Table of Contents | Previous | Next |
Chapter Three: Meeting the Demand for Child Care Subsidies
This chapter discusses state trends in the use of child care subsidies from 1997 to 2002.1 It offers two ways to facilitate cross-state comparisons: by using a standard metric of children living in families with incomes below 62% of state median income (SMI), and by describing the extent to which states provides subsidies to children in families with incomes below their own eligibility ceilings. After showing use trends from 2000-2002 for the study counties, the chapter describes how the 17 states in the study allocated the increased subsidy funding described in the previous chapter to meet the needs of TANF families, families transitioning from TANF, and non-TANF families. It also shows the distribution of types of arrangements that were subsidized and how these changed.
Summary of Findings
-
From 1997 to 2002, the number of children served by child care subsidies nearly doubled in the average state. Eleven of these 14 states experienced a growth rate of over 50 percent. Five—Indiana, Louisiana, New Mexico, New York, and Texas—more than doubled their child care subsidy caseload in this time period. For most states, most of the growth occurred between 1997 and 1999. In fact, in four states, the number of children receiving child care subsidies decreased between 2000 and 2002.
-
Although most states showed tremendous growth in the numbers of children receiving subsidies between 1997 and 2002, the proportion of low-income children served varied widely across the states, ranging from 12 percent in Virginia to 37 percent in Washington in 2002.2
-
In 2002, seven of the 14 states reporting data served 25 percent or more of the families with employed parents and incomes below state eligibility guidelines. In 1997, no states in this study met this threshold. States differed in the extent to which they provided subsidies to families who were potentially eligible for subsidies according to their state’s own income rules. In Illinois, more than half of children who were eligible for subsidies under state income rules received them in 2002. By contrast, in Minnesota, New Mexico, Texas, and Virginia, less than 20 percent of potentially-eligible children received subsidies.3
-
On the county level, between 2000 and 2002, changes in the number of children receiving subsidies varied widely, from an increase of 50 percent in rural counties in New Mexico, to a reduction of 27 percent in a rural county in Massachusetts. The level and the direction of change at the county level often were not in keeping with change at the state level over the same time period, partly because the lower numbers served at the county level make percentage change figures more volatile.
-
Within the subsidy caseload, children in non-TANF families and families transitioning from TANF constituted the great majority of subsidy users. In 12 of the 14 states that reported subsidy use data by TANF status, children in non-TANF families and families transitioning from TANF accounted for two-thirds or more of the child care subsidy caseload.
-
Although TANF children were a small and decreasing proportion of the child care subsidy caseload, in most states the proportion of children receiving subsidies whose families were on the TANF cash assistance caseload grew in the period from 1997 to 2002. In 2002, in 14 of the 15 states that reported data in sufficient detail, more than 10 percent of children from TANF families received child care subsidies. In 1997, only four of 13 states reported that more than 10 percent of children from TANF families were receiving subsidies.
-
In the majority of states in the study, more than half of all the subsidized arrangements were in centers. A much smaller proportion of subsidies supported care by relatives; in 13 of the states 20 percent or less of subsidized arrangements were care with relatives, either in the relative’s home or the child’s own home. These two trends not withstanding, there was tremendous variation among the states.
Child Care Subsidy Use
In this section, we describe the changes in the number of children who received child care subsidies in all the states that participated in the study except California, which was unable to provide state-level use information in sufficient detail. After describing state-level subsidy use from 1997 to 2002, we then provide county-level information for the period between 2000 and 2002.
State-Level Patterns of Subsidy Use
From April 1997 to April 2002, the number of children who received subsidies grew substantially in nearly all of the states in the study (see Exhibit 3-1). The average growth rate in subsidy use across the 14 states that could report data for all years was 89 percent. Eleven of the 14 states that reported subsidy use numbers for all four years experienced a growth rate of more than 50 percent. (State-by-state total child care enrollments and rates of growth are shown in Appendix Table 3-A.)
