Allowance for Loan Losses- - CUNA Councils · 2015-05-12 · Implementation Guidance 1. Because of...
Transcript of Allowance for Loan Losses- - CUNA Councils · 2015-05-12 · Implementation Guidance 1. Because of...
Allowance for Loan Losses-Preparing for CECL
CUNA CFO Council Annual Meeting
May 2015
New Orleans, La.
Presented By:
Mike Sacher, Sacher Consulting
Presentation Objectives
• Provide overview of CECL
– How different from incurred loss model?
• Summarize key provisions of proposed standard
• Review example provided in proposed standard and
discuss go-forward strategies
• Note: Not possible in one hour presentation to get too
micro. Goal of presentation is to provide overview and
“food for thought.”
2
What is CECL?
• Acronym for Current Expected Credit Losses
• CECL based on an estimate of “lifetime contractual cash flows not
expected to be collected.”
• Current GAAP based on “Incurred” Loss Model”
– Event giving rise to the loss must have already occurred or
loss expectation must rise to level of probable
• CECL will likely result in significant increase to the balance of the
Allowance for Expected Credit Losses
– No longer referred to as Allowance for Loan Losses
• “Day 1” losses will be recognized under CECL
3
Is CECL Based on Fair Value?
• No!
• No consideration of interest rate vs. market
• When impaired, uses contractual rate for DCF purposes,
which is not a FV indicator
• FV may be either higher or lower than ACL analytics
reflect
4
Why CECL?
• Incurred loss model viewed as flawed due to:
– “Delayed recognition” of losses
– Reflective of “incurred current credit events”
• CECL
– Recognizes “expected credit risks”
– Requires consideration of a broader range of reasonable &
supportable information to inform credit loss estimates
5
Basis for Conclusion #16
• “The Board believes that it is potentially misleading to
investors to allow the balance sheet to reflect an amount
greater than the present value of cash flows expected to
be collected for instruments measured using amortized
cost, which would be the result of an approach that
recognizes only some of the contractual cash flows not
expected to be collected (that is, only those expected to
occur in the next 12 months).”
• I recommend you read the “Basis for Conclusions”
section of the Proposed ASU-very interesting!6
When CECL?
• Proposed ASU was issued for comment in 2012, and comment period ended
4/30/13
• According to FASB website (as of 4/28/15):
– Last update was on 3/18/15
– Identified Next Steps
The FASB will continue redeliberations on the CECL model, considering
feedback received through comment letters and outreach activities.
• Once finalized, transition period?
• Initial adoption will be recognized by a cumulative effect adjustment
• Don’t panic-there will be time to prepare
7
What does CECL apply to?
• Applies to financial instruments measured at amortized cost
– Loans “held for investment” (vast majority of all CU loans)
– Debt securities classified as HTM (other guidance for AFS)
– US Treasury securities
– MBS
– Other HTM
• New ACL method for HTM provides opportunity to “recover” previously
recognized write downs as conditions improve
• Includes guidance for Purchased Credit Impaired (PCI) assets
• Loan commitments not measured at F/V through Net Income
8
$ Impact of CECL?
• My sense is ACL (Allowance for Credit Losses) could
end up 150% to 200% of today’s balance (assuming
today’s balance is not overstated).
• Why?
• What about new loans originated?
– When do you assess ACL requirements?
9
Thoughts About Complexity
• Overall, the concepts of this proposed ASU are pretty
straight forward and logical…
• The challenge may be in data collection by appropriate
“pools” of loans.
– Plenty of lead time to get this right, but need to start thinking
about this now
• Disclosure requirements may be one of most difficult
aspects-ensure you review this area of the proposed
ASU!
