AUDIT REPORT WELLS FARGO HOME MORTGAGE NON-SUPERVISED DIRECT ENDORSEMENT LENDER DES MOINES, IA 2004-KC-1003 July 16, 2004 REGION 7 OFFICE OF AUDIT KANSAS CITY, KS Table of Contents Issue Date July 16, 2004 Audit Case Number 2004-KC-1003 TO: John C. Weicher, Assistant Secretary for Housing-Federal Housing Commissioner, and Chairman, Mortgagee Review Board, H FROM: Ronald J. Hosking, Regional Inspector General for Audit, 7AGA SUBJECT: Wells Fargo Home Mortgage, Des Moines, IA We have completed an audit of Wells Fargo Home Mortgage, a non-supervised direct endorsement lender approved to originate Federal Housing Administration (FHA) insured loans. We selected Wells Fargo for audit because Wells Fargo is the leading lender of FHA loans nationwide; and during 2001 and 2002, Wells Fargo had submitted more late requests for insurance endorsement than any other lender. Our audit objectives were to determine whether Wells Fargo's late requests for endorsement complied with HUD's requirements, and whether Wells Fargo originated FHA insured single-family mortgages according to HUD regulations, procedures, and guidance. We reviewed Wells Fargo’s FHA loans submitted for endorsement for a two-year period to ensure that late requests for endorsement met HUD’s requirements for timely borrower payments before submission to HUD. We concluded that Wells Fargo improperly submitted 2,325 loans, totaling $265,381,849 for late endorsement during that period. In addition, we reviewed Wells Fargo’s underwriting of 74 FHA defaulted loans that were originated and/or sponsored by Wells Fargo under HUD’s 203(b), 203(k) or 234(c) programs. We concluded that Wells Fargo did not originate 61 loans, totaling $6,664,470, in accordance with HUD’s requirements. Our report contains two findings with recommendations requiring action by your office. In accordance with HUD Handbook 2000.06 REV-3, within 60 days please provide us, for each recommendation without management decisions, a status report on: (1) the corrective action taken; (2) the proposed corrective action and the date to be completed, or (3) why action is considered unnecessary. Additional status reports are required at 90 days and 120 days after report issuance for any recommendation without a management decision. Also, please furnish us copies of any correspondence or directives issued because of the audit. Should you or your staff have any questions, please contact me at (913) 551-5870. Table of Contents Management Memorandum THIS PAGE LEFT BLANK INTENTIONALLY 2004-KC-1003 Page ii Table of Contents Executive Summary We have completed an audit of Wells Fargo Home Mortgage (Wells Fargo), a non-supervised direct endorsement lender approved to originate FHA insured loans. We selected Wells Fargo for audit because Wells Fargo is the leading lender of FHA loans nationwide; and during 2001 and 2002, Wells Fargo had submitted more late requests for insurance endorsement than any other FHA lender. Specifically, Wells Fargo originated and/or sponsored 224,930 FHA loans totaling over $25.5 billion in mortgage value in 2001 and 2002, approximately 38% more loans than the next highest lender. Wells Fargo submitted 53,558 of these loans as late requests for insurance endorsement, totaling nearly $6 billion in mortgage value, during this same two-year period. The 53,558 loans were nearly three times that of the next highest lender, which submitted 19,700 late requests for endorsement. Our audit objectives were to determine whether Wells Fargo's late requests for endorsement complied with HUD's requirements, and whether Wells Fargo originated FHA-insured single- family mortgages according to HUD regulations, procedures, and guidance. Wells Fargo Home Mortgage improperly submitted 2,325 Wells Fargo Improperly loans, with mortgages totaling over $265 million, for insurance Submitted Late Requests endorsement when the borrowers had delinquent payments for Endorsement prior to loan submission to HUD. Wells Fargo did not have adequate controls to ensure its employees followed HUD’s requirements regarding late requests for insurance endorsement. These inappropriately submitted loans significantly increased the risk to the FHA insurance fund. Wells Fargo Home Mortgage did not adhere to HUD Wells Fargo Did Not requirements and prudent lending practices when processing Follow HUD 61 of the 74 loans we examined for compliance. The 61 loan Requirements when files contained at least one of the following deficiencies: Processing Loans unsupported assets, unsupported income, inadequate qualifying ratios, inadequate documentation, unallowable fees charged to the borrowers, derogatory credit information, underreported liabilities, potential fraud indicators, and improper approval method followed when using an automated underwriting system. Wells Fargo also improperly submitted 4 of the 74 loans as late requests for insurance endorsement, but did not follow HUD regulations when submitting the insurance requests. The deficiencies occurred because Wells Fargo’s management did not take appropriate action to ensure that staff adhered to HUD/FHA requirements when originating FHA loans. As a result, HUD lacks assurance that the mortgagors qualified for the 61 FHA-insured loans totaling $6,664,470. Page iii 2004-KC-1003 Table of Contents Executive Summary We provided results of our late endorsement testing and loan Coordination Regarding file reviews to Wells Fargo during the audit, and received and Audit Results evaluated its responses. We also held meetings and discussions with Wells Fargo throughout the audit, and held an exit conference with Wells Fargo on May 27, 2004. Wells Fargo provided written comments to our findings on July 2, 2004. We incorporated excerpts of the comments into our report as appropriate. The complete text of the comments is contained in Appendix F. We recommend that the Assistant Secretary for Housing- Recommendations Federal Housing Commissioner, and Chairman, Mortgage Review Board, take appropriate administrative action against Wells Fargo Home Mortgage based on the information contained in the findings. This action should, at a minimum, include requiring indemnification for 2,375 loans. 2004-KC-1003 Page iv Table of Contents Table of Contents Management Memorandum i Executive Summary iii Introduction 1 Findings 1. Wells Fargo Improperly Submitted Late Requests for Endorsement 5 2. Wells Fargo Did Not Follow HUD Requirements when Processing Loans 15 Management Controls 31 Follow Up On Prior Audits 33 Appendices A. Schedule of Questioned Costs and Funds Put to Better Use 35 B. HUD Requirements 37 C. Late Endorsement Scope and Methodology 39 D. Improper Late Requests for Endorsement 43 E. Loan Processing Deficiencies Chart 45 F. Auditee Comments 47 Page v 2004-KC-1003 Table of Contents Abbreviations: FHA – Federal Housing Administration HUD – Department of Housing and Urban Development OIG – Office of Inspector General 2004-KC-1003 Page vi Table of Contents Introduction HUD approved Wells Fargo Home Mortgage, Inc. (Wells Fargo) as a non-supervised direct endorsement lender on May 15, 1987. Wells Fargo originates Federal Housing Administration (FHA) insured loans, Veterans Administration loans, and conventional loans. Between January 1, 2001 and December 31, 2002, Wells Fargo originated and/or sponsored 224,930 loans totaling over $25.5 billion under FHA programs. Wells Fargo Home Mortgage, Inc. is a wholly owned subsidiary of Wells Fargo and Company, and has its headquarters in Des Moines, IA. Wells Fargo has a presence nationwide, through over 1,600 mortgage stores and bank locations. It operates over 40 processing centers nationwide, which perform loan processing, underwriting, and closing processes. It also operates three post-closing processing sites, located in Minneapolis, MN; Frederick, MD; and Bloomington, MN. Wells Fargo originates about 50 percent of its FHA mortgage loans through its own network of mortgage stores and bank locations, and the remaining 50 percent of its business is derived from loans it purchases at some point in the origination process or after the loans are closed and insured. Mortgage loans originated by Wells Fargo’s own network of stores and banks begin with a potential homebuyer completing an application