|
Generational Differences In U.S. Public Spending, 1980–2000
The balance between spending on children and spending on the elderly is important in evaluating the allocation of public welfare spending. We examine trends in public spending on social welfare programs for children and the elderly during 1980–2000. For both groups, social welfare spending as a percentage of gross domestic product changed little, even during the economic expansions of the 1990s. In constant dollars, the gap in per capita social welfare spending between children and the elderly grew 20 percent. Unlike spending for programs for the elderly, spending for children’s programs suffered during recessions. Public discussion about the current imbalance in public spending is needed.
Recent budgetary proposals continue to transfer federal responsibility for social welfare programs to the states. The passage of the 1996 Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) marked a dramatic shift in social policy by ending standard entitlement to public assistance for low-income children and families and by devolving federal oversight of traditional welfare mechanisms to the states.1 There are now proposals to turn Medicaid, the single largest insurer of U.S. children, from an entitlement with an open-ended funding commitment from the federal government into a state "block grant" with a predetermined allotment of federal funds that would also allow states to cap enrollment and spending for optional programs within Medicaid.2 In contrast, Social Security and Medicare are federally administered through trust funds buffered from short-term economic fluctuations.
In comparison with other countries, social indicators for U.S. elders are quite good, whereas indicators for children are poor, and evidence suggests that social welfare spending can, to some extent, reduce poverty.3 Despite recent declines, U.S. child poverty rates are among the highest in the world.4 Over the past twenty years, the absolute number of children living in poverty has been nearly three times the number of elderly living in poverty, and poverty rates for children have consistently exceeded those for the elderly (Exhibit 1
In a 1991 study, Ted Benjamin and colleagues found that during the recession of the early 1980s, real spending on children declined, and spending on the elderly grew modestly. They concluded that "during times of economic downturn or stagnation combined with government fiscal crisis...programs for children are hit much harder by the budgetary axe than programs for the elderly."9 In this study we evaluate trends in social welfare spending for children and the elderly from 1980 to 2000 and the relationship of national economic trends to social welfare spending patterns.
Methods. Following the approach of Benjamin and colleagues, we examined the major components of public social welfare spending that directly benefit children and the elderly.10 Generally, we considered children to be those younger than age eighteen, and the elderly, those older than age sixty-five. For programs that directly benefit children, we included primary and secondary education (K–12), Aid to Families with Dependent Children (AFDC) and Emergency Assistance/Temporary Assistance for Needy Families (TANF), other child welfare, nutritional support, maternal and child health programs, medical spending under Medicaid and the State Children’s Health Insurance Program (SCHIP), and those portions of Social Security payments (Old Age, Survivors, and Disability Insurance [OASDI] and Supplemental Security Insurance [SSI]) that benefit children. For programs that directly benefit the elderly, we included medical spending under Medicare and Medicaid, Older Americans Act expenditures, veterans’ and railroad workers’ pensions, and those portions of Social Security (OASDI and SSI) and food stamps that benefit the elderly.11 Private social welfare expenditures, such as those from philanthropic organizations, and indirect social welfare benefits, such as those provided through the tax system, were not included. In cases where data sources were discrepant or not up to date, we contacted the relevant government agency and individuals responsible for producing annual statistical reports to obtain the most recent and accurate data. Total social welfare spending was calculated as the sum of all relevant components for both groups. Expenditures were adjusted to 2000 U.S. dollars using the medical component of the Consumer Price Index (CPI) for Medicaid, Medicare, and SCHIP and the overall CPI for all other programs.12 National economic trends were classified as follows: 1980–1984 recession, 1985–1989 growth, 1990–1994 recession, and 1995–2000 expansion. Unlike expansion, inflation is present during periods of growth. To relate changes in per capita spending to national economic trends, percentage changes were calculated for each time period.13 Per capita social welfare spending for children and that for the elderly were analyzed separately. Data sources. Data on primary and secondary education are from the 2001 Digest of Education Statistics and the National Center for Education Statistics.14 Data on TANF/AFDC, Social Security, SSI, and veterans’ pensions are from the U.S. Department of Health and Human Services (HHS) Social Security Administration and Department of Veterans Affairs.15 Railroad retirement and Older Americans Act data are from the U.S Railroad Retirement Board and volumes compiled for the U.S. House of Representatives Committee on Aging.16 Data on food stamps; Women, Infants, and Children (WIC); and child nutrition are from the U.S. Department of Agriculture.17 Data on other maternal and child health programs and child welfare are from the HHS Administration for Children and Families. Medicaid, SCHIP, and Medicare data are from the Centers for Medicare and Medicaid Services (CMS). Poverty and population data are from the U.S. Bureau of the Census. CPI, medical CPI, and GDP data are from the Bureau of Labor Statistics.
