France Needs 45 Years to Grow Its Way Out

Julien Reszka · 2026-01-21T09:41

Cutting spending when you already spend 57% of GDP gives you six times less growth than the same cut at 20%. France is on the wrong part of the curve.

6x — Times more growth a country at 20% spending gains from a 1-point cut compared to a country at 50%, derived from the power law fit with alpha=1.5 — 6x Times more growth a country at 20% spending gains from a 1-point cut compared to a country at 50%, derived from the power law fit with alpha=1.5 Derived from cross-country power law fit, World Bank data 2005-2023

Government Size Has No Sweet Spot showed that cross-country data fits a power law better than the Armey Curve. The curve falls monotonically as spending rises. That seems like good news: cut spending, gain growth.

France at 57% spending sits on the flat end of the power law curve where cuts yield little growth

The bad news is in the shape of the curve.

A power law is steep at low spending levels and flat at high ones. The marginal growth gain from cutting one point of spending is proportional to spending^(-(alpha+1)). With alpha = 1.5, the exponent on the marginal term is 2.5. A country at 20% spending gains roughly (50/20)^2.5 = about 6 times more growth from the same cut than a country at 50%.

Singapore at 15% would gain more still. France at 57% is near the flat end of the curve.

This creates a trap. The cure is real. The payoff is small. The disruption is large. That combination is what makes it politically impossible to cut.

Three paths out:

Cut fast and deep: a large one-shot consolidation (10+ points) moves the country to a steeper part of the curve. Canada cut roughly 10 points in three years in 1995-1997.
Hold spending flat and let GDP growth do it. The ratio falls by about 2% per year as a proportion when real GDP grows at 2%. This is the Sweden post-1993 playbook - and it took 15 years.
Do nothing. The ratio stays high. Growth drag compounds. Debt rises. Future cuts become more expensive.

The math on the slow path is brutal.

France currently spends around 57% of GDP. At 0% real spending growth and 1.5% real GDP growth - optimistic for France's recent record - the spending-to-GDP ratio falls by about 1.48% per year as a proportion of its current level. To reach 30% takes roughly 45 years. To reach 20% takes over 90.

Canada and Sweden succeeded because they combined spending freezes with structural reforms that pushed real GDP growth above 2%. France has not done either.

The conclusion from the model is uncomfortable. Incremental annual cuts of 0.5 to 1 point are arithmetically real but economically invisible. They do not move the country far enough along the power law curve to produce a detectable growth response in a political cycle.

Cut fast, or wait 40 years. There is no comfortable middle path that produces visible results.

Myth: Cutting 5 points of public spending delivers roughly 5 points worth of growth improvement — Reality: The power law is steep at low spending and flat at high spending. The same cut delivers six times less growth improvement at 50% than at 20%. France is on the flat part of the curve. — **Myth:** Cutting 5 points of public spending delivers roughly 5 points worth of growth improvementPower law model, R2=0.42, World Bank cross-country data 2005-2023

When evaluating a fiscal consolidation plan, ask how many percentage points it moves spending and over how many years. A 1-point cut over 10 years in a high-spending country is noise. A 10-point cut in five years is signal.
Post on X

Discussion

Is the fiscal consolidation plan you are evaluating large enough and fast enough to move the country to a meaningfully steeper part of the curve?
Post on X