Countries Spending the Most on Education: Data Ranked

Education spending is often treated as the clearest signal of how seriously a country takes its future. The reality is messier. The country at the top of the World Bank's ranking spends 16.4% of its GDP on education, yet its economy is so small that the sum works out to only a few hundred dollars per resident. Meanwhile, several of the best-performing school systems in the world spend modest shares of GDP.

This article ranks countries by government education spending as a share of GDP, using World Bank data, and then does what a raw ranking cannot: it puts dollars next to percentages, compares spending with PISA mathematics scores, and explains why more money does not map neatly onto better results. Every figure comes from a source anyone can check.

Key takeaways

• On World Bank data, Kiribati spends the highest share of GDP on education among countries with recent figures, at 16.4% (2023), followed by Tuvalu (12.9%) and the Federated States of Micronesia (11.6%, 2020)
• Applied to 2025 GDP per capita, Kiribati's share equals roughly US$420 per resident, against more than US$4,000 per resident in Sweden, Norway, and the United States
• Among rich countries, Sweden (7.3%) and Iceland (7.3%) spend the most as a share of GDP, while Japan (3.3%), Ireland (2.9%), and Singapore (2.2%) spend the least
• PISA 2022 mathematics scores show no consistent link with spending: Singapore scored 575 on 2.2% of GDP, while Norway scored 468 on 5.4%
• The United States spends more per student than the OECD average, according to the National Center for Education Statistics (NCES), but its PISA 2022 mathematics score (465) was below the OECD average of 472

How education spending is measured

A single "education spending" number does not exist. Three different measures are in common use, and they rank countries differently.

Government expenditure as a share of GDP is the measure used in this ranking. It counts what national and local governments spend on education and divides it by the size of the economy. The World Bank publishes it as indicator SE.XPD.TOTL.GD.ZS, drawing on data from UNESCO's Institute for Statistics. It is the broadest cross-country series available, with data for most economies, though the most recent year differs by country.

Spending from all sources adds the money households and private institutions pay, such as tuition fees. The OECD publishes this for its members and partners. In 2019 the United States spent 6.0% of GDP on education from all sources, against an OECD average of 4.9%, according to NCES. Norway was highest at 6.6%, followed by Chile (6.5%), Israel (6.2%), Australia (6.1%), and the United Kingdom (6.0%). The lowest were Ireland (3.2%), Luxembourg (3.3%), and Lithuania (3.5%) (NCES: Education expenditures by country).

Spending per student divides total spending by the number of pupils. It is the number closest to what a child's school actually has to work with. NCES reports that in 2019 the United States spent US$15,500 per full-time-equivalent student at the elementary and secondary level, against an OECD average of US$11,300, and US$37,400 at the postsecondary level, against an OECD average of US$18,400.

Each measure answers a different question. Share of GDP asks how much of the national economy goes to schools. Spending per student asks how much each child receives. All-sources spending asks how much society, public and private, invests overall. A country can rank high on one and low on another, which is the central reason this ranking should be read with care.

The top 20: government spending on education, % of GDP

The table lists the 20 highest values in the World Bank series, each using the country's most recent available year. Years differ, so the figures are not a snapshot of a single year. American Samoa is a territory rather than a sovereign state, and Venezuela's figure is more than 15 years old; both are shown because they appear in the data, but neither should be read as a current measure.

Rank	Country	% of GDP	Year of data
1	Kiribati	16.39	2023
2	American Samoa (territory)	14.72	2006
3	Tuvalu	12.85	2023
4	Micronesia, Fed. Sts.	11.56	2020
5	Namibia	9.08	2024
6	Algeria	8.98	2024
7	Cuba	8.44	2022
8	Solomon Islands	8.29	2023
9	Botswana	8.06	2020
10	Marshall Islands	7.70	2022
11	Vanuatu	7.64	2024
12	Mauritania	7.61	2024
13	Bolivia	7.54	2024
14	Sweden	7.32	2022
15	Iceland	7.31	2022
16	Venezuela	6.87	2009
17	Kyrgyz Republic	6.83	2023
18	Tunisia	6.73	2023
19	Lesotho	6.62	2024
20	Moldova	6.56	2023

Source: World Bank, Government expenditure on education, total (% of GDP), indicator SE.XPD.TOTL.GD.ZS, most recent non-empty value per country, retrieved October 2026. See Sources.

Three things stand out. Only two countries in the top 20, Sweden and Iceland, are wealthy. Pacific island states, counting American Samoa, fill six of the top ten places. And the list contains almost no countries with outstanding school results, which is an early hint that share of GDP is a poor proxy for educational quality.

Several countries often assumed to be top spenders do not appear. Norway, for example, is at 5.43% in the World Bank series for 2022. Finland is at 6.38%, Denmark at 6.36%, and Belgium at 6.28%. All four are wealthy countries with strong reputations for schooling, but they sit below the top 20.

