Skip to navigation

Skip to main content


The Legacies of Welfare Reform's Immigrant Restrictions -- Michael Fix

The Legacies of Welfare Reform's Immigrant Restrictions

Michael Fix

Wendy Zimmermann

The Urban Institute

August 1998

Overview 1I. Long-term Legacies and Early Reversals 2II. Immigrant Eligibility in The Wake of Reform 6III. Beyond Eligibility: Welfare Reform's Normative and Institutional Legacies. 15IV. Conclusion. 27References 29

Michael Fix directs the Urban Institute's Immigration Studies Program. Wendy Zimmermann is a research associate in the Urban Institute's Population Studies Center. The opinions expressed here are those of the authors and do not reflect the views of the Urban Institute, its board, or funders. Support for this paper has been provided by the Andrew W. Mellon Foundation and the William and Flora Hewlett Foundation. The authors would like to thank Karen Tumlin for her insightful comments

The 1996 welfare reform law marked a sharp exclusionary turn in U.S. immigrant policy with new, unprecedented bars imposed on legal immigrants' receipt of public benefits. However, in the two years following passage of the law SSI and food stamps have been restored to some immigrants, while states and the federal government have generally sought to soften the law's impacts.

Despite these developments the welfare reform law remains a watershed in U.S. immigrant policy. Nearly three fourths of non-citzens who were receiving food stamps have not had their federal eligibility restored, including working age non-citizens. The loss of benefits will affect the total resources available to the families of these non-citizens. As one in ten American children lives in a mixed status family with at least one non-citizen parent and one citizen child, the restrictions' impacts will affect many poor citizen children. Immigrants entering the U.S. after the passage of the law on August 22, 1996 remain barred from receiving federally-funded Medicaid and TANF for 5 years and food stamps and SSI until citizenship.

Welfare reform also ushered in a number of normative and institutional changes that, when taken together, could limit immigrants' use of public benefits for which they remain eligible. Administrative data from Los Angeles County indicate that between January 1996 and January 1998 monthly approvals of AFDC/TANF and Medicaid cases for legal immigrants fell by 71 percent. During the same period, the number of monthly approvals for citizen cases did not change. These shifts in immigrant approvals occurred despite the fact that legal immigrants' eligibility for the benefits did not change in California. The results, then, could be read to suggest that welfare reform, along with related policy changes, is having a profound chilling effect on eligible immigrants' use of public services.

I. Long-term Legacies and Early Reversals
Like many Western industrialized nations, the United States has sought to reduce its welfare expenditures by restricting the benefits extended to immigrants. Few Western countries though, have reversed their immigrant eligibility policies as sharply as the United States. The 1996 Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) signaled a striking departure from an inclusionary social welfare policy that made legal immigrants eligible for public benefits on largely the same terms as citizens, to one that systematically discriminates against most non-citizens.

Yet welfare reform's immigrant restrictions have been substantially scaled back in the two years following enactment. In August 1997, one year after PRWORA's passage, Congress reversed course and restored SSI eligibility for most immigrants who were in the United States when the 1996 law was enacted. At the same time, the states -- which had been granted new authority to deny legal immigrants public benefits -- were taking the high road and extending Medicaid and cash welfare benefits to these immigrants. Then in June, 1998 the Congress voted to restore federal food stamp eligibility to immigrants who are elderly, disabled, or under 18 and were in the U.S. when the law passed.

Despite these reversals, welfare reform's immigrant provisions remain a watershed in social welfare policy because they leave a number of far-reaching, largely overlooked legacies. In the first place, the law still restricts non-citizens' and especially new non-citizens' eligibility for public benefits, representing a now-official policy of immigrant exceptionalism. Second, the law signals changing societal norms regarding immigrants' use of public benefits -- norms that are influencing the law's implementation and that appear to be affecting the willingness of immigrants to apply for benefits for which they remain eligible. Third, the law revolutionizes the governance of immigrant policy. It does so by devolving immigrant eligibility decisions to state and local governments; by imposing new verification and reporting requirements; and by fusing immigrant and social welfare policy, as issues of immigrant eligibility are explicitly layered onto all decisions to create or expand social programs.

Taken together, these legacies introduce deep structural changes in membership, social policy and federalism that will be hard to undo, even if the anti-immigrant attitudes and policies of the 1990s begin to recede.

Welfare Reform's Immigrant - Specific Provisions: An Overview. As enacted, welfare reform transformed immigrant policy by:

barring most immigrants from food stamps and Supplemental Security Income (SSI) -- cash assistance for the poor elderly and disabled. These include "current" immigrants who were already in the United States at the time the law was enacted, and new or "future" immigrants who had yet to enter.

barring new immigrants for five years from "federal means-tested benefits," defined to include Temporary Assistance for Needy Families (TANF) and Medicaid;

giving states the option of barring current immigrants (i.e. those in the U.S. on or before August 22, 1996) from TANF, Medicaid, and the Social Services Block Grant. The law also gave states the option of barring new immigrants (arriving after August 22) from TANF and Medicaid following the mandatory five-year bar;

exempting some legal immigrants with strong equities from the benefit restrictions. These include refugees during their first several years in the United States, legal immigrants who have worked for 10 years or whose spouse or parents have done so, and non-citizens who have served in the U.S. military;

barring "unqualified immigrants" from all "federal public benefits" and requiring the public agencies that dispense them to verify the legal status of applicants. Unqualified immigrants not only include undocumented immigrants but other groups with authority to remain in the United States without permanent residence.

