Child Care Arrangements for Children Under Five

Jeffrey Capizzano, Gina Adams, Freya L. Sonenstein

Number B-7 in Series, "New Federalism: National Survey of America's Families"

The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.

---

Note: To see data on child care arrangements for children under five from the 1999 National Survey of America's Families, click here.

Over the past decade, child care has emerged both as an issue of public concern and a key component of U.S. social policy. The large number of mothers with preschool children in the workforce has made America's families more reliant on nonparental care and raised public awareness of early care and education as an issue of public policy. Many children now spend at least some time in child care during their critical developmental years. As a result, child care centers, family child care homes, relatives, and nannies have become essential to working families with children. Policymakers also have recognized the importance of child care because of the role it plays in helping parents work and because of the impact it can have on the development of children. In 1996, for example, policymakers considered child care a key factor in helping welfare recipients attain self-sufficiency, and accordingly, the federal welfare reform legislation—the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA)—highlighted child care as a work support mechanism.

Examining Child Care at the State Level

Significant aspects of child care policy have historically fallen under the purview of state governments. For example, states establish many of the child care subsidy policies—such as reimbursement rates for child care providers and copayment rates for recipients of child care assistance—and also regulate child care quality. The changes to federal child care policy outlined by PRWORA further expanded the role of states in child care policymaking by ending the child care entitlement for welfare recipients, consolidating the four major federal child care assistance programs into a single block grant, and increasing child care funding to states.

Yet relatively little is known about child care patterns in individual states or how they vary across states. Most of what is known about the use of child care is gathered from nationally representative surveys that are not designed to capture state-level child care patterns. It is likely that large variations exist across states due to differences in the costs and supply of child care as well as variations in labor force patterns and child care policies. A better understanding of state-specific child care behavior will help state policymakers effectively target their child care policies and identify the likely impact of policy changes.

This brief, therefore, provides information on the primary child care arrangements used by children under five with employed mothers nationally and across a number of states. Also, because child care experiences tend to vary for children of different ages and incomes, we examine infants and toddlers separately from three- and four-year-olds, and children from families with incomes above 200 percent of the federal poverty level (FPL) separately from those with incomes at or below 200 percent of the FPL.

The National Survey of America's Families

Data from the 1997 National Survey of America's Families (NSAF)¹ are used to examine primary child care arrangements. The NSAF oversampled households with income below 200 percent of the FPL and collected child care information on a nationally representative sample of children as well as on representative samples of children in 12 states.² For randomly selected children in the sample households, interviews were conducted with the person most knowledgeable about each child. From these interviews, data were collected about the types of care used and the number of hours that the child spent in each form of care.³ Since the mother was most often the most knowledgeable adult, the term "mother" is used here to refer to this respondent.⁴ This analysis focuses only on children under five whose mothers were interviewed during the nonsummer months.⁵ We also restrict our analysis to preschool children whose mothers are employed.

The primary child care arrangement is defined as the arrangement in which the child spends the most number of hours while the mother is at work. We group the arrangements in the following primary arrangement categories: center-based child care (child care centers, Head Start, preschool, prekindergarten, and before- or after-school programs); family child care (care by a nonrelative in the provider's home); baby-sitter or nanny (care by a nonrelative in the child's home); relative care (care by a relative either in the child's or provider's home); and parent care (for those children whose mother did not report a nonparental child care arrangement while she worked).⁶

Where Are Preschool Children Cared For?

The National Picture. Nationwide a large percentage (76 percent) of preschool children with employed mothers are regularly cared for by someone other than their parents. For more than half of preschool children with employed mothers, the primary child care provider is not related to the child (figure 1). Thirty-two percent of children are in center-based child care arrangements, while about half as many (16 percent) are in family child care. A relatively small percentage of children (6 percent) are regularly cared for by a baby-sitter or nanny in the child's home.⁷

In contrast, less than half of preschool children with employed mothers are cared for primarily by relatives or by parents. Twenty-three percent of preschool children have a relative as the primary child care provider—9 percent in the child's home and 14 percent in the home of the relative—while 24 percent of children are in parent care.⁸