Why small Pacific economies top the list

Kiribati's 16.39% is the highest in the table, and the reasons are as much about arithmetic and geography as about priorities. Kiribati comprises 32 atolls and other islands plus one raised coral island, Banaba. Its 811 square kilometers of land are spread across more than 3.4 million square kilometers of ocean (Wikipedia: Kiribati). Its population was 136,488 in 2025 by the World Bank's count.

Delivering education across scattered islands is expensive per student. A school on a small atoll needs a building, a teacher, supplies, and transport for all of them, whether it serves 30 children or 300. These fixed costs do not shrink with the roll, so per-pupil costs rise as communities get smaller and farther apart.

Percent of a small number

The second reason is that GDP itself is tiny. Kiribati's GDP per capita was about US$2,559 in 2025, compared with US$94,594 in Norway and US$90,027 in the United States. A very high percentage of a very small economy still produces a modest sum.

The table below applies each country's education-spending share to its 2025 GDP per capita. Because the spending years (2021 to 2023) differ from the GDP year, the results are rough illustrations rather than official figures, but the ordering is clear.

Country	Spending, % of GDP	GDP per capita, 2025 (US$)	Approximate education spending per resident (US$)
Kiribati	16.39 (2023)	2,559	419
Tuvalu	12.85 (2023)	6,041	776
Marshall Islands	7.70 (2022)	8,489	654
Sweden	7.32 (2022)	63,133	4,621
United States	5.42 (2021)	90,027	4,880
Norway	5.43 (2022)	94,594	5,137
Iceland	7.31 (2022)	98,323	7,187

Source: World Bank indicators SE.XPD.TOTL.GD.ZS and NY.GDP.PCAP.CD; products calculated by SciHub101. Per-resident figures are per person in the whole population, not per student.

Iceland spends roughly 17 times as many dollars per resident as Kiribati, even though Kiribati devotes more than twice the share of its economy. That gap is the reason ranking by percent alone tells you about commitment and constraint, but not about what a classroom actually receives. For the other side of this equation, see our ranking of countries with the highest GDP per capita, and for a primer on the denominator, What Is GDP and Why It Matters.

Kiribati also relies heavily on fishing license fees and international aid, with Australia, New Zealand, and Japan providing most foreign aid (Wikipedia: Kiribati). Some education costs, such as secondary or tertiary study abroad, may be financed from outside the government budget, and the World Bank's government expenditure series would not capture that.

Beyond the islands: priority and structure

Namibia (9.08%), Algeria (8.98%), Botswana (8.06%), Mauritania (7.61%), Bolivia (7.54%), and Cuba (8.44%) are not small islands. Their placement suggests that high shares can reflect policy choices, small or resource-dependent economies, or high costs of reaching a young and dispersed population. The table cannot tell these explanations apart.

Cuba is the most discussed case. Its 8.44% in 2022 reflects a state-run system in which education is a core public function. Whether such spending achieves strong outcomes cannot be settled from this table. The PISA comparison later in this article covers only countries for which we could confirm 2022 scores, and Cuba is not among them, so no claim is made about its results.

Data quality is a further caution. Spending figures for low- and middle-income countries are often reported with delays and sometimes revised. Some entries here date from 2020, and one from 2006. The fact that Somalia appears at essentially zero in the same dataset, which is probably a reporting artifact, is a reminder that the series is only as good as the national reports behind it.

Wealthy countries: spending versus PISA results

For rich countries, the question people usually care about is whether spending buys better results. The PISA assessment, run by the OECD every three years among 15-year-olds, offers one yardstick. The table compares each country's government spending share with its PISA 2022 mathematics score. The OECD average for mathematics in 2022 was 472.

Country	Govt spending, % of GDP	Year	PISA 2022 mathematics
Singapore	2.19	2024	575
Japan	3.34	2021	536
South Korea	5.41	2022	527
Estonia	5.21	2022	510
Switzerland	4.86	2022	508
Canada	4.84	2022	497
Netherlands	5.18	2022	493
Denmark	6.36	2022	489
Belgium	6.28	2022	489
United Kingdom	5.91	2021	489
Austria	5.28	2022	487
Australia	5.06	2022	487
Finland	6.38	2022	484
Sweden	7.32	2022	482
New Zealand	5.21	2023	479
Germany	5.24	2022	475
France	5.32	2022	474
Norway	5.43	2022	468
United States	5.42	2021	465
Iceland	7.31	2022	459
Israel	5.93	2022	458

Sources: World Bank indicator SE.XPD.TOTL.GD.ZS for spending; PISA 2022 scores as compiled on Wikipedia: Programme for International Student Assessment, which cites the OECD. Values checked against that table in October 2026.

