Evaluating the Evaluation Bank Evaluation Programs Doug Potts, MAI, VP Commerce Bank A Few Questions • True/False? – Most loan transactions require appraisals. • True/False? – Evaluations are killing the appraisal profession. • True/False? – Reg B will eventually create more business for appraisers. • True/False? – Reg B will - in time - increase available consumer credit. A Primer on Bank Lending • Major Purposes • New Loans (new collateral) • Subsequent Transactions (existing collateral) • Problem Loan Management (“exiting” collateral) • Major Loan Categories • Consumer purpose lending (GSE / HELOC) • Commercial business lending (C & I) • Commercial investment lending (CRE) • Unclassified lending – the “junk drawer” Bank Portfolios--Risks & Rewards • • • • Where does most of a bank’s workload sit? Do renewals generate new $ for added cost? New loans generate growth in profits 80/20 rule with a vengeance • Loan profitability sits in the 20% • But the workload sits in the 80% • Renewed loans generate $0 added profit • But comprises the bulk of a bank portfolio LOAN STRATIFICATION v PROFITABILITY $4,000 6,000 $3,500 5,000 $3,000 # of Notes $2,500 $2,000 3,000 $1,500 2,000 $1,000 1,000 $500 $- - <$100M <$250M <$500M Loan Size Bracket <$1MM > $1MM Earnings MM 4,000 Job Stratification by Function Appr Eval NM Eval Valid 23% 48% 29% The Point Is …. • When you combine the 80/20 rule with the $0 profit rule …… • You need a cost-effective solution to meet • Underwriting need • Regulatory requirements • The key is “adequacy” The Purpose of Evaluations • Regulations require every loan get a value • Only some deals require an appraisal. • But they also provide for cost effectiveness • Spread thin but broadly • Load balance tool • Balance risk & reward • Risk sensitized • For loan size (e.g. <$250M de minimis) • For repayment reliance (CRE v C&I v Consumer) • For portfolio concentration & workload (new v renewal) Program Basics • External Vendor Procured • Proliferating Options • Growing array of vendors • Growing array of evaluation styles & products • Coverage & Fulfillment • Internally Developed • • • • Licensed or Unlicensed Team Members Qualifications & Training Data Sourcing & Inspections QA / QC Controls Evaluation – Scope of Work • • • • • Interagency Appraisal & Evaluation Guidelines Flexibility v rigidness When & where to use Defining the problem Creating the solution What will your regulator permit? 2010 IAG Requirements • • • • • Should contain sufficient information Details analysis, assumptions, conclusions Should be documented in the credit file or reproducible. Identify the location of the property. Provide a description of the property and its current and projected use. • Provide an estimate of the property’s market value in its actual physical condition, use and zoning designation as of the effective date of the evaluation (that is, the date that the analysis was completed), with any limiting conditions. • Describe the method(s) the institution used to confirm the property’s actual physical condition and the extent to which an inspection was performed. 2010 IAG Requirements #2 • Describe the analysis that was performed and the supporting information that was used in valuing the property. • Describe the supplemental information that was considered when using an analytical method or technological tool. • Indicate all source(s) of information used in the analysis, as applicable, to value the property, including: – External data sources (such as market sales databases and public tax and land records); – Property-specific data (such as previous sales data for the subject property, tax assessment data, and comparable sales information); – Evidence of a property inspection; – Photos of the property; – Description of the neighborhood; or – Local market conditions. 