Number A-15 in Series, "New Federalism: Issues and Options for States"
The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.
Note: This report is also available in PDF format which many find easier when printing.
mmigrants, though never the focus of the nation’s long-running debate over welfare reform, were thrust into the center of this important social experiment by the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA), enacted in August 1996.
1 As initially passed, PRWORA’s restrictions on immigrant access to benefits, particularly Supplemental Security Income (SSI) benefits and food stamps, were expected to account for $23 billion—almost half of the total federal savings that welfare reform was expected
to generate.
The Balanced Budget Act of 1997 restores SSI and derivative Medicaid benefits to all elderly and disabled immigrants receiving SSI at the time PRWORA was enacted and to all legal immigrants in the United States on the date of enactment who become disabled in the future. The budget act also restores additional benefits to refugees and expands the group of immigrants treated as refugees. 2
This restoration of benefits represents a major victory for immigrant advocates. But these reinstated benefits do not alter many of the law’s far-reaching effects on immigrants, on the nation’s immigrant policy, or on the new role state and local governments will play in shaping the policies that govern immigrant integration.
In this brief we argue that even a reformed PRWORA embodies a fundamental shift in the locus of the nation’s immigrant policy from the federal government to states and localities. (Immigrant policy covers the government rules and investments designed to promote the social and economic integration of newcomers. 3) The new policy fragments what had previously been uniform national rules set by Congress and the courts regarding noncitizens’ eligibility for public benefits. Further, because states are granted far greater power than before to deny public benefits to noncitizens, welfare reform deepens the differences between how citizens and noncitizens are treated and redefines noncitizens’ membership in society.
By drawing the kind of bright line between legal immigrants and citizens that was formerly drawn between illegal and legal immigrants, welfare reform tightens the circle of full membership within our society. 4 By conditioning access to the safety net on citizenship, welfare reform elevates the importance of citizenship in a nation where its value has been limited largely to exercising political rights, holding some government jobs, and obtaining certain immigration privileges. 5
Welfare Reform and Immigrant Policy
Unlike other high-immigration countries such as Australia, Canada, and Israel, the United States up to now has had only a limited set of federal policies specifically targeted to immigrants. These have included policies that help refugees resettle, assist immigrants and others to learn English, and seek to offset the impacts of immigration on local schools. A few states with substantial immigrant populations, such as Massachusetts, have supplemented federal programs with their own funds. In general, though, these federal and state initiatives have been limited both in scope and funding.
In contrast to this skeletal set of express federal immigrant policies, the nation’s de facto immigrant policy gives legal immigrants eligibility for mainstream social programs such as AFDC, Medicaid, and loans for higher education on largely the same terms as citizens. Further, the federal courts have barred states from discriminating against legal immigrants in state or federal public benefit programs. These rulings have generally promoted legal immigrants’ access to social welfare programs while circumscribing the states’ role in immigrant policy design.
Welfare reform dramatically reorders these arrangements. By restricting legal immigrants’ eligibility for benefits, the law transforms U.S. immigrant policy from one that has historically treated legal immigrants and citizens alike to one that, with some exceptions, mandates that legal immigrants be barred from "federal means-tested benefit programs." (As of this writing the scope of "federal means-tested programs" has been limited to Temporary Assistance to Needy Families (TANF) and Medicaid.) Further, the law marks the first time in modern history that Congress has explicitly authorized states to discriminate against legal immigrants in the administration of their public benefit programs. 6
These broad shifts in immigrant policy are taking place within the context of a relatively inclusive immigration policy. The United States now admits roughly 800,000 legal immigrants annually who are eligible to become citizens: more than to all nations in Western Europe combined and more than at any previous point in our history. While recent legislation makes it more difficult for low-income U.S. citizens or legal permanent residents to sponsor immigrants, the last (104th) Congress expressly rejected efforts to reduce high legal immigration levels. As a result, welfare reform marks a broad pullback in U.S. immigrant policy at a time of high, sustained immigration to the United States, creating what can be viewed as a mismatch between the nation’s immigration and immigrant policies.
In What Ways Does Welfare Reform Redefine Immigrant Policy?
