France spends more in absolute terms than some of its European neighbors, but its relative position depends heavily on the indicator chosen, and it remains below the OECD average for the share of total public spending devoted to education. According to the national DEPP indicator, France devotes 6.8% of GDP to education in 2024; according to the OECD indicator with a restricted scope (primary to higher education excluding preschool), this figure is 5.4%. The French school system embodies a paradox that PISA data makes increasingly difficult to ignore each year: scores are close to the organization’s average, following a decline significantly steeper than the OECD average in mathematics and markedly steeper in reading comprehension: 19 points lost between 2018 and 2022, compared to 10 points on average in the OECD, and social background continues to weigh on results more heavily than elsewhere. Yet a window is opening: the declining trend in student enrollment liberates margins that nothing compels to absorb through the status quo.
The Essentials
- France devotes a significant share of its national wealth to its schools (5.4% according to the OECD indicator, 6.8% according to the national DEPP indicator), yet without correcting the weight of social background on school results.
- French PISA scores are close to the OECD average but have declined more sharply than this average in mathematics and even more sharply in reading comprehension, a sign that overall spending does not compensate for allocation failures.
- The equity lever is documented: investing early, in early childhood and disadvantaged schools, produces returns superior to undifferentiated increases in resources.
- The decline in student numbers offers a rare opportunity to redirect resources without increasing the overall budget envelope.
- The choice between targeted reallocation and passive budget absorption will determine whether this window closes without result or initiates lasting rebalancing.
France Spends More and Achieves Less Than Its Neighbors
The reasoning seems simple: a country that invests more in its schools should obtain better results. France illustrates why this reasoning is incomplete. According to the OECD indicator, France spends 5.4% of its GDP on education, a share exceeding that of many comparable countries; taking into account the national DEPP indicator, which encompasses a broader scope, the gap is even greater. These resources are not neglected: teachers are trained, infrastructure exists, school enrollment is nearly universal from age three. Yet, when measuring what fifteen-year-old students can do, France finds itself at the average, following a 21-point decline in mathematics between 2018 and 2022—a decline significantly steeper than the OECD average of 15 points across 81 tested countries, with only 14 experiencing a larger drop—and a marked decline in reading comprehension as well.
This relative slippage is not a cyclical accident. It reflects an allocation problem. French spending is calibrated on a logic of uniformity: with few exceptions, a student in a rural secondary school receives the same resources as a student in a disadvantaged suburban school. Priority education policies exist and constitute a genuine step forward, but their scope remains limited relative to the scale of disparities. Recent OECD surveys confirm that France is among countries where social mobility is most constrained by family background, a result unexplained by the overall amount of spending, but by how it is distributed over time and space.
Social Background and School Trajectories in France
The gap between a student whose parents are senior managers and a student from a working-class or lower-income background widens in France well before entering secondary school. Education researchers have documented this for a long time: a substantial part of achievement inequality is already constituted upon entry to first grade. Available vocabulary, familiarity with writing, capacity for sustained attention—these cognitive and social resources are acquired or lacking in the earliest years of life. Primary school amplifies them more than it corrects them.
International comparison reveals that this mechanism is not inevitable. Nordic countries, and to a lesser extent Canada and Estonia, manage to produce systems where social background weighs significantly less on results. Their common point lies less in a higher level of spending than in concentrating resources where impact is strongest: early childhood and schools serving the most vulnerable populations. American economist James Heckman has quantified this differential return for over two decades: one dollar invested before age five in supporting disadvantaged children produces long-term returns in health, employment, and productivity far superior to the same dollar spent later in schooling. The principle applies in France.
This observation does not mean that the rest of school spending is useless. It means it is poorly weighted. France has built a system that treats equity as a marginal adjustment at the periphery of a uniformist model, rather than as the organizing principle of allocation. The result is an expensive system, relatively equitable in appearance—free education is universal, school assignment rules exist—but profoundly unequal in its real effects. An article in this journal on French budgetary choices recalls that this logic of uniformity runs through all social policies: protecting everyone at the same level often amounts to protecting insufficiently those who need it most.
Priority Schools: A Good Idea Under-Financed
French priority education policy, priority education zones, then REP and REP+ networks, is an explicit response to this problem. Its principle is sound: concentrate additional resources on schools serving the most disadvantaged students. Its execution remains insufficient relative to the stated ambition.
Additional resources allocated to REP+ schools are real: smaller classes, bonuses for teachers, coordination hours. But the resource gap between a priority school and an ordinary school in a wealthy area remains modest relative to what international comparisons suggest is necessary. More fundamentally, teacher assignment follows a seniority logic that tends to send the least experienced teachers to the most difficult schools, a well-documented mechanism that directly contradicts priority policy objectives.
The problem is also qualitative. Smaller classes are insufficient if pedagogy is not adapted to the specific needs of students arriving with significant language delays or concentration difficulties. Work on reading and explicit instruction shows that certain methods produce far superior results with vulnerable populations, and these methods are not systematically deployed in schools that need them most. The question of how joins the question of how much.
Demographic Decline as a Lever, Not as Passive Savings
For several years, the number of students enrolled in France has been declining. This trend, linked to falling birth rates beginning in the mid-2010s, will accelerate in coming years. Fewer students means, mechanically, budgetary room to maneuver: at constant budget, per-student spending increases. At reduced budget, consolidating public finances, declining enrollment allows maintaining quality without increasing funding.