package through a Home Mortgage Consultant. The Consultant gathers documentation to begin the loan consideration process, and provides tentative approval of the borrower, pending further review. The Consultant prepares a loan application package and forwards the documentation to a Mortgage Specialist. The Mortgage Specialist gathers additional information, such as income and credit documentation, and reviews the loan for initial decisions on whether Wells Fargo should approve the loan. The Mortgage Specialist forwards the loan package to an Underwriter, who reviews the loan documents and approves or denies the borrower’s application. Once a loan is approved, the Underwriter forwards the package to a Closing Specialist, who ensures that the loan is ready for closing. Once a loan closes, it is forwarded to one of the three post-closing sites. The Post-Closing Services staff prepares the necessary documents and HUD case binders to send to the appropriate HUD Homeownership Center for review and FHA insurance endorsement. Wells Fargo also purchases loans from mortgage brokers. A potential homebuyer completes an application package through a mortgage broker, and the broker gathers documentation and prepares the loan application package for submission to Wells Fargo. Once Wells Fargo receives the application package, an underwriter reviews the loan documents and approves or declines the application. After the loan closes, the loan file is sent to one of the three post-closing sites for the Post-Closing Services staff to prepare the necessary documents and HUD case binders to send to the appropriate HUD Homeownership Center for review and FHA insurance endorsement. Our audit objectives were to determine whether Wells Audit Objectives Fargo's late requests for endorsement complied with HUD's requirements, and whether Wells Fargo originated Federal Housing Administration-insured single-family mortgages according to HUD regulations, procedures, and guidance. Page 1 2004-KC-1003 Table of Contents Introduction During our audit, we reviewed the automated payment Audit Scope histories for 90,277 loans submitted for endorsement more than 66 days after closing to determine whether the borrowers were delinquent before Wells Fargo submitted the loans for endorsement. Wells Fargo could not provide an additional 1,284 automated payment histories subject to our testing. According to Wells Fargo, these loans were serviced by other servicers; therefore, we were unable to test these loans. We also reviewed all documents in the HUD and Wells Fargo loan files for 74 loans to evaluate Wells Fargo’s compliance with HUD’s loan origination requirements for Federal Housing Administration insured loans. We selected the 74 loans from a population of 11,558 Wells Fargo loans that had gone into default within the first two years to determine whether the files contained adequate support to justify approval of the loans. To achieve our objectives, we extensively relied on computer-processed data provided by Wells Fargo and data contained in HUD’s Single Family Data Warehouse system. We assessed the reliability of this data, including relevant general and application controls and found them to be adequate. We also conducted sufficient tests of the data. Based on these tests and assessments, we concluded that the data was sufficiently reliable to be used in meeting our objectives. Specifically, we relied on the loan payment histories and loan status data contained in Wells Fargo’s systems. We also relied on various dates in HUD’s systems, including loan closing dates, Notice of Rejection dates, and endorsement dates. We used the mortgage amount and claims status from HUD’s systems for information purposes only. We performed our audit work from August 28, 2003 through May 14, 2004. We conducted the fieldwork at Wells Fargo offices in Minneapolis, MN, and Des Moines, IA. The audit covered the period January 1, 2001 through December 31, 2002; and we extended the review, where appropriate, to include other periods. We interviewed HUD management and staff to obtain Audit Methodology background information on Federal Housing Administration requirements and Wells Fargo. We interviewed Wells 2004-KC-1003 Page 2 Table of Contents Introduction Fargo’s management and staff to determine its process for originating Federal Housing Administration insured loans and submitting them for endorsement; and to determine how it processed borrower payments on FHA mortgages. We reviewed HUD’s rules, regulations, and guidance for proper origination and submission of FHA loans. We also reviewed Wells Fargo’s policies and procedures, and training manuals, to gain an understanding of how its processes are supposed to function. We reviewed Wells Fargo’s Quality Control Plan, two years of monthly quality control review reports, and audits of its operations to identify the types of deficiencies disclosed through these reviews. We also reviewed examples of HUD reviews of Wells Fargo loans to understand what kinds of deficiencies HUD had previously reported regarding Wells Fargo. We conducted the audit in accordance with generally accepted government auditing standards. We provided a copy of this report to the Chief Executive Officer and President of Wells Fargo. Page 3 2004-KC-1003 Table of Contents Introduction THIS PAGE LEFT BLANK INTENTIONALLY 2004-KC-1003 Page 4 Table of Contents Finding 1 Wells Fargo Improperly Submitted Late Requests for Endorsement Wells Fargo Home Mortgage (Wells Fargo) improperly submitted 2,325 loans with mortgages totaling over $265 million, for insurance endorsement when the borrowers had delinquent payments prior to loan submission to HUD. Wells Fargo did not have adequate controls to ensure its employees followed HUD’s requirements regarding late requests for insurance endorsement. These inappropriately submitted loans significantly increased the risk to the Federal Housing Administration insurance fund. HUD Handbook 4165.1, Revision 1, requires that loans HUD Requirements submitted for insurance endorsement more than 60 days after closing meet certain late request standards. These standards include ensuring that the borrower has made, within the calendar month due, all loan payments up to the time of submission, or at a minimum, made six consecutive monthly payments within the calendar month due. Appendix B provides details of the HUD late request for endorsement requirements. We considered each loan entered in HUD’s data systems as Loan Universe to Test being received more than 66 days after the loan closed to be subject to the late request for endorsement requirements. This timeframe allows 60 days plus 6 days for HUD processing time and weekend mail time. Using HUD’s Single Family Data Warehouse system, we obtained a list of Wells Fargo’s 224,930 loans having a loan closing date from January 1, 2001 through December 31, 2002. Of these 224,930 loans, we determined that 91,561 loans, totaling $10,867,762,028 in original mortgage amounts, were required to comply with HUD’s late request for endorsement procedures. Appendix C provides details of the universe of loans tested in our review for improperly submitted loans. In performing our tests to determine whether Wells Fargo Testing Methodology complied with HUD’s endorsement requirements, we compared HUD and Wells Fargo loan data. Wells Fargo could not provide the automated payment histories for 1,284 of the 91,561 loans. According to Wells Fargo, these loans were serviced by other servicers; therefore, we were unable to test these loans. For the remaining 90,277 loans, Page 5 2004-KC-1003 Table of Contents Finding 1 we tested for the presence of delinquent payments prior to submission. Appendix C provides details of the methodology used to perform our review for improperly submitted loans. Our automated analysis of the payment histories provided by Improperly Submitted Wells Fargo and endorsement data from HUD’s systems Loans showed that for the 90,277 loans tested, Wells Fargo had submitted 3,371 loans for endorsement even though the borrowers had delinquent payments prior to submission. Subsequent to endorsement, 1,046 of the 3,371 loans were paid in full and no longer represent a risk to the FHA insurance fund. Because these loans are