2010 IAG Requirements #3 • Include information on the preparer when an evaluation is performed by a person, such as the name and contact information, and signature (electronic or other legally permissible signature) of the preparer. Let’s Re-Arrange the IAG Ideas Evaluations have 5 major elements • Define the problem to be solved • Get needed research elements • Make needed inspections as appropriate • Selecting the right data • Doing the math & calculations Evaluation Scope Defining the Problem Calculating an Answer Selecting Appriorate Data Research & Descriptions Inspection & Descriptions Eval Content - Preliminaries • Defining the Problem • • • • • Complexity Effective date(s) – what is this? Limiting conditions Can the premise be prospective? Can the zoning be prospective? • Basic research & descriptions • • • • Property specific records Neighborhood & market trends Current & projected use Current zoning designation Eval Content – Inspection Elements • • • • • Methods to confirm physical condition Describe extent of inspection Evidence of inspection Descriptions inspection details - granularity Property photos More on Inspections • “Methods to confirm physical condition” • “Describe extent of inspection” • What about? • Out of territory property • Borrower or lender representation? • Shelf life of the representation/inspection? • Origination v Renewal? • Flexible standard? • Credit Quality? Pass v non-pass? Eval Content –Selecting Data • • • • “As applicable” data External sales databases etc. Local market conditions Comparable sales Eval Content – Calculate an Answer • Provide value in actual physical condition • Is that the only condition allowed? • Describe analysis performed • Describe supporting information • Data sourcing & selection • What valuation technique? What methods? • Analytical method or technological tool • Analysis performed & supporting information • Supplemental information Eval Content - Fingerprints • • • • • “When performed by a person” Name / contact info Actual signature Electronic signature Is the evaluator qualified? • What about state restrictions? Evaluation Scope Defining the Problem Calculating an Answer Selecting Appriorate Data Research & Descriptions Inspection & Descriptions Limitations • Shelf life of evaluations • Tailoring for loan-specific issues • LTV adequacy • Lease structures • Environmental impacts & offsets • As-is value v prospective value Evaluation Options & Risk What program to develop and use? “You get to pick your problems, you don’t get to pick no problems….” • Internal Program • External Vendor Program • Appraisal Only – “no evaluations” • Hybrid? • What about BPO? The Nexus -- Risk v Regulation • Triangulating competing elements • • • • Accuracy Adequacy Efficiency Cost effectiveness • Does an evaluation need to be accurate? • • • • Is accuracy always cost effective? Is precision always necessary? What about low LTV transactions? Using a shotgun to shoot a fly… External v Internal Eval Programs • Consider the business model -• Collateral class – conventional v oddball • Market coverage – urban v tract suburb v rural • External Model • Consistency & uniform product – predictable • Within its limits – highly reliable system • Fulfillment failure risk • Internal Model • Flexibility & adaptability • Precision subordinated to practicality of business • Disclosure Risks External Products • • • • • • Appraiser-centric or broker-centric Available options How to compare them? How to review them? Fulfillment issues E & O dynamics Vendor List • Product type – residential v commercial • Geographic coverage • • • • National Urban coverage Rural coverage Appraiser- v broker-centric • Branch from national appraisal firm • Local vendor based approach Some Options Name Type Coverage Best Concentration Performed by Boxwood Comml Natl Urban/Suburb Appraiser Mountainseed Comml Natl Urban/Suburb Appraiser Clear Capital Comml & Res Natl Urb/Sub/Rural Broker Collateral Analytics Res 1-4 Natl Urban/Suburb AVM Corelogic Comml & Res Natl Urb/Sub/Rural Appr Brkr AVM DataQuick Res 1-4 Natl Urban/Suburb Appr Brkr AVM First American Comml & Res Natl Urban/Suburb Appr Brkr AVM Inside Valuation Comml & Res Natl Urban/Suburb Broker ProTeck Res 1-4 Natl Urb/Sub/Rural Appr Brkr AVM Valligent Res 1-4 Natl Urban/Suburb Appr ValueNet Res 1-4 Natl Urban/Suburb Appr AVM Veros Res 1-4 Natl AVM Comparison Methods 1334928.3 -- -- 1004000579 2/21/2013 Y,N 121026-52344A 121026-52450A 10/26/2012 YY 11/19/2012 Sanitized, MN 2/12/2012 Y,Y 2/12/2012 Y,Y 12/23/2011 Y,Y Y,Y,Y,Y,Y Y,Y,Y,Y,Y Y,Y,Y,Y,Y Y,Y,Y,Y,Y Y,Y,Y,Y,Y Y,Y,Y,Y,Y Y N Y - narrative N N Y N N N N N N Y Y Y Y Y Y N N N N N N Y Y Y Y Y Y N N Y Y Y Y Sep Rpt Sep Rpt Y Y Y Y