State Patterns. An examination of individual states, however, reveals that national estimates of child care mask sizable state variation in the use of specific primary child care arrangements (figure 1). The focus below is on the states with the greatest differences in the use of each form of care.⁹ Specifically, findings show that:

• Almost 40 percent of preschool children are in center-based care arrangements in Alabama, Minnesota, and Mississippi (39, 38, and 38 percent, respectively). These percentages are twice as large as the percentage of children in center-based care in California (19 percent).
• The percentage of preschool-age children in family child care ranges from 20 percent in Wisconsin to 10 percent in Massachusetts.
• Mississippi, with 32 percent of preschool children of employed mothers in relative care, has more than twice the proportion of children in this form of care as Minnesota (13 percent).
• Over 30 percent of children are in parent care in California, Washington, and Massachusetts (34, 33, and 32 percent, respectively), while less than one in five children are in this form of care in Mississippi and Alabama (15 and 17 percent, respectively).

While these findings highlight the differences in the use of each type of primary child care arrangement across states, states also differ from each other in the distribution of children across all arrangements. This diversity is found even among states that are similar on one dimension of child care (table 1). For example, while two states—Alabama and Minnesota—both have large percentages of children in center-based care, the distribution of children across the other arrangements in each state is quite different. Indeed, Alabama has over twice the percentage of children in relative care as Minnesota (27 percent, compared with 13 percent). Examples like this demonstrate the enormous diversity of child care utilization across states and point out the unique child care challenges that state-level policymakers and administrators face in making decisions about their states.

Child Care for Children of Different Ages

Parents often make different choices about child care arrangements for their preschool children based on the age of the child. Preschool children of different ages have varying developmental needs, and certain forms of child care are often more readily available for children in particular age groups. Therefore, to provide a better understanding of the child care choices made by employed parents, the types of arrangements parents chose for their infants and toddlers (children under three) are examined separately from choices made for three- and four-year-olds.

The Nation Overall. Nationally, infants and toddlers are more likely to be with relatives and in parent care, while three- and four-year-olds are more often found in center-based arrangements (table 1). Among infants and toddlers, 27 percent are in relative care and another 27 percent are in parent care, while smaller proportions are found in center-based care (22 percent) and family child care (17 percent). Among the older preschoolers, relative and parent care are used less often as primary arrangements (17 percent and 18 percent of the children, respectively), while more three- and four-year-olds are found in center-based care than in any other arrangement (45 percent). Similar to infants and toddlers, only 14 percent of three- and four-year-olds have family child care as their primary arrangement.

Variation across the States. These national patterns generally hold true across the individual states. The largest proportions of infants and toddlers are in parent or relative care across most states, while the highest percentages of three- and four-year-old children are in center-based arrangements in every state (figures 2 and 3). The proportion of children in each primary care arrangement, however, varies widely from state to state for both age groups.

Infants and Toddlers. For children under three, center-based care arrangements are usually not the most common form of care, although there are some states in which a significant proportion of very young children have centers as their primary arrangement. Relative care and parent care, in contrast, are major forms of child care for infants and toddlers across most states, though the proportion of children in these care arrangements varies. Specifically, the data show that:

• In Minnesota and Michigan, more than one in four infants and toddlers are in center-based arrangements (29 and 28 percent, respectively). California, on the other hand, has the smallest percentage of young children in center-based care (9 percent), and states like New Jersey and New York also have relatively small percentages of young children in centers (13 and 14 percent, respectively).
• Across states, the percentage of infants and toddlers in family child care ranges from as high as 20 percent in Wisconsin to as low as 9 percent in Massachusetts.
• While most states have close to one in three very young children in relative care, Mississippi has the largest percentage of children in this form of care (38 percent). Florida and Minnesota, on the other hand, have less than half this percentage in relative care (18 and 14 percent, respectively).
• California has the highest percentage of children in parent care (42 percent), while Mississippi, Michigan, and Texas have the smallest percentages of very young children in this form of care (19, 21, and 21 percent, respectively).

