(Withdrawn) The Squeezed Middle: Effects of Universal Pre-Kindergarten on the Use of Center-Based Care and Child Care Expenditures

Background: This paper examines the heterogeneous effects of universal pre-kindergarten (UPK) on center-based care enrollment and child care expenditures by household income, with a special focus on middle-income children. The cost of center-based care is not ignorable in the United States, even for families with moderate income. Middle-income families’ income levels are not low enough to be eligible for targeted child care programs such as child care subsidies and Head Start, but not high enough to afford high-quality early childhood education programs in the private market. As a result, families with young children in the middle of the income distribution may experience a “child care squeeze.” In fact, Reardon (2011) documents that since the 1980s, the gap between the 90th and 50th percentile has increased, while the gap between the 50th and 10th percentile has been decreasing, which indicates that middle-class children are losing ground. UPK’s universality provides an equal opportunity of early education and care for all age-eligible children regardless of their household income. Upon this background, this paper documents trends in the use of center-based care by household income and examines whether UPK availability affects middle-income children’s center-based care use and child care expenditures.

Method: I use restricted data from the Early Childhood Program Participation (ECPP) Survey of the National Household Education Surveys Program (NHES) from 1995 to 2019 and employ a two-way fixed effects model, taking advantage of cross-state variation in the timing of the implementation of UPK. To examine heterogeneity in the effects of UPK by household income, I add interaction terms between the policy variables and income quintile groups. Importantly, I address a recent methodological challenge of a two-way fixed effects model where time-varying treatment effects can lead to biased estimates raised by Goodman-Bacon (2021). As this issue has been rarely discussed in the early education and care policy literature, this paper contributes to the methodological discourse in the literature.

Results: I find that UPK is on average associated with a 16 percent increase in the use of center-based care. The heterogeneous analysis by household income reveals that the effects of UPK on center-based care use are larger for low- and middle-income families than for the most affluent families. On other hand, UPK has overall little effects on child care expenditures. I also find no significant heterogeneity by household income in the effects of UPK on child care expenditures.

Conclusions and Implications: Results highlight that UPK offers middle-income children the opportunity to attend center-based care, which is known to generate higher developmental benefits than informal care, with no significant change in their child care expenditures. Given the current administration’s proposal to expand UPK as well as the ongoing discussion about the cost-effectiveness of targeted versus universal child care programs, this paper sheds light on important policy implications.