urban institute nonprofit social and economic policy research

Federal and State Funding of Children's Programs

Read complete document: PDF

PrintPrint this page
Share on Facebook Share on Twitter Share on LinkedIn Share on Digg Share on Reddit
| Email this pageE-mail
Document date: March 01, 1998
Released online: March 01, 1998

This report is part of The Urban Institute's Assessing the New Federalism project, a multi-year effort to monitor and assess the devolution of social programs from the federal to the state and local levels. Project codirectors are Anna Kondratas and Alan Weil. The project analyzes changes in income support, social services, and health programs. In collaboration with Child Trends, Inc., the project studies child and family well-being.

The project has received funding from the Annie E. Casey Foundation, the Henry J. Kaiser Family Foundation, the W. K. Kellogg Foundation, the John D. and Catherine T. MacArthur Foundation, the Charles Stewart Mott Foundation, the Commonwealth Fund, the Robert Wood Johnson Foundation, the Weingart Foundation, the McKnight Foundation, and the Fund for New Jersey. Additional funding is provided by the Joyce Foundation and the Lynde and Harry Bradley Foundation through a subcontract with the University of Wisconsin at Madison.

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.

The authors would like to thank Larry Thompson for his guidance, support, and wisdom over the many drafts of this paper. The authors would also like to thank Deborah Ellwood for generously providing her expertise in this area. Lisa Bernhardt at the DHHS for providing most of the fifty-state spending data, and Alan Weil, Ana Kondratas, Julia Matsen at the Congressional Budget Office, Iris Lav at the Center on Budget and Policy Priorities, and Carol Cohen at the Finance Project for their helpful comments.

Notes: As always, those wishing to print this report may find it easier to use the PDF Version.


States vary widely in the proportion of their population in need of governmental assistance and in their ability and willingness to finance services from their own revenues. These differences lead to wide variations in state spending on children.1 The federal government takes these differences into account when operating federal matching and fully federally funded programs, ranging from cash assistance to health care financing. Some of these programs are designed to help reduce differences in state needs and abilities; others provide assistance directly to families with children regardless of where they reside.

The federal government structures many of these programs with the intention of targeting a higher level of federal dollars to states with more children in need of services and/or less ability to raise revenue and a lower level of federal dollars to states with fewer children in need and/or a high revenue base. The result of this system is that the former group of states receives a greater proportion of federal funds relative to their state spending levels than the latter. In theory, these different levels of federal to state expenditures should lead to more equal state spending levels per poor child.

Federal programs influence total government spending patterns in several ways. In calculating what each state receives, some federal programs factor in the wealth (i.e., revenue) of a state, others focus entirely on a state's residents' needs, and some factor in both. In addition, some of these financing mechanisms require a state to share in a program's costs, making federal funding contingent on a state's spending levels. These financing systems lead to the following questions: How well has the federal government factored in a state's wealth and its residents' needs? Do federal programs tend to narrow state differences in spending on children? Finally, how does a state's willingness to pay for services affect the level of expenditures across the states?

In this paper we address these questions and, in particular, the role of the state and federal governments in narrowing state spending differences among noneducation children's programs.2 We found that there are significant differences across states in spending from state funds. We also learned that the federal government's funding mechanisms do substantially reduce the spending differences among states. However, major differences continued to exist. This expenditure variation was most apparent in spending from matching programs, because of states' differing willingness to spend monies on children's programs and because the funding structure does not directly account for the extremely high needs of some states. Finally, we found that the program in which differences in spending among the states were greatest was the old Aid to Families with Dependent Children (AFDC) program.

In this paper, we first analyze the variation among the states in need, capacity, and willingness to spend. Second, we describe and analyze the two federal funding structures—matching dollars and fully federally funded programs—to understand whether and how these programs help to equalize children's spending relative to a state's need and ability to finance services. We then discuss a third factor, a state's willingness to pay for services, and how it influences total spending. Finally, we focus on specific programs to analyze whether variations in spending occur within certain programs and to study the effects of the three indicators on program spending.

Although states spend considerable amounts on education programs, this analysis focuses solely on noneducation children's programs. In federal fiscal year (FFY) 1995, the programs covered in this analysis accounted for $126 billion in federal and state spending. We have included over twenty programs in our analysis. Medicaid, the old AFDC, the Earned Income Tax Credit (EITC), and Food Stamps were the four largest programs, accounting for $89.6 billion and 69 percent of the total. Figure 1 illustrates the distribution of children's spending by the major programs in 1995. A list of the programs and a description of how the data were collected and computed are given in the Appendix.

Variation in State Need, Capacity, and Willingness

Before analyzing spending differences, we first address how states vary in their economic and demographic composition. We look at state differences in need, ability to raise revenue, and willingness to spend.

Child Need

States vary in their proportion of children in need of services. Table 1 lists child poverty rates for the 50 states. In the median state, 18.6 percent of the children lived in families with incomes below the poverty line. In the 10 states with the highest rates of childhood poverty, an average of 28.5 percent of the children were poor, while in the 10 states with the lowest childhood poverty rates, only 12.7 percent were poor. If state spending targeted to poor children were to be the same for all poor children, those states with a larger fraction of their population living in poverty would have to spend more than those states with a small fraction of poor children. For example, on average, a high poverty rate state would have to spend 2.24 (28.5/12.7) times as much as a low poverty rate state of the same size in order to spend the same amount per poor child.

As part of our analysis, we focus on how much each state spends per child in households with incomes below the poverty line. This lets us standardize state spending relative to need when we examine expenditure differences across states. We used the number of children living in poverty, rather than the number of children in a state, as a measure of need because it is more reflective of the population served by the programs covered here. While some programs, such as Medicaid and the EITC, also serve low-income children above the poverty level, most programs assist children living below the poverty line.

Fiscal Capacity

Fiscal capacity also varies considerably across states. A state's fiscal capacity represents its potential to raise revenue for all public functions, including programs benefiting children. We use per capita personal income as an indicator of fiscal capacity because it is the most widely used and most easily understood measure of a state's ability to fund services.

We realize that states try to export their tax burden, and a few succeed. For example, if a state has a large oil or tourist industry it can raise large amounts of revenue without taxing its own citizens. In these cases, per capita income actually falls short of a state's true fiscal capacity.3 However, for the most part, overall taxes end up being paid by a state's own residents.

As shown at the bottom of table 1, in 1995 the 10 most affluent states had an average per capita income level of $26,972 compared to $18,231 for the 10 poorest states. With these varying fiscal capacities, states would have to tax their residents at very different rates to provide a similar level of spending. The 10 poorest states would have to tap their resources at a 48 percent higher rate than the 10 most affluent states to provide the same level of per capita expenditures.

Relationship between Need and Capacity

Not surprisingly, table 1 also shows that many of the states with higher rates of child poverty tended to have low fiscal capacities. This inverse relationship made it even more challenging for a high-need state—one with above-average poverty rates—to access the revenue needed to finance expenditures for its poor children.

We studied this inverse relationship by analyzing the combined effects of both child poverty and fiscal capacity. We measured a state's level of personal income (i.e., revenue-raising ability) per poor child. This method enabled us to determine whether a state's fiscal capacity or child poverty exacerbated or balanced out the effects of the other indicator. More importantly, it helped us determine which states have the highest and lowest ability to spend on children's programs. In table 1, the 50 states are listed in order of ability to spend from lowest to highest. For example, table 1 illustrates that New York, although it had a very high poverty rate, placed close to the median in personal income per poor child because of its high fiscal capacity (28 percent above the median). In contrast, Utah, with a very low child poverty rate, ranked thirtieth in personal income per poor child because it had one of the lowest fiscal capacities. Table 1 also shows that the "high-ability states," those with the 10 highest levels of personal income relative to children in poverty, had on average $760,296 of available personal income per poor child. This amount was almost three times higher than the 10 "low-ability states," those with the 10 lowest levels of personal income relative to children in poverty, which had an average of $255,367 of personal income per poor child.

Fiscal Effort

A third factor influencing state expenditure levels is a state's willingness to spend. We use fiscal effort as an indicator of a state's willingness to invest in public services. Fiscal effort is defined as revenue from state and local taxes divided by state personal income. Unlike need and fiscal capacity, fiscal effort is a matter for the state to decide.

