assigned to each rating grade for its IRB portfolios within the calibration

Assigned to each rating grade for its irb portfolios

This preview shows page 161 - 162 out of 222 pages.

assigned to each rating grade for its IRB portfolios within the calibration process. Calibration is performed individually for each rating method. PD values reflect the 12-month probability of default based on long-term average default rates. In addition to the PD values, the bank assigns margins of conservatism depend- ent on the granularity of portfolios and relevant data history. Calibration of PD values is validated annually in line with all the rating methods validations. Internal ratings take into account all available essential infor- mation for the assessment of counterparty default risk. For non- retail borrowers, internal ratings take into account financial strength of the counterparty, possibility of external support, com- pany information, and external credit history information, where available. For the wholesale segment, internal ratings also take into account market information such as access to capital markets linked to external ratings or credit spreads. The willingness of the market to provide funds to the counterparty can be derived from these variables. For retail clients, internal ratings are based main- ly on behavioural and application scoring, but they also utilise demographic and financial information, supplemented by credit bureau information, where available. Rating ceiling rules on credit quality are applied based on membership in a group of economically related entities and country of main economic activity. Country ceilings are applied for cross-border financing facilities. Internal teams of specialists (‘Competence Centers’) provide internal rating models and risk parameters and develop them further. Rating development follows an internal methodology that is formalised into a group-wide methodology and documentation standard. Rating models are developed based on relevant and most-accurate data covering the respective market. In such way, Erste Group has established highly predictive rating models covering its entire core region. All scorecards, whether retail or non-retail, are regularly validat- ed by the central validation unit based on group-wide standard methodology. Validations are provided using statistical techniques in respect to default prediction performance, rating stability, data quality, completeness and relevancy, as well as, last but not least, the review of documentation and user acceptance. The results of this validation process are reported to the management and regu- latory bodies. In addition to the validation process, the Group applies a monthly monitoring process on the performance of rating tools, reflecting the month-to-month new defaults and any early delinquencies. A Holding Model Committee is established as an elementary steering and control body for the model development and mainte- nance process. The Holding Model Committee reports to the CRO Board. All new models and modifications of existing mod- els in the group (rating models and risk parameters), as well as methodology standards, are reviewed by the Holding Model Committee. The Holding Model Committee ensures group-wide
Image of page 161
Image of page 162

You've reached the end of your free preview.

Want to read all 222 pages?

  • Fall '13
  • Capital requirement, Tier 1 capital, Erste Group Bank AG, Erste Group

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

Stuck? We have tutors online 24/7 who can help you get unstuck.
A+ icon
Ask Expert Tutors You can ask You can ask You can ask (will expire )
Answers in as fast as 15 minutes