Distributional spending. The distribution pattern of social welfare spending on children and the elderly changed little from 1980 to 2000 (Exhibit 2  ). Education constituted the bulk of public spending for children over this period; it grew from 70.3 percent of total social welfare spending on children in 1980 to 81.2 percent in 2000. Children’s income assistance provided through AFDC (renamed TANF in 1996) represented the second largest portion. The decline in the relative proportion of funds devoted to income assistance and in the real dollar value of total spending in this area occurred predominantly after the passage of the 1996 PRWORA (Exhibit 3).18 There were declines in total food-stamp spending on children after 1996 and in the real dollar value of monthly food-stamp benefits. For the elderly, Social Security and Medicare represented the majority of social welfare spending distribution from 1980 to 2000, and spending in both areas grew consistently.
"Real" spending and as percentage of GDP. Social welfare spending in constant dollars (adjusted for inflation) for both children and the elderly grew over the study period. For children, real spending (in 2000 dollars) grew from $281 billion in 1980 to $459 billion in 2000. For the elderly, real spending grew from $380 billion in 1980 to $649 billion in 2000. However, as a percentage of GDP, social welfare spending for both groups was relatively unchanged (Exhibit 4 ).
Per capita spending. We next examined trends in per capita social welfare spending to account for differences in population growth. Per capita social welfare spending grew from $4,464 per child in 1980 to $6,380 per child in 2000 (Exhibit 5 ). Per capita spending for the elderly grew from $15,404 in 1980 to $19,702 in 2000. The gap in per capita spending between the two groups grew nearly 20 percent—from about an $11,000 difference in 1980 to a more than $13,000 difference in 2000.
Differences in per capita spending between children and the elderly grew during the recessions of the early 1980s and the early 1990s. As previously demonstrated by Benjamin and colleagues, spending on children’s programs suffered the most in 1980–1982 during the recession, when there was a 5.2 percent decrease in real total per capita total social welfare spending.19 In the ensuing period of growth, per capita public spending trends converged because of increased growth in spending on children. Real total per capita social welfare spending grew 15.8 percent for children and only 0.2 percent for elders. Despite this overall growth in per capita spending for children, most areas of social welfare spending on children declined in real dollars. Increases in spending on children through Medicaid and food stamps made up for declines in other areas. In the recession of the early 1990s, differences between per capita public spending on children and the elderly increased as growth of spending on children failed to keep pace with that for the elderly. During this recession, real total per capita total social welfare spending on children grew 6.0 percent; on elders, 7.7 percent. During the recent economic expansions of the late 1990s, total per capita social welfare spending rose 9.4 percent for children and 2.9 percent for the elderly. However, spending on children through AFDC and TANF dropped precipitously after passage of the 1996 PRWORA. At the same time, although medical spending on the elderly kept pace with inflation, medical spending on children fluctuated. From 1995 to 1997 it declined 22.5 percent and then increased 13.7 percent from 1998 to 2000 as Medicaid expansions and SCHIP were implemented.
Explaining the gap.
We found that differences in the growth rates of medical spending accounted for the majority of the widening gap during the study period. Per capita medical spending on children remained relatively flat, while per capita medical spending on elders grew (Exhibit 6
Public policy and economic trends. Changes in public policy enacted in response to national economic trends have had great influence on the allocation of public spending.20 Spending on children’s programs suffered the most between 1980 and 1982 during the recession, when there was a 5.2 percent decrease in real per capita spending. To some extent, this large reduction was attributed to an unprecedented increase in the size of the population under age eighteen (which had been steadily decreasing since 1967). Substantial cuts in income-tested programs for children were authorized by the Omnibus Budget Reconciliation Act (OBRA) and the Economic Recovery Tax Act of 1981.21 The result was that Medicaid and AFDC spending for children declined 10.7 percent and 7.2 percent, respectively, during this recession. Because of changes in eligibility criteria, Social Security payments to surviving children decreased even more steeply, by 19.3 percent. The Family Support Act (FSA) of 1988 took effect just as the economic growth of the mid-1980s ended, largely in response to growing public dissatisfaction with increases in welfare caseloads.22 The FSA reflected bipartisan consensus in which liberals achieved a broader safety net and conservatives achieved stronger work requirements for welfare recipients. Our analysis shows that social welfare spending on poor children through AFDC peaked in 1993 and then steadily declined in large part because of implementation of the FSA. The FSA served as a precursor to the dramatic shift in social policy that occurred with passage of the 1996 PRWORA. Although this legislation decoupled Medicaid and food-stamp eligibility from eligibility for cash assistance to preserve these