Access to early childhood education in Australia

Research Report No. 24 – April 2013

You are in an archived section of the AIFS website. Archived publications may be of interest for historical reasons. Because of their age, they may not reflect current research data or AIFS' current research methodologies.

Appendix A: Access to and use of early childhood education and care services: An international overview

Peter Moss

[The following is an overview article on access to and use of early childhood education and care services, prepared for the project by Peter Moss.]

This paper provides a brief overview of some issues in early childhood education and care (ECEC) services in two groups of countries:

  • 30 member states of the European Economic Area, which consists of the 27 member states of the European Union (EU), plus three of the four members of the European Free Trade Association (EFTA; the fourth member, excluded in this overview, is Liechtenstein); and
  • 3 predominantly English-speaking, non-European countries: Canada, New Zealand and the United States of America.

Of these 33 countries, 27 are member states of the OECD. The six that are not members are: Bulgaria, Cyprus, Latvia, Lithuania, Malta and Romania. The overview does not include the seven other OECD member states: Australia, Chile, Israel, Japan, Korea, Mexico and Turkey.

The paper addresses three issues:

  • systems of ECEC in these 33 countries;
  • access to ECEC services in these countries, including universal access and entitlement; and
  • how access may be measured.

The paper centres on Table A1, which provides information about the 33 countries, together with this text, which provides a commentary on the table as well as some additional material. The table covers four main areas:

  • the type of ECEC system in each country;
  • compulsory school age;
  • entitlement to ECEC; and
  • attendance rates at formal ECEC services, given separately for children under and over 3 years.

In addition, the paper has a short annex about the relationship between entitlement to ECEC and parental leave. While the paper's main focus is the relationship between the upper end of ECEC and compulsory (formal) schooling, its brief being to "be primarily concerned with early childhood education in the year prior to commencing formal compulsory school", a complete view of the issue of entitlement should consider how ECEC relates to policy areas at each end of its age range, which includes parental leave policy at the lower end.

Table A1: Details of ECEC in 30 European countries, North America and New Zealand (2011)
Country Type of ECEC system Compulsory school age Entitlement to ECEC Attendance rate at ECEC (2009)
Under 3 years old (< 30 hrs/wk | 30+ hrs/wk) 3 years old to CSA (< 30 hrs/wk | 30+ hrs/wk)
European Economic Area member states (EU + EFTA, exc. Leichtenstein)
Austria (F) Unitary - Part (Welfare) 6 years 5 years: PT 10% (7% | 2%) 80% (58% | 21%)
Belgium (F) Split (Ed. dominant) 6 years 2.5 years: FT 33% (17% | 16%) 99% (30% | 69%)
Bulgaria Split (Ed. dominant) 7 years None 8% (1% | 7%) 55% (7% | 48%)
Cyprus Split (Ed. dominant) ** 4.8 years (school starts @ 5.8 years) 4.8 years: PT 22% (8% | 14%) 81% (40% | 41%)
Czech Republic Split (Ed. dominant) 6 years None 3% (3% | 0%) 64% (28% | 36%)
Denmark Unitary - Full (Education) * 7 years (6 years on voluntary basis) 6 months: FT 73% (10% | 63%) 84% (12% | 72%)
Estonia Unitary - Full (Education) 7 years 18 months *** (3 years: FT) 25% (4% | 21%) 92% (9% | 84%)
Finland Unitary - Full (Welfare) * 7 years (6 years on voluntary basis) Birth: FT 27% (6% | 21%) 78% (20% | 57%)
France Split (Ed. dominant) 6 years 3 years: FT 41% (16% | 25%) 95% (48% | 47%)
Germany (F) Unitary - Part (Welfare) 6 years 3 years: PT 19% (7% | 12%) 89% (48% | 40%)
Greece Split (Ed. dominant) ** 5 years (school starts @ 6 years) 5 years: PT 11% (4% | 7%) 58% (33% | 25%)
Hungary Split (Ed. dominant) ** 5 years (school starts @ 6 years) Birth *** (3 years: FT) 7% (2% | 5%) 74% (17% | 57%)
Iceland Unitary - Full (Education) 6 years None 41% (4% | 38%) 99% (7% | 92%)
Ireland Split (CC dominant) * 6 years (4 years on voluntary basis) 3.25 years: PT 20% (15% | 5%) 87% (74% | 13%)
Italy Split (Ed. dominant) 6 years None 25% (9% | 16%) 92% (20% | 73%)
Latvia Unitary - Part (Education) ** 5 years (school starts @ 7 years) 5 years: FT 15% (2% | 13%) 74% (7% | 67%)
Lithuania Unitary - Part (Education) 7 years None 10% (1% | 9%) 54% (4% | 51%)
Luxembourg Split (Ed. dominant) ** 4 years (school starts @ 6 years) 4 years: PT 34% (22% | 12%) 71% (46% | 26%)
Malta Split (Ed. dominant) 5 years None 8% (4% | 4%) 77% (30% | 47%)
Netherlands Split (CC dominant) * 5 years (4 years on voluntary basis) 4 years: PT 49% (43% | 6%) 87% (75% | 12%)
Norway Unitary - Full (Education) 6 years 1 year: FT 36% (8% | 27%) [2007] 81% (17% | 63%) [2007]
Poland Split (Ed. dominant) ** 6 years (school starts @ 7 years) 6 years: PT 3% (0% | 2%) 38% (8% | 31%)
Portugal Split (Ed. dominant) 6 years 5 years: FT 36% (2% | 34%) 81% (8% | 73%)
Romania Split (Ed. dominant) ** 5 years (school starts @ 6 years) 5 years: PT 5% (4% | 1%) 63% (44% | 19%)
Slovenia Unitary - Full (Education) 6 years 1 year: FT 31% (4% | 27%) 90% (16% | 73%)
Slovakia Split (Ed. dominant) 6 years None 3% (1% | 2%) 76% (13% | 63%)
Spain Unitary - Part (Education) 6 years 3 years: FT 36% (18% | 18%) 94% (50% | 44%)
Sweden Unitary - Full (Education) * 7 years (6 years on voluntary basis) 1 year: FT 63% (26% | 37%) 94% (29% | 65%)
Switzerland (F) Split (Ed. dominant) 6 or 7 years None 27% (23% | 4%) 74% (65% | 9%)
United Kingdom Unitary - Part (Education) * 5 years (4 years in N. Ireland)
(4 years on voluntary basis)
3 years: PT 35% (31% | 4%) 91% (70% | 21%)
Other countries
Canada (F) Split (CC dominant) 6 years None 24% (2006) 57% (2006)
New Zealand Unitary - Part (Education) * 6 years (5 years on voluntary basis) None 38% (2008) 95% (2008)
United States (F) Split (CC dominant) 6 years None 31% (2005) 58% (2005)