These two interpretations are radically different in their consequences for equity. If freed margins are absorbed by class closures uniformly distributed or returned to the general budget without redirection, the system retains its structural flaws. If they are deliberately redirected toward disadvantaged schools and to the earliest years of schooling, they constitute an opportunity that ordinary budgetary constraints make extremely rare: improving equity without increasing total spending.
Several OECD studies emphasize this window opened by school demographics. It will not open indefinitely. School demographics do not decline forever, and political pressure over class closures, which affect already fragile rural territories, is real. The question posed to decision-makers is thus one of timing as much as direction: seizing the window implies deciding now on allocation priorities, before margins are absorbed by institutional inertia.
Decline in Student Numbers by 2035: Three Scenarios According to Choices Made
Two trajectories are open for the coming decade, and they are not symmetric in their effects on social mobility.
In the first scenario, savings induced by declining enrollment are deliberately redirected toward the earliest years of schooling—enhanced preschool, support for children under three in working-class neighborhoods—and toward schools serving the most disadvantaged populations. This requires modifying rules for teacher assignment to value experience in difficult positions, substantially increasing resource gaps between schools according to the social index of students served, and investing in continuing education in pedagogical methods documented as effective with vulnerable populations. Signals to watch in this trajectory are readable: the evolution of per-student spending in preschool in priority zones, and the evolution of PISA score gaps according to families’ socioeconomic status. If the gap narrows between two PISA survey cycles, reallocation produces its effects. The horizon for observable results on intergenerational mobility is longer, one to two decades, but first pedagogical signals appear well before.
In the second scenario, budget savings linked to declining enrollment are absorbed by reducing the public deficit without sectoral redirection. Class closures follow the geography of demographics rather than educational needs. Priority schools receive the same proportional adjustments as others. The system then retains its current configuration: overall spending exceeding that of many comparable countries, undifferentiated allocation, and a weight of social background on school trajectories remaining among the highest of advanced democracies. This scenario is most likely by default, not because it would be deliberately chosen, but because institutional inertia in a centralized school system is a well-documented phenomenon.
Structural reforms—modifying assignment rules, reweighting allocations, imposing pedagogical methods—encounter corporate resistance and political constraints that have defeated several previous attempts.
The two trajectories distinguish themselves less by amount than by governance. France has the financial means to improve the equity of its school system without spending more in volume. The productive constraint weighing on the French economy makes this approach all the more relevant: a more equitable school system constitutes, in the long term, a gain in productivity and cohesion. The political decision remains to be made—by whom, in what institutional configuration, before the demographic window closes.
Lessons from Foreign Experience
Comparative experiences do not lack for feeding reflection. Estonia, often cited in PISA rankings, built its system around explicit pedagogy and demanding continuous teacher training, while maintaining overall spending below that of France. Canada, and Quebec in particular, significantly reduced gaps according to social background by strengthening early childhood services and providing schools in disadvantaged areas with substantially superior resources. Poland, before the decade of regressive reforms in the 2010s, improved its PISA scores notably in one decade by reforming pedagogy and teacher training rather than spending level.
These examples share a characteristic: they treat equity as a central organizational variable, not as a marginal adjustment. They invest in what happens in the classroom as much as in the structures framing it. And they accept that schools are not treated identically so that students are treated equally.
France has the institutions to conduct this type of transformation: a centralized National Ministry of Education, an inspectorate capable of evaluating practices, pilot experiments in certain REP+ networks showing that different results are achievable. OECD work on intergenerational mobility indicates moreover that countries that have progressed best on this indicator over the past two decades are those combining early investment and resource targeting—two dimensions on which France has documented room for progress and, now, a budgetary window to exploit them.
Sources
- OECD, Economic Surveys France 2026, chapter on public finances and long-term growth
- OECD, Intergenerational social mobility across OECD countries, March 2026 (OECD report, no guaranteed URL)
- OECD, PISA database, comparative results by country and cycle (available at ocde.org/pisa)
- DEPP – Education spending France 2024: €197.1 billion, or 6.8% of GDP
- OECD – Education at a Glance 2024
- DEPP / Ministry – PISA 2022 France
- OECD – Economic Surveys: France 2026
- OECD – Intergenerational social mobility (Working Paper No. 1858, March 2026)
- DEPP – School enrollment projections to 2035
- INSEE – Inequalities in school achievement at elementary level
- DEPP Information Note no. 25-52 – Education spending 2024
- OECD – PISA 2022, Volume I
- DEPP – Information Notes nos. 23.48 and 23.49 (PISA 2022 France)
- DEPP – State of Schools 2025 (OECD indicator 5.4%)
- Éduscol – History of priority education (ZEP, REP, REP+)
- Senate – Report on teacher assignment and seniority
- DEPP Information Note no. 23.48 – PISA 2022 mathematics
- DEPP – Enrollment projections to 2035 (April 2026)
- Senate, report on equality of opportunity (2020)
- IH2EF / DEPP – PISA 2022 Results France
- DEPP Note no. 23.18 – International comparisons of education spending
- INSEE – Social inequalities in school education
- OECD / Senate – Intergenerational mobility France
- Observatoire des inégalités / PISA 2022 – Social inequalities
- Ministry of National Education – Budget and finances of the education system
- ENS Lyon / Heckman – Return on investment in early childhood aid and education