3 Immigration, Immigrant Integration and Welfare Policy in the United States

By Madeleine Sumption and Michael Fix, Migration Policy Institute

Introduction

The United States provides a narrower safety net than most other advanced industrialized nations, and welfare benefits for noncitizens are even more limited. The approach to immigrant integration is for the most part laissez-faire and targeted spending on immigrants is minimal, although in many cases immigrants are eligible for a wide range of non-targeted public services and benefits in fields such as education and child care. Some state governments make up for some of these gaps with their own state-level programs, but on the whole immigrants in the United States receive very little in the way of public assistance.

Nonetheless, immigrant welfare receipt remains a politically sensitive topic. Almost 15 years after a major piece of welfare reform legislation dramatically reduced welfare eligibility for citizens and especially noncitizens, the long-term consequences of the US welfare rules towards immigrants are still being digested. The economic crisis of 2007-2009 has created further uncertainty, and it may be some years before the role of welfare for immigrants in the recent downturn is fully understood. Meanwhile, immigrants’ fiscal contribution remains central to debates about proposed immigration reform legislation and the potential legalization of the United States’ large unauthorized immigrant population.

This report describes the US approach to immigrant integration and welfare receipt and examines its consequences. Section 2 provides a brief profile of immigration to the United States. Section 3 explains the US welfare system and immigrant eligibility for the largest programs. Section 4 examines targeted spending on immigrants, as well as some important non-targeted spending that reaches large numbers of immigrant families. Section 5 examines the consequences of the US approach to welfare and immigrant integration.

Immigrants and Immigration in the United States

The United States is the largest immigrant-receiving nation in the developed world. 38 million foreign born make up about 12.5 percent of the population, and the country has typically received more than 1 million net new immigrants per year in recent periods of economic growth.1 For most of US history, immigrants arrived primarily from Europe. This share has declined in recent decades, however, as immigration grew from other continents. Following the «bracero» guestworker program for agricultural workers from Mexico during and after World War II, immigration from Latin America increased substantially. Meanwhile, a 1965 amendment to immigration law removed 40 year-old policies that had heavily restricted immigration from outside of Europe, allowing increasing levels of «chain migration» from other regions. These developments alongside economic forces including globalization have led to large-scale inflows since the 1970s, creating a substantial foreign-born population primarily from Latin America (54 percent), Asia (27 percent), and Europe (13 percent).

The US immigration system in recent decades has been characterized by liberal family unification policies and high levels of illegal immigration, alongside a more limited employer-driven economic immigration stream. About two thirds of legal permanent immigrants arrive on family visas for spouses, parents, children or siblings of US citizens (or for the spouses and children of legal permanent residents). Humanitarian immigrants have received legal permanent residence at a rate of about 125,000 per year over the past decade—a large absolute number by international standards, but a relatively modest share of all green cards (16 percent in 2008 and 2009).2 Legal employment-based immigration is also modest and is heavily skewed towards highly-skilled workers (either on permanent visas or on a range of temporary visas, some of which allow the holder eventually to apply for lawful permanent residence (LPR) status. Supply of employment-based green cards has lagged demand over the past decade, and to some extent as a result, a large unauthorized population (approximately one third of all immigrants) provides a highly labor-motivated immigration stream at the lower end of the labor market.3

A quick profile of the immigrant population in the United States reveals some basic trends that should inform the discussion of immigrant integration and welfare use. Immigrants are overrepresented at both the high and low ends of the skill spectrum (and underrepresented in the middle), but only if one considers formal education alone. Broader measures of skill such as the occupations that immigrants perform and the wages they earn suggest that the picture is much more balanced.4 In particular, the US labor market has provided a mechanism for foreign-born workers with low education levels (and in many cases low English proficiency) to earn decent wages in jobs with room for upward mobility.5 Unlike in many European countries, immigrant unemployment is relatively low and over the past decade has rarely diverged from native unemployment by more than a percentage point.6