The table was sorted by test score, and it shows no tidy relationship. Of the six highest-scoring countries, the highest spender, South Korea, is at 5.4%, while the top scorer is at 2.2%. Sweden and Iceland, the highest spenders in the table, are below the OECD average in mathematics in the case of Iceland and just above it for Sweden. Norway, Iceland, and the United States all score below the 472 average with spending shares between 5.4% and 7.3%.

Why Singapore and Japan need a caveat

Singapore's 2.19% and Japan's 3.34% stand out because they sit alongside strong scores. These are real data points, but two caveats apply. Government spending alone understates what societies in these countries invest in education, because families also spend money on schooling, including tutoring and fees, and that spending is excluded from this series. And GDP in a high-income city-state differs from GDP in a larger economy in ways that distort the ratio.

A fairer conclusion is that among countries that already fund schools at a basic level of adequacy, additional spending shows a weak relationship with test results, and what the money is spent on matters more than the amount.

The United States: high spending per student, middling results

The United States is a frequent subject of this debate because it spends a lot per student. NCES reported US$18,614 of spending per pupil in public elementary and secondary schools in 2020-21, in constant 2022-23 dollars, of which US$16,280 was current expenditure, US$1,833 capital outlay, and US$501 interest on school debt. Total spending was about US$927 billion. Real current spending per pupil rose about 13% between 2010-11 and 2020-21 (NCES: Fast Facts).

On the OECD's 2019 data the United States spends 3.5% of GDP on elementary and secondary education, against an OECD average of 3.4%, so its share of GDP is close to the average. Its per-student spending is well above average because GDP per person is so high. At the postsecondary level the gap is much larger: 2.5% of GDP versus the OECD's 1.5%, and US$37,400 per student against US$18,400 (NCES).

The World Bank's government series gives the United States 5.42% for 2021, which can look low next to some peers, but on the OECD's all-sources basis the figure is 6.0%, above the OECD average of 4.9%. Which number is "right" depends on the question being asked. The United States is neither a low spender nor an extreme one.

What the average hides

A national average per-pupil figure says nothing about how money is divided across districts, schools, or student groups. The PISA 2022 mathematics score for the United States was 465, below the OECD average of 472, but its reading score of 504 and science score of 499 were above the corresponding OECD averages of 476 and 485. National averages blend very different school experiences.

The national figures above cannot show how large funding gaps are between American districts, so none is quoted here. They do support a more limited point: high average spending, by itself, does not guarantee high average scores.

Finland: what the data do and do not show

Finland is the most-studied example in discussions of education spending, and it is easy to tell the story wrongly. Finland's government spends 6.38% of GDP on education, mid-table among rich countries, and its PISA 2022 mathematics score of 484 sits modestly above the OECD average of 472.

Finland's reputation comes from earlier results. In the 2003 PISA study it ranked first in science and reading and second in mathematics. By 2012 it had fallen to sixth in reading, twelfth in mathematics, and fifth in science, and it has since been displaced from the top (Wikipedia: Education in Finland). Any explanation of Finland's success therefore needs to account for a decline, not only a peak.

Features of the Finnish system

Several features of the system are documented:

• Teacher qualifications. Both primary and secondary teachers must hold a master's degree, and entry to teacher training is highly competitive.
• Late start. Compulsory schooling begins at age 7 and covers the nine-year comprehensive school. Since September 2021, upper secondary education has also been compulsory.
• Mostly public schools. Schools up to university level are almost entirely funded and run by municipalities. Private schools are few. They receive a state grant comparable to a municipal school's, cannot charge tuition, and cannot select students.
• Light testing. The comprehensive school has no high-stakes tests, and grading in the early years starts with verbal assessments. Upper secondary students take a nationally graded matriculation examination.

Finland also reduced education funding by a total of €1.5 billion over 2011 to 2018, according to the same source, a period that overlaps with its slide in international rankings. The data do not prove that cuts caused the decline, since many things changed at once, but they show that Finland's results did not rise in step with spending.

The analogy that fits is a garden. Money buys good soil, seeds, and water, but the harvest also depends on skilled gardeners and on the climate. Finland's gardeners, its well-qualified teachers, and its policy of keeping the whole system within the public sector appear to matter more to its reputation than the budget line does. The link between how teachers treat students and how students perform is explored in The Oak School Pygmalion Experiment.

What money does buy, and what it does not

If spending shares tell us little about test scores, what do they tell us? A few things are reasonable to say from the data in this article.

Spending buys access and adequacy at the low end. Where schools lack buildings, textbooks, trained teachers, or electricity, additional money tends to matter a great deal. The top of the ranking is full of countries where that is a live issue. Our ranking is not designed to measure this, but the logic is straightforward.