Fiscal effort measures how intensively a state has tapped its available resources (i.e., total personal income). By determining total tax revenue relative to available personal income, we can account for differences in fiscal capacity across states. States and localities raise revenue for children's and other government services from three major sources: personal income tax, sales tax, and property tax.4

Table 1 shows the wide variations in fiscal effort. The fifty-state median effort was $115 per $1,000 of personal income. This level varied from a high of $155 per $1,000 of personal income in New York to a low of $94 per $1,000 of personal income in Alabama.

Relationship between Fiscal Effort, Fiscal Capacity, and Need

We found that a state's fiscal effort was not closely related to the average incomes of its residents. Figure 2 shows that of the 25 states that had a fiscal capacity above the median, 13 also exerted a fiscal effort above the median while 12 exerted an effort below the median. Of the 25 low fiscal capacity states (i.e., below the median), 12 exerted high efforts, and 13 exerted efforts below the median. This large variation and weak relationship between effort and capacity illustrate that how much revenue a state raises (i.e., how much it taxes its residents) is not solely dependent on a state's ability to raise resources (per capita personal income), but also on its willingness to finance government services.

We also found that a state's fiscal effort had an inverse relationship to its child poverty rate. States with high poverty rates tended to exert a low effort, while states with low poverty rates tended to exert a high effort. As figure 2 shows, of the 25 states with child poverty levels above the median, only six exerted fiscal efforts that were also above the median. In contrast, 19 out of the 25 states with poverty rates below the median exerted high levels of fiscal efforts (i.e., above the median).

Another indicator of the large variation in willingness among the states was the similarity in overall fiscal effort by the high- and low-ability states. Both groups placed close to the median in tax revenue per $1,000 of personal income. The high-ability states—those with high levels of personal income per poor child—raised an average of $116 per $1,000 of personal income compared to $113 per $1,000 for low-ability states. This suggests that there must be other variables, independent of a state's poverty rate and resident's average income, that determine a state's willingness to spend.

Federal Funding Mechanisms

The federal government attempts to address the state variations in need, capacity, and willingness to spend by assisting in the funding of children's programs. It provides aid to states' child populations through two major types of federally financed programs: matching programs and fully funded programs. These programs use different financing structures to attempt both to target funding to poor, high-need states and to narrow state differences in the amount of spending per poor child. The programs are grouped in the Appendix by financing structure.

Matching Programs

Matching programs require states to spend some of their own monies in order to receive federal funds for the specified programs. We define state matching spending as all state expenditures required to draw down federal money for the 12 largest federal matching programs. In FFY 1995, total matching expenditures were $66.6 billion, with $37 billion in federal funds and $29 billion in state spending.5 There are two types of financing structures: open-ended and closed-ended matching programs.

Open-ended matching programs, such as Medicaid, Foster Care, and the old AFDC program, enable states to obtain an unlimited amount of federal money as long as they match it with state dollars. This group of programs accounted for $58.5 billion of total matching expenditures. AFDC, Medicaid, and Foster Care together accounted for about 90 percent of all matching spending. The open-ended nature of these programs is due to the entitlement status which enables all eligible children to receive assistance.6

The state match for most open-ended programs is based on the Federal Medical Assistance Payment (FMAP), which is inversely related to a state's per capita income.7 For example, Mississippi, with the lowest per capita income, had a 21.4/78.6 percent state/federal matching rate. In contrast, Connecticut, with the highest per capita income, had a 50/50 matching rate (no state can have a higher share than 50 percent). For every dollar spent on FMAP-related expenditures, Mississippi would have spent 21.4 cents and Connecticut would have spent 50 cents. Because of the use of this FMAP system, poor states pay less per federal dollar than affluent states, and additional assistance is available when more people qualify.

States control total spending on open-ended matching programs by determining the level of benefits provided in the various programs. This state discretion means that a wealthy state could receive more federal money per capita than a poor state if it chose to set benefit levels much higher than the poor state.

Closed-ended matching programs work under the same principle as open-ended programs, requiring states to share in total costs. However, the matching rates for programs such as Child Welfare Title IV-B Parts 1 and 2 and the Maternal and Child Health (MCH) Block Grant are usually the same for all states. Also, total federal outlays are capped with poor, high-need states receiving a larger proportion than affluent, low-need states. For example, Child Welfare IV-B Part 1 funds are distributed to states on the basis of their population under 21 years of age and their per capita income. MCH funds are allocated to states based on the percentage of the nation's low-income children residing in the state. Since federal funds for these closed-ended programs are limited and targeted, states do not have the same ability to influence total spending as in open-ended matching programs.

Because of the different matching rates, states receive very different levels of total open- and closed-ended federal matching dollars relative to state dollars. Since these financing structures are primarily based on per capita income, we analyzed the 10 states with the highest fiscal capacity and the 10 states with the lowest. In FFY 1995, the 10 poorest states spent an average of $106.5 million and received $246.9 million in federal aid. In contrast, the 10 most affluent states spent an average of $984.8 million and received $1.024 billion in federal aid. Hence, for every dollar spent, the poor states received $2.32 in federal aid while the rich states received only $1.04.

Programs Funded Entirely by the Federal Government

Another set of programs which serve a very important role in providing assistance to low-income children are those that are financed completely by the federal government. These "fully federally funded" programs provided an additional $59.4 billion for services to children in 1995, accounting for 47 percent of total state and federal spending on programs directed to low-income children.

The fully federally funded programs examined in this paper target money to children in need. This structure helps reduce the financial burden in those states that have a high concentration of child poverty. These programs can be separated into two groups: federal block grants and federally financed programs, and federally funded grants to individuals.

Federal block grants and federally financed programs include programs such as Head Start, School Lunch, and the Child Care Development Block Grant (CCDBG). Outlays for these programs are established at the beginning of the fiscal year, meaning a state's allocation is capped. In addition to poverty levels, outlays are also sometimes based on the level of federal assistance when the block grant was established. For example, the Head Start allocation formula was based on three factors: the state's allocation in 1981, the number of children in poverty under age five, and the number of AFDC families with children under 18.8

Federally funded grants to individuals, such as the SSI, Food Stamps, and EITC programs, are different from block grants in that the federal government sets all of the policies and does not require states to use any of their own dollars for services.9 Also, these programs, unlike block grants and outlays, are not capped, so federal spending rises automatically when more children and families qualify.

Spending on fully federally funded programs is tied closely to state need. The 10 states with the highest child poverty levels received an average of $15.76 per $1,000 of personal income in federal aid. This amount was more than twice that received by the states with the lowest poverty levels, only $7.59 per $1,000 of personal income.

Does Federal Funding Narrow State Spending Differences?

These various federal fund mechanisms have the potential to direct proportionately more federal dollars to those states with high need and low fiscal capacity, thus narrowing state spending differences. We compared the spending levels of the 10 states with the highest levels of personal income per poor child to the 10 states with the lowest levels of personal income per poor child, referred to as the 10 "high- and low-ability" states. We first looked at the matching program expenditure structure to understand its effects on differences in spending levels between these two groups of states. Then we analyzed spending differences when the other federal financing structures are included along with total matching spending.

State Matching Spending

Table 2 shows that in FFY 1995, the median amount the 50 states spent on federal matching children's programs was $3.39 per $1,000 of personal income. The table also shows that the median 50-state spending for these programs was $1,717 per poor child. Spending levels of state funds relative to both fiscal capacity and child need varied considerably across the 50 states. For example, New York spent $4,649 per poor child, more than 11 times as much as Mississippi's spending level of $411 per poor child.

Table 3 shows that the top 10 high-ability states spent an average of $3,304 per poor child, or 92 percent above the 50-state median. In contrast, the 10 lowest ability states' average spending level of $767 places them 55 percent below the median. Table 3 also shows that in state-only dollars the 10 high-ability states spent 46 percent more per $1,000 of personal income than the low-ability states ($4.38 to $3.01). This difference means that even when accounting for the different levels of personal income (i.e, capacity), the high-ability, wealthy states still spent more than the low-ability, poor states.10

Total Matching Spending

With the federal matching dollars added in, the median matching spending per poor child rose to $4,162, over twice the average state fund's spending level. In some states, this inclusion increases the total by even greater factors: in Mississippi and New Mexico, state plus federal dollars is almost four times greater than the state's share of total expenditures per poor child.