But despite the United States’ impressive record of integrating large numbers of immigrant workers, immigrants’ economic assimilation is far from complete. First, poverty rates remain higher among immigrants than among the US born. After years of falling poverty during the economic boom of the 1990s, immigrant families’ poverty rates have increased somewhat in the past two years to 20 percent, compared to 12 percent for native households.7 Second, huge variations exist in the economic security of different immigrant groups. Hispanic immigrants are particularly vulnerable to changes in economic fortunes, and the substantial proportion among them who are unauthorized face massive barriers to upward mobility.8 Third, while immigrants have traditionally concentrated in just a few US states, over the past two decades greater geographical dispersion has created rapidly growing immigrant populations in areas with little history of immigration and few institutions designed to facilitate immigrant integration. Finally, the growing numbers of the children of immigrants in US schools (who make up almost a quarter of children under the age of 18)9 face the combined barriers of low English proficiency, low-income family backgrounds, and residential and educational segregation. These children are increasingly failing to match the academic achievement of their non-immigrant peers.10

Immigrant eligibility for general welfare programs

The current welfare architecture in the United States was largely framed by a major piece of legislation in 1996, the Personal Responsibility and Work Opportunity Act. This legislation represented a fundamental shift towards a «welfare to work» approach, significantly reducing eligibility for cash-based welfare for low-income families, and instituting lifetime limits on the duration of federal benefits for individuals and households.

At the same time, the 1996 legislation dramatically reduced immigrants’ eligibility for federally-funded, means-tested welfare programs. Most notably, it barred legal immigrants from receiving federal money through most welfare programs until at least 5 years after they receive a green card. The new laws on immigrant welfare receipt represented a departure from previous law under which most legal immigrants were eligible for welfare benefits on largely the same basis as citizens, although elements of the new approach can be identified even before the first major federal legislation on immigration in 1882.11 This philosophy is that immigrants should be self-sufficient and that public benefits should not create an incentive for immigration to the United States. As the 1996 welfare reform legislation states very clearly, immigrants should «not depend on public resources to meet their needs, but rather rely on their own capabilities and the resources of their families, their sponsors, and private organizations».12 In addition, public support for restrictive welfare policies has arisen not just from the substantial financial savings that could arise from denying immigrants benefits (40 percent of the savings from welfare reform were expected to come from the restrictions on immigrants), but also from the perception that immigrants already get a «good deal» by moving to the United States.13

Basic characteristics of the US welfare system

The safety net in the United States comprises a patchwork of programs aimed at the elderly and at low-income families and the unemployed. Most programs not aimed at the elderly are either restricted to or prioritize families with children. The federal and state governments share financial responsibility for administering and funding welfare programs, especially in the wake of the 1996 welfare reform. States often have considerable discretion over the management of most programs, leading to substantial variation in eligibility and benefits. This is particularly true of provision to immigrants, as we describe shortly. In many cases, federal funding provides a baseline level of benefit, and some states supplement this with their own funds.

This report focuses on seven major programs that together make up the lion’s share of the US safety net.

• Temporary Assistance for Needy Families (TANF) is a benefits program aimed primarily at families with children. States have broad discretion over how they use fixed federal funding, and huge differences in benefit levels and eligibility exist across the United States. However, various federal laws apply to all state-run TANF programs. These include the requirement that participants undertake work-related activities (such as job search or vocational training), and that no household receive federally-funded benefits for more than 60 months over their lifetime.

• The Earned Income Tax Credit (EITC) is a refundable tax credit that supplements income for working individuals with low annual earnings. The program is designed primarily for families with children, but single workers on very low incomes can earn some benefits. In 2009, the maximum EITC benefit was $5,657 for a married couple with three children earning between $12,550 and $21,450; payments are phased out as income rises, and cease once family income reaches $48,000.14

• Supplementary Security Income (SSI) is a cash benefits program for elderly, blind or disabled individuals living on very low incomes—broadly speaking, groups that are not expected to work. The program is administered federally but most states supplement federal SSI benefits with their own funds. Average SSI payments were approximately $500 per person per month in March 2010.15

• The Food Stamp Program (also known as the Supplemental Nutrition Assistance Program) provides funding for low-income households to buy food. It provides an average of $133 per month to each household member.16 The program is federally funded and administered by states.

