France devotes 57.2% of its GDP to public spending in 2025, second in the European Union, with a deficit exceeding 5% and debt standing at 115.6% of GDP at the end of 2025 (annual data Eurostat/INSEE), before reaching 117.5% at the end of March 2026 (quarterly data INSEE). Yet its health and education spending per capita surpasses that of countries that achieve better school and health results, a gap that the OECD documents without mincing words. The real constraint is not to spend less in absolute terms, but to move away from a model where the volume of spending has long served as policy.

The essentials

  • France spends 7.4 percentage points of GDP more than the eurozone average (49.8%), without translating this gap into superior performance on health, education, or administrative productivity indicators.
  • The Ravignon report (May 2024) quantifies the cost of territorial complexity at 7.4-7.5 billion euros, mapping duplications between the central state and local authorities.
  • Clarifying competencies between levels of government is the most documented lever, but the least exploited politically.
  • The deficit reaches 5.1% of GDP and debt 117.5% at the end of March 2026 (Eurostat / FIPECO), which reduces margins for gradual reform and forces explicit choices.
  • Restoring fiscal trust, a necessary condition for durable tax consent, requires accounting for the use of every euro as much as for its amount.

Spending more has never guaranteed spending better

The figure is well known but rarely measured against what it produces. France shows 57.2% of GDP in public spending according to Eurostat, a level surpassed only by Finland in the EU, with a deficit of 5.1% and sovereign debt at 117.5% of GDP at the end of March 2026, according to FIPECO consolidated data. The eurozone average stands at 49.8%. This 7.4-point gap represents, on a GDP base of approximately 2,800 billion euros, some 200 billion in additional annual spending compared to the European standard.

French public services are not commensurate with this financial effort. On PISA indicators, France ranks in the upper-middle range of the OECD, surpassed by countries that spend less per student. On healthy life expectancy, Spain, the Netherlands, and Sweden perform better with comparable or lower public health budgets as a share of GDP. On user satisfaction with administrative services, France regularly ranks below Germany, Sweden, or Estonia.

This gap between volume and performance transcends recent cyclical trends. Adolphe Blanqui, a 19th-century liberal economist often confused with his brother Auguste because of the family’s shared surname from the Nice area, already formulated the idea that spending efficiency takes priority over its volume. Current data show that France has pushed the experiment to its conclusion: it has reached a level of taxation and spending where the marginal return of each additional euro is structurally negative. Adding resources no longer solves the problems submitted to it.

Territorial complexity costs more than the services it provides

The Ravignon report, published in May 2024 at the government’s request, is one of the most precise works ever produced on the question. It does not prescribe an austerity cure. It maps duplications. And what it finds is instructive: between the central state, regions, departments, intercommunalities, and communes, dozens of competencies are exercised in parallel by multiple levels of government, without coordination or clearly assigned responsibility.

Employment policy illustrates the mechanism well. Pôle Emploi, now France Travail, coexists with local missions, local insertion plans, regional apprenticeship advisers, and departmental insertion schemes. Each level has its staff, its offices, its information systems. The beneficiary, meanwhile, circulates from one window to another. The Ravignon report quantifies the cost of this territorial complexity at 7.4-7.5 billion euros in total, of which 6 billion euros for local authorities and 1.5 billion for the state, without these duplications corresponding to additional services rendered to users.

This is a line-by-line identification of redundant spending.

The political problem is that each duplication has its constituency, its network of elected officials, and its captive users. Clarifying competencies would require levels of government to accept relinquishing prerogatives they have often painstakingly acquired. This is what is called a public good with diffuse benefits and concentrated costs: everyone would gain from rationalization, but those who lose functions feel it immediately, while those who gain clarity only notice it later.

Countries that succeed have settled on competencies

Comparative experience provides concrete benchmarks. Scandinavian countries have an expensive state—Denmark spends more than 50% of GDP—but it is intelligible: each level knows what it does, and the user knows whom to contact. The Swedish reform of the 1990s, carried out after a major budgetary crisis, did not simply cut spending: it refocused the state on its essential missions, clearly transferred certain competencies to municipalities, and introduced mechanisms for systematic evaluation of public policies. Swedish public debt, which exceeded 70% of GDP in 1995, fell below 35% in twenty years, without public service indicators collapsing.

Germany offers another model. German federalism is often criticized for slow decision-making, but it has a virtue: the Länder are fiscally responsible and their competencies are constitutionally bounded. A Land that spends poorly sees it in its budget, not in the federal government’s. This traceability makes political trade-offs more visible and more honest.