Early Implementation: Mitigating the Impacts. To date, many of the choices made by welfare reform's implementers have sought to soften the law's potentially devastating effects on immigrants. Upon signing welfare reform into law, President Clinton announced that he wanted the immigrant provisions scaled back, and their subsequent history has largely - but not entirely - followed his lead.

The systematic softening of the law's impacts can be traced to several converging developments. First, it reflects the forging of strong new interest-group coalitions made up of immigrant advocacy organizations like the National Immigration Law Center as well as more broad-based policy groups such as the Center on Budget and Policy Priorities. Benefit restorations have also been driven by extensive media coverage of the initial impacts of SSI and food stamp cuts on immigrants. The policy reversals have drawn on a delayed Congressional recognition of the nation's changed demographics, and on bipartisan concern within high immigration states that the costs of federal restrictions would be passed through to state tax payers. And, of course, the restoration of federal benefits has been carried out against the economic back-drop of unexpectedly large federal budget surpluses.

How have the draconian immigrant measures originally enacted been softened? At the

federal level, Congress and the Administration:

restored SSI eligibility to all qualified immigrants who were receiving benefits when PRWORA passed and to those who were in the United States as of August 22, 1996 who eventually qualify as disabled;

restored food stamp eligibility to elderly and disabled immigrants and to legal immigrants under 18 who were in the U.S. when the law passed;

defined "federal means-tested public benefits" quite narrowly, limiting it to TANF, Medicaid, SSI, food stamps and now, apparently, the new state Child Health Insurance Program (CHIP). Early legislative proposals would have extended these bars to anywhere from 60 to 100 federal programs;

extended the period of time refugees are eligible for SSI, Medicaid, and food stamps from five to seven years after their arrival;

expanded the definition of qualified alien to include certain immigrants and their children who are the victims of domestic violence. Certain Cuban and Haitian immigrants and Amerasians were also made eligible for benefits on the same terms as refugees;

exempted private, non-profit organizations from verifying the immigration status of applicants for services or benefits; and

excluded several child-oriented public benefit programs from those defined as "federal public benefits," thereby exempting them from verification requirements. (Two such exempted programs are Head Start and the Maternal and Child Health Care program.)

At the same time, in what marks a striking development in fiscal federalism, many of the states made unexpectedly generous implementation decisions. Among other things, they developed their own food stamp replacement programs, usually for elderly or disabled immigrants and for immigrant children. Eighteen states, including all major immigrant-receiving states have so far enacted such programs. All states chose not to bar unqualified immigrants, including undocumented aliens, from the food supplement program for Women, Infants and Children (WIC) -- despite their new-found powers to do so. Finally, almost all states retained the eligibility of current immigrants for TANF, Medicaid and services provided under the Social Service Block Grant. In a few instances, states -- most notably California -- have even provided state-funded TANF and Medicaid to new or "future" immigrants who are barred from the federally funded programs for five years. They have done so despite the fact that the full costs of these programs are borne by state treasuries.

II. Immigrant Eligibility in The Wake of Reform
These efforts to blunt welfare reform's impacts aside, many immigrants have still lost eligibility for public benefits.

Food Stamps. Recent federal food stamp legislation restores eligibility to an estimated 250,000 non-citizens or roughly one quarter of the approximately 935,000 non-citizens who lost federal benefits. The restoration leaves out future immigrants altogether. In fact, this bill's $818 million price tag represents less than one-quarter of the estimated $3 billion, five-year savings from the immigrant food stamps cuts. As the Congressional food stamp proposal was modeled on the SSI restorations, it covers only those who were 65 or older when the law passed, those under 18, and the disabled. Since SSI is a program for the elderly and disabled, nearly all current immigrants regained their eligibility for SSI. But by following the SSI population "model," the Congress left out the largest group of needy immigrant food stamp recipients: working-age adults.

Further, most large immigrant receiving states were already providing state-funded food replacement programs to children and the elderly. While these states will certainly appreciate the federal government's resuming its payments, few immigrant families in these states will see an increase in their food stamps.

With the restoration of federal food stamps, the state food programs, in their current form, become a less important element of the safety net for legal immigrants. Still, federal food stamps for immigrants do not resume until November, 1998, and state programs offer a bridge in the interim. In addition, some states could shift the funds they are now spending on benefits for the elderly, disabled and child immigrant populations to working age non-citizen adults, as California has already done.

The state food replacement programs offer a window onto the post-devolution world of immigrant policy set in motion by welfare reform. A key fact here has been the limited reach of the state replacement programs: only about 200,000 of the 935,000 immigrants who were expected to lose eligibility for federal food stamps by August 1997 had their eligibility restored under the state programs.

