Limitations of Implementing an Expected Credit Loss Model

42 Pages Posted: 18 Jan 2023 Last revised: 16 Jun 2023

See all articles by Jannis Bischof

Jannis Bischof

University of Mannheim - Accounting and Taxation

Rainer F. H. Haselmann

Goethe University Frankfurt - Faculty of Economics and Business Administration; Leibniz Institute for Financial Research SAFE

Frederik Kohl

University of Mannheim - Accounting and Taxation

Oliver Schlueter

Deutsche Bundesbank

Date Written: December 2, 2022

Abstract

The IFRS 9 loan impairment rules require banks to estimate their future credit losses using forward-looking information. We use supervisory loan-level data on German banks' internal rating models to investigate how banks apply their reporting discretion and adjust their lending upon the announcement of the new rules. Our research design exploits a cutoff for the level of provisions at the investment grade threshold based on banks' internal borrower rating. We find that banks required to adopt the new rules assign better internal ratings to the same borrowers compared to banks that do not apply IFRS 9 around this cutoff. This pattern is consistent with the strategic use of the increased reporting discretion inherent to rules requiring forward-looking loss estimation. At the same time, banks also reduce their lending exposure to precisely those borrowers at the highest risk of experiencing a rating downgrade below the cutoff. The lending change thus mitigates some of the negative effects of increased reporting opportunism on banks' crisis resilience.

Keywords: Bank Accounting, CECL, Expected credit losses, IFRS 9, Impairments, Loans

JEL Classification: G01, G21, G28, K23, M41

Suggested Citation

Bischof, Jannis and Haselmann, Rainer F. H. and Kohl, Frederik and Schlueter, Oliver, Limitations of Implementing an Expected Credit Loss Model (December 2, 2022). LawFin Working Paper No. 48, TRR 266 Accounting for Transparency Working Paper Series No. 124, Available at SSRN: https://ssrn.com/abstract=4325220 or http://dx.doi.org/10.2139/ssrn.4325220

Jannis Bischof (Contact Author)

University of Mannheim - Accounting and Taxation ( email )

Mannheim, 68131
Germany

Rainer F. H. Haselmann

Goethe University Frankfurt - Faculty of Economics and Business Administration ( email )

Mertonstrasse 17-25
Frankfurt am Main, D-60325
Germany

Leibniz Institute for Financial Research SAFE ( email )

(http://www.safe-frankfurt.de)
Theodor-W.-Adorno-Platz 3
Frankfurt am Main, 60323
Germany

Frederik Kohl

University of Mannheim - Accounting and Taxation ( email )

Mannheim, 68131
Germany

Oliver Schlueter

Deutsche Bundesbank

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
163
Abstract Views
823
Rank
301,150
PlumX Metrics