Modelling Corporate Probability of Default : A Possible Supervisory Benchmark Model

Szenes, Márk and Dabi, Zsófia (2020) Modelling Corporate Probability of Default : A Possible Supervisory Benchmark Model. FINANCIAL AND ECONOMIC REVIEW, 19 (3). pp. 52-77. ISSN 2415-9271 (print); 2415-928X (online)

Preview

Text fer-19-3-st2-szenes-dabi.pdf Download (796kB) | Preview

Abstract

In recent years, supervisory bodies around the world have lost some of their confidence in the estimations of credit risk parameters at banks applying the internal ratings-based methodology. Supervisory experience shows that differences in risk metrics and ultimately in regulatory capital requirement levels stem primarily from inconsistencies in the modelling techniques applied and the various methodological approaches, rather than from any actual differences between the inherent risks of bank portfolios. To avoid this unwanted effect, in its supervisory review of banks’ internal capital adequacy assessment process, the Central Bank of Hungary (Magyar Nemzeti Bank, MNB) aims at specifying the necessary capital requirements by developing and applying harmonised benchmark models. This study shows how it is possible to estimate a probability of default (PD) for corporate portfolios, which is based on large banks’ corporate default rate data series and available corporate financial data, uses a harmonised methodology that factors in differences between the credit quality ratings of various customers, and is suitable for the supervisor’s calculation of the capital requirement for any given bank. Nonetheless, there may also be other factors in addition to individual financial data (e.g. qualitative expert elements, sector information) that may affect credit quality; identifying these may be one of the objectives of benchmark model development.

Actions (login required)

Edit Item