Two explanations stand out for the fact that so few immigrants were covered by state-funded food benefits. First, like the federal restorations, many state programs left out working-age adults. Second, most (thirty-two) states had no replacement program. While the states without programs were not by and large the major immigrant-receiving states, immigration has been rising rapidly to non-traditional destinations. The number of immigrants settling in the 40 states with the smallest immigrant populations rose by 50 percent from 3.7 to 5.6 million between 1990 and 1996 alone. So the number of immigrants losing eligibility in these states is far from negligible.

Impacts on Mixed Families and Citizen Children. One little-noticed population edged out of the safety net by welfare reform's immigrant provisions is the U.S.-born citizen children of legal and undocumented immigrants whose households may see a decline in total resources. Contrary to popular belief, families do not fall neatly into two categories: those composed of citizens and those made up of non-citizens. According to the 1997 Current Population Survey (CPS), one in ten American children lives in a mixed status family: a family with at least one non-citizen parent and one citizen child. In places where immigrants are concentrated, the numbers are even more striking. (see Figure 1) Almost two-thirds of children in families with incomes below 200 percent of poverty in Los Angeles, and one-third in New York, live in such mixed-status families. Although citizen children retain eligibility, these families will have less money to spend on food when their non-citizen parents lose benefits. Neither the federal restorations nor many of the state food stamp replacement programs solve this problem since they do not cover working-age adults. As a result, immigrant-targeted food stamp cuts will be felt by more children -- and by more citizen children -- than the framers of federal policy probably intended.

Figure 1. Children in families of mixed citizenship*,

March 1997

*Here, a family of mixed citizenship is defined as one with at least one non-citizen parent and at least one citizen child.

Source: March 1997 Current Population Survey

But given that the non-citizen children of immigrants remain eligible for food benefits under state programs, why provide aid to working-age adults who, in the words of Texas Senator Phil Gramm, should be coming to America "with their sleeves rolled up and ready to work, not to get food stamps"? The fact is, many recipients are already working: about half of all non-citizen parents in food stamp households (as well as citizen parents) had worked during the past year.

Further, while the welfare reform law severely limits the food stamps that able-bodied childless adults can receive, those living in areas with high unemployment who are unable to find jobs retain their eligibility. In one of the clearest examples of discrimination against immigrants embedded in current policy, this exception is not made available to legal immigrants who lose their food stamps. They lose their benefits whether there is work to be found or not.

Impacts on New or "Future" Immigrants. Perhaps the brightest line drawn by welfare reform is not the one separating legal immigrants from citizens, but the line dividing non-citizens in the United States as of August 22, 1996 from the new immigrants arriving after that date.

Welfare reform barred most new legal immigrants from TANF and Medicaid during their first five years in the United States. New immigrants also remain ineligible for SSI and food stamps since the federal restorations did not affect this group. States are free, however, to use their own funds to provide benefits to these new immigrants arriving after welfare reform.

New immigrants' access to public benefits is further constrained by new sponsor-deeming requirements, under which the income of an immigrant's sponsor who signs an affidavit promising to support the immigrant, is deemed available to the immigrant when determining eligibility for benefits. In practice deeming usually works to effectively deny assistance. The new law extends the deeming period from 3 years until citizenship and expands deeming to Medicaid. Under previous law, deeming applied only to TANF, food stamps and SSI.

In contrast to their generous treatment of current immigrants, most states have followed the federal government's lead and not provided substitute benefits for new immigrants. Most of the comparatively small number of states that are providing TANF and Medicaid to new or "future" immigrants have imposed sponsor-deeming requirements, which will limit the number of immigrants who receive assistance.

The size of this new immigrant population is growing rapidly -- by roughly 800,000 to 1 million persons per year -- and the implications of its exclusion from the social welfare state will become less and less abstract over time, especially at the state level.

Figure 2: Refugees in the United States by Duration of Residence and Poverty:

1996-1997 Averages

Population Number Percent
Total non-citizen refugees 1,661,969

Non-citizen refugees residing in the U.S. for more than 5 years 981,283 59%

Under 125 percent of poverty 388,386 40% Non-citizen refugees residing in the U.S. for more than 7 years 702,179 42%

Under 125 percent of poverty 287,328 41%

Source: March 1996 and March 1997 Current Population Survey (CPS), compiled by The Urban Institute

Note: The CPS does not identify refugees. These figures include non-citizens born in countries that send large numbers of refugees, including Afghanistan, Albania, Cambodia, Cuba, Czechoslovakia, Ethiopia, Haiti, Hungary, Iran, Iraq, Laos, Nicaragua, Poland, Romania, Somalia, U.S.S.R., Vietnam, Yugoslavia, Thailand. Periods of entry vary by country.

New Time Limits on Refugees' Eligibility for Benefits. Welfare reform retains the preferential treatment historically extended to refugees by exempting them for a limited time from federal bars to food stamps, SSI, Medicaid, and TANF. Refugees' eligibility for food stamps, SSI and Medicaid terminates seven years after arrival (and for TANF, 5 years after arrival) if they do not meet another exemption and do not naturalize., The new time limits may leave many poor refugees without access to the safety net. Approximately 59 percent of the 1.6 million non-citizen refugees have been in the United States for more than five years and 40 percent have been here more than seven years. About two-fifths of each group has incomes below 125 percent of the federal poverty level-- the income cut-off for food stamps is generally 130 percent of poverty (Figure 2). A recent Urban Institute study of New York's immigrant population revealed that even after five years in the United States refugees' incomes trailed those of all other classes of immigrants -- including the undocumented. The study also found that the principal source of refugees' incomes was welfare payments. Further, refugees who come to the U.S. fleeing political persecution, do not have financial sponsors to whom they can turn for support if they lose public benefits.

