Cecl automation banking book analytics v3
1 like327 views
Nexx Consultants is a Canadian management consulting firm specializing in banking book risk and regulatory compliance, offering expertise to both financial institutions and regulatory authorities. The document elaborates on the Current Expected Credit Loss (CECL) requirements, methodology for lifetime expected credit losses, and highlights the challenges of compliance while emphasizing the firm's cloud-based CECL ALLL Analyzer tool. Nexx aims to provide affordable, high-quality consulting services to assist smaller financial institutions in navigating complex regulations.
Cecl automation banking book analytics v3
- 1. Copyright©2018NexxConsultants Disclaimer Nexx Consultants is incorporated under the Laws of Canada, Canadian Corporation #: 002369533, US Employer Identification #: 98-1260328. The information, data and approaches provided herein are an outcome of our research and content expertise. Although we may refer to third party materials and/or analyze their impact, the content is the outcome of Nexx's proprietary knowledge and is rightfully owned by us. This work is copyright protected and legally privileged. Please do not distribute this presentation without the prior written consent of Nexx or its authorized affiliates. Nexx has made best efforts to ensure that this material is complete in its entirety. However, we do not warrant its completeness, accuracy, usefulness or satisfaction with all requirements. Copyright © 2018 Nexx Consultants BankingBook Analytics: CECL ALLL Analyzer April, 2018 Consultancy | Training | ContentA Nexx Consultants Company
- 2. Copyright©2018NexxConsultants Content 2 What is CECL1 2 Development Approach 4 CECL ALLL Analyzer3 BankingBook Analytics – An Introduction Nexx Consultants – An Introduction5
- 3. Copyright©2018NexxConsultants Content 3 What is CECL1 2 Development Approach 4 CECL ALLL Analyzer3 BankingBook Analytics – An Introduction Nexx Consultants – An Introduction5
- 4. Copyright©2018NexxConsultants Current Expected Credit Loss (CECL) requires determination of Lifetime Expected Credit Loss • Lifetime ECL is defined as a “credit loss” estimate of the present value of all cash shortfalls over the expected life of the financial instrument. The cash shortfall is the difference between the present value of the cash flows due to an entity, in accordance with the contract, and the present value of the cash flows that the entity expects to receive. Lifetime expected credit losses result from all possible default events over the expected life of a financial instrument • The “one-year expected credit loss” is the portion of lifetime ECL that results from default events within the year following the reporting date. Hence, the estimate must reflect an unbiased and probability-weighted number of credit losses determined by evaluating a range of possible outcomes. This method follows best practices based on IFRS 9 ECL modeling, and satisfies the criteria for a CECL- compliant impairment model • In an attempt to soften the burden for smaller institutions, the CECL guidelines state explicitly that complex models are not required for smaller institutions. Nevertheless, finding a simpler approach that does not carry a harsh penalty in loss reserve levels or in auditor and examiner review is not obvious 4
- 5. Copyright©2018NexxConsultants Allowance for Credit Loss: Some key upgrades 5 • CECL requires an organization’s current estimate of all expected credit losses over the contractual term • Broadens the information considered when measuring credit losses to include forward-looking information • Increases usefulness of the financial statements by requiring timely inclusion of forecasted information in forming expectations of credit losses • Increases usefulness of the financial statements by requiring timely inclusion of forecasted information in forming expectations of credit losses • Increases comparability of purchased financial assets with credit deterioration (PCD assets) with originated and non-PCD assets • Increases users’ understanding of underwriting standards by requiring additional information about credit quality indicators by year of origination (vintage) • For available-for-sale deb securities, aligns the income statement recognition of credit losses with the reporting period in which changes occur by recording credit losses (and subsequent reversals) through an allowance rather than a write-down