Dividing the Immigrant Population into Qualified and Unqualified Immigrants
For the first time, welfare reform segments the immigrant population into broad categories for the purpose of receipt of public benefits: qualified and unqualified immigrants. This approach is intended to simplify the laws governing aliens’ rights to benefits. But simplifying the law in this way expressly relegates several classes of immigrants lawfully in the United States (such as applicants for asylum or adjustment of status) to the same unqualified status as the undocumented. It thereby blurs distinctions between immigrants here with the government’s consent and those here without it.
Restricting Access to the Social Safety Net: Current Immigrants
The budget act not only restores SSI and derivative Medicaid benefits to immigrants receiving SSI at the time PRWORA was passed, it will extend SSI to legal immigrants in the United States at the date of enactment who become disabled at a future date. However, the budget agreement does not restore any eligibility for immigrants losing food stamps. An estimated 1.4 million current noncitizen food stamp recipients, or 77 percent of current noncitizen food stamp recipients, will lose benefits under welfare reform. 7 Many immigrants losing food stamps will not be able to increase work to make up for lost benefits, as 17 percent of all noncitizens receiving food stamps are children, 17 percent are 60 or older, and 7 percent already work full time.
Restricting Access: Future Immigrants
Because of the immediate hardships faced by current immigrants losing benefits, less attention has been given to welfare reform’s broader exclusion of future immigrants from safety net programs. Future immigrants are not only excluded from SSI and food stamps, they will be barred for five years after entry from Medicaid, TANF, and any other programs defined by the attorney general as federal means-tested public benefits. Unless immigrants naturalize or work 10 years in covered employment, their exclusion can continue after the five-year federal bar. This prolonged exclusion is accomplished by "sponsor deeming"—that is, by ascribing the income of the immigrant’s sponsor to the immigrant for the purposes of qualifying for means-tested benefits.
It is important to recognize that deeming usually disqualifies applicants for benefits. Under welfare reform, states have the ability to impose sponsor deeming on future immigrants for federal, state, and local programs. States also have the option of providing coverage to future immigrants without deeming. While this population of "future immigrants" is now small, it will grow rapidly given the nation’s high immigration levels. As the future immigrant population grows, so too will the law’s impact on legal immigrants and the communities in which they settle.
Shifting the Burden of Support to Immigrants’ Families
Broad restrictions on immigrant access to benefits, coupled with new support obligations for immigrant sponsors, shift responsibility for the provision of safety net services from the federal government to the immigrant family. The new immigrant policy will force the families of future immigrants to absorb the health and other costs of family members—even where costs are unpredictable, such as those that arise when future immigrants become disabled after they enter the country.
Exempting Education and Human Capital Programs
Although welfare reform restricts immigrant access to certain safety net programs, it expressly retains immigrants’ eligibility for education and human capital programs. The legislation explicitly states that legal immigrants will remain eligible for Head Start, means-tested elementary and secondary programs, federal higher education assistance, and services provided under the Job Training Partnership Act.
How Does Welfare Reform Redefine State/Federal Roles in Immigrant Policy?
The new welfare reform law is striking not only for the new lines that it draws between classes of immigrants but also for the power that it vests in
states to draw those lines. The law is also notable for the greater financial responsibility that falls to states choosing to extend benefits to noncitizens. In addition, the law expands states’ responsibilities to police benefit programs to ensure that unauthorized immigrants do not receive benefits. Welfare reform, then, transfers much of the governance of immigrant policy from the federal to the state and local levels.
The case for and against devolution of immigrant policy to the states mirrors arguments for and against devolution generally. That is, proponents of devolution argue that lodging immigrant policy with state and local governments allows them to target benefits and services more efficiently to the needs of local populations. The case for localized policymaking is particularly strong given differences across immigrant populations in terms of their native languages, English language capacity, educational credentials, family structure, employment and social networks, and culture. At the same time, giving states the ability to restrict access to benefits allows them to make more cost-effective choices about the allocation of scarce financial resources. This reordering of financial responsibility may be particularly significant in an era of continuing high immigration marked by limited cost sharing on the part of the federal government. In addition, giving states the ability to condition aid on citizenship aligns the interests of both the federal and state governments in encouraging naturalization.