In addition, adding federal matching dollars to the state funds decreased expenditure differences between the ten high- and the ten low-ability states. Relative to need, the expenditure gap between the two groups narrowed to 184 percent from the 331 percent gap using state-only money. However, in actual dollars the spending difference between the two groups was still large, with the 10 high-ability states spending an average of $7,041 per poor child compared to $2,477 by the 10 low-ability states.

These continued disparities in spending per poor child are due in part to the federal matching funding structure. While it provides a higher match rate to poorer states, this matching mechanism still requires states to spend monies in order to receive federal funds, and many low-ability states are either unable or unwilling to spend significant state monies on these programs.

Figure 3 shows the average children's spending level by funding source in the 10 high- and 10 low-ability states. The figure helps illustrate how the matching structure affects the expenditure levels of these two groups of states. The federal matching funding provided the low-ability states with 2.23 times as much as their state-only spending, while the high-ability states received almost the same amount of federal matching dollars as state dollars (see previous section). However, since the low-ability states spent so little on the matching programs, this high level of federal matching dollars relative to state-only money only worked out to an average of $1,710 of additional money per poor child ($2,477 - $767). In contrast, the additional federal dollars provided the high-ability states with an extra $3,737 per poor child ($7,041 - $3,304). This difference illustrates that, since these programs require states to first spend monies in order to receive the federal matching funds, states' decisions influence overall spending levels. Therefore, although the federal money reduced the large state-only spending differences in percentage terms, the spending differences that remain are still significant. This result raises two questions: Is the matching structure factoring in the child poverty needs of poor states at a high enough level, and are the states with high poverty rates willing to spend only a limited amount of funds on these programs?

Total Spending

Spending under fully federally funded programs flows into low-capacity, high-need states, however, without regard to state spending decisions. This reduces the variation among states in spending per poor child. When fully federally funded programs are included, the high-ability states spent only 82 percent more than the low-ability states, down from 184 percent. This inclusion also raises the 50-state median to $8,571 per poor child. As figure 3 shows, total spending in the 10 states with the lowest ability was $6,588 per poor child compared to $12,003 in the 10 states with the highest ability.

Figure 3 also shows how the low-ability states rely on the fully federally funded programs to a much larger degree than the high-ability states. Of the funds spent on children's programs, the low-ability states received 62 percent from fully federally funded programs, whereas the high-ability states received only 41 percent of their funds from these sources.

Supplementing and Targeting

Federal dollars did not completely eliminate spending differences for poor children in a state. They did, however, supplement state spending per poor child in poor states with high needs at a higher level than in affluent, low-need states. In addition, the fully federally funded programs increased spending among the states with the highest levels of need even more dramatically. For example, although federal funds increased spending per poor child in the 10 lowest-ability states by 761 percent, Mississippi's expenditures per poor child increased by over 1,400 percent and Alabama's expenditures increased by over 1,100 percent over their state-only funding levels. In contrast, the federal funding structures increased spending per poor child in the 10 states with the highest ability by only 263 percent.

The additional federal dollars also significantly increased the fiscal capacity of the low-ability states. For example, table 3 shows that the fully federally funded program dollars provide the low-ability states with an additional $16.75 per $1,000 of personal income, whereas the high-ability states only receive an average of $6.56 in extra expenditures per $1,000 of personal income. In addition to expanding the low-ability states' capacity, the federal government provides the two groups of states with an almost equal number of fully federally funded dollars per poor child ($4,962 to $4,111).11

Willingness to Spend

When we studied the degree to which federal programs narrowed state differences in spending on poor children, we saw that spending differences did decrease but that variations in expenditures continued to exist. It may be that the federal government has not accounted for the considerable need of certain states at a rate high enough to make up for the state spending differences. In addition, expenditure differences may be due to the federal funding system's reliance on a state's willingness to pay. We addressed a state's willingness to pay for services by analyzing how fiscal effort affects a state's level of total matching spending per $1,000 of personal income.12 This expenditure measurement enabled us to analyze how much a state spent on children's services relative to its capacity and regardless of its need.

Fiscal Effort Relative to Matching Spending

At the beginning of this paper, we illustrated that very little relationship exists between fiscal effort, fiscal capacity, and a state's child poverty level. Consequently, fiscal effort varied considerably across high-ability and low-ability states. Differences in willingness to spend within the two groups of states help to explain variations in the ability of federal matching dollars to target states with high child poverty rates and low per capita income levels. For example, Texas and South Carolina, both low-ability states, exerted such low fiscal efforts that they received some of the lowest levels of total matching spending per $1,000 of personal income. In contrast, Massachusetts and Vermont, both high-ability states, exerted above-median fiscal efforts, enabling them to receive a higher level of total matching spending compared to the low-ability states.

These variations in matching expenditures relative to fiscal effort within the high- and low-ability states illustrate how spending is influenced not only by fiscal capacity and child poverty levels but also by a state's willingness to spend. The federal government, by requiring states to pay a portion of total costs, has built fiscal effort into its financing structures. This structure rewards states that are willing to pay. For example, several of the high-ability states had above-average total matching spending levels due in part to their high fiscal efforts. In addition, New Mexico and Arizona, both low-ability states, had very high levels of total matching spending (44 and 20 percent above the median) per $1,000 of personal income partly because they exerted high efforts (15 and 8 percent above the median). Interestingly, these two states still fall far below the median (32 and 17 percent respectively) in total matching spending per poor child. This result again illustrates that the federal funding system, even though it has targeted dollars to high child poverty states, might not account for need at a high enough level.

Categorical Spending

We analyzed how the interaction between the federal funding systems and states' efforts affects the level of spending in various programs. We studied three welfare programs for children: AFDC, Food Stamps, and Medicaid. Tables 4 and 5 show state spending variations for the three programs. The tables show that median total state and federal AFDC spending per poor child was $1,264. This amount constituted close to a third of total state and federal matching expenditures. Combined AFDC and Food Stamps spending per poor child was almost double this amount. Medicaid expenditures were also an important portion of spending on children, comprising approximately 37 percent of total state and 38 percent of total matching spending.13

AFDC and Food Stamps

AFDC expenditure amounts varied more across states than any other matching program. As mentioned earlier, total state and federal matching program expenditures per poor child were 185 percent greater in the high-ability states than in the low-ability states. However, as table 6 shows, in the AFDC program, the 10 states with the highest ability spent an average of 286 percent more in total federal and state dollars per poor child than the lowest-ability states.

It is important to note that state policy discretion played a role in these large differences. AFDC spending depends in part on the level at which each state sets its benefits. A state with a high benefit level and a high matching rate (e.g., 50 percent) could receive a higher proportion of federal dollars than a state with a low benefit level and low matching rate. This discretion demonstrates the important role a state's willingness to spend played in equalizing spending differences. Food Stamps, in contrast, are entirely federally funded and provide assistance to poor families, regardless of their state of residence, so the Food Stamps program provides additional federal dollars to residents of low-ability states and states that do not exert a high effort. In a sense, Food Stamps, as a federally funded program, serves as a substitute for a state's investment in children.

When AFDC and Food Stamp spending is combined, the spending gap between the high- and low-ability states declines to 113 percent. This decrease is also due to the unique interaction between AFDC and Food Stamps. While Food Stamps are not directly affected by state policy decisions, a child's or family's benefit level for Food Stamps is directly affected by AFDC benefit decisions. A family with a lower AFDC benefit level would have received a larger Food Stamp allocation. While AFDC state policy decisions affect Food Stamp spending, the opposite relationship could also hold true. Federally established Food Stamp eligibility and benefits levels could actually cause the large variations in AFDC state spending. Some states might lower their AFDC grants because they realize Food Stamps will serve as substitute income for any AFDC benefit reduction.