Abbreviations used in this table: F = Federal state; Ed. = Education; CC = Child care; PT = Part-time; FT = Full-time; CSA = Compulsory school age

Data are from 2009 unless otherwise indicated. Numbers are rounded.

Type of ECEC system:

Unitary = government responsibility integrated across all ECEC services. Full = system is fully integrated from birth to CSA, including access, regulation, funding and workforce. Part = government responsibility integrated, but not all of the other dimensions, i.e., access, regulation, funding and workforce. Education = government responsibility integrated in education; Welfare = government responsibility integrated in welfare or health.

Split = government responsibility for ECEC split between two ministries; all other dimensions split between two systems (education or welfare). CC dominant = child care provision/welfare dominant in ECEC system; Ed. dominant = school-based provision/education dominant in ECEC system.

Compulsory school age:

This mostly coincides with the child's age when starting at primary/elementary school. In some cases, children can enter primary/elementary school before the compulsory school age on a voluntary basis; countries where this occurs are shown using a "*" before the CSA, followed by the age at which children can enter school voluntarily.

In some countries, it is compulsory to attend an ECEC service for a period before starting school. In these cases, the age given is for compulsory attendance at ECEC and is preceded by "**", followed by the age at which children start in primary/elementary school.

Entitlement to ECEC:

Shows the age from which entitlement is operational. "***" indicates when the age of legal entitlement is not operational because of insufficient places, with the age following showing the actual age when entitlement is operational. PT = entitlement for 20 hours a week or less; FT = entitlement for longer than 20 hours a week.

Attendance rates:

Figures for European countries are for "formal" arrangements and include four types of child care and education: education at preschool; education at compulsory school; child care at centre-based services outside school hours (before/after) and child care at a day care centre. Therefore, formal arrangements include all kinds of care organised/controlled by a structure (public/private). Care provided by child-minders without any structure between the carer and the parents (direct arrangements) have been excluded from the definition of "formal" care in order to take into account only child care recognised as fulfilling certain quality patterns. The duration refers to average number of hours during a usual week (Eurostat, 2011).

The attendance figures for Canada, New Zealand and the United States are drawn from different sources compared to the European country data, which all come from European Union Statistics on Income and Living Conditions (EU-SILC). Data between these three countries and European countries is not, therefore, comparable; formal services, for example, may include a wider range of provisions, such as all registered child-minders carers at the home of the child.

The figures in square brackets show the proportion of children attending on average for fewer than 30 hours a week and the proportion attending for 30 hours a week or more; e.g., in Austria, 10% of children under 3 years attend formal services - 7% for fewer than 30 hours week, 2% for 30 hours or more.

Sources: Bennett, 2010 (type, compulsory school age, entitlement); Eurostat, 2011 (attendance rates excluding "other countries"); Oberhuemer, Schreyer, & Neuman, 2010; OECD, 2011 (attendance rates "other countries").

A.1 Systems of ECEC

Early childhood education and care services are defined, in part, by the age at which children move into compulsory primary schooling. It can be seen from Table A1 that 6 years is the most common compulsory school age. However, in a few countries (e.g., Bulgaria, Estonia, Lithuania, Sweden, parts of Switzerland) compulsory school age is 7 years; while in a few others (e.g., Netherlands, the UK), it is 5 or even younger. But, as the paper will later discuss, compulsory school age and the transition from ECEC to primary schooling do not always coincide, either because a period of attendance at an ECEC service has itself been made compulsory or because children can enter primary schooling on a voluntary basis before compulsory schooling.

Types of ECEC systems

The main way in which ECEC systems are classified in Table A1 (see "Type of ECEC system", column 2) is according to whether administrative and policy responsibility for ECEC is split between two government departments (usually welfare, or health and education), or whether administrative and policy responsibility is integrated within one department (usually education; welfare in one case, Finland).A1 Historically, every country has started out with a "split" system of ECEC, which, broadly stated, has involved a division between "(child) care" services in welfare or health, with provision mainly made in some form of non-school centre and family day care; and "early education" services in education, with provision that is often school-based. Some countries, however, in recent decades have moved towards a more or less integrated system (for further discussion of this development, see Kaga, Bennett, and Moss, 2010).

Each of these categories is further divided into two sub-groups. The countries with integrated responsibility - labelled "Unitary" in Table A1 - divide into those whose ECEC systems are fully integrated along four key dimensions: access, regulation (including curriculum), funding and workforce; and those where one or more of these key dimensions is not fully integrated. In the case of the former countries, the ECEC system now operates as a seamless whole; in the latter countries, important elements of the former split remain, despite administrative and policy integration.

The countries that have not integrated responsibility - labelled "Split" in Table A1 - divide between those where education provision is dominant, providing ECEC services for a majority of children, usually with most children from 3 years of age attending some form of school provision for a three-year period; and those where child care provision is dominant, mainly because educational provision is limited to a period of two years or less, and child care provision provides for children up to 3 years and, in some cases, for a substantial proportion of children over 3 years. In the former "education dominant" group, early education is more likely to be provided in separate "nursery schools" (the French école maternelle being a well-known example). While in the latter "child care dominant" group, early education is more likely to consist of nursery or kindergarten classes attached to a primary school.

