ONE
Introduction
Ludovica Gambaro, Kitty Stewart and Jane Waldfogel
In recent decades, the provision of early childhood education and care (ECEC) has risen up the policy agenda right across the globe. One central driving force has been the increasing labour force participation of women, which has created a growing demand for childcare services for preschool age children. Governments have encouraged and subsidised these services for a number of reasons. Concerns about child poverty have figured prominently in countries such as the UK, while the need to tackle social exclusion among immigrant groups has been a factor in many continental European countries. Demographic change has also been important. Rising rates of lone parenthood have created fiscal pressure in countries where the state has traditionally stepped in in the absence of a male breadwinner. In some countries, for example Germany, the policy goal of facilitating workâfamily balance has been coupled with that of promoting fertility rates.
At the same time, there has been growing interest in the value of early education from a child development perspective. Research has increasingly underlined the importance of what happens in a childâs early years for their later life chances (for a review, see Almond and Currie, 2011). Evidence from a wide range of countries indicates that children who have had exposure to preschool education do better at school, and that the benefits are long lasting (Heckman et al, 2010; Ruhm and Waldfogel, 2012).
Two points in particular emerge clearly from the research into the impact of ECEC. The first is the importance of quality: children stand to gain much more where the quality of provision is higher (Shonkoff and Phillips, 2000; Blau, 2001; Baker and Milligan, 2008; Sylva et al, 2011). What is understood by quality can vary between (and within) countries but broadly, children appear to do best in settings in which adults interact with children in a responsive, sensitive and stimulating way. If care is low quality the expected benefits do not materialise, and some provision may even be damaging to childrenâs prospects. Thus, while at its best childcare is far more than just somewhere to park children while parents are working, not all settings will promote childrenâs cognitive, social and emotional development: what happens in a setting is crucial.
The second point is that ECEC appears to make the most difference to children from disadvantaged backgrounds (Ruhm and Waldfogel, 2012). There are a number of possible reasons for this. Children from higher-income backgrounds are more likely to have access to books and educational toys at home, and more likely to be taken on trips to museums and parks which widen their horizons and stimulate their thinking; the added value of attending an early education setting will be greater for children who do not already enjoy these benefits. Many children from immigrant families will speak a different language with their parents, so early education gives them exposure to the national language before they start school. Other children may hear only a limited range of vocabulary at home, may live in cramped conditions with less space to run or play physical games, or may have parents whose attention is distracted by younger siblings or financial pressures.
Thus while childcare provision makes it possible for parents â or more particularly mothers â to go out and work in the paid labour market, children themselves can gain from high-quality early education and care. In principle, this looks like a win-win situation: high-quality care can help tackle income poverty in the short term (and gender inequality in the medium term) while improving childrenâs life chances by preparing them for future learning.
However, delivering on this potential âdouble dividendâ poses clear challenges. Most significantly, high-quality provision is expensive. In the context of scarce resources, the policy goal of making ECEC more universally available is often prioritised over improving quality (Kamerman and Kahn, 2001; West, 2006). But focusing predominantly on ensuring availability and affordability is not adequate if we are interested in fostering child development.
This book brings together eight country studies and examines the issues governments face when they try to expand early childhood provision and make sure that it is an equitable and high-quality service. Our starting point was the UK, where investment in services for young children over the last 15 years has gone a long way towards improving both the accessibility and the quality of early years provision, but continuing challenges remain. Our idea for this comparative volume grew from a belief that the experience of a diverse set of countries would offer new perspectives on how these challenges might be addressed.
Of course, systems of provision and the policies underpinning them vary substantially across countries. Some countries, for example France, have a strong nursery education tradition and offer a school-based service to all children age three and above, while providing parents with generous and universal childcare subsidies for younger children (Martin, 2010). In the Nordic countries, on the other hand, services for children under school age have developed since the 1960s, albeit at varying speeds, and are rooted in the pedagogical tradition, with hardly any distinction between children of different age groups (Leira, 2002). By contrast, English-speaking countries have been historically characterised by low levels of ECEC and by a more marginal role for the state, with perhaps the exception of Australia (OâConnor et al, 1999; Brennan, 2002).
