Although border and asylum issues remain front and center in virtually all U.S. immigration news, a key question lurking just beneath the surface is now arising in several different contexts: Can states authorize noncitizens to work?
The question burst into view recently, when thousands of recently-arrived immigrants in New York found themselves unable to work after their arrival in the United States. Both New York State Governor Kathy Hochul and New York City Mayor Eric Adams forcefully criticized the federal government for failing to provide work authorization quickly to the recently-arrived immigrants. State lawmakers then introduced legislation that would provide work authorization to people residing in New York who are seeking asylum within 45 days of their having applied—far more quickly than federal law allows.
The urgency surrounding that particular crisis seems to have passed, as the federal government subsequently granted expedited work authorization to nearly half a million Venezuelans through the Temporary Protected Status program, but the issue will likely recur before too long. Indeed, Adams was quick to note that the federal government’s decision facilitating work authorization for Venezuelans still left thousands of other recently-arrived refugees living in the city unable to work because they are from other countries.
Of course, recently-arrived refugees are hardly the only immigrants unable to work lawfully in the United States. Approximately 11 million undocumented people reside in the country, most of whom have been present for more than a decade without work authorization. A different state employment proposal—to benefit them, among others—has come from the Midwest. Even as some Republicans decry the large number of recent arrivals, others have noted that their states are experiencing a severe labor shortage that immigration law could help solve. Six months ago, the Republican governors of Utah and Indiana jointly wrote an op-ed in the Washington Post arguing that their states need more immigrant workers. They proposed a fascinating idea—to “let states sponsor immigrants,” essentially by having the federal government deputize states with the power to run their own immigrant worker programs. Versions of that proposal have been discussed for years. Utah actually enacted one more than ten years ago, though it never went into effect after the Obama administration sued to enjoin different provisions of the law. That same state-sponsored immigrant worker idea continues to circulate in Washington.
A third state employment proposal—one both more modest and closer to fruition than the first two—also garnered significant attention in the last year. It comes from California, where a coalition of undocumented university students, their professors, and legal academics (including me) have argued that the University of California (UC) already has legal authority to open employment opportunities to all students regardless of their immigration status. Public universities around the country have been closely watching the UC’s movement on the issue ever since, no doubt because there are now several hundred thousand undocumented students attending universities around the country, many of whom need to work to finance their educations and, in some cases, to complete their graduate degrees.
That states are looking to enact their own immigrant employment policies is a positive development. Congress has remained unable to pass any substantial immigration legislation for more than 20 years, including during the first two years of the Biden administration, when Democrats controlled both houses. As the prospects of federal legislative reform recede yet again, state and local policy that fosters immigrant-inclusion can play an important role in ameliorating the harms arising from our dysfunctional immigration system. It can also serve as an important political counterweight to anti-immigrant measures that some states have adopted, including a flurry of laws and practices that some Republican-led states have adopted since the Biden administration began.
To succeed, however, policymakers and advocates advancing pro-immigrant measures at the state level must have a nuanced understanding of the legal footing on which the various state employment proposals rest. The three proposals now receiving sustained attention all propose state government policy to facilitate the hiring of immigrants not currently authorized to work under federal law. But beyond that they differ in important ways—they respond to different policy problems, seek employment for different immigrant populations, and rest on distinct legal theories. Each of them warrants a closer look.
Historical Background on Employment Authorization for Immigrants
Someone generally familiar with immigration law might reasonably ask how states can enact any policy specifically about immigrants at all. The U.S. Supreme Court has described the power to regulate immigration as “unquestionably exclusively a federal power.” DeCanas v. Bica, 424 U.S. 351, 354 (1976) (citing, inter alia, Chy Lung v. Freeman, 92 U.S. 275 (1876)). So, isn’t immigration policy the federal government’s business alone?
No, it is not. As closer scrutiny of the law reveals, the federal government’s exclusive power over immigration exists primarily with respect to the power to admit and deport people, not the power to regulate their affairs while in the United States. States have long enacted various measures to make state-run programs open to undocumented immigrants, including drivers’ licenses, professional licensing privileges, in-state tuition for university students, and health insurance. Some states have also established worker-protective legislative frameworks that explicitly include protections for undocumented workers. California, where one in every eight people and nearly one in every four immigrants live, now makes a whole range of government programs open to all its residents regardless of status.
As to employment specifically, prior to the 1980s the federal government hardly regulated noncitizen employment at all, and instead left room for states to do so. When a California statute limiting noncitizen hiring was challenged on federal supremacy grounds in the 1970s, the Supreme Court sided with California, ruling that its statute was permissible absent “any specific indication in either the wording or the legislative history of the [federal Immigration and Nationality Act] that Congress intended to preclude even harmonious state regulation touching on [noncitizens] in general, or the employment of illegal aliens in particular.” De Canas v. Bica, 424 U.S. at 358.
