OECD countries spent an average of USD 13 331 per child on early childhood education (ECE) in the 2022 financial year, with most of the expenditure (USD 11 483 on average) coming from government sources. Total expenditure per child at this level slightly exceeds expenditure per full-time student at primary level.
Although enrolment numbers in pre-primary education (ISCED 02) have remained stable between 2015 and 2022 on average across the OECD, governments of OECD countries have increased funding at this level by 23%. As a result, government expenditure per child in pre-primary education increased by 24% on average between 2015 and 2022 across OECD countries.
At pre-primary level, three-quarters of funds go to public institutions while the remainder go to either government-dependent or independent private institutions, closely mirroring the distribution of enrolment across institution types.
Education at a Glance 2025
In equivalent USD converted using PPPs, direct expenditure on educational institutions
Note: Expenditure per child is based on headcounts rather than full-time equivalent students.
1. Year of reference differs from 2022.
2. Data do not cover day care centres and integrated centres for early childhood education.
For data, see Table C2.1. For a link to download the data, see Tables and Notes section.
Context
Investment into early childhood education (ECE) can ensure better access and higher-quality care for young children. Furthermore, expenditure dedicated to ECE can be a policy lever to increase women’s labour-market participation rates and promote equity by ensuring strong foundations for further learning for disadvantaged children. Widely available and affordable early childhood education is also used a means to increase birth rates and limit demographic decline.
Interpreting ECE funding requires different consideration than at other levels of education. Young children require close and frequent adult supervision, raising the cost of provision. The ages covered by ECE contains a mix of compulsory and non-compulsory (but sometimes free) years of education, the structure of which varies across countries. Furthermore, while primary and secondary students generally attend full-day programmes, ECE programmes vary widely in how long children attend each day. Yet data on hours of participation are limited, making it impossible to calculate the full-time-equivalent (FTE) enrolment counts that are the basis for the per-student expenditure data used in other chapters. As a result, expenditure per child at the early childhood educational development and pre-primary levels is based on total spending divided by headcount enrolment, rather than FTE students. This may distort comparisons, as countries with shorter attendance hours can appear to invest less per child. In addition, not all countries report finance data for early childhood educational development, limiting most analysis to pre-primary education. The data also only cover programmes meeting ISCED classification criteria, such as requiring at least 2 hours of educational activities per day and 100 days a year (see the Definitions section in Chapter B2). The figures in this chapter do not capture provision without such explicit educational requirements, or home-based or informally organised care, underestimating the scale of early childhood education in countries where such arrangements are common.
Other findings
On average, OECD countries dedicate the equivalent of 0.59% of GDP to the education of children aged 3 to 5. The countries spending the highest amount on this age group are Iceland (1.05% of GDP), Norway (0.87%) and Israel (0.85%).
In the last decade, 12 countries lowered the starting age of compulsory education to include one or more years of pre-primary education. In Bulgaria, Czechia and Lithuania, this resulted in increased spending on pre-primary education but in the rest the change had little impact on pre-primary expenditure. As enrolment rates were already high in most of these countries, the reforms served to formalise existing levels of attendance rather than prompting significant changes in enrolment.
Systems that rely more on private funding are not more generous overall: there is no correlation between the amount of expenditure per child and the share of that amount funded by private stakeholders (mostly families).
Distribution of sources of funding for pre-primary programmes
The vast majority of funding for pre-primary institutions comes from government sources (around USD 10 500 on average across OECD countries compared to around USD 1 500 coming from private sources), as shown in Figure C2.1. This breakdown is consistent with funding patterns observed at primary, secondary and post-secondary non-tertiary levels (see Table C1.1). In Belgium, Bulgaria, Luxembourg and Romania, 95% or more of funding was disbursed by the government in 2022. Of these, over 90% of funding in Bulgaria, Luxembourg and Romania was spent on public institutions (Table C2.3). Meanwhile, Belgium’s expenditure on pre-primary education is split nearly equally between public and government-dependent private institutions, illustrating the government allocation of funding set by the School Pact of 1958 (Franken and Leivens, 2022[1]).
In contrast, Australia, China, Portugal and the United Kingdom report shares of funding coming from private sources that are relatively larger compared to other countries. While this might reflect high out-of-pocket childcare costs paid by families, it is not necessarily the case. Figure C2.1 presents final sources of funds, which means that government transfers to households (e.g. childcare subsidies or voucher schemes) are included in private expenditure, because they are channelled to providers through families. As a result, the private share in some countries may partly include publicly supported financing mechanisms to make childcare more affordable for disadvantaged families. This is the case for Portugal, where relatively high childcare fees paid by families are offset by similarly high amounts of benefits and rebates (OECD, 2022[2]).