Following this four-way typology, countries in the table can be grouped as follows:

  1. unitary, fully integrated: Denmark, Finland, Iceland, Norway, Slovenia, Sweden
  2. unitary, not fully integrated: Austria, Estonia, Germany, Latvia, Lithuania, Spain, UK, New Zealand
  3. split, education dominant: Belgium, Bulgaria, Cyprus, Czech Republic, France, Greece, Hungary, Italy, Luxembourg, Malta, Poland, Portugal, Romania, Slovakia, Switzerland
  4. split, child care dominant: Ireland, Netherlands, Canada, United States.

These different types of system have implications for access. As OECD's (2006) Starting Strong review concluded, "variations in access and quality are lessened [when policy is integrated] under one ministry" (p. 49). Whereas countries in groups 2, 3 and 4 may provide universal access or entitlement to children over the age of 3, only countries with fully integrated unitary systems (group 1) extend the same access conditions to children both over and under 3 years: universal access from at least the age of 1 year is one of the defining features of this group.

It will be immediately obvious that countries are not randomly grouped. The unitary, fully integrated group is predominantly Nordic; the split, education dominant group is mainly Continental European; while the split, child care dominant group is made up of predominantly Anglophone countries. There is also substantial correspondence with some typologies of welfare state regimes, particularly those that divide countries into "social democratic", "conservative" and "liberal" regimes (Esping-Andersen, 1990).

Compulsory age for starting education

As already noted, there is some variation in compulsory school age (see column 3 of Table A1), though 6 years is the most common. Several countries have, however, moved in recent years to make a period at ECEC compulsory, usually one year preceding the start of school, and usually justified in terms of ensuring children's "school readiness". Countries that have lowered the age of compulsory attendance in this way are mainly from central, eastern and southern Europe and include: Cyprus, Greece, Hungary, Latvia, Luxembourg, Poland and Romania.A2 Other countries offer early admission to school on a voluntary basis, in some cases (such as Denmark and Sweden) providing a special "pre-school" class for these early entrants. These trends are weakening the relationship between the onset of compulsory education and the transition from ECEC to primary school.

Attendance rates

The issue of entitlement to a place in an ECEC service (see column 4 of Table A1) is discussed under "Access to ECEC services". The last part of Table A1 (column 5) provides information on attendance rates. Here the discussion on definitions and comparability of attendance rates, given in the notes section under the table, is important. For example, the data for European countries are comparable in the sense that they are based on countries collecting and classifying data to an agreed, common format, but they do not include certain providers, such as many child-minders and carers coming into a child's own home; by contrast, the data for the three non-European countries come from individual, national datasets (i.e., non-comparable sources) and may include some of the carers omitted in the European data. Both sets of data should exclude provision made informally, especially by grandparents and other relatives, which in many countries continues to make a major contribution to child care, especially for children under 3 years (though, it is interesting to note, relatives and friends have virtually disappeared as providers of such regular care in the Nordic countries, where parents have access to well-run and highly subsidised formal services for children from 12 months or younger) (Bennett & Moss, 2010).

Children under 3 years of age

Table A1 shows clearly how attendance rates for children under 3 years are generally far lower than for children 3 years and older. One reason for this is that most countries provide entitlements to paid postnatal leave, which reduces demand for ECEC among parents with very young children, especially where leave is paid at a high earnings-related rate. (For further information on leave entitlements, see Moss, 2011; see the Annex for a discussion of the relationship between leave and ECEC entitlements.) Another and more important reason is that ECEC services for children under 3 years are generally less available and more costly than services for children 3 years and over; as the next section demonstrates, entitlement to use such services often does not exist. However, having said this, attendance for children under 3 years varies considerably between countries, ranging from Denmark's very high rate of 73% to 5% or less in four countries.

Countries can be categorised into three groups, according to their ECEC attendance rates:

  1. high (33% or over): Denmark, France, Iceland, Luxembourg, Netherlands, Norway, Portugal, Spain, Sweden, UK, New Zealand;A3
  2. medium (11-32%): Belgium, Cyprus, Estonia, Finland, Germany, Greece, Ireland, Italy, Latvia, Slovenia, Switzerland, Canada, USA; and
  3. low (10% or less): Austria, Bulgaria, Czech Republic, Hungary, Lithuania, Malta, Poland, Romania, Slovakia.

Spotting commonalities between the countries in each group is not so easy as in the previous categorisation of ECEC structures. The "high" group includes four Nordic countries, but not Finland - the only Nordic country with a 3-year parental leave option - and also includes two southern European countries (Portugal and Spain) and two Anglophone countries (UK and New Zealand). The "low" group mostly consists of central and eastern European countries, formerly in the Soviet-dominated bloc, whose low levels of provision in part reflect the running down of services for under 3s after regime changes, due to financial pressures and a widespread reaction against employment among women with young children (for a discussion of this process in the Czech Republic and Hungary, see chapters on these countries in Kamerman & Moss, 2009). The "medium" group contains a very disparate group of countries.

A further way of looking at attendance rates is to consider how, in each country, the overall rate of attendance is divided between shorter and longer hours of attendance, defined here as under 30 hours a week and 30 hours a week and over respectively. The same overall attendance rate in two countries may disguise a very different pattern of usage when hours of attendance are taken into account. This analysis is only possible for European countries, given the data available. But among those countries, there are large differences. For example, to take the extremes, in Denmark, 73% of under 3s attend formal services, 86% of whom attend for 30 hours a week or more (63% out of 73%); while at the other extreme, in the Netherlands, the pattern is reversed, with 49% of under 3s attending formal services, and only 12% going 30 hours a week or more (6% out of 49%). Attendance for the full day is the norm in Denmark, while attendance for part-time hours is the norm in the Netherlands.

