EBA’s Revised Definition of Default

On July 2nd, the European Banking Authority (EBA) published a Consultation Paper proposing amendments to its 2016 Guidelines on the application of the definition of default (DoD). As part of the consultation process, open until 15 October 2025, the credit risk specialists at Zanders will submit a formal response, leveraging our extensive experience in DoD regulation and implementation.

In this article, we share our perspective on three of the EBA’s proposed amendments, focusing on the potential impact and implementation challenges for institutions:

• We expect that a shorter probation period for forbearance measures (that only alter the repayment schedule leading to a NPV loss not greater than 5%) are expected to provide incentives for banks to opt for those types of measures rather than the most sustainable ones.

• We recommend the EBA to implement EU wide DoD guidelines they considered for payment moratoria (similar to the one for Covid), whereas the EBA proposes not to. Zanders would approve permanent moratoria guidelines, as it clarifies if governmental moratoria introduced for climate risk related natural disasters should be regarded as forbearance.

• We are oncerned that the proposal to consider material arrears on non-recourse factoring exposures up to 90 (instead of 30) DPD as technical past due situations could result in an undesired increase in the percentage of IFRS stage 1 exposures migrating directly to stage 3 (impairment).

The following chapters elaborate on these three proposed amendments in more detail.

Forbearance

The first amendments addressed in the EBA’s consultation paper (CP) are related to forbearance. The supervisory authority explains that an increase of 1% threshold for a diminished financial obligation (DFO) to 5% was considered for certain forbearance measures. This follows from the European Commission’s mandate that the update of the EBA guidelines on DoD“… shall take due account of the necessity to encourage institutions to engage in proactive, preventive and meaningful debt restructuring to support obligors.”¹

In the EBA’s current DoD guidelines (DoD GL), a forbearance measure leading to a 1% or more DFO results in a default classification, which could discourage institutions from applying these measures. However, the CP purposefully proposes to exclude an increase to a 5% DFO threshold, since institutions can already implement strict(er) forbearance definitions (i.e. for concession, financial difficulty) to prevent undue default classifications. Instead, the EBA proposes to shorten the probation period from 12 to 3 months for forbearance measures that: (1) only lead to suspensions or postponements and not e.g. changes to the interest rate or exposure amounts and (2) leading to less than 5% DFO loss.

This treatment will likely incentivize institutions to choose forbearance measures in scope of the shorter probation period, rather than the ones that would be optimal for a “sustainable performing repayment status” of the obligor. The latter would be in line with the EBA’s own requirements on the management of forborne exposures (Par. 125 EBA/GL/2018/06). Furthermore, the fact that the EBA does not set the “predefined limited period of time” for the measures in scope could lead to RWA variability, as some institutions may apply the shorter probation period to longer duration forbearance measures than others. For example, if Bank A sets the limited period of time to 6 months, they can apply the shorter probation period more often compared to Bank B, which sets the period of time at 3 months. Finally, it appears as if the proposal of the banking authority aims at favouring granting forbearance measures in scope to obligors with short-term (rather than structural) financial difficulties. That is, the EBA explains that the forbearance measures in scope of the shorter probation period treatment “… would most likely be viable for obligors in temporary financial difficulties”. The shorter probation period would then lead to a return to a performing status earlier for obligors to which the forbearance measures in scope are extended, which leads to a better RWA for these obligors. Alternatively, a distinct probation period (or even higher DFO threshold) could be proposed for obligors in short-term financial difficulties, as defined in Paragraph 129(A) of the EBA’s guidelines on Management of Forborne Exposures. This would also achieve the EBA’s goal, without influencing institutions’ decision about which forbearance measure to apply.

It should be mentioned that while a large RWA impact is not anticipated from the establishment of a distinct probation period, there will likely be a significant implementation burden associated with the change. This is because, as multiple forbearance measures are usually adopted in tandem, different probation periods must be traced concurrently. The implementation of this modification would need to be retroactive, though, as credit risk models will need to be recalculated using adjusted historical data in order to account for this change. In the past, retroactively modifying the probationary period has proven to be a time-consuming and expensive problem.

Legislative payment moratoria

In light of the COVID-19 crisis, the EBA published guidelines in 2020 on handling payment moratoria introduced by governments as a means of financial aid in the context of forbearance. For certain COVID-19 measures allowing e.g. a grace period in scope of the guidelines, EBA/GL/2020/02 and amendments in .../08 and …/15, would not in itself require institutions to classify the exposures as forborne.

