Federal Policy for Immigrant Children: Room for Common Ground?

Mark Greenberg, Ron Haskins, and
Ron Haskins Senior Fellow Emeritus - Economic Studies

Shawn Fremstad
Shawn Fremstad Director, Inclusive and Sustainable Economy Initiative

June 1, 2004


Policymakers and analysts agree on the need to improve the well-being of children in immigrant families in the United States—for example, in the areas of public benefits, education, and economic mobility—but disagree about how to address the problems. The authors of this policy brief are no exception. Ron Haskins, Senior Fellow at the Brookings Institution and Senior Editor of The Future of Children, seconds the decision of Congress in the 1996 welfare reform law to make noncitizens ineligible for public assistance and Medicaid. He emphasizes the need to tie public benefits for immigrant families to work through such policies as education and training and the earned income tax credit for families with children. Mark Greenberg, Director of Policy at the Center for Law and Social Policy, and Shawn Fremstad, Deputy Director of the Welfare and Income Support Division at the Center for Budget and Policy Priorities, argue that noncitizen families should have the same eligibility for public assistance as citizen families and support greater financial aid for early childhood education and other forms of schooling. The hope of all three authors, however, is that researchers and public officials will continue to search for common ground to improve life for children of immigrant families, most of whom will grow up as Americans.


In 2000, about 11 percent of the U.S. population—32 million people—was foreign-born. Many of these immigrants are parents of children who are either immigrants or, more commonly, U.S. citizens themselves. As a result, one in every five children in the United States now lives in an immigrant family.

Children from immigrant families will play an important role in the nation’s future. These families have accounted for large shares of U.S. population and employment growth and will continue to do so. Most children of immigrants will be lifelong U.S. residents and will become part of the nation’s workforce. Some will serve in the U.S. military; some will be manual workers; others will become teachers or doctors. These children will play a crucial role in the viability of Social Security and Medicare—in fact, the Congressional Budget Office reports that the Social Security shortfall will increase if immigration falls below current levels. Even apart from the humanitarian reasons that some observers see as an adequate basis for attention to the circumstances of immigrant children, policies that help these children become successful adults are squarely in the national interest.

Although the United States is the prototypical “nation of immigrants,” it is worth noting that increased migration is a global trend. Most other developed nations are seeing similar increases in immigration; immigrants comprise more than 15 percent of the population in more than 50 counties and account for large proportions of population and employment growth in most developed countries. Some of these countries, Canada in particular, have done more than the United States to develop “immigrant integration” policies to maximize the economic and social benefits of immigration.

This is a favorable time for seeking consensus on U.S. policies affecting immigrant children. Even though issues such as policy toward undocumented immigrants and English-only requirements can be divisive, immigration generally has not been a major electoral issue in recent years. Public opinion research shows that the share of Americans who think that immigration is good for the nation increased in the year following 9/11. Moreover, about four out of every five Americans think that lawfully admitted immigrants should be treated the same way as U.S. citizens under the law. The relative lack of political intensity on immigration—particularly on policy that relates to the treatment of immigrants who have already been legally admitted to the United States—provides an opening to consider appropriate policies.

Children in Immigrant Families

Children of immigrants face several difficulties that children of native-born parents do not, including adapting to cultural norms that may differ from those of their parents and learning a language that may not be spoken at home. The poverty rate of children in immigrant families is 21 percent, as against 14 percent for children in native-born families. Nearly half of children in immigrant families have family incomes below 200 percent of poverty compared with only 34 percent of native children. Given the evidence that poverty is detrimental to the development of children, poverty among children of immigrants is reason for concern. Children of immigrants are more likely than children of native-born parents to be in families worried about or encountering difficulties paying for food. They are nearly twice as likely to lack health insurance, and they are more than four times as likely to live in crowded housing.

Research by Don Hernandez has shown that children of immigrants are at increased risk for poor developmental outcomes. Hernandez has calculated the share of children exposed to four risk factors—having a mother without a high school diploma, being economically deprived, living in a linguistically isolated household, and living in a single- parent family—that correlate with poor development. Some 67 percent of children born to immigrants had one or more of these risk factors and 17 percent had three; only 45 percent of children born to native parents had at least one risk factor and only 4 percent had three.