If we exclude countries with low levels of attendance overall (10% or less), the remainder can be divided into three groups:

  1. longer hours of attendance > shorter hours (by more than 5 percentage points): Denmark, Estonia, Finland, France, Iceland, Italy, Latvia, Norway, Portugal, Slovenia, Sweden;
  2. longer and shorter hours similar (within 5 percentage points): Belgium, Germany, Greece, Spain; and
  3. shorter hours of attendance > longer hours (by more than 5 percentage points): Cyprus, Ireland, Luxembourg, Netherlands, Switzerland, United Kingdom.

All six countries with fully integrated unitary systems appear in group 1, which also includes many countries with relatively low part-time employment rates for women. At the other end of the spectrum, group 3, are three countries (Ireland, Netherlands and the UK) with high levels of part-time employment, and also substantial numbers of 2-year-olds attending part-time "playgroup" services.

Children 3 years of age and older

Moving to attendance rates for children 3 years and over, the much higher figures in every country reflect a trend, at least in Europe, towards near universal attendance for this age group; and as compulsory schooling age is mainly 6 years, this is a trend also towards three years of attendance at ECEC. Again, the 33 countries can be divided into three groups:

  1. high (90% or over): Belgium, Estonia, France, Iceland, Italy, Slovenia, Spain, Sweden, UK, New Zealand;A4
  2. medium (70-89%): Austria, Cyprus, Denmark, Finland, Germany, Hungary, Ireland, Latvia, Luxembourg, Malta, Netherlands, Norway, Portugal, Slovakia, Switzerland; and
  3. low (below 70%): Bulgaria, Czech Republic, Greece, Lithuania, Poland, Romania, Canada, United States.

This distribution, again, presents no clear-cut patterns. Two Nordic countries are in the "high" group, but three are in "medium"; and two Anglophone countries (New Zealand, UK) are in the "high" group, while two (Canada and the United States) are in the "low" group. What is noticeable is that four continental western European countries with strong 3-year "nursery education" services (Belgium, France, Italy, Spain) are in the "high" group; while five central and eastern European countries (Bulgaria, Czech Republic, Lithuania, Poland and Romania) are in the "low" group.


Taking an overview of attendance rates for children under 3 years and 3 years and over, and taking account of both total attendance and the balance between shorter and longer hours, the lead countries for ECEC attendance are three of the five Nordic states (Denmark, Iceland and Sweden), together with Belgium and France. The countries that lag most are Bulgaria, Greece, Lithuania and Poland.

A.2 Access to ECEC services

The concept of access is, as the second report of OECD's (2006) Starting Strong review reminds us, "a complex one" (p. 75). For a start, universal access or entitlement need not equate with or require complete coverage:

Universal access does not necessarily entail achieving full coverage, as there are variations in demand for ECEC at different ages and in different family circumstances. Rather, it implies making access available to all children whose parents wish them to participate. (p. 14)

Furthermore, parental demand may not be uniform across or within countries; it may vary between different groups and between different areas, and it may change over time. Employment patterns and the availability of substitute policies (e.g., parental leave) can affect the call for and use of ECEC services; so too can changing parental expectations and understandings of good parenting and what a good childhood is. This mix of factors is well exemplified by the high attendance at ECEC in Nordic countries like Denmark and Sweden, which reflects not only high parental employment rates and high levels of good-quality provision, but also a high level of parental satisfaction with the well-developed and accessible system of ECEC services in these countries. Two Swedish researchers (Lenz Taguchi & Munkammar, 2003), in a report for UNESCO, highlighted the related factor of changing expectations among Swedish parents:

Enrolling children from age one in full-day pre-schools has become generally acceptable. What was once viewed as either a privilege of the wealthy for a few hours a day, or an institution for needy children and single mothers, has become, after 70 years of political vision and policy-making, an unquestionable right of children and families. (p. 27)

Then, again, to be accessible - to make access to ECEC a realistic option - services have to meet certain conditions. For example, they need to be free or available at a price all parents can afford, which in turn calls for public investment: affordability, as OECD's (2006) review noted, is "often an issue, and in many countries, low-income groups are excluded in practice from access to centre-based services" (p. 46). Services need to be equally available to all age groups and to provide an offer that parents need and want, in terms of, for example, quality, opening hours and type of provision. In sparsely populated areas, they need to be within reasonable travelling distance. More generally, services need to be equally available in all parts of a country, and this may be an especially difficult issue where there is strong decentralisation of responsibility for ECEC:

Experience from the OECD reviews suggests that devolution of powers and responsibilities may widen differences of access and quality between regions. In the devolution process, it seems important to ensure that early childhood services are part of a well-conceptualised national policy, with, on the one hand, devolved powers to local authorities and, on the other, a national approach to goal setting, legislation and regulation, financing, staffing criteria, and programme standards (OECD, 2006, p. 13)

Last but not least, ECEC services need to recognise and be responsive to the diversity of children and families and their needs:

Access is often inappropriate for children with special needs and/or additional learning needs, so much so that directors of centres may not allow them to enrol, or parents - seeing the difficulties involved for their children - simply desist. If access is achieved, classes may be far too large for these children, or appropriately trained staff may not be available to take them in charge. Similarly, group sizes, care and pedagogical approaches may be unsuitable for very young children in early education systems established along school lines (OECD, 2006, p. 82).

Taking these points into account, and looking at the conclusions of Starting Strong, as well as the overview of 33 countries in Table A1, a few broad generalisations can be made. First, at present the access "winners" are children over 3 years, at least in Europe where:

the concept of universal access for 3- to 6-year-olds is generally accepted. Most countries provide all children with at least two years of free, publicly-funded provision before they begin primary schooling. In fact, with the exception of Ireland and the Netherlands, [among the countries reviewed] such access is generally a statutory right from the age of 3 years, and in a handful of countries from an earlier age. Early education programmes in Europe are often free, and attached to schools (OECD, 2006, p. 77).