Impacts on Immigrants Who Do Not Naturalize. One often overlooked population excluded from public benefits is immigrants who seek to naturalize but cannot do so. By limiting the access that non-citizens have to benefits, welfare reform introduces a strong new incentive to naturalize. The power of this incentive will be especially strongly felt among working-age immigrants ineligible for food stamps and new immigrants arriving after August 22, 1996 whose eligibility for all major benefit programs is conditioned on naturalization.

Making citizenship the threshold for access to the welfare state has unquestionably contributed to increases in demand for naturalization. It is well known that annual applications for naturalization have risen from 200,000 in 1991 to 1.3 million in 1996. The inducements to naturalize may have also expanded demand within segments of the immigrant population that earn less, have lower educational credentials, and speak less English than earlier cohorts that applied for citizenship. Many of these non-citizens confront significant barriers to naturalizing. These barriers include waiting periods that now average 18 months across the country and reach 24 months in large immigrant-receiving cities such as Miami. During this waiting period many immigrants remain ineligible for benefits. In addition, the fee for naturalization is $95 per person, a non-trivial amount for those on public benefits. Immigrants must pass an English and U.S. civics test in order to naturalize -- tests that can prove quite difficult for those who have not had opportunities to learn English or who have low literacy levels in their own language.

Impacts on Unqualified Immigrants. By barring unqualified immigrants from all "federal public benefits," welfare reform not only excludes undocumented immigrants from virtually all public benefits, it also excludes various classes of immigrants who are arguably here with the nation's consent. These include applicants for asylum and recipients of temporary protected status.

While illegal immigrants have been eligible for few public benefits, a number of child-oriented services have been made available to immigrants regardless of legal status. Although the reach of new restrictions on "federal public benefits" is less broad than some observers anticipated ( the Head Start and Maternal and Child Health programs are excluded) unqualified immigrants will no longer be able to use services provided under the Social Service Block Grants (SSBG) which provide two key types of service: child protective and child care services.

Despite the fact that providing benefits to illegal immigrants is beyond the bounds of current political discourse, the undocumented represent a large and growing share of the immigrant population and a powerful demographic reality. Over one-third of the recent immigration flow and twenty percent of all foreign-born residents are undocumented. According to Urban Institute estimates, there are over one million undocumented immigrant children under age 18 in the United States. While lower birth rates and increased job production in Mexico, along with stepped-up enforcement of immigration controls, could eventually slow the flow of illegal immigrants, these demographic and economic changes are not likely to be felt in the short run. Moreover, new requirements that immigrants' sponsors have incomes that exceed 125 percent of the poverty level could foreclose legal entry for many low-income immigrants. The new sponsorship requirements will likely shift a share of the immigration flow from legal to illegal channels. According to our earlier estimates, over one-third of all families headed by a foreign-born person, and half headed by an immigrant from Mexico or Central America, would not meet the new sponsor income requirements.

The size of the undocumented population is not only increasing because of new entrants, but because the path to gaining legal residence has become longer and more difficult. Changes under the 1996 Illegal Immigration Reform and Individual Responsibility Act (IIRIRA) have made it considerably more difficult for immigrants to acquire legal status if they have spent more than six months in the United States illegally. At the same time, strict new fingerprinting requirements introduced to screen immigrants for their criminal history have created a backlog not only in naturalization but also green card applications. Together, these developments mean that being undocumented may become less of a transitional status than it has been historically, making the class of immigrants barred from public benefits larger and more permanent. These issues affect a significant share of the new immigrant population -- at least twenty percent of new immigrants have entered the U.S. illegally at either their first or last trip to the U.S.

III. Beyond Eligibility: Welfare Reform's Normative and Institutional Legacies.
The immigrant provisions of welfare reform ushered in a host of changes that go beyond simple shifts in eligibility for public benefits. New norms, new complexity, new and inexperienced decision makers, fewer due process protections, and expanded verification and reporting requirements could counteract the efforts to blunt welfare reform's impacts described above. These shifts and the chilling effects on benefits use that they introduce, may end up having more far-reaching impacts than eligibility changes and stand as welfare reform's real legacy for immigrant policy.

New Verification and Reporting Requirements. In addition to shifting decision-making and fiscal responsibilities to states, the law mandates that a host of federal, state and local institutions will have to verify their clients' immigration status for the first time. Since verification is required for all benefits classified as "federal public benefits," the reach of the new responsibilities is directly tied to the way in which federal officials define this simple but consequential phrase. The potentially broad sweep of the federal requirement has been circumscribed by federal implementers. In addition, federal regulations make clear that agencies need only verify the immigration status of the individual who will receive (versus applies for) the benefit. Agencies, then, need not verify the status of undocumented parents applying for benefits for their citizen children.