Opponents of transferring immigrant policy raise several counterarguments. Welfare reform gives states the power to shape the meaning of citizenship—a power that flows from the states’ new authority to define the rights and benefits due to legal noncitizens. Many argue that this power is better left to the federal government, as states could deploy their new authority in ways that would effectively discourage new immigrants, or classes of immigrants, from entering and settling—thereby fragmenting the country’s uniform immigration policies. Further, given immigration’s highly concentrated character, devolution of immigrant policy may only exaggerate the disproportionate fiscal burdens already borne by the few states in which immigrants are concentrated and which, for political or other reasons, extend benefits to noncitizens.
Devolution of immigrant policy—like other areas of social welfare policy—also invokes the specter of a race to the bottom, in which states seeking to avoid becoming welfare magnets provide their residents with fewer benefits than they would otherwise choose to do. Finally, noncitizens cannot vote and are increasingly members of racial and ethnic minorities, making them particularly vulnerable to state and local discretion.
New Areas of State Discretion
What, then, are some of the new powers that welfare reform vests in states? The law allows states to decide for the first time:
- Whether they will provide certain federal benefits (TANF, Medicaid, and other means-tested benefits) to current legal immigrants;
- Whether they will fund participation of future immigrants in federal means-tested programs during the five-year federal bar;
- Whether they will provide federal means-tested public benefits to future immigrants after five years;
- What mix of state- and locally funded services (e.g., general assistance (GA) or indigent care) they will extend to current and future legal immigrants; and
- What tools (e.g., bars, deeming, state residency requirements) they will use to restrict immigrant access to state-funded services.
New Limits on State Authority
While welfare reform generally provides states with greater authority when it comes to determining legal immigrants’ eligibility for public benefits, it simultaneously reduces state discretion in developing policies that deal with illegal immigrants. First, PRWORA requires a state wishing to provide most services to undocumented and other "not qualified" immigrants 8 to enact a new law that explicitly announces the state’s intent to do so. Existing laws do not count; the laws must be enacted after welfare reform’s passage. Some legal scholars have questioned whether this requirement unconstitutionally infringes upon states’ sovereignty by forcing them to pass laws in order to govern in an area of policy the federal government has not preempted. 9
Second, states are barred from retaining "sanctuary laws" that prohibit state or local officials from reporting illegal immigrants to the Immigration and Naturalization Service (INS). This prohibition was challenged by New York City Mayor Rudolph Giuliani, who claimed that it violates principles of federalism and jeopardizes public health. 10 Giuliani’s challenge was rejected by a federal district court in July and has been appealed.
Third, to restrict illegal immigrants’ access to public resources, welfare reform requires that states verify the immigration status of applicants for all "federal public benefits programs which provide direct federal expenditures" including services, contracts, and grants. 11 Further, by making citizenship, not simply legal presence, a prerequisite for receiving certain public benefits, states must now determine whether applicants are citizens, not simply that they are legally present.
Fourth, state agencies administering federal housing, SSI, and TANF programs must report to the INS the names, addresses, and other identifying information of aliens whom the state "knows are unlawfully in the United States." 12
Each of these new constraints on state discretion reflects an expansion of federal immigration control objectives into the day-to-day operations of state and local government agencies. This trend has been evolving since the passage of the 1986 Immigration Reform and Control Act, which required that agencies administering selected federal benefit programs verify the legal status of noncitizens. Welfare reform substantially accelerates it.
New State Spending
The new immigrant restrictions are likely to generate new state and local costs, exacerbating current intergovernmental fiscal inequities that flow from immigration. 13 Under current fiscal arrangements, immigration generates a net surplus for the federal government, a net deficit for local governments, and mixed results at the state level, depending on an individual state’s spending and tax structure. This is because most of the taxes paid by immigrants flow to the U.S. Treasury (e.g., FICA), while many of the services used by immigrants (e.g., education) are paid for by local governments.
According to Congressional Budget Office estimates for the 1997–2002 period, the federal government would have provided $23 billion less to state and local governments as a result of welfare reform’s immigrant restrictions than if no reform had passed. 14 Restoration of federal SSI and Medicaid benefits will reduce these savings by about $11.4 billion, according to early estimates. Still, the fiscal impacts of PRWORA’s immigrant provisions are likely to be felt at the state and local levels. Noncitizens remain barred from food stamps; since these are fully funded by the federal government, any replacement cost will be fully borne by state or local governments. The fiscal effects of these and other federal benefit curbs will be disproportionately felt by the six states that together account for three-quarters of the immigrant population—if they choose to extend benefits to noncitizens. In fact, California and New York alone account for half of all noncitizen food stamp participants. 15
It could be argued that one-time fiscal gains from welfare reform will help offset these losses. However, the two states with the largest immigrant populations, California and New York, stand to gain substantially less than most others from the expected short-term TANF windfall. (This is because each state’s TANF block grant is based on the size of its welfare caseload in the early to mid-1990s. Thus, states like California and New York that have not experienced steep declines in their welfare rolls get a smaller short-term financial windfall.)