10
Main Provisions (Pg. 2 of PASU)
1. Impairment based on current estimate of contractual cash flows
not expected to be collected at reporting date
2. Removes existing “probable” threshold
3. Broadens the range of information that must be considered in
measuring the ALL
– Historical loss information on similar assets (past events)
– Current conditions
– Reasonable and supportable forecasts that affect the
expected collectibility of remaining cash flow
11
Main Provisions (Pg. 2 of PASU)
4. Estimated losses would always reflect both:
– Possibility that a credit loss results AND
– Possibility that no credit loss results
5. “Accordingly, the proposed amendment would prohibit an entity
from estimating expected credit losses solely on the basis of the
most likely outcome (that is, the statistical mode)”
– More later…
6. Must consider prepayments in estimating future contractual cash
flows
12
1. Information on past events (historical experience of
similar assets)
2. Current conditions
3. Reasonable & supportable forecasts and their
implications for expected credit losses
4. Include quantitative & qualitative factors specific to
borrowers
– Borrower’s creditworthiness
– Current point and forecasted direction of economic cycle
13
Consider both internally and externally
available data
Consider both internally and externally
available data
5. Adjust historical experience metrics for current available
information
6. As forecast horizon increases, degree of judgment
increases
7. Consider available information without undue cost
and effort that is relevant to estimating contractual
cash flows.
– “Use estimation techniques that are “practical and
relevant”
14
Forecasting Extended Future Periods
See FAQ 13. Both of following okay
• “Revert to unadjusted historical averages for future periods
beyond which an entity is able to obtain reasonable and
supportable forecasts”
• “Assuming that economic conditions will remain stable for future
periods beyond which an entity is able to make or obtain
reasonable and supportable forecasts (that is, freezing the furthest
reasonable and supportable forecast and utilizing that forecast for
the remaining future periods”
15
Pooling of Loans
• Pool financial assets with similar risk characteristics. Be
careful not to broaden categories of dis-similar
characteristics
– 1st TD vs. 2nd TD
– New vs. Used auto
– Indirect vs. Direct
– Subprime vs. Prime
– Etc.
16
Forecasted Losses Might be Lower than
Historical Averages
• FASB Q&A #14
All things equal, when the current and expected economic
conditions facing the entity are more favorable than the economic
conditions that existed for the period over which historical
statistics were developed, the Board expects that the current
expected credit loss estimate would be lower than the historical
average.
17
Time Value of Money
1. Estimated losses shall reflect the time value of money.
2. DCF is one way of capturing time value concept.
3. If estimating losses using DCF model, use effective interest rate
for discount rate
4. Loss ratio method commonly used by credit unions is okay (see
par. 825-15-55-3)
18
Neither Worst-Case or Best-Case Scenario
1. Reflect both possibility that a credit loss results and the
possibility that no credit loss results
2. Probability weighted calculation that considers
likelihood of more than two outcomes not required
3. Prohibited from estimating credit losses based solely on
the most likely outcome (the statistical mode)
4. No ACL required if no loss would be recognized upon
default (adequately collateralized loans)
19
FASB Q&A #11:
“That is, even when the most likely outcome (which is a
common way to determine estimated cash flows) is zero
credit loss, an entity would be required to establish an
allowance for expected credit losses based on the risk of
credit loss for similar assets with a similar credit rating.
That risk of credit loss exists equally in portfolios and
individual assets.”
– In other words, still have to consider less likely outcomes and
provide some estimate in the event that a less likely outcome
might occur20
FASB Q&A #11:
“…In practice, however, the Board does not believe that the
prohibition against the statistical mode will conflict with many
commonly used approaches to estimating credit losses. Specifically,
many measurement methods (such as a loss-rate method, a roll-rate
method, a probability-of-default method, and a provision matrix using
loss factors) rely on an extensive population of actual loss data as an
input when estimating credit losses. Those methods already typically
rely on a statistical mean (that is, the average outcome) as opposed
to the statistical mode when estimating credit losses.
21
Accrual of Income (825-15-25-10)
1. “…An entity shall cease its accrual of interest income
when it is not probable that the entity will receive
substantially all of the principal or substantially all of the
interest.”
2. Beware of NCUA guidance on 90 days past due
cessation of accrual
22
Timing of Charge-Off (825-15-35-1)
• An entity shall directly reduce the cost basis in a financial asset (or portion of a
financial asset) within the scope of this Subtopic in the period in which the entity
determines that it has no reasonable expectation of future recovery.
– This is an important point and addresses a broad diversity in practice in
credit unions.
• Recovery of a financial asset previously written off shall be recognized by
recording an adjustment to the allowance for expected credit losses only when
consideration is received. In this context, a recovery means that an entity has
received consideration in satisfaction of some or all contractually required
payments following a write-off of the financial asset.
