factor is particularly weighty in government decisions regarding national security and foreign affairs. Of course, here, too, some scholars argue that the court tipped the balance too far toward the executive branch. But that has been a consistent strand in stay determinations and other remedies cases, making the contrary determination stand out in bold relief, as Steve Vladeck noted in an op-ed on Aug. 26.

Moreover, if the stay standard is to have integrity and consistency, it cannot merely defer to pro-enforcement government decisions. As the Supreme Court noted in Regents in vacating the Trump DACA rescission, government decisions not to enforce are also worthy of deference. That is why the court in Regents held that then-Secretary Duke had failed to give proper consideration to the feasibility of merely deciding not to remove DACA recipients, even without granting those individuals eligibility for work authorization. Indeed, as the Supreme Court held in Heckler v. Chaney, some enforcement decisions are so embedded in executive decisions about resource allocation that they are not subject to judicial review at all. Because Mayorkas also mentioned resource allocation in his June 1 memorandum, similar deference, if not outright insulation from judicial review, should apply here, as well.

The Better Part of Valor: A New MPP Termination Decision?

While judicial decisions in this matter have departed markedly from precedent, the simplest course of action for Mayorkas may be to start from scratch, resume MPP in some form and issue a new memorandum. In this new memo, Mayorkas could mention the Texas agreement, although that agreement has now been officially terminated. Mayorkas could also discuss the Trump administration’s October 2019 memo on MPP and more concretely address MPP’s impact on border arrests. In addition, a new memo could provide more detail on in absentia removals, without addressing matters that would be unduly embarrassing for Mexican officials.

A new memo could also discuss an important August asylum initiative undertaken by Mayorkas and Attorney General Merrick Garland. This proposed rule flowed from an earlier proposal by former Immigration and National Service (ICE predecessor) Commissioner Doris Meissner. The Mayorkas asylum plan called for asylum officers to take over more of the adjudication currently done by immigration judges. The explanation for the rule suggests that this plan will accelerate adjudication by using the more informal asylum officer approach. Asylum officers also have the training to understand refugee claims and sort out well-founded from unfounded allegations. The program promises to reduce the backlog of far more than 1.3 million cases in immigration court and streamline enforcement of the Immigration and Nationality Act, while maintaining or even raising the rate of accurate asylum decisions. This program was not part of the record for the district court or the Fifth Circuit, but it will be part of the record in any future court cases if Homeland Security decides to go “back to the drawing board” and issue a new memo on ending MPP.

A new explanation of reasons for halting MPP would discuss both the new asylum adjudication proposed rule and the possibility of a pilot program involving asylum adjudication in Mexico. This last measure would show a concrete commitment to regional asylum adjudication mechanisms, which would be an important step forward.

In sum, the courts’ response to the Biden administration’s decision to terminate MPP has conflicted with established precedent on judicial deference in the arena of foreign affairs. This pivot may be part of a “new normal” that has deleterious consequences for the separation of powers. As the Biden administration waits to see how the courts respond, it needs to be proactive. A new memo on MPP would demonstrate that the Biden administration is making the appropriate adjustments to the policymaking process.