Institutions not mandated to verify citizenship and immigration status may nonetheless be compelled -- or believe they are compelled -- to do so. Although charitable non-profit organizations are exempt from the verification requirements, the law sets up an inherent conflict: unqualified immigrants are barred from federal public benefits but organizations that provide assistance do not have to enforce the bar. Further, the law requires verification only for federal public benefits but it bars unqualified immigrants from federal, state and local public benefits. To comply with the law, then, states must also screen for immigration status when providing their own public benefits.

California in particular has taken an expansive approach to implementing this new bar on state public benefits, extending welfare reform's reach beyond welfare offices to a wide array of agencies and service providers. The state will not only screen for immigration status in public assistance programs but will require all holders of California professional or commercial licenses to prove that they are citizens or legal residents before their licenses can be renewed. The new requirements will affect over three million people who are licensed by the state (the California Division of Motor Vehicles alone renews over 573,000 commercial licenses annually). As a result, state bureaucrats who license professions ranging from acupuncturists to jockeys to pest control inspectors, will be forced to distinguish immigrants who are legally present from those who are not.

New verification rules have been accompanied by new requirements that certain state agencies report undocumented immigrants to the INS. State agencies administering federal housing assistance, SSI and TANF must report any alien the state "knows" is unlawfully in the United States to the INS at least four times a year. These new verification and reporting requirements are being introduced at the same time that welfare reform announces new bars on state and local "sanctuary laws" that barred city employees from asking about immigration status or reporting immigrants to the INS. Taken together, these new verification and reporting mandates could have potentially far-reaching effects on immigrants' willingness to trust public and private community institutions. In theory, the requirements could keep eligible citizens and legal immigrants from applying for, or receiving benefits if they fear that contact with a state or local agency could lead to the deportation of an undocumented relative -- again underscoring the complicated issue of mixed status families.

New Complexity. Implementing new verification and reporting requirements leads to another of welfare reform's legacies: increased complexity. In fact, few immigration-related laws have required such an immense dedication of energy and resources to interpret their meaning. It is axiomatic that complexity leads to decision errors, especially when thrust upon inexperienced new actors.

In the first place, a benefit-providing agency must wade through the more than 70 pages of the INS' draft verification guidelines and proposed rules to determine whether it must verify immigration status at all. If the answer appears to be "yes," staff will have to understand and implement the many new legal status distinctions the law draws among different groups of immigrants. In addition to the distinction between "qualified" and "not qualified" immigrants, eligibility may depend, among other things, on whether an immigrant entered before or after August 22, 1996, whether they or their family members have worked 40 quarters, and whether an immigrant's sponsor has signed an old or a new affidavit of support. Constantly changing immigrant eligibility rules also complicate implementers' jobs.

The intricate nature of the new eligibility rules and exemptions may also lead to a misunderstanding of the law on the part of immigrants themselves. Particularly following the partial restoration of eligibility for food stamps and SSI, many immigrants have been left confused over whether they qualify for benefits. The full complexity of the new law has yet to be seen, however, as litigation over the law's many controversial provisions and differing state and federal interpretations has only just begun.

Changing Social Norms. Welfare reform sends a number of implicit and explicit messages that reflect changing attitudes towards immigrants' use of public benefits. Some of the simple, and misleading, messages communicated by this very complex law include:

immigrants -- including legal immigrants -- are not entitled to benefits;

legal immigrants will not be able to naturalize if they get benefits;

legal immigrants will not be able to re-enter the United States if they receive benefits;

legal immigrants will have to repay benefits to re-enter the country;

legal immigrants will not be able to sponsor relatives if they receive benefits.

The shifting norms have already led to an overly broad interpretation of the law's restrictions on benefit use. Specifically, the State Department, the Immigration and Naturalization Service, California and other states have misapplied public charge laws to require repayment of benefits by immigrants trying to re-enter the country, obtain a green card or naturalize. Since then, California has been ordered to pay back the funds improperly collected. While the long-run chilling effects of changing norms and this controversial use of public charge have yet to be seen, data on declines in approved applications for public benefits in Los Angeles County provide a window onto their early impacts.

Compounding Effects of Other Policy Changes. Welfare reform's immigrant restrictions are not operating in a vacuum. Many changes occurring in other policy domains potentially limit immigrants' access to public benefits and could compound the challenges that public service providing organizations face. One example is the shift toward Medicaid managed care, which could limit immigrants' choice of health care providers by assigning them to health maintenance organizations (HMOs) that may or may not be conveniently located or offer culturally competent care. Medicaid managed care also appears to be placing additional financial burdens on public hospitals, making it harder for them to provide services to uninsured legal immigrants who no longer qualify for Medicaid.

Immigrants are not only affected by the provisions of welfare reform that specifically apply to non-citizens. Non-citizens who remain eligible for Temporary Assistance to Needy Families (TANF) will be subject to new time limits and work requirements that apply to citizens and non-citizens alike. Increased emphasis on job placement rather than education and training may have a particularly profound impact on immigrants and refugees who lack English and other job skills. It remains to be seen whether states will take into account the specific language and cultural needs of immigrants as they design their new welfare-to-work programs. Will they, for example, count English as a Second Language (ESL) classes as a work activity under TANF? In New York, there are already reports of immigrant welfare recipients being pulled out of ESL classes and put into janitorial jobs, which require few English skills and provide little opportunity for advancement.

