Elsevier

Early Childhood Research Quarterly

Volume 55, 2nd Quarter 2021, Pages 52-63
Early Childhood Research Quarterly

The Effects of Universal Preschool on Child and Adult Outcomes: A Review of Recent Evidence from Europe with Implications for the United States

https://doi.org/10.1016/j.ecresq.2020.10.009Get rights and content

Highlights

  • Targeted high-quality preschools improve development of disadvantaged children.

  • Little reliable evidence on benefits of universal high-quality preschools in the US.

  • More reliable evidence using quasi-experimental variation is available from Europe.

  • High-quality universal preschool benefits disadvantaged children but not more advantaged children.

  • The European findings have implications for US preschool policy.

Abstract

Public preschool programs can be universal, open to all age-eligible children, or targeted, with eligibility limited to children from lower-income families. The effects on children of targeted programs have been intensively studied in the US, with results showing substantial beneficial impacts on child development and subsequent adult outcomes for disadvantaged children. However, there is little reliable evidence on the medium and long run effects of universal preschool programs in the US. This paper reviews studies from Europe that have exploited quasi-experimental variation to estimate the causal impact of universal public preschool eligibility and enrollment on child and adult outcomes. The evidence shows that these programs provide substantial short and long run benefits to disadvantaged children, but relatively modest benefits to more advantaged children. The implications of the European evidence for the issue of universal versus targeted programs in the US are discussed.

Introduction

Young children from disadvantaged families lag their more advantaged peers in developing cognitive and behavioral skills (Duncan & Magnuson, 2013). These socioeconomic gaps begin early in childhood and are attributable to differences in the home and external environments in which children are raised. The environment experienced early in life can have profound consequences for a child’s wellbeing and success in childhood and throughout life. There is considerable public and scholarly interest in early interventions in the lives of disadvantaged children as a means of improving their developmental outcomes and prospects for success in life.

There are two main theoretical arguments in support of this approach, as opposed to later remedial interventions: (1) There are sensitive and critical periods in childhood during which investments in child development are relatively productive. These periods are concentrated in early childhood (Knudsen, Heckman, Cameron, & Shonkoff, 2006). (2) There are dynamic complementarities whereby investments in early childhood development make later investments more productive (Elango, García, Heckman, & Hojman, 2016). The benefits to children, families, and society from high-quality early interventions in the lives of disadvantaged children have been shown to be very large relative to the costs (Elango et al., 2016). Hence public investment in early childhood education (ECE) is widely viewed as an effective and socially efficient way to improve the development prospects of disadvantaged children and thereby help to reduce social inequity.

Preschool programs supported by public funding come in two varieties: targeted and universal. Eligibility for targeted programs is restricted mainly to children from low-income families, while universal programs are open to all age-eligible children. Head Start is the leading example of a targeted program in the United States, with eligibility for the most part limited to children in families with income less than the poverty level. There is a substantial amount of experimental and quasi-experimental evidence that high-quality targeted preschool programs in the US have had large beneficial effects on the health, education, employment, earnings and other long run outcomes of enrollees. This evidence is discussed at length in several recent reviews (Baker, 2011, Elango et al., 2016, Council of Economic Advisors, 2016, Duncan and Magnuson, 2013, Hotz and Wiswall, 2019, Phillips et al., 2017).

These reviews also point out that evidence on the effects of high-quality universal preschool programs on child outcomes in the US is limited. A handful of states initiated and/or greatly expanded preschool programs in the 1990s, and such programs now exist in 44 states and the District of Columbia (Friedman-Krauss et al., 2019). There are three difficulties in evaluating the effects of state-level preschool programs on child outcomes. First, there is substantial variation across states in the quality and characteristics of the programs. About half are universal, available to any child in the state who meets the age requirement. Others are targeted to children from disadvantaged families. Some are rated as high-quality and others are not. This variability makes it difficult to define an appropriate set of treatment and comparison states for use in a quasi-experimental evaluation of the effects of universal programs in the states that have instituted such programs. Second, there is considerable variation in program quality within states, making it difficult to define exactly what the “treatment” is (Phillips et al., 2017). Third, most of these programs were initiated in the relatively recent past, limiting the amount of information that can be learned about their long run effects.

Another source of evidence on the effects of universal preschool programs is from studies conducted in several European countries, including Denmark, Germany, Italy, Norway, and Spain. These studies evaluate the impacts of universal national programs funded by central governments, which typically aim to impose and enforce uniform high-quality standards throughout a given country. There are differences across countries in the standards, and there is no direct evidence on the uniformity of quality within each country, but most of these programs would be considered high quality by US standards, as discussed below.

There were major reforms in several countries over the past four decades that significantly expanded funding for these preschool programs. Most of the funding is provided by national governments, but the programs are managed by local or regional governments. In countries with major reforms, there was often substantial variation across local jurisdictions in the speed at which they were able to expand the supply of preschools in response to the reforms. This provides the opportunity for quasi-experimental difference-in-differences (DD) research designs. The presumed relatively uniform quality of the national programs ensures a well-defined “treatment” and the slow-expanding municipalities provide a natural comparison group. In other countries, there was no specific reform that could be used as a quasi-experiment, but there was random or quasi-random rationing of slots at the local level, providing another useful source of variation. Several of these programs have been in place long enough to provide evidence on medium and long-run impacts.

This paper reviews evidence from Europe on the effects of universal preschool programs on child, adolescent, and adult outcomes. This evidence is informative for the US because there is strong popular support for such a program in the US (see Child Care Aware of America, 2018), but proponents often cite evidence from evaluations of targeted programs when arguing in favor of a universal program. Evidence from evaluations of targeted programs is of limited value in projecting the impact of a universal program. The early targeted demonstration programs in the US (Perry Preschool, Abecedarian, and others) were of exceptionally high quality and served extremely disadvantaged children without good alternative sources of care. It is doubtful that their effects can be extrapolated to more typical high-quality programs serving children from all families in a very different social and economic environment today (Baker, 2011). Evidence from evaluations of Head Start is also of limited value for inferring the impact of a universal program, since Head Start does not serve more advantaged children.

This paper builds on the reviews cited above, complementing them by incorporating evidence from several recently published articles that evaluate effects of universal public programs in the countries mentioned above. This review differs from previous reviews in focusing explicitly on results from recent studies of medium and longer run effects in European countries.

Section snippets

Methods

The programs discussed in the articles surveyed here are universal center-based preschool programs. Each article asserts that preschool quality in the country of interest is relatively high and uniform across centers. However, none of the articles presents direct evidence on quality as measured by an observation-based instrument. In some cases they support their assertions by describing the curriculum and standards governing child-staff ratio, group size, and teacher training. Several of the

Average effects

Table 3 presents estimates of treatment effects from each article. The third column shows that six of the seven articles report statistically significant (at the 5% level) beneficial treatment effects on at least one outcome. Six report beneficial and statistically insignificant effects for at least one outcome as well. One reports a statistically significant adverse effect on one outcome. The terms beneficial and adverse are used instead of positive and negative, because for some of the

Comparison to Evaluations of Universal Preschool in the US

There has been a considerable amount of research evaluating the effects of state-funded universal preschool in the US, but much of it is not as informative as the evidence from Europe due to limited opportunities for research designs that are well-suited to estimating causal effects (Phillips et al., 2017). The most common research design used to study the impact of state-funded universal preschool in the US is to compare outcomes of children who were eligible for a program in a given year,

Discussion

To illustrate the policy implications of these findings for the US, it is useful to consider two specific US proposals for public preschool that have recently captured attention, one for a universal program and the other for a targeted program. The proposals provide concrete examples, including cost estimates, and are used here purely for illustrative purposes. This discussion is not intended to provide support for either program or for one over the other. Senator Elizabeth Warren’s Universal

References (32)

  • Council of Economic Advisors

    Inequality in early childhood and effective public policy interventions. Chapter 4

    Economic Report of the President

    (2016)
  • N. Drange et al.

    Early childcare and cognitive development, evidence from an assignment lottery

    Journal of Labor Economics

    (2019)
  • G.J. Duncan et al.

    Investing in Preschool Programs

    Journal of Economic Perspectives

    (2013)
  • S. Elango et al.

    Early childhood education

  • C. Felfe et al.

    Can’t buy mommy’s love? Universal childcare and children’s long-term cognitive development

    Journal of Population Economics

    (2015)
  • M.D. Fitzpatrick

    Starting school at four, the effect of universal Pre-Kindergarten on children’s academic achievement

    The B.E. Journal of Economic Analysis & Policy, Advances

    (2008)
  • Cited by (12)

    View all citing articles on Scopus

    Helpful comments were provided by attendees at the Research Forum of the Crane Center for Early Childhood Research and Policy, Ohio State. I am also grateful to Michael Baker, Nabanita Datta Gupta, Bill Gormley, Andrea Ichino and coauthors, and Marianne Simonsen for useful comments. I am solely responsible for the content, including opinions.

    View full text