Published by The Lawfare Institute
in Cooperation With
On Nov. 26, Judge Michael H. Simon of the U.S. District Court for the District of Oregon issued a preliminary injunction against the operation of President Trump’s October immigration proclamation. The proclamation denies entry into the United States for otherwise qualified visa applicants unless they are likely to receive “approved health insurance” within 30 days of entry or can pay for their health care. (See my earlier analysis of Simon’s temporary restraining order.) In his core ruling that the uninsured ban clashes with the Immigration and Nationality Act (INA), Simon was correct. However, some aspects of his statutory and constitutional analysis are flawed.
Simon’s ruling over-values one piece of legislative evidence from 2013 and also addresses a constitutional issue—whether the presidential proclamation rested on an unconstitutional delegation of legislative power—that Simon did not need to decide.
Trump’s ban on uninsured immigrants relies on 8 U.S.C. § 1182(f), which permits the president to bar entry of foreign nationals that would be “detrimental to the interests of the United States.” Trump cited this same INA provision in support of his travel ban, which the Supreme Court upheld in Trump v. Hawaii. The ban on the uninsured would bar any immigrant without an “approved” plan, defined as a plan not supported by a government subsidy. This prohibits Affordable Care Act (ACA) recipients from sponsoring otherwise qualifying relatives; only those with private insurance can sponsor newcomers. In turn, this means that many would-be lawful immigrants cannot become part of the ACA’s patient pool.
The ban’s restrictions would substantially reduce the number of citizens who could pursue the INA’s key goal of family reunification. In practice, the ban bars the entry of spouses, adult children, parents and siblings of many U.S. citizens who would otherwise be eligible for visas, as well as the spouses of lawful permanent residents (LPRs). Families in which the primary breadwinner works for a salary at or just above the minimum wage often lack employer health insurance plans; without an employer plan, sponsoring relatives becomes effectively impossible under the ban. The ban would not, however, bar the admission of refugees or the minor children of U.S. citizens and LPRs.
Simon noted that the uninsured ban undermines both the structure of the INA and provisions of other statutes, such as the ACA, that provide for the eligibility of lawfully present foreign nationals for subsidized health insurance. While Simon alluded to this point, he did not fully flesh out its doctrinal import, which is worth exploring further. Consider that Justice Antonin Scalia and Bryan Garner, in their classic study of statutory interpretation, “Reading Law,” highlighted the venerable canon that holds that statutes in pari materia (i.e., addressing the same topic) should be interpreted “harmoniously” (p. 252). Through harmonious interpretation, courts vindicate a value integral to the passage of legislation: the principle that “the body of the law should make sense” (p. 252). Scalia and Garner recounted that Justice Felix Frankfurter had warned in an earlier study of statutory construction that a judge would fail to honor this precept if, in parsing one law enacted by Congress, “[her] eye is closed to considerations evidenced in affiliated statutes” (p. 252).
In fact, the Trump proclamation undermines the ACA’s plan. As Simon observed (and the insightful and exhaustively researched amicus brief by California and other states attests), the ACA repeatedly mentions the eligibility for subsidized insurance of foreign nationals (noncitizens) who are lawfully present in the United States—the very population that the uninsured ban targets. For example, 42 U.S.C. § 18051(e)(1)(B) expressly provides for eligibility of lawfully present noncitizens.
In barring lawfully present immigrants from ACA enrollment, the uninsured ban also conflicts with a core policy goal of the ACA. Several ACA provisions demonstrate the policy behind the ACA that the Supreme Court acknowledged in King v. Burwell: The ACA aimed to broaden the risk pool of qualified individuals.
As Chief Justice John Roberts explained in his opinion for the court in King, all insurance hinges on the prudent dispersion of risk. A given individual’s insurance premiums never pay for the full cost of treatment, should that person require extensive medical attention. If all insured individuals were at high risk of needing treatment during the same period of time, the insurance plan would fail. Rather, a sustainable insurance plan depends on receiving premiums from a substantial number of individuals who will not need care for the time period in question. An optimal public health insurance risk pool thus includes both healthy—often younger—individuals who pay into the system but rarely need care, and less healthy—often older—individuals who require more care.
