Skip to navigation

Skip to main content

 

The Integration of Immigrant Families -- Michael Fix, Jeffrey Passel, Wendy Zimmermann

THE INTEGRATION OF IMMIGRANT FAMILIES


THE URBAN INSTITUTE


Michael Fix

Jeffrey Passel


Wendy Zimmermann

Paper Prepared for:

Immigration and the Changing Face of Rural California

Focus on the Napa Valley

October 5-6, 2000
University of California Davis

The authors would like to express their appreciation to Laureen Laglagaron for her expert assistance. An earlier version of this paper was presented in June 2000 at a Miami meeting of the National Conference of State Legislatures.

Support for the research and writing of this report was provided by the Andrew W. Mellon Foundation, the Annie E. Casey Foundation and the U.S. Office of Refugee Resettlement.

Opinions expressed are those of the authors and do not necessarily represent the views of the Urban Institute or any of its sponsors.


Table of Contents


EXECUTIVE SUMMARY..................................................................................................... i

I. Introduction................................................................................................................. ii

II. Selected demographic trends .................................................................................... iii

III. Selected integration trends......................................................................................... iv

IV. Trends in public spending on immigrants and their families..................................... vi

V. Towards a reform agenda: Selected policy design considerations.......................... viii

VI. Selected directions for the future.................................................................................. x

I. INTRODUCTION.............................................................................................................. 1

II. WHY DISCUSS THE INTEGRATION OF IMMIGRANT FAMILIES?..................... 3

Continuing mismatch between immigration and immigrant policies................................ 3

Averting unintended consequences..................................................................................... 3

Population dispersal........................................................................................................... 4

Shifting political climate.................................................................................................... 4

Strong economy................................................................................................................... 5

Deeper knowledge base...................................................................................................... 5

Policy targets: PRWORA and ESEA reauthorizations....................................................... 5

III. HOW IS THE IMMIGRANT FAMILY CHANGING?................................................ 7

High flows........................................................................................................................... 7

Dispersal............................................................................................................................. 9

Changing origins and rapid rise in Mexican migration.................................................. 11

Increased share of undocumented.................................................................................... 12

Large share of immigrants in families............................................................................. 14

Predominance of mixed status families............................................................................ 15

IV. HOW WELL ARE IMMIGRANT FAMILIES FARING?........................................ 18

Change across generations............................................................................................... 19

Language acquisition among school age children........................................................... 21

Household income growth................................................................................................ 21

Benefitting from the boom economy................................................................................. 22

Increased naturalizations................................................................................................. 24

Increased poverty among school-age children................................................................ 26

Segregation of LEPs in schools........................................................................................ 27

High drop out rates........................................................................................................... 28

High but rapidly falling rates of welfare use................................................................... 29

Rising uninsurance rates.................................................................................................. 31

V. INTEGRATION POLICY: SELECTED CONCEPTUAL AND DESIGN ISSUES. 33

Determining our expectations for immigrant family integration.................................... 33

Determining the reach of antidiscrimination principles................................................. 34

Limiting immigrant families’ support obligations........................................................... 36

Apportioning intergovernmental roles and responsibilities............................................ 38

Targeting integration policies to discrete populations................................................... 39

Choosing between mainstream and targeted programs................................................... 41

Identifying strategies for leveraging the private sector.................................................. 42

Assessing the merits of a national office for immigrants and refugees........................... 43

VI. WHAT DOES THE FEDERAL GOVERNMENT SPEND ON IMMMIGRANT FAMILY INTEGRATION?................................................................................................ 45

Federal spending on targeted programs.......................................................................... 45

Impact assistance.............................................................................................................. 47

Targeted services.............................................................................................................. 48

Refugee resettlement..................................................................................................... 48

Refugee education......................................................................................................... 49

Federal bilingual education......................................................................................... 50

Migrant education........................................................................................................ 51

ESL for adults.............................................................................................................. 52

Spending on immigrant families in mainstream federal programs.................................. 54

State and local spending on immigrant integration......................................................... 56

VII. SELECTED FUTURE POLICY ISSUES.................................................................. 58

1. Safety net....................................................................................................................... 58

2. Education...................................................................................................................... 58

3. Employment................................................................................................................... 59

4. Housing and community development.......................................................................... 60

5. Creation of institutions focused on immigrant integration......................................... 60

References........................................................................................................................... 61

Figures and Tables
Figure 1. Current Levels Are High....................................................................................... 8

Figure 2. Immigrant Numbers at Peak – Percentage is Not............................................... 9

Figure 3. Concentration is High, But New Centers Emerge in ‘90s................................. 10

Figure 4. Most Legal Immigrants from Latin America and Asia...................................... 11

Figure 5. Rapid Growth of Mexican Population................................................................ 12

Figure 6. Legal Status of Immigrants................................................................................. 13

Figure 7. Undocumented Population Returns to Pre-IRCA Levels.................................. 13

Figure 8. Household Type by Nativity of Household Head: 1998.................................... 15

Figure 9. “Mixed” Families Are Common.......................................................................... 16

Figure 10. Large Share of Low-Income Families with Children are Mixed..................... 16

Figure 11. Integration is Dynamic....................................................................................... 22

Figure 12. Immigrant Unemployment Rate Declines......................................................... 23

Figure 13. Immigrant Wages Rise Slower.......................................................................... 23

Figure 14. Employer-Provided Health Insurance for Immigrants Decreases.................. 24

Figure 15. Naturalizations Increase in Wake of IRCA..................................................... 25

Figure 16. More Immigrant Children are Poor.................................................................. 27

Figure 17. LEPs are Linguistically Segregated................................................................. 28

Figure 18. Immigrant Welfare Use Declines Faster than Citizens’................................. 29

