What Public Accounts Can Establish about the Displacement of the Public Debt Burden
Accounting Identities, Independent Benchmarks, and Limits (France, 1995-2025)
Stéphane Lalut
October 4, 2026
Abstract
What can public accounts establish about the burden of French public debt over time, across generations, households, countries, and levels of government? This working paper, written in French, checks every decomposition against an independent benchmark, reconciles three public readings deemed contradictory, and tests the AWP-03 framework against the accounts: supported within a wealth perimeter, conditional or beyond reach elsewhere.
anthropy
public debt
debt dynamics
intergenerational burden
data verification
JEL : H63, H62, H72, E62, C82
Résumé
Que permettent d'établir les comptes publics sur la charge de la dette française, dans le temps, entre générations, entre ménages, entre pays et entre échelons ? Ce working paper contrôle chaque décomposition par un témoin indépendant, rapproche trois lectures publiques réputées contradictoires et met à l'épreuve des comptes la lecture de l'AWP-03 : soutenue dans un périmètre patrimonial, conditionnelle ou hors de portée ailleurs.
Anthropy is the hypothesis that social systems displace disorder rather than resolve it.
This working paper is written in French; an English translation will follow. It tests the question “who pays the debt?” against public accounts. It does not settle it: it establishes what thirty years of official French series can and cannot show about five displacements of the burden — over time, toward later generations, across households, across countries, and across levels of government.
A verification rule
An accounting decomposition is exact by construction: its residual absorbs any error. The paper therefore checks each decomposition against a benchmark the computation does not use — the ratios Eurostat itself publishes — validated algebraically and then seen to reject an actual error: GDP converted into ECU before 1999, which put the 1995 debt at 57.5% of GDP instead of 57.8%. Each check covers one class of errors only; the paper lists them.
Three public readings reconciled
A 2014 citizens’ audit attributed 38% of the 1980-2012 rise in debt to the snowball effect; the French Court of Audit reduces it to 8.8 points over 1978-2024; the iFRAP Foundation computed in 2024 that interest represented 53% of the debt accumulated since 1974. Windows and measures explain much of the gap: in Eurostat series the effect accounts for 43% of the rise in 1996-2012, has subtracted 1.74 points a year since 2017, and cumulative gross interest measures a different object from its net effect.
The AWP-03 framework put to the test
The paper delivers a statement-by-statement verdict on the anthropic reading of debt proposed in AWP-03: the transfer toward future generations is supported only within a wealth perimeter; the effect of the interest-growth differential holds as accounting mechanics, not as a transfer; the asymmetry between mobile and captive households lies beyond what the accounts can test, which show only that the incidence of the same effort depends on the instrument.
What drove the rise in French public debt since 1995?
Primary deficits: from 1995 to 2025 the ratio rose from 57.8% to 115.6% of GDP, of which primary deficits account for 50.5 of 57.7 points, interest and nominal growth effects having nearly offset each other over thirty years. AWP-09 (Stéphane Lalut, 2026, DOI 10.5281/zenodo.23143030) recovers this result with an independent benchmark.
Why do opposite public readings of the debt all rest on official accounts?
Because they measure neither the same window nor the same object. In 1996-2012 the snowball effect accounts for 43% of the rise in the ratio, close to the 38% found by the 2014 citizens' audit; since 2017 it has subtracted 1.74 points a year, the figure given by the French Treasury; cumulative gross interest, without the growth that offsets it, measures yet another object.
Do public accounts tell who pays the debt?
They locate flows, accounting positions, and adjustments, not the final economic burden. They show that less than one-fifth of deficits since 1996 matched a net increase in non-financial public assets, that more than half of the debt is held by non-residents, and that the household incidence of the same fiscal effort reverses with the instrument chosen.