no longer insured, we did not conduct any further research of these loans, and neither did Wells Fargo. The remaining 2,325 loans are still insured and pose a risk to the FHA insurance fund. Appendix D provides a summary of the improper submissions. The following chart shows the delinquency status of the 2,325 loans, according to Wells Fargo, as of March 31, 2004: Number Mortgage Percent Delinquency Status of Loans Amount of Total Current 788 $85,962,013 33.89% 0 to 30 Days 362 $41,694,523 15.57% 31 to 60 Days 154 $17,735,387 6.62% 61 to 90 Days 93 $11,353,722 4.00% 91 to 120 Days 74 $8,115,183 3.18% 121 to 150 Days 50 $5,862,515 2.15% 150+ Days 637 $75,645,114 27.40% Paid in Full/Transfer 120 $14,136,759 5.16% No Status Provided 47 $4,876,633 2.03% Totals 2,325 $265,381,849 100.0% As of March 31, 2004, HUD had paid claims on 214 of these loans, with original mortgage amounts of $22,273,669. HUD had sold 60 of these properties, with a loss to HUD of $1,826,668. For the remaining 154 loans having original mortgage amounts totaling $16,732,526, HUD had paid claims and loss mitigation costs of $15,948,230 on these loans, with an indeterminate loss at this point. HUD cannot identify the loss on these loans until the properties are sold. These loans represent an increased risk to the insurance fund. 2004-KC-1003 Page 6 Table of Contents Finding 1 As of March 31, 2004, the insurance had been terminated without a claim on 293 of these loans. Of these 293 loans, 291 loans, totaling $39,148,152 in original mortgages, were refinanced to another FHA loan. Because Wells Fargo improperly submitted the original FHA loan, the improper endorsement also applies to the refinanced loan. Therefore, we have included these 291 loans as improperly endorsed loans. The remaining 2 loans were terminated for reasons other than FHA refinancing; therefore, they no longer represent a risk to the insurance fund. HUD’s data for two additional loans, totaling $278,222 in original mortgage amounts, was inconclusive. The data indicated that these loans were each streamline refinanced to another FHA loan; however, HUD’s data also showed that each original FHA loan was still active. Of the remaining 1,818 loans, 1,814 loans hold active insurance and total $203,154,699 in mortgage amounts. HUD has incurred $228,801 in partial claims and losses on these loans. HUD’s data systems did not show an insurance status on the remaining four loans, totaling $527,107. In the results of the testing for loans improperly submitted for late endorsement, we included four loans that are also reported in Finding 2 as having underwriting deficiencies. These four loans have original mortgage amounts totaling $506,567. Wells Fargo did not have an adequate control environment Inadequate Control to ensure that its employees followed HUD’s submission Environment requirements. In addition, Wells Fargo was not prepared to handle the sudden increase in mortgage volume that occurred in 2001 and 2002 due to low interest rates and unprecedented refinance activity. Wells Fargo’s Post-Closing Department was responsible for submitting closed loans to HUD for endorsement. During the first quarter of 2002, Wells Fargo experienced a backlog of $1.8 billion in uninsured government loans that had been closed for more than 60 days. Wells Fargo significantly increased staffing, including hiring temporary personnel, to handle the mortgage volume increases. According to Wells Fargo senior managers, Page 7 2004-KC-1003 Table of Contents Finding 1 temporary personnel did not receive the in-depth training needed to provide accurate processing and adherence to FHA submission requirements. Further, Wells Fargo’s Post-Closing management team lacked sufficient experience and expertise to respond timely to the significant loan origination volume. Senior managers also told us that during periods of 2001 and 2002 Wells Fargo’s linear loan processing procedures caused a significant time gap between when loans were closed and ultimately submitted to HUD for endorsement. By the time the loan file reached HUD, the payment history was potentially outdated, and Wells Fargo did not have procedures in place to review more updated histories to ensure that borrower payments were current and within the HUD requirements for insurance endorsement. Senior managers described improvements Wells Fargo has made to resolve the deficiencies identified during 2001 and 2002. Specifically, Wells Fargo has implemented procedures to expedite delivery of files to HUD, and for staff to review the most recent payment history to ensure that the borrower’s payments meet the endorsement requirements before sending the file to HUD. In addition, all employees on the government insuring staff continue to receive mandatory HUD endorsement training, and must complete Wells Fargo’s intensive endorsement certification training. Wells Fargo has also implemented a document imaging system for processing loans, which allows parallel and quicker processing of FHA loans. Wells Fargo believes the automated process is more reliable and less likely to involve human errors of sending case files to HUD for endorsement that do not meet HUD’s requirements. Wells Fargo needs to indemnify HUD for the loans improperly submitted, and implement controls that provide assurance employees are following HUD’s requirements when submitting loans for endorsement. The indemnifications should include protecting HUD against future losses from the 2,105 loans that are actively insured. The indemnifications should also include the two loans with an inconclusive status, as well as the four loans with an unknown insurance status, if HUD determines that any of these six are actively insured. Wells Fargo should also 2004-KC-1003 Page 8 Table of Contents Finding 1 reimburse HUD for losses on the 214 loans for which HUD has already paid a claim. Auditee Comments Excerpts from Wells Fargo’s comments on our draft finding follow. Appendix F contains the complete text of the comments. “ . . . WFHM takes pride in our ability to partner with HUD in providing funding to FHA customers. We also want to insure and have complete confidence that controls, policies and procedures are in place to protect HUD and Wells Fargo from any unknown or unpredictable liability. WFHM has made significant strides in improving all of our critical processes since 2001 and 2002. And, we will continue to monitor and embrace improvements in the area of insuring. “ . . . Throughout 2001 and 2002, WFHM was aware of and had procedures in place to comply with HUD’s late endorsement payment history requirements. However, during that period record loan volume presented WFHM and the entire mortgage industry with unprecedented challenges that resulted in periods of backlogs. “The explanation for most of the late endorsement payment history errors has been, simply, human error in the frequently complex task of analyzing payment histories. The extreme volume demands significantly increased our hiring and staffing, and necessitated the hiring of many temporary staff. . . . “We also agree that our processing procedures, in place at the time, added to our backlogs. . . .As a result, a secondary reason for the errors in submitting late endorsements to HUD were delays in WFHM processes that resulted in a gap that allowed some borrowers to miss a payment after we had analyzed the case but prior to HUD receiving the submission. . . . “The increased volumes immediately put Post Closing into action to develop and execute enhanced training programs, as well as technology, process and organizational design changes to enhance and greatly improve the process from a manual and linear to automated and parallel processes. . . . Page 9 2004-KC-1003 Table of Contents Finding 1 “These changes have been successfully implemented, resulting in significant improvements in our processes and the protection afforded HUD against unexpected losses. . . . “The number of loans cited in the OIG’s finding is based on a preliminary assessment that the computerized data used was sufficiently reliable for meeting the objectives of the audit. However, during the audit proceedings it became clear that a final determination of whether a loan was submitted properly and within the requirements