How are schools doing, and what do they cost?
Short answer: Year 9 pupils had an average of 229 grade points out of a possible 340 in 2025, about the same as ten years earlier, and 84 percent became eligible for a vocational programme in upper secondary school. Upper secondary schools have become better at getting pupils through: 71 percent graduated within three years in 2025, up from 65 percent in 2015. Three out of four compulsory school teachers are certified and qualified in their subject. A pupil in a municipal compulsory school cost 145,500 kronor in 2025, and 16 percent of compulsory school pupils attend an independent school.
University research, block grants and institutions are on Research and innovation.
Grades in year 9 and upper secondary diplomas.
Sweden
Qualified teachers, cost per pupil and independent schools.
Sweden
Which schools do better than their pupils' background predicts?
The Swedish National Agency for Education's SALSA model calculates the result a school would be expected to reach given its pupils' background: parents' education, the share of newly arrived pupils and the share of boys. The deviation is the difference between the actual and the expected result. A plus means the school does better than schools with similar pupils.
| School | Actual | Expected | Deviation |
|---|
All schools are on the map's school layer.
Where can you study for the job you want?
Type a programme, an occupation or a school. The search covers upper secondary programmes, adult vocational education, higher vocational education, labour market training, folk high schools and university programmes. If you pick a municipality at the top, its programmes come first, then the rest of the county.
| Education | Locality | Starts | Map |
|---|
The offerings come from the Swedish National Agency for Education (Susa-navet and upper secondary admissions), the occupations from the Swedish Public Employment Service occupational barometer. On the map every programme shows its provider, study pace, application deadline and job prospects.
Has student aid kept up?
In 2026, full student finance for full time studies is 2026 13 592 kronor per four weeks, of which 4,120 kronor is grant and 9,472 kronor loan. CSN pays out for 40 weeks a year, roughly ten of the year's twelve months. Rent and food cost money in summer too, so spread over the whole year it comes to 11,327 kronor a month.
Student finance against a standard budget
| Post office | Kronor per month | Source |
|---|
Food assumes all meals are cooked at home. The Consumer Agency's items under May be added total 3,180 kronor a month 2026; electricity and water are given in kronor only since 2022. Each year uses that year's student aid, Consumer Agency amounts and rents. The Consumer Agency's age group and items have changed over the years: the mobile phone moved 2021 to household media and the food calculation was redone 2026. Where the Swedish National Union of Students has no rent for a year, Statistics Sweden's median rent for a newly let one-room flat in the municipality is used. The rent for a student flat is an average or a lowest price from the Swedish National Union of Students' housing report. The sublet one-room flat is the average asking rent on Qasa in June 2026, compiled by Zmarta. The student card is the full price for 30 days, in Skåne calculated as Skånetrafiken's student discount of 25 percent on the price for the small zone and large city; from 1 July to early December 2026 the price is halved with a state grant. Over the year, income is 135,920 kronor spread over twelve months, rent for the months it is paid, course literature for ten months and the rest for twelve.
Surplus or shortfall year by year
The same budget for the chosen city and housing every year, with that year's student aid, standard amounts and rent. Above zero the student aid covers the budget, below zero it does not.
Left in a study month Left per month over the year
Over time
Student aid from CSN consists of a grant and a loan. The amounts are recalculated every year from the price base amount, so in fixed prices they barely move for long stretches. The big increases came with reforms: in 1989 the grant was raised sharply, in 2001 today's system arrived, in 2015 the loan was raised and in 2018 the grant. In 2023 and 2024 the amounts rose with the high inflation, because the price base amount did.
The income limit is how much you may earn in a calendar half year before your student aid is reduced. Since 2014 it has been 193.71 percent of the price base amount for full time studies, and for every krona above the limit the aid drops by 61 öre. During the pandemic the limit was removed entirely from January 2020 to June 2022, so that students could work, in healthcare among other places, without losing their student aid.