• Medicaid is a health insurance program for low-income families, pregnant women, the elderly, and disabled individuals. Note that the United States does not have a universal healthcare system, but relies primarily on private insurance. Medicaid, therefore, provides insurance to some groups who would not otherwise be covered. The program is funded jointly by states and the federal government, but states have some discretion over eligibility and benefits provided under the program.

• The State Children’s Health Insurance Program (SCHIP) provides medical insurance for children in low-income families not eligible for Medicaid (because they are above the income threshold). The program is jointly funded by states and the federal government.

• Unemployment Insurance (UI) is a contributory program which replaces an average of about half of a worker’s previous wages if they become unemployed, for a period of up to 26 weeks.17 To be eligible, an individual must have worked for four of the last five quarters. The program is state-funded, but during recessions the federal government provides additional assistance, including an extension of benefits for those who have exhausted the 26-week limit.18

These are the major programs. Others exist but we do not examine them in any detail in this report. These include additional food-based programs for Women, Infants and Children (WIC), free and reduced priced school lunches for children in low-income families, emergency medical services, and public health-related medical interventions such as immunizations.

Immigrant eligibility for the main welfare programs

Complicated rules govern immigrant welfare receipt. Immigrant eligibility varies across programs and across states. When federal funding is involved, certain groups of immigrants are consistently eligible or ineligible; but state programs often make up for gaps in federal provision. States also have the right to deny benefits to immigrants who are eligible according to federal rules, but in practice few do this. To generalize, the «hierarchy» of eligibility is approximately as follows (in decreasing order of access):

• Naturalized citizens (who receive the same benefits as US-born citizens);

• Refugees and asylees;

• Legal permanent residents who have held a green card for at least five years;

• Legal immigrant children;

• Legal permanent residents of less than five years tenure;

• Temporary workers and students;

• Unauthorized immigrants.

Table 3.1 shows the basic federal eligibility rules for the major programs. As we describe later, however, the federal government does not have the «last word», and states can provide additional benefits using their own funds (even to unauthorized immigrants). Some programs are more inclusive than others. Emergency medical care and free or reduced price school lunches, for example, is available to all immigrants regardless of legal status; EITC and Unemployment Insurance also provide relatively broad coverage. Most of the traditional welfare programs (such as TANF, SSI and food stamps) are not available to immigrants with less than 5 years tenure as a lawful permanent resident.

A few points are worth noting. First, naturalization is the clearest way to secure access to welfare. In order to naturalize, immigrants must have been a legal permanent resident for 5 years (or for 3 years if they received their green card by marrying a US citizen); they must pass background checks and a citizenship test in English. Many of the eligible but non-naturalized immigrants are thought to have limited English proficiency.19

Second, while in principle no distinction exists between family and employment-based legal permanent residents, in practice the waiting period is longer for employment-based immigrants, since most spend several years as a temporary immigrant before receiving a green card. Legal temporary immigrants on work visas are essentially ineligible for all programs.20

Finally, all legal immigrants are eligible for contributory, non-means tested programs such as social security and disability insurance, so long as they have made sufficient contributions while in legal status, and are still authorized to live in the United States at the point that they claim the benefit (in other words, this excludes temporary workers and visitors whose work and residence authorization is conditional on retaining the job for which they received a visa). Most legal immigrants of age 65 and older who have been resident for 5 years or more are also eligible for the main national health insurance program for the elderly, Medicare.

