Answers are grounded in this paper and cite their sources.
Randomized trial examines improvements in macro-fiscal forecast accuracy in France, suggesting a modest impact of HCFP.
For more than two decades, France has produced macro-fiscal forecasts that have tended to be overly optimistic, despite the fiscal discipline framework imposed by the European Union. This study draws on data from the Stability Programmes between 1998 and 2024 to trace the evolution of macroeconomic and fiscal forecast errors and to assess whether the creation of the French High Council of Public Finance (Haut Conseil des finances publiques, HCFP) in 2013 is associated with improved forecast accuracy. Descriptively, forecast errors, particularly those related to growth, debt, and expenditure, declined markedly after 2013, both in magnitude and direction. An interrupted time-series (ITS) analysis identifies a robust structural break in public expenditure forecast errors around 2013, while growth, revenue, deficit, and debt series do not exhibit statistically significant breaks. To address the concern that the observed improvement could simply reflect France becoming easier to forecast, we complement the ITS with a within-France difference-in-differences (DiD) comparing the French government’s Stability Programme errors to forecasts produced by the IMF (World Economic Outlook) and the European Commission (AMECO) for the same target years, using target-year fixed effects to absorb common shocks. The ITS, the forecaster DiD, non-parametric Wilcoxon rank-sum tests, and a permutation-based ITS with randomly assigned break dates converge on a robust improvement for expenditure (of about 0.5 pp), moderate evidence for debt, and no HCFP-specific effect for GDP, revenue, or the fiscal balance. The paper concludes that, despite its merely advisory mandate, the HCFP plausibly contributes to fiscal transparency on the expenditure side, though the magnitude of its disciplining effect remains modest and heterogeneous across fiscal aggregates.
No takes yet. Share an insight, caveat, or question.
Théo Metz (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: