Despite lower income groups being more often cost burdened, this does not reflect higher costs of child care. In fact, lower income groups spend less on child care than their higher income counterparts—perhaps reflecting an inability to afford full time, center based or licensed care; costs that are offset by child care subsidies for which their higher income counterparts are not eligible; or a preference for informal care by family or friends.7 For instance, whereas poor families spend an average of $2,547 on child care annually, families at or above five times the poverty line spend an average of $10,423 per year (see Table 1).
Cost Burden Varies By Family Characteristics
Twenty percent of married couples and 40 percent of single parents with young children who have child care costs pay more than 10 percent of their income for child care (see Table 2). Shares vary by race, with 23.4 percent of families headed by non-Hispanic whites being cost burdened compared with 36.5 percent of black, non-Hispanic-headed families, and 32.4 percent of families headed by Hispanics. More than one-third of the least-educated families (37.5 percent) devote more than 10 percent of their income to child care, compared to 21.9 percent of those headed by a college graduate. Finally, families headed by a full-time, year-round worker are slightly less likely than families headed by someone working fewer hours to be cost burdened.8
Child Care Cost Burden Varies Across the Nation
Child care cost burden is a significant challenge in all regions of the country (see Table 3), but more prevalent in the Northeast and the South. In metropolitan areas, an estimated 27.4 percent of families with young children are cost burdened by child care expenses, compared to 23.1 percent of families in nonmetropolitan areas.9
Figure 2 maps the child care cost burden by state. The burden is highest in states across the East Coast, including New Jersey, Georgia, and Maryland, where more than 33 percent of families are cost burdened by child care expenses. The high average child care costs in Eastern states like New Hampshire, Connecticut, New Jersey, and Maryland (and the District of Columbia) may be a function of the higher cost of living in these states.10 States with lower child care costs like West Virginia and Utah also have relatively low percentages of families who are child care-cost burdened.
Cost Burden Varies By Family Characteristics
The tables presented here display average child care costs among families with young children who had any child care expenses in the past year. We find that, in general, the families who tend to spend the most on child care are also the least likely to be cost burdened.
Families who are better off financially have more options when purchasing child care—and may more often select more expensive placements. They may also purchase more hours of care each week. Because of their high incomes, however, such care is not a large proportion of family income overall. Additionally, families lower on the socioeconomic ladder might report lower median child care expenses because subsidies (for which higher-income families are not eligible) defray some of the costs, although these subsidies may not always be generous enough to alleviate a child care cost burden.
These analyses indicate that the burden of child care costs is common among American families. Nearly one in three families with young children has some child care costs, spending an average of more than $6,500 annually on child care.11 For 26.8 percent of these families—a group disproportionately comprised of poor and low income families and families headed by a single parent or someone with less than a high school education—the cost amounts to more than 10 percent of their incomes. That more than half of poor families face a child care cost burden is striking, and suggests a need for policies to address the affordability of child care for very low income families.
Because our analyses explore only costs formally reported in the survey data, they do not capture the complex reality of families’ child care decision making. For instance, our analyses include only those families with young children who had any child care expenses, and as such do not include families providing their own child care because the income they could earn from work does not justify the expense. Nor can we address the informal arrangements families may make to juggle work demands and child care needs. Finally, these data do not allow for exploration of child care availability, quality, and characteristics, all of which are important components of child care policy.
The data for this project are from the Current Population Survey (CPS) Annual Social and Economic (ASEC) supplement from 2012 to 2016, the most recent years for which data are available. We use five years of data to ensure sufficient sample sizes for state level analyses. All data about income and expenses refer to the previous calendar year (for example, the 2014 survey refers to income and expenses from 2013), and all income values are inflation-adjusted to reflect 2015 dollars. Note that a redesign of the ASEC in 2014 included changes to its income-related questions, with the revised items administered to five-eighths of the sample. As such, slight inconsistencies between survey years after and prior to the redesign may exist, although these differences are minimal in their impact.12
In the CPS, parents are asked whether they pay for child care and how much they pay. This amount is summed for all children to reflect the total amount the family spends on child care. Because parents are asked how much they actually pay for child care costs, it is reasonable to assume that parents are reporting child care expenses net of any child care subsidy received. However, since parents are not asked specifically about subsidies in the CPS, it is impossible to know for certain whether families include subsidy payments in their calculations. In our analysis, the “family” is equivalent to the Supplemental Poverty Measure (SPM) unit, which includes all people in a household related by blood or marriage, cohabiting partners and their children, and unrelated children (such as foster children) who are cared for by the family.13 All analyses are weighted using the Census Bureau’s Supplemental Poverty Unit weight.
Readers should be cautious when comparing estimates between groups because the CPS is asked of a sample of the population, rather than the total population. Although some estimates may appear different from one another, it is possible that any difference is due to sampling error. Further, in some cases very small differences may be statistically significant due to the large sample size of the CPS. Nonetheless, all differences discussed in this brief are statistically significant (p<0.05).
