What Finland’s PISA Decline Actually Shows About Equity

Finland has been education’s most-cited example for twenty years, usually in the same four notes: small classes, little standardised testing, respected teachers, happy children. The citation has outlived the data behind it. PISA 2022 — the OECD’s most recent round — says something more specific, more uncomfortable, and considerably more useful to anyone actually running a classroom.

What is still true

Start with what has not changed, because it is the part worth borrowing.

In Finland, the variation in mathematics performance between schools accounts for about 9% of the total variation — among the smallest of the 80 systems the OECD measured. Nearly all the difference in how Finnish 15-year-olds perform sits within schools rather than between them. In plain terms: which school a Finnish child attends still matters remarkably little.

The socio-economic picture is also, by international standards, good. Advantaged students outscored disadvantaged students by 83 points in mathematics, against an OECD average gap of 93. Socio-economic status explained 12% of the variation in Finnish mathematics performance, against 15% across the OECD. And 12% of Finland’s disadvantaged students scored in the top quarter of their own country’s results — what PISA calls academically resilient — against an OECD average of 10%.

By PISA’s own definition, Finland remains one of a small group of highly equitable systems. That should be said plainly before anything else, because the rest of this post is about a trend, not a collapse.

What has changed

Over the 2012–2022 period, Finnish students’ performance fell by more than 20 points on average in mathematics, reading and science. The OECD puts that figure in a way worth repeating: more than the typical gain a student makes, around the age of 15, over an entire school year.

It would be easy to file that under “the pandemic”, and easy to be wrong. The OECD’s own country note is careful here: the steep drop between 2018 and 2022 “confirmed and reinforced a decline that began much earlier.” Something was moving before 2020.

And the decline has not fallen evenly. Between 2018 and 2022, the gap between Finland’s strongest 10% and weakest 10% widened in all three subjects. In mathematics, the OECD notes, almost all students became weaker — but low achievers declined by more than high achievers did. Compared with 2012, the share of Finnish students performing below the baseline level of proficiency in mathematics rose by 13 percentage points.

Then the sentence that matters most for anyone who cites Finland as an equity model:

Between 2012 and 2022, the gap in mathematics performance between the top and the bottom 25% of students in terms of socio-economic status widened in Finland, while the average gap across OECD countries remained stable.

Read that twice. The OECD average held steady. Finland’s did not.

Why this is not a story about Finland

Performance fell in nearly all OECD countries in 2022; Finland is not an outlier in the fall. What makes it instructive is the combination: a system built explicitly around equity, still measurably more equitable than almost anywhere else, and losing ground at the bottom while the international average holds.

That combination retires a comfortable idea — that equity, once designed into a system, stays there. It does not. It is a running total, and when it slips it does not slip uniformly. It slips at the bottom first, quietly, in the students who were already furthest from the average and who are least likely to generate a complaint.

There is also a lesson about what to copy. The Finnish features that get quoted at conferences — the absence of high-stakes testing, the status of the profession, the shortness of the school day — are national policy settings that no individual teacher and few individual schools can adopt. The finding that actually travels is the diagnostic one: watch the distance between your weakest students and everyone else, and watch it over years rather than terms. Finland’s own numbers only revealed the problem because someone kept measuring the same thing for a decade.

What a teacher can take from this

Very little of the above is actionable at national level by the people who read it. Some of it is actionable in a classroom.

If the loss concentrates among the students furthest behind, then the practical question is not “how do we become more like Finland” but “what actually works for the students who are falling”. That question has better answers now than it did when Finland’s reputation was formed — evidence syntheses on what helps the lowest attainers in an ordinary mixed classroom are more specific, and more boring, than the folklore version of school reform. Explicit teaching in small steps. Scaffolding that is designed to be removed. Grouping that forms for a purpose and then stops. It is unglamorous work, and it is the work the data points at.

That is the subject of our week on Every Child is Special, which is built on exactly that evidence base rather than on any country’s reputation.

Sources

• OECD, PISA 2022 Results (Volume I and II) — Country Note: Finland, 2023.
• OECD, Education GPS — Finland, Student performance (PISA 2022).
• Finnish Ministry of Education and Culture, “PISA 2022: Performance fell both in Finland and in nearly all other OECD countries.”