Slide 1: CECL – Using a Reasonable and Supportable Forecast
CECL – Using a Reasonable and Supportable Forecast July 2019
Slide 2: Speakers
Speakers Chris Henkel Robby Holditch Sohini Chowdhury Senior Director, Director, Regulatory and Director, Regulatory and Enterprise Risk Solutions Accounting Solutions Accounting Solutions Moody’s Analytics Moody’s Analytics Moody’s Analytics CECL – Using a Reasonable and Supportable Forecast 2
Slide 3: Today's Discussion Points
Today’s Discussion Points » CECL Overview: What’s Changing? » Recent Updates: Real-life Impact » Estimating Expected Credit Losses (“ECL”): A Refresher » Understanding and Defending Your Reasonable and Supportable Forecast » Concluding Remarks and Q&A CECL – Using a Reasonable and Supportable Foreca
Slide 4: CECL Overview
1 CECL Overview
Slide 5: What is CECL
What is CECL FASB, ASU No. 2016-13, June 2016 Financial Instruments—Credit Losses (Topic 326) CECL means CURRENT EXPECTED CREDIT LOSS Lifetime loss estimate from origination which replaces “incurred loss” model, where: “The measurement of expected credit losses is based on relevant information about
Slide 6: Changes Under CECL
Changes Under CECL Applies to all banks, savings associations, credit unions » Scope: financial instruments measured at amortized cost basis – Loans held for investment – Debt securities held to maturity – Debt securities available for sale* – Off balance sheet exposures (Loan commitments, Letters o
Slide 7: Summary: Your CECL Formula
Summary Your CECL Formula = Adjustments Adjustments for Historical for Current or or Reasonable loss Economic experience & Conditions Supportable Forecast* *326-20-30-9 - An entity is not required to develop forecasts over the contractual term of the financial asset or group of financial assets. Rat
Slide 8: Defining What is Acceptable…
Defining What is Acceptable… There are a few elements that are required to be incorporated when using any methods… 326-20-30-9 An entity shall not rely solely on past events to estimate » Historical Information expected credit losses…. When an entity uses historical loss information, it shall consid
Slide 9: New disclosure requirements!
New disclosure requirements! 326-20-50-11 An entity shall disclose all of the following by portfolio segment and major security type: a) A description of how expected loss estimates are developed b) A description of the entity’s accounting policies and methodology to estimate the allowance for credi
Slide 10: From AICPA Banking Conference 2018
From AICPA Banking Conference 2018 “ CECL is sensibly designed… Prepares need to present economic assumption, ” perhaps in a tabular format…. SEC Remarks at the 2018 AICPA Banking Conference CECL – Using a Reasonable and Supportable Forecast 10
Slide 11: Economic Assumption Tabular Example
Economic Assumption Tabular Example Moody’s Example CECL – Using a Reasonable and Supportable Forecast 11
Slide 12: Recent Updates
2 Recent Updates
Slide 13: Potential Timeline Changes
Potential Timeline Changes » There are currently 4 initiatives underway that could alter the timeline for the implementation of CECL. All 4 are being actively monitored but could yield changes to the implications of the new standard. » Those in House and Senate are in committee and require committee
Slide 14: Issuer Legislation Summary
Issuer Legislation Summary Status FASB Private Company Council The PCC within the FASB is considering a The implication is that this (PCC) proposal for change within it’s standard issuance process would push the non-SEC effective date changes that would make the effective date on new filing PBE’s (m
Slide 15: Regulatory Capital Changes
Regulatory Capital Changes NR 2018-142 FOR IMMEDIATE RELEASE December 21, 2018 Agencies Allow Three-Year Regulatory Capital Phase In for New Current Expected Credit Losses (CECL) Accounting Standard The federal bank regulatory agencies approved a final rule modifying their regulatory capital rules a
Slide 16: Proposed Changes to Call Report FFIEC 031, 041 & 051