2. Most recent work on child care costs has compared state-level average costs to state-level median incomes, in order to identify places with child care costs that are disproportionately high compared with family earnings (see, for example, the Economic Policy Institute’s state by state calculator at http://www.epi.org/child-care-costs-in-the-united-states/). Other research has looked at the relationship between child care costs and family characteristics (see Linda Laughlin, “Who’s Minding the Kids? Child Care Arrangements: Spring 2011” U.S. Census Bureau, 2003, p70–135; Kristin Smith and Nicholas Adams, “Child Care Subsidies Critical for Low-Income Families Amid Rising Child Care Expenses,” Policy Brief No. 20 (Durham, NH: Carsey Institute, 2013.). In this brief, we focus specifically on the characteristics of the child care cost-burdened population, and approach this exploration from the perspective of families, rather than state averages.
3. See the Proposed Rule by the Health and Human Services Department establishing this benchmark of affordability at https://www.federalregister.gov/documents/2015/12/24/2015-31883/child-care-and-development-fund-ccdf-program; Elise Gould and Tanyell Cooke, “High Quality Child Care Is Out of Reach for Working Families,” Washington, DC: Economic Policy Institute, 2015 (see http://www.epi.org/publication/child-care-affordability/); “Policy Proposal on Child Care, Early Education, After School Care,” Make It Work Campaign, 2015 (www.makeitworkcampaign.org/wp-content/uploads/2015/05/Caregiving-Proposal-For-web-5.13.15-FINAL.pdf). Given more recent recommendations by the U.S. Department of Health and Human Services that affordable child care not exceed 7 percent of family income, these estimates are conservative determinations of the child care cost burden. This 10 percent benchmark of affordability was also identified in Hillary Clinton’s proposal on early childhood education. See Hillary Clinton’s official campaign website: http://www.hillaryclinton.com/issues/early-childhood-education/.
4. Note, however, that in general median values are lower suggesting high values skew the mean value.
5. In all analyses, we cap child care costs at the value of total gross income to account for extreme outliers in families spending far more on child care than is earned in a given year. These families are not typical, and this topcoding has no impact on median values presented throughout this brief nor on the share of families that are child care-cost burdened.
6. Poverty is a family-level construct that compares total family income to a threshold based on family size and number of children. Families with total incomes below their assigned threshold are considered poor. For these analyses, families with total incomes below twice their assigned threshold are considered low income. The official poverty line for a family of four with two children was $24,036 in 2015.
7. See Smith and Adams (2013) for evidence that those with subsidies indeed pay less overall for child care.
8. Many of the family characteristics discussed thus far are interrelated (for example, higher educational attainment is associated with higher family income), making it difficult to tease out the impacts of any individual family-level indicator. To address this, we conducted a logistic regression predicting cost burden greater than 10 percent, controlling for families’ region of residence, metropolitan status, number of children under age 6, age of youngest child, and percent of poverty threshold, as well as the head’s marital status, race, educational attainment, and full-time/full-year work status (results not shown but available upon request). This analysis allows us to isolate the effects of any given indicator on the odds that a family experiences child care cost burden. We find that, net of other factors, having more children under 6 in a family, having a family head who is single, having a family head who works full time year round, and having family income below the federal poverty threshold all independently increase a family’s odds of being child care-cost burdened. Although similar to findings in Table 2, it is important to note that net of income, families headed by a full-time, year-round worker are more likely to be cost burdened than their counterparts who work less often. It is possible that although these families have higher earnings (reducing their risk of being cost burdened), they may require more annual hours of child care than their peers, thus driving up their overall risk of being cost burdened.
9. Note that this breakdown excludes the 0.75 percent of the sample for whom metropolitan status is not identified. These families are included in all other calculations in this brief except here.
10. Cost of Living Data Series, Second Quarter 2016; see https://www.missourieconomy.org/indicators/cost_of_living/.
11. This estimate is somewhat lower than those recently reported from the Economic Policy Institute (EPI). This discrepancy likely stems from methodological differences in that the EPI reports focus on the cost of full-time care only (see Elise Gould et al., “The Economic Policy Institute’s 2015 Family Budget Calculator: Technical Documentation,” Working Paper #299, 2015).
12. See Jessica L. Semega and Edward Welniak, Jr., “The Effects of the Changes to the Current Population Survey Annual Social and Economic Supplement on Estimates of Income,” proceedings of the 2015 Allied Social Science Association Research Conference, www.census.gov/content/dam/Census/library/working-papers/2015/DEMO/ASSA-Income-CPSASEC-Red.pdf.
13. For more on child care expenses and SPM family units, see https://cps.ipums.org/cps/resources/spm/p60-241.pdf.