Despite their higher poverty and hardship rates, low-income children in immigrant families are less likely to receive public assistance than other low-income children. Although citizen children of immigrant parents are not subject to any of the public assistance eligibility restrictions that apply to noncitizens (see below), they are less likely to receive income assistance and food stamps than citizen children with native-born parents. In 1999, only about 8 percent of low-income families headed by lawful permanent residents received income from Temporary Assistance for Needy Families (TANF), compared with 12 percent of low-income citizen families. Similarly, low-income noncitizen children are about half as likely to participate in Medicaid as low-income citizen children. This gap existed before passage of the noncitizen eligibility restrictions in 1996, but has widened in recent years.

What Accounts for Higher Poverty In Immigrant Families?

Children in immigrant families have a high rate of poverty largely because their parents are much more likely to hold low-wage jobs than other parents. In 2002, nearly half of foreign-born workers were low-wage workers. Randy Capps has found that immigrants account for a disproportionate share of the nation’s low-wage workforce—14 percent of all workers, but 20 percent of all low-wage workers. And immigrants are about 50 percent more likely than natives to earn less than the minimum wage.

Poor education and weak English proficiency both contribute to immigrants’ lower wages. Among immigrant lowwage workers, most (62 percent) have limited proficiency in English, and nearly half have not completed high school. Controlling for differences in education and other factors, immigrants who are proficient in English earn about 17 percent more than immigrants who are not. Other factors such as discrimination, geographic and linguistic segregation, and lack of bargaining power also likely affect immigrants’ earnings.

Marital status does not appear to contribute to immigrants’ higher poverty rate. Low-income immigrants are more likely to be married than low-income natives. Hernandez has shown that only about 15 percent of immigrant children live with a single parent, as against 26 percent of native children (although the share of two-parent immigrant families declines in subsequent generations). Moreover, the Urban Institute has found that among children in two-parent families in 1999, those in immigrant families were three times as likely to be poor as those in native-born families—18 percent as compared with only 6 percent.

Nor are immigrants’ higher poverty rates attributable to low employment rates. Immigrant fathers are just as likely to work full-time, year-round, as are native fathers. About 80 percent of both groups of fathers are employed fulltime. Thus, policies designed to increase labor force participation and hours of work are unlikely to have a significant impact on immigrant poverty.

Public Benefit Policy

The 1996 welfare reform law made most legal noncitizens ineligible for TANF and Medicaid during their first five years in the United States. It also restricted their eligibility for food stamps and supplemental security income (SSI). The restrictions directly affected the eligibility of children who are not citizens and also likely affected citizen children of immigrant parents. In some of the programs, particularly food stamps, their participation fell at a much faster rate than that of citizen children of native-born parents. Four developments softened the effects of the 1996 provisions. First, most states chose to use state funds to restore some or all benefits. Second, only a few states elected to deny TANF and Medicaid to noncitizens for more than five years. Third, citizenship rates increased in the 1990s. And, finally, Congress lifted several of the original restrictions.

Even so, there is little question that the restrictions increased hardship for many noncitizens. George Borjas found that “food insecurity” increased among noncitizens in states that did little to provide replacement benefits, while it declined among noncitizens in other states. Leighton Ku and Matt Broaddus found that noncitizen children living in states without state-funded health care programs for noncitizens are much less likely to have health insurance than noncitizen children living in states that have such programs and that the difference is attributable to lower rates of public coverage (private coverage rates for noncitizen children are similar in the two groups of states).

The authors of this brief do not agree on whether legal noncitizens should have the same eligibility for public assistance as citizens. Haskins largely supports the current restrictions. He believes that the long-standing public charge provision in immigration law, which bars entry to immigrants without means of support, is the correct policy and is consistent with the restrictions. If noncitizens fall on hard times before they gain citizenship, many have sponsors who have signed a legally binding document requiring them to provide support. Thus, the primary line of defense against destitution for noncitizens should be sponsors.

Along with others who oppose welfare eligibility for noncitizens, Haskins holds that America offers immigrants one of the best bargains in the world: immigrants who come to America are privileged to enjoy vast individual freedom, to live in a society governed by principles of law and moral behavior, and to join one of the world’s most prosperous economies. All the nation requires in return is that noncitizens obey the law and avoid receiving certain federally funded public benefits until they become citizens.

Greenberg and Fremstad argue that legal noncitizens should qualify for public assistance on the same terms as citizens. Immigrant children should have the same access to public programs that reduce hardship and improve life chances as the citizen children they sit next to at school. If it is in the national interest for all children to be ready for school and grow up to be the most productive citizens they can be, it is shortsighted to deny noncitizen children the same access to health care, adequate nutrition, and stable housing as citizen children. Greenberg and Fremstad agree that sponsors should assist the immigrants they sponsor, but they note that public assistance and private sources of support are not mutually exclusive for citizens and should not be mutually exclusive for legal immigrants.

