Chapter 31 — Immigration Policy: The Permanent Crisis and the Competing Visions

"On immigration, the United States has no policy in the formal sense. It has a 1965 statutory architecture, fifty years of accumulated executive workarounds, a court system that cannot keep up, a political system that punishes any legislator who votes for a comprehensive bargain, and an electorate that wants tighter enforcement and more legal pathways at the same time. The crisis is not that the system is broken. The crisis is that the system, to all available evidence, is functioning exactly as the political incentives reward." — Composite paraphrase, drawing on Cristina Rodríguez, Reihan Salam, and Gabriel Chin

31.1 Why this chapter is hard to write

Immigration is, by most measures, the most politically fraught policy area in modern American politics. Polling on it is volatile but consistent in one respect: substantial majorities of Americans want both stricter enforcement at the border and a workable pathway for long-term unauthorized residents. Congress has not enacted a comprehensive immigration reform since 1986. Every few years a bipartisan group of senators announces a deal; every few years it dies. Three different presidents have governed primarily by executive action, and a fourth — Trump in his second term — is governing through the most aggressive interior-enforcement push in postwar memory. Federal courts are the venue where most of the actual policy is being made.

That makes immigration unusually hard to teach for two reasons. The first is that the empirical claims and the normative claims are tangled. Some of what is contested — whether immigrants commit crimes at higher rates than the native-born, whether immigrants are net contributors to public finances, whether wage effects on similar-skill native workers are large or small — is settled in the academic literature. Some of what looks like empirical disagreement is actually disagreement about values: how much weight to give to the claims of long-term residents, how to weigh humanitarian considerations against rule-of-law considerations, what kind of country one wants. The second difficulty is that the political terrain itself is moving fast. The 2024 election turned, in part, on immigration; the new administration's enforcement push is reshaping the empirical picture month by month.

This chapter tries to do three things. It states the legal architecture and the empirical landscape clearly. It steel-mans the three main families of policy positions — restrictionist, comprehensive-reform, and pro-expansion — at the level the smartest advocates of each present them. And it is honest that the policy stalemate has substantive causes beyond bad faith. Both parties have constituencies that benefit from the status quo. Both parties have constituencies that are punished by it. The political-economy explanation for the cycle of failed reform — that as long as the issue benefits one side electorally, comprehensive reform faces electoral headwinds — is not a partisan accusation. It is a structural feature of how the issue has come to function.

Read this chapter looking for the gap between the rhetorical positions ("secure the border, no amnesty"; "welcome the stranger, no deportations") and the operational realities (a border wall already exists across most of the high-traffic sectors; large-scale deportation has constitutional and capacity constraints; mixed-status families make purely abstract policy talk impossible). The substance of immigration policy is in the operational details.

31.2 The empirical landscape

A coherent immigration discussion requires a shared set of numbers. Most political argument on the topic proceeds without one. Begin here.

31.2.1 Total immigrant population

As of 2025, the United States has a total population of roughly 335 million. The foreign-born share — anyone living in the country who was not a US citizen at birth — is approximately 14.5 percent, or about 48 million people. That share is historically high. It matches the early-1900s peak of about 14.7 percent, reached in 1910, and is roughly triple the 4.7 percent share recorded in 1970. The country is, in this respect, more like the immigrant America of Ellis Island than the relatively closed America of the mid-twentieth century.

The 48 million foreign-born residents are not a single legal category. They break down approximately as follows:

• Naturalized US citizens: roughly 24 million. These are former immigrants who have completed the naturalization process. They vote, hold passports, and are citizens for all legal purposes.
• Lawful permanent residents (green-card holders): roughly 13 million. They have permanent legal authorization to live and work in the US, can travel internationally, and are eligible to apply for citizenship after meeting residency requirements.
• Other lawfully present non-citizens: roughly 5 to 6 million. This category includes students on F-visas, H-1B and other employment-visa workers, asylum seekers awaiting hearings (a large and growing share), Temporary Protected Status (TPS) recipients, parolees admitted under specific humanitarian programs, and various other lawful temporary statuses.
• Undocumented: estimated at 11 to 13 million. The estimate is made by demographers (Pew, the Center for Migration Studies, the Migration Policy Institute, and the Department of Homeland Security itself) using residual methods — comparing total foreign-born population from the Census American Community Survey with administrative records of legal admissions. Estimates from different sources vary by roughly two million; the standard error is non-trivial.

The undocumented share has been roughly stable since approximately 2008. The composition has changed: the share from Mexico has declined as Mexican migration has fallen, and the share from Central America and increasingly South America has risen. But the total number is not as different from 2008 as much current rhetoric implies.

31.2.2 Recent border dynamics

The volatility in the immigration debate of the past four years is largely about the flow at the southwestern border, not the stock of long-term residents.

From roughly 2010 through 2020, apprehensions at the southwestern border ran in the range of 300,000 to 800,000 per year — high by historical standards but stable. Beginning in 2021, the numbers rose sharply. Apprehensions reached approximately 1.7 million in fiscal 2021, 2.4 million in fiscal 2022, and 2.5 million in fiscal 2023. Fiscal 2024 began at similar levels and then fell sharply in the second half.

In the early months of the second Trump administration (January 2025 onward), apprehensions at the southwestern border declined by approximately 90 percent from fiscal-2023 peaks, driven by a combination of Mexican enforcement cooperation, expedited removal authority being applied more widely, and the practical message to potential migrants that the door had closed. As of the latest available data, apprehensions at the southwestern border are running at the lowest sustained levels in approximately a decade.

The political interpretation of this rise and fall is contested. Pro-restriction advocates argue that the rise from 2021 was driven by the rollback of Trump-1 enforcement policies and the perceived weakening of border deterrence; the fall under Trump-2 demonstrates that enforcement works. Pro-immigration advocates argue that the rise was driven by Western Hemisphere economic and political collapses (Venezuela, Haiti, Cuba, Honduras) for which US policy was not the principal cause, and the fall reflects bilateral cooperation with Mexico more than novel enforcement. Both readings have empirical support.

31.2.3 What the numbers do not tell you

It is important to mark what the headline numbers do and do not capture. Apprehensions are encounters between Customs and Border Protection (CBP) and migrants — they are not the same as successful entries or deportations. A single migrant may be apprehended multiple times. A migrant who is encountered, processed, and released into the interior pending an immigration-court hearing is counted as an apprehension. A migrant who turns themselves in and asks for asylum is also counted. The headline numbers are loose proxies for what the public conversation is actually trying to discuss.