Ten years later, Congress altered the landscape by passing the Immigration Reform and Control Act of 1986 (IRCA). That landmark statute is best-known for having granted amnesty to about three million people who had lived in the United States unlawfully for at least five years. But it also created the first federal prohibition on the hiring of undocumented people, as well as the requirement that employers verify eligibility for employment when hiring workers.
Although IRCA created a substantial new federal system regulating noncitizen employment, it did not entirely crowd out the states. Since 1986, states have continued to pass laws regulating the employment-related activity of noncitizens from time to time, and the Court has upheld some of those laws while striking down others, depending on how they were crafted. When Arizona enacted a state law making it a crime to work without federal work authorization, the Supreme Court struck it down, because it conflicted with Congress’s judgment that “it would be inappropriate to impose criminal penalties on unauthorized employees.” Arizona v. United States, 132 S. Ct. 2492, 2505 (2012). In contrast, when Kansas accomplished largely the same result by prosecuting undocumented immigrant workers using a law on its books that made it a crime to use a false Social Security number, the Court upheld it, explaining that IRCA did not “exclude a State from the entire field of employment verification.” Kansas v. Garcia, 140 S. Ct. 791, 797 (2020).
While the changing composition of the Court no doubt plays some role in explaining why Arizona lost in 2012 while Kansas won eight years later, these cases also illustrate that how a state crafts its employment provisions may play a critical role in determining whether courts uphold them. As this history illustrates, policymakers and advocates would do well to pay careful attention to the legal theories underlying the various state employment proposals when considering whether to adopt them.
Three Modern Proposals
So how do the three most prominent proposals for state employment authorization stack up against that legal framework?
A. State Employment Authorization Justified by Emergency
New York’s proposal is both the most ambitious and the least likely to survive legal challenge. It would provide expedited employment authorization to recently-arrived refugees as a way of easing the burden they impose on the state’s social services infrastructure. It seeks to do this by enacting legislation at the state level, without any federal legal action. While the proposed legislation is not entirely clear, it appears to empower the New York State Labor Commissioner to grant work authorization to anyone “in the process of filing” an asylum application who is “residing in New York within forty-five days of application.” That key provision appears to conflict with federal law—regardless of whether the “application” refers to the asylum application or instead the application for employment authorization. Under federal statute, people seeking asylum “shall not be granted [employment] authorization prior to 180 days after” they apply for asylum, more than four months after the New York legislators would like. 8 U.S.C. 1158(d)(2) (emphasis added). As the Supreme Court explained in Arizona, “state laws are pre-empted when they conflict with federal law. This includes cases where compliance with both federal and state regulations is a physical impossibility, and those instances where the challenged state law stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress.” 132 S. Ct. at 2501. Because it authorizes people to work when federal law prohibits such authorization, this legislation appears to be unconstitutional on its face.
While the proposed legislation itself includes no proposed findings or other indications that might tell us how its proponents would defend its legality, one conceivable argument might derive from the state’s emergency powers. This past spring, the state of New York declared an emergency with respect to the large numbers of refugees arriving in the state (following New York City’s announcement of its own emergency declaration the prior August). The state just extended its declaration for another month. By their terms, those orders do not purport to justify employment authorization, but the concept that states could declare emergencies as a way of justifying greater immigration-related policymaking is not new. Since the Biden administration began, several Republican-led states have purported to justify various immigration enforcement policies based on emergency declarations. Those moves build on Trump-era federal policy that implausibly invoked emergency powers to justify various immigration rules that Congress refused to enact, including funding for the border wall and Executive Order 13769 (often referred to as the “Muslim Ban”).
State and local governments have broad powers in emergencies, but it is highly doubtful that such power could support state employment legislation like the bill proposed in New York. Leaving aside the question whether the arrival of thousands of immigrants into New York could be characterized as the kind of emergency the statute contemplates, federal law already addresses such immigration emergencies. The Immigration and Nationality Act provides the Attorney General with authority to address “an actual or imminent mass influx” of noncitizens arriving by land or sea that “requir[es] an immediate Federal response.” 8 U.S.C. 1103(a)(10). Under that provision, the federal government has authority to deputize state and local law enforcement officers “to perform or exercise any of the powers, privileges, or duties conferred or imposed” by the immigration laws, including those related to work authorization. Thus, here (as elsewhere), the federal government appears to have exclusive authority to determine that a mass influx of immigrants has occurred and delegate greater authority to states on that basis.