However, despite different starting points, all governments are confronted with the pressure of providing early education and care and face dilemmas regarding the policy instruments to be used to achieve this goal. This book explores how services are organised and how policies are designed in different countries. The aim is not that of favouring âfast policy transferâ; rather, the idea underpinning the book can be described as âcontextualised policy learningâ (Mahon, 2006), whereby attention is given to how policies work on the ground and to the contexts in which they are embedded. We asked each of our authors to address the same question: How does your country ensure access to high-quality early childhood education and care for disadvantaged children? Their answers are rich in policy detail and empirical evidence, and offer new ideas and insights â although, as often as not, they highlight common policy challenges rather than identifying clear solutions.
The rest of this introductory chapter takes the following structure. We begin by discussing current evidence on the impact of early education and care for childrenâs outcomes. We then turn to explore the purpose of this book in more detail. We discuss our rationale for choosing the eight countries and consider some broad similarities and differences between them, drawing on international data. Finally, we provide a brief overview of each of the country chapters, highlighting the key policy issues that arise in each one.
Why does it matter? What we know about early education and care and childrenâs outcomes
A growing body of evidence points to the importance of ECEC for child development, and hence its potential impact on longer-term educational, employment and wider social outcomes. Initially, such evidence came from US evaluations of small-scale trials, including the Perry Preschool project, which provided high-quality early childhood education to a randomly selected group of disadvantaged children in Michigan. Studies which have followed the Perry children into their forties have found long-term gains attached to enrolment in the programme, including improvements in educational attainment, employment and earnings, as well as social benefits such as reduced criminal activity (see, for example, Karoly et al, 2005; Heckman et al, 2010). Waldfogel (2006) discusses evaluations of other similar experimental programmes, all of which point to substantial gains in cognitive achievement.
These evaluations have been influential, despite small sample sizes, because the randomised project design allows us to be confident that identified effects are causal and do not simply reflect hidden differences between families, such as differing parental attitudes to education. However, while they offer solid guidance for the impacts of small, high-quality interventions on very disadvantaged children, they are less helpful in relation to universal or large-scale programmes (Baker, 2011). In this respect, evidence from European countries is illuminating. This strand of research has made use of regional variations in service provision, birthday cut-offs or rigorous econometric techniques to get close to identifying causal effects in the absence of randomised design (see Ruhm and Waldfogel, 2012 for a review). For example, studies in France, Norway and Denmark have exploited variation in local provision to examine the effects of the expansion of universal preschool programmes during the 1960s and 1970s, and found positive benefits for attainment and later labour market participation (Havnes and Mogstad, 2011; Bingley and Westergaard-Nielsen, 2012; Dumas and Lefranc, 2012). More recently, universal prekindergarten programmes have been implemented in a number of US states, and the related research points to positive short-term effects on childrenâs literacy and maths scores and on socio-emotional development (Gormley et al, 2005, 2008; Magnuson et al, 2007a, 2007b; Wong et al, 2008).
In England, the best available evidence comes from an observational study, the Effective Provision of Preschool Education (EPPE) project, which observed children in a range of different preschool settings in 1997 and tracked their progress on into compulsory schooling. Children who had attended preschool had higher levels of cognitive and social-behavioural outcomes on entry to primary school than children who had not (Sylva et al, 2004). Follow up studies found that positive effects were still apparent at the end of primary school (Sylva et al, 2008). Higher-quality preschool continued to predict maths, science and social-behavioural outcomes at age 14 (Sylva et al, 2012b).
Analysis of the Organisation for Economic Co-operation and Developmentâs (OECD) education survey, the Programme for International Student Assessment (PISA), also indicates that early education can have lasting effects. In nearly all OECD countries, 15-year-olds who had attended pre-primary education outperformed those who had not: even after controlling for socioeconomic background, a year of preschool was associated with a test score improvement of 33 points, close to the 39 points linked to a year of formal schooling (OECD, 2011a). Using these same data, Mostafa and Green (2012) estimate that if Sweden and the UK had had universal preschool programmes in place in the early 1990s, Sweden would been seven places higher up the OECD league table in PISA 2009, and the UK twelve places higher up.
Aside from the generally positive impact of early education on later outcomes, two further findings emerge clearly from the research in this field. The first is that the quality of provision matters; not all formal provision is alike. Studies that assess both the quality of provision and childrenâs outcomes are relatively rare, and the available evidence is largely observational, but findings are very consistent: children make more progress in settings with high âprocess qualityâ, meaning settings where interactions between adults and children are warm and responsive. This is in turn associated with particular structural features, most notably staff qualifications and child-to-staff ratios (see, for example, Ruopp et al, 1979 for a rare example of experimental research in this area, and literature reviews in Shonkoff and Phillips, 2000; Vandell and Wolfe, 2000; Blau, 2001).
