Disproportionate Guarantee: An “Expired” Commitment Still Counts (Cass. com., 8 July 2026, No. 25-16.540)

French Supreme Court, Commercial, Financial and Economic Chamber, 8 July 2026, Appeal No. 25-16.540, published in the Bulletin

What happened in this case?

The facts are so commonplace that they will speak to many company directors. On 7 September 2011, a bank grants a company a business loan of €70,000. Two individuals, a man and a woman, stand as joint and several guarantors (cautions solidaires) for all sums due under this loan, up to €84,000, covering principal, interest and, where applicable, penalties or late-payment interest.

As is all too often the case, the borrowing company runs into difficulties: it is placed in court-supervised reorganisation (redressement judiciaire), then in compulsory liquidation (liquidation judiciaire). The bank then turns against the guarantors and sues them for performance of their commitments in 2021, some ten years after the deed was signed.

Before the Metz Court of Appeal, the guarantor being pursued obtains an initial success: the judges find that their commitment was manifestly disproportionate on the date it was entered into. Under former Article L. 341-4 of the Consumer Code, the bank should therefore not be able to rely on it. But the text reserves one scenario: the creditor regains its right to sue if the guarantor’s assets, at the time they are called on, enable them to meet their obligation. This is what is known as the “return to better fortune” (retour à meilleure fortune).

And it is precisely on this second stage of the analysis that the court of appeal goes off the rails. To hold that the guarantor was able to meet their obligation on the date of the summons, it excludes from their indebtedness several earlier guarantee commitments: two guarantees taken out in May 2010 with another bank for a term of 72 months, a guarantee from 2006 securing a €210,000 loan with a term of nine years, and an aval (aval) given to secure a €80,000 cash-flow credit granted at the end of 2012 and expiring at the end of 2013. Its reasoning: all these commitments were “expired” on the date of the summons, so the guarantor was “released” from them. Consequence: the guarantor is ordered to pay €38,815.33 in principal, plus interest at the statutory rate.

The guarantor lodges an appeal on points of law. The Commercial Chamber of the French Supreme Court, in a judgment of 8 July 2026 intended for publication in the Bulletin, quashes the Metz judgment on this point and refers the case to the Nancy Court of Appeal.

What does Article L. 341-4 of the Consumer Code provide?

Former Article L. 341-4 of the Consumer Code, resulting from the Act of 1 August 2003 known as the “Dutreil Act”, is one of the most protective provisions ever adopted in favour of individual guarantors committed to professional creditors. It provides that a professional creditor may not rely on a guarantee contract entered into by an individual whose commitment was, when it was concluded, manifestly disproportionate to their assets and income, unless the guarantor’s assets, at the time they are called on, enable them to meet their obligation.

This text was recodified in 2016 as Articles L. 332-1 and L. 343-4 of the Consumer Code, before being replaced, for guarantees entered into from 1 January 2022, by new Article 2300 of the Civil Code (code civil) resulting from Ordinance No. 2021-1192 of 15 September 2021. But for all earlier guarantees (and there are still very many in ongoing litigation, as this case shows, where the deed dated from 2011), it is indeed the former text that applies, with its sanction most favourable to the guarantor: complete discharge.

How is disproportion assessed on the day of signing?

The case law is settled: disproportion is assessed by reference to the guarantor’s overall indebtedness. The French Supreme Court recalls this in the judgment commented on here: disproportion “must be assessed taking into account the guarantor’s overall indebtedness, including that resulting from earlier guarantee commitments, provided that these guarantees are not, in whole or in part, extinguished”.

In concrete terms, to measure whether a guarantee is disproportionate, one weighs, on the one hand, the amount of the commitment entered into and, on the other, the guarantor’s assets and income, from which all their debts are deducted: personal loans, mortgages, but also guarantees previously given in favour of other creditors. A director who accumulates personal guarantees over the course of the bank facilities granted to their companies thus sees their available financial capacity shrink with each new commitment.

The judgment of 8 July 2026 adds an important methodological clarification: the amount of these earlier commitments “means the sums remaining due under the principal obligation that they secure”. In other words, neither the face-value ceiling of the guarantee nor a token amount is taken into account: one counts what the guarantor actually risks having to pay in respect of the secured debt on the day of the assessment.