Got-aways — migrants observed entering the country who are not apprehended — are estimated by CBP using sensors and air surveillance. Estimates ran in the range of 300,000 to 600,000 per year during the 2021–24 period, on top of apprehensions. Visa overstays, as discussed below, are the source of approximately two-thirds of the long-term unauthorized population — and they do not appear in border-crossing numbers at all.

A reader who wants to understand immigration should look at: the foreign-born share (long-run trend), the green-card admissions (legal-immigration policy), the apprehension numbers (border flow), and the unauthorized population (long-run accumulation). No single number tells the whole story.

31.3 The legal architecture

US immigration law is built on a 1965 statute, modified extensively but never replaced. Knowing the architecture is necessary to understand why the politics is what it is.

31.3.1 The Immigration and Nationality Act of 1965

The Immigration and Nationality Act (INA) of 1965 — the Hart–Celler Act — replaced a previous system of national-origins quotas (the 1924 Johnson–Reed Act, modified in 1952) that had heavily favored Northern and Western European immigration. The 1965 act eliminated the national-origins quotas and substituted a system based on family relationships, employment skills, and refugee status. Roughly 70 percent of green cards went to family-sponsored categories, with the rest split among employment, refugee, and a small "diversity" allotment.

The architects of the 1965 act, including Senator Edward Kennedy, expected that the new system would produce immigration patterns roughly similar to the previous European-dominant patterns, just without the explicit national-origins formula. They were wrong. Eliminating the national-origins quotas dramatically changed the source countries of US immigration. By the 2000s, immigration was overwhelmingly from Asia and Latin America rather than Europe. Whether one regards that change as a beneficial diversification or as an unintended demographic transformation is a values question; the empirical fact of the change is not contested.

The 1965 act has been amended repeatedly. The 1986 Immigration Reform and Control Act (IRCA), under President Reagan, granted legal status to roughly 2.7 million unauthorized residents in exchange for stronger employer sanctions and increased border enforcement. (The legalization happened; the enforcement, by most accounts, did not.) The 1990 amendments raised employment-based admissions and created the diversity lottery. The 1996 IIRIRA (Illegal Immigration Reform and Immigrant Responsibility Act) tightened enforcement, expanded grounds for deportation, and made it harder for unauthorized residents to legalize through marriage. The Patriot Act and the 2005 REAL ID Act added security-related modifications. None of these has been a comprehensive replacement of the 1965 framework.

31.3.2 The legal pathways

Within the 1965 framework, there are five main pathways to lawful permanent residence (the green card):

1. Family-sponsored (~70 percent of green cards). Immediate relatives of US citizens — spouses, parents, and minor children — are admitted without a numerical cap. Other family categories (adult children, siblings) are subject to per-country caps and global numerical caps. The result is severe backlogs for some categories from some countries. A US citizen's adult unmarried child from Mexico, the Philippines, or India may face a wait of fifteen to twenty-five years for an available visa. Sibling preferences from the Philippines run beyond two decades. The wait is a function of the cap structure, not of administrative slowness. From a country with low demand (most of Europe, most of Africa), the same category may have no wait.

2. Employment-based (~14 percent of green cards). The five EB categories range from EB-1 (extraordinary ability, multinational executives, outstanding researchers) through EB-5 (investors). EB-2 and EB-3, which include most skilled workers, are also subject to per-country caps and accumulate enormous backlogs for high-demand source countries. An Indian software engineer with employer sponsorship may face an EB-2 wait of fifty years or more under current cap allocations — long enough that it is not a wait but a denial. This is the core reform demand of the country-cap reform advocates: that visas not be capped by birth country.

3. Refugee admissions. The President sets the annual ceiling in consultation with Congress. Trump-1 lowered the ceiling to approximately 15,000; Biden raised it to 125,000. Actual admissions typically run far below the ceiling, because of resettlement-pipeline capacity. Refugee admissions are determined abroad, before the refugee enters the United States.

4. Asylum. Asylum is filed by individuals who are already in the United States (or at a port of entry) and who claim a credible fear of persecution in their home country on protected grounds (race, religion, nationality, political opinion, or particular social group). There is no numerical cap on asylum approvals. The system is administered by USCIS asylum officers and by the immigration courts (housed administratively in the Department of Justice). The immigration-court backlog stood at approximately 3.7 million pending cases as of 2025. Wait times for hearings can run several years. During that wait, asylum applicants are typically released into the interior with work authorization after 180 days.

5. Diversity lottery. Approximately 50,000 visas per year are awarded by random selection to applicants from countries with low recent immigration to the United States. Critics argue this category is an arbitrary use of scarce visa numbers; defenders argue it preserves source-country diversity that would otherwise be crowded out by high-demand countries.

The H-1B specialty-occupation visa, frequently discussed in immigration debates, is a temporary work visa, not a green card. The annual cap is 85,000 new H-1Bs (65,000 plus 20,000 for advanced-degree holders). Demand vastly exceeds supply; in recent years, USCIS has received roughly 750,000 registrations for the lottery, meaning a nine-in-ten chance of rejection per registration. H-1B holders may eventually transition to employment-based green cards, where they encounter the per-country backlogs.

31.3.3 INS to DHS: the 2002 reorganization

Before 2002, immigration enforcement and services were administered by the Immigration and Naturalization Service (INS), housed in the Department of Justice. After September 11, 2001, the Bush administration and Congress reorganized federal immigration administration. The Homeland Security Act of 2002 abolished INS and split its functions into three component agencies inside the new Department of Homeland Security:

• United States Citizenship and Immigration Services (USCIS) — adjudicates applications for visas, green cards, naturalization, asylum, and other benefits. The "services" side of immigration.
• Immigration and Customs Enforcement (ICE) — interior enforcement, including detention, deportation, work-site enforcement, and Homeland Security investigations.
• Customs and Border Protection (CBP) — border enforcement, ports of entry, and the Border Patrol.

The reorganization was intended to fix coordination problems revealed by the 9/11 attacks (specifically, that several hijackers had overstayed visas without consequence). It also separated benefits adjudication from enforcement, which has had downstream consequences: a person applying for an immigration benefit (USCIS) is in a different agency from the one that might deport them (ICE), and information-sharing across these agencies is fraught.