Figure 19. Low-Income Immigrant Families with Kids Use Less Welfare....................... 30

Figure 20. Nationwide Chilling Effects.............................................................................. 31

Figure 21. Enrollment in Adult Education Programs: 1994-98......................................... 53

Table 1. Targeted Federal Spending on Immigrants......................................................... 46

Table 2. Emergency Immigrant Education Program (EIEP) Funding: 1985-2000.......... 47

Table 3. Federal Refugee Resettlement Program Funding: 1990-2001........................... 49

Table 4. Federal Bilingual Education Funding: 1991-2001................................................ 51

Table 5. Federal Funding for Migrant Education Program: 1990-1999........................... 52

Table 6. Federal Funding for Adult Education Programs: 1992-1996.............................. 54

I. Introduction.
Historically, few domains of public policy have been as driven by family considerations as immigration policy. It may be well known that legal admissions based on family unification goals account for close to three-quarters of all legal immigration to the United States. But it is less widely recognized that a significant share of employment based admissions and humanitarian migrants (i.e., refugees and asylees) are premised on family grounds.

Despite immigration policy's family orientation, recent policies have imposed new burdens and risks on the immigrant family. These policies include new restrictions on legal immigrants= access to health, nutrition, and other public benefits; the imposition of new support burdens on immigrant families; increased risks of deportation for legal noncitizens; and new income requirements that citizens and noncitizens alike must meet before they can unite with their immediate relatives.

In this paper we address five questions:

< First, what are some of the most striking recent immigration trends?

< Second, how well are immigrant families faring — or put differently, what trends do we see in immigrant family integration?

< Third, what direction is public spending taking on immigrants and their families?

< Fourth, what design and conceptual issues do we confront in considering the merits of an immigrant family integration agenda?

< Finally, as we look to the future, what are some of the most important reform issues facing immigrant families?

II. Selected Demographic Trends.
< Immigration flows to the United States are quite high and growing. Annual immigration flows have tripled over the past generation with more immigrants entering the United States during the 1990s than during any other decade.

< High recent immigration levels have meant that immigrants= impacts are no longer just being felt by the six states that have traditionally received most immigrants — they are now being felt by nontraditional receiving areas as well. These communities not only find themselves with more immigrants, but, in the wake of welfare reform, they find themselves with more power to determine their rights to benefits and with more responsibility for financing the services they decide to offer.

< While the national origins of immigration flows have changed dramatically over the past thirty years, perhaps the most striking development of the last decade has been a remarkably sharp rise in Mexican migration to the United States. Today Mexicans represent 30 percent of both the total immigrant population and the annual immigrant flow. Between 1990 and 1999 alone, the Mexican population rose by 70 percent from 4.3 to 7.2 million persons, making Mexico the single largest source of both legal and undocumented immigration.

< This spike in Mexican migration is correlated with a steady increase in undocumented immigration to the United States. In 1998 almost a quarter (22 percent) of the nation=s foreign-born population was undocumented. That said, according to a study of New York=s undocumented population, almost two-thirds of children in families headed by an undocumented adult was a U.S.-born citizen.

< These large legal and undocumented immigration flows have led, not surprisingly, to big institutional impacts. Currently 20 percent — or one in five school age children in the United States — is the child of an immigrant: a share that has tripled since 1970. By way of comparison, only 16 percent of the school age population is black. Nonetheless, immigrant and language minority children have been largely absent from national debates over educational opportunity — representing perhaps the most important example of the continuing mismatch between our immigration and our immigrant policies.

III. Selected Integration Trends.
These policy mismatches would be of little interest if immigrant integration were proceeding smoothly on all fronts. While our research indicates that the integration story remains a largely successful one, some troubling counter-trends can be seen.

< The average household incomes of legal immigrants who have been in the United States for 10 years or more slightly exceed those of natives.

< Cross-generational integration trends for immigrants who are 20 to 30 years old reveal that by the second generation immigrants overall end up roughly even with — or in some instances exceeding — third generation whites along a number of important dimensions. These include their: (1) educational attainment; (2) labor force participation; (3) household incomes; and (4) hourly wages. While we see substantial divergence across groups, with Asians typically doing better than whites and Hispanics doing worse, these are, in general, hopeful, solid results.

< In some respects, immigrants have benefited strongly from the recent economic boom as their unemployment rates fell faster than natives’ from 1996 to 1999. Wage trends, however, tell a different story as natives= wages rose 50 percent faster than immigrants= between 1996 and 1999. In short, the current boom has produced significant employment, but not wage gains for immigrants.

< At the same time though, we see a sharp rise in poverty among the foreign born. As recently as 1980, immigrants= poverty rates only narrowly exceeded natives=. But by 1998, immigrant poverty rates were almost double natives. Along similar lines, between 1970 and 1995 fully 60 percent of the growth in child poverty in the United States overall could be ascribed to the children of immigrants.

< We have found disturbing trends in the nation=s schools — where half of limited English proficient children attend schools where a third or more of their fellow students are also LEP.

This means that they are going to schools that are not just ethnically and economically segregated, but are linguistically isolated.

< We see the share of families that are divorced or separated doubling from the first to the second generation (equaling the rates of non-Hispanic white natives). Along the same lines, we see the share of unmarried parents also rising rapidly from the first to the second generation (with the second generation=s rates exceeding white natives by a third).

Mixed Status Families and Welfare Reform.

With these broad, demographic and integration trends as context, we turn to the demography of the immigrant family.

< According to data from the Census Bureau, 85 percent of immigrant families with children are mixed immigration status families — i.e. families where at least one parent is a noncitizen and one child is a citizen.

