Number A-2 in Series, Assessing 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 available in its entirety in the Portable Document Format (PDF).
The Personal Responsibility and Work Opportunity Reconciliation Act, passed in 1996 by a Republican Congress and signed by a Democratic President, was hailed by many as a defining moment in the nation's progress toward devolving responsibility for social programs from the federal government to the states. A major part of this Act replaced federal responsibility for setting basic national standards and matching state spending for cash assistance to low-income families with a fixed block grant to states and major increases in state discretion over eligibility, benefit levels, and program rules. Other parts of the Act tightened eligibility standards for Supplemental Security Income (SSI) and the Food Stamp Program.
This brief looks at the 1996 reforms from a budgetary perspective, focusing on federal and state spending on public assistance generally. One important measure of the division of responsibility between the federal and state governments is their relative spending. A budgetary perspective, thus, provides several reality checks on the amount of devolution encompassed by the new reforms, and frames these changes within the broader budget trends that have driven much of the devolution debate.
Highlights of the Story
We find that the 1996 legislation, even when fully implemented, will make very little difference to federal spending on public assistance as a share of Gross Domestic Product (GDP). We consider federal spending as a percentage of GDP to indicate spending in relation to national economic growth.
In addition, despite one of the perceived intents of the block-grant funding structure, financial responsibility for safety net spending could become increasingly centralized at the federal level, not diffused through block grants, even as management and cost-control functions shift increasingly to states and localities.
Finally, a full assessment of the extent of devolution must look far beyond social safety net spending to the more dramatic trends in overall federal budget shares that have been taking place over the past 40 years—some toward decentralization, others in the opposite direction.
A Historical Perspective on Federal Social Program Spending
Federal social program spending trends over the four decades leading up to the 21st century highlight three major points.
Federal spending on Aid to Families with Dependent Children (AFDC) has been a declining part of the federal social safety net for the past 25 years.
Federal spending on AFDC through 1996, and on its successor program, Temporary Assistance to Needy Families (TANF), through 2002, appears at the bottom of figure 1.1 After reaching a high of about one-half of one percent of GDP in 1972, federal spending in this category declined through the early 1980s relative to economic growth. Since then it has remained remarkably stable at around one-fourth of one percent of GDP.
The projected impacts of all the 1996 reforms show a slight decline in total federal spending on non-Medicaid public assistance to low-income families relative to economic growth between 1998 and 2002. Much of this decline is not attributable to the legislation, however, but to the caseload implications of constant economic growth projections that were made before the 1996 block grant legislation and are unchanged by it. Also, nearly all of the reductions due to the legislation are due to cuts in SSI and food stamp eligibility, not AFDC/TANF changes.
The federal part of the jointly federal-state funded AFDC program has been overtaken in importance during the past 30 years by purely federal assistance programs. Food and nutrition assistance (mainly food stamps), SSI, housing assistance, and the Earned Income Tax Credit (EITC) together have been a far more important source of assistance to low-income Americans than AFDC for several decades.2 By the turn of this century, these entirely federal forms of assistance for low-income families together will account for about 1.5 percent of GDP, compared with an AFDC/TANF share of about 0.2 percent of GDP.
Medicaid—also funded jointly by federal and state governments—is the single dramatic exception to the projected pattern of steady or even declining federal spending for safety net programs. As a result of explosive growth in the early 1990s, federal spending on Medicaid by the year 2002 will be about as high as federal spending on all other safety net programs combined. While the Medicaid projections for 1994-2002 in figure 1 may be too high (see brief no. 5 in this series), revised estimates would not change the fundamental message that federal expenditures on health care for low-income Americans are following a dramatically different spending trend than is true for other parts of the safety net.
Shifting Roles within a Block Grant Reform: Potential Paradox
Under TANF's block grant structure, the amount of federal funding a state will receive to assist low-income families with children will be fixed for the next six years on the basis of recent federal spending on AFDC in that state.3 Because of the funding formula, almost all states will receive more federal funding than previously for at least the first year of TANF (see brief no. 1 in this series).