The Role of State Governments

In response to the minimal federal funding, many state governments have stepped in to provide a somewhat wider safety net to a greater proportion of immigrants. Among this number are some of the largest immigrant destination states such as California, Illinois, and New York. For example, several states fund TANF benefits for immigrants who are ineligible at the federal level;21 a smaller number also provide food stamps to most ineligible legal immigrants.22 Medical programs are particularly common. Largely as a result of state contributions to medical assistance for otherwise ineligible immigrants, participation in medical programs for adults has not decreased since welfare reform.23 For the most part, any additional state funding for ineligible immigrants goes to legal permanent residents who have not yet spent 5 years in the country. However, a handful also fund healthcare insurance for low-income children even if they are unauthorized (these include major immigrant destinations New York state and Illinois).24 Note that by the standards of many European nations, even the expanded benefits that states provide on top of federal levels are relatively minimal.25

States nonetheless provide a substantial share of funding for immigrant welfare. One consequence of this system is vulnerability to cuts in benefits or eligibility when state budgets are strained—for example in times of recession. Many states now face significant budget deficits, and this has put pressure on «discretionary» programs supporting immigrant welfare. For example, the major immigrant-receiving state of California provides among the most generous welfare benefits in the country, but at the onset of the recession the governor proposed to eliminate funding for several programs supporting immigrants not eligible for federal money. These cuts did not materialize largely because the federal American Recovery and Reinvestment Act (ARRA), a stimulus package passed in early 2009, provided substantial payments to states in order to support services such as education and healthcare. When this funding runs out, however, cuts may again be on the table.

Boks 3.1 Immigrant Participation in Welfare: To What Extent Do Immigrants Rely on Welfare in the United States?

How significant is immigrant participation in welfare programs? Various studies have come to different conclusions, and methodology matters (for example, whether to consider immigrants as individuals or as households, whether to include their children in the calculation, and which programs to examine). Most importantly, it matters whether analysts compare immigrants to the US born, or whether they compare low-income immigrants to low-income US born.

Because immigrants are significantly more likely to have low incomes, they participate in welfare programs at higher rates. However, controlling for immigrants’ low incomes, welfare use is relatively low. Estimates from 2004 suggest that among low-income legal permanent resident families with children, 4 percent used SSI, 5 percent used TANF, 22 percent used food stamps, and 59 percent used Medicaid (Figure 3.1). With the exception of Medicaid, these percentages are somewhat lower than among equivalent US-born families. Note that although these numbers were larger before the 1996 welfare reforms, even before 1996 immigrant use of welfare was lower than for the US born in the same categories.

Kilde: Randy Capps, Michael Fix, and Everett Henderson, «Trends in Immigrants’ Use of Public Assistance After Welfare Reform» in Michael Fix (ed) Immigrants and Welfare (New York: Russell Sage Foundation, 2009)

Publicly-funded integration programs targeted at immigrants

The United States does not have an immigrant integration «program», and no single federal authority oversees immigrant integration. Integration policy comprises an uneven patchwork of programs providing language instruction, education and training, employment services, and citizenship instruction. These programs either vary dramatically by state, or focus narrowly on just a few groups. Proposals for a more coordinated federal policy have been made (and have appeared in previous unsuccessful legislative proposals). For the most part, however, integration policy does not take center stage in the US immigration debate and the approach to immigrant integration remains laissez faire, relying on the efforts of non-profit and grass-roots organizations and on the integrating potential of the US labor market in which immigrants participate at high rates.

A few features of the US approach to immigrant integration policy are worth noting. First, federal programs targeted specifically at immigrants receive very little funding and are narrowly focused on two groups: refugees and agricultural workers (the great majority of whom are foreign born). Targeted spending under these programs has fluctuated substantially over the past decade.26 Similarly to welfare policies, various initiatives at the state and local levels fill some of the gaps but certainly do not add up to a comprehensive program. Again, huge variations exist in the availability of services for immigrants across US states.

Second, while eligibility rules for integration-related initiatives differ by program and by geographical area, on the whole they tend to be significantly more inclusive than welfare benefits. Unauthorized immigrants are eligible for several of the programs, particularly those provided through public schools (which are required by law to admit children regardless of their legal status). Indeed, a few programs designed for agricultural workers serve a population that is roughly 50 percent unauthorized. In addition, the 5 year bar on eligibility for legal permanent residents that applies to federally-funded means-tested welfare does not apply to integration policies.