France has attempted several times to clarify its territorial organization. The NOTRe law of 2015 eliminated the general clause of competence for regions and departments, in hopes of reducing duplications. The reform remained halfway: competencies were partly redistributed on paper, but human and financial resources did not follow the same logic. Expected savings did not materialize, because administrative rationalization requires sustained political will over at least two electoral terms, a horizon incompatible with the French electoral calendar.

Administrative AI: Real contributions and structural limits

The digital transformation of the state opens an additional avenue. Estonia has become the world reference by refounding its processes before computerizing them, without massively eliminating civil servants. 99% of Estonian public services are accessible online, including voting. The cost per administrative act has been divided several times since the 2000s. The country devotes less than 40% of its GDP to public spending for governance indicators clearly superior to France’s.

Generative AI opens perspectives in the same direction: automation of repetitive tasks, faster file processing, personalization of responses to users. Several French administrations are already experimenting with processing assistants in tax services and CAF offices. But these tools amplify the existing organization; they do not reform it. A redundant automated process remains a redundant process. This is why the question of who truly benefits from administrative modernization cannot be treated as a purely technical question.

Digital transformation is a necessary but insufficient condition. It can reduce unit costs of administrative acts, but if the savings generated are not reallocated or if redundant structures are not eliminated, they simply produce a digital state as thick as the paper state.

Fiscal trust, the blind spot of spending debates

There is a register that numerical comparisons struggle to capture: that of tax consent. In France, the mandatory levy rate reaches 43.5% of GDP according to FIPECO, the highest in the OECD along with Denmark. But whereas Danes trust their state to use that money, the French express growing distrust of public institutions. This distrust is not irrational: it partly reflects the concrete experience of users who do not understand where their money goes, who find themselves facing multiple windows for a single problem, and who perceive an administration more preoccupied with managing itself than serving them.

As this article on French budgetary trajectory analyzed, the problem exceeds accounting. Naming what one decides to finance and what one chooses not to finance anymore is a political act that engages the democratic legitimacy of taxation. A state that does not account for how spending is used progressively erodes its own foundation.

Rebuilding fiscal trust passes through three practices that the highest-ranking countries have in common: systematic evaluation of public policies with publication of results, clarity of the link between tax and service rendered, and accountability of levels of government on their results rather than on their resources. France has the tools—the Court of Accounts, Parliament, the High Council of Public Finances—but their use remains advisory where it should be binding.

The blind spots of the Ravignon report

The Ravignon report identifies the potential for savings. It does not say how to achieve them politically. This is the limitation of every good governance report: it describes the problem with precision and leaves to politicians the difficult part.

The billions of euros in duplications mapped do not transform into savings by decree. They require negotiations between the state and local authorities, personnel transfers, closures of structures whose agents are often local voters. They also require temporal coordination: savings arrive eventually, transition costs are immediate. In a context of a deficit at 5.1% and pressure from markets on French sovereign debt, pressure whose mechanisms have been precisely documented, the government is tempted by faster trade-offs: hiring freezes, cuts to investment, postponement of structural reforms.

Labor economists call this phenomenon the short-term bias in public finances: governments cut where it is visible and quick—investment, training—and preserve where adjustments are slow and conflictual. Spending thus remains high, but degrades qualitatively. One pays for structures, less for services.

France must undertake a reform of efficiency over a sufficient duration to produce measurable results before budgetary constraints force less intelligent cuts. The Ravignon report provides a roadmap. The next budgetary deadlines will indicate whether the political will exists to seize it, or whether it will, once again, be postponed to the next government.

Sources

  1. FIPECO, The definition, level, and distribution of public spending
  2. Ravignon Report, Reducing Public Spending Without Lowering Service Quality, May 2024 (French Government Publications)
  3. Eurostat, Public Administration Spending as % of GDP, 2025
  4. OECD, Government at a Glance 2025 (OECD Publishing)
  5. FIPECO, Data on French deficit and public debt, updated April 2026
  6. INSEE – Deficit and public debt 2025 (IR no. 78, March 2026)
  7. FIPECO – Public Spending and Mandatory Levies 2025
  8. Eurostat – Deficit and spending eurozone 2024 (October 2025)
  9. Ravignon Report – Ministry of Interior (May 2024)
  10. INSEE – Debt Q1 2026 (June 2026)
  11. IRDEME – Efficiency of OECD health systems