Three- and Four-Year-Olds. For children ages three and four, center-based arrangements are the most popular form of primary child care arrangement in every state. While relative care and parent care are generally less prevalent among these children, there are states that have high percentages of children in these forms of care. The data show that among three- and four-year-olds:

• Mississippi and Alabama have the highest percentages of children in center-based arrangements (60 and 58 percent, respectively), while in some states like California and Wisconsin, only around one in three three- and four-year olds are in center-based care (31 and 33 percent, respectively).
• Wisconsin has the highest percentage of three- and four-year-olds in family child care (20 percent), while much less than half the proportion of this age group is in this form of care in Texas (7 percent).
• Twenty-four percent of these children are in relative care in Alabama. Minnesota, on the other hand, has only 11 percent of three- and four-year olds in relative care.
• In Washington and Massachusetts, around 30 percent of three- and four-year-olds have parent care as their primary arrangement (32 and 27 percent, respectively). In contrast, Alabama and Mississippi (8 and 10 percent, respectively) have very low percentages of children in this age group in parent care.

Child Care for Children from Families of Different Incomes

The child care arrangements of low-income children are an important policy priority for lawmakers. With welfare reform's work requirements expected to increase the demand for child care, and with federal and state governments investing to provide child care subsidies to low-income families, the types of child care arrangements used by these families are of increasing interest to policymakers. Therefore, it is important to look separately at the arrangements of the low-income population¹⁰ and how these arrangements compare with those of higher-income families.

Nationally, we find that low-income children are less likely to be in centers than higher-income children (26 percent, compared with 35 percent) and are more likely to be in relative care (28 percent, compared with 20 percent) and parent care (28 percent, compared with 21 percent) (table 1). Low- and higher-income children, however, are almost equally likely to be in family child care (14 percent, compared with 17 percent).

Differences across States. There is great variety in the primary child care arrangements used by low-income children across the states, as well as in the arrangements used by higher-income children. In fact, no consistent pattern exists across states for either income group—different forms of care are prevalent in different states (see table 1). Low-Income Families. A closer look at low-income families across states, for example, shows that:

• In Massachusetts, Mississippi, and Florida, roughly a third or more of low-income children are in centers (38, 34, and 32 percent, respectively). Conversely, California, Wis-consin, Michigan, New York, Minnesota, and Washington—those states with the smallest proportions of low-income children in center-based care—have about one in five children in this type of arrangement (17 to 21 percent) (figure 4).
• New York has the highest percentage of low-income children in family child care—17 percent. In contrast, Washington has only 7 percent of low-income children in family child care.
• In Mississippi, Michigan, and Alabama, about two in five low-income children use relatives as their primary care arrangement (43, 39, and 38 percent, respectively)—roughly twice the percentage of low-income children in relative care as Florida (21 percent).
• Washington has by far the highest percentage of low-income children using parent care, with almost three times as many children in this form of care (41 percent) as in Mississippi (14 percent).

Higher-Income Families. Among higher-income children, findings show that:

• The proportion in center-based care ranges from more than two in five in Alabama and Minnesota (50 and 44 percent, respectively) to less than one in five in California (20 percent) (table 1).
• Wisconsin has the largest percentage of higher-income children in family child care, with 22 percent. Conversely, New York and Massachusetts have only around 10 percent of higher-income children in this form of care (9 and 10 percent, respectively).
• Though children in higher-income families generally have relatives as their primary care arrangements less frequently than low-income children, some states—including New Jersey, Massachusetts, California, Michigan, Texas, and Wisconsin—have as many as one in four higher-income children in this form of care (all between 24 and 26 percent). Minnesota has by far the lowest percentage of higher-income children in relative care (6 percent).
• The use of parent care ranges from over 30 percent of higher-income children in California and Massachusetts (34 percent each) to as low as 12 percent in Texas.

Differences within Individual States. The differences that exist between low- and higher-income populations in the use of child care arrangements are also of interest to policymakers. Different patterns of child care use between the two populations within a particular state may reflect the ability of higher-income families to pay more for child care, the inability of low-income families to access certain kinds of care, or different child care preferences between the two income groups.