Y - inf/eq/sup N - no analysis Y - ranking Y - AVM Y - AVM Y - ranking Y - comps Y - comps Y - comps Y Y Y - comps N N Y Y Y Y N N Y Sep Rpt Sep Rpt N N N Sep Rpt Sep Rpt Y Y - MLS Y Y Y Y Y - MLS Y Y Y Y Y - MLS Y Y Y Y Y - MLS Y Y Y Neighborhood Description Local Market Conditions Preparer Name & Signature N N N Sep Rpt Sep Rpt N Y Y Y Y Y N Y Y N Y Y Y Order portal avail? Y Y Y BOK, Soverign, Bank of West; State Farm Y Y Order # Date Property Location Property Description Property Current Use Property Projected Use Property Market Value Est "As-Is" Property Zoning Designation Effective Date of Value 130219-53537A Address; assmt tax ID Bldg size, bldg age, lot size, bed, bath Bldg Use Type (SFR, MFH, Condo, Offc, Retail, Ind, etc.) Proposed construction or renovation, change in bldg use Must state "value" not "price" Local zoning citation "as is" = Date of Drafting "as proposed" = Future Date Limiting Conditions Physical Condition Confirmation Describe Method of Analysis Supporting Information - Sales Supporting Information - Listings Describe Supplemental Information Indicate External Data Sources Indicate Property Specific Data Indicate Evidence of Property Inspect Property Photos Inspection or not; proof of inspect Unknown -- is a display grid sufficient? Unknown -- is a display grid sufficient? Unknown -- is a display grid sufficient? Sales history; assmt Other banks using? Independent 3rd party validation avail? Validation Frequency? Can product be upgrade to appraisal? Compliance with C & R Optional items: PNC; Suntrust No No NA Monthly to Annual Yes extra fee Yes extra fee Yes extra fee Cert Appr Signs Yes extra fee non-appr revs non-appr revs Cert Appr Signs Evaluation Review Needs • • • • General v specific Checklist only? Part of the purpose is speed/efficiency Drags on velocity • Secondary data confirmations • Secondary corroborating data • Secondary inspections Evaluation Review Questions 1. 2. 3. 4. 5. 6. 7. 8. Adequately identifies the collateral property (e.g. legal description, tax number, street address, etc.)? Adequately describes the collateral’s current and (if applicable) projected use? Provides an estimate of the collateral’s market value in “As is” condition & use & (if applicable) zoning as of the effective date of the Alternate Valuation? Indicates all source(s) used in the analysis, as applicable, to value the property (e.g. external data sources, property specific data, photos, neighborhood description, description of local market conditions)? Describes the analysis that was performed and the supporting information that was used in valuing the property? Evidence of a sufficient property inspection (e.g. photographs, comments)? Includes all other relevant information (described as needed in Comments below)? Report includes preparer name and signature? A Primitive Example Here’s an Interesting Wrinkle This review intends to fulfill Regulation B requirements for Residential 1- to 4-unit dwellings. Transactions secured by 1- to 4-unit dwellings that qualify for an evalution by law may be valued for xxxx by outside providers. The valuation document under review may be labeled “Appraisal;” but because of its limited content, it is reviewed under “evaluation” content standards outlined in the Interagency Appraisal & Evaluation Guidelines (effective Dec-2010). Vendor Management • Fulfilling possibilities • • • • • List of serviced counties How many active vendors How many jobs by county last year Avg fee by county How many fulfillment failures by county • Contracts – NDAs - Processing • The follow-up process • What changes can be incorporated to improve? • Portals & ownership To AMC or Not to AMC • • • • • • • Geographic coverage Panel control Customary & Reasonable Fees Waterfall process - upgrades Portals Invoicing Compliance challenges • Would it just be cheaper to bring it in house? Automated Valuation Modeling • • • • • Products & Options Hit Rates Confidence Intervals & FSD AVM Cascades & Selection Tables Program Benefits • Speed – returns in moments • Cost effective – even in a cascade/waterfall AVM Validation - Challenges • • • • • • • Regulatory compliance – modelling policy Sample spread Sample size Sample cost Sample frequency Validations methods applied to each event