of HUD could not be accurately made by the analysis of the HUD computerized data alone. The HUD computerized software used was not able to fully capture the payment history analysis information and/or the data available in the systems was incomplete. In many cases, supplementing the CPH information was required before an accurate conclusion could be reached. “Based on our additional analysis, we agree with the finding in the report that some of the loans submitted by WFHM are not insurable. The further research on the CPH analysis showed that 944 of the loans—or less than 30 percent of those originally cited by the OIG report—did not meet the HUD requirements at time of submission nor do they meet the subsequent payment requirements thereafter. As a result, WFHM needs to work with HUD to establish indemnification parameters for these loans improperly submitted. “We do not agree, however, that the other loans identified by the OIG require full indemnification from Wells Fargo, as our further analysis to date has provided us with the following results: o We have determined that while 876 of the loans identified during the audit did not meet the HUD requirements at time of submission, they did meet the six-month sequential current payment requirement thereafter. As a result, they would have been appropriately approved for insurance if resubmitted to HUD at that time and should not be subject to indemnification from WFHM. o Without further information and analysis of the 361 refinanced FHA loans identified in the OIG report, 2004-KC-1003 Page 10 Table of Contents Finding 1 we cannot state definitively at this time if Wells Fargo should be responsible for indemnification. The report states that because WFHM improperly submitted the original FHA loan, the improper endorsement would also apply to the refinanced loan. However, our in-depth research and analysis on several of these refinanced loans has brought to light a number of factors that may prove that conclusion to be not entirely true, including: This group includes loans that were not FHA Streamline Refinances but rather FHA Rate/Term or FHA Cash-Out Refinances. This group includes loans that are now paid in full and no longer present risk to the FHA insurance fund. This group includes loans that were on the indemnification list twice; once for the original loan and once for the refinanced loan. Given the examples mentioned here, we believe these loans will need to be reviewed on a loan-by- loan basis to determine indemnification responsibilities. o Finally, we are in the process of conducting an in- depth analysis of the CPH information on the remaining loans. The majority of this remaining group was service released prior to endorsement, requiring Wells Fargo to work with many external loan servicers to pull complete payment histories. . . o “As we complete this research and pull internal and external CPH information on these loans, WFHM will work directly with HUD officials on a loan-by- loan basis to finalize the insurability or indemnification of this remaining population.” OIG Evaluation of We commend Wells Fargo for its commitment to ensure Auditee Comments that it has controls, policies, and procedures in place to protect HUD’s interests. Based on the initiatives Wells Fargo described to us during the audit and in its formal Page 11 2004-KC-1003 Table of Contents Finding 1 response, it seems that Wells Fargo is improving its processes for submitting FHA loans to HUD for endorsement. As for the data testing issues, we recognize that a significant amount of supplemental research by both Wells Fargo and our office was necessary to verify the automated data testing in many cases. We maintain that the data used was sufficiently reliable to identify exceptions, which was the objective of the testing. However, other factors affected the accuracy of the preliminary results. Major factors, such as incomplete automated payment histories, the misapplication of borrower payments, and instances of incorrect loan setup by Wells Fargo caused false positives in the test results. Therefore, supplemental research was necessary to verify the preliminary assessments in many cases. Wells Fargo agreed that it had improperly submitted loans to HUD for endorsement, and should indemnify these loans, but did not agree with the total number of loans we questioned in this finding. In particular, Wells Fargo did not agree that it should fully indemnify 876 loans that had late payments at the time of submission. Wells Fargo believes that if it had held these loans and not submitted them for endorsement until the borrower had made six timely payments, the loans would have met HUD’s six- month rule and qualified for endorsement. We do not agree with this approach. While these loans may have qualified for endorsement at some point in the life of the loan, HUD requires that the loan be current at the time of submission, and these 876 loans were not. Therefore, these 876 loans were improperly submitted and should be indemnified. As for the refinanced FHA loans, we conducted additional testing of the 361 loans and identified 70 loans that were, in fact, properly submitted. We have removed these 70 loans from this finding, but continue to include the remaining 291 refinanced FHA loans. Wells Fargo stated that it was continuing to analyze the payment history information on the remaining loans, and would work with HUD on a loan-by-loan basis to finalize the insurability or indemnification of these loans. We encourage Wells Fargo to follow through with this process, 2004-KC-1003 Page 12 Table of Contents Finding 1 and work with HUD for a final resolution to all of the improperly endorsed loans identified in this report. Recommendations We recommend that the Assistant Secretary for Housing- Federal Housing Commissioner, Chairman, Mortgagee Review Board: 1A. Take appropriate administrative action against Wells Fargo for not complying with HUD’s requirements, including requiring Wells Fargo to indemnify 2,105 loans with active insurance, totaling $242,302,851; including $228,801 in partial claims and losses incurred on these loans. Also, indemnify any of the two loans (totaling $278,222), and the four loans (totaling $527,107), that HUD determines are actively insured. 1B. Take appropriate administrative action against Wells Fargo for 214 loans for which HUD has already paid a claim, including requiring Wells Fargo to reimburse HUD for these claims and related losses. HUD has paid $15,948,230 for 154 properties not yet sold, and incurred losses of $1,826,668 on the 60 sold properties. 1C. Verify that Wells Fargo has developed and implemented improved controls over FHA loans to ensure that its employees follow HUD’s submission procedures for late requests for endorsement. Page 13 2004-KC-1003 Table of Contents Finding 1 THIS PAGE LEFT BLANK INTENTIONALLY 2004-KC-1003 Page 14 Table of Contents Finding 2 Wells Fargo Did Not Follow HUD Requirements when Processing Loans Wells Fargo Home Mortgage (Wells Fargo) did not adhere to HUD requirements and prudent lending practices when processing 61 of the 74 loans we examined for compliance. The 61 loan files contained potential fraud indicators, unsupported income, unsupported assets, underreported liabilities, inadequate documentation, derogatory credit information, inadequate qualifying ratios, unallowable fees charged to the borrowers, and improper approvals from an automated underwriting system. Wells Fargo also improperly submitted 4 of the 74 loans as late requests for insurance endorsement, but did not follow HUD regulations when submitting the insurance requests. The deficiencies occurred because Wells Fargo’s management did not take appropriate action to ensure that staff adhered to HUD/FHA requirements when originating FHA loans. As a result, HUD lacks assurance that the mortgagors qualified for the 61 FHA-insured loans totaling $6,664,470. HUD Requirements HUD Handbook 4155.1, Revision 4, Change 1, “Mortgage Credit Analysis for Mortgage Insurance on One-to-Four- Family Properties” requires mortgagees to determine the borrowers’ ability and willingness to repay the mortgage debt, and thus, limit the probability of default or collection difficulties. Four major elements are typically