Although courts are unlikely to uphold state laws directly authorizing noncitizens to work where federal law forbids it, it does not follow that proposing such measures is entirely pointless. One could argue that the Biden administration would not have granted TPS for Venezuela—thereby allowing nearly 500,000 people to quickly obtain work authorization—if state and local actors in New York had not proposed their own employment measures. Such proposals can give “voice” to state and local demands in a way that speech alone cannot. On the other hand, if such emergency state laws were upheld, they could also set precedent for more draconian responses. If a large influx of immigrants constitutes an emergency in New York, it may do the same in Texas, where the state might respond very differently.
In any event, what matters most for present purposes is that policymakers and advocates be keenly aware of whether the legislation they are proposing is primarily designed to express a demand, or instead intended to survive legal challenge.
B. State Immigrant “Sponsorship” Based on Federal Permission
In contrast to New York’s proposal, the Utah/Indiana concept would allow state governments to “sponsor” immigrants—both people already in the United States and those who might come from abroad—to work for businesses in their states. The governors propose to do this through federal legislation, though other versions of the idea rest on a combination of state legislation and federal regulatory action. Under either theory, the goal is to improve labor market conditions in those states facing severe worker shortages.
While this idea has some surface-level similarity to the proposal in New York, it ultimately rests on very different legal footing. Whereas the New York legislation purports to authorize state employment policy in conflict with federal law, the Utah/Indiana proposal requires federal action—whether legislative or administrative—that would delegate to states the authority to sponsor immigrants for employment. It would only become reality if either Congress or the executive branch gave explicit permission to the states. For that reason, it would not conflict with federal immigration authority, and simply does not test the limits of state authority over immigration policy in the way the New York proposal does.
Nonetheless, it is not free of legal risk, at least in the form based on executive branch action. Any Utah/Indiana style-proposal not involving legislation would likely rely on the Attorney General’s regulatory authority under IRCA, which states that “[a]s used in this section, the term ‘unauthorized [noncitizen]’ means, with respect to the employment of [a noncitizen] at a particular time, that the [noncitizen] is not at that time either (A) [a noncitizen] lawfully admitted for permanent residence, or (B) authorized to be so employed by this chapter or by the Attorney General.” 8 U.S.C. 1324a(h)(3) (emphasis added). Just as the Attorney General has long granted work authorization to people awaiting adjustment of status or granted deferred action, the Utah/Indiana model would allow qualifying undocumented people in requesting states to receive permission to work under regulations implementing Section 1324a(h)(3).
As those who have closely followed the litigation over the Deferred Action for Childhood Arrivals (DACA) program may recall, the meaning of the phrase “or by the Attorney General” in 8 U.S.C. 1324a(h)(3) has been the subject of some dispute. Since the Obama administration, the federal government has pointed to the regulations implementing that provision as the legal basis for providing employment authorization to recipients of deferred action under DACA. That was also the government’s argument in support of DACA’s ill-fated successor, the Deferred Action for Parents of Americans and Lawful Permanent Residents (DAPA) program. DAPA would have granted temporary protection from deportation and work authorization to well over three million unlawfully-present immigrants. But the Fifth Circuit struck it down, finding that Congress would not have delegated to the executive branch the authority to operate a program granting “lawful presence” and work authorization to so many people. In reaching that result, the Fifth Circuit also rejected the Obama administration’s reliance on Section 1324a(h)(3) to support the program, albeit without any analysis of the statute’s text. Texas v. United States, 809 F.3d 134, 183 (5th Cir. 2015). That decision was affirmed by an equally divided Supreme Court (though we cannot know why, because the Justices issued no opinions in the case). Since that time, the Fifth Circuit has also ruled DACA’s work authorization component unlawful, relying extensively on its DAPA decision to reach that result. Texas v. United States, 50 F.4th 498 (5th Cir. 2022).
Of course, that the Fifth Circuit held that view does not mean the Supreme Court would agree. DACA survived its last Supreme Court challenge, and the Supreme Court has reversed the Fifth Circuit in important immigration cases twice in the last two years. Moreover, there are important differences between the Utah/Indiana proposal and the DAPA and DACA programs. Most obviously, at least some versions of the Utah/Indiana proposal need not provide protection from deportation along with work authorization, which is part of what troubled the Fifth Circuit. The program also could be smaller—presumably the federal government could dictate its size by choosing how many immigrants to let states sponsor. This could matter even in the Fifth Circuit, as that court pointed to DAPA’s size as part of its rationale for striking it down (though it also found DACA unlawful, despite acknowledging that it was substantially smaller than DAPA).
