Definition of Default

Regulations

In 2016, the European Banking Authority published the final report on the Guidelines on the application of the definition of default. This guideline, commonly referred to as the guidelines on the new Definition of Default poses a challenge for banks in multiple ways. The first challenge is that their internal processes need to be adjusted to account for the new definition of default. For most banks, a default event is a trigger to take additional actions to ensure that credit risk remains acceptable. The second challenge is that the complexity of the new definition of default is a large burden on IT implementation capacity. All IT processes need to be updated to incorporate the new definition of default. The third challenge for banks is that their credit risk models need to be recalibrated or redeveloped to incorporate the new definition of default. That is, IRB credit risk models have been calibrated to the long-run average default rate according to the old definition of default, which changes after one historically applies the new DoD. Furthermore, since IFRS9 credit risk models are backtested on their ability to accurately predict point-in-time default rates, these Loan Loss Provisioning models should also be recalibrated.

RiskQuest has considerable experience in incorporating the new definition of default in credit risk models of various banks. There are two approaches to implementing the new definition of default into the IRB models, both of which RiskQuest has applied in practice. The first approach is the One-Step Approach, where the new DoD is directly incorporated in the models, using full redevelopment ensuring compliance with all other recent published guidelines for IRB credit risk models together with historically backscored new Definition of Default. The second approach to implementing the new DoD is the Two-Step Approach. The core of the Two-Step Approach is that the new definition of default is implemented before the required date of the 1st of January 2021, after which the IRB models can be updated partly using the data collected from the implemented new DoD.

RiskQuest’s role in the incorporation of the new DoD in credit risk models starts from creating a methodological framework and way-of-working that can be consistently applied across models. This covers both IRB and IFRS9 credit risk models and incorporates internal policies and standards as well as regulation. It is of great importance that this methodological framework and way-of-working is broadly endorsed by all stakeholders, which RiskQuest ensures by working with various relevant departments in the bank.

Recalibration

Regardless of the approach to incorporate new DoD in the credit risk models, the second step is to create a list of defaults under the new Definition of Defaults by applying the default definition retrospectively to the historical portfolio. Given the updated days past due counter, the newly introduced unlikely to pay (UTP) triggers, the newly introduced UTP indicators and the possibility to introduce additional UTP triggers this is no easy feat, requiring multiple data sources and computations. RiskQuest’s experience in historically backscoring new DoD greatly helps in efficiently and correctly computing defaults under new DoD historically.

The last step in the process is to incorporate the backscored new DoD in the credit risk models. Dependent on if the One- or Two-Step Approach is followed, the models are recalibrated or redeveloped, fully incorporating the new Definition of Default. RiskQuest recalibrates or redevelops the models together with the internal modelling teams, ensures alignment with stakeholders, communicate with model validation and regulatory institutions. The incorporation of new DoD in the models ends with approval from the regulatory institutions in case of the IRB models and model validation in case of the IFRS 9 models. Given the substantial experience RiskQuest has in dealing with the new Definition of Default, it is guaranteed that the new Definition of Default is correctly incorporated in the credit risk models. Are you interested in working on these kind of topics, feel free to look at our vacancies.