23
Disclosure Issues
• Voluminous Allowance disclosures still applicable
• Disclosure requirements will increase with finalization of
this ASU
• See next slide
24
• The Board decided to require that the credit quality indicators for all classes of financing
receivables (excluding revolving lines of credit such as credit cards) that are disclosed
under current GAAP be disaggregated by year of the asset’s origination (that is,
vintage year).
• The Board decided to limit the disaggregation by vintage year to no more than five
annual reporting periods, with the balance for financing receivables originated before the
fifth annual reporting period shown in aggregate.
• Other guidance for interim reporting (n/a for credit unions)
• The Board directed the staff to perform outreach to understand the usefulness, costs,
and operability issues with preparers, users, auditors, and financial institution regulators
for this disclosure requirement, and on the basis of the feedback received, the Board
may reconsider the alternatives to the roll-forward disclosure requirements at a
future meeting.
• Above as per 2/11/2015 FASB Action Alert
25
Implementation Guidance
1. Because of the subjective nature of the estimate of expected credit losses, this
Subtopic does not require specific approaches or specific policy elections in
this regard.
2. Rather, an entity has latitude to develop estimation techniques that are applied
consistently over time and aim to faithfully estimate expected credit losses by
using the key principles in this Subtopic.
3. An entity is not required to utilize a probability-weighted discounted cash flow
model to estimate expected credit losses.
4. Similarly, an entity is not required to reconcile the estimation technique it uses
with a probability-weighted discounted cash flow model.
26
Implementation Guidance
• Collateral Dependent Financial Asset
– A financial asset for which the repayment is expected to be
provided primarily or substantially through the operation (by
the lender) or sale of the collateral, based on an entity’s
assessment as of the reporting date
• Are the following Collateral Dependent Financial Assets?
– Residential real estate loan?
– Auto loan?
– Member business loan?
– Real estate loan in foreclosure?27
Implementation Guidance
• Expected losses on loan commitments must be included
in calculation.
• Can’t simply use one-year loss rate adjusted for
remaining contractual term since loss occurrence is not
linear!
• Think about “probability of default” separately from “loss
given default.”
28
Sacher Thoughts
1. Loss ratio approach used by most CU’s is a STARTING
POINT.
– Often not adequately disaggregated
– Only a snapshot of past xx months (often 12-24 months)
– Does not measure losses over remaining life of portfolio
– Often does not consider current how current conditions
impact historical loss measures (even though supposed to
consider Q&E)
– Does not quantify forecasts of future conditions
29
Sacher Thoughts
Loss Ratio Not the Answer!
825-15-55-24
“It typically would be inappropriate to estimate the expected credit losses for a long-
term asset by multiplying an annual loss rate (that is, the net amount written off in a
12-month period divided by the average amortized cost) by the remaining years of
the asset’s contractual term because loss experience is often not linear.”
“That is, for certain types of lending, credit losses are low shortly after origination,
rise rapidly in the early years of a loan, and then taper to a lower rate until maturity.
When estimating expected credit losses under this Subtopic, the loss rate should be
commensurate with the current credit risk of the financial asset.”
30
Sacher Thoughts
2. CU’s should consider the need for static pool analysis
a. Proper identification of unique pools, such as
Used vs new auto
Direct vs. indirect
1st mortgage vs. 2nd
Condo vs. house
b. Cumulative loss data for each pool by vintage year
c. See par. 825-15-55-19 for discussion of this issue
31
Sacher Thoughts
3. CU’s need to think thru the types of loan portfolio
metrics that meet the requirements of the ASU:
– Updated FICO & FICO migration
– Updated LTV/CLTV
4. What about FICO odds charts for consumer loans?
– If we can quantify probability of default, and combine with
loss given default (the magnitude of loss upon default), then
we have the basis for complying with key ASU requirements
32
Sacher Thoughts
5. CU’s need to identify appropriate general economic
indicators that might need to be impounded into the
CECL equation:
– National & local unemployment trends and forecasts
– Unique field of membership impacts (strikes, layoffs, etc.)
6. Do we need outside analytics provider to manage this
process?
– Probably not for straight forward portfolios
– Maybe yes for larger, more sophisticated portfolios
33
What About Auto Loans?