The point to be emphasised here is that universal access in Europe is not limited to the year prior to compulsory schooling, but is well on the way to coverage of a two- or (usually) three-year preschool period. Outside Europe, however, the OECD (2006) Starting Strong review concluded that "provision is generally much weaker" (p. 77).

By contrast, and clearly shown in Table A1, access for children under 3 years is much lower, in Europe and elsewhere. Lower levels of access - because there are fewer services and/or they are more costly - are associated with increased inequality, particularly for children from lower income and migrant backgrounds. A revealing analysis of European Union Statistics on Income and Living Conditions (EU-SILC) data (a source discussed further in A.3), this time for 2005, shows the extent of inequality in access to formal services for children under 3 years, depending on their mothers' level of education (Table A2).

Table A2: Comparison of attendance rates in Netherlands, United Kingdom, Denmark and Sweden for children under 3 years, by mother's level of education, 2005
Mother's level of education Netherlands (%) United Kingdom (%) Denmark (%) Sweden (%)
Low 16 13 75 52
Medium 37 30 72 56
High 59 39 70 47

Source: Unpublished secondary analysis of EU-SILC data, cited in Bennett and Moss (2010)

In the Netherlands and the UK - countries with marketised systems and no entitlement for children under 3 years - inequality of access is marked, both in absolute terms and compared with access to formal services in Denmark and Sweden - countries with high and sustained levels of public support and entitlement for children under 3 years. Although the unequal access in the Netherlands and the UK is highlighted here, the same pattern is apparent in most non-Nordic European countries.

To get a full, comprehensive and comparative picture of access, taking account of all the issues raised above, would require access to large, detailed and multinational datasets that simply do not exist at present. The best that can be done for the moment is to consider, in the broadest of terms, the overall situation on access in the 33 countries, taking account of whether an entitlement to an ECEC place has been established, and levels of provision. Once again, the situation in the 33 countries considered here can be reduced to four broad groups:

  • entitlement to ECEC:
    • with obligatory attendance;
    • without obligatory attendance;
  • no entitlement to ECEC:
    • with universal access; and
    • without universal access.

Entitlement to ECEC

Entitlement with obligatory attendance

As already noted, seven of the 33 countries have made ECEC attendance compulsory for 1 to 2 years of the period immediately before the start of primary school. The other side of compulsory attendance is that such attendance must be free. In all cases, this obligatory attendance is for shorter hours, so that entitlement is limited in terms of both duration and hours of attendance.

Entitlement without obligatory attendance

In these cases, government is required by law to provide a place in an ECEC service if requested to do so by a parent. Usually countries only introduce such entitlements when they are confident there is sufficient capacity to meet demand. But in two of the 33 countries covered (Estonia and Hungary), though laws exist that entitle children under 3 years to ECEC provision, these entitlements cannot be met due to a shortage of places.

In addition to the seven countries where attendance is obligatory, 16 out of the 33 countries provide an entitlement to attend ECEC for a period of time. This entitlement is:

  • 5 years in Austria, Portugal and 4 years in the Netherlands - in each of these three cases, entitlement is only for the year immediately before compulsory schooling;
  • 2.5-3.5 years in Belgium, France, Germany,A5 Ireland, Spain and the UK; also, in effect, from this age in Estonia and Hungary, even though the law refers to an entitlement from an earlier age, because there are insufficient places to meet demand; and
  • 1 year or less in Denmark, Finland, Norway, Slovenia, Sweden.

As just noted, the most common age at which entitlement comes into force is around 3 years, giving 2 to 3 years of "entitled" attendance. Four of the five Nordic states, plus Slovenia, have taken entitlement further to 12 months or younger, though, to reiterate an earlier point, this does not mean that every child attends a formal service from these young ages. Entitlement in these five countries means the possibility of attending services on a part-time or full-day basis. Entitlement starting at 3 years or above is more likely to be for part-time hours; that is, 20 hours a week or less, though it is longer in some countries.

The period of attendance covered by entitlement is usually free of charge to parents, removing the issue of affordability. However, parents in the four Nordic countries and Slovenia do have to make a financial contribution, though that contribution is kept low by a high level of public funding. A partial exception is Sweden, where there is a free period of attendance of 525 hours a year for 3-5 year olds, with the remaining attendance period (before 3 years and beyond 525 hours a year) attracting a charge.A6

The free attendance is related to the rationale for entitlement, which is educational; that is, entitlement is seen as providing children with access to early education, and education is usually treated as a free (at time of use) public good. The Nordic countries and Slovenia, which require a parental contribution, have fully integrated all their ECEC services (in education, except for Finland where, until 2013, welfare remains responsible for all ECEC), treating care and education as inseparable; in other words, they no longer distinguish between "child care" and "education", and the entitlement can be seen as being both for children and parents and both for child care and educational reasons. Sweden provides an important example of this thinking. Entitlement for children from 12 months was initially introduced in 1995, but only for children with employed or studying parents or who had special needs. Following the transfer in 1996 of government responsibility for the integrated ECEC system from welfare to education, entitlement was extended to all children, irrespective of their parents' employment status.

No entitlement to ECEC

This leaves ten countries with no entitlement to ECEC access at any age, for compulsory or voluntary attendance: Bulgaria, Iceland, Italy, Lithuania, Malta, Slovakia, Switzerland, Canada, New ZealandA7 and the United States. These countries can be divided into two more groups.

No entitlement but access for all parents wanting a service

In a few countries, provision of ECEC services is so extensive that all children whose parents choose to send them have access to these services, even though there is no formal entitlement. This situation applies to services for children 3 years and over in Iceland and Italy, with attendance in Italy being free of charge.

No entitlement, no universal access

The situation of having a shortage of places (compounded sometimes by the cost of what are available) applies in a number of countries, especially for children under 3 years. In such circumstances, children may have to remain in full-time maternal care or else parents make widespread use of informal arrangements, such as relatives.