The emphasis here is on pre-primary education (ISCED 02), as data availability across OECD countries is better at this level; 21 OECD countries do not report finance data for early childhood educational development (ISCED 01). It is also worth noting that Figure C2.1 refers to expenditure per child, rather than per full-time equivalent student (which is the standard measure used for primary to tertiary education). This means that a child who spends eight hours a day in pre-primary education and one that spends four hours a day will both count as one child, whereas if they were in primary education, the first would count as one full-time equivalent, the second as 0.5.
Investing in children aged 3 to 5
Figure C2.2 compares countries based on their expenditure on educational institutions for children aged 3 to 5 as a percentage of GDP. Measuring spending on 3-5 year-olds allows comparisons to be made between countries regardless of where they draw the line between early childhood education and primary education. In addition, unlike measures that focus on a particular level of education, it is not affected by differences in the age composition of the target population. Age greatly influences expenditure spent per child – younger children require higher adult-child ratios, leading to higher personnel costs per child. But children start and complete different stages of early childhood education such as pre-primary at different ages in different countries, creating different population make-ups at each level (see Chapter B2). All other things being equal, spending per child in pre-primary education will be higher in countries where primary education starts relatively early and the average age of children in pre-primary education is relatively young. Comparing expenditure for a defined age group such as 3-5 year-olds addresses this and offers a clearer basis for cross-country comparison, as it limits the effect of national enrolment age policies.
The age group of 3-5 year-olds is also of particular interest, as it marks a point where national policies on compulsory education diverge. As of 2023, compulsory education starts after the age 5 in 22 out of the 38 OECD countries. For these countries, public spending on 3-5 year-olds as a percentage of GDP could naturally appear lower, not necessarily due to limited investment, but because enrolment rates may be lower among children who have not yet reached the official starting age of compulsory schooling.
Figure C2.2 indicates that total education expenditure on 3-5 year-olds averages 0.59% of GDP across OECD countries, with the highest shares in Iceland (1.05% of GDP), Norway (0.87%), Israel (0.85%) and Sweden (0.82%), all over one-third more than the OECD average (Table C2.1). In addition to reflecting these countries’ economic priorities, these high ratios of investment in ECE could also be partially shaped by the geographical distribution of their populations: operational costs (e.g. administration and capital goods such as the construction of ECE centres) increase as population densities fall, because fixed costs must be borne regardless of the number of children enrolled.
One important caveat is that, in common with the other data in this chapter, Figure C2.2 does not fully capture types of early childhood education that do not fit ISCED criteria. In countries where institutional settings without explicit educational components or home- or family-based arrangements are common, it will underestimate investment in the early years.
In per cent, direct expenditure on educational institutions
1. Data do not cover day care centres and integrated centres for early childhood education.
For data, see Table C2.1. For a link to download the data, see Tables and Notes section.
Over the past decade, public investment in early childhood education (ECE) has increased steadily across OECD countries. For instance, as shown in Figure C2.6 below, government expenditure in pre-primary education rose in most countries between 2015 and 2022 (in constant prices). This has often accompanied higher enrolment rates at this level (see Chapter B1). One potential driver of this has been the move in several countries in the last decade to lower the starting age of compulsory education. Box C2.1 explores changes in the patterns of ECE spending among countries that have recently lowered their starting age of compulsory education.
This box explores how lowering the starting age of compulsory education might have affected patterns of public expenditure. In the past decade, 12 OECD and partner countries have implemented such reforms: Belgium (2020), Bulgaria (2017 and 2021), Costa Rica (2018), Czechia (2017), Finland (2015), France (2019), Greece (2020), Hungary (2015), Lithuania (2016), Romania (2020), the Slovak Republic (2021) and Sweden (2018). These reforms lowered the starting age of compulsory education by one to three years, affecting pre-primary programmes (for more details see Table B2.1 in Education at a Glance 2024 (OECD, 2024[3])).
Figure C2.3 shows trends in government expenditure on pre-primary education (dark blue line). The red dotted line indicates the final year before the reform was implemented, to distinguish trends before and after the policy change. The figure also includes expenditure patterns for early childhood educational development and primary education to provide context. In most countries, spending at all three levels followed parallel trajectories before the reform, suggesting common drivers such as macroeconomic conditions or broader education funding trends. This helps strengthen the case for any post-reform changes in pre-primary spending having been driven by the reform itself.