Third, with the exception of the federal program for refugee resettlement, integration policies in the United States tend to focus much more on ongoing support for low-income immigrant families than on helping immigrants immediately after arrival. Meanwhile, the highly skilled and those with incomes above a certain threshold are expected to «fend for themselves». There is no national framework for credential recognition, for example, although various non-profit organizations help skilled immigrants to gain recognition for foreign qualifications.

Finally, several policies and initiatives that are not labeled as «immigrant integration» but that affect the resources and amenities that immigrants can access, have a collective impact that is perhaps more important than that of targeted integration programs. The largest of these is the collection of policies concerned with English language proficiency.

The following few pages describe the nature of the services available to immigrants under various different categories, either as targeted programs or as broader initiatives that are not necessarily thought of as immigrant integration measures.

The integration program for refugees

The broadest federal integration program in the United States provides assistance to refugees. The Office of Refugee Resettlement, which runs the program, had a budget of $715 million in 2009.27 These funds are used to provide cash and medical assistance, employment and skills training, job placement services, language instruction, and support for victims of torture. In many ways, the refugee program is the closest analogue to the concept of «settlement services» found in some European countries. It provides much more comprehensive set of services and information for newly-arrived immigrants than other programs but, of course, is focused narrowly on the refugee population.

Programs for Agricultural workers

Various programs attempt to address high poverty rates, poor health and limited access to education among agricultural workers. Only one quarter of hired farm-workers are US citizens (including naturalized citizens) and roughly half are unauthorized.28 As a result, these programs are essentially targeted at immigrants. They provide a range of education, social and health services, pre-Kindergarten child care, adult education and parenting classes. The largest of these programs, Migrant Education, consisted of $380 million in grants to states in 2008.29

Language Learning

Approximately 10 percent of children in US schools are English language learners (ELL), and the number of ELL children has grown rapidly over the past decade. Most of these children have parents who speak Spanish.30 Here, the most important policies are again not conceptualized as targeted «immigrant integration» measures, though they may have this effect. The No Child Left Behind Act of 2001 (NCLB) for the first time imposed a federal requirement that schools must demonstrate how students with limited English proficiency have progressed in English. Schools must report the performance of English language learners as a separate subgroup, and are held accountable for these results. States receive grants to help their schools to meet this requirement.31 Because this policy is tied to the public schooling system, no restrictions apply on the basis of legal status. In addition, immigrants also stand to benefit from provisions in the No Child Left Behind Act that provide funding for schools in disadvantaged areas—another example of non-targeted spending that could be particularly important for immigrant families.

21.6 million adults in the United States also have limited English proficiency.32 Given the extraordinary importance of English language ability in the US labor market, English language instruction, is one of the single most important integration policies. English language classes for adults are jointly funded at the state and federal level as part of a wider non-targeted program of adult education and workforce development for both immigrants and the US born. Total spending on adult education grants was $554 million in 2008,33 and states can divide this between programs (including language instruction) as they choose. These funds are quickly exhausted, however, and language classes are heavily oversubscribed.34 Oversubscription could worsen significantly if any new immigration reform legislation includes a legalization program that requires an English test in order for unauthorized immigrants to qualify for legal permanent residence. (At the time of writing, the prospects for immigration reform in the United States remained unclear).

Language access and information

The United States does not provide information or «orientation» services at the federal level to newly arriving immigrants, but the US government requires that agencies using federal money must ensure access for individuals with limited English proficiency—most of whom are of course immigrants. In addition, various state and local governments have aimed to expand immigrants’ access to public services by employing bilingual staff and/or interpreters. New York state has recently provided funding for translation and interpretation services in public schools for LEP parents. And Illinois has a multilingual outreach program which aims to inform immigrant families about social services.35

Citizenship preparation

The federal government provides modest funding in grants (totaling approximately $7 million in 2010)36 for providers of citizenship preparation services in the form of English language classes, civics instruction, teaching materials, and assistance with citizenship applications. Additional civics or citizenship-based classes are provided under the same framework as the adult education programs described above. Some states such as New York and Illinois also provide citizenship preparation services. Overall, however, funding for these programs is modest; it has also proved particularly vulnerable to cuts in the recession.