[D] |
In the majority of the states, the numbers of children who received subsidies increased in each succeeding year during this period, but the gains were not steady. For the period between 1997 and 1999, the gains were dramatic; nine of the 14 states that could report subsidy use for both years showed increases of over 40 percent, with Indiana, Louisiana, New Mexico, and Texas showing gains of over 100 percent. Tennessee had the lowest growth rate—9 percent (see Exhibit 3-2). The rate of growth slowed considerably between 2000 and 2002 (see Exhibit 3-3). In that period, only Ohio showed a growth rate of over 30 percent and, in four of the states, subsidy caseloads actually decreased.
[D] |
[D] |
After describing the general growth in the number of children served by all sources of subsidies, a next step is to describe the growth using a common metric so that it is possible to compare states of different sizes. (See Exhibits 3-4 and 3-5.) To make appropriate comparisons between states with relatively large populations, such as New York, and states with small populations, such as New Mexico, we used an estimate of the number of children under age 13 living in families with incomes below 62% of state median income (SMI), with all parents in the household employed, as well as all children under age 19 with disabilities.4 We chose the figure of 62% of SMI because that is where the average state set its eligibility ceiling.5 This is not an estimate of eligible children served, since, under current law, states are allowed to set their eligibility ceiling as high as 85% of SMI. In some states in our study, such as Minnesota and North Carolina, children are eligible for subsidies when families have incomes that are higher than 62%. In other states, such as Alabama, Massachusetts, and New Jersey, children in families with incomes at 62% of SMI are ineligible. It is also important to note that, in reality, the number of children with incomes below 62% of SMI is different each year because changing numbers of families enter or leave the labor force and family incomes change from year to year. However, data limitations for this report make it necessary to use an estimate for a single time period to compare use numbers from several years.
Exhibit 3-4 shows that the average percentage of low-income children served in the study states, calculated as described above, increased from 15 percent in April 1997 to 24 percent in April 2002. The percentage peaked in 2001, at 25 percent. Throughout the five-year period, the proportion of children served in the study states varied widely; for instance, in April 2002, the proportion served varied 12 to 37 percent. Exhibit 3-5 shows individual state figures for April 1997 and April 2002. (See also Appendix Table 3-A.)
| 1997 | 1998 | 1999 | 2000 | 2001 | 2002 | |
|---|---|---|---|---|---|---|
| Range | 5%-20% | 13%-26% | 11%-31% | 10%-33% | 11%-36% | 12%-37% |
| Average | 15% | 18% | 22% | 22% | 25% | 24% |
| % Change From Previous Year | 20% | 22% | 0% | 14% | -4% |
[D] |
States that initially served smaller proportions of these low-income children tend to show the greatest percentage gains. This is the case in Texas. The state ranked among the lowest of the study states in terms of the proportion of children served by subsidies in 2002, when calculated using the denominator of children in families below 62% of SMI, yet experienced a 210 percent increase in numbers served between 1997 and 2002.
However, states that served relatively high percentages in 1997 also showed dramatic percentage gains. For instance Illinois, New York, and Michigan all showed relatively high rates of increase in numbers of children served between 1997 and 2002, ranging from 92 to 131 percent, even though they were serving relatively high percentages of low-income children in 1997.
A second way to think about state patterns in the numbers of children served by subsidies is to compare the proportion of children served who may be eligible under each state’s rules. These comparisons make it possible to see the extent to which states are meeting the needs of families that they have targeted for subsidies. Exhibit 3-6 shows the average percentages of children served under state income rules across the study states for the years of the study, together with the range, and Exhibit 3-7 shows individual state patterns. In 2002, seven states in the study provided child care subsidies to an estimated 30 percent or more of the children who were potentially eligible for subsidies under state rules. This is quite a change from 1997, when only Tennessee served more than 30 percent of potentially state-eligible children.
