In a decision on Dec. 31 in Doe #1 v. Trump, the U.S. Court of Appeals for the Ninth Circuit upheld President Trump's Oct. 2019 proclamation barring entry of immigrants without "approved" health insurance (a policy often referred to as the "uninsured ban"). Judge Daniel Collins, a Trump appointee, wrote the panel's opinion, holding that the uninsured ban was within the president's authority under the Immigration and Nationality Act (INA). a George W. Bush appointee, Jay Bybee, joined in that opinion; Judge A. Wallace Tashima, a Clinton appointee, dissented. The panel's ruling raises the stakes even higher for President-elect Joe Biden's first days in office, since Biden has said he will move quickly to withdraw several of President Trump's immigration proclamations.
The uninsured ban relies on 8 U.S.C. § 1182(f), which empowers the president to bar entry of foreign nationals who would be "detrimental to the interests of the United States.” This same provision supported President Trump's 2017 travel ban, which the Supreme Court upheld in Trump v. Hawaii. The uninsured ban would bar many immigrants who lack an "approved" insurance plan, defined as a plan not supported by a government subsidy.
The proclamation's definition of an "approved" plan likely stems from President Trump's well-known opposition to the Affordable Care Act (ACA). As this explainer describes, the ACA seeks to provide comprehensive health coverage. ACA subscribers can obtain a wide range of needed treatment, including preventive care. By providing this broad range of care, Congress hoped to promote the overall health of Americans and thus lower health-care costs over time. Congress specifically rejected the model favored by President Trump, which promotes bare-bones insurance plans that are less expensive than comprehensive plans. These bare-bones plans lower costs by sharply reducing coverage. Many bare-bones plans only provide coverage for catastrophic health conditions. Typically, bare-bones plans do not cover preventive care. Approved plans under the proclamation include an array of bare-bones insurance policies that provide minimal benefits and are thus incompatible with the comprehensive coverage that the ACA encourages.
In practice, the ban bars the entry of many who would otherwise be eligible for visas—spouses, adult children, parents and siblings of many U.S. citizens, as well as the spouses of lawful permanent residents (LPRs). This is because families already in the U.S. in which the primary breadwinner works for a salary at or just above the minimum wage often lack employer health insurance plans. And without an employer plan, it becomes effectively impossible to sponsor relatives under the act unless the noncitizen can personally afford, out-of-pocket, a plan designated as "approved" under the ban.
The core of the Ninth Circuit's Doe opinion is a broad reading of 8 U.S.C. § 1182(f). In his opinion, Judge Collins relied heavily on Chief Justice Roberts's opinion for the Supreme Court in Trump v. Hawaii. In that case, Roberts observed that § 1182(f) "exudes deference in every clause." For the Doe majority, that broad textual grant of power was sufficient. Citing Chief Justice Roberts's broad view of § 1182(f) in Hawaii, Judge Collins asserted that the president merely had to make a finding that entry of the immigrants identified in the proclamation would be "detrimental." Once the president made that finding, Collins asserted, the court owed the president deference.
The Doe panel's deference rested on President Trump's finding that lawful immigrants were "three times more likely than U.S. citizens" to lack health insurance and that uncompensated health care costs strain the health-care delivery system. Also following Hawaii, Judge Collins explained that the uninsured ban was subject to periodic review by the executive branch. At least in theory, that provision for periodic review ensured that the ban would last only as long as U.S. economic conditions warranted.
To uphold the uninsured ban, the Doe court rejected arguments made by the challengers and Judge Tashima in his dissent that the ban conflicted with other statutory proclamations and departed from longtime practice under § 1182(f). On the latter point, I served as co-counsel (along with Loeb & Loeb's Neil Nandi and Laura McNally and Penn State's Shoba Sivaprasad Wadhia) on an amicus brief that I discuss below. On the former point, the Doe court found no conflict between the ban and provisions of the Affordable Care Act (ACA)—an issue I unpacked in an earlier post and I discuss further below.
Why might there be tension between the ban and the ACA? As Chief Justice Roberts explained in his opinion for the Supreme Court in King v. Burwell, Congress sought in the ACA to promote sustainable access to health care through a pillar of sound insurance: pooling high- and low-risk subscribers. Risk pooling works by enrolling healthy individuals who pay more in premiums than they withdraw in benefits. Enrollment of these low-risk subscribers balances out less healthy individuals—such as the elderly and those with preexisting conditions—who are more likely to claim benefits (King v. Burwell at 2493). Several ACA provisions authorize enrollment of lawfully present noncitizens. And experts have noted that encouraging enrollment of immigrants, many who are likely to be young and healthy, served the risk-pooling at the heart of the statute.
Testimony given by public health experts to Congress before it enacted the ACA noted noncitizens' edge in personal health. At a 2008 hearing, one public health official explained that immigrants are "by and large healthier than American born residents"; they "live longer, have less chronic disease, and use less medical resources per capita." Indeed, the "healthy immigrant effect" is a recurring refrain in the public health canon. Enrollment of immigrants thus fit the ACA's risk-pooling plan. By refusing to classify ACA plans as "approved," the uninsured ban undermined the ACA's risk-pooling foundation.