• See Example 3 from Proposed ASU
• Entity B is a lending institution that provides retail financing to consumers
purchasing new or used farm equipment throughout the country.
• The four-year amortizing loans it originates are secured by the farm equipment
purchased by the borrowers with proceeds from the loan using a relatively
consistent range of loan-to-collateral-value ratios at origination.
• The underlying farm equipment collateral is repossessed and sold at auction by
Entity B when the borrower becomes 90 days past due
• Entity B tracks these loans on the basis of the calendar-year of origination. The
following pattern of credit loss experience has been developed based on the
ratio of the amortized cost basis in each vintage that was written off because of
credit loss and the original amortized cost basis, shown as a percentage
34
What About Auto Loans?
35
• In estimating expected credit losses on the remaining outstanding loans at December 31,
20X9, Entity B evaluates its historical loss experience. It notes that the majority of losses
historically emerge in the second and third year of the loans.
• It notes that historical loss experience has worsened since 20X3 and that loss
experience for loans originated in 20X6 has already equalled the loss experience for
loans originated in 20X5 despite the fact that the 20X6 loans will be outstanding for one
additional year as compared with those originated in 20X5.
• In considering current conditions and reasonable and supportable forecasts, Entity B
notes that there is an oversupply of used farm equipment in the resale market that is
expected to continue, thereby putting downward pressure on the resulting value of
equipment. It also notes that severe weather in recent years has increased the cost of
crop insurance and that this trend is expected to continue.
36
What About Auto Loans?
What About Auto Loans?
• On the basis of these factors, it estimates that cumulative
loss experience on the remaining vintages outstanding
will be 4.6 percent, 4.8 percent, 5.0 percent, and 5.1
percent for loans originated in 20X6, 20X7, 20X8, and
20X9, respectively.
• These rates would be applied to the amortized cost in
each category, and the effects of those changes would
be recognized currently in net income as an adjustment
to the allowance for expected credit losses
37
What About Unsecured Loans?
• See Example 4 in PASU-homogenous unsecured loans
• Past due loans evaluated individually
• Current loans evaluated on a pool basis using “loss rate” approach
based on loans with similar risk characteristics
• Loss rate reflects cash flows the entity does not expect to collect
over the life of the loans in the pool
• Can’t simply be an annual loss ratio
• What about loss indicators (odd charts data) provided by credit
agencies?
38
What About Mortgage Secured Loans?
• No examples provided in literature
• If this were 2006-2007 timeframe, what impacts to prior
loss estimates?
• Impacts at end of “great recession?”
• See slide # 15 re: forecasting beyond reasonable
timelines
39
What About MBL’s?
• See Example 1 in PASU
– Uses historical loss rates by credit grade assigned to loans,
updated for current conditions.
– Assumed no change in current conditions nor any changes to
future losses (based on a 5-year loan term)
• See Example 2 in PASU
– Baseline historical loss rates similar to Example 1
– Add additional credit risk factors that reflect current conditions
(mgt. evaluation of current point in economic cycle, collateral
values, underwriting standards, etc.)
– Also 5-year loan terms, so less emphasis on future indicators 40
Reference Material
• FASB Project Update: Accounting for Financial
Instruments-Credit Impairment– http://www.fasb.org/jsp/FASB/FASBContent_C/ProjectUpdatePage&cid=1176159268094
• See also Basis for Conclusion Section of Proposed ASU
• FASB FAQ on Proposed ASU– http://www.fasb.org/cs/ContentServer?c=Document_C&pagename=FASB%2FDocument_C%2FDocumentP
age&cid=1176162305167
• FASB’s CECL Model: Navigating the Changes, Crowe
Horwath– http://www.crowehorwath.com/folio-pdf/FASBCECLModel-NavigatingChanges_FIA15904.pdf
41
42
Contact Information
Michael J. Sacher, CPA
Office: 310.459-9313
Cell: 310.880-5323
Fax: 310.454-7160
• Sup. Committee Strategic Planning &
Education
• Supervisory Committee Audits
• Merger Strategy & Negotiation • Allowance For Loan Losses Methodology &
Analysis
• Accounting Standards Implementation • Temporary CEO/CFO Assignments
• Internal Control Evaluation/Design • Whistleblower Reporting Thru PMYCU.Com,
LLC