| 1997 | 1998 | 1999 | 2000 | 2001 | 2002 | |
|---|---|---|---|---|---|---|
| Range | 7%-27% | 11%-34% | 16%-48% | 14%-53% | 15%-55% | 16%-54% |
| Average | 16% | 22% | 24% | 25% | 25% | 26% |
| % Change From Previous Year | 38% | 9% | 4% | 0% | 4% |
| Source: Information provided by the study states, drawn from their ACF-800 and ACF-801 reports to the U.S. Department of Health and Human Services and additional sources. |
[D] |
A closer look at individual state patterns shows that, while in most states larger increases occurred in the first half of the study period, there was substantial variation among states. Exhibit 3-8 shows, for each state, the percentage of potentially state-eligible children served each year between 1997 and 2002 for which we have data. It also shows the number of children served as a percentage of the number below 62% of SMI. (For additional information, see Appendix 3-A.)
[D] |
County-Level Information
All but two of the states in the study provided information about child care subsidy use in the 25 study counties for the years 2000-2002.6 Exhibit 3-9 shows the percentage change in numbers of children receiving subsidies at the county level for these years. The county-level use data shows great differences in percentage changes among the counties for this period; the range, in fact, is much wider than at the state level. (See Exhibit 3-2 for comparison and Appendix Table 3-B for further detail.)
[D] |
In addition to showing a wider variation than at the state level, the change that occurred at the county level was not necessarily consistent with overall change in the home state. For instance, for the period of 2000-2002, Tennessee experienced an 11 percent increase in subsidy use. However, county-level data for that state shows a very mixed picture: Shelby experienced a 12 percent decrease in the numbers of children receiving subsidies, one rural county group experienced no growth, and the other rural county group showed a 15 percent increase. Similarly, Franklin, Massachusetts experienced the greatest percentage decrease in subsidy use (27 percent) among the counties for which data were available for 2000-2002, yet the state of Massachusetts reported an increase of 36 percent over that time period. Exhibit 3-10 provides a comparison of the change in subsidy use at the county level over the two-year period, compared with change at the state level.
| County | 2000-2002 Percentage Change, County Level |
2000-2002 Percentage Change, State Level |
|---|---|---|
| MOBILE, AL | 11% | 24% |
| COOK, IL | 2% | 3% |
| MADISON, IN | -20% | 2% |
| FRANKLIN, MA | -27% | -1% |
| WAYNE, MI | 5% | 10% |
| HENNEPIN, MN | 25% | 5% |
| ITASCA, KOOCHICHING, PENNINGTON, MN | -8% | 5% |
| UNION, NJ | -2% | -11% |
| DONA ANA, NM | 33% | 22% |
| GRANT, HIDALGO, LUNA, NM | 50% | 22% |
| ALAMANCE, NC | 10% | -3% |
| JOHNSTON, NC | -14% | -3% |
| MECKLENBURG, NC | 2% | -3% |
| HAMILTON, OH | 17% | 32% |
| BEDFORD, COFFEE, MARSHALL, TN | 15% | 11% |
| FAYETTE, HARDEMAN, HAYWOOD, LAKE, LAUDERDALE, TN | 0% | 11% |
| SHELBY, TN | -12% | 11% |
| HARRIS, TX | 26% | 16% |
| ARLINGTON, VA | -20% | NA |
| KING, WA | 3% | 13% |
| Note: State and county numbers were unavailable for California. County numbers were unavailable for Louisiana and New York for at least one of the two years. Source: Information provided by the study states, drawn from their ACF-800 and ACF-801 reports to the U.S. Department of Health and Human Services and additional sources. |
The Composition of the Child Care Subsidy Caseload
As we noted in Chapter 1, children eligible for subsidies live in families with several different circumstances: with parents in approved employment or job preparation activities and receiving TANF cash assistance; with employed parents transitioning from TANF; or in low-income (non-TANF) families with parents eligible for subsidies because they are working or in approved education or training activities. Many states balance the needs of these three groups, but in general make TANF families and those transitioning from TANF among their highest priorities.