To conclude, this discussion has emphasised that the concepts of "access" and "entitlement" may overlap, but that they are not synonymous. Universal access, on paper, may not convert into universal access in reality if services are too expensive, inadequately distributed or unresponsive to the diverse needs of children and parents. Universal access may exist without the accompaniment of a legal entitlement to access; sufficient places exist to meet demand. While, in a few cases, legal entitlement may exist without universal access due to an insufficiency of places, this renders the entitlement meaningless and effectively nullifies it. Mostly, though, countries with an entitlement have universal access; "universal" in this case meaning not that all children attend but that all children attend whose parents wish them to. In a few cases, though, entitlement and universal access are backed by the state requiring that all children do attend ECEC for a period of time before formal primary schooling.

A.3 How access may be measured

There are two broad approaches to the measurement of access, which can also contribute to an assessment of the effects of entitlement on attendance. Both can be illustrated by the case of England, where an entitlement to free part-time attendance for 3 and 4 year olds at ECEC, to receive early education, was introduced for 4-year-olds (in 1998) and for 3-year-olds (in 2004). It should be noted that this entitlement can be delivered by a variety of public and private providers in a variety of types of provision.

The first method is based on administrative returns from local authorities and providers. The latest published information on the "provision of education for under fives in the maintained, private, voluntary and independent sectors" in England is for January 2011 (Department for Education [England], 2011) and draws on three Census exercises:

  • The Early Years Census: Local authorities (LAs) are "under a statutory obligation to send individual child-level information for every funded three and four year old child in a private, voluntary or independent provider. The return provides child-level information about the number of three and four year olds benefiting from some free early education in private, voluntary and independent providers. Also, the return records the aggregate numbers of three and four year olds taking up early education places in those private and voluntary providers and independent schools that are registered with LAs and receive some government funding for delivery of the free entitlement" (p. 5; original emphasis).
  • The School Census: This collects data from schools on "the number of three and four year olds benefiting from some early years education in maintained [i.e., publicly funded] schools, at January of each year" (p. 5).
  • The School Level Annual School Census: this collects information on "the number of three and four year olds benefiting from some early years education in independent schools" (p. 5).

These three sources are used to produce an annual report called Provision for Children Under Five Years of Age in England, though its remit is in fact limited to 3- and 4-year-olds receiving the publicly funded early education entitlement.

Tables produced for the Department of Education (England) report (2011) cover numbers of "three and four year olds taking up or benefiting from early education places" (p. 6) by type of provider and by local authority area (of which there are 152). Take-up is also expressed in terms of percentage of the population of 3- and 4-year-olds who attend early education, based on mid-year estimates and projections of population, provided by the UK Office of National Statistics. For example, it is estimated that in England in January 2011, "the number of 3 and 4 year olds benefiting from some free early education (where each child is counted once …) was 1,224,465 or 95% of the 3 and 4 year old population" (p. 1). Put another way, it appears that, on the Census dates, only about 5% of children were not taking up the two-year part-time entitlement.

The second method for measuring access is to use large-scale national sample surveys targeted on specific groups, in this case, the eligible population for the entitlement to early education: parents with at least one child aged 3 or 4 years. This approach was adopted by the previous (Labour) government, to monitor take-up of the newly introduced entitlement to early education. Annual surveys were commissioned over six years (running from 1997 to 2002) of parents of three and four year old children and their use of early years services, drawing a large sample of 3- and 4-year-olds from government-held Child Benefit records, then interviewing their parents to gather information on usage of ECEC services for 3- and 4-year-olds, as well as on parental perceptions and attitudes (for the last survey, in 2002, see Bell and Finch, 2004).

Because this survey was focused specifically on use of ECEC by 3- and 4-year-olds, it was able to cover a wide range of related items, not only on access to and use of services, but on issues such as parents' understanding of and attitudes towards ECEC, relationships between parents and providers, parent perceptions of children's transition into school, and learning activities at home. Such surveys also score over administrative data in that they can include and examine families not using ECEC.

Further surveys were commissioned by the English government in 2007 and 2008 to examine parents' use of child care and early years services in England. These surveys were still targeted, but rather more widely, having a broader remit, to cover parents with children aged up to and including 14 years, though oversampling families with 2-4 year olds "in order to enable a more detailed analysis of the take-up of early years education by this age group" (Speight, Smith, La Valle et al., 2010, p. 1). A spin-off from the second of these surveys, was a study of the take-up of the universal early education entitlement by disadvantaged families, going further into the relationship between universal access and actual usage. The conclusions from this secondary analysis work (Speight, Smith, Coshall, & Lloyd, 2010) - cited at some length here to demonstrate the potential of such targeted survey work - included that:

  • children from lower income and larger families (i.e., with three or more children), those whose mothers did not work and those whose mothers did not have any academic qualifications were less likely to receive early years provision;
  • children from lone-parent families were more likely to receive early years provision than those from two-parent families when the analysis took account of differences in work status, income and other socio-demographic characteristics between these two types of families;
  • the take-up of early years provision was similar for boys and girls, and it did not vary by whether the child had a longstanding illness or disability or by whether they had special educational needs;
  • compared with children whose mothers were white, children of black African, Pakistani and Bangladeshi mothers were substantially less likely to receive early years provision; however, once the analysis controlled for differences in socio-demographic profile between families from different ethnic backgrounds, the effect of ethnicity on uptake of early years provision was no longer significant; and
  • children living in deprived areas were less likely to receive early years provision than those living in more affluent areas.

Overall, the authors (Speight, Smith, Coshall, & Lloyd, 2010) concluded that:

there was a strong association between the level of multiple disadvantage experienced by the family and their take-up of early years provision. In families experiencing no disadvantage only 3% of children were not attending any early years providers, whereas the proportion was higher for families experiencing some or a lot of disadvantage, with the highest figure (13%) found among the most disadvantaged families.