The interaction between AFDC and Food Stamps illustrates the important role federal funding plays in decreasing spending differences. While AFDC state spending varied considerably, Food Stamp expenditure levels were similar across the high- and low-ability states ($1,538 and $1,250 per poor child respectively). However, Food Stamps spending constituted three times the AFDC expenditure level in the low-ability states and less than 60 percent of AFDC spending in the high-ability states. Consequently, since the low-ability states received a large amount of Food Stamp dollars relative to their AFDC spending level, their total AFDC and Food Stamp spending increased more than that of the high-ability states. This difference caused the overall AFDC and Food Stamps spending gap between the 10 high- and low-ability states to decrease.


Medicaid expenditures on children also did not follow the total matching spending trends noted earlier. In contrast to AFDC, state discretion caused smaller state differences. High-ability states spent an average of 107 percent more per poor child than the low-ability states. This variation was a lot smaller than the total state and federal matching spending differences between these two groups of states (185 percent) and the total AFDC spending differences (289 percent).

Despite significant variation in spending in the Medicaid program, federal policies have helped decrease the variations among the states. In particular, the federal government has expanded child eligibility, forcing low spending states to increase their costs. In addition, the federal government has required states to set reimbursement rates at a level that meets the expenses of an "economically and efficiently operated hospital." This policy has caused state hospital reimbursement rates to vary less among the states than they would have otherwise.

Reliance on Different Programs

Figure 4 shows total spending in the high- and low-ability states by type of program. The figure shows that the two groups of states relied on different programs for their funding. Among the high-ability states, the two largest programs were the Medicaid and AFDC programs, which together accounted for 43 percent of the funding. The Food Stamps and EITC programs provided the high-ability states' residents with only 27 percent of the total funding. The low-ability states' residents, on the other hand, received most of their assistance through the Medicaid, EITC, and Food Stamps programs. These three programs accounted for 61 percent of their funding. The most striking difference between the two groups of states is that the low-ability states received a much greater proportion of their funding from the Food Stamps and EITC programs (42 percent) than did the high-ability states. These two programs are 100 percent federally funded and do not require a state match. In addition, the EITC is administered through the federal tax system, and states have no role in this program. Lastly, as discussed above, the program which varies the most between the two groups of states is the AFDC program.

Conclusions and Implications

The federal government plays an important role in funding programs for low-income children. It has targeted dollars to states with the least ability to pay and the highest need for services, and has reduced the discrepancy in spending on children between the highest- and lowest-ability states. When federal funding is included, high-ability states spend only 1.82 times as much as low-ability states, rather than 4.3 times as much. Without these various funding mechanisms, low-capacity, high-need states would have to exert a much larger fiscal effort to provide similar services as provided in high-capacity, low-need states. Or, children would likely receive very different levels of assistance depending on where they lived.

Besides addressing need and capacity of states, the federal government has expected states to assist in financing children's programs. In fact, the federal government appears to be balancing need and capacity with state participation, since approximately one-half of the funding it distributes for children's programs is in matching programs and one-half is in fully federally funded programs.

Even with the high level of federal government funding, there are still significant spending differences among the states. This result is due to the extremely high child poverty levels of some states, which means that even with high levels of federal assistance relative to state spending, these states were still unable to fund services equally. Differences are also caused by the structure of federal matching programs in which federal funds are allocated based on a state's willingness to spend. By using this funding mechanism to distribute funds, the federal government was less effective in targeting money to high child poverty states, since some states did not spend enough to pull down sufficient federal funds. Lastly, some of these state spending variations may reflect differences in the cost of living among the states and, to a lesser extent, the cost of providing services.

If the only goal were to further narrow state spending on children's programs, the federal government could mandate spending on children as is done in the Medicaid program, spend more on fully federally funded programs, or decrease the required match for low-ability states. These last two remedies, however, would go against the federal government's other goal of achieving state participation in the funding of children's programs. In addition, it may be that reducing the match does not increase total spending, as it may lead some states to spend even less on these programs.

It is not clear what determines the amount states are willing to spend on children's programs. Contrary to what one might expect, a state's fiscal effort is not dependent on the wealth of its residents and is actually slightly inversely related to its child poverty rate. Thus, there must be other factors, independent of a state's need and capacity, that determine a state's willingness to spend on children's programs. This difference in willingness to spend was most evident in the old AFDC program, where expenditure amounts varied across the states more than in any other matching program. The fact that states showed the most variation in spending on this program is probably due to its nature and structure. In the old AFDC program, states were given flexibility to determine eligibility levels and benefit amounts and therefore were more able to determine the overall cost of the program. The AFDC program also provided individuals grants for housing costs, and thus the variation in grant levels was probably reflective of the varying costs of housing across the country. In addition, since AFDC was the main welfare program, its funding was probably more reflective of the states' differing political attitudes toward welfare programs and thus varied more than state funding of other programs.

The Personal Responsibility and Work Opportunity Reconciliation Act of 1996 converted AFDC into a block grant known as Temporary Assistance for Needy Families (TANF). This new block grant, unlike other fully federally funded programs, does not target states with high need. Instead, it sets allocations to states' historic AFDC spending levels. Since the old AFDC program actually exacerbated the differences in spending between rich and poor states, TANF freezes into place the wide expenditure differences among states. In addition, because states have greater flexibility in spending TANF monies and the new law requires states to maintain spending at only 75 percent or 80 percent of its prior levels, it is possible we will see even greater discrepancies in state spending.

This switch to a block grant, therefore, does not address the fact that the AFDC was the program least able to target high-need states. In the future, however, this change could lead to federal funding policy revisions that level out spending differences. Since the federal government can now determine overall TANF spending, Congress could decide in future years to restructure the funding allocations to target states with high child needs. Although currently this outcome appears unlikely, Congress has enacted similar financing changes in the past to other block grant programs.14 If TANF funding became more targeted, overall spending variations would be likely to decrease.

Programs, Data Sources,
and Methodology

Matching Programs

All sources, unless otherwise noted, are from the Office of Legislative Affairs and Budget (LAB) in the U.S. Department of Health and Human Services (DHHS) for FFY 1995.


Adoption Assistance: Provides payments to parents who adopt a special needs child who is AFDC- or SSI-eligible. Also provides one-time assistance to parents adopting non-AFDC- or SSI-eligible children. Federal reimbursement rate is FMAP for payments, 50 percent for administration, and 75 percent for training.

AFDC: Provides monthly cash assistance payments to low-income families. Eligibility based on income and assets. Spending includes assistance payments and administration and excludes child support retained revenue. Federal reimbursement rate is FMAP for payments, 50 percent for administration.

AFDC Emergency Assistance: Provide emergency payments to AFDC-eligible families who have depleted their resources. Federal reimbursement rate is FMAP.

AFDC/JOBS Child Care: Available to AFDC recipients who need child care in order to accept employment, remain employed, or participate in employment activities. Spending does not include administration. Federal reimbursement rate is FMAP.

Child Support: Provides states with funds to enforce child support orders determined in court. The program requires the state to provide services to both AFDC and non-AFDC families. Federal reimbursement is 66 percent for most administrative costs and 90 percent for information systems.

Foster Care: Provides maintenance payments to AFDC-eligible children who are removed from their homes and placed in foster care homes or other facilities. Spending includes benefits, administration, and training. Federal reimbursement rate is FMAP for payments, 50 percent for administration, and 75 percent for training.

Medicaid: Provides health care to low-income persons. Only spending for persons under the age of 19 is included in our analysis. Data is based on Urban Institute calculations using data from the Health Care Financing Administration 64 and 1082 forms. Federal reimbursement rate is FMAP.

Transitional Child Care: Available to former AFDC recipients who need child care in order to continue working. Spending does not include administration. Federal reimbursement rate is FMAP.


AFDC JOBS: The Family Support Act of 1988 required states to have an employment, education, and training program to help prevent long-term welfare dependency. Federal reimbursement is set at various rates, including the FMAP, for different portions of a state's allocation. Spending is based on FSA-331 and ACF-332, Administration for Children and Families, DHHS.

At-Risk Child Care: Provides funding to low-income families who are not enrolled in AFDC, need child care in order to work, and would be at risk of becoming eligible for AFDC without child care. Reimbursement is based on the FMAP, capped at state allotment.