Greenberg and Fremstad also disagree with Haskins’s interpretation of the public charge provision, which they believe offers little historical support for the post-1996 restrictions. Treasury Department documents from the period when the public charge provision was first enacted describe a public charge as someone who is unable to “earn a living” because of “accident, bodily ailment, or disease, or physical inability.” Similarly, case law from the nineteenth century holds that “persons in full possession of their faculties, sound in body, neither paupers, vagrants, nor criminals, and in all respects competent to earn a livelihood” are not public charges. Thus, they believe the public charge provision has little applicability to the current public assistance system that helps families get back into the labor force after losing work and meet basic needs when employers fail to pay a living wage.

They also note that the principle that a nation’s laws should welcome immigrants by extending the same helping hand to them that is extended to citizens is one that compassionate conservatives should support. President Bush has proposed restoring food stamps and SSI benefits for certain immigrants, has given states the option to extend state child health insurance (SCHIP) prenatal care benefits to undocumented women, and has called on the nation to “build a culture of life in which the sick are comforted, the aged are honored, the immigrant is welcomed, and the weak and vulnerable are never overlooked.” Even the Center for Immigration Studies, which sides with Haskins on most other immigration policy issues, argues that legal immigrants who are required to pay taxes and meet other civic obligations should have the same opportunity to obtain public assistance as other taxpayers.

Greenberg and Fremstad urge a return to the traditional principle of equal treatment of citizens and legal immigrants that prevailed in public assistance programs before 1996. Haskins argues that a repeal of the federal restrictions would be wrong in principle, would be prohibitively expensive, and is unlikely as long as Republicans control either the presidency or at least one House of Congress.

Greenberg and Fremstad concede that an immediate lifting of all special immigrant restrictions is unlikely but believe that bipartisan support exists for measures similar in scale to Congress’s recent restoration of food stamp benefits for many legal immigrants, which expanded President Bush’s proposal to restore food stamps to certain legal immigrants. Congress could start by lifting the restrictions on TANF, which would cost the federal government nothing; restoring Medicaid for children and pregnant women (a measure that has passed the Senate with bipartisan support); and passing bipartisan legislation to extend SSI benefits for refugees.

Haskins argues for a more limited set of modifications that ties eligibility directly to work. He points out that federal law already reflects the policy of helping immigrant workers. The 1996 reforms explicitly declared legal immigrants eligible for a host of programs that provide education and training. Even more important, parent workers who are legal residents of the United States and authorized to work also are eligible for the earned income tax credit (EITC), a wage supplement paid to low-income workers through the tax code. Increasing federal EITC payments (which provide up to $2,500 for one child or up to $4,200 for two or more children, depending on income), encouraging more states to offer their own EITC payments, and making sure eligible immigrant families receive the payments would increase the financial security of immigrant families.

Haskins offers several other ways to reward work by immigrant parents. States, at their option, could be permitted to use TANF dollars to provide any of a broad range of work supports to legal immigrants who are ineligible for TANF, Medicaid, and food stamps (though Haskins would still oppose providing cash assistance, even to those working). An additional step would be to make the noncitizen children of working immigrant families eligible for Medicaid and SCHIP.

Greenberg and Fremstad view the Haskins proposal as inconsistent with the traditional principle that government should not have a separate, stricter set of rules for legal immigrants as long as immigrants have the same obligation to pay taxes and meet other civic duties as citizens. In addition, they note that Congress rejected tying food stamp eligibility to work during the debate on the food stamp restorations in 2002 and that there is no precedent for having separate work requirements that apply only to immigrants. The Haskins proposal also would have the effect of maintaining ineligibility for the noncitizens, such as victims of domestic violence and people who are temporarily incapacitated, who could most benefit from help in transitioning to work. Fremstad and Greenberg believe that public assistance should help immigrants and citizens alike when they lose jobs and need assistance to reenter the workforce and meet their basic needs.

Education Policy

Three policies could address the educational deficits of immigrant children. First, achieving the goal of equality of educational opportunity requires good public schools. Many immigrant children attend underperforming schools, and even better-performing schools may not have effective programs for English-language learners. Under the No Child Left Behind (NCLB) law, assessment results and state progress objectives must be broken out by student groups based on poverty, race and ethnicity, disability, and limited English proficiency. These and other provisions of NCLB hold promise for immigrant children, though Greenberg and Fremstad, along with many in immigrant communities, have serious concerns that insufficient federal funding may limit NCLB’s effectiveness.