According to the California amicus brief in Doe, immigrants promote sound risk-pooling. Studies show that immigrants as a group are likely to be “younger, healthier, and below-average users of healthcare” than the overall insured population. This means immigrants contribute more in premiums than they take out in benefits. Without these lawfully present immigrants participating in the ACA, the program will receive less money in premiums and pay out proportionately more money in reimbursements for medical treatment, making the program unsustainable. In this fashion, the exclusion of lawfully present immigrants from the ACA risk pool will make it more difficult to fund the provision of affordable care under the ACA for older, sicker individuals who utilize health care at higher-than-average rates.
The proclamation skews responsibility for health care costs in another way that clashes with the ACA. Without access to insurance under the ACA, noncitizens could still enter the country, if they were able to purchase plans that are acceptable under the proclamation. For example, immigrants could buy unsubsidized bare-bones plans, which are considered “approved” under the proclamation. Increased reliance on bare-bones plans, which leave many conditions uncovered, adversely affects states, counties, and localities, since noncitizens lacking coverage for some conditions will have to seek treatment in public hospitals run by these subfederal entities. Those entities will then have to absorb the cost of providing such treatment. Imposing such costs on states, counties and localities is precisely the kind of inequitable risk allocation that Congress passed the ACA to remedy.
Under the in pari materia canon, a court should reconcile the INA with the ACA’s express inclusion of lawfully present noncitizens; this would mean striking down the uninsured ban. Only that approach would, in Scalia and Garner’s apt phrase, “make sense” of the two statutes. That reconciliation is achievable here, since the INA has no express bar on noncitizens who utilize benefits under the ACA. The bar on uninsured noncitizens is purely a presidential measure. There is no logically sound way to shoehorn the uninsured ban into the ACA’s comprehensive structure. To do so, a court would have to argue that Congress intended to give with one hand and take away with the other: granting lawfully present noncitizens access under the ACA in order to promote the risk-pooling at the heart of the ACA’s statutory scheme, while authorizing the president under the INA to deny admission to this very same group. Since the INA does not expressly bar individuals who have insurance under the ACA or other subsidized plans, the in pari materia canon suggests that a court should view the proclamation as exceeding legislatively delegated executive power.
While Simon gestured toward the argument in the previous paragraph by mentioning the ACA and the risk-pool arguments explained fully in the excellent California amicus brief, his decision does not fully develop the in pari materia point. The failure to develop that argument is a weakness in the opinion. Perhaps an appellate court will have the chance to remedy this omission.
Another issue with the uninsured ban is its clash with the general/specific canon, which presumes that specific provisions, such as the INA’s public charge provision, govern more general grants of authority, such as § 1182(f). As Scalia noted in RadLAX Gateway Hotel, LLC v. Amalgamated Bank, the general/specific canon has particular relevance to specialized areas of law, such as immigration or bankruptcy, where Congress “has enacted a comprehensive scheme and deliberately targeted specific problems with specific solutions.” In that context, Scalia and Garner explained (p. 183), “the specific provision comes closer to addressing the very problem” that Congress sought to address.
Here, the narrow provision is the INA’s 8 U.S.C. § 1182(a)(4), which bars the admission of a foreign national who is “likely … to become a public charge” (i.e., dependent on the government). The public charge provision, which Congress included in the immigration statute well over a century ago and amended in 1996, seems to impose a “totality of the circumstances” test entailing consideration of myriad factors such as “age;” “health;” “family status;” “assets, resources, and financial status[;]” and “education and skills.” After extensive notice and comment under the Administrative Procedure Act (APA), the Department of Homeland Security published a final rule in August on the INA’s public charge provision that takes a similarly flexible approach to noncitizens’ lack of health insurance, viewing it as “one factor in the totality of the circumstances.” It is notable that on health insurance, Homeland Security’s approach after an elaborate round of APA notice and comment was more flexible than the approach taken in the proclamation.
In contrast to the Homeland Security rule, the uninsured ban flags lack of unsubsidized health insurance as categorically barring admission of an otherwise qualified foreign national. In setting up this stark litmus test, the uninsured ban ignores the more nuanced inquiry established under the public charge provision and instead relies on the general language of 8 U.S.C. § 1182(f), terming noncitizens without “approved” health insurance “detrimental to the interests of the United States.” That reliance on § 1182(f)’s general authorization flouts the general/specific canon’s strong preference for concreteness as a sound guide to legislative design.