The block grant funding structure presents states with very different incentives to spend state funds from those facing them previously. Under AFDC, any additional dollar a state chose to spend on cash assistance for low-income families with children would be matched by federal funds (a one-for-one federal match for the richest states and an even larger match for poorer states). Under TANF, any additional dollar a state chooses to spend—to cover groups that are ineligible for TANF assistance (such as legal immigrants), raise benefit levels, or relax work requirements—will be state-only dollars.
This powerful new incentive to avoid additional state spending on low-income families with children makes it entirely possible that the state share of total spending on family cash assistance under TANF may fall—particularly given the TANF rule that states have to maintain only 75 to 80 percent of their own previous spending.4 Declining state TANF spending could also mean that the state share of all non- Medicaid public assistance could fall.
An additional factor increases the likelihood of such a scenario. State spending on Medicaid has been growing rapidly (figure 2). Given its dominant place in state and local public assistance, any growth in Medicaid provides inherent pressure to crowd out state spending in other safety net areas, including cash assistance.
Depending on its design, a future block grant reform of Medicaid could further this tendency toward lower state spending in relation to the overall size of the economy. Should the two predominant areas of state-financed assistance—AFDC/TANF and Medicaid— shrink, we could see the paradoxical result that the financing of low-income assistance becomes more concentrated at the federal level, even as responsibility for program design and implementation further devolves to the states.
Devolution beyond Public Assistance
Other realms of government activity may also be affected by devolution. Figure 3 shows that only two major components of the federal budget are projected to grow faster than the growth of the economy over the next few years: Medicare (the program that finances health care for the elderly and disabled) and Medicaid. All other major components are projected to remain constant or fall as we enter the 21st century. Soon thereafter retirement costs will rise dramatically as the baby boomers begin to retire.
The rapid increase in Medicare, retirement spending, and Medicaid has already been absorbing a larger share of the federal dollar—at the expense of most other categories of the budget—for some time now (figure 4). The one major exception has been non-Medicaid public assistance, which grew as a share of the federal budget from 1962 to the present (with the exception of the 1980s). It is now scheduled to fall over the next few years.
The broader budgetary perspective is important for several reasons. Changes in government functions like community development, energy, the environment, transportation—and, of course, defense spending—may have a much larger impact on local economies than changes in public assistance. Maintaining highways and other public infrastructure, for example, may require significantly more budgetary devolution than public assistance, if a larger state and local role is needed as the federal government devotes less of its resources to these areas.
The declining significance of the non-public assistance, nonretirement parts of the federal budget has increasing implications for how the budget accommodates public assistance versus spending on the elderly. Except for interest and perhaps a modest defense budget, most federal dollars will soon be shared among retirement, health, and non-Medicaid public assistance. These three items accounted for 20 percent of federal spending in 1962 and are projected to reach over 60 percent of federal spending by 2002 (figure 4).
Inevitably, these three categories will be forced to compete with each other, since for the share of any one of the three to grow noticeably, the combined share of the other two must decline. At almost no point in the past half century has this type of choice been required—mainly because a declining share for defense was able to offset in percentage terms almost anything else that the federal government pursued.
Sorting through the budget numbers leads us to conclude that devolution so far has been very selective in its impact. It has seemed to focus mostly on the young, both as the federal government has backed out of education and through scheduled changes in the funding of public assistance, most of which is spent on families with children.
Apart from Medicaid, these changes have been small as a percentage of GDP, though they still may play themselves out in unanticipated ways. To the extent that it is occurring, however, any decentralization of assistance for the young has been more than offset by centralization of assistance for the elderly and near-elderly.
This points to a final possible implication of the current stage of fiscal devolution. Federal resources could shift increasingly to states with higher percentages of elderly and near-elderly residents, and away from states with larger percentages of the young.
Notes
1. Figures for 1997-2002 are projections provided by the Congressional Budget Office.
2. Three of these forms of assistance—SSI, housing assistance, and the EITC—have state counterparts in some states, but these state-only programs are generally small.
3. Each state gets the federal share of its AFDC spending either averaged over 1992-1994, or in 1994, or in 1995, whichever is highest.
4. For a good discussion of potential state spending responses to a block grant world, see Howard Chernick and Andrew Reschovsky, "State Responses to Block Grants: Will the Social Safety Net Hold?" The LaFollette Policy Report, Spring/Summer 1996.
Note: This report is available in its entirety in the Portable Document Format (PDF).
PDF