Voglio sapere a riguardo di…

Federal Register / Vol. 86, No. 128 / Thursday, July 8, 2021 / Notices 2.5 Fair Lending The Bureaus fair lending supervision program assesses compliance with the Equal Credit Opportunity Act ECOA 67
and its implementing regulation, Regulation B,68 as well as the Home Mortgage Disclosure Act HMDA 69 and its implementing regulation, Regulation C,70 at banks and nonbanks over which the Bureau has supervisory authority.
Examiners found that supervised institutions engaged in violations of HMDA and Regulation C, and ECOA
and Regulation B.

lotter on DSK11XQN23PROD with NOTICES1

2.5.1 HMDA Examination Findings 2018 & 2019 Data The Bureau continues to examine mortgage originators, including bank and nonbank financial institutions, for compliance with HMDA and its implementing regulation, Regulation C.
Regulation C requires financial institutions to collect and report data regarding applications for covered loans that they receive, covered loans that they originate, and covered loans that they purchase each calendar year.71
Recent examinations identified HMDA
violations due to inaccuracy of HMDA
data submitted by financial institutions, including fields newly added to the HMDA loan application register LAR
beginning in 2018. In October 2015, the CFPB issued a final rule 2015 HMDA
Rule that included changes to the types of institutions that are subject to Regulation C; the types of transactions subject to Regulation C; the specific information that covered institutions are required to collect, record, and report;
and processes for reporting and disclosing data.72 For HMDA data collected on or after January 1, 2018, certain covered institutions were required to collect, record, and report data points newly added or modified by the 2015 HMDA Rule.
Specifically, the 2015 HMDA Rule added new data points for Applicant or Borrower Age, Credit Score, Automated Underwriting System information, Unique Loan Identifier, Property Value, Application Channel, Points and Fees, Borrower-paid Origination Charges, Discount Points, Lender Credits, Loan Term, Prepayment Penalty, Nonamortizing Loan Features, Interest Rate, and Loan Originator Identifier as well as other data points. The 2015 HMDA Rule also modified several existing data 67 15

U.S.C. 16911691f.
CFR part 1002.
69 12 U.S.C. 28012810.
70 12 CFR part 1003.
71 12 CFR 1003.4a.
72 80 FR 66128 Oct. 28, 2015.
68 12

VerDate Sep<11>2014

16:47 Jul 07, 2021

Jkt 253001

points.73 Most of the additions and modifications to the HMDA LAR fields within the 2015 HMDA rule became effective January 1, 2018. Examinations evaluating data reported in 2018 and 2019 were the first examinations in which the Bureau reviewed the accuracy of the data in HMDA LAR
fields added by the 2015 HMDA Rule.
The CFPBs HMDA examinations include transaction testing of a sample of the institutions HMDA LAR and review of its CMS as it relates to HMDA.
Transaction testing consists of comparing a sample of the institutions HMDA LAR to source documents from the loan files corresponding to each LAR entry LAR line or row of the data and assessing whether or not the LAR
entry is accurate. When errors are identified, examiners evaluate the number of errors relative to the resubmission threshold, which is the data accuracy standard used in the CFPBs examinations. Specifically, the HMDA interagency resubmission thresholds provide that in a LAR of more than 500 entries, when the total number of errors in any data field exceeds four, examiners should direct the institution to correct any such data field in the full HMDA LAR and resubmit its HMDA LARs with the corrected fields.74 These resubmission thresholds are included in the CFPBs HMDA examination procedures.75
2.5.2 2018 & 2019 HMDA LAR Errors Examiners identified widespread errors within 2018 HMDA LARs of several covered financial institutions.
To date, examiners have not identified widespread LAR errors within institutions 2019 LARs. In several examinations, examiners identified errors that exceed the HMDA
resubmission thresholds. In general, examiners identified more errors in data fields collected beginning in 2018
pursuant to the 2015 HMDA rule than for other fields. For example, the fields with the highest number of identified errors across several institutions were the newly required Initially Payable to 73 See the CFPB HMDA Summary of Reportable Data chart 2015, https
files.consumerfinance.gov/f/201510_cfpb_hmdasummary-of-reportable-data.pdf.
74 LARs of 500 entries or fewer have a resubmission threshold of three errors. CFPB
Examination Procedures, updated April 1, 2019, available at https files.consumerfinance.gov/f/
documents/cfpb_supervision-and-examinationmanual_hmda-exam-procedures_2019-04.pdf.
75 For more information about CFPB HMDA LAR
transaction testing and samples, refer to the CFPB
HMDA Examination Procedures, updated April 1, 2019, available at https
files.consumerfinance.gov/f/documents/cfpb_
supervision-and-examination-manual_hmda-examprocedures_2019-04.pdf.

PO 00000

Frm 00041

Fmt 4703

Sfmt 4703

36115

Your Institution field and the Debt-toIncome Ratio field.
2.5.3 Root Causes of HMDA Data Errors In several examinations in which examiners identified numerous errors, the root causes of the HMDA violations were deficiencies in the institutions CMS. The CMS deficiencies included the institutions board and management oversight, policies and procedures, training, monitoring and audit, and the institutions service provider oversight.
Many of the widespread or systemic errors related to problems within the institutions data mappingthe data transfers from operations-based systems, such as loan origination systems, to data storage systems that populate the HMDA LARs. For example:
Examiners determined that numerous errors within the Credit Scoring model fields were caused by data transfer deficiencies in which institutions extracted data from credit scoring models then transferred them to systems that reported inaccurate codes and descriptions of the credit scores.
Examiners identified errors within the Rate Spread field and observed that these errors occurred due to data mapping or data transfer deficiencies.
Institutions allowed erroneous software updates within their loan processing systems to result in inaccurate Rate Spread values reported on their HMDA
LARs. Examiners determined that service provider oversight deficiencies resulted in institutions failure to correct the erroneous data transfers.
Examiners identified inaccurate values for the debt-to-income ratio. The institutions acknowledged the errors and stated the fields reported incorrectly were the result of a change made to the programming of their loan origination system.
Many of the widespread or systemic errors were caused by misinterpretation of Regulation C requirements or the institutions specific policy. For example:
Examiners determined that employees at one institution misinterpreted the institutions policies and procedures for calculating the ages of applicants and co-applicants.
Examiners determined that these errors were caused by deficiencies in the institutions monitoring and audit function.
Examiners determined that an institutions senior management misinterpreted HMDA and Regulation C, concluding erroneously that the Origination Charges, Discount Points, and/or Lender Credits fields should be reported as Not applicable. For
E:FRFM08JYN1.SGM

08JYN1