ING Direct 2015 Annual Report Download - page 263

Contents
Report of the
Executive Board
Corporate
Governance
Consolidated
annual accounts
Parent company
annual accounts
Other
information
Additional
information
Additional Pillar III information - continued
The relatively low RWA density for Sovereigns and central banks is driven by certain sovereign entities, which are rated between 1-4
and whose exposures receive a regulatory risk weight of nil %.
Changes in risk parameters since last reporting date
The table below shows the changes in risk parameters since last reporting date in percentages. This should be read in conjunction with
the table in the paragraph ‘Disclosure of model outcomes’, above.
2015
2014
Sovereigns
Institutions
Corporate
Secured by
Res. Mortgage
Other retail
Total
Total
Average PD
13%
–36%
–20%
–16%
–12%
–17%
–10%
Average LGD
0%
–1%
2%
–11%
7%
–2%
–5%
READ
14%
–8%
15%
1%
–6%
5%
7%
RWA
–3%
–38%
15%
–13%
1%
–1%
6%
RWA density
–15%
–33%
0%
–14%
7%
–6%
–1%
Includes the AIRB portfolio only; excludes securitisations, equities and ONCOA.
Over the course of 2015 the risk profile of the AIRB portfolio improved with the average PD decreasing with 17%. The PD decrease seen
in the Institutions exposure class was mainly driven by portfolio composition changes with most notable a strong reduction related to
a Deposit Guarantee Scheme exposure.
The increase in the average PD of the Sovereigns portfolio of 13% is mainly a result of increasing customer deposits in Japanese Yens
and consequently higher placements with the Bank of Japan for which the county rating has deteriorated, resulting in not being 0%
risk weighted placement anymore. The LGD increase in the Corporates portfolio is mainly a result of the Commercial Property Finance
LGD model update which now incorporates more conservatism. The improved PD values in the Corporates exposure class improved
with 20% for which the main drivers were the restructuring of a few files within the Real Estate Finance portfolio and an improving risk
profile of the Corporates segment within the Netherlands. This was partially offset by pressure on oil and commodity prices in
combination with economic and political turmoil in Russia and Ukraine which resulted in an upward trend of PD values in mainly the
Structured Finance portfolio partly offsetting the overall downward trend. The PD and LGD improvement in the Secured by Residential
mortgage portfolio is primarily driven by the Dutch mortgage portfolio, where economic conditions improved especially with respect
to house prices and employment in the Netherlands and was accompanied with a more efficient arrears management. On top of the
improving economies in most of the key retail countries, the Residential Mortgage portfolio and the SME portfolio PDs have showed the
most significant improvements.
Disclosure of estimated and actual loss parameters
ING has dedicated AIRB credit risk models per business unit, segment and country. An independent Model Validation department
periodically reviews all AIRB models for compliance including back testing when possible. If a model is considered not to be robust or
the back-testing indicates insufficient performance, than the model is either re-calibrated or re-developed. All model
recommendations from Model Validation department are tracked via iRisk, the same internal database that management uses to
track issues detected by the Internal Audit department, incidents and non-financial risk issues. All significant model changes are
submitted to the ECB and implemented after regulatory approval. On average, 94% of the AIRB credit risk models in the validation
cycle have had ‘No to Remote’ (57%) and ‘Minor’ (37%) model deficiencies.
Changes in AIRB risk parameters in %
ING Bank Annual Report 2015 261