evaluated in assessing a borrower's ability and willingness to repay the mortgage debt: o Stability and adequacy of income; o Funds to close; o Credit history; and o Qualifying ratios and compensating factors. Chapter 3-1 of the same HUD handbook notifies mortgagees that HUD expects the application package to contain sufficient documentation to support the lender's decision to approve the mortgage loan. Appendix E details the deficiencies identified on each loan reviewed, including the specific HUD requirements not met when processing the loan. HUD Handbook 4165.1, Revision 1, “Endorsement for Insurance for Home Mortgage Programs (Single-Family),” requires the lender to follow late request for endorsement procedures if the mortgage is submitted to HUD for endorsement more than 60 days after closing. Page 15 2004-KC-1003 Table of Contents Finding 2 Our examination of 74 loans totaling $7,991,071 originated Loans Did Not Comply by Wells Fargo from January 1, 2001 through December 31, with HUD Requirements 2002 disclosed origination deficiencies in 61 of the 74 cases. Wells Fargo’s loan files contained at least one of the following deficiencies: unsupported assets, unsupported income, inadequate qualifying ratios, inadequate documentation, unallowable fees charged to the borrowers, derogatory credit information, underreported liabilities, potential fraud indicators, and improper approval method followed when using an automated underwriting system. Detailed descriptions of the deficiencies noted are presented below, and a table summarizing the deficiencies is presented later in the Finding. Wells Fargo’s loan files contained insufficient verification of Unsupported Assets the assets that the borrower(s) claimed to close the mortgage loan. HUD Handbook 4155.1, Revision 4, Change 1, Chapter 2-10 requires mortgagees to verify all funds for the borrowers’ investment in the property. Wells Fargo did not sufficiently verify the assets needed to close for 37 of the 74 loan files reviewed. The majority of the deficiencies related to insufficient verification of gift funds, cash on hand, or funds held in bank accounts. For example, in two instances, Wells Fargo entered unsupported depository funds in Loan Prospector (an automated underwriting system), and gained approval for the loans. For FHA case #137-1559390, the Loan Prospector Feedback Certificate showed that Wells Fargo reported $2,000 in depository funds held by the borrower. However, Wells Fargo did not obtain bank statements to support that the borrower possessed the $2,000. For FHA case #251-2829840, Wells Fargo overstated the depository assets. The co-borrower claimed $921 in bank funds, confirmed by a bank statement dated 5/10/02. However, Wells Fargo obtained the subsequent bank statement, dated 6/10/02, which showed a balance of only $78. The Loan Prospector evaluation of 7/11/02 used $920 (in addition to other bank funds) to approve the loan even though more recent documentation held by Wells Fargo showed that the bank balance was only $78. Wells Fargo approved borrowers for FHA mortgages using Unsupported Income calculated monthly income that the loan file documentation 2004-KC-1003 Page 16 Table of Contents Finding 2 did not support. HUD Handbook 4155.1, Revision 4, Change 1 requires mortgagees to establish the anticipated amount of income, and likelihood of its continuance, to determine the borrowers’ capacity to repay the mortgage debt. Mortgagees may not use any income in calculating the borrower’s income ratios that cannot be verified, is not stable, or will not continue. Wells Fargo did not obtain sufficient evidence to support the monthly income used to qualify the borrowers in 36 of the 74 loan files reviewed. Often the income was calculated incorrectly or there were insufficient pay stubs or verifications of employment presented in the loan file. For example, FHA case #281-2811892 showed that Wells Fargo approved the loan based on a calculated monthly income of $2,870. Wells Fargo computed the income using 26 pay periods in a calendar year, however, the borrower was paid bi-monthly, on the 15th and 30th of the month. Employees paid on such a pay schedule incur only 24 (not 26) pay periods in a year. Using the incorrect calculation method, Wells Fargo overstated the borrower’s income by more than $200 per month. Wells Fargo’s loan files showed the borrowers’ qualifying Inadequate ratios exceeded HUD’s allowable limits. HUD Handbook Qualifying Ratios 4155.1, Revision 4, Change 1, Chapter 2-12 allows a mortgage payment-to-income ratio of 29 percent and a total debt-to-income ratio of 41 percent; however, those loans approved by Loan Prospector do not have to meet these criteria. Chapter 2-13 lists compensating factors lenders may use to allow borrowers to exceed the ratio limits, if the compensating factors are specifically identified and properly documented in the case file. In 30 of the 74 loans reviewed, excluding loans approved by Loan Prospector, we identified ratios that exceeded the allowable limits. Mortgagees may allow borrowers to exceed the ratio limits if the lender can demonstrate adequate compensating factors; however, in these 30 cases, Wells Fargo either did not document any compensating factors or the compensating factors presented were not adequate. For example, FHA case #271-8771051, Wells Fargo manually underwrote and approved the loan even though the borrower’s housing ratio of 36 percent and debt ratio of Page 17 2004-KC-1003 Table of Contents Finding 2 47 percent significantly exceeded HUD’s limits of 29 and 41 percent respectively. The underwriter did not document any compensating factors for allowing the borrower to exceed HUD’s financial ratio limits. Wells Fargo did not properly document decisions made and Inadequate actions taken when processing the loan documents. HUD Documentation Handbook 4000.4, Revision 1, Change 2, Chapter 2-5 identifies one of the underwriter's specific responsibilities as the coordination of all phases of the underwriting of the mortgage loan. This enables the underwriter to ensure that prudent underwriting procedures were followed. The handbook also requires the mortgagee’s underwriter(s) to personally review the application documents and certify that the documents are in compliance with the applicable requirements. HUD Handbook 4155.1, Revision 4, Chapter 3-1 notifies the mortgagee that HUD expects the application package to contain sufficient documentation to support the lender's decision to approve the mortgage loan. The loan files reviewed, both those files submitted to HUD and those maintained by Wells Fargo, contained incomplete, incorrect, and outdated documentation; and in many instances did not contain the documentation necessary to demonstrate that Wells Fargo processed the mortgage loan in accordance with HUD requirements. We identified material deficiencies of inadequate documentation in 21 of the 74 loan files reviewed. For example, Wells Fargo processed FHA case #105- 0040061 even though multiple loan documents contained discrepancies regarding the property address purchased and endorsed. HUD’s systems showed the insured address as “532” Kathwood Drive, as did the HUD-1 Settlement Statement, sales contract, appraisal, building permits and inspections, builder warranty, and the Conditional Commitment for Direct Endorsement. However, the mortgage note, Application and Addendum, non-profit gift letter, Director Endorsement Approval, and the late request for endorsement letter to HUD identified the address as “534” Kathwood Drive. The loan file contained no evidence that Wells Fargo recognized and resolved the discrepancy. Therefore, it is unclear what property is related to the mortgage insured by FHA. 2004-KC-1003 Page 18 Table of Contents Finding 2 While we are questioning only those loans with material documentation deficiencies, we noted many instances in which the loan file Wells Fargo submitted to HUD did not include required documentation. However, the loan file Wells Fargo maintains contained documentation supporting that Wells Fargo had performed the required procedure(s). For example, HUD Handbook 4155.1, Revision 4, Paragraph 2-5-A requires lenders to examine HUD's "Limited Denial of Participation List" and the