- 6. Copyright©2018NexxConsultants Non-credit impaired Key requirements § Non-credit impaired § Gross carrying amount using Expected Loss § ALLL to be calculated as the difference between the book value of an asset and the discounted expected cash flows of the asset using the Effective Interest Rate for the product (EIR) Data & Calculation Approach § Point in time determination of risk parameters Where, vk = discount factor for horizon k = Forward looking PD within the 12 moth period between k and k-12 = Forward looking LGD within the 12 moth period between k and k-12 !"#!"#$ % ! & = Exposure at default in dollars for account that default in 12 month period between k and k-12 LT = Expected remaining lifetime of the account LOL ECL= ∑ &!"# !$%& '#!'%& ( ! ) ∗ )*#!'%& ( ! ) ∗ !"#!'%& ( ! ) '#!'%& ( ! ) )*#!'%& ( ! ) 6
- 7. Copyright©2018NexxConsultants Credit impaired Key requirements § Credit impaired § Net carrying amount: This is calculated by multiplying the average ECL balance for assets by the portfolio EIR Data & Calculation Approach § Point in time determination of risk parameters ECL = "#$#%&' ∗ *+, ! *+, ! = Lifetime LGD for defaulted accounts 7
- 8. Copyright©2018NexxConsultants Content 8 What is CECL1 2 Development Approach 4 CECL ALLL Analyzer3 BankingBook Analytics – An Introduction Nexx Consultants – An Introduction5
- 9. Copyright©2018NexxConsultants CECL Modelling Framework 9 PD EaD LGD PIT adjustment/Dual calibration Alignment required Current balance and limit Current collateral value Current LGD Basel II 12 month PDs 12 months Forward Looking PD Lifetime Forward Looking PD Life of Loan definition Macroeconomic model Lifetime Forward Looking Adjustment Lifetime FL EaD Lifetime EaD Adjustment Lifetime Forward Looking LGD Lifetime Forward Looking LGD Adjustment Amortization Forecast collateral values CECL Components
- 10. Copyright©2018NexxConsultants PD Analysis - Our approach 10 Portfolio Example Asset Classes PD Modelling approach Forward PDs Delivery • Standalone • Residential Real Estate, Commercial Real Estate, Corporate Loans, Project Finance, etc. • Cox-regression for Survival function: • PD is given by: • Interpolation and Extrapolation ! = !! + $ − $! ∗ (!" − !!)/($" − $!) • Consulting and customized models development • Cohort/pool- based • Lines of Credit • Retail pools • Loans without IRB ratings • Kaplan-Meier estimate of hazard functions to remove potential biases in the data (e.g., Censored data) • Software • Consulting • Securities/Bonds • Externally rated bonds • Transition matrices • Simulation of loss distribution • Markov Models • Extrapolation assuming memoryless process • Software • Consulting ey Challenge: Lifetime PD Term Structure Method 2: Regression Modelling Relationship between historical PDs and behavioural factors Analyse statistical significance and business intuitiveness of factors Linear regression with logistic transform Cox regression for survival function ln 𝑃 𝑿 1−𝑃 𝑿 = 𝛽0 + 𝛽𝑖 𝑋𝑖 𝑝 𝑖=1 , 𝑆 𝑿 𝑡 = 𝑒ℎ0 𝑡 𝑒 𝛽 𝑗 𝑋 𝑗 𝑝 𝑗=1 where 𝑃𝑿 = 𝑃 𝐷 𝑿 = 1 and𝐷 𝑿 = 0 𝑖𝑓 𝑑𝑒𝑓𝑎𝑢𝑙𝑡 𝑑𝑜𝑒𝑠 𝑛𝑜𝑡 𝑜𝑐𝑐𝑢𝑟 1 𝑖𝑓 𝑑𝑒𝑓𝑎𝑢𝑙𝑡 𝑜𝑐𝑐𝑢𝑟𝑠 . where ℎ0 𝑡 is the empirical hazard function, estimated non-parametrically. Under this model, the probability of default at outcome period 𝑡 is given by: 𝑃𝑿 𝑡 = 1 − 𝑆 𝑿 𝑡 . onsiderations Consider transforms of variables to ensure stationarity Different regression formats (linear regression with logistic transform preferred by most banks) Term structure developed through including a month on book variable 10 ethod 2: Regression Modelling Relationship between historical PDs and behavioural factors Analyse statistical significance and business intuitiveness of factors Linear regression with logistic transform Cox regression for survival function n 𝑃 𝑿 1−𝑃 𝑿 = 𝛽0 + 𝛽𝑖 𝑋𝑖 𝑝 𝑖=1 , 𝑆 𝑿 𝑡 = 𝑒ℎ0 𝑡 𝑒 𝛽 𝑗 𝑋 𝑗 𝑝 𝑗=1 where 𝑃𝑿 = 𝑃 𝐷 𝑿 = 1 and𝐷 𝑿 = 0 𝑖𝑓 𝑑𝑒𝑓𝑎𝑢𝑙𝑡 𝑑𝑜𝑒𝑠 𝑛𝑜𝑡 𝑜𝑐𝑐𝑢𝑟 1 𝑖𝑓 𝑑𝑒𝑓𝑎𝑢𝑙𝑡 𝑜𝑐𝑐𝑢𝑟𝑠 . where ℎ0 𝑡 is the empirical hazard function, estimated non-parametrically. Under this model, the probability of default at outcome period 𝑡 is given by: 𝑃𝑿 𝑡 = 1 − 𝑆 𝑿 𝑡 . nsiderations Consider transforms of variables to ensure stationarity Different regression formats (linear regression with logistic transform preferred