While no one knows how the Supreme Court will ultimately rule on DACA, let alone how it might rule some day on the hypothetical Utah/Indiana proposal, for present purposes what bears emphasis is that advocates and policymakers should understand that the fates of these two very different ideas may well be intertwined, as they ultimately rest on the same provision of federal law. If the Supreme Court upholds DACA, it should not be hard to design an executive branch version of the Utah/Indiana proposal that would survive as well. Conversely, a decision striking down DACA would likely make the path to survival for the Utah/Indiana proposal far more uncertain.
C. States as Employers
The third state employment proposal to garner significant attention in the recent past comes from the University of California. That proposal would permit the University to provide educational employment opportunities to all students regardless of their immigration status. The need for such opportunities has become pressing in the university context because the DACA program has been effectively closed to new applicants for most of the last five years under court orders that bar the processing of new applications. In addition, because the Biden administration did not update the program’s eligibility date, no one who arrived in the United States after June 15, 2007, is eligible even to apply. As a result, there are now nearly 300,000 students in higher education across the country who do not have DACA, and therefore lack the ability to obtain federal work authorization.
The UC adopted a groundbreaking policy to address this problem in May, and is now poised to take further steps to implement that resolution. Of course, UC’s proposal does not entirely solve the problems arising from DACA’s slow demise, as it does not purport to protect students from deportation. But it would permit undocumented students to accept employment within the University—whether as teaching assistants or medical residents so they can complete their advanced degrees, or at the cafeteria, bookstore, or other jobs that help pay tuition and living expenses.
The UC policy rests on a fundamentally different legal theory from those undergirding the other state employment proposals. It rests on a novel interpretation of the same 1986 federal law prohibiting the hiring of undocumented people—that the law does not apply to state governments when they act as employers. That theory finds support in IRCA’s text, as it prohibits any “person or other entity” from hiring undocumented workers, and goes on to specify that “entity includes an entity in any branch of the Federal government,” without mentioning state governments. 8 U.S.C. 1324a(a)(1), (7). That plain language argument is bolstered by federalism principles. Whom state governments choose to hire is traditionally a matter for them to decide, as De Canas explains. While the federal government can regulate in that area—as it has done in Title VII, the Fair Labor Standards Act, and the Americans with Disabilities Act, among other statutes—it must do so explicitly. Indeed, all of those statutes (and several others) specify that they apply to state governments or governmental entities, whereas IRCA does not. Although this theory is unquestionably novel, many prominent legal scholars endorsed it last year in an open letter and accompanying legal memo that I co-authored.
California’s state employment proposal is more modest than those from New York and Utah/Indiana in two respects. First, it would not authorize people to work in the private labor market. In its current form it applies only to undocumented students without DACA. An aggressively pro-immigrant state or local government could apply the theory more broadly by opening many (or even all) of its government jobs to everyone regardless of status, but even such measures would only permit state or local government entities to do such hiring; they would give no additional hiring authority to employers in the private market—where the vast majority of people in the U.S. economy work. Second, unlike the first two state employment proposals, California’s proposal does not require any new federal or state legislation or even regulation. The theory rests on the view that federal law already permits state government entities to choose whom they hire irrespective of immigration status. So long as state law contains no separate prohibition, state employers are already free to adopt the proposal.
The UC proposal is no doubt an attempt to give “voice” to the need for undocumented students to work, but it is far more than that, given the strength of the legal arguments supporting it. And it is not an attempt to advance immigration policy through federal-state cooperation, as is the Utah/Indiana model. Rather, it involves an attempt to get state actors to act unilaterally, within their own limited sphere, to improve conditions for one group of undocumented immigrants—those pursuing higher education.
Conclusion
Given on-going Congressional inaction on immigration policy, the urgent needs of newly-arrived asylum seekers, and the political demands of the long-resident undocumented population that will only grow as DACA becomes increasingly irrelevant, we can expect the calls for state-led pro-immigrant policies to grow. While supporters may demand all of these proposals, and opponents reject them all, it behooves everyone interested in the area to understand precisely where each one fits in the ever-shifting immigration policy landscape. Employment proposals resting on state emergency powers are very unlikely to succeed in the courts, as they conflict with federal law; but they can nonetheless play an important role in articulating state and local political demands. In contrast, proposals that rest on the permission of the federal government do not suffer from that flaw. They offer great promise, but their success requires cooperation (rather than mere inaction) from Washington, and even then their survival may depend on what the courts ultimately decide about the legality of DACA itself. Finally, proposals limited to employment by state and local government entities could survive even if the other two types of programs fail, though they rest on a novel legal theory, and ultimately would provide support to a necessarily-limited population of undocumented immigrants.