What proportion of subsidies supports children from non-TANF families? The states in the study were asked to indicate the numbers of children in their child care subsidy caseload that fell into the categories of TANF, transitional (if a formal program existed), or Non-TANF families. Exhibit 3-11 shows the proportion of children receiving subsidies in each of these family categories in April 2002. (We include information on “transitioning” families for those states that have such a category, but it is important to remember that, in the states without such a distinction, they are counted among the “non-TANF” population.) With the exception of Tennessee, in all the states with subsidy use data reported by TANF status, children from non-TANF and transitioning families constituted the majority of children receiving subsidies. Even in Tennessee, non-TANF children accounted for nearly 44 percent of the child care subsidy caseload. (See Appendix Table 3-C for more details.)
[D] |
Exhibit 3-12 provides subsidy caseload information by TANF status for 1997 and 2002, for the 12 states for which both years of data were available. In all these states except Tennessee and Minnesota, children from TANF families accounted for a smaller proportion of the child care subsidy caseload in April 2002 than in April 1997. In nine of the states the proportion of children who were from TANF families dropped substantially—by more than 100 percent. (For more-detailed information, see Appendix Table 3-C.)
[D] |
Meeting the Demand for Subsidies for Families on TANF
Although TANF children represented a decreasing share of the numbers of children in subsidy caseloads, paradoxically, over time a larger proportion of children from TANF families were receiving child care subsidies. With the passage of Federal welfare reform, many state and local policymakers expected that the TANF program’s emphasis on work would significantly increase the demand for child care subsidies by those receiving cash assistance. And, indeed, between 1997 and 2002, in all states in the study, the proportion of children in TANF families who received subsidies increased, in some cases dramatically.
How can it be that children from TANF families represented a decreased share of the child care subsidy caseload, and yet a larger proportion of TANF children received subsidies in 2002 than in 1997? Much of the explanation lies in the dramatic decrease in the overall TANF caseload. Even though a higher proportion of children on TANF were receiving subsidies in 2002 than were in 1997, the absolute number decreased substantially. This section briefly describes TANF caseload trends and their effect on the use of child care subsidies.
TANF Caseload Trends
In many of the study states, TANF caseloads decreased substantially between 1997 and 2002 (Exhibit 3-13). In 13 of the 17 states, TANF caseloads decreased more than 40 percent in this time. (For more information, see Appendix Table 3-D.)
[D] |
Trends in Numbers of Children on TANF Cash Assistance Who Received Subsidies
Data reported by 11 of the states make it possible to estimate the percentage of children living in TANF families who received child care subsidies between 1997 and 2002. Exhibit 3-14 shows that many states served a significantly higher percentage of the TANF child caseload with child care subsidies in 2002 than in 1997.7 In some states, the increase was very substantial. For instance, in Illinois, Louisiana, and Minnesota, the percentage doubled or nearly doubled; in Texas, the percentage of TANF children receiving subsidies increased more than three-fold.
[D] |
State Patterns of Child Care Receiving Subsidies
Overall State Patterns
A final way to describe the subsidy patterns among the states is to do so in terms of the types of care subsidies purchase. Exhibit 3-15 depicts the proportions of the different types of care supported by each state, and across the 15 states reporting data in April 2003. These states were asked to report on all of the arrangements that were paid for with all sources of funding used for child care. We asked that states be as precise as possible and to indicate whether their reports were generated by their administrative systems, informed estimates, or a combination of the two approaches.
In nine of the 16 states that reported use data in 20028, more than half of the subsidized arrangements were in centers. A much smaller proportion of subsidies supported care by relatives; 13 of the states had 20 percent or less of their subsidized arrangements with relatives, either in the relative’s home or the child’s own home. In eight of the 15 states, less than 1 percent of subsidized care was provided by a non-relative in the child’s own home. These two trends not withstanding, as the exhibit shows, there was tremendous variation among the states. For instance, in Michigan, 16 percent of subsidized arrangements were in centers and 66 percent of children were with a relative (in the relative’s or child’s home) or with a non-relative in the child’s home. By contrast, in North Carolina 85 percent of arrangements were in centers and 3 percent in relative care and .02 percent of subsidized care occurred with a non-relative in the child’s home. For detailed state information on the distribution and growth rates of all types of care, see Appendix Table 3-E.