Employment programs

No immigrant-targeted employment services exist at the federal level, except under the refugee resettlement program described earlier. Legal immigrants are eligible for some relatively limited programs available to all US workers, although immigrants may lack the basic language skills required to benefit from the classes and are underrepresented among the beneficiaries of these programs.

State-level policies

Some states — notably Illinois — have begun to develop a more comprehensive approach to immigrant integration, aiming to provide a range of targeted services in areas such as English language learning, employment training, credential recognition, and information services. In recent years a few states established advisory committees or panels to examine immigrant integration policies and develop recommendations for more effective state-level measures; but the recent recession and resulting budget cuts have often prevented these efforts from coming to fruition.37 In addition, these state-level efforts are far from widespread.

Some non-targeted state-level policies are also worth mentioning. For example, Illinois and Oklahoma run universal pre-Kindergarten child care and education programs for which immigrant families (including the unauthorized) are eligible. In addition, various states’ public universities and colleges offer unauthorized immigrants the lower tuition fees available to «in-state» residents, potentially expanding unauthorized children’s access to higher education.38

The Consequences of the US Approach to Immigration and Welfare

If Norway combines relatively restrictive admissions policies with generous welfare and integration policies for immigrants who gain admission, the United States’ approach is perhaps the opposite. Admissions policies are relatively generous (at least when it comes to family-based immigration), but welfare rules are restrictive and targeted integration spending extremely limited. This approach is consistent with the US philosophy that immigrants should be self-sufficient and should not rely on public programs. It is also consistent with a higher tolerance for inequality across the board, and a somewhat lower propensity to redistribute income than in many other countries.

What have been the consequences of this approach? Have restrictive welfare rules led to high immigrant poverty? Have they created the right «incentives» for would-be immigrants? What have been the impacts on immigrant-receiving communities?

Impact on immigration to the United States

The hypothesis that welfare benefits serve as a «magnet» for new immigrants receives little empirical support in the United States. Immigrants have traditionally had low participation in public welfare programs (even before welfare reforms introduced highly restrictive rules), and their location decisions appear to be driven primarily by employment opportunities and existing social networks.39 As a result, it seems unlikely that US policies restricting welfare recipiency among immigrants will have had a large impact on the «quality» of immigrants coming to the country. Rules that prevent immigrants from sponsoring a relative’s green card if their income is at less than 125 percent of the poverty line are likely to have affected the inflow of immigrants into low-income families, although it is difficult to quantify the impact of this provision directly.40

Impacts on Immigrant Integration and Poverty

The impact of restrictive welfare rules on immigrant households is subject to debate. On the one hand, poverty among immigrants fell after welfare rules became more stringent in 1996, although the booming economy over this period is responsible for much of this trend. On the other hand, poverty remains higher among immigrants than among the US born; while non-citizens have lower levels of access to medical and dental care, shoulder more of the cost of medical care themselves, and face greater food insecurity and hunger.41 Despite somewhat more inclusive policies towards children, noncitizen children also fare worse than citizens according to various measures of health and educational attainment,42 suggesting that some form of investment in their wellbeing is necessary. In addition, evidence shows that the recent recession has increased poverty rates among immigrant families.

In all of these cases, it is very difficult to determine what would have happened under a different policy regime, and direct measures of the impact of 1996 welfare reform are very limited. Three important arguments about immigrants and US welfare are worth noting, however. First, the complex rules and patchwork of programs and eligibility requirements may have led to «chilling effects» whereby eligible immigrants do not apply for benefits either because they believe they are not eligible, or because they are concerned they will jeopardize their legal status or application for citizenship. In other cases, unauthorized parents are unwilling to apply for benefits on behalf of their US citizen children, for fear of revealing their legal status. Interestingly, participation in welfare programs after the 1996 reforms declined significantly even among groups (such as refugees) that experienced little or no change in eligibility—an effect that could arising from confusion or fear, or from an improved economic outlook at falling poverty rates among immigrants.43