< 75 percent of the children in immigrant families are citizens.

< In New York, 70 percent of families with children headed by undocumented immigrants contain citizen children.

< 60 percent of all low-income children in Los Angeles and 30 percent of New York=s low-income children live in mixed status families.

The prevalence of these mixed status families holds profound policy implications. Policies intended to restrict benefits to noncitizens inevitably spill over to citizen members of these immigrant households. We have found strong evidence that chilling effects on both noncitizens= and citizens= use of benefits have emerged following welfare reform. In many states, low enrollment in the Child Health Insurance Program (CHIP) is being blamed in large part on the reluctance of immigrant families to enroll their citizen children. We also found that between March 1994 and March 1997:

< Despite the fact that few immigrants had lost eligibility because of their status, welfare use among noncitizen families with children fell more sharply than their native counterparts. By 1997, noncitizen families= use rates were almost half those of citizen families. As a result, immigrants accounted for a disproportionate share of overall welfare caseload decline.

IV. Trends in Public Spending on Immigrants and their Families.
We can view federal spending on immigrant families through two lenses: spending that is more or less targeted to newcomer groups and spending on mainstream programs in which immigrants participate.

Targeted programs. There are three defining characteristics of targeted federal spending on immigrant programs. First, they are modest in scale. Total spending is roughly $1.6 billion — despite the fact that the foreign born constitute 10 percent of the nation=s overall population. Second, the programs that make up the nation=s immigrant policy have been created in an ad hoc manner and fall short of representing a coherent integration agenda for immigrant families. Third, after declining sharply through the 1980s and most of the 1990s, federal spending on targeted programs has grown significantly in the past 2 to 3 years, presenting potential opportunities for immigrant families.

Targeted spending programs are exemplified in some respects by the Emergency Immigrant Education Program (EIEP). The EIEP provides funds to school districts based on the number of foreign-born students in U.S. schools for 3 years or less. Though spending for the EIEP is higher than it has ever been ($185 per student) it remains far below the $500 per student that is authorized. Moreover, reimbursement rates per student are less than 10 percent of the reimbursement payments made to states to offset the costs of incarceration under the State Criminal Alien Assistance Program ($184 versus $2,307).

Mainstream Programs. While spending on targeted programs may have been rising it is dwarfed by federal spending on mainstream programs. To illustrate, while 300,000 limited English proficient (LEP) students were served in 1999 under the targeted Bilingual Education Act programs, 1.5 million LEP students were served under the $8 billion mainstream Chapter I Program for the disadvantaged.

We note several recent trends within mainstream social welfare programs that should hold promise for immigrants and their families. First, we have seen the creation of important new programs for low wage families. These include the $20 billion CHIP program and the $4 billion Welfare-to-Work program. Second, we see sharply rising funding levels within a number of family programs such as Head Start and the consolidated Child Care Development Fund. Third, eligibility for some federal programs such as Medicaid has been expanded to working poor families whose incomes excluded them from participating in the past.

Despite these trends, though, immigrant families= access to these programs may lag for several reasons. The most powerful is the continuing effect of PRWORA=s exclusion of immigrants arriving after August 1996 from the 5 core, federal social welfare programs: Food Stamps, Supplemental Security Income, TANF, Medicaid and the Child Health Insurance Program. At the same time, implementation practices have historically led to under -representation of immigrants in some programs such as Head Start.

State and Local Spending. Like the federal government, states and localities are also spending more on a wider range of assistance than before welfare reform. Federal restrictions have shifted greater costs to states and localities as they step in to provide assistance, such as food stamps, that was previously the sole responsibility of the federal government. This cost-shift deepens the inequity that existed prior to welfare reform, with the federal government receiving the lion=s share of the taxes paid by immigrants and states and localities paying for most of their services.

V. Towards a Reform Agenda: Selected Policy Design Considerations.
As we begin to consider possible policies and practices that take the immigrant family more fully into account, we confront a number of design challenges:

< First, we need to determine what our expectations of immigrant integration should be. What should we expect of immigrants during the period after arrival? Over the life course? By the second generation? How should our expectations vary depending on the immigrant’s status at entry: whether he or she is a refugee? A family versus an employment immigrant?

< Second, how can policy better account for the mixed status family and the fact that policies aimed at noncitizen adults spillover to citizen children? What are the policy implications of the fact that most children in families headed by an undocumented immigrant are citizens?

< Third, when is it legitimate to treat legal immigrants differently than citizens? Here we would only note that by making citizenship, versus legal status, a prerequisite for participating in public programs, the United States seems to be evolving into an outlier among Western industrialized nations.

< What limits should be set for the support obligations imposed on the immigrant family? While many might agree that the family should be the first line of support for newcomers, we quickly face the issue of compelling poor noncitizen but not citizen families to pay for health insurance coverage. Even before welfare reform went into effect, immigrants and their children not only had far fewer average health care visits than natives, they also made dramatically fewer emergency room visits (0.2 versus 0.7 per year).

< Fifth, we need to determine when immigrants are better served by mainstream policies and institutions versus targeted policies and dedicated institutions (such as offices of refugee resettlement). Here we would only note that a recent Urban Institute study of immigrants in secondary schools contends that while the institutional challenges raised by newcomer populations seem at first blush to be quite idiosyncratic, the reforms that best addressed them were universal in character. (Such reforms included professional development for monolingual classroom teachers; extended time for learning, and expanded school accountability.)

VI. Selected Directions for the Future.
Below we present a short and inevitably arbitrary list of policy issues that we believe hold far–reaching significance for the integration of immigrant families.