31.3.4 The immigration courts

A controversial piece of the architecture is the Executive Office for Immigration Review (EOIR), which houses the immigration courts. EOIR is in the Department of Justice, not in the judicial branch. Immigration judges are DOJ employees who serve at the pleasure of the Attorney General. The Board of Immigration Appeals, also in DOJ, hears appeals.

This is unusual. It is sometimes called an Article I tribunal, by analogy to bankruptcy and tax courts, but it is not even that — it is an executive-branch administrative system. The Attorney General has authority to certify cases to themselves and overrule immigration-court decisions. (Attorneys General have used this power across administrations to alter substantive immigration law.)

Reform proposals for an Article III immigration court — making the immigration judges life-tenured federal judges — have been put forward by both liberal and some conservative legal analysts, but have made little progress. The current system is faster, cheaper, and more responsive to executive priorities than an Article III court would be — which is precisely why each administration has been reluctant to change it, and why critics worry that the lack of judicial independence compromises the rule of law.

31.3.5 Asylum and the credible-fear standard

Asylum law as practiced today derives largely from the Refugee Act of 1980, which incorporated the international refugee definition into US law. To establish asylum, an applicant must demonstrate a "well-founded fear of persecution" on one of five protected grounds. Most asylum cases come down to one of two: political opinion or membership in a particular social group.

The "particular social group" category has produced a great deal of litigation. Courts have, at various times, recognized social groups including women fleeing female genital mutilation, gay men in some countries, victims of gang violence (in some readings), and victims of domestic violence (also in some readings). Different administrations have taken different positions on whether particular social-group claims should be read broadly or narrowly. Trump-1 narrowed the reading; Biden broadened it; Trump-2 has narrowed it again, with extensive use of the AG's certification power.

The credible-fear interview is the gateway to the asylum process. A migrant who claims asylum at the border or after apprehension is interviewed by an asylum officer to determine whether they have a "credible fear" of persecution that warrants further proceedings. The threshold is intentionally low. In recent years, approximately 70 to 85 percent of credible-fear interviews resulted in a positive finding, allowing the applicant to enter the asylum system and remain in the country pending a hearing. Final asylum-grant rates after full hearings are much lower — historically around 20 to 30 percent, though varying by source country and individual judge.

The empirical question debated within the asylum system is whether the gap between credible-fear approvals and final asylum grants reflects (a) the system filtering frivolous claims through a long process, (b) merit-claim filtering (migrants who would qualify for asylum but lose for procedural or capacity reasons), or (c) abuse of the system as an entry mechanism. All three explanations have empirical support for different cases.

31.4 The undocumented population

About 11 to 13 million people in the United States are unauthorized residents. The statistical profile of this population is consistent across multiple data sources and is necessary for any serious policy analysis.

31.4.1 Length of residence

The undocumented population is predominantly long-term resident. Median time in the United States is approximately 15 years. Roughly two-thirds have been in the country for more than a decade. The image of the undocumented population as recent arrivals is empirically wrong as applied to the existing stock — it is more accurately a description of recent border crossers, who are a smaller fraction.

31.4.2 Country of origin

About 85 percent are from the Americas. Mexico remains the largest single country of origin, though its share has fallen sharply over the past two decades — from roughly 60 percent of the unauthorized population in 2007 to roughly 40 percent in 2024. Central American countries (Guatemala, Honduras, El Salvador) and increasingly South American countries (Venezuela in particular) account for a growing share. Asian countries — China, India, and the Philippines — together represent approximately 10 percent.

31.4.3 Mode of entry

Approximately one-third of the unauthorized population entered the country without authorization (typically by crossing the southwestern border). Approximately two-thirds entered legally — on a tourist visa, student visa, or work visa — and then overstayed. This is one of the most important and most underdiscussed facts about the unauthorized population. Most policy attention focuses on border enforcement, but most of the long-term unauthorized population is the result of visa overstays, which border enforcement does not reach.

31.4.4 Mixed-status families

About half of unauthorized residents live in mixed-status families — households containing at least one US citizen, typically a US-citizen child born in the country (the birthright-citizen child of unauthorized parents). There are approximately 4 to 5 million US-citizen children with at least one unauthorized parent.

This fact is at the heart of why purely abstract enforcement policy is so difficult to operationalize. Removing an unauthorized parent of a US-citizen child does not remove the family from the country. It either separates the family or, if the citizen child accompanies the parent, removes a US citizen from their country of citizenship — sometimes a young child who has never lived elsewhere. The 2018 family-separation policy exposed how quickly policy abstractions translate into concrete moral horror; renewed concerns under Trump-2's expanded interior enforcement focus on the same dynamic.

31.4.5 Economic profile

The undocumented population is concentrated in particular industries: agriculture (estimates suggest roughly half of crop workers are unauthorized), construction (15 to 25 percent of certain trades), hospitality and food service, and home-care services. The Pew Research Center estimates approximately 8 million unauthorized workers in the US labor force, or roughly 5 percent of total US employment.

Federal income tax payments by unauthorized workers are estimated at approximately $13 billion annually**, paid through wage withholding using Individual Taxpayer Identification Numbers (ITINs). State and local taxes total an additional **$10 billion. Most of this is paid against benefits the workers will never collect — Social Security taxes are paid into accounts that cannot be claimed without legal status.

Estimated lifetime economic output of the current unauthorized population, on standard input-output methodologies, runs in the range of $1.5 to $2 trillion in present value, with substantial variation by methodology. This number is contested, and reasonable analysts disagree about the right way to compute it (gross product vs. net product, fiscal vs. economic, inclusion of crowd-out effects on native-born workers). The fact that the number is large — that the unauthorized population is economically substantial — is not contested.

31.5 The empirical questions: what the research actually shows

A small number of empirical questions appear repeatedly in immigration debates. On most of them, the academic literature has converged. Stating the convergence is not partisan; ignoring it is.

31.5.1 Wage effects on US workers

Does immigration depress wages of native-born workers? The robust answer from the academic literature: small effects on aggregate wages, modest negative effects for similar-skill native workers (especially less-educated native workers), and positive effects for complementary workers and consumers.

The two principal disputants in this literature are George Borjas (Harvard) and David Card (Berkeley, Nobel laureate 2021), with Giovanni Peri (UC Davis) representing a synthesis position. Borjas's work, using the "national skill-cell" approach, finds that immigration has reduced wages of native-born workers without high-school diplomas by roughly 5 to 8 percent over the past several decades. Card's work, using the "spatial correlation" approach with the 1980 Mariel boatlift as a key natural experiment, finds smaller and statistically less robust effects, sometimes near zero. Peri and others have found that high-skill immigration produces positive complementarity effects (immigrants raise productivity of native-born workers in the same industry).