governmentwide General Services Administration List of Parties Excluded from Federal Procurement or Nonprocurement Programs" and document the review on the Mortgage Credit Analysis Worksheet submitted to HUD. In 9 of the 74 files reviewed, the HUD file contained no evidence that Wells Fargo reviewed the lists; however, the Wells Fargo file contained evidence that it had performed the required reviews. Wells Fargo charged borrowers fees that were specifically Unallowable Fees prohibited on FHA mortgage loan closings. HUD Charged to Borrowers Homeownership Center Reference Guide, Chapter 2-15, "Closing Costs and Other Fees" notifies mortgagees that all closing costs associated with a HUD-insured loan, including Paid Outside of Closing items, must be itemized on the HUD- 1 Settlement Statement for Real Estate Settlement Procedures Act compliance. The reference guide provides a list of typical settlement/closing fees and identifies whether the individual fees are allowable on FHA loan closings. HUD-1 Settlement Statements on 16 of the 74 loans reviewed identified charges to the borrowers that were not allowed by HUD regulations, or were not adequately explained so as to allow a reviewer of the loan package to know whether the fee was allowable. Borrowers paid such fees as: government administrative fees, administrative fees, and tax service and desk review fees to Wells Fargo. Borrowers also incurred charges for: administrative fees paid to the seller’s realtor; and express package handling, delivery/courier fees, and fax/wire fees on non-refinance loans. For example, the HUD-1 Settlement Statement for FHA case #052-2091042 showed that the borrower paid a $400 government administrative fee to Wells Fargo. Because the borrower also paid origination fees, Wells Fargo should not Page 19 2004-KC-1003 Table of Contents Finding 2 have charged separate fees for document preparation, such as the government administrative fee. Wells Fargo did not properly evaluate the borrowers’ past Derogatory credit performance and ensure that borrowers demonstrated Credit Information financial responsibility. HUD Handbook 4155.1, Revision 4, Change 1, Chapter 2-3 identifies past credit performance of the borrower(s) as the most useful guide in determining the attitude toward credit obligations that will govern the borrowers’ future actions. Specifically, Wells Fargo did not adequately verify that significant credit deficiencies were resolved before approving and closing the FHA loans. We identified deficiencies such as unpaid judgments, State tax liens, and significant delinquencies on federal student loans. For example, FHA case #071-0911907 contained a credit history revealing two civil judgments and one State tax lien filed against the borrower; however, Wells Fargo did not adequately verify that the borrower had satisfied the debts. The only documentation in the case file was a letter from the borrower claiming that the State tax lien was paid in full, and the letter did not address the civil judgments. We also noted another loan in which the borrower was participating in a formal debt management plan during the time of the FHA loan processing. The borrower’s credit history showed multiple historical collection accounts being repaid through the repayment agreement, which began on March 15, 2001. The repayment agreement required specific approval from the debt assistance agency before the borrower was to incur any new debt. However, Wells Fargo did not verify that the borrower had either satisfied the debts of the agreement, nor did it obtain written approval from the debt assistance agency for the borrower to incur the new debt of the FHA loan. Wells Fargo’s loan files contained evidence of liabilities that Underreported Liabilities Wells Fargo did not consider when approving the borrowers for the mortgage loan. HUD Handbook 4155.1, Revision 4, Change 1, Chapter 2-11 requires mortgagees to consider all recurring obligations, contingent liabilities, and projected obligations that meet HUD’s specific stipulations when evaluating a loan application. 2004-KC-1003 Page 20 Table of Contents Finding 2 Wells Fargo did not consider all of the borrowers’ recurring, contingent, and projected liabilities when approving 8 of the 74 loans reviewed. For example, FHA case #045-5590735 showed that the borrower had four outstanding federal student loans. HUD requires lenders to include any debt payments scheduled to begin within 12 months of the FHA loan closing in the evaluation of the borrower’s liabilities. However, Wells Fargo did not include debts of $404 per month in two student loan repayments due to begin within 11 and 11 ½ months of the FHA loan closing. Wells Fargo’s loan files contained discrepancies that could be Potential Fraud Indicators indicative of fraud. HUD Handbook 4000.4, Revision 1, Not Resolved Change 2, Chapter 2-4-C requires mortgagees to employ underwriters who assume responsibility for awareness of the warning signs that may indicate irregularities, and an ability to detect fraud, as well as the responsibility that underwriting decisions are performed with due diligence in a prudent manner. Social Security Number discrepancies or credit report alerts related to the legitimacy of the Social Security Number provided by the borrower were evident in 4 of the 74 loan files reviewed. Although there may have been a legitimate explanation for these discrepancies, Wells Fargo did not follow up to determine whether the borrower applied for the mortgage using a proper Social Security Number. For example, FHA case #381-6156610 contained documents on which the borrower used two different Social Security Numbers. The pay stubs provided to support the borrower’s income showed a different Social Security Number than the credit report and other loan documents. Wells Fargo provided no evidence that it had researched and adequately resolved the discrepancy. In addition, the loan file did not contain a copy of the borrower’s Social Security Administration card, or any other verification of the proper Social Security Number. Wells Fargo improperly processed loans and may have Improper Loan Approval used abbreviated underwriting procedures when it Method Followed when incorrectly considered the loans approved by Loan Using an Automated Prospector, causing less scrutiny of the loans than required. Underwriting System HUD Mortgagee Letter 98-14 approved the use of Loan Page 21 2004-KC-1003 Table of Contents Finding 2 Prospector on FHA mortgage applications. Loan Prospector either concludes that the borrower’s credit and capacity are acceptable or refers the loan application to an individual underwriter for further consideration. Wells Fargo represented to HUD that Loan Prospector approved 2 of the 74 loans when the loan documentation showed that Loan Prospector had not approved the loans. Without Loan Prospector approval, Wells Fargo was required to follow traditional, more stringent underwriting procedures in processing the loan. Loan Prospector had not approved the loans, but referred them to an underwriter for traditional review. In both cases, FHA case #093-5284768 and #491- 7350357, Loan Prospector rated the loans as “Refer” decisions, not “Accept” decisions. Wells Fargo improperly submitted late requests for Improper Late Requests insurance endorsement on 4 of the 74 loans. This type of for Endorsement deficiency is discussed in more detail in Finding 1. The following table summarizes the individual categories of loan deficiencies previously described. Number of Percent of Type of Non-Compliance Instances Loans Potential Fraud Indicators 4 5.4% Unsupported Income 36 48.6% Unsupported Assets 37 50% Underreported Liabilities 8 10.8% Inadequate Documentation 21 28.4% Derogatory Credit Information 11 14.9% Inadequate Qualifying Ratios 30 40.5% Unallowable Charges to Borrower 16 21.6% Improper Approval from Automated Underwriting 2 2.7% Improper Late Requests for Endorsement 4 5.4% The deficiencies in the above chart are not independent of one another. The counts in the chart do not total the 61 loans that had at least one material deficiency because many of the loan files contained more than one deficiency. Appendix E provides a chart of loan processing deficiencies of loans with material deficiencies. 