1. Safety net.

< While the Congress has restored eligibility to legal immigrants for a number of benefit programs, working-age immigrants remain ineligible for food stamps. Further, the rapidly increasing populations of legal immigrants arriving after 1996 remain ineligible for the five core federal means-tested programs: food stamps, SSI, Medicaid, CHIP and TANF. Should these benefits be restored? If so, how should their restoration be financed? What outreach would need to be done to overcome the chilling effects of earlier exclusions?

< Litigation in California and New York has underscored the fact that newcomers often do not have equal language access to public benefit programs or to publicly financed employment and work support services. How frequently is such access denied? When is it prohibited? And what practical remedies exist to overcome language access issues?

< Reauthorization of the federal welfare reform law in 2001 will unquestionably address the issue of the $7 billion state TANF surplus that has arisen as a result of declining caseloads and sustained federal spending at 1996 levels. Should Congress authorize the states to use these surpluses to offset the state costs of extending safety net services to legal immigrants who were barred from federal benefits?

2. Education

The reauthorization of the large Elementary and Secondary Education Act either this year or next and the rapid adoption of content and performance standards in schools, raise a number of important issues when it comes to the education of immigrant youth. These include:

< How effectively are limited English proficient and immigrant students served by mainstream education programs such as Chapter 1, Head Start, and Perkins Vocation Education programs?

< Are schools taking into account the needs of LEP and immigrant students as they introduce standards-based education reforms and the high stakes testing that accompanies those reforms?

< Given the rapid growth of immigrant children in the nation’s schools, what strategies can we adopt that will equip classroom teachers so they can communicate more effectively with English language learners? What best practices have been introduced in the area of professional development?

< What fiscal policies might correct the mismatch between the grade distribution of LEP students (many of whom are in secondary schools) and the availability of resources to promote English language learning (which is heavily concentrated in elementary schools)?

< How can we expand language and literacy instruction not just to children, but to their parents? What promise and lessons does the Department of Education’s Even Start Program offer?

3. Employment

< Given the strength of the economy, and the ready access that immigrants now have to jobs, what policies might accelerate their economic mobility? Should policymakers’ primary focus be on post-employment services that promote skill and language acquisition? What incentives need to be provided to employers to offer such services?

< Given the concentration of immigrants in low wage sectors, does it make sense to expand federal and state enforcement of regulatory programs (wage and hour, occupational safety and health) aimed at low wage industries?

4. Housing and Community Development

< Can we increase immigrant home ownership by adopting asset accumulation models, individual development accounts, or other strategies for establishing credit worthiness? What role can the public sector play?

< To what extent does the continuing withdrawal of Section 8 subsidized units from housing markets affect immigrant families? What if any local policies have successfully expanded housing opportunities for low-income newcomer families?

5. Creation of Institutions Focused on Immigrant Integration

< Would it be advisable to create a National Commission on the Integration of Immigrant Families that would frame a national and state policy agenda on these issues?

< Should a national Office for Refugees and Immigrants be created? Should it be housed, as the Office of Refugee Resettlement now is, within the Administration of Children and Families?

I. INTRODUCTION.
The purpose of this paper is to help frame a discussion of the integration of immigrant families within the United States: their progress, the receptivity of the communities within which they settle, the nation’s policies (or lack thereof), and possible future directions.

This paper proceeds from two straightforward assumptions. One is that the nation may be ready for a period of constructive engagement on the issue of how best to integrate immigrant families into U.S. society. The second is that there is a basic mismatch between the nation’s essentially liberal, if highly regulated immigration policies and its historically laissez faire immigrant policies. That is, despite the fact that the nation admits more immigrants who are on track for citizenship than any other country, U.S. immigrant integration policies have essentially been ad hoc and small scale.

We begin by noting several reasons for starting a discussion of the integration of immigrant families now. We then proceed to examine some of the demographic imperatives for an integration agenda as well as selected trends in immigrant integration. With these trends in mind, we proceed to explore some of the conceptual and design issues that should inform an integration agenda for immigrant families. We then document trends in recent spending on immigrant integration and conclude by briefly discussing several substantive areas and issues that we believe bear further work.

Because the reach of the paper is rather broad, we should note several of its limitations. In the first place, the paper relies heavily on analyses conducted by the Urban Institute, and as such our themes and findings are drawn less from the rich literature on integration than our own institution’s analyses. Second, we acknowledge that our demographic measures and our metrics of integration do not include several important trends, such as political participation. Third, despite our own past emphasis on the merits of disaggregating the immigrant population by legal status, duration of residence, national origin and the like, we have been compelled to present more aggregated findings than we might have preferred, owing to data and resource limits.

Finally, the reader will find that we do not advance a firm, narrowly drawn definition of immigrant family integration. Rather we acknowledge that the term will have differing meanings for differing people. We do believe, however, that it involves more than an accounting of immigrants’ mobility over time, encompassing notions of community change as well.

II. WHY DISCUSS THE INTEGRATION OF IMMIGRANT FAMILIES?

We believe that there are several powerful reasons to expand the quite limited current national discussion of immigrant family integration.[1] They include:

Continuing mismatch between immigration and immigrant policies. In the first place, despite unprecedentedly high numbers of entrants, few mainstream institutions — -- schools, the military, departments of transportation, child welfare agencies — have directly confronted the significance of immigration-driven demographic change for their policies and programs.

Schools may offer the best example. In 1997, one in five school-age children in the U.S. was the child of an immigrant: a share that had tripled since 1970. (By way of contrast, 16 percent of school-age children in the United States are black.) Despite their numbers, debates over educational opportunity — including vouchers, high stakes testing, standards of learning, and the like — rarely take into account the needs of the children of immigrants (Ruiz-de-Velasco and Fix, 2000).