benefits for needy families, multiple reports at the national, state, and local levels documented enrollment declines in all three programs.23 Our analysis confirms that declines in welfare case-loads were accompanied by declines in total public spending in TANF, Medicaid, and food stamps. Subsequent efforts to remedy inappropriate disenrollment from Medicaid, combined with the movement to Medicaid managed care, Medicaid privatization, and implementation of SCHIP, led to increased outreach efforts to enroll all eligible children in Medicaid.24 As a result, spending levels returned to their 1993 levels by 1999. Devolution to the states. As the U.S. population ages and elderly people continue to have increasing and unmet medical needs, our observation that the growth of per capita medical spending on the elderly has exceeded the growth of per capita medical spending on children is not surprising. In an attempt to preserve health care benefits for elderly people in the future, the Balanced Budget Act (BBA) of 1997 included provisions to ensure the solvency of federally guaranteed Medicare trust funds.25 In contrast, current budgetary proposals continue to transfer funding responsibility and determination of eligibility requirements for social welfare programs, including Medicaid, from the federal government to the states, discounting the needs of 8.5 million uninsured U.S. children.26 This trend is particularly alarming in light of the fact that children represent an increasing proportion of the population living in poverty and that poverty has been persistently and repeatedly shown to be associated with poor health outcomes.27 Moreover, given that a large proportion of adult morbidities, such as cardiovascular disease and cancer, are associated with preventable childhood precursors, such as obesity and smoking, increasing and stabilizing investments in child health merit serious consideration.28 Study limitations. This study has several limitations. First, some programs providing benefits to both groups, such as public housing and the low-income home energy assistance program, do not provide data classified by age. These programs were not included in our analysis. Second, our analysis did not include indirect social welfare benefits for children and elders, such as those provided through tax subsidies. The Earned Income Tax Credit (EITC), a refundable federal income tax credit for low-income working individuals and families, underwent modifications in 1993 and 1996 that led to large increases in the number of recipients and refunded amounts.29 Even though the total amount refunded grew from $1.3 billion in the 1980s to more than $26 billion in 2000, this would not offset the $190 billion gap in social welfare spending between children and the elderly in 2000.30 Furthermore, to avoid bias, consideration of tax policies designed to benefit children (such as refundable credits and nonrefundable deductions) warrants consideration of tax policies designed to benefit the elderly (such as tax-deferred retirement accounts and the Tax Credit for Elderly and Disabled).31 Estimating benefits to the elderly through the tax system would require complex statistical modeling that relied on untested assumptions about retirement account contributions, earnings, and distributions. Taking these issues into consideration, indirect social welfare benefits through the tax system were not included in our analysis. Further research in this area is needed. Despite these limitations, our findings provide evidence that unlike spending on elders, social welfare spending on children is vulnerable to downturns in the U.S. economy. Persistent disparities in the proportion of children versus elders living in poverty and the importance of investments in preventive health argue for increased and stable social welfare spending on children’s programs. The continued transfer of responsibility for funding of social welfare programs from the federal government to the states endangers the stability of funding for all children’s social welfare programs, especially health insurance. Whereas the elderly are afforded a basic guarantee of support regardless of economic and political changes, social welfare spending on children is left vulnerable to these fluctuations. In the current period of economic stagnation, U.S. spending on children’s social welfare programs and health care is likely to be further compromised.
The authors are all members of the Pediatric Generalist Research Group at the Children’s Hospital of Philadelphia, senior fellows at the Leonard Davis Institute of Health Economics, and associate scholars at the Center for Clinical Epidemiology and Biostatistics at the University of Pennsylvania School of Medicine. Susmita Pati and Ron Keren are assistant professors of pediatrics in the Division of General Pediatrics and Evaline Alessandrini, in the Division of Emergency Medicine, at the Children’s Hospital. Donald Schwarz is the Mary D. Ames Associate Professor of Child Advocacy and the Craig-Dalsimer Division Chief of Adolescent Medicine there and vice chairman of the Department of Pediatrics at the Penn School of Medicine. Susmita Pati is supported in part by a grant from the Anne E. Dyson Foundation. Ron Keren and Evaline Alessandrini are supported by Career Development Awards no. K23 HD 043179 and no. K23 HD01320-01A1, respectively, from the National Institute of Child Health and Human Development. The authors gratefully acknowledge Jennifer Loftus for her assistance with the preparation of this manuscript.
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Study Data And Methods
Findings