There were differences in the types of providers attended by children depending on the level of disadvantage experienced by their families. Children from the most disadvantaged families were more likely to receive early years provision at nursery classes and less likely to receive it at playgroups/pre-schools, day nurseries and childminders, whereas attendance at reception classes and nursery schools did not vary by level of multiple disadvantage. (pp. 2-3)

An alternative application of the survey method to gather information on access and take-up of entitlement is to add questions on these subjects to non-targeted general surveys. While the range of items to be covered will be more limited, due to the other demands being placed on the survey, this may prove less costly and ensure long-term data. The EU's information on attendance rates across member states is currently based on this method, working through the EU-SILC, which aims to collect "timely and comparable cross-sectional and longitudinal multidimensional microdata on income, poverty, social exclusion and living conditions" for each EU member state" (Eurostat, 2010). Data collection is linked to an existing survey in each member state, and harmonised methods and definitions are applied to establish reliable comparisons between member states. The information required for EU-SILC from each member state falls into five core areas, one of which is labour market data that includes "child care" (though in effect this picks up attendance at all formal ECEC services, including nursery schooling).

EU-SILC and similar general surveys have the potential to link up data on use of ECEC services to other items included in the survey. Such linkages, however, need resources and capacity for secondary analysis, meaning that some of the potential of this method may not be realised. The routinely published information on ECEC from EU-SILC is limited to attendance at formal services by country, age of child and hours of attendance. However, much more use could be made of the dataset, as Table A2 illustrates.

From these examples, it might be concluded that the issue is not so much how to measure the extent to which universal access or entitlement is taken up, but what other questions are thought to be important in relation to access and entitlement. Who uses access opportunities and who does not? Why do users use ECEC and why do non-users not do so? Is use higher in some areas than others? How are ECEC services used and how would parents like them to be improved? It is the range and type of questions posed that will determine the best means to be used to measure access and entitlement.

A.4 Concluding comments

As early childhood education and care moves up national policy agendas, levels of provision are increasing and it is becoming more common for countries to provide an entitlement to such services. Entitlement is usually for two to three years from the age of 3 years to compulsory school age, and in a few cases overlaps with making ECEC attendance compulsory for a period immediately before school. Provision and attendance for children under 3 years lags behind, though it is also increasing in most countries. However, only a few countries have extended entitlement to children under 3 years, and those are countries with fully integrated systems of ECEC, well-developed services, and systems of parental leave that are coordinated with ECEC entitlement.

Administrative data can give a broad overview of actual attendance rates, including the extent to which entitlement is taken up by parents for their children. But other methods, particularly surveys, are needed to get a fuller picture of which children and families are using or not using ECEC services, and also their reasons for doing so. This fuller picture is important if there is a political and policy concern that ECEC services are accessible to and accessed by all children and families. It is also important that disadvantaged groups or groups with additional needs are not under-represented, as high overall levels of attendance may mask concentrations of non-attendance among those groups that policy most wants to reach.

A.5 Annex: Relationship between entitlement to parental leave and ECEC

Most discussions of policy frameworks to support employed parents with young children highlight two key policies: parental leave and ECEC. Table A3 looks at the relationship between these two policy areas, and in particular at whether they are coordinated in the sense that an entitlement to leave leads immediately into, or coincides with, an entitlement to ECEC. The table includes 27 countries, mostly but not wholly corresponding to the 33 countries in Table A1. Data have been taken from the 2011 annual international review of leave policies and research produced by the international network on leave policies and research (Moss, 2011).

Table A3: Relationship between leave and ECEC entitlements, 2011
  Child's age (months) at: Gap between …
End of leave End of well-paid leave Start of ECEC entitlement end of leave & start of ECEC end of well-paid leave and start of ECEC
Austria 24 1.8 • 5 years PT 36 months 58 months
Belgium 34 4.0 •• 2.5 years No gap 26 months
Canada (2006) 12 None - - -
Québec 13 13 - - -
Croatia 12 + bonus 2 12 + bonus 2 - - -
Czech Republic 36 5.1 - - -
Denmark 14 14 •• 6 months No gap No gap
Estonia 36 18 • 18 months (* 3 years) No gap 18 months
Finland 36 9 + bonus 1 •• End of parental leave No gap No gap
France 36 3.5 • 3 years No gap 33 months
Germany 36 13.3 + bonus 2 • 3 years PT No gap 20 months
Greece 19-60 6-12 • 5 years PT 41 months 48-54 months to no gap
Hungary 36 24 • Birth (* 3 years) No gap 12 months
Iceland 15.2 9.0 - - -
Ireland 15.7 6.0 • 3.25 years PT 24 months 33 months
Italy 13.7 + bonus 1 3.7 - - -
Luxembourg 13.8 1.8 • 3 years PT 22 months 34 months
Netherlands 14.3 2.3 • 4 years PT 34 months 46 months
New Zealand 12.5 3.2 - - -
Norway 36.7 12.2 •• 12 months No gap No gap
Poland 41.1 5.1 • 6 years PT 31 months 67 months
Portugal 36 + bonus 1 6 + bonus 1 • 5 years 24 months 54 months
Slovenia 14.2 11.5 •• End parental leave No gap No gap
Spain 36 5.1 • 3 years No gap 31 months
Sweden 18 13.2 •• 12 months No gap No gap
Switzerland 3.2 3.2 - - -
United Kingdom 18.5 1.4 • 3 years PT 18 months 35 months
USA - - - - -


Well-paid leave = Period of leave paid at least at 66% of normal earnings.

Bonus = A bonus of additional months of leave is available to a family where part of the basic entitlement is shared by mothers and fathers. This is intended to provide an incentive for fathers to take leave.

- = No entitlement to ECEC; • = Entitlement to ECEC, but only from 3 years or older; •• = Entitlement to ECEC from below 3 years of age; PT = entitlement for 20 hours a week or less; * = There is an obligation to provide a place, but this obligation cannot be met due to shortages of provision. This shortage usually applies to children under 3 years, and the bracketed figure indicates when the entitlement can usually be met in practice. The gap in the next column refers to the later age; that is, when the entitlement is implemented in practice.

