Controlling Loan Portfolio Risk Using Credit Migration Models
Credit union executives struggle to determine the proper amount that needs to be set aside in their Allowances for Loan and Lease Losses (ALLL).
The Office of the Comptroller of the Currency has issued a series of Policy Statements intended to guide financial institutions in the calculation of the Reserve for Loan Losses consistent with GAAP.
The most recent Policy Statement, published in 2006, identifies Credit Unions and the NCUA as subject to its provisions.
In this policy statement, two specific directives are included that address calculation of the reserve for individually identified loans.
- “For purposes of this policy statement, the term “estimated credit losses” means an estimate of the current amount of loans that is probable the institution will be unable to collect given facts and circumstances of the evaluation date.”
- “Under FAS 114, an individual loan is impaired when, based on current information and events, it is probable that a creditor will be unable to collect all amounts due according to contractual terms of the loan agreement. It is implicit in these conditions that it must be probable that one or more future events will occur confirming the fact of loss. Thus, under GAAP, the purpose of the ALLL is not to absorb all the risk in the loan portfolio, but to cover probable credit losses that have already been incurred.”
Thompson Consulting and Training (TCT) recently initiated a longitudinal study to quantify the probability of actual loss for loans that become 60 or more days delinquent. Loan data from client credit unions is gathered monthly. The study so far has covered a time period of several years. The study is ongoing. As with any ongoing study, findings may vary over time.
But, here is what has been observed to date:
- In a normal environment, with consistent collection activity, between 25% and 35% of unsecured loan balances that become 60 or more days delinquent are eventually charged off
- 80% to 90% of total delinquencies and charge-‐offs are attributable to loans that have impaired FICO Scores (dropped 2 or more grades)
- Less than 10% of total delinquencies and charge-‐offs are attributable to loans with an unchanged FICO score
- The most effective way to manage charge-‐offs is to identify and control loans associated with declining FICO scores
- The largest segment of delinquent loans comes from loans that were originally “B” or “C” grades and subsequently dropped to “D” or “E”
- The most valid indicator of impending delinquency and loan loss for any loan is a declining FICO score
From this study and other research, TCT concludes that an empirical, statistically validated Credit Migration model which includes a reporting system that identifies loans (individually and categorically) that are experiencing declining FICO scores will significantly contribute to:
- Correct placements to ALLL
- Controlling delinquencies and charge-‐offs more effectively than relying on other reporting systems or subjective methods
- The most up-‐to-‐date and effective lending policies and procedures including Concentration Risk policies
- Insight into lending problems that lie ahead and corrective actions needed
TCT is based in Boise, Idaho. Dr. Randy Thompson, CEO of TCT, designs his training and management tools using empirically based, statistically validated methods and research. TCT has provided services to credit union professionals for 25 years. TCT’s clientele include credit unions, credit union trade associations, and federal and state regulatory agencies. TCT can be reached at (208) 939 8366 or tctconsult.com
About the Author
Dennis Child is a 40 year veteran credit union CEO recently retired. He has been associated with TCT for 25 years. Today, Dennis enjoys providing solutions and training for credit union managers. He also uses his financial credentials and advisory skills to assist the Boomer generation plan and prepare for their retirement years. He and his wife, Geri, live in Logan, Utah. Dennis can be reached at firstname.lastname@example.org.