Welfare reform's goals of restricting immigrant access to public benefits are interwoven with expanded, and arguably invalid, uses of public charge doctrine as well as confusion over the its correct application. Immigrants seeking to obtain a green card must prove they are not likely to become a pubic charge. Use of public benefits may, but does not necessarily, affect their chances of getting a green card. No public charge test, however, is applied to an applicant for naturalization and a legitimate us of public benefits should not therefore affect an immigrant's chances of becoming a U.S. citizen. As mentioned, federal and state officials have invoked the public charge doctrine to condition immigrants' adjustment to legal status (including getting a green card) and legal immigrants' reentry into the United States on the repayment of public benefits they previously received. This confusion over the consequences of benefits receipt and the misapplication of public charge rules may be magnifying the chilling effect that welfare reform has had on immigrants' use of public services.

Among the many changes introduced by the 1996 Illegal Immigration Reform and Immigrant Responsibility Act (IIRIRA) is a substantial reduction in the due process protections (such as judicial review) extended to non-citizens facing deportation or removal. These reduced protections could further chill immigrants' willingness to seek public benefits for which they remain eligible if it means greater exposure to expected deportation.

Chilling Effects: Data from Los Angeles County. We have suggested above that changing norms, new verification and reporting requirements, decision errors that may flow from increased complexity, overly broad use of public charge doctrine, and reduced due process protections could have a chilling effect on eligible immigrants' access to public benefits. Recent data provided by the Department of Public Social Services (DPSS) of Los Angeles County appear to support this contention.

As Table 1 indicates, between January 1996 and January 1998 there was no change in the number of newly approved citizen-headed cases or families for AFDC (now called TANF or CalWORKs in California) and Medi-Cal in Los Angeles County. During that same time period, though, the number of non-citizen-headed cases approved for TANF and Medi-Cal fell 52 percent from about 3,000 approved cases per month in early 1996 to about 1,500 cases per month for most of 1997. (Overall denial rates during the period held steady.) The fall-off in applications was equally steep for the citizen children of immigrants: monthly approved applications for TANF and Medi-Cal fell by half while there was a slight increase among the children of citizens (Table 2). This sharp decline in approved cases occurred despite the fact that there was no change during this period in immigrants' eligibility for these two programs under California law. As the table clearly indicates, this drop reflects a fairly steady decline over the course of the two years and is not a one-time phenomenon.

The decline in approved cases is greater for households headed by legal immigrants (-71 percent) than for households headed by illegal immigrants (-34 percent) who are applying for assistance for their eligible citizen children. In both instances, though the principal beneficiaries are typically citizen children. Similar patterns appear in the monthly approvals for county-funded General Relief and Medi-Cal only.

One explanation for the sharp decline in approved immigrant cases while there is no change in newly approved citizen cases could be high levels of naturalization. That is, it may be that some immigrants are still applying for benefits but as citizens rather than non-citizens, though it is unlikely that naturalization accounts for such steep declines.

Table 1. Cases Newly Approved for AFDC/TANF with Medi-Cal in Los Angeles County by Immigration Status of Household Head*:

January 1996 - January 1998


Total Monthly Approvals**


Percent of Total


Percent of Total

Legal Immigrants

Percent of Total Undocumented Immigrants*** (Unaided)

Percent of Total
January 1996 7,332 4,085 56% 3,177 43% 1,545 21% 1,632 22%
February 1996 7,681 4,468 3,168 1,650 1,518
March 1996 7,970 4,623 3,283 1,588 1,695
April 1996 7,754 4,626 3,068 1,542 1,526
May 1996 7,450 4,520 2,873 1,403 1,470
June 1996 6,680 4,119 2,490 1,218 1,272
July 1996 7,900 4,726 3,089 1,548 1,541
August 1996 5,693 3,570 2,066 1,036 1,030
September 1996 7,953 5,078 2,808 1,420 1,388
October 1996 7,767 5,072 2,621 1,307 1,314
November 1996 5,833 3,833 1,951 963 988
December 1996 5,781 4,037 1,702 529 1,173
1/96 - 1/97 Percent change -17% -1% -38% -59% -19%
January 1997 6,060 4,042 67% 1,967 32% 641 11% 1,326 22%
February 1997 5,691 3,923 1,731 564 1,167
March 1997 5,938 4,173 1,705 532 1,173
April 1997 5,846 4,272 1,491 457 1,034
May 1997 5,089 3,765 1,255 444 811
June 1997 5,262 3,819 1,387 441 946
July 1997 5,667 4,148 1,433 476 957
August 1997 5,689 4,170 1,457 472 985
September 1997 6,109 4,413 1,619 505 1,114
October 1997 6,543 4,774 1,696 489 1,207
November 1997 4,899 3,554 1,270 368 902
December 1997 6,257 4,466 1,702 454 1,248
January 1998 5,669 4,072 72% 1,519 27% 450 8% 1,069 19%
1/97 - 1/98 Percent Change -6% +1% -23% -30% -19%
1/96 - 1/98 Percent Change -23% 0% -52% -71% -34%

*Data are provided by "first adult" in household. This person is typically the parent. Each case usually includes one or two parents and their child(ren).