by most banks) Term structure developed through including a month on book variable 10 ECL model component - PD Key Challenge: Lifetime PD Term Structure Method 1: Cohort Analysis Years From Origination CumulativeDefaultRate • Conditional for survival • Kaplan-Meier estimate of hazard functions to remove potential biases in the data ℎ 𝑡, 𝑠 + 𝑡, 𝑿 = 𝑓𝑠,𝑿 𝑡 𝑆𝑠,𝑿 𝑡 − 1 ℎ 𝑡, 𝑠 + 𝑡, 𝑋 = default hazard of a customer 𝑡 months after observation 𝑓𝑠,𝑿 𝑡 = P(account defaults exactly 𝑡 months after observation ) 𝑆𝑠,𝑿 𝑡 = P(default does not occur within the first 𝑡 months) Cumulative Default Rate Years From Origination MarginalDefaultRate Marginal Default Rate Considerations • Requires a long time series of data (loan lifetime; segmentation and a full economic cycle • Impacted by calendar based events (e.g. change in business policy) • Develop lifetime PD given Stage 2 IFRS 9 PD for all accounts Basel II 12 Month PD 12 months Forward Looking PD Life-time Definition Life-time PD structure Lifetime Forward Looking Adjustment PD IFRS 9 EAD for all accounts Amortisation profile Current balance and limitEAD IFRS 9 LGD for all accounts Forecast collateral values Current LGD 12 month / Lifetime Forward Looking LGD LGD Lifetime Forward Looking PD 12 months Forward Looking Adjustment Alignment Required Bucket 1 and Bucket 2 DefinitionsMacroeconomic Model Current collateral value IFRS 9 12 Month PD 12 month / Lifetime Forward Looking EAD 12 month/ Lifetime FL Adjustment 12 month/ Lifetime FL Adjustment
- 11. Copyright©2018NexxConsultants EaD Analysis - Our approach 11 Portfolio Example Asset Classes Life of loan estimate Key challenge • Amortizing • Residential Real Estate, Commercial Real Estate, Corporate Loans, Project Finance, etc. • Usually contractual !"#$%&'#() "%)#! = !"#$%&'#()! ,"-.-(%)-/( 0%1%(2# !"#$%&'#() "%)#! = 3 + ∑ 6-7-" #$% +∑ 89&9& '$% 6#$:/#;;-2-#() ;/" '%2"/ <%"-%=1# 7#$>%2"/ <%"-%=1# 8'$:/#;;-2-#() ;/" 1/%( 1#<#1 <%"-%=1# &'$?/%( 1#<#1 2ℎ%"%2)#"-)-2 • Revolving • Lines of Credit • Retail pools Mean Residual Life = ()(%+%/-! (!/∝)#) %01(!) - t Where, 8 %, 7 -B C$$#" -(2/'$1#)# D%''% ;C(2)-/( Exposure = Used limit + K1 x (Unused limit) K1 = Sometimes referred as DDF – Drawdown factor or CCF – Credit Conversion factor Where K= UtilizationDefault – Utilization-t Limit-t – Utilization-t
- 12. Copyright©2018NexxConsultants LGD Analysis - Our approach 12 Portfolio Key challenge • Collateralized and unsecured • Ex-post where the index i runs over all types of collateral and the Value of Collateral before Defaulti • Ex-ante !"# = %!" &'!, ) . !"##$%. 1 − -./0 23'0 . 1 + % 6789: -;<'< Here, the Black-Scholes formula is given by with the standardized normal distribution N, the lognormal distribution as implemented, 1 1 ,0 n i ii Max Recovery Rate Value of Collateral before Default LGD Max Exposure at Default = æ ö× ç ÷= - ç ÷ è ø å ( ) ( )( ) 2 1 2 N( d ) N( d ) , lognorm 1,ln , exp( ) 1 BS VtL F VtL VtL s s - - × - = - ( ) ( ) 1 2 ln ln ; 2 2 VtL VtL d d s s s s = + = -
- 13. Copyright©2018NexxConsultants Challenges to comply with CECL 13 Lack of data Ratings Expert judgement Model design Implementation Peoples and skills System and processes Timelines Uncertainty Correctness Stakeholders’ expectations of compliance Regulatory expectation Industry practice Auditor and assurance Management Finance Challenges BBA Value proposition Model-design based on data availability Best-practice approach Reviewed and tested by audit firms Knowledge transfer Consulting only on as needed basis (Plug and Play solution)
- 14. Copyright©2018NexxConsultants Content 14 What is CECL1 2 Development Approach 4 CECL ALLL Analyzer3 BankingBook Analytics – An Introduction Nexx Consultants – An Introduction5
- 15. Copyright©2018NexxConsultants BB Analytics CECL ALLL Analyzer* Our cloud-based solution 15 Survival Analysis Calibration CECL Calculator CECL Solution by BB Analytics • Key benefits o Cloud-based configurable solution (plug & play – no consulting required) o No data integration required o Based on best practice approach o Synergizes development of risk data (can be extended to Stress Testing) Module 1 Module 2 Module 3 * Patents pending
- 16. Copyright©2018NexxConsultants An overview of the software solution 16 Loans Credit impaired Non-credit impaired P e r f o r m i n g N o n - p e r f o r m i n g Lifetime Expected Credit Loss Macroeconomicadjustments ALLL Calculation