Comparing low- and higher-income children in each state reveals large differences in the use of child care arrangements. Consistently, the major differences occur in the use of center-based care; higher-income children are much more likely to be in centers than low-income children. Low-income children are instead found more frequently in relative and parent care. Specifically, data show that:

• The largest gaps in the use of centers occur between low- and higher-income children in Alabama and Minnesota, where the percentage of higher-income children in centers is 26 and 24 percentage points greater than for the low-income population (figure 5). Michigan, Texas, and New York also have sizable differences. Massachusetts is the only state in which low-income children are in center-based care in larger proportions than higher-income children (a difference of 14 percentage points).
• It appears that in most states children of low-income and higher-income families are placed in family child care in similar proportions (table 1). Mississippi has the largest difference between the low- and higher-income children in the use of family child care (9 percentage points greater among higher-income children).
• Generally within the states, low-income families rely on relatives more than do higher-income families. In three states—Minnesota, Mississippi, and Alabama—low-income families have much higher percentages of children in relative care than higher-income families (26, 22, and 18 percentage points greater, respectively).
• The use of parent care among low-income children is greater than among higher-income children in most states. For example, the percentage of low-income children in parent care in Texas is considerably larger than that of higher-income children (17 percentage points greater). Massachusetts, in contrast, is the only state in which the percentage of higher-income children in parent care is considerably greater than that of low-income children (12 percentage points greater).

Conclusions

This brief documents differences in the utilization of child care arrangements across states for preschoolers of varying ages and incomes. Three important findings emerge. First, while parents work, a large majority of preschool children, regardless of age or income, are regularly cared for by individuals other than their parents. This fact is true nationally and in every state examined here, and it emphasizes the importance of child care in the lives of America's families. The high utilization of child care reinforces the need for policymakers to pay close attention to the experiences of children while they are in child care.

Second, this brief shows how the availability of state-specific child care data can illuminate vast differences in child care experiences across states. The NSAF estimates reveal, for example, that Alabama, Florida, Minnesota, Texas, and Mississippi have the largest proportions of children under five in center-based care, while the largest proportion of preschool children in California, Massachusetts, and Washington are cared for by parents.

Finally, common conceptions of where children are placed in care sometimes do not hold. While infants and toddlers are more likely to be cared for in less formal child care arrangements (such as relatives and parents), and three- and four-year-olds are more likely to be in center-based care, there are clear exceptions. In Michigan and Minnesota, for example, almost 30 percent of infants and toddlers are in center-based care, and in Washington, close to 33 percent of three- and four-year-olds have parent care as their primary arrangement. Also contrary to expectations, low-income children are not always more likely to be in less formal arrangements and higher-income children are not always more likely to be in center-based care. Indeed, more low-income children are in center-based care than any other kind of arrangement in Massachusetts, while parent care is the predominant form of care for higher-income children in both California and Massachusetts.

While this brief documents differences that exist in the types of care used across states, it is only an initial step to a better understanding of state-level patterns of child care. The next step is to explore why these state differences exist and whether they are due to such factors as differences in parental work patterns, the supply of child care, family structure, cultural preferences, and child care policies. As it is apparent that there is no single factor that can explain state variation, future research using multivariate analysis is necessary to illuminate how these forces are associated with the state differences.

The findings that do emerge from this brief, however, highlight both the reliance of America's families on nonparental care and the differences in the types of nonparental care used across states. They reinforce the importance of continuing to explore state differences in child care through state-specific data, such as the National Survey of America's Families, and to emphasize the challenges facing policymakers across the country as they work to develop policies to support the child care choices of families within their states.

---

Notes

The authors thank James Barsimantov for his excellent research assistance as well as Alan Weil, Stefanie Schmidt, Linda Giannarelli, Joan Lombardi, Sandy Hofferth, and Lynne Casper for helpful comments on earlier versions of the brief.

1. The National Survey of America's Families is a national survey of over 44,000 households and is representative of the noninstitutionalized, civilian population of persons under age 65 in the nation as a whole and in 13 focal states. The survey focuses primarily on health care, income support, job training, and social services, including child care.

