ENCS
Tuesday, 11 August 2026
Your paper, your pace.

Teachers: what has actually been tried, and who is left in front of the class

Across the OECD, 7.1% of secondary teachers did not meet their own country's qualification standard in 2022/23. Four countries tried to change that number. One of them moved the money from training into the third and fifth year of teaching, and lost 275 maths teachers doing it.

Checked on

All ten problems

Across the OECD in 2022/23, 7.1% of secondary teachers did not meet the training and certification standard their own country sets for the grade and subject they teach. That is the OECD’s Table D8.2, published in Education at a Glance 2025, and it is the number this page is argued against. It runs from 0.0% in Hungary, Japan and Korea to 21.6% in Sweden. Two things about it before we use it. It is defined by each country’s own rules, so a government can improve it by changing the rules. And it says nothing about the posts nobody filled at all: Austria had 6.0% unqualified teachers and 5.1% of secondary posts standing empty at the start of the year.

Czechia is not in that table, so here is the national equivalent, measured differently and not directly comparable: 10% of teaching in basic schools and almost 7% in upper secondary schools was done by unqualified teachers in 2024, up from under 5% in 2016. In the Karlovy Vary region the basic-school figure is 22%. That gap between 10 and 22 is the whole point of this page. National averages are where shortages go to hide.

The pressure is mostly demographic and it is easy to check. Over a third of OECD teachers were 50 or older in 2023, and only 9% of secondary teachers were under 30. In Lithuania the share of older secondary teachers went from 44% to 59% in a decade, in Portugal from 31% to 56%. Across 19 countries with data, 6.5% of qualified teachers left the profession in 2022/23, from under 3% in France, Greece, Ireland and Israel to over 10% in Denmark, Estonia and Lithuania. Below are four governments that did something and then measured — including one that moved the money from the start of a teaching career to the middle of it.

How this is measured

One number, named and sourced. Everything on this page is argued against it, so you can check us.

Share of secondary school teachers who are not fully qualified — who do not meet all the training and certification requirements their own country sets for the grade and subject they teach %

OECD, Education at a Glance 2025, Chapter D8 and Table D8.2, reference year 2022/23; published September 2025

Who is actually ahead

Ranked on the number above — not on reputation.

Share of secondary school teachers who are not fully qualified — who do not meet all the training and certification requirements their own country sets for the grade and subject they teach
Country Result Why
Japan 0.0 % None at all, and 0.2% of secondary posts unfilled at the start of the year. But the education ministry's own first shortage survey counted 2,558 unfilled teaching posts nationwide on the first day of the 2021 school year.
Hungary 0.0 % Also zero: Hungarian rules do not allow a teacher without the qualification. The pressure appears in the age profile instead — 49% of Hungarian secondary teachers were 50 or older in 2023, and 5% were under 30.
Poland 0.6 % One of the lowest measured anywhere. Poland's secondary teachers aged 50 or over went from 26% in 2013 to 40% in 2023, and in primary schools from 23% to 44% — the renewal problem is ahead of it, not behind.
United Kingdom 2.8 % England only. Down from 3.7% in 2014/15, with 0.4% of secondary posts unfilled at the start of the year. Recruitment is the weak point: secondary teacher training ran 11% below target for 2025/26.
Netherlands 3.9 % Down from 6.2% in 2014/15. But 2.3% of Dutch secondary posts were unfilled at the start of 2022/23, so the shortage shows up as an empty chair rather than an unqualified teacher in it.
Austria 6.0 % Nearly double the 3.1% of 2014/15, and 5.1% of secondary posts unfilled at the start of the year — second only to Sweden's 5.3%; the OECD reports these as the only two countries above 3%. Of Austrian secondary teachers who resigned in 2022/23, 63% had taught less than five years.
France 9.1 % For 2021/22, up from 7.7%. Against 1.3% in French primary schools: this is a secondary-school shortage, and the national average of the two would hide it.
Sweden 21.6 % The highest in the table, though down from 25.5% in 2014/15. The OECD notes Sweden's definition is the strictest of any country here — it also counts teachers working outside the subject or grade their certificate covers.

What has actually been tried

Real countries, real policies, and what the numbers did afterwards — including where it went wrong.

Czechia

Czechia has legislated its way at this twice. The 2004 Act on Pedagogical Staff obliged teachers without the qualification to obtain it. Then in 2017 the country changed school funding from money per pupil to money per taught hour, effective 2020, and schools prepared early by splitting classes and adding optional lessons. From 1 September 2023 an amendment allowed a holder of a relevant master's degree to teach for up to three years while completing the pedagogical study.

