Economic Capital
FinPricing offers:
Four user interfaces:
- Data API.
- Excel Add-ins.
- Model Analytic API.
- GUI APP.
Financial business is exposed to many types of risks due to the nature of business. To guard against the risk, financial institutions must hold capital in proportion to the potential risk. Market risk economic capital (EC) is intended to capture the value change due to changes in market risk factors. It is an internal capital reserve to cover unexpected loss due to market movement.
Market Risk Economic Capital (EC) is an expression of the capital requirement for business within the Capital Markets Trading Product Framework, relative to the economic risks in its underlying business activities. EC is intended as a forward‐looking estimate of the Maximum Potential Loss in economic (or market) value that could be incurred over a specified time interval with a defined confidence level.
In accordance with the Economic Capital Corporate Standard, EC should be estimated for a 1‐year horizon at a 99.95% confidence level. Risk metrics that are inputs to the EC are calculated at 99% (or 99.9%) confidence level and at various horizons (1 day, 2 weeks, or 1 month). As such, the risks must be scaled to the EC confidence level and risk horizon.
A base EC measurement that reflects the EC requirement for current risk positions under prevailing market conditions. This estimate assumes that both the portfolio composition and prevailing market environment will persist over the EC forecast period. The base EC should be calculated for each business date.
A stressed EC measurement reflects the current risk positions under a stressed market environment. Stressed EC is calculated at the same frequency as stressed VaR, on a weekly basis at least.
In the current Market Capital at Risk (MCaR) model framework, the P&L distribution is driven by a single factor stochastic volatility model. Due to the lack of data for calibration, this model presents some challenges for model parameter selection.
Market risk EC calculations are calibrated to a 99.95% confidence and a 1‐year horizon. Risk metrics that are inputs to the EC are calculated at 99% or 99.9% confidence levels and at various horizons (e.g., 1 day, 2 weeks, 1 month, or 1 year for IRC). As such, the risks must be scaled to the EC confidence level and risk horizon.
Line management is responsible for the risk/return performance of their business unit(s). As such, they are primarily accountable to determine what action should be taken if a business experiences adverse revenue performance. Risk management has a secondary accountability to identify where risk/return profiles should be flagged, or where revenue performance requires a management action plan.
The liquidity horizon is the estimated time to close or hedge the market risk positions. Liquidity horizons are an input into various metrics used within Capital Markets Trading. To ensure consistency in approach and assumptions, the process below should be followed to estimate liquidity horizons.
Liquidity assessments may be reassessed during the year if market conditions significantly change and/or the line of business (LOB) starts trading new products or unwinds a certain strategy. Once approved, the liquidity assessments are maintained in the master EC application, which is owned and maintained Market Risk.
The scaling approach computes an aggregate scaling factor that maps VaR(1day,99%) to VaR(1year,99.95%) by capturing the liquidity period and management intervention effects. As such, the scalar is dependent on liquidity and management assumptions.
The sum of the liquidity horizon and management action horizon is denoted by T and the aggregate, T-dependent, scaling factor by SC(t,p; 1year,99.95%), where t is the horizon at which the risk metric is calculated (normally one day or 10 days) and p = 99%. We assume that the portfolio risk possesses the same tail behavior. More exactly, we work under the hypothesis that the loss distribution tail behavior of up to and including time T and after time T when liquidity and management effects (referred to as the T-effect hereafter) come into play is the same.
References |