[D] |
The states also reported their subsidized arrangements according to their regulatory status. Exhibit 3-16 shows that in 12 of the 16 states, more than 60 percent of the arrangements supported occurred in regulated settings; in half of the states, 80 percent or more of subsidized care was regulated. In Ohio, where 100 percent of subsidized child care is considered “regulated,” the state requires license-exempt arrangements to submit to its regulatory standards, although in these cases, parent inspections may substitute for inspections from the regulatory agency.
Changes in the Proportions of Care Between 1997 and 2003
How did patterns shift from 1997 to 2002? Exhibit 3-17 shows each state’s distribution of arrangements by type in 1997 and 2002 for the 13 states that could supply data in both years. These show that there were shifts in distribution of subsidized arrangements between the two points in time, but these shifts did not always move in one direction. For instance, between 1997 and 2002, four of the states showed a 10 percent or greater increase in the proportion of care subsidized in centers, while, in five states, the change in the proportion of subsidized center slots was between plus or minus 10 percent and, in four states, the change was 10 percent or more. In the case of child care that goes largely unregulated—care by relatives and care in a child’s home by a non-relative—there was a consistent decrease. With the passage of the 1996 legislation, policymakers and others were concerned that welfare reform would result in increasing numbers of subsidized children in these forms of child care. However, the proportion of subsidies that went to the combined category of all relative arrangements and care by a non-relative in the child’s home stayed the same or decreased in nine of the 13 states that could report data for the two time periods. For more information, see Appendix Table 3-E.
[D] |
[D] |
***
This chapter described the study states’ use of the subsidy funding in terms of numbers served, families, TANF status, and the types of arrangements purchased. The following chapter describes the state policies that guided these decisions.
1 Please note that data on state expenditures reported in the prior chapter is for the period ending in 2001 while data on subsidy use is for the period ending in 2002. The year difference is because the final data collection occurred in 2002, at which point the most recent data were reported by States. While states could report April 2002 subsidy use, they were able to report complete financial information for the prior year only. (back)
2 To facilitate comparisons, we computed the numbers of children served as a percentage of those who lived in households with family incomes below 62% of SMI, as described in the previous chapter. We chose the 62% cut-off because this was where the average state set its income ceiling in 2002. For more information, see the text in this chapter following Exhibits 3.3, as well as the relevant sections in Chapter 2. (back)
3 The income level at which the state sets eligibility for subsidies does, of course, affect the proportion of eligible children served. (back)
4 The number of children in families below 62% SMI was estimated by the Urban Institute using a simulation model and data on income, employment, and disability status from the combined March 1995, March 1996, and March 1997 Current Population Surveys, which cover calendar years 1996-1998. (back)
5 See Child Care and Development Fund Report of State Plans FY 2002-2003, which summarizes the CCDF State plans for the 50 states, Washington, D.C., the U.S. Virgin Islands, and Puerto Rico. December 2002. Published by the Child Care Bureau, U.S. Department of Health and Human Services. (back)
6 California and New York provided insufficient county-level information. (back)
7 Our calculation is different from those used by most states. States generally calculate the use of subsidies according to the number of parents with earnings or who are in approved education and training programs. Because these programs differ greatly, children in similar families are included in one state’s calculation but excluded in another. To facilitate cross-state comparisons, our estimate takes into account all children living in families receiving TANF assistance by taking the reported number of TANF families and multiplying it by the average number of children in family. (We have excluded child-only TANF cases.) (back)
8 California did not supply data on child care use for 1997-2003. (back)
| Table of Contents | Previous | Next |