Child Welfare (Title IV-B Part 1): Provides funding to support states' efforts to keep families together, reunify families, and find children adoptive homes. There are no income guidelines. Federal reimbursement is 75 percent for all services, capped at state allotment.

Family Preservation (Title IV-B Part 2): Provides family preservation services to children and families at risk or in crisis. Federal reimbursement is 75 percent for all services, capped at state allotment.

Maternal and Child Health (MCH): Provides states with funds to develop and administer programs for the care of mothers and children. Reimbursement is $4 for every $3 spent by the state, capped at state allotment. Federal data comes from the Office of Operations and Management, DHHS.

Fully Federally Funded Programs

All sources, unless otherwise noted, are obligations from the Office of Management and Budget, Budget Information of the United States, FFY 1997.

Federal Block Grants and Fully Federally Financed Programs

Child Nutrition Programs: This includes expenditures for the School Lunch, School Breakfast, and Child and Adult Care Food programs.

Child Care Development Block Grant (CCDBG): Provides states with funds to improve quality and availability of child care. Twenty-five percent of funds must be used for early childhood and before- and after-school child care. Data is FFY 1995 allocations reported by LAB, DHHS.

Head Start: Provides comprehensive child development services to primarily low-income children ages 3 to 5.

JTPA Titles IIB and IIC: Provides employment and training funds for economically disadvantaged youth ages 16 to 21. IIB is for summer job training and IIC is for year-round training.

WIC: Provides supplemental food, health care referrals, and nutrition education at no cost to low-income pregnant and postpartum women, infants, and young children up to 5 years of age.

Federally Funded Grants to Individuals

EITC: This program enables parents with modest earnings, including AFDC parents who leave welfare because of work, to receive a cash supplement. Data are for tax year 1994 from the Internal Revenue Service and collected by the Center on Budget and Policy Priorities, and include payments to all families, including individuals.

SSI for Children: Provides benefit payments to needy blind and disabled children. Spending is an estimate for the federal fiscal year, based on spending in June and December of each year. Includes federal spending and also state supplements for states in which the state supplement is federally administered. Data is from the Office of Research, Evaluation, and Statistics, Social Security Administration (SSA).

Food Stamps: Provides benefit payments for purchase of food items. Includes only payments to households with children. Based on Urban Institute tabulations using Food Stamp Quality Control data and tabulations by Food and Consumer Service, U.S. Department of Agriculture.

Other Sources

Per Capita Personal Income and Total Personal Income: U.S. Department of Commerce, Bureau of Economic Analysis, 1995 calendar year estimates revised as of October 1996.

Child Poverty Rates: Current Population Survey (CPS), three-year average (March 1994-March 1996 where 1994 is the center year) edited using the Urban Institute's TRIM2 microsimulation model.

Number of Children in Poverty: U.S. Census Bureau, Office of Statistics, ST-96-10 Estimates of the Population of the U.S. for Selected Age Groups. Total number of children 17 and under in a state was multiplied by its percentage of children in poverty (see Child Poverty Rates for explanation of percentages).

Total Tax Revenue (Fiscal Effort): U.S. Census Bureau, Government Division, Federal, State, and Local Government Finances. Revenue raised by state and local governments from the public, excluding charges, liquor store revenue, insurance trust revenue, utility revenue and money received from issuance of debt, liquidation of investments, and agency and private trust transactions. Data are from FY 1994, the latest year available for total state and local revenue. Total revenue is divided by total state personal income for calendar year 1993.


Decision Rules

This paper includes spending on noneducation children's programs. We developed three main rules for defining programs that would be included in the analysis: (1) We included spending on programs that were explicitly designed to help or "treat" children. Examples of these programs include child care, Head Start, and foster care. (2) We included spending on adults in programs where they received monetary assistance or services only because of the presence of a child. Examples of these programs include WIC, AFDC, and EITC. Although the EITC is provided to low-income individuals as well as families, the income limits are such that few individuals qualify. In addition, we included Food Stamp expenditures for families with children under this rule since it is used as an augmentation or substitution for the AFDC program. (3) We included spending on children's portion of expenditures in programs that benefited both children and adults. These programs include Medicaid spending and SSI.

We also excluded programs that were designed for adults but indirectly benefit children such as JTPA IIA. We generally included administrative expenditures for a program if the majority of such spending was for service provision. For example, much of the spending on AFDC administration is for eligibility determination workers, and therefore administration was included in our AFDC numbers.

Unfortunately, we did not include three important federal programs under these decision rules: the Title XX Social Services Block Grant (SSBG), Mental Health Block Grant (MHBG), and housing subsidies. For SSBG, states have the discretion to use funds for both children and adult populations. Currently, no accurate data exist on state spending by population on SSBG. We excluded housing subsidies and MHBG for similar reasons. Most programs do not collect data on expenditures for child or family populations.

Data Collection Process

We compiled expenditure information on these children's programs from various federal departments, including the Department of Health and Human Services and the Office of Management and Budget. State expenditures for matching programs were computed based on a state's matching rate or the various matching rates within a given program. We performed this computation because matching program reporting systems, in most cases, included federal expenditure totals, but we did not ask states to submit information on a state's matching expenditures (AFDC did require state data).

This state expenditure total provides a very good estimate but is still a lower-bound indicator of what states spent on children's programs. It does not include state expenditures for state-initiated programs or for expansion of services to children not eligible for federal programs. For example, states tend to invest state-only money to expand services in child care and child welfare. In addition, states pay for almost all the costs of juvenile justice programs. A future Assessing the New Federalism report, focusing on 13 states, will analyze total state spending on children's programs by collecting data directly from the state budget offices.

For purposes of this report we counted obligations to states for fully federally funded programs as actual state expenditures, even though states usually had up to two years to use the funds. For example, the federal government reported state expenditure outlays for JTPA IIB and IIC but did not know how much money the state spent in a given year.

In the case of the JOBS program, this difference between obligations and actual expenditures was readily apparent. DHHS had two expenditure reporting systems for JOBS, one that presented information on federal obligations to states by each matching rate and another that provided data on combined state and federal expenditures per month. In many instances, large discrepancies existed between these two sources because several states had drawn down money in a given fiscal year but had spent it in another. For example, a state could have drawn down $50 million in FFY 1995 and $30 million in FFY 1996. Yet it could actually have spent only $25 million in SFY 1995 and $35 million in SFY 1996 and have held the remaining $10 million for the subsequent state fiscal year. We therefore developed a methodology that cross-referenced the two reporting systems and resolved these differences. We first multiplied the average monthly combined federal and state expenditure total by 12 to arrive at a 12-month total. Then we used the other DHHS reporting system on obligations to disaggregate the 12-month total into state and federal funds.


1. This paper expands on previous reports by Steven Gold, Elizabeth Davis, Deborah Ellwood, et al., and Carol Cohen and Martin Orland, which also analyzed state spending on programs for children and families. These papers focused on how spending patterns changed between 1980 and 1992.

2. Equal spending does not mean equal services or even equal standards of living. However, because of the lack of comprehensive data on the cost of providing services and the cost of living across the states, we consider only expenditures when comparing state spending on noneducation children's programs.

3. Economists have developed measures which account for states' abilities to export tax burdens. Total Taxable Resources (TTR) and Representative Tax System (RTS) are two of these indicators. See Steven Gold et al., How Funding of Programs For Children Varies Among the 50 States, Center for the Study of the States (Albany, New York: The Nelson A. Rockefeller Institute of Government, State University of New York, January 1996), pp. 8-10, 16.

4. We exclude charges from our analysis because children's services are primarily funded with general purpose revenue.

5. We did not account for state investment in state-initiated programs or on state expansions to cover children not eligible for federal programs. Therefore, the state totals should be considered very good estimates but still a lower-bound indicator of what states spent on children's programs. A future Assessing the New Federalism report focusing on 13 states will analyze total state spending on children's programs.

6. Our analysis was based on FFY 1995 data. With the passage of the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA), the AFDC program was converted into a block grant beginning in FFY 1997. Under the PRWORA, states receive a block grant based on historic AFDC expenditure levels, and the act requires states to maintain their state spending at 75 or 80 percent of prior levels.