** Citizens and noncitizens do not necessarily add to the total because this total includes "other" cases, including those in which the immigration status of the household head is unknown.

***Undocumented immigrants are not eligible for AFDC/TANF or regular Medi-Cal. These cases typically include citizen children who are eligible for assistance.

Note: All AFDC recipients are automatically eligible for Medi-Cal. Most also receive federal food stamps (excluding ineligible immigrants). Data include both single- and two-parent families.

Source: Administrative Data from the Los Angeles County Department of Public Social Services

Table 2. Children Newly Approved for AFDC/TANF with Medi-Cal in Los Angeles County by Immigration Status of Household Head*:

January 1996 - January 1998


Total Monthly Approvals**

Citizen Children of Citizen Adults

Percent of Total

Citizen Children of Non-Citizen Adults

Percent of Total

Citizen Children of Legal Immigrant Adults

Percent of Total

Citizen Children of Undocumented Adults

Percent of Total
January 1996 11,669 6,148 53% 5,144 44% 2,741 23% 2,403 21%
February 1996 11,954 6,586 4,984 2,803 2,181
March 1996 12,489 6,917 5,178 2,689 2,489
April 1996 12,330 6,926 5,075 2,761 2,314
May 1996 11,467 6,644 4,418 2,288 2,130
June 1996 10,343 6,100 3,882 2,086 1,796
July 1996 12,476 7,161 4,815 2,531 2,284
August 1996 8,952 5,496 3,147 1,630 1,517
September 1996 12,572 7,761 4,415 2,348 2,067
October 1996 12,606 7,846 4,359 2,294 2,065
November 1996 9,197 5,834 3,111 1,613 1,498
December 1996 9,269 6,161 2,868 1,025 1,843
1/96 - 1/97 Percent change -17% +2% -39% -56% -19%
January 1997 9,725 6,245 64% 3,160 32% 1,203 12% 1,957 20%
February 1997 8,872 5,799 2,831 1,068 1,763
March 1997 9,285 6,208 2,819 1,000 1,819
April 1997 9,085 6,356 2,446 860 1,586
May 1997 8,088 5,643 2,193 914 1,279
June 1997 8,437 5,880 2,325 871 1,454
July 1997 9,111 6,360 2,442 886 1,556
August 1997 9,142 6,454 2,395 900 1,495
September 1997 9,936 6,858 2,788 998 1,790
October 1997 10,714 7,454 2,924 932 1,992
November 1997 8,052 5,615 2,136 684 1,452
December 1997 10,476 7,176 2,982 916 2,066
January 1998 9,460 6,501 69% 2,693 28% 921 10% 1,772 19%
1/97 - 1/98 Percent change -3% +4% -15% -23% -9%
1/96 - 1/98 Percent change -19% +6% -48% -66% -26%

* Data are provided for children by "first adult" in household. In most cases, this adult is the child's parent.

** Columns do not add to the total because this total includes "others," including non-citizen children and those for whom immigration status is unknown.

Note: All AFDC recipients are automatically eligible for Medi-Cal. Most also receive food stamps (excluding ineligible immigrants). Data include both single- and two-parent families.

Source: Administrative Data from the Los Angeles County Department of Public Social Services

The Fusion of Immigrant and Social Welfare Policy. In the wake of welfare reform, virtually all federal benefit or support programs must be classified as being either a "federal public benefit" or a "federal means-tested benefit" for the purposes of determining immigrant eligibility. By forcing policy-makers to classify all benefit programs in these terms, virtually all social policy actions will need to confront questions of immigrant policy and membership -- an imperative that takes on new meaning in an era of budget surpluses. States, which must bar unqualified immigrants from their own state and local public benefits, will need to take immigration issues more explicitly into account when they address benefits policy.

This new classification system will inevitably lead to controversial results. The new state Children's Health Insurance Program (CHIP) offers a good example. Like Medicaid, the funds this block grant provides to states to expand poor children's health coverage have been defined as a "federal means-tested benefit." This means that a state that funds a health program for poor residents with a combination of federal CHIP funds and its own revenues will be subject to the same immigrant restrictions as the Medicaid program. So some state health care programs that were once available to all legal non-citizens may now be barred to new immigrants. By effectively restricting the states' ability to serve their poverty populations, the law defeats its own federalist goal of giving states more authority over how to spend their money.

Moreover, the new five year bar on means-tested benefits paradoxically restricts immigrants' access to the programs they need most at the time their need is greatest. In the case of the children's health insurance program, new qualified immigrants are ineligible even though poor immigrants are nearly twice as likely to be uninsured (52 percent versus 26 percent) as poor natives. Further, immigrants' incomes are lower during their first years after arrival than they are after being in the U.S. for five or ten years. Although new immigrants' sponsors have pledged to support them, paying for health care for an uninsured immigrant who falls ill can be prohibitively expensive.