- 17. Copyright©2018NexxConsultants Content 17 What is CECL1 2 Development Approach 4 CECL ALLL Analyzer3 BankingBook Analytics – An Introduction Nexx Consultants – An Introduction5
- 18. Copyright©2018NexxConsultants BB Analytics develops risk and capital applications using SaaS approach Low cost and accelerated delivery 18 Traditional solutions • Require significant FTE for implementation and develop input and output ETL (Extract, Transform and Load) layers • These layers and intermediate databases that are painful to feed and maintain • Back-end integration with core-banking system can lead to reconciliation challenges • Solutions often appear as black-box BankingBook Analytics • Immediate deployment and doesn’t require integration with source or risk systems. Our solutions are ready-to-use, hosted centrally in a secure cloud, and are directly accessed from your web browser without any installation • Our clients benefit from savings in overhead, IT infrastructure and implementation expense: o Typical risk and capital projects require significant investment in infrastructure and complex integration
- 19. Copyright©2018NexxConsultants Key offerings Clients use our applications with zero downtime 19 Key attributes and value transfer of bundled consulting Solutions Description Value transfer • Retail Pricer • Corporate Pricer • Develop floor and ceiling for your product-based retail pricing. Create segmented view of pricing and assess impact of pricing adjustment using bid- response pricing approach • Model Builder • Retail Lending Specialist • Mid-market & Wholesale Lending Specialist • IRB models development toolkit • Suite of retail lending models with dual ratings (PD & LGD) • Suite of mid-market and wholesale models (PD &LGD), covering CRE, Corporate, Project Finance • Auto-Validation • Basel II Credit Risk Capitalization • Internal Capital Adequacy Assessment • Internal Liquidity Assessment Process • CET 1 Stress Testing • Credit Cycle Scenario Analyzer • IFRS 9 Impairment Analyzer • CECL ALLL Analyzer • Operational Risk Capital Model • Interest Rate Risk in Banking Book • Several regulatory models in compliance with regulatory requirements Regulatory Compliance as a Service Credit Rating as a Service Credit Pricing as a Service
- 20. Copyright©2018NexxConsultants Content 20 What is CECL1 2 Development Approach 4 CECL ALLL Analyzer3 BankingBook Analytics – An Introduction Nexx Consultants – An Introduction5
- 21. Copyright©2018NexxConsultants Our brief history Since inception the firm has been involved in several high profile consulting engagements Roll-out of training practice for credit quants Incorporation and launch of consulting practice Launch of risk and regulatory content and knowledge distribution platform Over 30 quants, data scientists “curated” and trained 2013 2014 2015 2016 Signing up of initial portfolio of consulting clients Founder team 21 Design and development of proprietary models building and validation tools 2017 Assisted a major Canadian bank with IRB effort and development of PD, LGD models across several asset classes Launch of BankingBook Analytics 2018
- 22. Copyright©2018NexxConsultants A quick background • Who we are o Established in 2013, Nexx Consultants is an independent, Toronto-based management consulting firm, specializing in Banking Book risk and regulatory compliance. We assist regulated financial institutions, risk and supervisory communities • Why Nexx Consultants o With both, regulations and business-mix becoming more and more complex, access to expert advice has become an important factor for successful compliance. While the regulations impact financial institutions equally, however the access to such advice remains “un-equal”, especially for smaller FI’s. The principal idea behind creation of Nexx Consultants is that of “inclusion”- helping relatively smaller financial institutions get access to tier-1 advice – advice that meets their budgetary requirements and addresses their needs with speed and transparency • What is our offering o Financial Institutions: § Regulatory compliance, bank-wide risk management framework, development of risk measurement models and analytics, risk-based and strategic pricing, capital management strategies, reporting and stakeholder communication o Regulatory Authorities § Benchmarking of regulatory architecture, financial stability, regulator-mandated stress testing, supervisory training • What makes us different o Offering leading practice, Tier-1 expertise o Delivering fixed price engagements o Providing more complete solutions o Transferring knowledge and upskilling capabilities 22
