CECL Workshop Discounted Cash Flow Method

Transcription

1 CECL Workshop Discounted Cash Flow Method Martin M. Caine, CPA Alan D. Lloyd, CPA MEMBER OF ALLINIAL GLOBAL, AN ASSOCIATION OF LEGALLY INDEPENDENT FIRMS 2017 Wolf & Company, P.C.

2 Introduction Martin Caine, CPA Member of the Firm, Wolf & Company, P.C Voice: (413) LinkedIn: martin-caine Alan Lloyd, CPA Senior Audit Manager, Wolf & Company, P.C. Voice: (215) LinkedIn: alan-lloyd 2

3 Agenda Discounted Cash Flow Model Key concepts and inputs Pros and cons Sample DCF model review Q&A 3

4 Key Concepts and Inputs From ASC : If an entity estimates expected credit losses using methods that project future principal and interest cash flows (that is, a discounted cash flow method), the entity shall discount expected cash flows at the financial asset s effective interest rate. When a discounted cash flow method is applied, the allowance for credit losses shall reflect the difference between the amortized cost basis and the present value of the expected cash flows. 4

5 Key Concepts and Inputs For variable rate loans benchmarked to an index: Projections of changes in the factor (e.g LIBOR, prime rate, etc.) shall not be made for purposes of determining the effective interest rate or estimating expected future cash flows. 5

6 Key Concepts and Inputs From ASC : An entity shall consider estimated prepayments in the future principal and interest cash flows when utilizing a method in accordance with paragraph (DCF). An entity shall not extend the contractual term for expected extensions, renewals, and modifications unless it has a reasonable expectation at the reporting date that it will execute a troubled debt restructuring with the borrower. 6

7 Key Concepts and Inputs Information that can be used: Internal, external or combination of both related to past events, current conditions, and reasonable, supportable forecasts Consider quantitative and qualitative factors in the environment Not required to search for all possible information that is not readily available without undue cost and effort 7

8 Key Concepts and Inputs Collateral-dependent loans: Regardless of the initial measurement method, an entity shall measure expected credit losses based on the fair value of the collateral when the entity determines that foreclosure is probable. An entity also shall consider any credit enhancements that are applicable to the financial asset when recording the allowance for credit losses. 8

9 Key Concepts and Inputs Changes in present value over reporting periods: May result not only from the passage of time but also from changes in estimates of the timing or amount of expected future cash flows. DCF users are permitted to report the entire change in present value as credit loss expense (or reversal of credit loss expense). Alternatively, an entity may report the change in present value attributable to the passage of time as interest income. 9

10 Key Concepts and Inputs Key Inputs: Loan Data: Loan number Unpaid principal balance (UPB), Book Balance Payment amount Maturity Date Segment Information (e.g. call code, product code, geography, risk rating, FICO scores, etc.) Payment structure Revolving status Available credit balances Government guarantee info Participation information, etc. 10

11 Key Concepts and Inputs Key Inputs: Assumptions: Prepayment rate (Conditional Prepayment Rate (CPR), Single Monthly Mortality (SMM)) Expected default rates Expected loss rates Timing to recovery 11

12 Key Concepts and Inputs Example SMM calculation that can be translated to CPR: TOTALS 25,376,213 11,933,614 76,169, % 13.85% Month Beginning Balance Contractual Payment Contractual Principal Contractual Interest Ending Balance Actual Payments Collected SMM CPR 1 300,000,000 3,109,152 1,984,152 1,125, ,846,416 6,153, % 11.6% 2 293,846,416 3,109,152 2,007,228 1,101, ,436,704 6,409, % 12.7% 3 287,436,704 3,109,152 2,031,265 1,077, ,918,095 6,518, % 13.4% 4 280,918,095 3,109,152 2,055,709 1,053, ,738,317 6,179, % 12.4% 5 274,738,317 3,109,152 2,078,884 1,030, ,315,888 6,422, % 13.6% 6 268,315,888 3,109,152 2,102,968 1,006, ,007,906 6,307, % 13.5% 7 262,007,906 3,109,152 2,126, , ,773,425 6,234, % 13.5% 8 255,773,425 3,109,152 2,150, , ,455,112 6,318, % 14.1% 9 249,455,112 3,109,152 2,173, , ,977,159 6,477, % 15.1% ,977,159 3,109,152 2,197, , ,575,062 6,402, % 15.2% ,575,062 3,109,152 2,221, , ,252,965 6,322, % 15.2% ,252,965 3,109,152 2,245, , ,830,298 6,422, % 16.0% 12

13 Pros & Cons Pros Less historical data required Works well for amortizing loans, balloons, fixed P&I, etc. Can be easily leveraged for other areas (e.g. Stress-testing, pricing, fair value calculations, etc.) Cons Significant amount of computational power required Does not work well for revolving loans Not as practical as some other models such as historical loss and vintage Useful if lack of loss history Back-testing to support and fine-tune assumptions Go-forward analysis instead of historical 13