7. The state matching requirement for the Child Support program is not based on the FMAP. Its federal reimbursement is 66 percent for most administrative costs and 90 percent for information systems. In addition, administrative spending for several matching programs is set at 50 percent for all states.

8. This was the formula before the passage of the PRWORA and the elimination of the AFDC program. It is unclear how the Head Start monies will be distributed in the future.

9. States may choose to provide a supplement to the federal SSI payment, and currently all but eight states provide them. States are required to maintain either their previous year's total supplementation expenditures or their SSP payment levels as of March 1983, or risk losing their Medicaid reimbursement. States also have the option of administering their supplementation payments themselves or contracting with the Social Security Administration. States also pay a portion of Food Stamp administrative expenses.

10. In this paper, we have only accounted for state matching spending, which means we cannot accurately conclude that low-ability states are "less willing" to spend than high-ability states. These high-ability states had to exert a higher spending effort to achieve the same results since they have a lower matching rate (e.g., 50/50 state-federal). In contrast, low ability states received a high matching rate (e.g., 30/70 state-federal), which meant they could exert a lower effort to receive the same level of total state and federal matching dollars as the high-ability states. Hence, we should expect variations in state matching spending per $1,000 across the high- and low-ability states.

11. High-ability states still received more per poor child than low-ability states due to the various funding formulas which take into account several measures including historical allocation levels and number of children in the state, as well as the number of children in need.

12. We compare fiscal effort to total matching spending, rather than state or federal alone, because the combined total accounts for the different levels of required investment and reimbursed federal dollars.

13. We only included Medicaid expenditures on children, which accounted for approximately 16.8 percent of total federal and state Medicaid spending in FFY 1995.

14. For example, in 1992, Congress continued to change and expand the funding for the Substance Abuse and Mental Health Block Grants. In addition to increasing state expenditures, it also restructured the funding allocations which had been based on historical spending levels to target additional dollars to states with a higher need for services.

Tables, Figures and Charts

TABLE 1: 50 State Child Poverty, Per Capita Personal Income,
Personal Income Per Poor Child and Fiscal Effort

(states listed by lowest to highest personal income per poor child)

  Child Poverty   Per Capita Personal Income   Personal Income Per Poor Child   Fiscal Effort Per $1,000
of Personal Income (a)
State % Rank   $ Indexb Rank   $ Indexb Rank   $ Indexb Rank
MISSISSIPPI 34.4 49   16,683 77 50   173,391 37 50   113.52 99 29
LOUISIANA 35.9 50   18,981 88 40   186,474 40 49   104.17 91 44
NEW MEXICO 29.8 48   18,206 84 47   209,959 45 48   131.70 115 5
WEST VIRGINIA 29.1 47   17,687 82 49   260,064 56 47   114.02 99 28
KENTUCKY 27.9 46   18,849 87 42   269,173 58 46   114.98 100 25
SOUTH CAROLINA 26.1 45   18,998 88 39   283,863 61 45   107.74 94 40
OKLAHOMA 24.6 39   18,580 86 44   284,135 61 44   109.27 95 37
ARKANSAS 24.4 37   18,101 84 48   286,540 61 43   106.21 92 43
TEXAS 25.8 43   21,206 98 30   291,168 62 42   107.96 94 39
ARIZONA 24.9 41   20,489 95 35   308,906 66 41   124.42 108 12
ALABAMA 23.8 36   19,181 88 38   319,726 68 40   94.32 82 50
IDAHO 19.3 29   18,906 87 41   335,112 72 39   115.13 100 24
CALIFORNIA 25.6 42   24,073 111 11   342,955 73 38   110.65 96 34
TENNESSEE 24.5 38   21,038 97 32   348,791 75 37   96.95 84 48
SOUTH DAKOTA 19.4 30   19,576 90 37   357,776 77 36   101.98 89 45
MONTANA 18.8 26   18,445 85 45   364,107 78 35   114.30 99 27
FLORIDA 25.9 44   23,061 106 20   383,336 82 34   107.66 94 41
MICHIGAN 22.0 35   23,915 110 16   413,489 88 33   124.48 108 11
OHIO 21.0 34   22,514 104 21   420,510 90 32   112.42 98 30
NORTH CAROLINA 20.1 32   21,103 97 31   427,864 92 31   114.95 99 26
UTAH 12.4 5   18,232 84 46   428,657 92 30   122.05 106 14
MISSOURI 19.8 31   21,819 101 24   429,061 92 29   96.16 84 49
GEORGIA 19.1 28   21,741 100 25   435,019 93 28   112.35 98 31
NEW YORK 24.6 40   27,678 128 4   452,322 97 27   155.36 135 1
ILLINOIS 20.8 33   25,225 116 8   464,146 99 26   110.32 96 35
KANSAS 17.3 22   21,841 101 23   470,469 101 25   117.31 102 20
OREGON 18.2 24   21,611 100 26   475,199 102 24   118.60 103 18
WYOMING 15.0 15   20,684 95 34   487,566 104 23   128.99 112 7
NORTH DAKOTA 14.2 11   18,625 86 43   493,151 106 22   119.13 104 17
INDIANA 16.3 19   21,433 99 28   516,386 111 21   111.35 97 33
IOWA 15.9 18   20,921 97 33   517,089 111 20   126.00 110 9
MAINE 15.8 17   20,105 93 36   517,254 111 19   125.26 109 10
PENNSYLVANIA 18.8 27   23,558 109 19   522,776 112 18   110.29 96 36
WASHINGTON 17.3 21   23,774 110 18   534,302 114 17   121.24 105 15
RHODE ISLAND 17.3 23   23,844 110 17   580,368 124 16   117.46 102 19
WISCONSIN 14.4 12   22,261 103 22   589,894 126 15   137.34 119 3
MINNESOTA 14.8 14   23,971 111 14   604,276 129 14   131.46 114 6
NEBRASKA 12.9 6   21,447 99 27   619,943 133 13   117.05 102 21
NEVADA 14.7 13   24,390 113 10   670,936 144 12   108.57 94 38
HAWAII 14.1 9   24,590 113 9   684,159 146 11   137.12 119 4
MASSACHUSETTS 17.2 20   28,021 129 3   697,386 149 10   116.39 101 22
VERMONT 12.1 3   21,231 98 29   697,594 149 9   128.63 112 8
MARYLAND 15.0 16   26,333 121 5   703,898 151 8   112.00 97 32
CONNECTICUT 18.4 25   31,776 147 1   719,285 154 7   122.99 107 13
VIRGINIA 13.7 8   23,974 111 13   724,402 155 6   101.28 88 46
COLORADO 12.4 4   23,961 111 15   754,981 162 5   107.17 93 42
ALASKA 10.0 1   24,002 111 12   772,862 165 4   141.76 123 2
DELAWARE 13.2 7   26,273 121 6   811,333 174 3   115.70 101 23
NEW HAMPSHIRE 11.8 2   25,587 118 7   854,353 183 2   99.79 87 47
NEW JERSEY 14.1 10   29,848 138 2   866,864 186 1   120.70 105 16
50 State Median 18.6     21,676 100     467,308 100     114.96 100  
Top Ten 12.7     26,972 124     760,296 163     134.36 117  
Bottom Ten 28.5     18,231 84     255,367 55     101.57 88  
        Source: Urban Institute calculations based on data from U.S. Department of Commerce, Current Population Survey, and the U.S. Census Bureau, Government Finance Division.
          a. Children in families with incomes less than the Federal Poverty Level as a percent of total children in the state.
          b. Based on 50-state median of 100
          c. Total tax revenue is FY 1994.
        Note: Child poverty and the number of children in poverty are 1994 estimates based on three-year averages.