- 23. Copyright©2018NexxConsultants Nexx Consultants offers independent, expert consultancy on risk and capital 23 Drawing on years of practical experience, our experts develop and refine the models, tools and processes you need to help ensure effective business operations and regulatory compliance. Affordability • Our professional services are priced to meet clients’ budgets. We truly partner with our clients to share the project risks, adopting a “we are in the same boat” mentality Independence • Regulators require advisors to avoid conflict of interest and offer honest advice. Nexx offers a balanced viewpoint that meets stakeholder demands, and also acts as an honest broker Best Practice Tools, Models and Deliverables • We apply advanced statistical techniques and develop models based on best practices. Our work draws on our team’s content expertise and global projects delivery experience. Our consulting bundles, model recipes and pre-built tools maximize the speed of delivery and lower development cost. We work hard to ensure we consistently deliver a high quality service Fixed price Contracts • Our business model operates on a fixed price contract basis, which ensures budget certainty, timely completion and zero cost overrun. Individual engagements are designed on the basis of “Appropriateness of Fit” within the organization, building capabilities and promotion of exchange of knowledge within our clients that outlast our involvement on the project
- 24. Copyright©2018NexxConsultants Risk and Capital Regulatory 19% Credit risk 43% Market risk 10% Liquidity risk 3% Interest rate risk in banking book 8% Integrated risk 12% Distribution of domain and team expertise Domain expertise Team expertise Credit quants 43% Market risk quants 11% Data scientists 8% Ex-Regulators 9% Generalists 11% Support 9% Operational risk 5% Statisticians 9% 24
- 25. Copyright©2018NexxConsultants We provide thematic coverage over entire spectrum of credit risk topics that address key challenges faced by industry today Data & Technology Credit Risk Measurement Performance Measurement Capital Strategies Policy, Governace and Culture • Do I have adequate data to develop models • How can my organization bridge data gaps • How can we design a robust risk data warehouse • What is the most appropriate technology solution for a bank our size • Are we creating or destroying capital • What is the minimum risk-adjusted return acceptable to my shareholders • How can we optimize return on our risk- weighted assets • What is the best-practice organization structure for our CRO organization • Who owns reputational risk and is accountable for it • How can we apply for IRB accreditation to OSFI • How can we develop risk centre of excellence and shared services • What are the most suitable model development approaches acceptable to the regulators • How can we independently validate models • How can we develop integrated planning framework • What is the most appropriate level of capital buffer for my instituion • Is it preferred to manage capital based on point-in-time or through-the-cycle data 25
- 26. Copyright©2018NexxConsultants Disclaimer Nexx Consultants is incorporated under the Laws of Canada, Canadian Corporation #: 002369533, US Employer Identification #: 98-1260328. The information, data and approaches provided herein are an outcome of our research and content expertise. Although we may refer to third party materials and/or analyze their impact, the content is the outcome of Nexx's proprietary knowledge and is rightfully owned by us. This work is copyright protected and legally privileged. Please do not distribute this presentation without the prior written consent of Nexx or its authorized affiliates. Nexx has made best efforts to ensure that this material is complete in its entirety. However, we do not warrant its completeness, accuracy, usefulness or satisfaction with all requirements. Copyright © 2018 Nexx Consultants Thank You