The interest rate, the debt and repayment.
The government sets the interest rate on student loans ahead of each year. It is based on the average government borrowing rate over the last three years, reduced by 30 percent because the interest is not tax deductible. When the Riksbank had a negative policy rate from 2015 to 2019, the government borrowing rate fell towards zero, and the student loan rate reached 0 percent in 2022. When rates rose with inflation in 2022 and 2023, the student loan rate followed. In 2023 the Riksdag decided on a surcharge to cover loans that are never repaid, so since 2024 the rate is a base rate plus a surcharge.
The interest goes to the state. In 2025 borrowers paid 5.8 billion kronor in interest on student loans, against 3.4 billion in 2024. CSN itself borrows the money from the Swedish National Debt Office and paid it 6.6 billion kronor in interest, against 3.2 billion the year before. The borrowers' rate is about 70 percent of the Debt Office's because it is not tax deductible, and the state covers the rest with appropriations: 1.7 billion kronor in 2025. The figures come from CSN's annual report for 2025, which only reports two years at a time.
Loans taken after 1 July 2001 are annuity loans: you repay equal amounts every year, starting six months after you last received student aid. Since 2022 the loan must be repaid within 25 years or by the year you turn 64, whichever comes first. Student aid can be granted up to and including the year you turn 60, but the number of weeks with a loan shrinks from age 41. Loans from 1989 to June 2001 are repaid at 4 percent of income, and older loans under rules of their own.
Independent schools, grades and falling results.
In the 1990s the Swedish school system changed more than its neighbours'. The municipalities took over responsibility in 1991, and the school voucher of 1992 let independent schools open with public money. No other Nordic country has gone as far: in Norway independent schools may not make a profit, and in Finland there are few of them. How things went when the state sold or opened other services to private providers is on Privatisation.
Results have fallen across the Nordic countries. In PISA 2025, Swedish 15-year-olds scored 466 points in reading and 464 in mathematics, the lowest results Sweden has had and 21 and 18 points lower than in 2022. Every Nordic country scored lower in reading and mathematics than last time. Finland is still ahead of Sweden in reading and science, while Norway and Iceland are behind in all three subjects.
Research gives the reforms a smaller role than the debate often does. The decline seems to have started before the reforms could affect pupils, and school choice has had small positive effects on the average. On the other hand, schools have become more different from each other, and schools grade with different degrees of generosity.
PISA 2025 in the Nordic countries, mean scores
| Country | Reading | Mathematics | Natural science |
|---|---|---|---|
| Sweden | 466 points | 464 points | 485 points |
| Finland | 474 points | 469 points | 504 points |
| Denmark | 460 points | 471 points | 478 points |
| Norway | 453 points | 452 points | 470 points |
| Iceland | 422 points | 450 points | 441 points |
| OECD average | 461 points | 463 points | 482 points |
Source: OECD, PISA 2025 Results (Volume I), table I.1. Differences of a few points between countries are often not statistically significant. In Norway 10.4 percent of students were excluded from the test, more than PISA allows, so the Norwegian results are less certain and may be too high.
-
Do pupils learn more in independent schools?
Municipalities where independent schools grew got slightly better results in year 9 and longer education later, but the gain only showed about ten years after the reform. It comes mainly from competition between schools, not from independent school pupils themselves learning more. In upper secondary school, independent school pupils instead scored slightly lower on tests marked outside the school, especially the weaker pupils.
Böhlmark and Lindahl (2015), Economica. Tyrefors Hinnerich and Vlachos (2017), Labour Economics.
-
Are grades too generous?
When the same national tests were marked by someone outside the school, the scores were lower, and the difference was larger in free schools: in upper secondary they graded their own pupils about 0.14 standard deviations more generously than municipal schools. Grades from different schools are therefore not fully comparable.
Tyrefors Hinnerich and Vlachos (2017), Labour Economics.
-
Did the 90s reforms cause the decline?