Averting unintended consequences. Integration as well as immigration policies can have unintended effects on immigrant families. In the case of welfare reform, for example, bars on immigrants= use of benefits appear to have had the effect of chilling not just noncitizen adults, but many of their citizen children’s use of health and other services. Although immigration policy is in some ways quite family–friendly, recent reforms aimed at toughening immigration controls have had the unexpected effect of separating some immigrant families, as noncitizens are deported for minor crimes committed years earlier. Similarly, the imposition of new income requirements for sponsors may be keeping some families apart. Presumably, such unintended effects could be minimized by a more deliberate set of policies, developed with an understanding of immigration flows, the mixed legal statuses within immigrant families and immigrant integration trends.

Population dispersal. Another reason to focus on the integration of immigrant families is that immigrants are increasingly moving to nontraditional receiving states and communities. These communities not only find themselves with more immigrant families, but, in the wake of welfare reform, with more power to determine immigrants= rights to benefits, and more power to shape their own integration policies. One corollary of this new authority is communities’ increased responsibility for financing the services they decide to offer.

Shifting political climate. The time for a greater focus on the integration of immigrant families may also be ripe because of the shifting political crosscurrents in immigration and immigrant policy since the strikingly anti-immigrant period of the mid 1990s. Since the mid 1990s, we have seen the Congress restore food stamps and Supplemental Security Income benefits to some legal noncitizens; authorize some Central Americans and Haitians to seek legal status who had become — or would soon become — deportable; and expand the number of temporary visas made available to high tech workers. At the same time though, we have seen California voters overwhelmingly support an initiative to severely limit the use of bilingual education, as well as efforts to revive Proposition 187. Meanwhile, most of the core provisions of the 1996 welfare and illegal immigration reform laws limiting legal immigrants’ rights remain in force. And though proposals to restore food stamps to noncitizen parents are pending in Congress, current debates suggest that legislators have not yet taken the reality of mixed status families to heart.

Strong economy. No doubt some of the more inclusive political actions of the past two years could be at least partially ascribed to the sustained economic expansion — an expansion that is not only creating new demand for immigrant workers, but may be allaying some natives= concerns about their own job security.

Deeper knowledge base. Another reason to begin discussing immigrant family integration is a broad, recent expansion in family and integration-related scholarship. Some examples include the Urban Institute’s own National Survey of American Families, the recent work by the National Research Council on the health and well-being of children in immigrant families and the work of Mary Waters, John Mollenkopf and Philip Kasinitz on second generation immigrants in New York City . The extensive research of Rubén Rumbaut and Alejandro Portes has contributed substantially to the literature in this area. This scholarship has deepened our understanding of immigration and its impacts, and we can expect a further expansion with the release of the 2000 Census and as other data sets with large immigrant samples become available.[2]

Policy targets: PRWORA and ESEA reauthorizations. The timing of this discussion of immigrant integration may also be right for political and policy purposes. Almost all federal aid programs for elementary and secondary education are scheduled to be reauthorized under the Elementary and Secondary Education Act (ESEA) in the 106th Congress. And the 1996 Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) — which had far-reaching impacts on the membership of legal noncitizens — is due for reauthorization in 2001. That reauthorization may be an opportunity for a first-principle discussion of the rights and entitlements of noncitizens. In addition to revisiting immigrant eligibility issues, the reauthorization of PRWORA offers a chance to address a number of implementation issues related to immigrant families and TANF. These include whether immigrant families have access to English as a Second Language, child care and transportation assistance, and whether public and private agencies providing this assistance have the linguistic and cultural capacity to serve immigrant families. Reauthorization may also address whether immigrant families are more likely than others to remain on the welfare rolls.
III. HOW IS THE IMMIGRANT FAMILY CHANGING?

Promoting family unity has long been the main imperative of U.S. immigration policy. About three-quarters of all immigrants coming to the U.S. enter through family unification channels, as close relatives of U.S. citizens or permanent residents. What is not widely understood, however, is the degree to which employment and diversity immigration is family driven as over half of all immigrants entering under these categories are actually the spouses and children of the primary beneficiaries.[3] Taken together, approximately 80 percent of all immigrant admissions in FY1998 entered to join family members in the United States or came as part of a family.[4] Family unity also plays an important role in humanitarian admissions as refugees with family members in the United States are given special preference. In fact, most refugees coming to the United States are joining family members.

While a detailed treatment of the flows of immigrant families to the United States is beyond the scope of this paper, we sketch below several immigration trends that point to the need for a more focused discussion of the integration of immigrant families.

High flows. The obvious place to begin is the fact that immigration to the U.S. is quite high. Annual immigration flows have tripled over the past generation with more immigrants entering the U.S. during the 90s than during any other decade (see figure 1).

Figure 1. Current Levels Are High


Source: Urban Institute Tabulations based on Immigration and Naturalization Service Data

Not only has the number of immigrants risen substantially, but the share of the total U.S. population the foreign born represent — now roughly 10 percent — has almost doubled since 1970. That said, the share remains substantially below the 15 percent that the foreign born represented at the turn of the last century ( see figure 2).

Figure 2. Immigrant Numbers at Peak – Percentage is Not


Source: Decennial Censuses and Current Population Survey

Dispersal. Over the past decade these high numbers have meant that immigration=s impacts have been felt beyond the six states that have been traditional receiving areas for immigrants. As figure 3 shows, during the 1990s, the immigrant population in what we term “new immigrant states” grew twice as fast (61 versus 31 percent) as the immigrant population in the six states that receive the largest numbers of immigrants.[5]

Figure 3. Concentration is High, But New Centers Emerge in ‘90s


Source: Urban Institute Tabulations of 1990 Census and March 1999 Current Population Survey

It could be argued that immigration and integration politics have been largely centered on the six traditional receiving states. But now, with greater numbers of immigrants settling in states without the experience or infrastructure to promote incorporation, integration issues may attract broader political attention. Further, it seems that the tensions and successes surrounding the settlement of immigrant families will be increasingly reported and judged through the lens of nontraditional receiving areas= experiences.

