## ICAAP Compliance: Credit Risk: Stress Test: Simple Sensitivity Analysis – Increase & Shift in NPL

*in*Basel II and III

Earlier we had looked at one way of stress testing credit risk under the Internal Capital Adequacy and Assessment Process (ICAAP). We will look at some simple sensitivity analysis for credit risk below, focusing on stressing non-performing loans (NPL).

# Simple Sensitivity Analysis

This measures the change in the value of the portfolio for the shocks of various degrees to different independent risk factors while the underlying relationships among the risk factors are not considered.

For credit risk we will look at the following risk factors for stress testing:

- forced sale value of collateral,
- non-performing loans (NPLs)

The stress test for credit shock assesses the impact of increase in the level of non-performing loans of the bank. This involves three types of shocks:

## Increase in NPL:

In this case we analyze the impact of 5%, 10% and 20% increase in the total NPLs downgraded to loss category having 100% provisioning requirements. 100% provisioning means that the bank will not get the loans back and they should be treated as a loss. The methodology used will be as follows:

### For 5% increase in total NPLS:

Increase in NPLs = Total NPLs * 5%

Increase in Provision = Increase in NPLs * 100%

Tax Adjusted Loss = Increase in Provision * (1 – Tax Rate)

Revised Capital = Total Capital – Tax Adjusted Loss

Revised risk weighted assets = Risk Weighted assets – Tax Adjusted Loss

Revised CAR (%) = Revised Capital / Revised risk weighted assets

Fall in CAR (% age points) = Original CAR – Revised CAR

*Note: The same procedure shall be followed for 10% and 20% increases in NPLs.*

### Shift in NPL:

In this case we increase the provisioning. We would analyze the impact of 50%, 80% and 100% downward shift in the NPL categories. For the first level of shock, 50% of substandard shall be categorized under doubtful and 50% of doubtful shall be added to the loss category. Then the Forced Sale Value of mortgaged collateral under each category is listed down. This is the current market value of the mortgaged collateral. The Forced Sale Value (FSV) of mortgaged collateral is subtracted from the corresponding NPL category, which is then multiplied by the %age provisioning for each category to get the provisions.

The provisioning for each NPL category is as follows:

- Substandard category: 25% provisioning is required.
- Doubtful category: 50% provisioning is required.
- Loss category: 100% provisioning is required.

The impact of shift in 50% NPLs to next categories with no change in total NPLs shall be accounted for as follows:

- Weighted NPLs (no shift) = Original NPLs in substandard category * 25%(provisioning for sub standard category) + Original NPLs in doubtful category * 50%(provisioning required for doubtful category) + Original NPLs in loss category * 100%(provisioning required for Loss category)
- Weighted NPLs (after shift) = Shifted NPL in substandard category + Shifted NPL in doubtful category + Shifted NPL in Loss category
- Shifted NPL in substandard Category = [Original NPLs in substandard category * 50%] * 25% (provision for substandard category)
- Shifted NPL in Doubtful Category = [Original NPLs in substandard category * 50%+ Original NPLs in Doubtful Category * 50%] * 50% (provision for doubtful category)
- Shifted NPL in Loss Category = [Original NPLs in Doubtful Category * 50% + Original NPLs in Loss Category] * 100% (provision for loss category)
- Increase in Provision = Weighted NPLs (after shift) – Weighted NPLs (no shift)
- Tax Adjusted Loss = Increase in Provision * (1 – Tax Rate)
- Revised Capital = Original Capital – Tax Adjusted Loss
- Revised risk weighted assets = Original risk weighted assets – Tax Adjusted Loss
- Revised CAR = Revised Capital / Revised risk weighted assets
- Fall in CAR (% age points) = Original CAR – Revised CAR

*Note: The same procedure shall be followed for 80% and 100% shifts in the NPLs to the respective downward category.*

We have discussed two techniques for carrying out a simple sensitivity analysis for credit risk. We will look at an additional approach below considering percentage falls in the Forced Sale Value (FSV) of mortgaged collateral.