The honest summary: immigration produces winners and losers in the US labor market. The winners are likely to be native-born workers in occupations that complement immigrant labor, consumers (because some prices are lower), and immigrant workers themselves. The losers are most likely to be native-born workers who compete most directly with immigrant labor — which, given the educational distribution of immigration, often means less-educated native-born workers. The progressive concern about wage effects on low-income workers and the conservative restrictionist concern about wage effects on the working class are partially the same concern, with partial empirical support.

31.5.2 Crime rates

Do immigrants — documented or undocumented — commit crimes at higher rates than native-born Americans? No. Across multiple methodologies, multiple data sources, and multiple time periods, the consensus finding is that immigrants commit crimes at lower rates than the native-born. This finding holds for both documented and undocumented immigrants, holds across most categories of crime, and has been replicated extensively (Sampson 2008; Adelman and Reid 2017; Light, He, and Robey 2020 on Texas administrative data; numerous others).

The data are subject to one important caveat: the comparison is to native-born Americans of similar age and demographics, since immigrants skew younger and male and crime is concentrated among young men. Even adjusting for age and sex, the immigrant rates are lower.

This empirical finding does not by itself answer the policy question. One can hold that immigrants commit crimes at lower rates than native-born Americans and still hold that admitting any specific individual who later commits a crime represents a policy failure. (Restrictionist advocates often make exactly this point: the relevant comparison is not between immigrants and native-born, since native-born Americans cannot be excluded from the country, but between admitting an immigrant and not admitting them.) The empirical claim is: the population effect of immigration on the US crime rate is, if anything, slightly negative. State the data; do not editorialize.

31.5.3 Public-services use and fiscal effects

Are immigrants net fiscal contributors? The empirical finding, replicated across CBO analyses and academic studies, is that immigrants on average pay more in taxes than they receive in benefits over their lifetimes, but that the fiscal pattern varies by characteristic (high-skill immigrants are strongly net positive; less-skilled immigrants are weakly net positive or close to neutral; lifetime fiscal calculations depend on whether second-generation effects are included).

The state-and-local pattern is more mixed. Federal tax revenues (Social Security, Medicare, federal income tax) accrue to the federal government; education costs and emergency medical costs (Medicaid emergency coverage, K–12 public-school costs for citizen children of immigrants) often fall on states and localities. The fiscal benefits of immigration are concentrated at the federal level; some of the costs are concentrated locally. This is one structural reason why state and local governments in high-immigration areas often advocate for federal compensation.

The CBO's 2024 analysis of the 2021–24 immigration surge estimated a net positive federal fiscal effect over the 2024–34 budget window of approximately $1 trillion (revenues exceeding outlays for the marginal new arrivals). This estimate is contested, in part because it relies on assumptions about labor-market participation and wage distribution. But the direction of the estimate — that the fiscal effect of working-age immigrants is positive at the federal level — is consistent with the broader literature.

31.5.3a A note on study quality and contestation

Before continuing, a brief note on how the empirical literature should be read. Several immigration questions have been the subject of decades of careful research, multiple replication studies, and high-profile methodological exchanges between leading economists. The findings summarized in this section reflect the current consensus among researchers who have published in top peer-reviewed journals. The findings are not unanimous; the standard practice in honest summaries is to report the central tendency and to flag the dispersion. Where the literature is genuinely contested — as it is on the precise magnitude of wage effects on the least-educated native-born workers — this chapter reports the contested zone rather than picking a side. Where the literature has converged — as it has on crime rates, on aggregate fiscal effects, and on cross-generational assimilation — this chapter reports the consensus. Readers who suspect the consensus has been mis-stated are invited to consult the National Academies reports, the Borjas–Card–Peri exchanges, and the underlying primary literature. The empirical claims in this chapter survive that scrutiny.

31.5.4 Cultural assimilation

Do immigrants assimilate? The empirical literature, including the National Academies of Sciences 2015 consensus report The Integration of Immigrants into American Society, finds strong assimilation patterns across generations on virtually every measurable dimension: English-language fluency, intermarriage, civic identification, residential integration, occupational integration. The pace varies by group (some groups assimilate faster, some slower), but the overall pattern is consistent across waves of immigration. By the third generation, descendants of immigrants are largely indistinguishable from the native-born majority on most cultural-integration measures.

This finding has been used by both sides. Pro-immigration advocates cite it to refute claims that immigration is producing unassimilated enclaves. Pro-restriction advocates point out that the rate of assimilation is observably slower for some groups than for the European immigrants of a century ago — particularly in residential integration and language retention — and argue that current rates of admission may exceed the country's assimilation capacity. Neither claim is incompatible with the underlying empirical finding. The mechanisms of assimilation (intermarriage, geographic mixing, education in shared institutions) operate now as they did then.

The honest summary: assimilation is happening. Whether it is happening fast enough, in a country whose institutional and cultural cohesion is also undergoing other strains, is a values judgment.

31.6 The policy debates

With the empirical landscape mapped, the policy debates can be examined in their actual structure rather than as abstract pro-and-con.

31.6.1 Border enforcement

The border-enforcement debate has a number of distinct components.

The border wall. Approximately 700 miles of physical barrier exist along the 1,950-mile southwestern border, in various forms — pedestrian fencing, vehicle barriers, concrete walls in some sectors. Construction continued under Bush-43, Obama, Trump-1 (~450 miles of new or replacement barrier), Biden (continued construction in some sectors, halted in others), and Trump-2 (renewed construction). The empirical effectiveness is mixed. Barriers in heavily trafficked urban sectors (San Diego, El Paso, Yuma) appear to deter crossings effectively; barriers in remote desert sectors are sometimes circumvented by tunneling, climbing, or crossing in unwalled stretches, with crossings redirected rather than prevented. The effectiveness debate is empirically narrow but politically symbolic.

Internal enforcement. Interior enforcement, conducted by ICE, has been the subject of intense policy variation across administrations. Workplace enforcement through Form I-9 audits and work-site raids was substantial under Bush-43, declined under Obama, rose under Trump-1, declined under Biden, and is rising sharply under Trump-2. E-Verify — the federal employment-eligibility verification system — has been adopted as mandatory by several states and is used by many federal contractors but is not federally mandated for all employers. Mandatory nationwide E-Verify is one of the recurring proposals in restrictionist reform packages and one of the principal objections of business groups in agriculture and hospitality.