What is the “return to better fortune”?

The guarantor’s discharge is not definitively acquired by the mere fact of the initial disproportion. Former Article L. 341-4 gives the creditor a second chance: if, at the time the guarantor is called on (that is, in practice, on the date of the summons for payment), their assets enable them to meet their obligation, the bank regains the right to rely on the guarantee.

Two snapshots of the guarantor’s assets are therefore taken: one on the day of signing, the other on the day of the call for payment. And it is on the second that the case turned. The Metz Court of Appeal had accepted the initial disproportion; it remained to be determined whether, ten years later, the guarantor had returned to better fortune. To answer in the affirmative, the Metz judges believed they could remove from the guarantor’s liabilities all their earlier “expired” commitments. It is this shortcut that the French Supreme Court censures.

Why is the Metz Court of Appeal censured?

What had the court of appeal held?

The court of appeal’s reasoning rested on an intuitive but legally incorrect equation: a guarantee entered into for a fixed term (72 months, 108 months) would be “expired” when its term arrived, and the guarantor would be “released” from it. The May 2010 commitments, granted for six years, had reached their term in 2016; the 2006 one, for a term of nine years, in 2015; the aval for the 2012 cash-flow credit had followed the expiry of the facility at the end of 2013. On the date of the summons, in 2021, none of this would any longer weigh on the guarantor’s assets, who could therefore, together with their assets and income, meet their obligation of €38,815.33.

What error does the French Supreme Court sanction?

Citing Articles 1134 and 2292 of the Civil Code, in their versions prior to the 2016 and 2021 reforms, and Article L. 341-4 of the Consumer Code, the Commercial Chamber lays down the rule in terms of principle: “in the absence of an express contractual stipulation limiting in time the creditor’s right to sue, the fact that the guarantor is called on to pay after the deadline of their commitment has no bearing on the guarantor’s obligation regarding the claim that arose before that date. This rule applies even where the guarantee is granted to secure a specific debt.”

It then draws the consequence in the present case: “the expiry of the term of a guarantee does not release the guarantor from their obligation to settle claims that arose earlier”, so that the 2010 commitments, “although they had reached their term, had to be taken into consideration in assessing whether [the guarantor’s] assets enabled them to meet the obligations resulting from the guarantee at issue”. By holding otherwise, the court of appeal failed to draw the legal consequences of its own findings: quashing.

The nuance is essential. The term of a fixed-term guarantee ends the obligation to cover (obligation de couverture): debts that arise after that date are no longer guaranteed. But it does not affect the obligation to settle: all debts that arose before the term remain owed by the guarantor, who may be pursued as long as the limitation period has not expired. An “expired” guarantee is therefore not an extinguished guarantee, and as long as it is not extinguished, it weighs on the guarantor’s overall indebtedness.

Obligation to cover, obligation to settle: what is the difference?

This distinction, forged by legal scholars and long established by case law, is the beating heart of the judgment. It deserves to be explained simply, because it determines the fate of many guarantors.

The diagram below summarises the chronology in a typical case:

Applied to the assessment of disproportion, this framework changes everything. When drawing up the statement of the guarantor’s assets, whether on the day the guarantee at issue was concluded or on the day they are called on, all earlier guarantees whose obligation to settle subsists must be entered as liabilities, that is, all those that secure debts that arose before their term and have not yet been paid. Only guarantees that are truly extinguished, because the principal debt has been repaid in full, because the guarantor has paid, or because a cause of extinction specific to them has occurred, drop out of the calculation.

The judgment further specifies that the rule applies “even where the guarantee is granted to secure a specific debt”. It therefore makes no difference whether the commitment secured a specific loan rather than a flow of future debts: the logic is identical, and the aval given to secure a cash-flow credit follows the same analysis.

What is the significance of the judgment for guarantors?

Does the principle apply at the time of conclusion as well as when payment is called for?