What is driving this dispersal? According to a recent Urban Institute analysis, the states to which immigrants migrated during the period 1995 to 1999 were not the states that extended more generous benefits to noncitizens following welfare reform. Moreover, some of the most generous states — most notably California — show a substantial net loss in immigrants during the period. In short, the analysis raises questions about the welfare magnet theory since immigrants’ settlement patterns appear to be driven more by the strength of local economies than the generosity of state welfare schemes.[6] (Passel and Zimmermann, 2000).

Changing origins and rapid rise in Mexican migration. Who is coming to the United States? It is widely recognized that the national origins of immigration flows have changed dramatically over the past thirty years — shifting from primarily European to Asian and Latin American sources (see, figure 4). But the degree to which Mexico accounts for recent flows may be less widely appreciated.

Figure 4. Most Legal Immigrants from Latin America and Asia


Source: Urban Institute based on Immigration and Naturalization Service data

As figure 5 indicates, Mexicans today represent almost 30 percent of both the total immigrant population and the annual immigrant flow. Between 1990 and 1999 alone, the Mexican population rose by 70 percent, from 4.3 to 7.2 million persons, making Mexico the single largest source of both legal and undocumented immigration. In fact, the Mexican population in the United States has almost doubled in the past decade; it has quadrupled since 1980; and has grown ten-fold since 1970.

Figure 5. Rapid Growth of Mexican Population


Source: Decennial Censuses and Current Population Survey

Increased share of undocumented. One corollary of increasing Mexican immigration is increased undocumented immigration. In 1994, 3 percent of the nation=s foreign born population was undocumented (Fix and Passel, 1994). By 1998, that share had risen to 22 percent (see, figures 6) and the total number of undocumented immigrants residing within the United States — 6 million — exceeded the highest estimates of the population=s size before enactment of the 1986 Immigration Reform and Control Act (see, figure 7). Further, it is estimated that anywhere from a quarter to a third of the current annual immigration flow is undocumented, returning the nation to patterns that held before IRCA=s legalization program in 1986.

Figure 6. Legal Status of Immigrants


Source: Urban Institute

Figure 7. Undocumented Population Returns to Pre-IRCA Levels


Source: Urban Institute

One large difference, though, between the current and prior policy contexts is the new restrictions imposed on undocumented immigrants= ability to adjust status following illegal immigration reform. As a result, the stock of the undocumented population may well grow faster than in the past as fewer illegal immigrants are able to convert to legal status. These developments are likely to exert continuing pressure to enact an IRCA-like legalization program as recently proposed by the AFL-CIO.

Although many imagine undocumented immigrants to be single adults, a 1998 Urban Institute study found that half of all undocumented immigrant-headed households in New York contain children.[7] Growing illegal immigration therefore raises the question of how an immigrant family integration agenda should take into account undocumented nonmembers and their mostly citizen children (see below).

Large Share of Immigrants in Families. The importance of using the family as a lens on immigrant integration is underscored by the fact that households headed by noncitizens are significantly more likely to contain children than those headed by citizens (55 versus 35 percent). Along similar lines, 44 percent of those living in foreign-born headed households are in families with children versus 33 percent of those in U.S. born-headed households (see figure 8).


Figure 8. Household Type by Nativity of Household Head: 1998

Predominance of mixed status families. The demographic phenomenon that holds perhaps the most far-reaching implications for an integration agenda is mixed status families. According to the census, 85 percent of immigrant families with children are mixed legal status families — i.e., families where at least one parent is a noncitizen and one child is a citizen. The metrics of mixed status families are striking:

< Nationwide, 1 in 10 U.S. children lives in a mixed status family (see figure 9);

< 75 percent of all children in immigrant families (those headed by a noncitizen) are citizens;

< 27 percent of all children in New York City, and 47 percent of all children in Los Angeles, live in mixed status families;

< In New York, 70 percent of families with children headed by undocumented immigrants contain citizen children.

Figure 9. “Mixed” Families Are Common


ource: Fix and Zimmermann 1999

Figure 10. Large Share of Low-Income Families with Children are Mixed


Source: Fix and Zimmermann 1999

Mixed status families are not just demographically important; they should be of central concern to social welfare policy, as:

< 15 percent of all poor children nationwide (i.e., under 200 percent of poverty) live in mixed status families (see, figure 10);

< 60 percent of the poor children in Los Angeles, and 30 percent of New York=s poor children live in mixed status families;

< 21 percent of poor uninsured children nationwide and fully one half of uninsured children in California live in mixed status families.

As we discuss later, these mixed status families present design challenges for policy-makers who seek to ration rights or benefits on the basis of citizenship status. On the one hand, the imposition of benefit restrictions for noncitizens tend to spill over to their citizen children. On the other hand, policies intended to extend benefits to noncitizen children are limited in their reach because most children in immigrant families are already citizens.

IV. HOW WELL ARE IMMIGRANT FAMILIES FARING?

Despite recent concerns about a decline in immigrant quality and slowing incorporation rates, our analyses suggest that the immigrant family integration story remains a largely successful one. However the data also reveal several emerging trends that are cause for concern.

We should say a word about how we are approaching the elusive term immigrant "integration." In this regard, we would only note that the data we present focus less on cultural measures of integration than on measures that we consider to be more directly correlated with economic and social mobility. We should also note that we use the term Aintegration@ not Aassimilation@ to reflect our expectation of continued diversity not homogeneity.