Sanctuary jurisdictions — cities, counties, and states that limit cooperation between local law enforcement and federal immigration enforcement — exist in roughly 200 jurisdictions across the country. The legal authority for sanctuary policies derives from the anti-commandeering doctrine (the Tenth Amendment principle that the federal government cannot require state and local officials to enforce federal law) recognized in Printz v. United States (1997). Sanctuary policies vary widely. Some prohibit local officials from inquiring about immigration status; some prohibit honoring ICE detainer requests for individuals in local custody; some allow cooperation only when a judicial warrant is presented. The Trump-2 administration has aggressively litigated against sanctuary policies; lower-court rulings have largely upheld the constitutional principle that the federal government cannot commandeer local enforcement, while finding some specific sanctuary provisions preempted on narrower grounds.

Asylum processing. The asylum system at the southwestern border was the principal pressure point of the 2021–24 surge and remains the principal pressure point of current enforcement debate. The Trump-1 Migrant Protection Protocols (MPP, "Remain in Mexico") required asylum applicants to await hearings in Mexico. The program was suspended by Biden, restored by court order in 2021, and ended after a Supreme Court ruling allowed termination. Title 42, the public-health expulsion authority used during COVID-19, allowed CBP to expel migrants without immigration-court proceedings; it was used by Trump-1 and continued by Biden until May 2023, when it was ended along with the COVID public-health emergency. Expedited removal — which allows DHS to remove certain migrants without immigration-court hearings — has been available for decades but was historically used only for migrants apprehended within 100 miles of the border within 14 days of entry. The Trump-2 administration has expanded expedited removal to nationwide application for migrants who cannot prove two years of continuous residence.

Mass deportation. The Trump-2 administration has stated a goal of removing "millions" of unauthorized residents. Operationally, the constraints are substantial. ICE has approximately 6,000 deportation officers nationwide. Detention capacity, even after expansion, runs in the low hundreds of thousands. Deportation requires either a removal order from an immigration judge or, for certain categories, expedited removal. As of 2025 data, ICE removals are running at substantially elevated levels relative to the Biden administration but are not, in a numerical sense, the "mass" deportation of the rhetoric. The capacity to remove ten million people in any reasonable timeframe does not exist; whether expansion of capacity should be a policy goal is itself contested.

31.6.2 Pathway debates

Whether to provide a legal pathway for some or all of the long-term unauthorized population is the second principal axis of the policy debate.

DACA — the Deferred Action for Childhood Arrivals program, established by the Obama administration in June 2012 — is examined in detail in the case study at the end of this chapter. As of 2025, approximately 580,000 active DACA recipients have work authorization and protection from deportation but no pathway to permanent legal status. The program has been in extended legal jeopardy. A series of Fifth Circuit rulings has held the original program unlawfully implemented; the Supreme Court has ruled on procedural questions but has not fully resolved DACA's substantive legal status. Any legislative codification has failed to pass.

Comprehensive immigration reform has been attempted repeatedly since 2001. The major attempts are:

• 2001–2002. The Bush–Fox negotiations, derailed by 9/11.
• 2006. The Senate passed S. 2611 (Hagel–Martinez), a comprehensive bill including legalization for long-term unauthorized residents and an expanded guest-worker program. The House refused to take it up.
• 2007. A bipartisan Bush–Kennedy–Kyl–McCain proposal was killed in Senate procedural votes. Immigration restrictionists in both parties opposed it.
• 2013. The "Gang of Eight" bill (S. 744) passed the Senate 68–32 with significant Republican support. It included a 13-year pathway to citizenship for the unauthorized, a doubling of border-enforcement funding, expansion of legal-immigration channels, and an E-Verify mandate. Speaker John Boehner declined to bring it to the House floor, where it would not have passed.
• 2024. The Lankford–Sinema–Murphy negotiation, examined as the second case study at the end of this chapter, would have provided enforcement authority and limited reform. It was rejected by the Republican Senate majority after Trump's vocal opposition.

The pattern is consistent across two decades and four administrations. Bipartisan reform proposals can be negotiated, but they cannot be passed. The political-economy explanation, which is becoming increasingly uncontested across observers from different ideological starting points, is that the issue benefits one or both parties electorally in ways that make non-passage more politically valuable than passage. Each party's activist base opposes the bargain its negotiators are willing to make. Each party can plausibly campaign against the other's failure to pass the bill that, were it to pass, would deprive both parties of a major electoral motivator.

This is a structural problem, not a problem of personalities. Until the underlying electoral incentive changes, the comprehensive-reform path will continue to be blocked.

31.6.3 Legal-immigration reform

Reform of legal immigration, separate from enforcement and legalization, is its own policy area.

Country-cap reform would eliminate the per-country numerical limits on employment-based green cards. Under current law, no single country can receive more than seven percent of total green cards in any category. For high-demand source countries (India, China, Mexico, the Philippines), the cap produces decade-long backlogs. The EAGLE Act and similar proposals have come close to passage repeatedly, with bipartisan support, but have died in conference or in the House. Country-cap reform is broadly supported by employment-based-immigration advocates and by Indian-American advocacy groups; opposition comes from advocates who want to reduce total employment-based immigration rather than reallocate it.

Refugee and asylum reform has come from both directions. Advocates of expansion want to raise refugee ceilings, increase processing capacity, and reduce the immigration-court backlog. Advocates of restriction want to tighten the credible-fear standard, expand expedited removal, narrow the "particular social group" category, and reduce asylum approvals. Some specific reforms — speeding up immigration-court processing, increasing the number of immigration judges — have bipartisan support but get caught in larger comprehensive negotiations.

H-1B reform has been proposed by advocates of expansion (raise the cap, exempt advanced-degree holders, allow easier transitions to permanent status) and by advocates of restriction (raise wage floors, reform the lottery, prevent body-shop abuse, prefer US workers). The Trump-1 RAISE Act would have shifted the family-versus-employment balance sharply toward employment by capping family categories and expanding skilled employment categories. Biden's reform proposals would have raised both family and employment categories. Both stalled.

The general pattern in legal-immigration reform is that piecemeal proposals occasionally make progress but tend to get folded into comprehensive packages, where they fail along with everything else.

