New Definition of Default—Recalibration of Credit Risk Models Using Bayesian Approach

Aneta Ptak-Chmielewska,Paweł Kopciuszewski

doi:10.3390/risks10010016

Aneta Ptak-Chmielewska, Paweł Kopciuszewski

Open Access

https://doi.org/10.3390/risks10010016

Copy DOI

Abstract

After the financial crisis, the European Banking Authority (EBA) has established tighter standards around the definition of default (Capital Requirements Regulation CRR Article 178, EBA/GL/2017/16) to increase the degree of comparability and consistency in credit risk measurement and capital frameworks across banks and financial institutions. Requirements of the new definition of default (DoD) concern how banks recognize credit defaults for prudential purposes and include quantitative impact analysis and new rules of materiality. In this approach, the number and timing of defaults affect the validity of currently used risk models and processes. The recommendation presented in this paper is to address current gaps by considering a Bayesian approach for PD recalibration based on insights derived from both simulated and empirical data (e.g., a priori and a posteriori distributions). A Bayesian approach was used in two steps: to calculate the Long Run Average (LRA) on both simulated and empirical data and for the final model calibration to the posterior LRA. The Bayesian approach result for the PD LRA was slightly lower than the one calculated based on classical logistic regression. It also decreased for the historically observed LRA that included the most recent empirical data. The Bayesian methodology was used to make the LRA more objective, but it also helps to better align the LRA not only with the empirical data but also with the most recent ones.

Highlights

Following the financial crisis, European Banking Authority (EBA) has established tighter standards around the definition of default (Capital Requirements Regulation—CRR Article 178, EBA/GL/2017/16) (EBA 2017) to achieve a higher comparability and consistency in models used for credit risk measurement and procedures and capital frameworks across banks and financial institutions
The initial deadline for the implementation was the 1st of January 2021. These deadlines were under discussion with European Central Bank (ECB) by many banks, and they are under further revision by ECB due to COVID—19 circumstances (EBA 2016b)
We provide a summary of the main changes implied by the new definition of default (DoD) and its impacts for credit risk modeling with specific emphasis for the institution under study in this article

Summary

Introduction

EBA has established tighter standards around the definition of default (Capital Requirements Regulation—CRR Article 178, EBA/GL/2017/16) (EBA 2017) to achieve a higher comparability and consistency in models used for credit risk measurement and procedures and capital frameworks across banks and financial institutions. Banks need to update their risk management practices and support pricing and accounting decisions related to the expected credit loss methodology (according to International Financial Reporting Standards—IFRS 9) (EU 2016) and those related to capital requirements models (according to Internal Ratings Based Approach -IRB and Internal Capital Adequacy Assessment Process—ICAAP) (ICAAP 2018) These extensive and very detailed guidelines often challenge information technology (IT) infrastructure, processes, data engines and data analytics, commonly used model risk platforms, model implementation and execution and automated solutions (BCBS 2013). The Bayesian method was applied to anticipate this simulated and empirical data for a basic credit risk measure: probability of default (PD) re-calibration Such example of Bayesian approach utilization in modeling and calibration of credit risk parameters is promising for trying to incorporate such approach in calibration of PD, using parameters estimated on empirical data for new DoD as prior information. Final section contains concluding remarks and suggestions for future research

Implications of the New Definition of Default—Literature Review

Proposed Recalibration and Re-Development Methods for Credit Risk Parameters

Probability of Default PD—Model Recalibration

PD Recalibration—Application on Real Data

Findings

Conclusions and Discussion