Simon’s decision noted the clash between the general statutory authority for the ban and the INA’s specific public charge provision. Moreover, to reinforce the perils of relying on a general provision such as § 1182(f) over the more specific inquiry in § 1182(a)(4), Simon also observed that Congress, in the course of deliberating about proposed legislation that would coalesce into the Illegal Immigration Reform and Immigrant Responsibility Act, had considered and rejected a categorical approach like the one in the proclamation. The rejected bill would have set the trigger for a finding that an immigrant was likely to become a public charge at the receipt of 12 months of public assistance. Congress’s skepticism about inserting a per se test in the statute is further evidence that a categorical test tied solely to obtaining an “approved” health care policy is inconsistent with Congress’s detailed scheme.
That said, Simon inappropriately doubled down on Congress’s rejection of per se tests in citing the decision by a bipartisan group of U.S. senators not to include a similar per se benefits test in proposed 2013 immigration reform legislation. Simon’s citation of the group’s decision ignores one key factor: That proposed legislation never became law. Therefore, the decision of the 2013 bill’s sponsors not to include a per se benefits test says nothing about the INA’s duly enacted provisions. The annals of Congress are littered with the shards of legislation that failed to pass. Proposals too frail to gain entrance into the final versions of unsuccessful legislation are far too weak to provide guidance on the more robust legislative products that actually made it into the U.S. Code. Simon’s analysis would have been even more compelling if he had omitted his citation to the ultimately fruitless 2013 legislative deliberations.
Simon’s statutory analysis was also less sharp than it could have been because he invoked constitutional arguments that were not necessary for his holding. According to the court, if the INA had permitted such a sweeping exercise of presidential authority under § 1182(f), the provision would have been an unconstitutional delegation of power from Congress to the executive branch. Arguments about constraining Congress’s ability to delegate matters to administrative agencies have surfaced on occasion for decades and are attracting renewed interest today, based on concerns voiced most recently by Justice Brett Kavanaugh (see his opinion attached to the Supreme Court’s denial of certiorari in Paul v. United States) and Justice Neil Gorsuch (see his dissent in Gundy v. United States).
This case, however, is a poor vehicle for a nondelegation challenge. Courts have historically accorded Congress broad deference regarding immigration. In addition, the court’s upholding of the travel ban in Trump v. Hawaii illustrates the difficulty of constitutional challenges to executive discretion over immigration. Unless Simon believed the statutory arguments against the uninsured ban are weaker than he acknowledged, his reliance on the nondelegation doctrine seems unnecessary at best and counterproductive at worst, since it distracted from those strong statutory arguments.
Moreover, if Simon invoked delegation to clarify why the court’s deference in Trump v. Hawaii did not also dictate a victory for the president in this case, Simon had more effective ways to explain that Hawaii did not apply. In Hawaii, Chief Justice Roberts, writing for the court, cited the national security and foreign affairs aspects of both the travel ban and earlier uses of § 1182(f). Roberts asserted that, in imposing the travel ban, President Trump and the ban’s interagency drafters aimed to “protect national security and public safety, and induce improvement” in vetting by affected foreign nationals’ home countries. Prior uses of § 1182(f) discussed by the chief justice—such as President Carter’s curbs on Iranian nationals in the United States during the Iranian hostage crisis, President Reagan’s limits on Cuban immigration to induce changes in the Castro regime’s policies, and President Clinton’s efforts to interdict asylum seekers on the high seas—had a similar national security/foreign affairs signature, enabling the United States to “retaliate for conduct by … [foreign] governments that conflicted with U.S. foreign policy interests.”
In contrast, the new insurance test does not even assert a national security or foreign policy rationale. Instead, it rests entirely on a concern for minimizing health-care costs. Yet, as I’ve highlighted above, the proclamation’s implementation of this concern clashes squarely with the ACA. Moreover, instead of emerging from an interagency process, as did the travel ban upheld in Hawaii, the proclamation undermines the more flexible approach taken by Homeland Security. Even without a dubious nondelegation argument, the case for judicial deference to the executive branch is far weaker here than in Hawaii.
The Justice Department will soon decide if it wishes to appeal Simon’s ruling to the U.S. Court of Appeals for the Ninth Circuit and, if so, whether it wishes to seek a stay of the injunction pending that appeal. If the government appeals and the Ninth Circuit denies a stay, the case may end up in the Supreme Court in several weeks (see Stephen Vladeck’s paper on the government’s efforts to speed up litigation at the Supreme Court). Then, the entire court will have a chance to weigh in on this important issue at the intersection of immigration and health care.