31.7 The positions, steel-manned

Three major position-clusters define the contemporary debate. Each contains internal variation. Each, in its strongest form, is a coherent position held by serious people. The book's task here is to present each at its strongest.

31.7.1 Restrictionist / pro-borders positions

The restrictionist position, in its strongest version, rests on several connected concerns:

Rule of law. A country that admits residents only through legal processes has a different character than one that tolerates large-scale unauthorized residence. The restrictionist case is that legalizing unauthorized residence — whether through DACA, comprehensive reform, or de-facto non-enforcement — undermines the meaning of legal status and creates incentives for further unauthorized migration. The rule-of-law concern is not reducible to xenophobia, and the strongest restrictionist writers explicitly distinguish between immigration levels (which are debatable) and immigration legality (which is not).

Sovereignty and border control. A nation-state that does not control who enters has, in the restrictionist view, surrendered a defining feature of sovereignty. The argument is structural rather than primarily empirical: even if every individual unauthorized entrant were beneficial in expected value, the loss of state capacity to control entry has its own costs.

Labor-market effects on existing low-income workers. The empirical literature, as discussed above, supports modest negative effects on wages for similar-skill native-born workers. The restrictionist case is that elite proponents of expanded immigration tend not to compete with immigrant labor and so are insufficiently attentive to these effects. George Borjas has been the most influential academic exponent. Reihan Salam in Melting Pot or Civil War? makes a related case from a more centrist starting point: that current immigration policy serves the interests of upper-middle-class consumers (cheaper services) and educated immigrants (legal pathways) at the expense of working-class native-born and working-class immigrants alike.

Public-services capacity. Local jurisdictions with high concentrations of immigrants — particularly low-skilled immigrants — bear costs (school capacity, emergency medical services, housing pressure) that federal-fiscal analyses do not capture. The restrictionist position is that the country can absorb high levels of immigration in some respects but not in all of them at once, and that the absorptive capacity of communities has been ignored.

Cultural cohesion. The strongest version of the cultural-cohesion concern, advanced by writers like Salam, Yuval Levin, and (with sharper edges) Mark Krikorian at the Center for Immigration Studies, is that rapid demographic change strains the institutional and informal mechanisms by which a country produces shared civic identity. The argument is not "immigrants do not assimilate" — the data show they do — but rather "high rates of admission can outpace the slow work of assimilation and produce visible community-level dislocation." This concern is not coextensive with nativism; some who hold it would favor higher legal immigration of certain kinds and lower of others.

The strongest restrictionist writers do not call for closed borders. They typically call for lower total numbers, stricter enforcement of legality, stronger employer accountability, and tighter asylum standards. They typically support some legalization of long-term residents in exchange for strong enforcement guarantees — what is sometimes called the "enforcement-first" position.

31.7.2 Pro-expansion positions

The pro-expansion position, also in its strongest version, rests on several connected concerns.

Humanitarian considerations. People migrate because their lives in their home countries are difficult, sometimes desperate. The strongest humanitarian case, advanced across a range of religious and secular traditions, is that the moral claim of a person fleeing persecution or extreme poverty is weighty, and the burden of justification falls on those who would refuse them.

Economic growth. The economic case for higher immigration, advanced by the Cato Institute, by libertarian economists including Bryan Caplan (Open Borders, 2019), and by immigration economists like Giovanni Peri and Alex Nowrasteh, is that immigration produces positive aggregate effects on growth, productivity, innovation, and entrepreneurship. The growth case is largely uncontested at the aggregate level; what is contested is the distributional pattern.

Demographic crisis. Most developed economies, including the United States, face long-run demographic challenges from below-replacement fertility. The pro-expansion case is that immigration is the only realistic mechanism for offsetting population aging in the absence of fertility recovery. Without continued immigration, US population would peak and decline by mid-century, with substantial fiscal and economic consequences.

Innovation and entrepreneurship. Immigrants found about a quarter of US new businesses and a substantially larger share of high-growth startups (Kerr and Lincoln 2010; Azoulay et al. 2022). High-skill immigration has been particularly important in technology and bioscience, where Indian, Chinese, and other foreign-born researchers are heavily represented in patenting, faculty positions, and startup formation.

Rule-of-law via sensible paths. The strongest pro-expansion writers do not advocate for unauthorized residence; they advocate for expanding legal channels so that what is currently unauthorized would be legal. The Caplan position, in its purest form, is that the moral and economic case for the status quo's restrictions is weaker than its defenders suppose, and that broader legal pathways would address the rule-of-law concern by removing the incentive for unauthorized residence in the first place. More moderate pro-expansion writers (the editorial tradition of the Wall Street Journal on legal immigration, much of Cato's work) favor large expansions of legal-immigration channels paired with stronger enforcement of remaining restrictions.

The strongest pro-expansion writers typically call for higher legal admissions, broader pathways for the unauthorized, expanded refugee admissions, and employment-based reform that prioritizes admitting high-skill workers without rationing by birth country. They are typically skeptical of large-scale interior enforcement and of arguments that immigration produces large net costs.

31.7.3 Comprehensive-reform / centrist positions

A third position cluster, less ideologically pure than the first two, has been the explicit goal of every major bipartisan reform attempt of the past two decades: a bargain that exchanges meaningful enforcement for meaningful legalization.

The standard bargain has these elements:

• Pathway to legal status for long-term unauthorized residents (typically with conditions: payment of back taxes, no criminal record, English-language requirements, a multi-year provisional period before full status).
• Strong enforcement provisions: mandatory E-Verify, increased border-enforcement resources, expanded interior enforcement against new arrivals, asylum-processing reform.
• Reform of legal-immigration channels: increased numerical caps, country-cap reform, modernization of family categories, expansion of skill-based admissions.
• Sometimes a guest-worker program for agricultural and other sectors with chronic labor shortages.

This bargain has been negotiated repeatedly. It has not passed since 1986. The political-economy diagnosis is that it cannot pass because the activist bases of both parties oppose the trade-off: progressive activists object to the enforcement provisions; conservative activists object to the legalization provisions. Centrist positions retain substantial public support, but the active veto players in each party's coalition can block the bargain.

The comprehensive-reform position, as a position, is held by a thinning band of legislators (the senators who negotiated the various bills), think-tank scholars (Cristina Rodríguez at Yale; the Niskanen Center; some at the American Enterprise Institute and Brookings), and editorial-page writers across the spectrum. It is distinctively unattached to the activist base of either party, which is one reason for its persistent failure.