2004-KC-1003 Page 22 Table of Contents Finding 2 Wells Fargo management did not take appropriate action to Inadequate Resolution to ensure that its staff adhered to HUD requirements when Identified Deficiencies originating FHA loans and submitting them for insurance endorsement. During 2001 and 2002, Wells Fargo quality control staff continually informed management of material loan origination deficiencies; however, management did not take quick and effective measures to resolve the deficiencies. HUD Handbook 4060.1, Revision 1, Chapter 6 requires mortgagees to have and maintain a written Quality Control Plan which provides for internal or external audits, or other independent reviews, of the mortgagee’s origination and servicing of insured mortgages. The Plan must also provide for periodic reports for senior management, which identify areas of deficiency. Senior management must initiate prompt and effective corrective measures to eliminate the deficiencies. Wells Fargo had a Quality Control Plan in place, and in accordance with its plan, conducted 66 monthly quality control reviews of FHA loans in 2001 and 2002; and reported the material deficiencies to management. Wells Fargo defined material deficiencies as significant deviations from the specific loan program parameters under which the loan was originated. Also, material deficiencies existed if the loan contained significant risk factors affecting the underwriting decision and/or contained misrepresentations. The following chart depicts the overall significance of the number of loans reviewed and the material deficiencies identified by quality control staff during the two-year period: Loans with Percent of Loans Material Loans Type of Review Reviewed Deficiencies Deficient Early Payment Defaults 601 364 61% Randomly Selected 5,725 1,221 21% Joint Ventures 279 52 19% Wells Fargo experienced the loan deficiencies for multiple reasons, including its reliance on manual processes that Page 23 2004-KC-1003 Table of Contents Finding 2 allowed inconsistent adherence to policies and procedures, its inability to identify the individual employees responsible for the poor loan originations, and the lack of enforcement of its expected adherence to Wells Fargo’s standards. Because Wells Fargo management did not take quick and adequate action to resolve the deficiencies, the poor originations increased when the mortgage industry experienced a significant rise in loan activity, beginning in mid-2001 and continuing into 2002. To manage the increase in loan activity, Wells Fargo relied on temporary, inadequately trained staff to support operations during this period, exacerbating the problems it was already experiencing. The monthly quality control reviews indicated that material loan origination deficiencies actually increased, rather than decreased, as the quality control staff continued to report problems to management. The following chart demonstrates the percentage of loans reviewed that had material deficiencies in 2001 as opposed to 2002: Type of Review 2001 2002 Early Payment Defaults 54% 66% Randomly Selected 18% 24% Senior managers responsible for overseeing the loan origination/underwriting processes at Wells Fargo offered additional reasons for loan originations that did not meet FHA requirements, including staff using policies and procedures that were not always clear or followed, and staff lacking the in-depth knowledge of unique aspects of the FHA loan programs. Wells Fargo told us it has taken steps to improve its loan origination processes. The senior managers provided extensive information regarding the actions Wells Fargo has taken to alleviate the loan origination deficiencies. According to Wells Fargo, it has improved its policies and procedures, increased its training of personnel, and tied its underwriting managers’ compensation to the error rates identified on their loans (i.e. pay for performance). Wells Fargo has also put processes in place that identify loan deficiencies quicker and more in depth, allowing Wells Fargo to act to correct the 2004-KC-1003 Page 24 Table of Contents Finding 2 problems immediately; and formed management teams to conduct on-site training sessions for personnel in its operations centers nationwide. Further, it has implemented recurring teleconferences and round-table meetings among management personnel to discuss loan processing issues and to develop action plans aimed at resolving the problems. Inadequate underwriting results in HUD insuring mortgages Impact of Inadequate that do not meet the minimum requirements. Improperly Underwriting originated loans increase the risk of loss to the HUD mortgage insurance fund. At the time we selected our sample, all 74 loans selected had been in default within two years of origination. The loan delinquency status on the 61 loans, per Wells Fargo as of March 31, 2004, is presented in the following table: Number Mortgage Percent Delinquency Status of Loans Amount of Total Current 6 $823,425 9.84% 0 to 30 Days 3 $471,201 4.92% 31 to 60 Days 5 $436,492 8.20% 61 to 90 Days 4 $363,778 6.56% 91 to 120 Days 2 $194,825 3.28% 121 to 150 Days 5 $507,330 8.20% 150+ Days 26 $2,577,388 42.62% Paid in Full/Transfer 5 $754,217 8.20% No Status Provided 5 $535,814 8.20% Totals 61 $6,664,470 100.0% This finding includes four loans with original mortgage amounts totaling $506,567 that are also reported as improper late requests for endorsement in Finding 1. To account for the overlap, our recommendations for this finding relate only to the other 57 loans with mortgage amounts totaling $6,157,903. In addition, 7 of the 57 loans, totaling $778,262, have terminated their FHA insurance without a claim, as of March 23, 2004. These seven loans are included in the above chart, but because these loans no longer represent a risk to the insurance fund, we have removed them from our recommendations. The original mortgage amount for the remaining 50 loans is $5,379,641. Page 25 2004-KC-1003 Table of Contents Finding 2 As of March 23, 2004, HUD has paid claims on 18 of the 50 remaining loans. These 18 loans had original mortgage amounts totaling $1,838,786. HUD has sold 4 of the 18 properties, incurring losses of $150,801. Further, HUD has paid claims and related losses of $1,331,639 on the other 14 loans, with an indeterminate overall loss at this point. The overall loss to HUD will not be known until the properties are sold. Wells Fargo needs to indemnify HUD for the 32 loans currently FHA-insured, totaling $3,540,855; and reimburse HUD $19,266 for losses already incurred on these loans. Wells Fargo also needs to reimburse HUD for claims and related losses incurred for the 18 loans for which HUD has paid a claim. Further, Wells Fargo also needs to implement controls that provide assurance that its employees are following HUD requirements when originating loans. Auditee Comments Excerpts from Wells Fargo’s comments on our draft finding follow. Appendix F contains the complete text of the comments. “ . . . While we agree that the error rate reported in the OIG’s sample of 74 defaulted loans would appear to be too high, it needs to be evaluated in this broader industry context. In addition, we are confident that the error rate of a small, adverse sample of 74 defaulted loans is not indicative of the overall quality of WFHM’s FHA originations. “Because of our commitment to leadership in FHA lending, WFHM has taken steps to improve its originations and underwriting processes and controls. We believe that these changes are producing positive results, as measurement and analysis by WFHM’s internal quality control group finds that the error rate in adhering to HUD requirements and prudent lending practices for FHA originations has improved significantly for the 2003-2004 period compared to 2001- 2002. We will continue to take actions necessary to ensure ongoing improvement in the quality of our FHA originations and underwriting processes. “WFHM agrees that we should work with HUD on indemnification parameters on 49 of the 74 loans cited in the study as not having adhered to HUD requirements and 2004-KC-1003 Page 26 Table of Contents Finding 2 prudent lending practices. . . . We believe the remainder of the loans cited by the OIG should require limited