Proposed Changes to Call Report FFIEC 031, 041 & 051 Draft Reporting Form Call Report Revisions Proposed This draft reporting form reflects revisions addressing the revised accounting for credit losses under the Financial Accounting Standards Board’s Accounting Standards Update No. 2016-13, “Financi
Slide 17: Estimating Expected Credit Losses ("ECL"): A Refresher
3 Estimating Expected Credit Losses (“ECL”): A Refresher
Slide 18: Common Methodologies to Estimate Credit Loss
Common Methodologies to Estimate Credit Loss » Loss Rate – Pool/cohort approach – Rating and loan type – “WARM” method » Probability of Default (“PD”) and Loss Given Default (“LGD”) – Mapping internal ratings to agency ratings – Use internal rating distribution and a central tendency of default › Im
Slide 19: Solving the Data Problem
Solving the Data Problem A sensible way to think about it… How should I segment my portfolio? Which methodologies are appropriate? Where can I find the data I need? CECL – Using a Reasonable and Supportable Forecast 19
Slide 20: Three types of data needed to derive CECL estimates
Regardless of the approach, you will need three types of data to derive CECL estimates 1. Data that captures the segment/pool’s historical loss experience 2. Data for adjusting historical loss data to reflect the current credit environment on instruments in the segment/pool Data can be used to model
Slide 21: Loss Rate Method Example
Loss Rate Method Example A group of loans have an amortized cost of $5M at the end of 2015. Year Amortized Cost Average Balance Observed NCOs 2015 $ 5,000 2016 $ 5,500 $ 5,250 $ 20 2017 $ 6,000 $ 5,750 $ 50 2018 $ 6,500 $ 6,250 $ 40 2019 $ 7,000 $ 6,750 $ 30 2020 $ 7,500 $ 7,250 $ 50 During 2016-202
Slide 22: PD and LGD Method
PD and LGD Method Commercial Real Estate Loan Commercial & Industrial Loan Quantitative Quantitative Qualitative Factors Qualitative Factors Factors Factors Master Rating Scale Quantitative Risk Qualitative Score Quantitative Risk Qualitative Score Measure (EDF%) (0–100) Rating Rating Measure (EDF%)
Slide 23: Example of a PD and LGD Rating Scale
Example of a PD and LGD Rating Scale A B C D E F G 5% 15% 20% 25% 35% 45% 55% 1 Pass 0.08% 0.00% 0.01% 0.02% 0.02% 0.03% 0.04% 0.04% 2 Pass 0.14% 0.01% 0.02% 0.03% 0.04% 0.05% 0.06% 0.08% 3 Pass 0.25% 0.01% 0.04% 0.05% 0.06% 0.09% 0.11% 0.13% 4 Pass 0.43% 0.02% 0.06% 0.09% 0.11% 0.15% 0.19% 0.24% 5
Slide 24: Asset Quality Statistics
Asset Quality Statistics Noncurrent loans are well below the …therefore so are the amount of loans long-run average… being charged-off Noncurrent Loan Rate Annual Net Charge-Off Rate 5.00% 2.50% 4.50% 4.00% 2.00% 3.50% 3.00% 1.50% 2.50% Avg. 1.59% 2.00% 1.00% Avg. 0.66% 1.50% 1.00% 0.50% 0.50% 0.00%
Slide 25: Food for Thought…
Food for Thought… It is acceptable to adjust historical loss information for current conditions and the reasonable and supportable forecasts through a qualitative approach rather than a quantitative approach…but is it really easier? Small changes to “Q Factor” assumptions may result in large changes
Slide 26: Understanding and Defending "Reasonable and Supportable" (R&S) Forecasts
4 Understanding and Defending “Reasonable and Supportable” (R&S) Forecasts
Slide 27: CECL Forecasting Requirements
CECL Forecasting Requirements Topic 326 guidance “The measurement of expected credit losses is based on relevant information about past events, including historical experience, current conditions, and reasonable and supportable forecasts that affect the collectability of the reported amount. An enti
Slide 28: 3 Ways of Satisfying the R&S Requirement
3 Ways of Satisfying the R&S Requirement 1. Reversion in inputs Revert to unadjusted historical average economic values R&S period < life of the loan 2. Reversion in outputs Revert to unadjusted historical average losses 3. Lifetime R&S R&S period = life of the loan CECL – Using a Reasonable and Sup
Slide 29: How to Incorporate Economic Forecasts in CECL?