Who owns the validation process? Trends in Vendor Mgmt • Fed management requirements are heavy • And getting heavier • Escalating compliance costs • Many banks are considering moving to inhouse programs • More product control • Compliance is no worse than vendor management • Cost is ≤ to vendor management Internal Evaluation Risks • Adequacy of Facts • Sufficient documentation • Potential inaccuracies • Adequacy of Opinions • Recognize all needed elements? • Disregard elements? • Value precision? • Does adequate alone = misleading? • Misleading to whom? • Potential inaccuracies acceptable – big picture • Risk is internal business decision Disclosure Risks • Does it matter if an opinion is not disclosed? • Can a consumer recognize the difference between an evaluation and appraisal? • Consumer wounded ego. • What about credit denial? • Consumer uses a disclosed eval in unintended ways • • • • • Eval walks to another bank – portability Divorce litigation or tax appeal Price representations & negotiations Personal financial statement Estate planning or donation What About Fed/State Compliance • The enforcement referral • The court subpoena • The patchwork quilt of state regulators • Mandatory v non-mandatory states • Scope of practice Issues • State enforcement risk to an internal program • Built-in exemptions • Fed requirements for qualified evaluators • What about AI Ethics/Standards? Comparing possible programs • • • • Outside vendor appraisal Appraisal-centric eval Internal team eval – non licensed Internal team eval –licensed Vendor Appraisal – Conventional Research Intensity Inspection Intensity Results Precision Delivered Cost Fulfillment Probability Threat to State Licensure Vendor Appraisal-Centric Eval Research Intensity Inspection Intensity Results Precision Delivered Cost Fulfillment Probability Threat to State Licensure Internal Eval – Licensed Staff Research Intensity Inspection Intensity Results Precision Delivered Cost Fulfillment Probability Threat to State Licensure Internal Eval – Unlicensed Staff Research Intensity Inspection Intensity Results Precision Delivered Cost Fulfillment Probability Threat to State Licensure Can BPOs Help? • Shifting risks from in-house to ….. • Out-house? • The importance of broker/agent skill • Local knowledge & customary thinking • Can BPOs function as in-lieu evaluations? • The IAG restriction? • Work arounds? • State brokers regulatory prohibitions • NAR lobbying in the States What is NAR Up To? • Challenges from Zillow’s sometimes misleading zestimates • RealtyAlliance v Local MLSs • The RET server: – Who owns the MLS data? What level of access? • RPR – the NAR AVM – 600 MLS servers feed it. – Not making money. Trying to grow • New NAR rules allow broker developed AVMs in all markets • Issuing rules to keep MLS data from being proprietary by local MLS systems • NAR pursuing inroads to allow brokers to prepare evaluations for financial institutions without a potential listing. What are Peers Doing? from CART CART -- How many internal evaluations last year? 2013 45 2014 CART - What does Bank permit for eval support? 2013 40 # of Banks # of Banks 35 30 25 20 15 10 5 1-100 101-1000 101-5000 >5000 0 Volume of Evals CART -- do you use the tax value w adj factor? 50 45 40 2013 2014 35 # of Banks 0 30 25 20 15 10 5 0 Never Up to $250M Up to $500M Up to $1MM Up to $5MM Over $5MM Other 2014 Dodd Frank & Reg B / ECOA • Key New Definitions • Borrower • Transaction • Collateral – what is a dwelling? • Notification v disclosure • Consumer purpose or • Secured by 1-4 Unit Dwelling • The Dwelling Trump Card • Mission creep from law to rule-making Reg B Notice & Disclosure LOAN PURPOSE V COLLATERAL CLASS PROBLEM LOAN PURPOSE COLLATERAL CLASS Consumer Business Reg B Mandated 1-4u dwelling Notice & Disclosure No Notice & Disclosure Reg B Exempt Everything else Notice / No disclosure No Notice / No disclosure Is This a Dwelling? Reg B Impacts Community Banks • Increases financing & liability risk • Forces disclosure of potentially inaccurate but otherwise adequate analysis • Requires deal restructure – financing unsecured to avoid documentation of value • Potentially inaccurate but otherwise adequate can be misconstrued or misused by borrower • Alternative is to order more