Solving the CECL Challenge Presented by Doug Wright CFO, Mission Federal Credit Union and Bryan W. Mogensen CPA, CLA (CliftonLarsonAllen LLP) Doug Wright • CFO, Mission Federal Credit Union • 31 years in the credit union, banking industry • 16 year CFO • Member, FASB CECL Transition Resource Group • MBA, Gonzaga U., BA, Univ of California, Berkeley
22
Embed
Solving the CECL Challenge - NAFCU - CECL... · CECL Implementation Plan & Timeline CECL Implementation Team - Done Developing basic CECL understanding - Done – External resources
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Solving the CECL Challenge
Presented by
Doug WrightCFO, Mission Federal Credit Union
andBryan W. Mogensen CPA, CLA (CliftonLarsonAllen LLP)
Doug Wright
• CFO, Mission Federal Credit Union
• 31 years in the credit union, banking industry
• 16 year CFO
• Member, FASB CECL Transition Resource Group
• MBA, Gonzaga U., BA, Univ of California, Berkeley
Bryan W. Mogensen, CPA• 25+ years working primarily with credit unions
• Speaker/Presenter at: – AICPA National Credit Union Conference
– ACUIA National Conference
– Various State League and Chapter Meetings
– CUES/CUNA
• Graduate of University of Wisconsin – Milwaukee, WI
• Member of: – AICPA
– Arizona Society of CPAs
• On the AICPA National Credit Union Planning Committee
Data Collection Insights Data required depends on methodologies used
Roll rate Vintage PD/LGD Discounted cash flow
Many segments may not have statistically valid sample sizes Reduce sample size by more statistically precise techniques (multi-variate
regression, iterative processing) Sample size/lack of data can also be tackled with industry data
Economic cycle “expectation” presents interesting questions Valid data back to pre-2007? If so, is 2005/6/7 - current reflective of a reasonable cycle? How does this change in the next 3 years?
It’s likely that additional refinements will be required regarding data for the first 2-3 years after implementation
Data Collection Recommendations
Identify the methodologies you are most likely to use Identify critical inputs required:Loan specific infoEnvironmental & economic info
Do your best with historical data, but develop capabilities to retain & evaluate required data on a go-forward basis
Collect as much “useable” info as possible on a loan-specific level
Identify how you are going to overcome small sample sizes
Modeling/Methodology Insights
• FASB committed to allowing flexibility, but largely “uninvolved” post-implementation
• Regulators profess flexibility, but may gravitate to more common approaches Field examiner knowledge Institution comparability Vendor concentrations
• Different methodologies may yield “better” results for different segments
Modeling/Methodology Insights
Model “platform” should include areas to input & describe:o Historical basiso Current adjustments: qualitative & environmentalo Reasonable economic forecastso Reversion to historicalo Other key assumptions: prepayments, contractual lives, extensions
Model platform built with future in mind; how is the program going to be maintained & updated?
Discounted cash flow specifics:o Usually based on another methodology (e.g. PD/LGD)o Given other factors equal, will often result in lower estimated allowance
because future losses are discounted
Modeling & Methodology Recommendations
• Explore industry whitepapers/webinars on specific methodologies• Engage with several vendors to view approaches• If building one internally, consider:
– Platform requirements as noted previously– Ongoing maintenance from quarter to quarter– Validity, backtesting & “auditability”– Regulatory “defense”
• Applicability for other uses– Credit management– Credit stress testing– Loan/relationship pricing
Buy vs. Build Insights Vendors have committed significant resources to developing
models Platforms are still under development Most have worked with auditors, but none has been audited Pricing varies, depending on model complexity and institution
asset size Build extends far beyond initial platform development:
o Ongoing collection of data, maintenance & updating of model for each quarter, and model refinement
o Validation and “auditability”o Regulatory scrutiny
Buy vs. Build Recommendations
Contact and consider several vendors Beware of “CECL Compliant” claims Look closely at validation/explanation capabilitiesConsider cost vs. “performance” tradeoffsDon’t rush into a decision If build, note prior insights If build, work closely with auditor, regulator and
other independent parties for feedback
Technical Challenges
1. Credit cards
2. Loans formerly known as TDRs
3. PCI loans
4. Debt securities
Credit Cards
Unfunded credit lines No allowance if unconditionally cancellable
by issuer• Provision for unconditionally cancellable
• Most CUs have this provision
If not cancellable, record liability for unfunded commitments
Credit CardsEvaluate portfolio into:Transactors
• Pay off monthly
• Earn rewards
• Lower charge-offs
Revolvers• Retain monthly balance
• Most net interest income earned by these
• Higher charge-offs
Credit Cards
• Evaluate weighted average life– Payments based on activity (FIFO)
– Payments applied based on CARD Act• Higher APRs paid first
– Either method acceptable
• Determine loss emergence period– Period from loss-triggering event to charge-off
Troubled Debt Restructures (TDRs)• Individually impaired concept removed from ASU
2016-13
• TDR concept retained
• Initial indications – TDR credit losses measured using various CECL models
• But, FASB indicated that allowance must still consider full “economic loss”?
• Discounted cash flow measures economic loss
• Pending further clarification
PCD/PCI Loans
• Purchased Credit Deteriorated – loans that have incurred “more than insignificant deterioration since origination”
• More loans qualify as PCD compared to Purchased Credit Impaired (PCI)
• Discounts associated with credit loss added to reserve upon acquisition
PCD/PCI Loans
• PCD adjustments applied prospectively– Adjust amortized cost basis of loan to reflect
addition of the ALLL
• Not required to reassess current PCI loans meet definition of PCD
PCD/PCI Loans
Can elect to maintain pool of current PCI
Noncredit discount/premium, after ALLL adjustmento Accreted to interest income using interest method
on effective interest rate after ALLL adjustment for credit losses
Debt Securities
AFS:• Replaces OTTI impairment concept
• Determine if decline in fair value due to credit or other losses– If credit – post as allowance through income
statement (provision)
– If other (interest) – post through OCI
Debt Securities
AFS Use CECL model
Use an allowance instead of direct write-off (permits reversals)
Credit risk – post as allowance through income statement (provision)
HTM Treatment similar to loans held in portfolio
CECL reserve
Debt Securities
• Complex models not expected with plain vanilla portfolios
• For those with credit risk, requires reserve– Some municipals, private-label MBS
– Reassess allowance monthly
– Can reverse credit allowance up to zero
– Cannot have negative credit loss
Transition Considerations
Model and ALLL align with loan growth• Loan and related PLL booked on Day 1
Forecasting/Q&E• Should align with your ALM models
Transition Considerations
• Possible effects to:– Timing of loan promotions
– Loan discounts/offers
– Defer all loan origination fees and costs• Direct and indirect
Transition Considerations
• Net income challenges?– If book loan and PLL in Q1 then have
entire year to recoup income through loan interest
– If book loan and PLL in Q4 then have more PLL in CY and lower loan interest