PrecisionLender uses your assumptions in order to determine the Risk Adjusted Return on Risk Adjusted Capital (RAROC) which is referred to as ROE in PrecisionLender. An important assumption is the capital associated with a particular risk rating and the amount of loan loss reserves.
In this article we will discuss the methods used to determine capital, loan loss reserves and guarantee factor used when a loan is guaranteed.
- Unmitigated Capital
- Amount of capital needed to reflect operational and market risk.
- There are events where a financial institution has limited ability to make decisions to protect itself.
- A tornado or hurricane that damages a branch
- International shocks that move financial markets
- The literature from the Basel Committee on Bank Supervision provides guidance on how this risk should be measured based on an institution’s gross interest income and average assets.
- Credit Capital
- Amount of capital required based on the credit risk of the loan.
- Will vary based on the risk grade of the loan and its duration.
- Using Basel guidelines credit capital is the amount of funds needed to ensure that a bank has 99.9% confidence it has sufficient reserves for a particular risk grade.
- The suggested method to calculate this, is by using the credit migration of the institution’s loan portfolio.
- Typically examined over a one year period
Calculating Credit Capital
- PrecisionLender includes advisory and consultation services.
- We partner with you in order to conduct an analysis of your credit experience in order to recommend the following assumptions for our clients to consider for adoption:
- Annual Loss (expected loss)
- Credit Capital (unexpected loss)
- Guarantor Factor
For a more detailed explanation on PrecisionLender’s approach in determining Annual Loss, Credit Capital, and the impact of Guarantors on loan profitability in respect to Probability of Default (PD) and Loss Given Default (LGD) assumptions, please see this How To Determine The Capital Calculation.
If you choose to use our advisory and consultation services to help configure your PrecisionLender assumptions, please work with your implementation team to complete the following exercise:
- Take a snapshot of your commercial and agricultural loan portfolio 12 months ago by risk grade.
- This should be based on the number of loans in each risk grade and not the loan amounts.
- Please do not include consumer loans or owner occupied residential mortgage loans, particularly those sold in the secondary market.
- Enter your risk grade descriptions in the "RG Desc" column.
- Enter the number of loans that were in each risk grade in the "# Loans @" column.
- Note: all cells with blue are editable inputs. Also, the dates shown in the sample below are for illustration purposes only, please use current dates if possible.
- Take a snapshot of the same group of loans today to determine what the current risk grades are for the same loans vs. 12 months prior.
- Ignore any new loans that were originated during the interim period.
- For loans that were paid off, refinanced or otherwise withdrawn during the interim period add those to the "Paid Off" column in the matrix.
- Please include any loans that migrated to a loss / charge off status during the period, even if there was only a partial charge off.
- The goal is to identify the loans from one year ago that are still on the books one year later and compare the risk grades between the two points in time.
- Using the columns labeled "To RG 1", etc., enter the number of loans from each risk grade that migrated to this risk grade during the year.
- If you have completed the matrix correctly, the total in the column labelled “# Loans @” should equal the amount in column labelled “Total.” This ensures that the client has accounted for all of the original group of loans.
- Return the completed migration matrix to PrecisionLender along with a copy of your current risk rating / underwriting guidelines or policy.
- Note: We value the protection of your client's Personal Identifying Information. Please do not include any loan level information containing Personal Identifying Information.
Using the Basel guidelines to obtain a 99.9% confidence level PrecisionLender uses the Markov chain approach to determine the appropriate probability of default which is a basis for determining the Credit Capital for each of the financial institution’s risk grade. We examine the percentage of loans within a certain rating that are still in that rating, have moved to a higher rating or have declined to a lower rating one year later. This is followed through until a loan hits a defaulted category (Substandard, OLEM, Doubtful, etc.). The percentage of loans that hit a default level for each year, is measured to determine a probability of default. We then apply the Basel formula for Advanced Approach Banks at a 99.9 confidence level to determine credit capital, this formula is:
Where LGD is loss given default; PD is probability of default, R is Rho which is the asset correlation and (M-2.5) is the Basel adjustment by maturity. We also use this type of approach to determine the loan loss for each year over a 10 year period. Finally there is a calculation as to the effectiveness of a guarantees offered on a loan. The results look similar to the sample below. As noted this is performed for each risk grade:
These results are used to calculate the total economic capital requirement and loan loss for a loan input into PrecisionLender. We also help the client determine the appropriate regulatory capital level. PrecisionLender is RAROC based, thus the denominator of this equation is capital.
The actual method used to determine the dollar capital requirement for a loan is explained in this article: How To Determine Credit Capital Calculation .