Probably not. The decline in both Swedish and international assessments seems to have started before the reforms could reasonably affect pupils. The study finds no support for the idea that municipalisation drove the decline, and school choice has had small positive effects on results on average.
Holmlund et al. (2014), IFAU report 2014:25.
-
Have schools become more different?
Yes. Differences in results between schools almost doubled between 2000 and 2016. Mostly this is because pupils with different backgrounds increasingly attend different schools, which both residential segregation and school choice contribute to. A larger share of free schools also goes with more sorting of pupils. The importance of family background for the merit rating has stayed roughly flat in recent years, but in most individual subjects it has increased, most of all in mathematics.
Skolverket (2018), report 467. Skolverket (2025), Equity at the end of compulsory school. Holmlund et al. (2014), IFAU. Residential segregation per municipality is on Segregation.
-
Does class size matter?
Yes, and for a long time. Pupils who ended up in smaller classes in years 4 to 6 had better results at ages 13, 16 and 18, stayed longer in school and earned more between ages 27 and 42. The study uses the rule of at most 30 pupils per class that applied before 1991.
Fredriksson, Öckert and Oosterbeek (2013), Quarterly Journal of Economics.
-
What does student aid do for students?
The student aid reform of 2001 raised both the grant share and the income ceiling. Students from homes without a tradition of higher education then started working more on the side and studied at a slower pace, while study pace only rose for students with an academic background. A higher income ceiling can thus widen gaps instead of narrowing them.
Avdic and Gartell (2015), Labour Economics.
Operating profit and dividends in the independent school groups.
Independent schools receive a per-pupil grant from the pupil's home municipality, and those run as limited companies may make a profit. This shows operating profit for four of the large groups, AcadeMedia, Engelska skolan, Kunskapsskolan and Thorengruppen, and for the whole industry in Statistics Sweden's business statistics. Operating profit is what remains before interest and tax. It can be paid to the owners, used to buy back shares or as group contributions, or stay in the company. The Swedish National Agency for Education publishes no statistics on profits, and dividends for the industry as a whole are not published.
Where the figures come from.
- Source: Kolada. Grade points, eligibility, upper secondary diplomas, teacher qualifications and cost per pupil per municipality, originally from the Swedish National Agency for Education.
- Older national figures for years Kolada lacks: the Swedish National Agency for Education's Descriptive Data reports and memos on final grades (merit rating, upper secondary leaving certificates, upper secondary teachers' degrees, cost 2000 to 2004) and SCB (cost per pupil 2007 to 2014). Every splice is marked in the chart.
- Our own processing of Kolada's data: the share of pupils in independent compulsory schools, calculated as pupils in independent schools in the municipality divided by all compulsory school pupils in the municipality.
- Swedish National Agency for Education, SALSA (CC0): actual and expected results per compulsory school, from the agency's open API.
- CSN (CC0): student aid amounts per half year from 1965, the interest rate from 1989, the income limit from 2001 and total student debt from 2009, from CSN's open data and statistical database. Rules from the Student Aid Act (1999:1395). Rent per square metre: SCB (CC0), average for rental flats in Sweden. SCB does not report student housing separately.
- Independent school profits: annual reports from AcadeMedia (2016/17 to 2024/25, and the year-end report for 2025/26) and Internationella Engelska Skolan (to 2019/20), and for later years and the other companies the annual reports filed with the Swedish Companies Registration Office, read here via allabolag.se. The industry: Statistics Sweden, Structural business statistics (CC0), SNI 85.2 and 85.3.
- Not included: the Swedish Schools Inspectorate's school surveys and third party compilations, since we have not been able to establish the terms for republishing them.
The year is the year the school year ends: 2025 is the school year 2024/25. Results and qualifications are counted by the pupil's home municipality, teachers and costs by the municipality where the school is. Costs are in current prices without adjustment for inflation. Each municipality has more on its municipality page, and the schools are a layer on the map.