Profits from Losses

Dyer_Graham_RGraham Dyer | Senior Manager, National Professional Standards Group | Grant Thornton

In the quarters leading up to CECL, your focus will be on developing and implementing a model and process that results in a GAAP-compliant estimate of the allowance for expected credit losses (AECL) that best accommodates your bank and loan portfolio. Your new AECL estimation process must then undergo scrutiny from, and win the approval of, your auditors. What will they look for?

Grant Thornton’s Graham Dyer, one of the National ALLL Conference’s most highly rated speakers in recent years, returns to the 2016 National ALLL Conference to help you chart a course to audit approval. Among the issues Dyer will address:

  • Technical merits: Will your estimation process, including modeling future economic conditions and the resulting expected losses, result in an AECL estimate that is compliant with the CECL standard?
  • Data: Have you assembled and built your estimation process around a sufficient amount and appropriate types of data?
  • Governance: Have you addressed risk management activities and internal controls over financial reporting?

Key Takeaways

  • Your analysis and identification of sources of past credit losses may lead to identification of ineffective lending practices.
  • The requirement to obtain and analyze lifetime loss data may provide new and important insights into the real drivers of credit losses.
  • Deferred compensation plans are becoming increasingly popular, and including credit loss-driven metrics is common.

Your institution generates high-cost, high-quality information that feeds into the ALLL estimation process. There may be untapped opportunities to use this information to improve operations and profitability.

Lowering credit losses:

The primary way you can affect the ALLL under CECL is by creating structural improvements in credit risk mitigation in your institution and your loans. The requirement to obtain and analyze lifetime loss data may provide the opportunity to new and important insights into the real drivers of credit losses for your portfolios. Lifetime loss data is a powerful resource for meaningful analysis, provided your data warehouse captures the right information to allow flexible analysis.

Quantitative analysis of historical loss patterns and correlation to origination source, underwriting terms, etc., which may become more common under CECL, can reveal patterns regarding sources credit losses, such as:

  • Loan products
  • Loan officers
  • Branches
  • Collateral type and LTV
  • Underwriting policies, including covenants
  • Loan administration and workout practices

As a result of your analysis and identification of sources of past credit losses, you may wish to change certain lending practices. All other things equal, lower credit losses mean a lower ALLL, more capital, and a more profitable institution. Also, institutions may plan, at least initially, to use industry or peer data in their CECL estimation process.

However, using the loss information of others in the long term does not provide as much opportunity to identify drivers of credit losses.

Compensation and credit loss performance:

Deferred compensation plans for lenders are becoming increasingly popular, and including credit loss-driven metrics is common.

Product selection and pricing:

Expected changes in macroeconomic factors can greatly impact expected returns, and should influence pricing.

Yield on a loan should have a yield sufficient to cover:

  • Costs of funds
  • Operating costs
  • Loan losses
  • Profit margin

Risk adjusted return on capital is one powerful way to analyze the appropriateness of pricing, a method that many previously considered too data intensive to utilize in the past. However, as much of the data that may be used to estimate the AECL is also useful for RAROC analysis, adoption of CECL may represent an opportunity to also integrate an analysis such as RAROC.

In summary, the ALLL methodology should be in a continuous feedback loop with the product management, loan underwriting, pricing, borrower relationship management, and HR functions. These functions directly affect the amount of the needed allowance. The flow of valuable information into the ALLL estimation process should affect the management of the bank.


Download the full 2016 National ALLL Conference Digest.

For audio of the session, click here.