Fiscal France: a country that levies a great deal and still explains the use of public money too poorly.
France does not only have a problem with the level of tax. It has a problem of consent, readability and confidence.
In 2025, compulsory levies represent 43.6% of gross domestic product, according to provisional INSEE data.[1] This aggregate rate does not by itself describe the taxpayer’s experience, which is a stack of taxes, contributions, local taxes, social levies and filing obligations.
243 low-yield taxes: the symptom of a system that piles up
The Court of Audit lists 243 low-yield taxes in 2024, bringing in a combined €5.98 billion.[2] Low revenue is not automatically enough to justify abolishing a tax: its purpose, its management cost, its behavioral effects and the existence of a replacement solution must also be examined.
The accumulation of narrow levies can nevertheless make the tax landscape less readable. Each scheme should therefore be evaluated with a date, a yield, an administrative cost and a measurable objective.
465 tax expenditures listed in the budget documents
The official term is “tax expenditure.” It refers to exemptions, allowances, tax credits, reductions, special rates or derogatory regimes compared with a reference tax norm. The budget document preparing the 2026 finance bill lists 465 tax expenditures appearing in volume II of the “Voies et moyens” annex to the 2025 finance bill, 82 of which are being phased out.[3]
These schemes may pursue legitimate objectives. The problem arises when they are extended without sufficient evaluation, when their cost is poorly documented, or when their beneficiaries and results are no longer compared with the other possible instruments.
Tax confidence is a political fact to be measured
The 2025 barometer of the Council on Compulsory Levies studies consent to taxation, the perception of how public money is used and the confidence placed in those responsible for managing levies.[4] The results should be read question by question and in their evolution, rather than summed up in a single formula.
The decisive point is not only the amount paid. It is the ability to understand what is levied, to identify the objective, to check the results and to challenge a decision when it is wrong.
The big blind spot: the full cost
A public scheme is not limited to the appropriation announced. Its full cost may include management, control, information systems, litigation, delays, indirect effects and duplication. A credible evaluation must make these components visible.
What should be done
The first tax reform should be a reform of readability and evaluation:
- publish for each levy its objective, its yield, its management cost and its review date;
- evaluate the most costly tax expenditures on the basis of their beneficiaries, their cost and their results;
- document the full cost of tax policies;
- organize the abolition, merger or transformation of schemes that have become useless;
- present official data and reform proposals separately.
Conclusion
Consent to taxation cannot be decreed. It depends on understanding, perceived fairness, the quality of service and proof that the money levied produces results.
The central question
Before asking for further effort, the state must be able to answer clearly: what has it done with the money already levied, with what results and at what full cost?