2. The states are Alabama, California, Florida, Massachusetts, Michigan, Minnesota, Mississippi, New Jersey, New York, Texas, Washington, and Wisconsin. Colorado is also a focal state in the Assessing the New Federalism (ANF) project but is not included in these analyses. Due to the late addition of Colorado to the ANF project, responses to the child care questions from a large number of Colorado respondents were received during the summer months and did not provide information on nonsummer child care arrangements, which are the focus of this analysis. Because of the small size of the nonsummer sample from Colorado, it is excluded from the analysis.

3. For more on NSAF survey methods, including the "most knowledgeable adult," see Dean Brick et al. (1999).

4. The mother of the child was the "most knowledgeable adult" for 83 percent of the children in the sample.

5. Because child care arrangements and hours spent in care can vary widely from the school year to the summer, the observations with data on child care relating to the summer months (June 12 to September 26) were not included in this analysis. The observations that are included are weighted to provide representative data on child care during the school year.

6. The survey did not include questions about parental care, which could include care provided by the other parent, the mother caring for the child while she worked, or care for the child at home by a self-employed mother. If the respondent did not report an arrangement, the child is assumed to be in one of these "parental care" categories. We are confident that this measure captures parental arrangements because the share of children of employed parents with parents as the primary arrangement in the NSAF (24 percent) is the same as the share of preschool children (24 percent) in the 1994 Survey of Income and Program Participation (SIPP) who were cared for primarily by their mother at work or their father while their mother was working (Casper 1997).

7. Because of the small percentage of children using nannies or baby-sitters as their primary care provider, this arrangement is not discussed in the following sections.

8. It is important to note that the data on primary care arrangements presented in this brief underestimate the use of any particular form of child care because the estimates do not reflect the extent to which each arrangement may be used as a secondary child care arrangement. For example, children with a center-based primary care arrangement may also use relatives, family child care, or nannies for shorter periods of time regularly each week. A later NSAF brief will look more closely at the use of multiple child care arrangements.

9. The states that have the highest and lowest percentages of children in a given arrangement are presented here. These states are statistically different from each other at the .05 level. Differences among other states not presented may or may not be statistically significant. In addition, one should be cautious in interpreting the actual point estimates because of the sizes of the state samples. Confidence levels around national point estimates averaged +/- 3 percentage points, and the confidence intervals around subpopulation point estimates within states were larger (+/- 7 percentage points for our state estimates of age and income subpopulations).

10. A low-income family is a family with an income equal to or below 200 percent of the poverty level (i.e., $25,258 for a family of two adults and one child in 1997).

---

References

Casper, Lynne M. 1997. "Who's Minding Our Preschoolers?" Current Population Reports P70-62 (fall 1994 update). U.S. Department of Commerce, Economics and Statistics Administration, Bureau of the Census. http://www.census.gov/prod/3/ 97pubs/p70-62.pdf.

Dean Brick, Pat, Genevieve Kenney, Robin McCullough-Harlin, Shruti Rajan, Fritz Scheuren, Kevin Wang, J. Michael Brick, and Pat Cunningham. 1999. 1997 NSAF Survey Methods and Data Reliability. Washington D.C.: The Urban Institute. National Survey of America's Families Methodology Report No. 1.

About the Authors

Jeffery Capizzano is a research associate in the Urban Institute's Population Studies Center. He has written and published in the areas of transportation and welfare policy and is currently working on an investigation of child care policy in 13 states.

Gina Adams is a senior research associate in the Urban Institute's Population Studies Center, where she is responsible for directing research on child care and early education. Her research efforts focus on policies and programs that affect the affordability, and supply of child care and early education, as well as on the child care arrangements of families.

Freya Sonenstein is currently the director of the Population Studies Center at the Urban Institute. She has over 20 years of experience conducting applied and behavioral research related to family and children's policy.

---

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.

Usage, posting and reprint of materials on the UI web site:

Most publications may be downloaded free of charge from the web site in PDF format. This information may be used and copies made for research, academic, policy or other non-commercial purposes. Proper attribution is required.

Copyright of the written materials contained within the Urban Institute website is owned or controlled by the Urban Institute. Posting UI research papers on other websites is permitted subject to prior approval from the Urban Institute—contact paffairs@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.