How to Incorporate Economic Forecasts in CECL? CECL does NOT require a specific approach Qualitatively leveraging the forecasts acceptable for smaller institutions No strict rules on number of scenarios, weights etc. But, – Using multiple scenarios mitigates the uncertainty from a single forecas
Slide 30: R&S Shorter Than Life of the Loan
R&S Shorter Than Life of the Loan Elect an R&S period, a reversion period and a reversion technique. Reversion to unadjusted historical averages can be – 1. IN INPUTS Over R&S period = Economic forecasts using the model Over reversion period = Economic forecasts artificially revert to unadjusted his
Slide 31: Input Reversion Example
Input Reversion Example Unemployment rate, %, US 10.0 Lookback Period = 20 qtrs. 9.0 Historical Unadjusted Average = 5.2% 8.0 R&S Period = 8 qtrs. 7.0 Reversion Period = 4 qtrs. Reversion Technique = Straight Line 6.0 5.0 4.0 3.0 Moody's Analytics Baseline User Defined CECL – Using a Reasonable and
Slide 32: R&S Shorter Than Life of the Loan
R&S Shorter Than Life of the Loan 2. IN OUTPUTS Over R&S period = Credit loss and economic forecasts using the model Over reversion period = Credit losses artificially revert to some unadjusted historical average After reversion period until the end of life = Credit losses set equal to unadjusted hi
Slide 33: Output Reversion Example
Output Reversion Example Monthly Loss Rate, % 0.20 0.15 0.10 Model-determined over 3yrs R&S period 0.05 Historical Unadjusted Loss Rate Immediate Reversion Gradual Reversion over 1yr 0.00 1 13 25 37 49 Age of loan (months) For illustration purposes only CECL – Using a Reasonable and Supportable Fore
Slide 34: R&S = Life of the Loan
R&S = Life of the Loan 3. LIFETIME R&S Possible only if BOTH a) and b) are satisfied a) Economic forecasts are R&S over the life of the loan b) Credit loss models produce reasonable estimates of losses over the life of the loan CECL – Using a Reasonable and Supportable Forecast 34
Slide 35: What Makes an Economic Forecast R&S Over Lifetime?
What Makes an Economic Forecast R&S Over Lifetime? It is produced by a model which: is based on sound, generally accepted economic theory incorporates inter-relationships and feedback effects • a shock to one factor impacts all other factors over time considers a range of possible outcomes provi
Slide 36: Structural Forecast Model: Set of Interlinked Equations
Structural Forecast Model: Set of Interlinked Equations The approach used by Federal Reserve, IMF, Central Banks, and Moody’s Analytics 10-yr yield Consumption Wages and salaries Banking sector Investment Labor force Monetary policy rate Government Employment Population Prices Exchange rates Exports
Slide 37: Integrated National, State, and Metro-Level Forecasts
Integrated National, State, and Metro-Level Forecasts Unemployment rate, % 12 National State 10 8 2019Q1 F 6 4 2 Metro 0 00 02 04 06 08 10 12 14 16 18 2019Q1 F CECL – Using a Reasonable and Supportable Forecast 37
Slide 38: Moody's Forecasts Cover a Range of Possible Outcomes
Moody’s Forecasts Cover a Range of Possible Outcomes Scenario Inventory US Real GDP, % change annualized BL Baseline Forecast (50th pctile) 8 CB Consensus Baseline 7 S0 Strong Upside (4th pctile) 6 S1 Stronger Near-Term Growth (10th pctile) 5 S2 Slower Near-Term Growth (75th pctile) 4 S3 Moderate Re
Slide 39: Which R&S Approach Should You Use?
Which R&S Approach Should You Use? Each has its pros and cons R&S Approach PROS CONS Reversion in Inputs or – Need economic forecast only – Have to defend choice of R&S period Outputs through R&S period – Have to defend choice of lookback period used (R&S period < life of loan) – Need credit loss mo
Slide 40: Concluding Remarks and Q&A
5 Concluding Remarks and Q&A
Slide 41: CECL – Using a Reasonable and Supportable Forecast
CECL – Using a Reasonable and Supportable Forecast 41
Slide 42: Globally and locally acknowledged for award-winning tools
Globally and locally acknowledged for award-winning tools to measure and manage risk ——— Steer sat sae pn Fen ut Faience caer Awards a FE scons En 2017 nent A PR ‘Winner “Res Solvency If Tech Solution CECL Technology Balance Sheet Compliance Risk Technology CLO Data Provider of Innovation in Custome
Slide 43: For more info on Moody's Analytics solution, visit our CECL site
Robby Holditch Chris Henkel Director Senior Director +1 (212) 553-2119 +1 (212) 553-4679 Robby.Holditch@moodys.com Chris.Henkel@moodys.com For more info on Moody’s Analytics solution, visit our CECL site: http://MoodysAnalytics.com/CECL-implementation
Slide 44: CECL – Using a Reasonable and Supportable Forecast
© 2019 Moody’s Corporation, Moody’s Investors Service, Inc., Moody’s Analytics, Inc. and/or their licensors and affiliates (collectively, “MOODY’S”). All NO WARRANTY, EXPRESS OR IMPLIED, AS TO THE ACCURACY, TIMELINESS, COMPLETENESS, MERCHANTABILITY OR FITNESS FOR ANY rights reserved. PARTICULAR PURP