Source: Moss (2011)

Nineteen countries have an entitlement to an ECEC service, but in most cases - 13 countries - this is only from 3 years of age or later (including Estonia and Hungary, which have a legal entitlement to ECEC at an earlier age, but shortage of places means the entitlement is not operational until around 3 years). Entitlement in this group is often limited to part-time places.

Only six countries have entitlement before 3 years: at 2.5 years in Belgium, and at 12 months or younger or at the end of parental leave in five countries: Denmark, Finland, Norway, Slovenia and Sweden, with full-time places being available in all cases. It is only in these five countries that policies are designed to ensure no gap between the end of well-paid leave and the start of an ECEC entitlement. Elsewhere, this gap is from 18 to 67 months that, if combined with countries that have no ECEC entitlement, emphasises the extensive lack of coordination between these two policy areas.


  • Bell, A., & Finch, S. (2004). Sixth survey of parents of three and four year old children and their use of early years services. London: Department for Education and Skills.
  • Bennett, J. (2010). Early childhood care and education regional report: Europe and North America. Paris: UNESCO. Retrieved from <>.
  • Bennett, J., & Moss, P. (2010). Working for inclusion: An overview of European Union early years services and their workforce. Edinburgh: Children in Scotland. Retrieved from <>.
  • Council of the European Union. (2009, 29 May). Council: Notices from European Union institutions and bodies. Official Journal of the European Union (C119/2-C119/10). Retrieved from <>.
  • Department for Education (UK). (2011). Provision for children under five years of age in England: January 2011 (Statistical First Release). London: Department for Education. Retrieved from <>.
  • Esping-Andersen, G. (1990). The three worlds of welfare capitalism. Cambridge: Polity Press.
  • European Council. (2002). Presidency conclusions: Barcelona European Council, 15 and 16 March 2002. Brussels: European Commission. Retrieved from <>.
  • Eurostat. (2010). European Union Statistics on Income and Living Conditions (EU-SILC): Description of dataset. Luxembourg: Eurostat. Retrieved from <>.
  • Eurostat. (2011). Formal child care by duration and age group. Luxembourg: Eurostat. Retrieved from <>.
  • Kaga, Y., Bennett, J., & Moss, P. (2010). Caring and learning together: A cross- national study of integration of early childhood care and education within education. Paris: UNESCO.
  • Kamerman, S., & Moss, P. (Eds.) (2009). The politics of parental leave policies. Bristol: Policy Press.
  • Lenz Taguchi, H., & Munkammar, I. (2003). Consolidating governmental early childhood education and care services under the Ministry of Education and Science: A Swedish case study. Paris: UNESCO. Retrieved from <>.
  • Moss, P. (Ed.) (2011). International review of leave policies and related research 2011. Retrieved from <>.
  • Oberhuemer, P., Schreyer, I., & Neuman, M. (2010). Professionals in early childhood education and care systems: European profiles and perspectives. Farmington Mills MI: Barbara Budrich Publishers.
  • Organisation for Economic Cooperation and Development. (2006). Starting strong II. Paris: OECD.
  • Organisation for Economic Cooperation and Development. (2011). OECD Family Database. Paris: OECD. Retrieved from <>.
  • Speight, S., Smith, R., Coshall, C., & Lloyd, E. (2010). Towards universal early years provision: Analysis of take-up by disadvantaged families from recent annual childcare surveys (Research Brief DFE RB066). London: Department for Education. Retrieved from <>.
  • Speight, S., Smith, R., La Valle, I., Schneider, V., Perry, J., Coshall, C., & Tipping, S. (2010). Childcare and Early Years Survey of Parents 2008 (Research Report No. DCSF RR136). London: Department for Children, Schools and Families. Retrieved from <>.


A1 Finland has recently announced that it will transfer responsibility for its integrated ECEC system from welfare to education at the start of 2013. All countries with fully integrated systems will then have those systems located in education.

A2 There are arguments for and against making ECEC compulsory. The argument in favour emphasises school readiness, especially for "disadvantaged" children whom, it is argued, are in particular need of additional support to be ready for primary school. Under a voluntary system, attendance may be over 90%, but the small group of non-attenders usually includes disproportionate numbers of "disadvantaged" children (see, for example, evidence from England discussed from page A11). The argument against emphasises parents' right to choose whether or not to send their children to ECEC and also the need for ECEC services to work on reforming themselves to become more responsive and more welcoming to attract all families, including those who feel alienated or excluded by unreformed services. Compulsory attendance may also remove the need for services to be self-critical and innovative.

A3 The European Union in 2002 set ECEC targets - usually referred to as the "Barcelona Targets" - for member states: "Member States should … strive, taking into account the demand for child care services and in line with the national patterns of childcare provision, to provide childcare by 2010 to… at least 33% of children under 3 years of age" (European Council, 2002, p. 12).

A4 The EU's Barcelona Targets, referred to in footnote 3, also specify that "Member States should … strive, taking into account the demand for child care services and in line with the national patterns of childcare provision, to provide childcare by 2010 to at least 90% of children between 3 years old and mandatory school age" (European Council, 2002, p. 12). More recently, the strategic framework for European cooperation in education and training, adopted by the EU in May 2009, set a benchmark to be achieved by 2020 that at least 95% of children between the age of four and the age for starting compulsory primary education should participate in early childhood education (Council of the European Union, 2009).

A5 Local authorities in Germany must, by 2013, ensure ECEC provision for children under 3 years for all parents who are employed or undertaking professional training, which will extend entitlement downwards in age.

A6 This charge is capped by a maximum fee. For example, the maximum that a parent with a child aged 1 to 6 years attending an ECEC service pays is A$185 a month; the maximum for a second child is A$125.

A7 New Zealand provides funding to services for 20 hours of early childhood education for 3 and 4 year olds. But this is only available to children attending services that "opt in" to the scheme, and attendance for this age group is not an entitlement.