Welfare reform’s immigrant provisions are also likely to stimulate other areas of state spending: several large immigrant states and cities are spending money to help immigrants through the naturalization process. And all states are likely to be forced to bear new administrative costs from expanded verification and reporting requirements.
New Local Immigrant Policies
The devolution of immigrant policy to the states is triggering a further "pass-through" devolution to counties and other local units of government. This pass-through devolution will drive immigration politics to lower levels of government, generating divergent immigrant policies not just across states but across counties. Because immigrants, and in particular poor immigrants, are highly concentrated in central cities,16 financially strapped urban counties like Los Angeles may find it particularly difficult to increase expenditures on newcomer populations.
Immigrant Policy: What Choices Are States Making?
In the wake of welfare reform, states face three broad sets of choices as they shape their immigrant policies. First, they must set eligibility criteria for state- and federally funded programs. Second, they must decide if they will spend state funds to offset the loss of federal benefits and promote naturalization. Third, they must determine how vigorously they will enforce new restrictions on legal and illegal immigrants’ use of benefits, as well as the support obligations of future immigrants’ sponsors.
Determining Eligibility for Benefits
Welfare reform empowers states to set eligibility criteria across a wide range of federal, state, and locally funded programs. States are implementing these new responsibilities by determining which populations they will exclude from benefits as well as how they will condition the benefits they extend to otherwise qualified populations.
Exclusions. As we have noted, states are authorized to exclude or bar current immigrants from TANF, Medicaid, Title XX Social Services Block Grant programs, and other federal means-tested programs. As of this writing, none of the major immigrant-receiving states has limited current legal immigrants’ access to TANF, Medicaid, or Title XX Block Grant programs. Fear of becoming a welfare magnet has not driven at least these states to withdraw benefits. In fact only one state, Alabama, has chosen to deny or limit immigrant participation in TANF.
States can also decide whether they will exclude future immigrants from TANF, Medicaid, and any other means-tested federal benefits after the five-year federal bar ends. In contrast to their treatment of current immigrants, many early state proposals (including states with substantial immigrant populations) do not extend either federal or state benefits to future immigrants.
Furthermore, states must determine whether they will exclude qualified legal immigrants from their own state- and locally funded benefit programs and, if they choose to do so, which immigrants will lose benefits. For noncitizen food stamp recipients and for virtually all future immigrants who fall on hard times, these state choices regarding state and local programs are likely to prove the most important. Maryland, for example, will provide state-funded food assistance to legal immigrant children but not to legal immigrant adults.
Finally, states must determine whether they will exclude illegal and other "not qualified" needy immigrants from state and federal benefit programs. For a narrowly circumscribed number of programs, including WIC, states may provide services without having to enact a law that expressly authorizes them to do so. For other types of state benefits, though, such as indigent care, it appears they must take on the politically difficult, federally mandated burden of enacting new legislation. Again, this new federal requirement raises constitutional concerns.
Conditions of Eligibility. States are also using their new legal authority to condition the receipt of benefits for immigrants who meet broad eligibility criteria. One important new power that welfare reform grants states is the authority to deem the incomes of immigrant sponsors. In considering how they will apply deeming, states have several options. These include determining which programs they will condition on sponsor deeming, whether to deem the sponsor’s full income, the length of the deeming period, and whether deeming will be applied to current and future immigrants. In Washington State, for example, the legislature has voted to make both current and future immigrants eligible for TANF, Medicaid, and Social Services Block Grant programs. However, the state will deem sponsor incomes for all immigrants, thereby limiting their actual receipt of public benefits.