We recognize that integration is not simply a function of the traits and efforts of the immigrant family. It is also a function of the context in which newcomers find themselves, including the economic, political and demographic trends that characterize the nation at the time of entry. Integration’s pace is also influenced by the fiscal and other strengths of the receiving community and its institutions such as schools.[8] And, of course, integration is influenced by the receptivity of the community to newcomers and the degree to which the community itself changes in response to migrant flows.

Finally, rather than draw our measures of integration from the rich store of data and analysis that have been developed over the years,[9] we focus largely on data recently developed by the Urban Institute that have not been widely reported. We believe that the data highlight important recent trends that should help frame a discussion about immigrant family integration.

The reader should also be aware that the measures of integration that we include are incomplete, as we have omitted a number of key issues (civic and political participation and homeownership, e.g.). We have also not yet included a number of measures that could be used to gauge the receptivity of the receiving community to immigrants.[10]

Change across generations. Jeffrey Passel and Jennifer Van Hook recently analyzed cross-generational integration trends. They explore differences between first generation, one and a half generation (i.e., immigrants who arrived in the United States before they were 10 years old), and second generation (i.e., U.S.-born children to one or more foreign born parents) 20 to 30 year olds captured by the 1995 to 1998 Current Population Surveys. They find that by the second generation immigrants overall end up doing as well as, or in some instances, better than third generation non-Hispanic white natives[11] in terms of their:

< educational attainment;

< labor force participation;

< wages; and

< household income.

There is, however, substantial divergence across the ethnic and racial groups that compose the immigrant population, with Asians typically doing better than whites, Hispanics doing worse, and black immigrants experiencing more mixed outcomes. However, the analysts find no difference in wage outcomes across racial and ethnic groups in the second generation when education is standardized. Overall, then, these trends appear to us to represent positive, hopeful results.

But a different picture emerges from an analysis of trends in family formation and cohesion across generations. On the one hand, we see high intermarriage rates among immigrants of all races and ethnicities. But at the same time we see that immigrants= cross-generational gains and economic integration are paralleled by an all-too-American pattern of immigrant family disintegration. Passel and Van Hook find that though first generation families are less likely than natives to be divorced, the share of immigrant families that are divorced or separated doubles from the first to the second generation, equaling the rates of non Hispanic white natives. Along similar lines, they find the share of unmarried parents also rises rapidly from the first to the second generation, with the second generation’s rates exceeding white natives by a third. (In each instance, the family dissolution and single parenthood rates of immigrants lag substantially that of native blacks.)

We see similarly negative cross-generational integration patterns regarding child health. A 1998 report by the National Academy of Sciences that examined a wide range of child health outcomes found that children in immigrant families Aare healthier than U.S.-born children in U.S.-born families.@ However, the report went on to conclude that A(T)his relative advantage tends to decline with length of time in the United States and from one generation to the next.@ (National Research Council, 1998; Rumbaut, 1999).

Language acquisition among school-age children. Cross-generational analyses of language acquisition among school-age children reveal more positive trends. There is a rapid, if expected, decline from the first to the second generation in the share of children that are limited English proficient. Further, LEP status varies quite widely among populations whose native language is not English. Hence we see that in both the first and second generation, Mexicans are twice as likely to be LEP as Asians.

Household income growth. Turning now to other types of integration measures, we find that the incomes of households headed by legal permanent residents who have lived in the U.S. for 10 years or more slightly exceeds that of natives (see, figure 11 ). The result owes to both rising incomes in the period following settlement and the fact that legal permanent residents= households are larger than natives= and contain more earners. The incomes of undocumented immigrants remain well below those of natives and rise little with time in the U.S.

Figure 11. Integration is Dynamic


Source: Urban Institute based on March 1997 Current Population Survey

Benefiting from the boom economy. It appears that, at least in some respects, immigrant families have benefited strongly from the recent economic boom. As figure 12 indicates, immigrants= unemployment rates fell faster than natives= from 1996 to 1999. Particularly steep declines are evident among foreign born Hispanic males. Despite the decline, immigrants= unemployment rates remained higher than natives= in 1999 (although they were substantially lower than native blacks).

Figure 12. Immigrant Unemployment Rate Declines


Source: Urban Institute Tabulations of March 1996,1999 Current Population Survey

Our analysis of wage trends during the same period (see, figure 13) tells a different story. Here we see that between 1996 and 1999 natives= median wages rose over 50 percent faster than immigrants=. In short, the current economic boom appears to have produced more significant employment than wage gains for immigrants.

Figure 13. Immigrant Wages Rise Slower


Note: Numbers are adjusted to 1999 dollars.

Source: Urban Institute Tabulations of March 1996,1999 Current Population Survey

A look at employees= health insurance rates confirms the fact that while immigrants may be finding jobs, the jobs they are finding are worse than natives=. We find that immigrants are less likely to hold jobs that carry employer-provided health insurance than natives and that the gap widened between 1996 to 1999 (see, figure 14).

Figure 14. Employer-Provided Health Insurance for Immigrants Decreases


Source: Urban Institute Tabulations of March 1996,1999 Current Population Survey

Increased naturalizations. One traditional measure of immigrant integration is naturalization. As figure 15 indicates, there has been a surge in petitions for naturalization as well as approved applications in the wake of IRCA=s legalization program, Proposition 187, and the 1996 welfare and illegal immigration reforms. These rapid increases are particularly prominent among some national origin groups that historically have shown little propensity to naturalize. To illustrate, in 1992 only 16 percent of Mexicans who had been legally admitted in 1977 had naturalized. Five years later, that share had doubled to 32 percent.[12] Colombians evidence similar patterns as 36 percent of 1977 entrants had naturalized by 1992. By 1997 however, 61 percent of the 1977 cohort had become citizens.