The OECD PISA analysis finds the strongest association between preschool education and later test scores in countries that have invested to improve the quality of provision (OECD, 2011a), while in England the EPPE results also underline the importance of quality. The study rated the quality of provision using the Early Childhood Environment Rating Scale (ECERS), which includes observation of classroom practice and interactions between staff and children. Higher quality was found to be strongly associated with more highly qualified staff, and with the presence of trained teachers in particular. The effect of preschool experience on outcomes at entry to school was greater where the quality of early education had been higher. By age 11, attendance at a low-quality preschool setting carried almost no benefits in comparison to non-attendance; and by 14, only the highest quality settings appeared to have left a mark (Sylva et al, 2011, 2012b).
There is also evidence that low-quality provision can have a negative impact. In Canada, studies examining a childcare subsidy programme that led to big increases in the use of non-parental childcare found significant negative effects on socio-emotional outcomes, health, and the vocabulary of young children. Researchers attributed these effects in part to declines in parental health and relationship quality (in turn linked to more hostile, less consistent parenting), and in part to the fact that most of the childcare taken up was informal and of poor quality (Baker and Milligan, 2008; Lefebvre et al, 2011).
The second clear finding is that gains are largest for children from low-income or immigrant households, and for those with less educated parents. Indeed, in many studies the positive effects are confined to these groups (Ruhm and Waldfogel, 2012). In England, the EPPE study finds a stronger effect of high-quality preschool on children from disadvantaged backgrounds (Sylva et al, 2011, 2012a). In the US, studies of prekindergarten and kindergarten expansion find larger effects for families with low levels of education, low-income, immigrant or non-English speaking backgrounds, and families from disadvantaged neighbourhoods (Gormley et al, 2005, 2008; Magnuson et al, 2007a; Figlio and Roth, 2009; Fitzpatrick, 2010; Dhuey, 2011). The studies cited earlier of preschool expansion in Denmark, France and Norway all find stronger effects for disadvantaged children, in particular the children of less educated mothers in both Denmark and Norway (Havnes and Mogstad, 2011; Bingley and Westergaard-Nielsen, 2012; Dumas and Lefranc, 2012). Studies by SpieĂ et al (2003) for Germany and Fredriksson et al (2010) for Sweden find that preschool attendance closes gaps in attainment at age 12 or 13 between children of immigrants and children with native-born parents.
One question on which research is less clear, however, concerns the benefits of ECEC for younger children â those under three years old. Most of the research cited so far focuses on preschool programmes for children aged three, four and five. There is less research into the impact of formal provision for under threes. A wide range of literature looks at the impact of maternal employment during a childâs first year and is fairly consistent in identifying negative effects for health, cognitive and socio-behavioural development, especially where mothers work full time, although the effects vary by the quality of alternative care provided, by the quality of maternal care, and also by the extent to which employment leads to increases in income (see the discussion in Waldfogel, 2006).
Studies of maternal employment at ages one and two generally find either positive or neutral effects for childrenâs cognitive outcomes, although long hours of group care have been linked to negative social and behavioural outcomes, particularly for boys, and again the quality of provision seems important (see Langlois and Liben, 2003; Waldfogel, 2006). These findings suggest that some exposure to high-quality group ECEC provision, perhaps part time, might have a positive impact on child development for one- and two-year-olds, as it does for older children, but research that has focused specifically on this question, largely for two-year-olds, has been inconclusive. For France, Goux and Maurin (2010) (making use of regional variation in availability) find that enrolment in nursery school at age two rather than three shows no significant relation to later school achievement, while an observational study by Caille (2001) finds only slightly less likelihood of children being held back a class in their later schooling if they started attending at age two rather than three, although results are larger for children of immigrants. For England, Sylva et al (2012a) find very little medium-term advantage associated with starting preschool at age two rather than three. On the other hand, in Germany, Felfe and Lalive (2011) find that centre-based care for nought- to three-year-olds is associated with small developmental benefits for the average child and larger and lasting benefits for children from lower-income families.
So while there is strong evidence that preschool attendance is beneficial once children reach three, it is less clear th...