vendor product to deflect risk • Limits credit access to community borrowers • Inflexible rules don’t sync with the physical collateral world • Added cost of vendor product • Added delay of vendor product Reg B Impacts Consumers • Increases confusion • Blurs distinctions from consumer to biz deals • Treats all “valuations” equally for disclosure • Increases cost • Vendor products more costly than internal products • Prioritizes precision over adequacy • Increases delay • Vendor products take additional time • Risk of fulfillment failure Homebuilder Impacts • Reg B intended to give each of thousands of individual consumers an individual appraisal. but instead …. • Reg B requires a handful of homebuilders to receive thousands of appraisals Hybrid Product – the BPO Eval • Can it “shift the risk” ? • How to qualify the valuation elements? • • • • Opine price or value -- both? IAG limitations Review based qualification of BPO opinion? Separate opinion using BPO elements only? • Convert into bank-issued estimate of value • Who inks it? • Qualification v state compliance risk • Wiggle words - disclaimers • E & O need? BPO Vendor – Evaluation Hybrid Research Intensity Inspection Intensity Results Precision Delivered Cost Fulfillment Probability Threat to State Licensure Appraisers Re-enter the Eval Space • USPAP & state restrictions • Economics • E & O coverage Techtonic Shifts • • • • • • • AI Standards v USPAP FRTs v RET? Where do evaluations fit? Should we cage USPAP into FRT only? HR 5148 (Leutkemeyer TILA High Risk Mortgage Bill) Federal pre-emption – supremacy clause See Gibbons v Ogden 22 US 1 (1824) See Gade v National Solid Wastes Mgmt Assoc 505 US 88.98 (1992) • See Florida Lime v Avocado Growers Inc v Paul 373 US 132, 142-43 (1962) The Pre-Flight Evaluation • The good-housekeeping seal of approval • Post-eval changes to loan – Impact on eval reliability • Ordering a subsequent appraisal – Valuation priority of place Foreclosure Evaluations • What is allowed – de minimis rules. • Risks v benefits of use • Can you manage the inspection? • How to manage accounting needs • Fair Value • Liquidation / Distress Value • Pre-foreclosure requirements • Post-foreclosure surprises The Forensic Evaluation • • • • • Multiple evaluations over time Same asset Allows a measure of market change Retrospective analysis to test an appraisal Showing editorial selection process • What comps were uncovered but not used Evaluation in Portfolio Monitoring • The Nexus of Accuracy v Adequacy • Existing collateral & the subsequent transaction • Transaction cost v flat profitability shift • Floor values on file as adequacy filter • Against current balance? Increasing balance? • Declining value? Physical deterioration? • Methods? • Owner occupied class • Investment class • Special cases Our Friend IRV Inc Value Rate Capitalization Modeling $9,000,000 $8,000,000 7.9% RO 7.0% RO $7,000,000 $6,000,000 $5,000,000 $4,000,000 Annual Expenses NOI $3,000,000 $2,000,000 $1,000,000 $0 2011Q1 2012 2013 Appraisal Data Prior Actual Prior Actual Forecast IRV Example Sales Trends $70.00 $60.00 $50.00 Price $/sf $40.00 $30.00 Orig Sales $20.00 Orig Avg Value Opinion @ Orig Update Sales $10.00 Update Avg Suggested Current Value $0.00 Date Trend Tech Specs APPRAISAL DATA IN FILE -- BUILDING SALES -- UNADJUSTED Address Sale Date Sales Low Sales High Sales Median Sales Avg Sales COV 3/27/07 3/25/09 2/8/08 2/5/08 0.8% Land Area (sf) Bldg Area (sf) Year Built $/SF $13,600,000 834,012 $21,600,000 2,178,000 $19,265,671 1,692,306 $18,722,866 1,494,233 16.2% 41.6% 330,190 500,000 403,960 408,151 16.6% 1981 2007 2007 1998.2 0.6% $ 41.19 $ 53.50 $ 44.79 $ 46.02 10.3% Sale Price Com m ents UPDATED BUILDING SALES Address Sales Low Sales High Sales Median Sales Avg Sales COV Land Area (sf) Bldg Area (sf) Year Built $/SF 12/1/09 $7,657,546 583,704 6/26/13 $23,531,680 3,242,171 10/25/11 $15,750,000 1,300,266 10/4/11 $15,672,307 1,606,602 1.6% 57.8% 76.0% 226,576 500,000 368,944 366,116 37.4% 1961 2009 1988.5 1986.75 1.2% $ 28.12 $ 50.54 $ 41.15 $ 40.24 25.8% Sale Date Sale Price Com m ents Gas Station Validation Gross Profit Multiplier Value Gas Station Performance Multiplier Trending Experiences • How is Reg B working for you? • How are evaluations working for you?