How did spending change after the reform?
Pre-primary expenditure generally continued along pre-existing trends, with no sharp increase at the time the reform was implemented. This suggests that in many of these countries, spending adjustments were either not directly tied to the reform or masked by other events occurring around that time. Expanding ECE participation may have already been a priority, with the reform simply formalising ongoing efforts.
Expenditure on pre-primary education increased after the reforms in Bulgaria and Lithuania, while Czechia experienced a modest but clear increase in pre-primary spending in the year of the reform and this upward trend continued through 2021.
There was no clear link between trends in spending and the number of years by which the starting age of compulsory education was reduced – even in countries that implemented more substantial reforms. For example, Czechia and Hungary reduced the starting age by two years, and France by three, while all other countries made only a one-year change. This suggests that even far-reaching reforms do not necessarily translate into substantial or immediate budgetary shifts.
Countries with recent changes in the duration of compulsory pre-primary education, in billions USD in constant prices
Note: The red dotted line marks the final year before the reform extending compulsory ECE was implemented. Bulgaria introduced two reforms to the starting age of compulsory education, in 2017 and again in 2021. Pre-primary education is highlighted in dark blue; other levels are shown for context. Given the variation in expenditure levels across countries, direct cross-country comparisons should be made with caution.
For a link to download the data, see Tables and Notes section.
What could explain the observed trends?
In systems where enrolment at the affected ages was already nearly universal, reforms only had a marginal effect on enrolment. In some cases, reforms strategically targeted disadvantaged subgroups – such as socio-economically at-risk children in Lithuania (Eurydice, 2023[4]) or migrant and low-educated families in Belgian cities (European Commission, 2020[5]) – thus affecting a relatively small number of children and limiting the need for substantial new investment.
Free access or legal entitlements to an ECE place may have also played important roles in shaping expenditure growth in the years prior to or around the reform. In Belgium, Bulgaria, Hungary, Lithuania, Romania and Sweden, free early education was already available to children below the new compulsory age (for more details see Table X1.3 in Annex 1). When it comes to entitlements, in Czechia, for instance, from 2018 municipalities were required to guarantee subsidised places for all children over age 3, upon parental request (European Commission / EACEA / Eurydice, 2025[6]). Such measures can drive higher spending even without major enrolment increases, as governments must ensure adequate provision in advance of implementation. Much of the investment may also have already occurred, limiting the need for a sharp rise in public spending at the time of reform.
The COVID-19 pandemic and countries’ fiscal situation might have shaped implementation and spending patterns. Belgium, Bulgaria, Greece, Romania and the Slovak Republic implemented reforms in 2020-21. Belgium and Bulgaria recorded increases in spending while Greece saw a decline. This might have been driven by economic factors: Belgium and Bulgaria entered the pandemic with stronger fiscal capacity, allowing for additional investment in pre-primary education despite the crisis. In Bulgaria, for example, a 2022 reform abolished fees for public nurseries and kindergartens to promote access and equity (OECD, 2023[7]). In contrast, Greece’s ability to scale up provision may have been constrained by limited public investment capacity (OECD, 2020[8]).
The timing and design of budget processes also matter. In some countries, the allocation of expenditure to ECE is not directly tied to changes in enrolment policy, meaning that reforms to lower the starting age of compulsory education may not immediately influence overall government spending. Shifts in political regimes or ruling parties can also affect budget priorities and the pace of fiscal adjustments, influencing how quickly and to what extent spending responds to enrolment reforms.
How did enrolment patterns evolve in the light of the reform?
Figure C2.4 shows enrolment trends in pre-primary institutions by age, with each line representing a different age group affected by the reform.
In most countries, enrolment rates were high (between 80% and 90%) even before the reform, limiting the room for additional increases. Reforms required only marginal adjustments, if any, in capacity or staffing, rather than large-scale system expansion.
Costa Rica is the only country where the reform was followed by a sharp increase in enrolment rates: from 60% to 80% among 4-year-olds in the year of the reform, rising to 90% the following year, and reaching near universal coverage by 2023. Among 5-year-olds, enrolment rates were stable prior to the reform and increased from around 80% to 90% in the year of the reform. A comparable, though smaller, shift occurred in Czechia, where enrolment rates among 5-year-olds rose from around 90% to 95% in the year of the reform.
In Greece, enrolment rates among 4-year-olds had been rising strongly prior to the reform and continued to increase after it, going from around 65% to 95% in ten years. This suggests that families’ or local authorities’ anticipation of the reform or the implementation of other policy reforms were the main drivers of increasing enrolment.