TABLE 2: Total Spending on Children's Programs in 1995

  State Funds   State & Federal Matching Funds b   Total State & Federal Funds c
  Per $1,000 Personal Income Indexa Per Poor Child Indexa   Per $1,000 Personal Income Indexa Per Poor Child Indexa   Per $1,000 Personal Income Indexa Per Poor Child Indexa
ALABAMA 1.85 55 593 35   5.85 62 1,870 45   22.45 117 7,177 84
ALASKA 8.18 242 6,322 368   16.95 181 13,101 315   25.05 131 19,361 226
ARIZONA 3.87 114 1,195 70   11.23 120 3,470 83   23.69 124 7,319 85
ARKANSAS 2.39 71 685 40   8.00 85 2,291 55   24.35 127 6,977 81
CALIFORNIA 7.65 226 2,622 153   15.61 166 5,352 129   25.63 134 8,789 103
COLORADO 2.97 88 2,243 131   6.48 69 4,895 118   13.56 71 10,239 119
CONNECTICUT 4.62 136 3,323 194   9.52 101 6,845 164   13.51 70 9,715 113
DELAWARE 4.28 126 3,471 202   8.97 96 7,280 175   16.51 86 13,398 156
FLORIDA 3.99 118 1,530 89   9.13 97 3,500 84   19.50 102 7,474 87
GEORGIA 3.37 99 1,464 85   8.84 94 3,844 92   21.20 111 9,223 108
HAWAII 6.79 201 4,648 271   14.03 150 9,599 231   22.63 118 15,482 181
IDAHO 1.95 58 653 38   5.95 63 1,995 48   16.08 84 5,387 63
ILLINOIS 4.84 143 2,249 131   9.93 106 4,608 111   18.37 96 8,526 99
INDIANA 2.83 84 1,461 85   7.35 78 3,798 91   15.73 82 8,122 95
IOWA 3.33 98 1,721 100   8.73 93 4,515 108   16.17 84 8,363 98
KANSAS 2.93 87 1,378 80   7.29 78 3,432 82   15.61 81 7,345 86
KENTUCKY 3.11 92 838 49   9.39 100 2,529 61   23.67 123 6,371 74
LOUISIANA 3.23 95 602 35   10.83 115 2,020 49   31.57 165 5,887 69
MAINE 4.07 120 2,107 123   10.92 116 5,650 136   20.31 106 10,507 123
MARYLAND 4.53 134 3,189 186   9.37 100 6,597 158   16.26 85 11,443 134
MASSACHUSETTS 5.67 168 3,956 230   11.61 124 8,097 195   16.63 87 11,597 135
MICHIGAN 5.03 149 2,081 121   11.68 124 4,828 116   19.78 103 8,179 95
MINNESOTA 4.69 138 2,831 165   10.48 112 6,336 152   16.77 87 10,134 118
MISSISSIPPI 2.37 70 411 24   9.44 101 1,637 39   36.42 190 6,315 74
MISSOURI 3.09 91 1,328 77   7.73 82 3,315 80   18.02 94 7,730 90
MONTANA 2.82 83 1,026 60   8.89 95 3,235 78   19.95 104 7,262 85
NEBRASKA 3.54 104 2,192 128   8.83 94 5,472 131   16.93 88 10,495 122
NEVADA 2.72 80 1,823 106   5.93 63 3,979 96   12.84 67 8,616 101
NEW HAMPSHIRE 3.13 93 2,678 156   6.61 70 5,646 136   11.32 59 9,668 113
NEW JERSEY 3.05 90 2,646 154   6.34 68 5,495 132   11.90 62 10,316 120
NEW MEXICO 3.82 113 803 47   13.49 144 2,833 68   31.21 163 6,553 76
NEW YORK 10.28 304 4,649 271   20.82 222 9,418 226   29.06 151 13,146 153
NORTH CAROLINA 3.62 107 1,551 90   9.88 105 4,225 102   20.87 109 8,932 104
NORTH DAKOTA 2.99 88 1,475 86   8.31 89 4,099 98   17.96 94 8,856 103
OHIO 4.07 120 1,713 100   10.23 109 4,302 103   19.26 100 8,098 94
OKLAHOMA 2.92 86 831 48   9.20 98 2,615 63   23.12 121 6,569 77
OREGON 3.80 112 1,808 105   9.73 104 4,625 111   18.69 97 8,883 104
PENNSYLVANIA 5.18 153 2,708 158   11.36 121 5,938 143   18.94 99 9,899 115
RHODE ISLAND 5.82 172 3,379 197   13.11 140 7,608 183   21.02 110 12,200 142
SOUTH CAROLINA 2.30 68 654 38   7.37 79 2,093 50   22.10 115 6,273 73
SOUTH DAKOTA 2.10 62 751 44   6.12 65 2,191 53   16.82 88 6,017 70
TENNESSEE 3.40 100 1,185 69   9.76 104 3,403 82   22.30 116 7,777 91
TEXAS 2.68 79 780 45   7.21 77 2,100 50   21.80 114 6,348 74
UTAH 2.72 80 1,165 68   9.10 97 3,901 94   18.63 97 7,984 93
VERMONT 4.98 147 3,473 202   12.61 134 8,798 211   21.57 112 15,046 176
VIRGINIA 2.41 71 1,743 102   5.05 54 3,656 88   12.76 67 9,246 108
WASHINGTON 5.08 150 2,716 158   10.94 117 5,844 140   18.40 96 9,829 115
WEST VIRGINIA 3.36 99 873 51   12.22 130 3,177 76   27.93 146 7,265 85
WISCONSIN 4.17 123 2,461 143   10.33 110 6,095 146   17.34 90 10,232 119
WYOMING 2.84 84 1,385 81   7.68 82 3,746 90   17.16 89 8,365 98
50 State Median $3.39 100 1,717 100   $9.38 100 4,162 100   $19.19 100 8,571 100
        Source: Urban Institute calculations based on data from DHHS, HCFA, USDA, IRS, OMB and SSA.
          a. Based on 50-state median of 100.
          b. Includes state spending plus federal macthing dollar spending.
          c. Includes total matching spending plus fully federally funded program spending.
        Note: Child poverty and the number of children in poverty are 1994 estimates based on three-year averages.

Table 3

TABLE 4: Total AFDC and Food Stamps Spending on Children's Programs, 1995
  Total AFDC Total AFDC and Food Stamps
  Per $1,000
Indexa Per
Indexa Per $1,000
Indexa Per
ALABAMA 1.12 42 357 28 5.60 97 1,789 73
ALASKA 7.41 280 5,726 453 10.40 180 8,035 326
ARIZONA 3.23 122 999 79 7.35 127 2,271 92
ARKANSAS 1.17 44 335 27 5.06 88 1,450 59
CALIFORNIA 8.49 321 2,913 231 11.35 197 3,894 158
COLORADO 1.69 64 1,274 101 3.69 64 2,789 113
CONNECTICUT 3.82 144 2,745 217 5.17 90 3,718 151
DELAWARE 2.11 80 1,710 135 4.17 72 3,383 137
FLORIDA 2.65 100 1,015 80 5.80 101 2,225 90
GEORGIA 2.94 111 1,277 101 6.59 114 2,869 116
HAWAII 6.07 229 4,156 329 10.44 181 7,142 290
IDAHO 1.50 57 502 40 3.83 66 1,284 52
ILLINOIS 3.25 123 1,507 119 6.08 105 2,822 114
INDIANA 1.64 62 849 67 4.18 72 2,158 88
IOWA 2.56 97 1,323 105 4.52 78 2,335 95
KANSAS 1.90 72 894 71 4.04 70 1,899 77
KENTUCKY 2.64 100 711 56 7.36 128 1,981 80
LOUISIANA 1.89 71 352 28 8.27 143 1,542 63
MAINE 3.77 143 1,951 154 6.87 119 3,552 144
MARYLAND 3.02 114 2,129 168 5.31 92 3,739 152
MASSACHUSETTS 4.04 153 2,820 223 5.59 97 3,896 158
MICHIGAN 4.72 178 1,951 154 7.47 129 3,088 125
MINNESOTA 3.45 130 2,086 165 5.20 90 3,140 127
MISSISSIPPI 1.71 65 296 23 8.93 155 1,548 63
MISSOURI 2.34 88 1,004 79 5.77 100 2,474 100
MONTANA 3.44 130 1,253 99 6.38 111 2,325 94
NEBRASKA 2.14 81 1,329 105 3.96 69 2,455 100
NEVADA 1.70 64 1,141 90 3.61 63 2,421 98
NEW HAMPSHIRE 2.10 79 1,791 142 3.37 58 2,876 117
NEW JERSEY 2.58 98 2,238 177 4.28 74 3,709 150
NEW MEXICO 5.66 214 1,188 94 11.15 193 2,341 95
NEW YORK 7.03 266 3,178 252 9.80 170 4,432 180
NORTH CAROLINA 2.41 91 1,033 82 5.09 88 2,178 88
NORTH DAKOTA 2.26 85 1,116 88 4.37 76 2,157 87
OHIO 3.47 131 1,458 115 6.63 115 2,788 113
OKLAHOMA 2.82 107 800 63 7.12 123 2,023 82
OREGON 3.31 125 1,573 124 6.21 108 2,951 120
PENNSYLVANIA 3.37 127 1,759 139 5.96 103 3,114 126
RHODE ISLAND 5.76 218 3,343 265 8.73 151 5,068 206
SOUTH CAROLINA 1.59 60 452 36 5.22 90 1,483 60
SOUTH DAKOTA 1.63 62 582 46 4.02 70 1,437 58
TENNESSEE 1.82 69 636 50 5.77 100 2,013 82
TEXAS 1.66 63 483 38 6.52 113 1,897 77
UTAH 2.06 78 882 70 4.21 73 1,807 73
VERMONT 5.06 191 3,530 279 7.91 137 5,519 224
VIRGINIA 1.50 57 1,088 86 3.83 66 2,771 112
WASHINGTON 4.83 183 2,581 204 7.42 129 3,964 161
WEST VIRGINIA 3.38 128 878 69 9.44 164 2,455 100
WISCONSIN 3.79 143 2,238 177 5.48 95 3,230 131
WYOMING 2.23 84 1,088 86 4.68 81 2,281 93
50 State Median $2.65 100 1,264 100 $5.77 100 2,465 100
        Source: Urban Institute calculations based on data from DHHS and Food Stamp Quality Control data and tabulations by Food and Consumer Service, USDA.
          a. Based on 50-state median of 100.
        Note: Child poverty and the number of children in poverty are 1994 estimates based on three-year averages.

