ING Direct 2009 Annual Report Download - page 246

Model Disclosures
Measurement
Economic capital for market risk is calculated using internally developed methodologies with a 99.95% confidence interval and a horizon
of one year, which represents extreme events and ING’s target rating. The Economic Capital for market risk for non trading portfolios is
calculated for each risk type, while for trading portfolios it is calculated on a portfolio level. The calculations for Economic Capital market
risk include Real Estate risk, foreign exchange rate risk, equity price risk, interest rate risk and model risks.
Real Estate price risk includes both the market risks in the investment portfolio and the development risk of ING Real Estate. The Real
Estate price risk for ING Real Estate is calculated by stressing the underlying market variables. The stress scenarios at a portfolio level take
into account all diversification effects across regions and Real Estate sectors. Also, the leverage of participations in the Real Estate
investment funds is taken into account.
For the Real Estate development process, in addition to market sale price risk, the risk drivers of market rent, investor yield and
construction delays are taken into account. Furthermore the risk model differs for each development phase (i.e., research, development,
and construction) to appropriately reflect the risk taken in each phase. Using correlations, all risk drivers, and stages are used to calculate a
possible market value loss representing the Economic Capital for market risk for the development portfolio.
For the direct market risks, the actual VaR (measured at a 99% confidence interval, a one day holding period and under the assumption of
an expected value of zero) of the trading and non-trading portfolios is taken as a starting point for the Economic Capital calculations for
market risk. To arrive at the Economic Capital for market risk, a simulation based model is used which includes scaling to the required
confidence interval and holding period. In determining this scaling factor, several other factors are also taken into account like the
occurrence of large market movements (events) and management interventions.
The economic capital for the equity investments is calculated based on the ECAPS system. Using Monte-Carlo simulation, the model
generates 20,000 possible ‘states-of-the-world’, by randomly simulating all risk drivers simultaneously. For each state-of-the-world, the
market value is recalculated and the 99.95% worst-case change in market value is the Economic Capital level.
Economic Capital for market risk for the mortgage portfolios within ING Retail Banking and ING Commercial Banking is calculated for
embedded option risk (e.g. the prepayment option and offered rate option in mortgages). The embedded options are hedged using a
delta-hedging methodology, leaving the mortgage portfolio exposed to convexity and volatility risk. The Economic Capital model for
market risk is based on the estimated 99% confidence adverse interest rate change.
While aggregating the different Economic Capital market risk figures for the different portfolios, diversification benefits are taken into
account as it is not expected that all extreme market movements will appear at the same moment.
The nature of market risk Economic Capital, evaluating the impact of extreme stress with a 99.95% confidence level, can sometimes be
difficult to evidence in a statistical sound manner with the available historical data. The Economic Capital figures disclosed by ING Group
are a best effort estimate based on available data and expert opinions.
OPERATIONAL RISK
Operational risk is the risk of direct or indirect loss resulting from inadequate or failed internal processes, people and systems or from
external events. It includes the risk of reputation loss, as well as legal risk, whereas strategic risks are not included. While operational risk
can be limited through management controls and insurance, operational risk incidents may have a substantial impact on the profit and loss
account of financial institutions.
The capital model, an actuarial model, consists of a combination of three techniques:
Loss Distribution approach (LDA), which applies statistical analysis to historical loss data;•
Scorecard approach, which focuses on the quality of risk control measures within a specific business unit;•
‘Bonus/Malus’ approach, which focuses on the actual operational incidents of a specific business unit.•
Risk management (continued)
2.1 Consolidated annual accounts
ING Group Annual Report 2009
244