Beyond adequacy, composition matters more than totals. A budget can go to teacher pay, smaller classes, early childhood programs, buildings, administration, or tertiary institutions. The OECD data cited by NCES show how different the mix can be: the United States spends relatively more at the postsecondary level than at the school level compared with the OECD average. A country's PISA results at age 15 are unlikely to respond to a university funding increase.

Teachers are the largest cost and probably the largest lever. Salaries are the biggest item in most education budgets. How teachers are recruited, trained, and supported varies sharply between countries, which is why direct comparisons of totals mislead. The science of how people actually learn, summarized in How Learning Actually Works, points to practice, feedback, and spaced review, not to expenditure as such.

Equity may matter as much as averages. An economy can spend heavily and still leave large groups behind if funds do not reach them. For related inequality patterns, see Countries With the Highest Income Inequality.

How to read an education spending ranking

A few practical cautions apply to this ranking and any similar one.

Check the year. The World Bank series mixes data from 2006 to 2024. Treat any figure older than five years as historical.

Check the definition. Government spending, spending from all sources, and spending per student are different measures. A country can appear in one list and be absent from another.

Remember the denominator. A share of GDP is only as meaningful as the GDP beneath it. Small, aid-dependent economies and high-income city-states both produce misleading ratios.

Separate spending from outcomes. PISA measures one age group, 15-year-olds, in a handful of subjects, and scores are not available here for most countries in the top-20 spending list. Test results measure something real, but not everything that schools do.

Look at related rankings. Spending is only one input. For results, see Countries With the Best Education Systems. For the university end, see Countries With the Most Top Universities and Countries with the Most International Students. For classroom tools, see AI Tools for Teachers.

Frequently asked questions

Q: Which country spends the most on education as a percentage of GDP? A: On World Bank data, Kiribati, at 16.39% of GDP in 2023. Tuvalu (12.85%, 2023) and the Federated States of Micronesia (11.56%, 2020) follow. American Samoa appears at 14.72% but with a 2006 figure and as a territory rather than a country.

Q: Which wealthy country spends the most on education? A: Among high-income countries in the World Bank series, Sweden (7.32% in 2022) and Iceland (7.31% in 2022) are the highest, followed by Finland (6.38%), Denmark (6.36%), and Belgium (6.28%). On the OECD's all-sources measure for 2019, Norway was highest at 6.6%, according to NCES.

Q: Does more education spending lead to better results? A: Not reliably among wealthy countries. In the PISA 2022 table above, Singapore scored 575 in mathematics while spending 2.19% of GDP on government education, and Iceland scored 459 while spending 7.31%. In poorer countries where basic needs are unmet, extra spending is more likely to matter.

Q: Why does Kiribati spend so much of its GDP on education? A: Part of the reason is geography: scattered atolls make schooling costly per student. Part is arithmetic: its GDP per capita was only about US$2,559 in 2025, so even 16.39% of GDP amounts to roughly US$420 per resident when applied to that figure, far less than in rich countries.

Q: How does the United States compare? A: NCES reports that in 2019 the United States spent 6.0% of GDP on education from all sources, above the OECD average of 4.9%, and US$15,500 per elementary and secondary student, against an OECD average of US$11,300. Its PISA 2022 mathematics score of 465 was below the OECD average of 472.

Q: Why did Finland fall in the PISA rankings? A: No single cause has been established. Finland ranked first in science and reading in 2003 and had fallen to sixth in reading and twelfth in mathematics by 2012. It also cut education funding by €1.5 billion between 2011 and 2018, but researchers do not agree on how much those cuts explain.

The bottom line

Education spending as a share of GDP is a meaningful but incomplete measure. The countries at the top of the ranking are mostly small, poor, or both, and a high percentage of a small economy can still be a modest sum per student. Among rich countries, the relationship between spending and PISA scores is weak, and some of the strongest performers spend less than the OECD average. The more useful questions are what the money buys, who it reaches, and how teachers are trained and supported. A ranking can tell you where a country's priorities lie. It cannot tell you whether a child in that country is learning.

Sources

• World Bank. Government expenditure on education, total (% of GDP), SE.XPD.TOTL.GD.ZS: selected countries and all countries.
• World Bank. GDP per capita (current US$), NY.GDP.PCAP.CD and population, SP.POP.TOTL.
• National Center for Education Statistics. Education expenditures by country and Fast Facts: Expenditures.
• Wikipedia. Programme for International Student Assessment (PISA 2022 scores, citing the OECD).
• Wikipedia. Education in Finland and Kiribati.

Related Reading

• Countries Spending the Most on Military R&D (2025)
• Countries Spending the Most on Healthcare (2025)
• Countries With the Most Top Universities (2025)
• Countries With the Best Education Systems (2025)
• Countries With the Most Tanks in the World (2025)