31.7.4 What charity to all sides actually requires

It is tempting, given the heat of the debate, to frame each side as motivated by a single bad impulse: restrictionists by xenophobia, pro-expansion advocates by indifference to citizens, comprehensive reformers by careerist centrism. This framing is wrong on every count.

Each position, at its strongest, addresses real concerns. The restrictionist concern with rule of law and labor-market effects on low-income native-born workers is an argument from values progressives also hold. The pro-expansion concern with humanitarianism, economic growth, and demographic crisis is an argument from values conservatives also hold. The comprehensive-reform concern with operational policy that actually closes the gap between law and reality is a structural recognition that the alternatives (perpetual stalemate, executive workarounds, judicial intervention) are worse for the rule of law than a workable bargain would be.

The genuine disagreement is how to weigh these concerns against each other when they conflict. That is a values question, not a fact question. A reader's answer depends on their priors about what kind of country they want, how much weight they give to which kinds of harm, and how they trade off humanitarian considerations against considerations of order. Reasonable people can answer differently. They will.

31.8 The deportation debate

The most operationally consequential immigration debate of the current moment is over deportation policy.

31.8.1 Mass deportation as a policy

The Trump-2 administration has stated a goal of deporting "millions" of unauthorized residents and has made the largest interior-enforcement push since the Eisenhower-era Operation Wetback. Whether the operational reality will match the rhetorical scale depends on several factors.

Capacity. ICE has approximately 6,000 deportation officers and approximately 41,500 detention beds (as of mid-2025, after expansion). Removing ten million people would require multiplying the operational capacity of ICE by approximately ten — a budgetary and personnel undertaking that, even with full congressional cooperation, would take years to build. The current administration has expanded military and federal-law-enforcement support for immigration enforcement; the constitutional and statutory limits on such use are themselves contested in ongoing litigation.

Legal process. Most removals require a removal order from an immigration judge. The immigration-court backlog is approximately 3.7 million pending cases. Expedited removal allows DHS to bypass immigration courts for certain categories, but does not apply universally. The administration has attempted to expand expedited removal to nationwide application; lower courts have upheld parts of the expansion and blocked others.

Practical economics. Several industries are heavily dependent on unauthorized immigrant labor. Agriculture, construction, hospitality, and home care employ millions of unauthorized workers. Large-scale removal of this labor force would produce substantial price increases and labor shortages. Industries with heavy unauthorized employment have, in past enforcement pushes (Georgia 2011, for example), produced documented harvest losses and price spikes. The administration's stated response is that native-born workers will fill the gap; the empirical question of how quickly and at what wages remains open.

Public opinion. Polling on deportation is bimodal. Substantial majorities support deporting recent unauthorized arrivals and unauthorized residents with criminal records. Substantial majorities oppose deporting long-term residents who have been in the country for years, particularly those with US-citizen family members. The political viability of mass deportation depends on which subgroup of the unauthorized population is the principal target — and the operational reality of large-scale enforcement is that targeting the more sympathetic subgroup is unavoidable.

31.8.2 Family separation

The 2018 zero-tolerance policy, which separated approximately 5,500 children from parents at the southwestern border, was the moment at which the operational reality of an enforcement-first policy collided with concrete moral horror. The policy was reversed under public and congressional pressure. The reverberations remain politically significant.

Renewed concerns under Trump-2 focus on the same dynamic: large-scale interior enforcement against the unauthorized population, which is heavily integrated into mixed-status families, will produce family separations. There is no enforcement strategy that combines large-scale removal with no family separation. The values question is whether the enforcement is worth the separation; the operational question is at what scale the country is willing to absorb the visible human consequences.

31.8.3 Constitutional limits

Several constitutional questions are being litigated in the current enforcement push.

Birthright citizenship. The Fourteenth Amendment's Citizenship Clause provides: "All persons born or naturalized in the United States, and subject to the jurisdiction thereof, are citizens of the United States and of the State wherein they reside." The doctrinal interpretation, established by United States v. Wong Kim Ark (1898), is that the clause confers citizenship on virtually all children born on US soil, including children of unauthorized parents. The Trump-2 executive order purporting to deny birthright citizenship to children of unauthorized parents was challenged in multiple courts. As of mid-2025, three district courts and two federal appellate courts had blocked enforcement of the order, holding it inconsistent with Wong Kim Ark and the plain text of the Fourteenth Amendment. The Supreme Court has not yet taken a case on the merits. The standard view among constitutional scholars across the ideological spectrum, including originalist scholars sympathetic to immigration restriction, is that the Wong Kim Ark reading is correct and that the executive order will not survive Supreme Court review.

Due process for removals. The constitutional question of what process is due to individuals facing deportation is doctrinally complex. Due process applies to individuals physically present in the country regardless of status (Mathews v. Diaz, 1976; Zadvydas v. Davis, 2001). The scope of process — particularly the question of whether expanded expedited removal can be applied to long-term residents who could not realistically prove residency — is in active litigation.

Use of military forces. The Posse Comitatus Act limits the use of federal military forces in domestic law enforcement. The administration's deployment of National Guard units, federalized in some cases, to support immigration enforcement has produced litigation over the boundaries of permissible use. Lower-court rulings have been mixed; the Supreme Court has not yet weighed in.

The constitutional architecture is being tested in real time. As of the writing of this chapter, the doctrinal lines remain largely as they have been for decades; the operational pressure is unprecedented in postwar American practice.

31.9 The political economy of stalemate

Pull together the threads. Why has comprehensive immigration reform failed so consistently for so long?

The answer is not that the technical challenge is intractable. Reform proposals have been drafted, scored, negotiated, and brought to the verge of passage repeatedly. The technical work has been done.

The answer is not that the public is unwilling. Polling consistently shows majority support for both stronger enforcement and a pathway for long-term residents. The center of public opinion is roughly where comprehensive reform proposals have always been.

The answer, instead, is structural. The activist base of each party — the voters and donors who turn out in primaries, who fund campaigns, who provide ground-game energy — opposes the bargain that the centrist legislators of each party are willing to make. Progressive activists oppose enforcement provisions; conservative activists oppose legalization provisions. Each party's activist base has effective veto power within its own party's congressional caucus. The result is that proposals that would pass on a free-vote basis cannot survive primary politics.