Another new condition of eligibility that states are imposing on legal immigrants is state residency requirements. Maryland has passed legislation that would provide TANF to current immigrants who have been residents of the state for 12 months or have moved from another state that provided cash benefits. It should be emphasized that these proposed residency requirements, which in the past have been struck down by the courts because they violated the constitutionally protected right to travel, are now being applied only to noncitizens.
Some states are conditioning benefits receipt on pursuit of naturalization. New Jersey, for example, makes receipt of GA beyond six months contingent on naturalization.
In sum, efforts to frame new eligibility criteria for noncitizens are leading to the de facto creation of new classes of immigrants, defined by their state of residence, the date of their arrival, the length of their residency within an individual state, the welfare policies in place in the state of their former residence (if applicable), and their progress through the naturalization process. While all these criteria may influence eligibility, none bears directly on indigence or need.
Financing Benefits for Those Losing Federal Services
The second broad area of state choice has to do with spending. Here states must decide whether to finance services out of their own revenues for (a) current immigrants losing food stamps and (b) future immigrants denied food stamps, SSI, SSI-Medicaid, and other "means-tested federal public benefits." States may decide which losses to offset, which immigrants to cover, and the duration of benefits. They must also decide whether to allocate federal funds provided under TANF or Social Services Block Grant programs to current immigrants.
Current state-level deliberations have highlighted several constraints in substituting state for federal benefits. One is the limited ability of GA programs to replace the benefits lost to legal immigrants. In general, GA cash benefits are substantially lower than either SSI or TANF benefits. Indeed, even in a state with a generous GA program (such as New York), immigrants no longer eligible for food stamps and SSI will replace less than 60 percent of these federal benefits by qualifying for state GA. Moreover, not every state or jurisdiction has a GA program in place. Nine states do not have a GA program at all, and nine other states 17 only have a program in certain counties, including the high-immigrant states of Illinois, Florida, and Texas.
Another emerging issue is the challenge states face in replacing federal food stamp benefits. Currently, most states do not operate food or nutrition programs that can be used to deliver assistance to noncitizens losing food stamps. Thus, states may be forced to create new programs from the ground up, incurring high costs associated with establishing the needed administrative and service delivery "infrastructure." One way around this problem is to provide aid to food pantries. This may be a viable alternative, but it is hard to ensure that food is received by target populations such as elderly or disabled immigrants.
Another way states can provide food assistance to noncitizens is by purchasing food stamps from the federal government and distributing them to legal immigrants. (State purchases were authorized by the FY 1997 Emergency Supplemental Appropriations Act.) Despite these changes, it now appears that when states do provide food assistance to noncitizens, they offer a lower level of support than federal food stamps and provide it to only selected groups of the formerly eligible immigrant population (e.g., the disabled elderly).
Enforcement
A third new area of choice is the commitments that states make to policing restrictions on legal and illegal immigrants’ benefits use and enforcing sponsors’ support obligations. One threshold decision states will make is whether to spend the money required to set up administrative systems that can enforce the affidavit of support signed by an immigrant’s sponsor. States are also facing choices regarding the scope of the verification they will undertake in benefit programs: Will they only verify the legal status and citizenship of those who volunteer that they are not citizens? Or will they undertake the much larger task of verifying the legal status of all applicants for benefits? How will they reconcile state privacy requirements with federal reporting demands? Will they report suspected illegal immigrants to the INS?
Conclusion
Although the recently enacted budget agreement restores SSI and derivative Medicaid to almost a half million noncitizens, welfare reform still represents a sharp reduction in the safety net available to current and especially future immigrants, as well as a fundamental shift in immigrant policy and its governance. Legal immigrants arriving in the United States will no longer be treated the same as citizens in many federal and state benefit programs, creating a new official policy of immigrant exceptionalism. At the same time, welfare reform fragments immigrant policy by transferring much of its formulation, funding, and enforcement from the federal government to the states.
Devolution will mean that the public resources dedicated to immigrants’ integration will be determined by the states and counties in which they live and pay taxes. And this in turn raises a number of questions that will be the subject of ongoing monitoring by the Urban Institute. Will the acrimonious national political debate over immigration transfer to the state and local levels along with the transfer of decisionmaking? How well will the interests of noncitizens, most of whom are minorities and who cannot vote, be represented at the state and local levels? Will immigrants secure more generous spending policies in places where they are concentrated and where their impacts may be more forcefully felt? Or will more generous spending be in jurisdictions where they are less concentrated and their fiscal and other impacts less visible? Finally, will states and the urban counties in which immigrants are concentrated have the fiscal capacity to extend benefits—even if they are politically inclined to do so?