What happened. Unqualified teaching fell from more than 15% at the start of the 2000s to under 5% by 2016, then climbed back to 10% in basic schools and almost 7% in upper secondary schools by 2024. The funding change added more than 4,000 teaching posts a year after 2018: full-time-equivalent teachers passed 150,000 in 2024, and pupils per teacher fell from 12.9 in 2018 to 12.0 in 2023.

The catch. Demand was created faster than universities could supply it. Teacher-training graduates ran near 4,000 a year historically, fell to about 2,500 after 2010 and are back near 3,000 — over 2010 to 2024 that is roughly 15,000 fewer than replacing retirements alone required. And the shortage is not national: 22% of teaching in basic schools in the Karlovy Vary region is unqualified, about double the country as a whole. The ministry's survey published on 8 January 2026 put the largest gaps in physics (1,463 posts at lower secondary), computing (1,413) and maths (1,220).

Koucký and colleagues, Faculty of Education of Charles University, 'Českému školství hrozí nedostatek učitelů', TAČR project for the Ministry of Education, February 2025; Czech Ministry of Education, extraordinary survey on the staffing of teaching, published 8 January 2026

United Kingdom

England tried moving the money later in the career. From 2018 the Department for Education piloted a phased maths bursary: a smaller grant during training, with the rest paid out in the third and fifth years of teaching. The upfront bursary fell from £25,000–£27,500 to £20,000–£22,000, with £5,000–£7,500 paid later. A parallel scheme paid retention money to early-career maths and physics teachers.

What happened. The evaluation by CFE Research and FFT Education Datalab found that recruitment onto maths teacher training fell by 10 to 15 per cent — an estimated 275 fewer maths teachers entering state schools in 2019/20 — while the deferred payments retained an estimated 94. The pilot lost about three teachers for every one it kept. NFER's 2025 review of five such schemes put their average effect at 5.1% lower leaving rates, not significant at the 5% level (p = 0.06), costing roughly £43,000 per additional teacher-year against £9,000–£13,000 for a training bursary.

The catch. The two findings do not cancel out, and this is the interesting part: the retention payments did work on retention. Eligibility for the maths phased bursary is associated with a 10.9% lower leaving rate, and that estimate is statistically significant. What broke was recruitment. NFER also warns that results here are sensitive to how eligibility is defined — its own estimate for the maths and physics retention payment is lower than an earlier study's, from the same data.

Worth and McLean, 'The impact of early career retention payments on teacher retention', NFER, June 2025; CFE Research and FFT Education Datalab evaluation, 2023, reported by Schools Week

Germany

Berlin stopped making teachers civil servants in 2004 to save money. Every other German state carried on doing it, and for nearly two decades Berlin lost staff to neighbours who could offer the status — Brandenburg is a commute away. From the 2022/23 school year Berlin reversed itself, offering civil-servant status to new hires, to teachers arriving from other states, and to existing staff in annual cohorts.

What happened. By the 2025/26 school year, 58.2% of Berlin's roughly 35,000 teachers held civil-servant status, up from 31.1% in 2021/22, with just under 12,000 serving teachers converted. The Senate created 35 full-time administrative posts, funded through 2026, purely to process applications and check credentials.

The catch. Eighteen years is a long time to test a saving. The status is not free either: German civil servants may not strike, and their pension and health costs land on the state budget decades later, which is why Berlin dropped it in 2004. The education senator's claim of better trainee retention was made without published figures, so what is demonstrated so far is the status, not yet what it bought.

Berlin Senate Chancellery, press release of 22 March 2022; news4teachers, 31 March 2026

Japan

Japan recruits teachers through a competitive examination and does not employ people who lack the qualification. It reports 0.0% non-fully-qualified staff to the OECD and 0.2% of posts unfilled at the start of the school year. In 2021 the education ministry, MEXT, ran the country's first national survey of teacher shortages, to find out what those figures were missing.

What happened. On 1 April 2021, the first day of the school year, 2,558 teaching posts were unfilled — 0.31% of the 836,079 required. The gaps sat in 1,897 of the 32,903 schools surveyed, or 5.8% of them: 1,218 posts in elementary schools, 868 in junior high, 217 in high schools and 255 in special-needs schools. Special-needs schools were the worst affected in proportion, with 13.1% short of a teacher.

The catch. The first day of the school year is the kindest possible moment to count. Posts fall vacant during the year through illness and parental leave, and this survey existed precisely because the tidy annual figures hid that. The 0.0% on the board is accurate — and it is also why the ministry had to commission a separate survey to see the problem at all.

