Financial institutions continue to prepare for the transition to accounting standard ASU 2016-13, commonly known as CECL. The new accounting standard requires institutions to recognize the full life of loan expected credit loss at the time of origination. As implementation dates creep closer, most are busy gathering data, building models or selecting 3rd party partners, and conducting parallel tests with current ALLL calculations. The results of these parallel runs are allowing institutions to determine which methodologies (DCF, Vintage, PD/LGD, Migration, WARM, etc.), work best for their situations. Regardless of which methodologies are ultimately utilized, an accurate life of loan calculation forms the foundation for accurate loss forecasts.
Simplistic Approach – Attrition Analysis
A simplistic approach to calculating the loan portfolio’s life is through an attrition analysis. Under this approach, the number of loan accounts that pay off in a given period as a percentage of the total number of active loans is calculated. The result is then annualized to determine the percentage of the portfolio that closes each year. This annual attrition rate is then converted to “life of loan” by calculating how long it would take for 100% of the portfolio to pay off. As an example, if the average annual attrition rate is 20%, then it would take 5 years for the entire portfolio to pay off, and 5 years would then be used as the life of loan assumption. The attrition analysis should be performed for multiple historical periods so that an average rate can be used. Loan prepayments are inherently built into the attrition analysis, and therefore don’t need to be calculated separately.
Robust Approach – Cash Flow Weighted Average Life Analysis
Weighted average life (WAL) is simply a loan’s contractual principal payback weighted by the month in which the payback occurs. This calculation is best performed at the instrument level, and while a spreadsheet-based approach may sufficiently handle the calculations, it most likely will make the process onerous and susceptible to errors. A better alternative is to leverage existing ALM software that is designed to handle cash flow analysis. For most ALM software applications, if proper data is imported into the system, principal cash flow reports are easily generated. Important data fields include:
- Payment Type (P&I, Interest Only, etc.)
- Payment Amount
- Origination Date
- Maturity Date
- Amortization Term
- Interest Rate
- Payment Frequency
- Accrual Method
Once the principal cashflow of individual loans is calculated, it can be grouped together into your various CECL loan pools and the weighted average life of the cashflows can be calculated. The following is an example of the WAL calculation on an $830M loan portfolio before applying prepayment assumptions.
Prepayment Rates Speed Up Cash Flow
In the example above, the calculated WAL is 7.61 years when a 0.00% constant prepayment rate (CPR) is applied. This highlights the importance of bank specific loan prepayment assumptions and the impact they have on weighted average life. Consistent with ALM best practices, institution specific prepayment speeds should be calculated and utilized. Prepayment analysis may be performed at the loan pool level but is more accurate when determined at the underlying instrument level. A manageable process used to determine prepayment speeds is outlined below.
- Determine a lookback period (12 months or longer).
- Calculate the contractual principal paydowns from the beginning of the lookback period forward to the current period.
- Calculate the actual principal paydowns on the same loan accounts used in step 2, over the same period.
- Calculate the difference between the actual paydowns and the contractual paydowns as a percentage of loan balances at the beginning of the lookback period; this additional paydown represents the prepayment rate.
The prepayment analysis should be routinely performed to keep the assumptions current. The graphics below illustrate the significant impact the prepayment speeds have on weighted average life.
As the industry migrates to the calculation of expected losses over the “life of the loan”, the estimated life of the loan portfolio becomes is critical. If an institution over or under estimates the life of its loan portfolio, it will also over or under estimate the magnitude of its expected losses. While there are simplistic methods, like an attrition analysis, that can be used to estimate the life, a more robust cash flow approach using institution specific prepayment rates will better satisfy auditors and regulators, and lead to better, more accurate estimated losses under CECL.
For more information about ProBank Austin’s CECL AdvisorPRO® services, please visit www.probank.com/what-we- do/financial-management/cecl-services/.