The Devolution of Immigrant Policy. Welfare reform marks a sharp departure from the federal government's exclusive role in determining immigrants' eligibility for public benefits. Before welfare reform, states were prohibited from setting their own eligibility standards for immigrants, as Supreme Court rulings from the 1970s held that the power to discriminate on the basis of alienage (i.e. citizenship status) was reserved to the federal government. The 1996 law essentially overruled this settled judicial doctrine and gave states broad new authority to determine immigrants' eligibility for benefits.

But welfare reform represents more than a simple transfer of authority to the states. It also can be viewed as a cost shift, as states choosing for political or other reasons to extend benefits to new or future immigrants must now pay the full costs. As the state food stamp experience indicates, where costs are shared between the states and the federal government, the states have been generous in retaining non-citizens' eligibility for benefits. But where there is no cost sharing they are less generous, and there is greater variation. And because it has been mainly the large immigrant-receiving states -- like California and New York -- that have extended state-funded food stamps, for instance, the inequitable and disproportionate share of immigration-related costs they now absorb is likely to widen in the future.

Welfare reform does not represent an expansion of state power in all instances, however. While welfare reform increases state power when it comes to determining legal immigrants' eligibility for public benefits, it reduces state discretion when it comes to framing policies that deal with illegal immigrants.

IV. Conclusion.
In sum, even though the federal government has scaled back some of the welfare law's more draconian restrictions, welfare reform's immigrant provisions leave a number of far-reaching legacies for the nation's immigrants and for immigrant policy. In the first place, almost three quarters of non-citizens who lost federal food stamps, most of them working age adults, have little prospect of having their eligibility restored. The effect of this continuing ban will be forcefully felt by citizens as well as non-citizens because of the large numbers of mixed families, i.e. families with non-citizen parents and citizen children. The rapidly expanding population of immigrants who arrived after August 22, 1996 have not benefitted from Congress' second thoughts about immigrant bars to public benefit programs. They remain excluded from most such programs at both federal and state levels.

Second, welfare reform's legacies go beyond eligibility restrictions on benefit programs to encompass the law's apparently profound chilling effects. In Los Angeles County, the number of monthly approved applications from legal immigrant-headed families for AFDC/TANF fell by over 70 percent between January 1996 and January 1998 despite the fact that there was no change in legal immigrants' eligibility for the program. During the same time period, there was no change in monthly approved applications for citizen-headed families. This decline in applications from immigrant families does not appear to be a one-time phenomenon, as 1,000 fewer AFDC/TANF cases headed by a legal immigrant are approved each month. These declines have translated into a commensurately sharp reduction in the number of citizen children of legal immigrants who are newly approved for AFDC/TANF in Los Angeles County.

Third, welfare reform has transformed the governance of immigrant policy by devolving authority to state and local governments. As the demographic significance of immigration continues to rise, responsibility for incorporating newcomers will increasingly fall to state and local governments whose capacity and inclination to respond generously will vary widely. Immigrants in states with strong economies, developed advocacy networks, and enough sympathetic voters may end up with substantially greater access to health and nutritional benefits than their counterparts in other states.


Committee on Ways and Means. 1996. 1996 Green Book: Background Material and Data on Programs within the Jurisdiction of the Committee on Ways and Means. November.

Clark, Rebecca L., Jeffrey S. Passel, Wendy N. Zimmermann, and Michael Fix. 1994. "Fiscal Impacts of Undocumented Aliens: Selected Estimates for Seven States." Washington, D.C.: The Urban Institute. September .

Congressional Budget Office. 1996. "Federal Budgetary Implication of the Personal Responsibility and Work Opportunity Reconciliation Act of 1996." Washington, D.C.: Congressional Budget Office. December.

Fix, Michael, and Wendy N. Zimmermann. 1997. "Welfare Reform: A New Immigrant Policy for the United States," Washington, D.C.: Immigrant Policy Program, The Urban Institute. April.

Jasso, Guillermina, et al. 1997. "The New Immigrant Survey (NIS) Pilot Study: Preliminary Results." August.

The Los Angeles Times. 1998. "Licensing to Require Legal Residency." February 28.

Magagnini, Stephen. 1997. "Suicide Illustrates Welfare Reform's Toll Among Hmong." The Sacramento Bee. November 9.

Passel, Jeffrey S., and Joan R. Kahn. 1998. "Immigration, Fertility, and the Future American Work Force." Washington, D.C.: The Urban Institute. March.

Passel, Jeffrey S., and Rebecca Clark. 1998. "Immigrants in New York: Their Legal Status, Incomes and Taxes." Washington, D.C.: The Urban Institute. April.

Schuck, Peter. 1998. "The Open Society: America is Pro-immigration After All." The New Republic. April 13.

The Wall Street Journal. 1998. "Cost of Highway Bill Endangers Plan to Restore Food Stamps to Immigrants." 3 April.

Wong, Doris Sue. 1997. "Food Pantries See Sharp Increase in Demand." The Boston Globe. November 29.

Zimmermann, Wendy, and Michael Fix. 1998. "Declining Immigrant Applications for Medi-Cal and Welfare Benefits in Los Angeles County." Washington, D.C.: The Urban Institute. July.

Zimmermann, Wendy, and Karen Tumlin. "Immigrants and Welfare Reform: State Policy Choices and Trends." Washington, D.C.: The Urban Institute, forthcoming 1998.