Figure 15. Naturalizations Increase in Wake of IRCA


Source: INS Statistical Yearbook

These rising naturalization rates can be ascribed to at least four phenomena:

< IRCA=s legalization of almost 3 million formerly undocumented immigrants;

< Immigrants= reactions to the political environment of the early, mid, and even late 1990s: symbolized by the broad voter approval of California=s Propositions 187 and 227;

< Greater tolerance of dual citizenship both within the United States and sending countries such as Mexico;

< The increased instrumental value of citizenship following welfare and illegal immigration reform. These greater Areturns to citizenship@ include expanded eligibility for public benefits, especially for noncitizens entering after August 22, 1996.[13]

The first two (legalization, fear-based responses) might be viewed as more or less one time or at least episodic events. The latter (dual citizenship, greater returns to naturalization), can be seen as the products of more enduring changes that should have a continuing, longer term effect on immigrants= increased propensity to naturalize. Overall, these differing sources of rising naturalizations raise the question whether naturalization in the post-Proposition 187/welfare reform era should be viewed as a metric of integration in the way it once was.

Growing numbers of naturalized immigrants will also have positive impacts on immigrant families. Although the number of immediate relatives of permanent residents who can be admitted in a given year is capped, there is no limit on admissions of citizens’ spouses and children. Hence we could see an increase in this type of family-related immigration flow.

Increased poverty among school-age children. In the first place, we see a sharp rise in poverty among the children of immigrants. As recently as 1970, poverty rates among immigrant children only narrowly exceeded non-Hispanic whites=. But by 1997, their poverty rate was more than double that of non-Hispanic whites=, rising from 17 to 39 percent (see, figure 16). Further, from 1970 to 1995 the overall child poverty rate rose from 14.7 to 20.4 percent. Roughly 60 percent of this growth in child poverty could be ascribed to the children of immigrants (Van Hook and Fix, 2000). (Much of this increase in immigrant poverty is probably due to the large recent growth in low-income Mexican and undocumented immigration that we discuss above.)

Figure 16. More Immigrant Children are Poor


Source: Van Hook and Fix 2000

Segregation of LEPs in schools. Another troubling trend we see among children in immigrant families is their segregation within schools. As figure 17 reveals, we have found disturbing trends in the nation=s schools, where half of limited English proficient children attend schools where a third or more of their fellow students are also LEP. This means that they are going to schools that are not just ethnically and economically segregated, but are linguistically isolated.

Figure 17. LEPs are Linguistically Segregated


Source: Van Hook and Fix 2000.

High drop out rates. This spatial segregation may be partially responsible for the high drop out rates that we see among immigrant children. Immigrant students= drop out rates exceed those of native students. While drop out rates for the second generation are lower than the first, they turn upward again for the third generation. (In constructing these drop out rates we have only included children who attended school in the U.S., i.e. those who have chosen to drop in at some point.) Mexican dropout rates for each of the first, second, and third generations are roughly double the national average. By contrast, first generation Asians drop out at a rate that is less than a quarter of the average for all foreign-born immigrants.[14]

High but rapidly falling rates of welfare use. We conclude this discussion by exploring recent trends in immigrant families’ use of public benefits. When we compare across all households, we see that noncitizen households were more likely than citizens’ to receive welfare in 1994, before welfare reform (13.9 versus 7.8 percent). They were also more likely to receive benefits in 1997, following welfare reform=s enactment and early implementation (9.0 versus 6.7 percent).[15] (See, figure 18).

Figure 18. Immigrant Welfare Use Declines Faster than Citizens’


Source: Fix and Passel 1999

However, the picture changes when the presence of children and poverty are taken into account. First, noncitizens= households are substantially more likely to contain children than citizens= (55 versus 35 percent). Second, noncitizen households are substantially more likely to be poor (i.e. under 200 percent of poverty) than citizens= (54 versus 31 percent). When we restrict our analysis to poor households that contain children, we find that noncitizen households used fewer benefits than citizens both before and after welfare reform. In fact, following reform, noncitizens= use of benefits is roughly half as high as citizens= (14.0 versus 25.8 percent) (See, figure 19).

Figure 19. Low Income Immigrant Families with Kids Use Less Welfare

Source: Fix and Passel 1999.

These results make plain that welfare reform resulted in steeper caseload declines between 1994 and 1997 among noncitizen families than citizen families. The rapid decline likely owes less to increased eligibility restrictions than to welfare reform’s chilling effects, stemming from confusion, concern about the consequences of using welfare, and, possibly, misinterpretation of the rules. We find that:

< Noncitizens= use of welfare (i.e. TANF/AFDC, SSI, G.A.) fell by over a third, from 13.9 to 9 percent while citizens= use of benefits fell much less: by 14 percent (see figure 20). As a result, immigrants accounted for a disproportionate share of overall caseload decline. That is, noncitizens accounted for 23 percent of the total caseload decline while they represented only nine percent of U.S. households.

< Between 1994 and 1997 we see equally steep declines in Food Stamps and Medicaid, despite the fact that during the period examined some noncitizens lost eligibility for Food Stamps but very few lost Medicaid eligibility: suggesting that eligibility changes in one program may affect use in others.

< Declines among refugees were as steep as among legal immigrants — again, despite the fact refugees were completely shielded from welfare reform=s restrictions during the period.

< We find that naturalizations did not account for the differing rates of decline between citizens and noncitizens and that income growth explained very little of the drop in benefit use among immigrants during the period.

Figure 20. Nationwide Chilling Effects


Source: Fix and Passel 1999