The decision’s clearest contribution lies in its field of application. That earlier guarantees that have not been extinguished count in the assessment of disproportion on the day of conclusion had already been affirmed in the case law. The judgment of 8 July 2026 expressly transposes this requirement to the second stage of the text: verifying the return to better fortune on the day the guarantor is called on. The symmetry is perfect: the same overall indebtedness, measured by the same method (the sums remaining due under the secured principal obligations), must be reconstructed at both dates.

For banks, proving the return to better fortune becomes appreciably more demanding. It is no longer enough to point to the arrival of the term of the guarantor’s earlier commitments in order to sweep them from the liabilities: it must also be established that the debts they secured are extinguished, in whole or in part. Yet the burden of proving the return to better fortune rests on the creditor. A bank that produces no evidence on the fate of the facilities secured by the earlier guarantees risks seeing its demonstration rejected.

How can “expired” guarantees be put to use in a case?

For the guarantor being pursued and their counsel, the judgment provides a user’s guide. First reflex: list all the guarantee and aval commitments entered into before the deed at issue, even old ones, even those that have reached their term. Second reflex: determine, for each of them, whether the secured principal debt arose before the term and whether it remained unpaid on the relevant dates, namely that of the conclusion of the guarantee at issue and that of the summons. Third reflex: quantify the sums remaining due on those two dates, because it is this amount, and not the ceiling of the commitment, that swells the overall indebtedness.

In the case decided, what was at stake was far from theoretical: the commitments set aside by the court of appeal secured, among others, a €210,000 loan granted in 2006 and an €80,000 cash-flow credit granted at the end of 2012, in addition to two loans granted in 2010 by another bank. Reinstated among the guarantor’s liabilities, they could tip the assessment of the return to better fortune, and therefore lead to the outright discharge of the guarantor, despite an order to pay “only” €38,815.33 in principal.

How can a guarantor who is being pursued defend themselves?

The judgment of 8 July 2026 is a reminder that a summons for payment served by a bank on a guarantor is never the end of the game. Several lines of defence must be systematically examined, in order.

First, manifest disproportion: for any guarantee entered into before 1 January 2022 towards a professional creditor, former Article L. 341-4 of the Consumer Code (then L. 332-1 and L. 343-4) provides a complete discharge if the commitment manifestly exceeded the guarantor’s assets and income at signing. The personal financial information sheet completed at the time, the tax assessments and the inventory of all earlier commitments, including those that have reached their term but have not been extinguished, as the judgment commented on here confirms, are the key documents for this demonstration.

Next, if the initial disproportion is established, the debate shifts to the return to better fortune: it is for the bank to prove that the guarantor’s assets, on the day of the summons, enable them to meet the obligation. The guarantor has every interest in relying on the persistence of their obligation to settle under their earlier commitments, in line with the judgment of 8 July 2026.

Finally, other grounds deserve examination depending on the case: the formalism of the handwritten statement for deeds subject to former Articles L. 341-2 and L. 341-3 of the Consumer Code, the failure to provide the guarantor with annual information, forfeiture of the right to interest, the duty to warn an unsophisticated guarantor, or the limitation of the bank’s action. Each of these grounds follows its own rules of proof and time limits; their combination makes the difference between a judgment suffered and a discharge obtained.

What should be taken from the judgment of 8 July 2026?

By affirming that a guarantee that has reached its term continues to weigh on the guarantor’s overall indebtedness as long as their obligation to settle subsists, the Commercial Chamber closes a loophole that professional creditors would readily have used to neutralise the sanction of disproportion. The solution, published in the Bulletin, applies at both stages of the analysis (conclusion of the guarantee and call for payment) and extends to guarantees of specific debts as well as to avals.

For guarantors of business loans, often directors who have accumulated personal guarantees over the years, the message is clear: a history of guarantees, even old and “expired” ones, is an evidentiary asset. Properly reconstructed and quantified, it can establish the initial disproportion and defeat the proof of a return to better fortune, and therefore lead to a complete discharge. These cases are won on precision: the dates on which the secured debts arose, the sums remaining due, the chronology of the terms. Support from a lawyer well versed in banking guarantee law makes it possible to turn this landmark judgment into a concrete result.

FAQ: Disproportionate Guarantees and Expired Commitments