Age groups affected by the extension of compulsory pre-primary education
Note: Each line represents a different age affected by the reform to compulsory education starting age. The red dotted line marks the final year before the reform extending compulsory ECE was implemented. Bulgaria introduced two reforms to the starting age of compulsory education: one in 2017 lowering it to age 5, and another in 2021 lowering it further to age 4.
For a link to download the data, see Tables and Notes section.
Overall, recent reforms to lower the starting age of compulsory education did not lead to major shifts in public expenditure or enrolment, but this does not imply they were ineffective. Rather, the limited fiscal and enrolment impacts in many countries reflect the fact that early childhood education was already a policy priority, with high participation rates and infrastructure largely in place prior to the reform. In such contexts, the reforms served to consolidate progress, extend legal guarantees or improve equity – particularly when targeted at disadvantaged groups. These findings suggest that the success of compulsory age reforms cannot be measured solely by short-term spending or enrolment changes, but must also consider broader policy goals, such as formalisation, inclusivity and long-term system development.
Pre-primary expenditure by type of institution
Figure C2.5 presents the distribution of total expenditure across public and private pre-primary institutions. It shows the extent to which funding is channelled into publicly governed or privately managed models (which may be heavily government funded, in the case of government-dependent private institutions). For most OECD countries, the breakdown of funding by type of institution mirrors very closely that of enrolment at pre-primary level (see Table B1.3). However, in Italy and Luxembourg, independent private institutions educate a relatively large share of children yet receive a smaller proportion of total spending. In Israel, government-dependent private institutions are underfunded relative to their enrolment share, while independent private providers receive double the share of funding compared to their share of students (Table C2.3).
The countries dedicating 10% of expenditure or more to independent private institutions are all those spending at or below the OECD average per child at pre-primary level. This may reflect a structural reliance on non-public provision for both enrolment availability and funding (Table C2.1).
The OECD average of the share of funding dedicated to public institutions is slightly higher for pre-primary education than for early childhood education and development. This is largely driven by differences in the distribution of enrolment patterns between public and private institutions across these two levels of education (see Table B1.3).
In per cent, direct expenditure on educational institutions
1. Year of reference differs from 2022.
For data, see Table C2.3. For a link to download the data, see Tables and Notes section.
Trends in pre-primary government expenditure per child
Figure C2.6 examines changes in government expenditure per child in pre-primary institutions between 2015 and 2022. A change in government expenditure on educational institutions per child can be driven by two factors: the change in total government expenditure (measured in constant 2020 prices) and the change in the number of children enrolled. In 25 countries with data available, government expenditure per child increased during this period. In 18 of these 25 countries, the increase was driven by a fall in the number of children enrolled. For instance, Japan’s government expenditure towards kindergartens and Kindergarten Departments of Special Needs Education Schools grew by nearly 40% despite the number of children enrolled in these programmes dropped by 30%. One possible reason for this expenditure growth is a 2019 reform that provides free early childhood education and care for children aged 3 to 5.
However, some countries show a different trajectory. Hungary, Sweden, the Republic of Türkiye and the United Kingdom experienced a decline in government expenditure per child between 2015 and 2022. In three of these – Hungary, Sweden and the United Kingdom – this decline was driven by a reduction in government spending (in constant prices). Italy also recorded a drop in public expenditure, although this was offset by falling enrolment, keeping spending per child stable overall (Figure C2.6).
In per cent, constant 2020 prices
Note: Expenditure per child is based on headcounts rather than full-time equivalent students.
1. Data do not cover day care centres and integrated centres for early childhood education.
2. Full-time equivalent is used in the calculation of expenditure per child.
For data, see Table C2.5 (available on line). For a link to download the data, see Tables and Notes section.
Educational institutions can be classified into two different categories: public and private. An institution is classified as private if its overall control and management rest with a non-governmental organisation (e.g. a church, trade union, business enterprise or foreign or international agency) and if most of the members of its governing board are not selected by a public agency. The terms “government-dependent” and “independent” are used to distinguish private institutions. A government-dependent private institution is a private institution that receives 50% or more of its core funding from government agencies, or one whose teaching personnel are paid by a government agency or by government directly. An independent private institution is a private institution that receives less than 50% of its core funding from government agencies and whose teaching personnel are not paid by a government agency.
For the definitions of direct government expenditure on educational institutions, and direct private expenditure on educational institutions, refer to Chapter C1.