Japan's Ministry of Education, Culture, Sports, Science and Technology (MEXT), first national survey of teacher shortages, 2021, reported by Nippon.com; OECD Education at a Glance 2025, Tables D8.2 and D8.3

What a machine would optimise for

This is arithmetic, not advice and not a prediction. We name the single number being maximised and follow it wherever it goes. The point is to see the shape of the answer a calculator gives.

The number being maximised

The single number is the share of secondary teachers who are not fully qualified: 7.1% across the OECD in 2022/23. Drive it to zero and nothing else enters the calculation.

  1. Change the definition before hiring anybody. 'Fully qualified' means whatever a country's own rules say it means. England counts 2.8% and Sweden 21.6%, and the OECD flags Sweden's definition as the strictest because it also counts teachers working outside their certified subject or grade. Apply England's rule in Sweden and thousands of teachers change category without one classroom changing.
  2. Delete the post rather than fill it. The measure is a share of teachers, so a lesson that is not taught contains no unqualified teacher. Czechia's pupils per teacher fell from 12.9 in 2018 to 12.0 in 2023; running that back would free about 7% of teaching posts, roughly the share currently taught by someone unqualified.
  3. Buy the scarce subject, not the popular one. Czechia's ministry counted 1,463 missing physics posts, 1,413 in computing and 1,220 in maths at lower secondary for 2025/26. One physics teacher removes more of the shortage than one history teacher, because there is no shortage of history teachers.
  4. Ignore the regions. Karlovy Vary runs 22% unqualified teaching against a Czech average of 10%. Moving one qualified teacher from Prague to Cheb improves the regional figure and leaves the national one exactly where it was, so a calculator watching only the national number has no reason to move anyone.

The arithmetic

OECD average, 2022/23: 7.1% of secondary teachers are not fully qualified. In a system of 100,000 secondary teachers that is 7,100 people, and there are two routes to zero. Route one recruits 7,100 fully qualified teachers who do not currently exist — Czechia produced about 3,000 teacher-training graduates a year against a historic 4,000. Route two removes 7,100 posts. On round numbers, Czech regional education had about 150,000 full-time-equivalent teachers and 12.0 pupils per teacher, so roughly 1.8 million pupils; at the 2018 ratio of 12.9 the same pupils need about 139,500 teachers, which is 10,500 posts fewer, or 7% of the workforce. Neither route changes a single lesson plan, and the indicator cannot tell them apart.

What the machine would miss

The most important part of this page. A number that goes up can still be paid for by somebody, and some things never make it into the number at all.

Who pays

The children at the back of the queue, and they pay in the only currency they have, which is the years they get. Unqualified teaching in Czech basic schools runs at 10% nationally and 22% in the Karlovy Vary region; in Japan, special-needs schools were short of a teacher in 13.1% of cases against 4.9% of elementary schools. In England the phased maths bursary cost an estimated 275 maths teachers who never entered a classroom in 2019/20 — those children have now left school, and no later correction reaches them. The bill for a policy tested on a cohort is always paid by that cohort.

What the number flattens

Everything that makes a shortage a shortage. France reports 1.3% in primary and 9.1% in secondary; averaged, the country looks mildly troubled instead of specifically short of secondary teachers. Subjects vanish the same way — the Czech ministry's 2026 survey found physics, computing and maths accounting for the largest gaps while other subjects had none. And the two ends of the table are not measuring the same object: the OECD warns that Sweden's 21.6% is high partly because Sweden counts out-of-field teaching, so a country that does not count it looks better without teaching any differently.

What a spreadsheet cannot see

Who is actually standing in the room, and for how long. A fully qualified geographer covering Year 9 physics is a green tick in this table and out of her depth at the whiteboard. A merged class, a dropped option, a headteacher taking Friday maths and a rotation of supply teachers all leave the post recorded as filled. Nor does the table see churn: 63% of Austrian secondary teachers who resigned in 2022/23 had been teaching less than five years, and across 19 countries 6.5% of qualified teachers left the profession that year. A class on its fourth teacher in three years counts exactly the same as a class that has had one.

Where does your country stand?

Open my country

Where the numbers come from

We are a newspaper, not a government and not an adviser. Nothing on this page is a recommendation to you or to anyone in office. It is what was tried, what it measured, and what a calculator would say if you let it loose on the same problem.

One good piece of thinking a day

The day's most worthwhile story, and the question underneath it. No spam, one click to leave.

One click to leave. We never sell or share your address.