The ban's interaction with the ACA plays out against the backdrop of today's dominant theory of statutory interpretation: textualism. Justice Antonin Scalia, textualism's foremost exponent, asserted that text was the most reliable guide to interpretation.. However, Scalia also acknowledged the importance of context, stating in Whitman v. American Trucking Associations that a statute's "words … are given content … by their surroundings" (p. 466). Judge Collins clerked for Justice Scalia in the 1990s and echoed this holistic dimension in his March 2019 confirmation hearing, telling senators that in reading statutes, a court must "look at the larger context of the problem." In Doe, Judge Collins payed insufficient attention to this holistic aspect of Scalia's approach. Adopting a holistic approach, the Doe majority would have held that the uninsured ban conflicted with the ACA's risk-pooling dynamic.
Judge Collins's opinion in Doe eschewed context, opting instead for a narrow view of the ACA's text. According to Collins, the ACA only authorized the enrollment of immigrants who "already are 'lawfully present in the United States'" (emphasis in original). (p. 29) In contrast, the ban only applies to foreign nationals who are seeking to enter the United States. Those individuals are not yet physically or lawfully present in the country. Citing this distinction, Collins maintained that the ACA and the uninsured ban occupy separate spheres, obviating any conflict between the statutes.
This move is superficially persuasive, but only if one views the ACA as an arid text divorced from any practical plan of action. Congress's repeated mention of lawfully present immigrants in the ACA did more than simply authorize immigrants' enrollment. It also signaled the importance of immigrants to the financial stability of the ACA's framework. And stability does not turn merely on current economic conditions: "short-term stability" is an oxymoron. Borrowing Scalia's evocative phrase in Wrigley, stability requires "courses of conduct." As Chief Justice Roberts's analysis in King v. Burwell showed, an insurance plan's ability to pool high and low risks over time is crucial to its success. An insurance plan that works for today but collapses tomorrow has insured very little.
This focus on stability should have informed the Doe Court's reading of the interaction of the ACA and the uninsured ban. Over time, blocking immigrants' entry because they lack "approved" insurance has the same effect as blocking enrollment in the ACA of immigrants who are already lawfully present. The ACA’s stability is bolstered by immigrants as a general population group, regardless of whether or not they are “already here.” In each case, the executive branch has undermined the ACA's foundational premise.
A more nuanced account of textualism would have acknowledged this conflict. Instead, the Doe majority focused myopically on today's lawfully present immigrants but refused to acknowledge the stability that future immigrants can provide for the ACA's framework. Admittedly, a version of textualism that dwells unduly on context sacrifices clarity and consistency. To quote Scalia and his co-author Bryan Garner's citation to the great U.S. Court of Appeals of the Second Circuit judge Learned Hand in Reading Law: The Interpretation of Legal Texts, the panel practiced a "sterile literalism" that failed to see the "forest for the trees." (p. 356).
The Doe court also discounted important parts of Chief Justice Roberts's opinion in Hawaii that stressed the national security aspects of the travel ban. Surveying past practice under § 1182(f), Roberts stated that presidents often used the provision to "retaliate for conduct by … governments that conflicted with U.S. foreign policy interests." As our amicus curiae brief on this point explained, one prominent use of § 1182 during the Reagan administration concerned an immigration dispute with Castro's Cuba. Other uses have included blocking war criminals, international cyber thieves and human traffickers. The uninsured ban departs from this strong trend and uses the provision for domestic economic reasons unconnected to national security or foreign policy interests. Judge Tashima cited this marked departure in his dissent.
However, the Doe majority viewed past practice as irrelevant and the foreign/domestic distinction as "unworkable," despite past presidents' clear preference for action based on national security. Here, the majority again took a literalist turn, suggesting that "all" restrictions under § 1182(f) could be viewed as domestic to a "greater or lesser degree." (p. 37) But that point distorted the challengers' argument against the ban. Of course, by definition, any restriction on immigration affects the number and identity of persons who are lawfully present in the country. To that extent, any conceivable restriction under § 1182(f) has some domestic component. But it is still possible to distinguish between foreign policy-driven restrictions—such as Reagan's Cuba measure—that have incidental domestic impacts, and restrictions driven by purely domestic concerns. The Doe majority's eagerness to conflate root causes of a measure and incidental effects is further evidence of the court's literalist approach.
But despite the court’s decision, at least one important step remains before the uninsured ban can take effect. Moreover, the looming presidential transition might bring respite for the petitioners.
Over 180 days have elapsed since the effective date of the proclamation on Nov. 3, 2019. Because of the additional time that has passed, under section 4 of the proclamation, Secretary of State Mike Pompeo will have to find that the proclamation is still necessary to protect U.S. workers. The challengers to the ban are likely to seek an en banc rehearing from the full Ninth Circuit. According to the Federal Rules of Appellate Procedure, the challengers have 45 days to petition for the rehearing. And Pompeo is unlikely to issue his proclamation while their petition is pending.
By the time the 45-day period expires, the new administration will be well in place. President Biden can and should withdraw the uninsured ban very early in his presidency, along with other Trump administration immigration changes such as the nonimmigrant visa ban. The success of the uninsured ban in the courts also holds an important lesson for Congress. It should pass legislation that would amend § 1182(f) to limit the president's power in this context (as a related example, see this bill passed by the House of Representatives to override President Trump's travel ban). Given the courts' deference in Hawaii and Doe, legislation is the best way to guard against a repeat of President Trump's signature sweeping use of this provision.