14 Sample DCF Model Review - Mortgage Excerpt from model for single mortgage loan: TOTALS 99,858 55, ,549 18,592 2,789 15, ,918 Month Beginning Balance Contractual Payment Principal Interest Prepayment Defaulted Principal Estimated Loss Estimated Recovery Ending Balance Cash Flows 1 275,000 1, ,030 2, ,723 3, ,723 1, ,018 2, ,468 3, ,468 1, ,005 2, ,236 3, ,236 1, , ,026 3, ,026 1, , ,839 3, ,839 1, , ,673 3, ,673 1, , ,529 3, ,529 1, , ,407 3, ,407 1, , ,307 3, ,307 1, , ,228 3, ,228 1, , ,170 3, ,170 1, , ,133 3, ,133 1, , ,117 3, ,771 1,393 1, ,373 1, ,373 1,393 1, ,

15 Sample DCF Model Review - Mortgage Assumptions and Other Data Used for single mortgage loan: ASSUMPTIONS & INPUTS DATA & OUTPUTS Contractual Rate (INPUT) 4.50% GL Bal ance 275,000 CPR 12.00% PV of Cas h Fl ows 272,010 Months to Maturi ty (INPUT) 360 Es ti mated CECL DCF Res erve 2,990 Annual Defaul t Rate 1.50% Res erve to Carryi ng Bal ance 1.09% Loss Rate (% of Default Amt) 15% Recovery Rate (% of Default Amt) 85% Loss to Carry Balance per Est. Loss 1.01% Months to Recovery 10 Expected Pri nci pal Defaul t 6.76% NOTE: Assumptions would require qualitative adjustments for reasonable, supportable forecasts (e.g. if rates have moved up substantially in recent months, then prepayment speed (CPR) may need to be decreased). 15

16 Excerpt from model for pool: Sample DCF Model Review - Pool TOTALS 169,479,009 35,934, ,727,983 12,793,007 1,535,161 11,257, ,399,312 Month Beginning Balance Contractual Payment Principal Interest Prepayment Defaulted Principal Estimated Loss Estimated Recovery Ending Balance Cash Flows 1 300,000,000 3,109,152 1,585,652 1,123,500 3,727, ,000 48, ,287,177 6,436, ,287,177 3,109,152 1,614,664 1,102,105 3,655, ,383 47, ,624,674 6,372, ,624,674 3,109,152 1,643,420 1,080,899 3,584, ,833 46, ,012,048 6,308, ,012,048 3,109,152 1,671,923 1,059,880 3,513, ,349 45, ,448,860 6,245, ,448,860 3,109,152 1,700,174 1,039,046 3,444, ,932 44, ,934,675 6,183, ,934,675 3,109,152 1,728,177 1,018,395 3,374, ,580 43, ,469,061 6,121, ,469,061 3,109,152 1,755, ,927 3,306, ,292 42, ,051,590 6,060, ,051,590 3,109,152 1,783, ,638 3,238, ,069 41, ,681,837 5,999, ,681,837 3,109,152 1,810, ,528 3,170, ,909 40, ,359,384 5,939, ,359,384 3,109,152 1,837, ,596 3,104, ,813 40, ,083,812 5,879, ,083,812 3,109,152 1,864, ,839 3,037, ,778 39, , ,854,709 6,172, ,854,709 3,109,152 1,891, ,256 2,972, ,806 38, , ,671,667 6,106, ,671,667 3,109,152 1,917, ,845 2,907, ,896 37, , ,534,278 6,041, ,666,330 3,109,152 3,090,533 13,730 7,192 4, , ,717 3,153, ,717 3,109, ,717 2, , , ,614-34,614 16

17 Sample DCF Model Review - Pool Assumptions and Other Data Used at pool level: ASSUMPTIONS & INPUTS DATA & OUTPUTS Contractual Rate (INPUT) 4.50% GL Bal ance 300,000,000 CPR 14.00% PV of Cas h Fl ows 298,163,136 Average Li fe to Maturi ty (INPUT) 120 Es ti mated CECL DCF Res erve 1,836,864 Annual Defaul t Rate 1.60% Res erve to Carryi ng Bal ance 0.61% Loss Rate (% of Default Amt) 12% Recovery Rate (% of Default Amt) 88% Loss to Carry Balance per Est. Loss 0.51% Months to Recovery 10 Expected Pri nci pal Defaul t 4.26% NOTE: Assumptions would require qualitative adjustments for reasonable, supportable forecasts (e.g. if rates have moved up substantially in recent months, then prepayment speed (CPR) may need to be decreased). 17

18 Sample DCF Model Review Discussion points 1. Any concerns with the complexity? 2. Any concerns with the required assumptions? 3. How can this analysis be improved? 3. Does anyone plan to use a DCF model for CECL? 18

19 Thank You Martin Caine, CPA Member of the Firm, Wolf & Company, P.C. Voice: (413) LinkedIn: martin-caine Alan Lloyd, CPA Senior Manager, Wolf & Company, P.C. Voice: (215) LinkedIn: alan-lloyd 19