Even though the EBA considered introducing guidelines for potential future moratoria, the CP proposes against these changes. As one of the arguments against new moratoria guidelines, the EBA remarks that moratoria in itself will not result in DFO loss of more than 1%, hence not leading to defaults. The EBA implies that introducing new moratoria guidelines would therefore be obsolete. The EBA is also worried about RWA variability that might arise if governments declare legislative moratoria for crises in their jurisdictions. That is, the EBA expects that intra-EU comparability of RWA across institutions, might be compromised.

Adding the considered guidelines describing when moratoria should lead to forbearance in the amended DoD GL is advisable, even though the EBA proposes in the CP to remove them. Zanders challenges that guidelines describing when moratoria do not lead to forbearance would not be necessary, because the 1% DFO threshold will not be met. That is, Zanders highlights moratoria guidelines would still decide when the forborne status should be assigned to exposures if moratoria are applied. This forborne status impacts the default status later on, both for performing and defaulted exposures. The reason is that if performing forborne exposures become 30 days past due within 24 months after receiving the forborne status, a defaulted status should be assigned. If the moratoria do not lead to a forborne status, these exposures should default after becoming 90 days past due on a material amount instead. Furthermore, for defaulted exposures, it is important to understand when moratoria result in the forborne status and when they do not. That is, in order for a forborne defaulted exposure to go out of default, a substantial payment and an extended cure period are needed. Zanders would therefore be in favor of EBA guidelines that specify when moratoria should result in a forborne status and when this is not necessary.

As for the RWA variability, as self-identified by the EBA, stringent criteria could be introduced prescribing what moratoria are in scope of the amended DoD GL. As described by the EBA as well, in light of climate risk related natural disasters, payment moratoria could occur more often as a governmental means of financial aid. In contrast to ad hoc rules for each specific crisis, such as observed during the COVID-19 pandemic, Zanders contends that permanently applicable moratoria instructions in the updated DoD GL will eventually lead to a more stable RWA impact when economic or natural catastrophes occur.

Days past due for non-recourse factoring

Paragraph 23(D) of the current version of the DoD guideline stipulates that in the specific situation of non-recourse factoring for which the arrears materiality threshold is breached, but none of the receivables is more than 30 days past due (DPD), should be treated as a technical past due situation. Non-recourse factoring refers to the situation where the institution (e.g. a bank) has bought receivables from its client (e.g. service provider) owed by the debtor (e.g. service consumer). The idea behind the 30 DPD is that the DPD counter might continue to increase due to a consecutive overlap in non-payments of invoices, lengthy administrative processes, and a low degree of control of the institution over the invoices.

The CP proposes to allow for up to 90 DPD to be considered technical past due situations, in correspondence to the industry requesting the EBA to be more lenient in the DoD guidelines for non-recourse factoring. This is motivated by the fact that many corporates have at least one invoice past due more than 30 days, while being rated investment grade.

Although Zanders understands  corporates’ need for more leniency, allowing for up to 90 DPD to be recognized as technical past due could make stage 2 obsolete for IFRS provisioning models. That is, if material arrears on non-recourse factoring exposures should be considered technical past due for situations up to 90 DPD, the said exposures will move from 0 DPD to 91 DPD in one day. The additional lenience would break the desired flow of exposures transitioning from IFRS stage 1 (performing), first towards stage 2 (significant increase in credit risk), before going to stage 3 (credit impaired). This stage migration effect could be mitigated by another stage 2 trigger: forbearance. However, the institution cannot apply forbearance measures to a sold invoice that is due to the institution’s client, rather than due to the institution itself. Therefore, as a stage 2 trigger, forbearance cannot compensate for the lack of the30 DPD in the particular scenario of non-recourse factoring risks.

Zanders proposes to find a balance between leniency on DoD guidelines and stage migrations, by increasing the 30 days threshold. The proposed number of days should be based on an analysis of non-recourse factoring portfolios from a representative sample of supervised institutions. This analysis should then strike a balance between the average observed days past due of invoices sold on the one hand and the representativeness of IFRS stage transitions on the other hand. Zanders is convinced that amending DoD GL based on this analysis will prevent the undesired impact on IFRS provisioning models and will better fit European corporate invoicing practice.

Conclusion

In this post we analysed 3 proposed amendments from the published Consultation Paper, in which the European Banking Authority (EBA) proposes amendments to its 2016 Guidelines on the application of the definition of default (DoD).  Alternatives are suggested for all 3 proposed amendments as the proposed amendments leave room for improvements .

Reach out to our experts John de Kroon and Dick de Heus, if you are interested in getting a better understanding of what the proposed amendments mean for your credit risk portfolio.

We monitor the progress of the Consultation Paper in the future. Keep a close eye on our LinkedIn and website for more information, or subscribe to our newsletters here.