A second policy to help immigrant children is early childhood education. Research shows that high-quality preschool programs can have lasting effects on school performance. But three- and four-year-old children in immigrant families are less likely than children in native-born families to participate in nursery school or preschool programs. Moreover, little is known about the quality of the programs they attend or about the extent of participation by younger children in immigrant families.

The authors agree about the need to expand early childhood education programming, but disagree about how to get there. Haskins argues that federal and state governments already spend more than $25 billion on preschool education and child care programs. He proposes giving states more flexibility in the use of all sources of federal funds for preschool programs if they agree to provide at least one year of high-quality preschool to all four-yearolds (including immigrants) with family income under, say, 125 percent of poverty. States would also be required to put up matching funds, to coordinate their programs with the public schools, to guarantee parent choice in selecting a preschool, and to evaluate the effects of their program. Congress should appropriate at least $100 million annually to provide additional funding to participating states.

Greenberg and Fremstad respond that existing early care and education programs are seriously underfunded and that a larger federal and state fiscal commitment is needed to assure that quality programs are available to children in immigrant families as well as other children. Child care funding is not a dedicated funding stream for preschool programs, and subsidies reach only a small fraction of eligible families. The picture has deteriorated in recent years as federal child care funding has remained flat while other federal and state sources have become less available. Simply allowing more discretion in the use of existing funds will mean, at most, increased services for some groups at the expense of reduced services for others. Further, there is a risk that broad discretion in state use of the funds would be paired with opportunities for states to avoid federal performance standards and other requirements. To avoid such results, a federal initiative to provide support for state prekindergarten programs should be combined with expanded support for care and education for both older and younger children. In efforts to promote language acquisition and immigrant integration, a national policy of waiting until a child reaches age four has no sound basis. Further, given that immigrant children are less likely to receive health, nutrition, and other needed services, any federal initiative must be structured to ensure that disadvantaged families have access to comprehensive services.

A third policy involves English proficiency. Most immigrant children live in homes in which the primary language is not English. If school instruction is conducted in English, those who do not speak English well can do poorly in many subjects. A debate has raged for decades about whether immigrant children should be taught in their native language or English. Because most Americans believe immigrants should learn English quickly, all questions of government support for official use of foreign languages to help non-English speakers face public opposition. Indeed, California’s ballot initiative on ending bilingual education in 1998 was approved by 61 percent of the voters.

Even here, there may be room for compromise. Robert Slavin and Alan Cheung’s recent review of rigorously evaluated programs found that bilingual reading programs paired with simultaneous English instruction produced the most rapid learning. Instructional language was an important factor, but the most decisive factors were phonics use, one-to-one or small group tutoring, and extensive reading. Thus, improving the educational practices of the schools attended by immigrant children—regardless of instructional language—should be the primary concern of those seeking to improve English proficiency.

Improving the Earnings Potential of Immigrant Parents

One important way to improve the living conditions of immigrant children is to raise the earnings of their parents by expanding access to higher education and promoting labor force advancement. In addition, Greenberg and Fremstad believe that providing legal immigrants with the same TANF eligibility as citizens would improve immigrant parents’ access to employment services and that allowing English language instruction to count toward TANF participation requirements would increase English proficiency. Haskins agrees with this recommendation as long as TANF funds are not used for cash benefits.

Changes to federal law could improve both access to and the quality of training and other workforce services for limited- English-proficient workers. Such reforms include:


structuring workforce-system performance standards so that they do not discourage providing services to limited-English-proficient persons;
encouraging the development of integrated training programs that combine job training and language acquisition; and
encouraging one-stop centers to structure career counseling, vocational assessment, and other services to meet the needs of limited-English-proficiency customers.


We share the belief that America faces an important challenge: how to improve the development and well-being of children in immigrant families. We agree about some policies for helping these children: more support for working families, expanded early education opportunities, and greater efforts to address English language acquisition. We disagree about other policies, notably, whether to restore the pre-1996 rules that provided immigrants with the same public-benefits eligibility as citizens and whether to give states broad discretion over federal child care and early education funding. We hope that in light of the importance of these issues for immigrant children and for the nation’s future, there will be renewed efforts in Washington to find common ground.