TABLE 5: Medicaid Spending on Children
  Per $1,000
Index* Per
ALABAMA 2.74 69 875 49
ALASKA 6.17 155 4,767 267
ARIZONA 5.40 135 1,667 94
ARKANSAS 4.31 108 1,234 69
CALIFORNIA 3.68 92 1,262 71
COLORADO 2.45 61 1,850 104
CONNECTICUT 2.42 61 1,739 98
DELAWARE 3.44 86 2,788 156
FLORIDA 4.99 125 1,913 107
GEORGIA 4.11 103 1,789 100
HAWAII 5.48 137 3,751 210
IDAHO 2.10 53 704 39
ILLINOIS 3.92 98 1,820 102
INDIANA 3.28 82 1,696 95
IOWA 3.99 100 2,062 116
KANSAS 2.45 61 1,152 65
KENTUCKY 4.02 101 1,081 61
LOUISIANA 6.26 157 1,168 66
MAINE 4.43 111 2,289 128
MARYLAND 4.11 103 2,890 162
MASSACHUSETTS 4.93 124 3,441 193
MICHIGAN 4.31 108 1,780 100
MINNESOTA 4.59 115 2,772 156
MISSISSIPPI 5.47 137 948 53
MISSOURI 2.96 74 1,268 71
MONTANA 2.83 71 1,031 58
NEBRASKA 3.52 88 2,180 122
NEVADA 2.66 67 1,785 100
NEW HAMPSHIRE 2.53 63 2,164 121
NEW JERSEY 2.11 53 1,833 103
NEW MEXICO 6.03 151 1,266 71
NEW YORK 6.79 170 3,070 172
NORTH CAROLINA 4.53 114 1,939 109
NORTH DAKOTA 2.77 69 1,365 77
OHIO 3.87 97 1,626 91
OKLAHOMA 4.04 101 1,149 64
OREGON 3.84 96 1,825 102
PENNSYLVANIA 4.03 101 2,107 118
RHODE ISLAND 4.23 106 2,455 138
SOUTH CAROLINA 3.99 100 1,134 64
SOUTH DAKOTA 2.84 71 1,017 57
TENNESSEE 5.77 145 2,013 113
TEXAS 4.00 100 1,164 65
UTAH 4.12 103 1,767 99
VERMONT 3.96 99 2,762 155
VIRGINIA 2.29 57 1,655 93
WASHINGTON 3.64 91 1,945 109
WEST VIRGINIA 6.44 161 1,675 94
WISCONSIN 3.98 100 2,348 132
WYOMING 2.96 74 1,445 81
50 State Median $3.99 100 1,783 100
        Source: Urban Institute calculations based on data from HCFA 64 and 1082.
          * Based on 50-state median of 100.
        Note: Child poverty and the number of children in poverty are 1994 estimates based on three-year averages.

TABLE 6: Program Spending Trends Per Poor Child
in the Ten Highest and Lowest Ability States — FFY 1996
Program 10 High Indexa 10 Low Indexa Difference
AFDC 2,505 198 649 51 285.8
Food Stamps 1,538 122 1,250 99 23.1
AFDC & Food Stamps 4,043 164 1,899 77 112.9
Medicaid 2,589 145 1,248 70 107.4
EITC 1,694 114 1,488 100 13.9
All Other 3,677 140 1,952 75 88.3
Total $12,003 140 $6,588 77 82.2%
        Source: Urban Institute calculations based on data from DHHS, HCFA, USDA, IRS and OMB.
          a. Based on 50-state median of 100.
        Note: Child poverty and the number of children in poverty are 1994 estimates based on three-year averages.

Source: Urban Institute calculations based on data from U.S. Department of Health and Human Services (DHHS), Health Care Financing Administration (HCFA), U.S. Department of Agriculture (USDA), Internal Revenue Service (IRS), Office of Management and Budget (OMB), and Social Security Administration (SSA).

Source: Urban Institute calculations based on data from DHHS, HCFA, USDA, IRS and OMB, and SSA.
Note: Total tax revenue is FY 1994.

Source: Urban Institute calculations based on data from DHHS, HCFA, USDA, IRS and OMB, and SSA.
Note: Child poverty and the number of children in poverty are 1994 estimates based on three-year average.

Source: Urban Institute calculations based on data from DHHS, HCFA, USDA, IRS and OMB, and SSA.

About the Authors

Toby Douglas is a research associate at the Urban Institute. His primary interests include state fiscal issues, welfare reform, and homelessness. For the Assessing the New Federalism project he has conducted case studies on income support and social services in California. He also has been analyzing spending on children's programs in the ANF case study states.

Kimura Flores is a research associate at the Urban Institute. Prior to coming to the Institute, she worked on the California state budget for the California Assembly Ways and Means Committee and, more recently, advised the Chief Financial Officer of the District of Columbia.

Assessing the
New Federalism

ssessing the New Federalism is a multi-year Urban Institute project designed to analyze the devolution of responsibility for social programs from the federal government to the states, focusing primarily on health care, income security, job training, and social services. Researchers monitor program changes and fiscal developments. In collaboration with Child Trends, Inc., the project studies changes in family well-being. The project aims to provide timely, nonpartisan information to inform public debate and to help state and local decisionmakers carry out their new responsibilities more effectively.

Key components of the project include a household survey, studies of policies in 13 states, and a database with information on all states and the District of Columbia, available at the Urban Institute's Web site. This paper is one in a series of occasional papers analyzing information from these and other sources.

Topics/Tags: | Children and Youth | Economy/Taxes | Poverty, Assets and Safety Net

    Usage and reprints: Most publications may be downloaded free of charge from the web site and may be used and copies made for research, academic, policy or other non-commercial purposes. Proper attribution is required. Posting UI research papers on other websites is permitted subject to prior approval from the Urban Institute—contact publicaffairs@urban.org.

    If you are unable to access or print the PDF document please contact us or call the Publications Office at (202) 261-5687.

    Disclaimer: 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. Copyright of the written materials contained within the Urban Institute website is owned or controlled by the Urban Institute.

    Email this Page