Second, immigrants may use welfare differently from the US born. Some researchers have argued, for example, that before 1996 immigrants used welfare in the initial stages of settlement as a means of smoothing income before they developed adequate social and employment networks, and that welfare dependency is less of a problem among immigrants than the US born.44 If this is the case — in other words, if integration occurs fastest within a more inclusive welfare system — removing new immigrants’ access to welfare could hurt integration in the long run. By contrast, other researchers have found that welfare restrictions increased labor supply among immigrant women, particularly the low educated and recently arrived.45 Whether or not welfare for immigrant adults reduces the incentive to work, however, restrictions on welfare receipt that reduce immigrant families’ incomes could damage the development of children in immigrant families. The evidence remains incomplete, but it remains plausible that welfare access for immigrants could serve as an investment in the future productivity of immigrants and especially their children; if this is true, the financial savings that result from restrictive welfare rules would appear less substantial.

Finally, the central goals of welfare reform for the US population—to reduce welfare dependency, to increase the incentive to work, and to encourage stable families—are less appropriate for immigrants in the United States than they may be for the US born. Immigrants are more likely to live in traditional two-parent families, and labor force participation among immigrants is already very high, particularly for some low-skilled, low-income groups such as unauthorized workers.

Impacts on Immigrant-Receiving Communities

As we have described in this report, government spending on immigrant integration and welfare benefits takes place primarily at the state and local levels. By contrast, immigration policy is the responsibility of the federal government. Not only are federal admission policies for family immigrants relatively generous, but the US does not admit most immigrants on the basis of their likely ability to integrate successfully.

As a result, some states such as California spend substantial sums on immigrant integration and welfare but do not have control over the number of immigrants they admit (because of both federal admissions policies and high levels of illegal immigration). Generally speaking, states do not receive «compensation» for these costs. However, some funds that remain available to reimburse states for the cost of uncompensated emergency medical services that they provide to immigrants without health insurance. And a small program of grants was made available to states in order to administer a one-off legalization policy in 1986. However, area-based policies to help destinations with new or concentrated immigrant populations are not widely used.

Conclusion

Spending on immigrants in the United States is low. To some extent, this follows naturally from the underlying philosophy of self-sufficiency and moderate aversion to redistribution. Immigrants’ fiscal contribution or burden loom large in the public debate over immigration, particularly when it comes to illegal inflows. In a recent poll, just over 60 percent of respondents said that they were «very concerned» that illegal immigration might put a strain on schools, hospitals and government services.46 In a 2008 poll, two thirds of American respondents thought that «illegal immigrants cost the taxpayers too much by using government services like public education and medical services».47 Meanwhile, paying taxes is one of the most commonly proposed requirements for legalization. Polling on legal immigrants’ use of public services is less common, although public opinion seems to be somewhat more forgiving in this regard. During the debate about an ultimately unsuccessful immigration bill in 2007, a Gallup poll found that concerns about overall immigrant use of welfare and public services ranked fourth among respondents’ reasons for opposing the bill—one of the main reasons for 11 percent of respondents.48

These trends suggest that restrictive welfare for immigrants arguably could increase the prospects of reform that would help immigrant families in other ways (for example, by legalizing the large unauthorized population). If immigrants received the same welfare benefits as the US born, support for current levels of immigration or for legalization might diminish.

On the other hand, the question arises whether the United States forgoes the benefits of investment in the future productivity of immigrants and their children. Almost one quarter of children under the age of 18 come from immigrant families, and many are US citizens. If US policies undermine these children’s development, already high levels of inequality could become more persistent. Simmering concerns exist that intergenerational mobility among immigrants may be lower than it has been in the past, and that their labor market prospects on arrival have declined so much that they are no longer able to catch up with their non-immigrant peers even in the second generation.49 All these things suggest that more coordinated, more concerted investments in immigrant integration might bring long-term rewards.