Business failures in France: catch-up or deterioration?

In 2025, 68,524 businesses entered safeguard, receivership or judicial liquidation proceedings in France, according to the series of the Institut national de la statistique et des études économiques (INSEE) derived from commercial court judgments. It is the highest figure since the series began in 1990; the pre-pandemic record dated from 1993, with 63,709 failures, and 2024 had already exceeded it.

The question is not whether the figure is a record — it is — but what it measures: the catch-up of businesses kept artificially alive during the health crisis, or a real weakening of the productive fabric? The two readings do not call for the same response.

From 1990 to 2019, France counted on average 54,801 failures a year, in a range from forty-three thousand to sixty-four thousand depending on the business cycle. Then the series breaks: 31,217 in 2020, 27,582 in 2021 — the historical low, in the middle of the health crisis — and 41,241 in 2022. Emergency aid, State-guaranteed loans and the deferral of tax and social deadlines suspended the usual selection.

The catch-up follows: 56,274 in 2023, 66,115 in 2024, 68,524 in 2025, or +3.6% over the last year. Cumulatively, the three freeze years (2020-2022) produced 64,363 fewer failures than the pre-crisis average; the three catch-up years (2023-2025) produced 26,510 more. The deficit is not filled: over six years, 290,953 businesses failed, that is 37,853 fewer than if the average 1990-2019 pace had simply continued.

What has changed is the denominator. The number of active businesses went from 5,480,094 in 2019 to 5,615,765 in 2024, in a series whose scope changed in 2021. Set against that stock, the 66,115 failures of 2024 represent 1.18% of businesses — against 0.93% in 2019 and 1.80% in 2009, at the worst of the financial crisis. The 2024 failure rate is therefore higher than in 2019, but markedly lower than in 2009.

What has also changed is the nature of businesses: 1,165,795 creations in 2025, 758,542 of them under the micro-entrepreneur regime, or 65% of the total. A large share of these units has neither employees nor bank debt; when they stop, they do not go through a court, and so do not enter the count of failures.

Failures usually rise when activity falls and margins are squeezed: in 2009, the −2.8% contraction and a non-financial corporations' margin rate down to 30.6% had pushed failures to 63,283. Nothing of the kind in 2025: growth is weak (0.8%) but positive, the margin rate stands at 32.2% — above its 2019 level (30.7%) — and business investment as a share of gross domestic product (GDP), 12.3% in 2024, also exceeds its pre-crisis level (12.1%).

In other words, the indicators of average business health did not deteriorate at the moment failures hit their record. That mismatch tilts the reading towards catch-up: the businesses failing in 2023-2025 are, in part, those that 2020-2022 had spared, while the average of the productive fabric was doing rather well.

Three episodes structure the series. The 1993 recession takes failures to their old record. The financial crisis brings them back to 63,283 in 2009, without exceeding it. And the health crisis produces the reverse of a crisis: a collapse of failures in 2020-2021, because the State suspended the cash constraint that usually triggers them.

The micro-entrepreneur regime, created in 2009, sent creations from 363,394 in 2008 to 864,557 in 2019 and 1,165,795 in 2025: it changed the reference population, and hence the meaning of an absolute number of failures.

In 2025, 68,524 failures against about 1,165,795 creations: the balance remains massively positive. The margin rate is 32.2%, growth 0.8%. The 2026 data will tell whether the catch-up is running out or the level is holding.