Notes
1. Title IV, Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (P.L. 104-193).
2. Balanced Budget Act of 1997 (P.L. 105-33).
3. Michael Fix and Wendy Zimmermann, "After Arrival: An Overview of Federal Immigrant Policy in the United States," in B. Edmonston and J. Passel, eds. Immigration and Ethnicity: The Integration of America’s Newest Arrivals, Urban Institute Press, 1994.
4. Alexander Aleinikoff, "Citizens, Aliens, Membership and the Constitution," Constitutional Commentary 9 (1990).
5. Constitutional scholar Alexander Bickel has written, "Remarkably enough ... the concept of citizenship plays only the most minimal role in the American Constitutional scheme." The Morality of Consent (New Haven: Yale University Press, 1975), p. 33.
6. These provisions are set out in companion immigration legislation intended to clarify and modify Title IV of the Personal Responsibility and Work Opportunity Reconciliation Act. The law is the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (P.L. 104-208).
7. Mike Stavrianos, Scott Cody, and Kimball Lewis, "Characteristics of Childless Unemployed Adult and Legal Immigrant Food Stamp Participants: Fiscal Year 1995," Food and Consumer Service, U.S. Department of Agriculture, February 13, 1997.
8. "Not qualified aliens" as defined by PRWORA include all aliens not falling into the "qualified alien" category and who are not nonimmigrants as defined by the Immigration and Nationality Act. Generally, this category includes all undocumented immigrants and PRUCOL immigrants.
9. Erwin Chemerinsky, "Memorandum on the Constitutionality of Section 411(d) of H.R. 3734," September 1996.
10. The suit claims that the new federal prohibition violates the Tenth Amendment and the Guarantee Clause of the U.S. Constitution because, among other things, it "usurps states’ and local governments’ administration of core functions of government." The City of New York v. The United States of America, USDC, SDNY, July 24, 1997.
11. Some programs, such as emergency Medicaid services, immunizations, testing and treatment for communicable diseases, and emergency disaster relief, will not be subject to verification.
12. It remains to be seen how broadly this knowledge requirement will be interpreted. Under current agency and congressional interpretation, such knowledge does not arise until the alien is under a final order of deportation. Only the food stamp program now requires reporting of household members. Charles Wheeler, "The New Alien Restrictions on Public Benefits: The Full Impact Remains Uncertain," Interpreter Releases 73 (September 23, 1996).
13. Eric S. Rothman and Thomas J. Espenshade, "Fiscal Impacts of Immigration to the United States," Population Index 58, no. 3 (1992).
14. Federal Budgetary Implications of the Personal Responsibility and Work Opportunity Reconciliation Act of 1996, Congressional Budget Office, December 1996, p. 27.
15. Stavrianos, Cody, and Lewis, 1997.
16. While immigrants represented 8 percent of the nation’s population in 1990, they constituted 16 percent of the population of the nation’s central cities. Sixty percent of immigrants who have lived in the United States for 10 or more years and who are poor live in central cities. Maria Enchautegui and Aaron Sparrow, "Ten Years and I Ain’t Making It: Poverty among Long-Term U.S. Immigrants," Journal for Children and Poverty 3, no. 1 (spring 1997): 67.
17. Cori Uccello, Heather McCallum, and L. Jerome Gallagher. "State General Assistance Programs, 1996" Urban Institute, October 1996.
Tables and Figures
About the Authors
Michael E. Fix is Director, Immigrant Policy Program, and a principal research associate at the Urban Institute’s Population Studies Center. His areas of special interest are legal and political analyses of regulatory, civil rights, immigrant, and immigration policy.
Karen Tumlin is a research assistant working on the Institute’s Assessing the New Federalism project.
The authors would like to thank Anna Kondratas, Freya Sonenstein, Alan Weil, and Wendy Zimmermann for their helpful comments on an earlier draft. The research was funded by the Andrew W. Mellon Foundation and the Urban Institute’s Assessing the New Federalism project. The opinions expressed herein are those of the authors and not those of the officers and trustees of the Urban Institute or the Mellon Foundation.
PDF