The deeper reason is that the issue benefits each party electorally more in unresolved form than in resolved form. A Democratic Party that legalized the long-term unauthorized population would lose a galvanizing motivator for Latino and immigrant voters. A Republican Party that secured the border to general satisfaction would lose a galvanizing motivator for working-class and culturally-conservative voters. Whether this electoral asymmetry is symmetric in magnitude is contested; that some asymmetric incentive operates is hard to deny.

This is not a thesis about bad faith. It is a thesis about how electoral incentives shape what gets passed. Members of Congress who personally favor comprehensive reform face primary challenges from those who do not. Presidents who personally favor comprehensive reform face activist coalitions that punish concessions. The status quo is sticky because the politics rewards stickiness.

The pathway out of this structural trap is not obvious. One possibility is that a generational shift in the activist base of one or both parties might rebalance the incentive. Another is that the operational consequences of unresolved status — visible to voters in ways that abstract policy debate is not — might eventually drive a settlement. A third is that judicial intervention will force the legislative branch's hand. None of these is certain, and each operates on a long time horizon.

Until then, the cycle of failed bargains will, on the evidence of the past two decades, continue.

It is worth pausing on one further mechanism that has reinforced the stalemate: the emergence of immigration as a media issue in which visual coverage drives public salience in ways that the textual policy debate cannot replicate. Footage of border crossings, of detention facilities, of family separations, and of arriving migrants has a directness that comprehensive-reform proposals do not. The visual coverage tends to be partisan in selection — outlets that emphasize border disorder cover crossings; outlets that emphasize humanitarian considerations cover detention conditions and family separations — and the asymmetric attention reinforces partisan information environments in which the same underlying events appear differently. The 2018 family-separation coverage, the 2021–22 Del Rio bridge images of Haitian migrants, the 2023 Eagle Pass scenes during the Texas–federal standoff, and the 2025 expedited-removal operations have all functioned as moments of intense salience that reshaped, briefly, the political incentives — and that, equally briefly, faded.

What makes immigration distinctive among policy areas is that the policy is continuous: every day, several thousand people enter the country, several thousand are processed, several thousand are removed, several thousand transition between statuses. The system is operating at large scale at all times. Comprehensive reform, when it finally arrives, will not be a discrete moment but a recalibration of an ongoing operation. The political class has tended to discuss immigration as if it were a one-time decision; the operational reality is that it is a permanent system whose settings are being adjusted, mostly without legislative authorization, every administration.

31.10 Your district and the immigration question

Every congressional district in the country has an immigration story. Some districts have foreign-born populations above 30 percent (parts of California, New York, Florida, Texas); some have foreign-born populations below 3 percent (parts of West Virginia, Mississippi, Wyoming). The local economic structure — agricultural employment, manufacturing dependence, service-sector employment, professional employment — shapes the immigrant footprint.

Look up your district's foreign-born share and country-of-origin distribution at the Census Bureau's American Community Survey. Look up your representative's voting record on immigration legislation at GovTrack. Look up the major employers in your district and the sectors of the local economy that depend on immigrant labor. The political position of your representative will, in most cases, reflect both the demographics of the district and the activist base of the party — sometimes pulling in the same direction, sometimes pulling apart.

The immigration debate is national in its rhetoric and local in its operational reality. Wherever you live, somewhere within driving distance is a worksite, a school, a hospital, a courthouse where immigration policy is being lived out. The next time the policy debate cycles into another comprehensive-reform attempt, the question of what your district would gain or lose from passage will be the question that, in aggregate across 435 districts, determines the outcome.

31.11 What this chapter has argued

The empirical landscape of US immigration is well documented. The country is more foreign-born than at any time since 1910. The undocumented population is roughly stable, predominantly long-term, mostly from the Americas, and substantially integrated into mixed-status families. Immigrants commit crimes at lower rates than the native-born; immigrants are net fiscal contributors at the federal level; immigration produces small aggregate wage effects with modest negative effects for similar-skill native-born workers; immigrant assimilation across generations is empirically robust.

The legal architecture is a 1965 statute modified extensively but not replaced, administered through three DHS component agencies, with adjudication housed in the Department of Justice. The legal pathways — family-sponsored, employment-based, refugee, asylum, diversity — are subject to caps and per-country limits that produce decade-long backlogs for high-demand categories.

The policy debates fall into three clusters: border enforcement, pathway and legalization, and legal-immigration reform. Each contains genuine empirical and normative disagreement. Each contains positions held in their strongest forms by serious people for non-pathological reasons.

The comprehensive-reform path has failed for reasons that are structural, not personal. The political economy of the issue rewards non-passage. Until the underlying incentive changes, the cycle of failed bargains will continue.

The current moment, with the most aggressive interior-enforcement push since the 1950s, is testing the constitutional architecture in real time. Birthright citizenship, due-process limits on removal, and the use of federal military forces in domestic enforcement are all in active litigation. The doctrinal lines are largely intact. The operational scale is unprecedented.

Read the chapter understanding that immigration is the policy area where almost everyone has a position, and where the gap between rhetorical positions and operational realities is at its widest. The country will continue to have an immigration policy whether it passes a new one or not. What the policy looks like will, on present evidence, be determined less by the comprehensive bargains that fail to pass than by the cumulative weight of executive actions, court rulings, and operational decisions that fill the space the legislature has vacated.

Whether that is a feature or a bug, given the institutional design of American government, depends on what one wants. The honest answer is that neither side wholly likes what is happening, and that the country has not yet found the path to anything different.

A final note on tone. This is the chapter in which the gap between rhetorical heat and analytical clarity is widest. There is a temptation, in writing about immigration, to lean toward the rhetoric of crisis or the rhetoric of welcome, depending on disposition. Both rhetorical modes obscure rather than illuminate. The crisis rhetoric makes the situation sound recent when much of it is decades-old; the welcome rhetoric makes the policy sound simple when it is operationally complex. The honest analyst keeps the empirical numbers in view, recognizes that the most heated rhetorical positions on each side are responses to real concerns held by other Americans, and refuses the satisfaction of declaring the question settled. It is not settled. The country has not yet decided what kind of country it wants to be on this question, and the period during which it will be deciding is likely to be longer than the political cycle is comfortable with. The chapter ends, accordingly, not with a conclusion but with a recommendation: read the empirical literature, examine your local immigration footprint, watch the legislative attempts as they cycle through, and form your own view from a position that takes seriously the strongest arguments on every side. That is the discipline this material requires.