Expenditure per child on educational institutions for early childhood education development and pre-primary levels is calculated by dividing total expenditure on educational institutions at that level by the corresponding sum of full-time and part-time enrolment, resulting in total expenditure on educational institutions per head count as opposed to per full-time equivalent student.
For an overview of the methodology, see Chapter C1. For more detailed information, please refer to the OECD Handbook for Internationally Comparative Education Statistics (OECD, 2018[9]). For country-specific notes, see Education at a Glance 2025 Sources, Methodologies and Technical Notes (https://doi.org/10.1787/fcfaf2d1-en).
For the data sources used in this Chapter, refer to Chapter C1. For additional details, see Education at a Glance 2025 Sources, Methodologies and Technical Notes (https://doi.org/10.1787/fcfaf2d1-en).
References
[5] European Commission (2020), Education and Training Monitor 2020 - Belgium, Publications Office of the European Union, Luxembourg, https://data.europa.eu/doi/10.2766/917974.
[6] European Commission / EACEA / Eurydice (2025), Key Data on Early Childhood Education and Care in Europe - 2025, Publications Office of the European Union, Luxembourg, https://eurydice.eacea.ec.europa.eu/publications/key-data-early-childhood-education-and-care-europe-2025.
[4] Eurydice (2023), Lithuania: Organisation of the education system and of its structure in Lithuania, European Commission, https://eurydice.eacea.ec.europa.eu/eurypedia/lithuania/organisation-education-system-and-its-structure.
[1] Franken, L. and J. Leivens (2022), “The end of the opt-out era in Belgian governmental schools?”, British Journal of Religious Education, Vol. 44/4, pp. 472-485, https://doi.org/10.1080/01416200.2021.1967110.
[3] OECD (2024), Education at a Glance 2024: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/c00cad36-en.
[7] OECD (2023), OECD Economic Surveys: Bulgaria 2023, OECD Publishing, Paris, https://doi.org/10.1787/5ca812a4-en.
[2] OECD (2022), “P.F.3.4: Childcare support”, in OECD Family Database, OECD, https://webfs.oecd.org/els-com/Family_Database/PF3-4-Childcare-support.pdf.
[8] OECD (2020), OECD Economic Surveys: Greece 2020, OECD Publishing, Paris, https://doi.org/10.1787/b04b25de-en.
[9] OECD (2018), OECD Handbook for Internationally Comparative Education Statistics 2018: Concepts, Standards, Definitions and Classifications, OECD Publishing, Paris, https://doi.org/10.1787/9789264304444-en.
To download the data for the figures and tables in this chapter, click StatLink above.
To access further data and/or other education indicators, please visit the OECD Data Explorer: https://data-explorer.oecd.org/.
Data cut-off for the print publication 13 June 2025. Please note that the Data Explorer contains the most recent data.
Table C2.1. Total expenditure and government expenditure on early childhood education per child as a percentage of GDP per capita and as a percentage of GDP (2022)
Note: Columns showing data on expenditure on educational institutions as a percentage of GDP are available for consultation on line.
1. Total expenditure on educational institutions includes payments by households outside educational institutions.
2. Data do not cover day care centres and integrated centres for early childhood education. Data include subsidies to households and transfers and payments to other non-educational private entities.
3. Year of reference 2021.
Table C2.2. Distribution of expenditure on early childhood educational institutions, by source of expenditure (2022)
1. Total expenditure on educational institutions includes payments of households outside educational institutions.
2. Data do not cover day care centres and integrated centres for early childhood education. Data include subsidies to households and transfers and payments to other non-educational private entities.
3. Year of reference 2021.
Table C2.3. Distribution of expenditure on early childhood education, by type of educational institution (2022)
1. Total expenditure on educational institutions includes payments by households outside educational institutions.
2. Data do not cover day care centres and integrated centres for early childhood education. Data include subsidies to households and transfers and payments to other non-educational private entities.
3. Year of reference 2021.
a – category not applicable; b – break in series; d – contains data from another column; m – missing data; x – contained in another column (indicated in brackets). For further control codes, see the Reader’s Guide.
For further methodological information, see Education at a Glance 2025: Sources, Methodologies and Technical Notes (https://doi.org/10.1787/fcfaf2d1-en).
Direct expenditure within institutions
Note: For notes on this table and a link to download the data, see Tables and Notes section above.
In per cent, expenditure within educational institutions, by level of education
Note: For notes on this table and a link to download the data, see Tables and Notes section above.
In per cent, direct expenditure within educational institutions, by level of education
Note: For notes on this table and a link to download the data, see Tables and Notes section above.