or no indemnification by WFHM.” OIG Evaluation of We commend Wells Fargo for its commitment to improving Auditee Comments its originations and underwriting processes. Based on the initiatives Wells Fargo described to us during the audit and in its formal response, it seems that Wells Fargo is making great strides in improving its origination and underwriting processes for FHA loans. Wells Fargo reviewed the 74 loans tested for technical compliance and agreed with the factual content of the findings. Wells Fargo also committed to repaying all of the unallowable fees either to the FHA borrower, or if a claim has been paid, to HUD. Although Wells Fargo agreed that all 61 questioned loans had one or more technical defects, it considered many of the errors to be minor and did not have a material effect on the loan approval and/or the performance of the loan. Wells Fargo identified 12 of the 61 loans that it contends possessed technical infractions that were of little consequence to the loan’s merit, and cited two examples. In the first case, involving deferred student loan repayments, Wells Fargo agreed that it did not follow HUD requirements. Specifically, it did not consider the $404 monthly loan payments beginning within 12 months of the loan closing when evaluating the borrower’s future debts and approving the loan. At the time of the loan closing, the borrower’s two student loans were short of the 12-month requirement by 4 days and 29 days. Wells Fargo does not believe that this specific issue made a material difference in the borrower’s ability to repay the loan. We disagree. The additional recurring debt of $404 per month would have increased the borrower’s debt ratio at the time of closing to 46.5 percent, well above HUD’s total debt limit of 41 percent, and therefore increased the risk of default. In the second case, involving housing and debt ratios, Wells Fargo agreed that it did not follow HUD requirements because the underwriter failed to specifically note compensating factors for exceeding the housing ratio Page 27 2004-KC-1003 Table of Contents Finding 2 guideline by .33 percent. The underwriter approved the loan with a housing ratio of 29.33 percent and a total debt ratio of 29.33 percent. We agree that the housing ratio only slightly exceeds HUD’s limit of 29 percent; however, this was not the only issue on this loan. Wells Fargo increased the borrower’s documented income by 15 percent without following HUD requirements for substantiating the borrower’s eligibility for having income “grossed up.” Specifically, Wells Fargo did not substantiate that the income was not taxable and was likely to continue for the next three years. Using the borrower’s supported monthly income, without the added 15 percent, the borrower’s housing ratio increased to 33.7 percent. HUD limits a borrower’s housing ratio to 29 percent of gross income, and without the unsupported income, the borrower significantly exceeded the HUD limit, and therefore increased the risk of default. Recommendations We recommend that the Assistant Secretary for Housing- Federal Housing Commissioner, Chairman, Mortgagee Review Board: 2A. Take appropriate administrative action against Wells Fargo for not complying with HUD’s requirements, including, requiring Wells Fargo to indemnify HUD for the 32 loans totaling $3,540,855, and any related losses incurred, on the loans in which Wells Fargo did not follow HUD loan origination requirements. HUD has already incurred $19,266 in losses on these loans. 2B. Require Wells Fargo to reimburse HUD for the $1,331,639 in claims paid for the 14 properties not yet sold, and reimburse HUD $150,801 in losses incurred on the 4 sold properties in which Wells Fargo did not follow HUD loan origination requirements. 2C. Verify that Wells Fargo has implemented an effective control environment that prevents Wells Fargo from submitting loans for FHA insurance endorsement that do not meet HUD requirements. 2004-KC-1003 Page 28 Table of Contents Finding 2 THIS PAGE LEFT BLANK INTENTIONALLY 2004-KC-1003 Page 30 Table of Contents Management Controls Management controls include the plan of organization, methods and procedures adopted by management to ensure that its goals are met. Management controls include the processes for planning, organizing, directing, and controlling program operations. They include the systems for measuring, reporting, and monitoring program performance. We determined the following management controls were Relevant Management relevant to our audit objectives: Controls • Controls over submitting loans to HUD for insurance endorsement. • Controls over origination of FHA loans. We assessed the relevant controls identified above. It is a significant weakness if management controls do not provide reasonable assurance that the process for planning, organizing, directing, and controlling program operations will meet an organization’s objectives. Based on our review, we considered the following items Significant Weaknesses significant weaknesses: • Wells Fargo did not have adequate controls to ensure it properly submits closed loans to HUD for endorsement (see Finding 1). • Wells Fargo did not have adequate controls to ensure it originates loans in accordance with HUD’s requirements (see Finding 2). Page 31 2004-KC-1003 Table of Contents Management Controls THIS PAGE LEFT BLANK INTENTIONALLY 2004-KC-1003 Page 32 Table of Contents Follow Up On Prior Audits This is the first Office of Inspector General Audit of Wells Fargo Home Mortgage. Page 33 2004-KC-1003 Table of Contents Follow Up On Prior Audits THIS PAGE LEFT BLANK INTENTIONALLY 2004-KC-1003 Page 34 Table of Contents Appendix A Schedule of Questioned Costs and Funds Put to Better Use Recommendation Type of Questioned Cost Funds Put to Number Ineligible 1/ Unsupported 2/ Better Use 3/ 1A $ 228,801 $242,302,851 1A $ 278,222 1A $ 527,107 1B $ 15,948,230 1B $ 1,826,668 2A $ 19,266 $ 3,540,855 2B $ 1,331,639 2B $ 150,801 1/ Ineligible costs are costs charged to a HUD-financed or HUD-insured program or activity that the auditor believes are not allowable by law, contract or Federal, State or local policies or regulations. 2/ Unsupported costs are costs charged to a HUD-financed or HUD-insured program or activity and eligibility cannot be determined at the time of audit. The costs are not supported by adequate documentation or there is a need for a legal or administrative determination on the eligibility of the costs. Unsupported costs require a future decision by HUD program officials. This decision, in addition to obtaining supporting documentation, might involve a legal interpretation or clarification of Departmental policies and procedures. 3/ Funds Put to Better Use are costs that will not be expended in the future if our recommendations are implemented. For this review, the funds put to better use consist of loans and guarantees not made because of indemnification. Page 35 2004-KC-1003 Table of Contents Appendix A THIS PAGE LEFT BLANK INTENTIONALLY 2004-KC-1003 Page 36 Table of Contents Appendix B HUD Requirements HUD Handbook 4165.1, Revision 1, requires the lender to Mortgagee Requests for follow late request for endorsement procedures if the Late Endorsement mortgage is submitted to HUD for endorsement more than 60 days after closing. The degree of risk to HUD must be no greater than existed at the time of closing in order for the mortgage to be endorsed. The late request procedures require the lender to provide a payment ledger that reflects the payments received, including the payment due for the month in which the case is submitted, if the case is submitted after the 15th of the month. The mortgage payments must not be delinquent when the loan is submitted for endorsement. The payment ledger must cover the entire period from the first payment due date to the date of submission for endorsement. Each payment must be made in the calendar month due. If a payment is made outside the calendar month due, the lender cannot submit the case for endorsement until six consecutive payments have been made within the proper calendar month due. Page 37 2004-KC-1003 Table of Contents Appendix B THIS PAGE LEFT BLANK INTENTIONALLY 2004-KC-1003 Page 38 Table of Contents Appendix C Late Endorsement Scope and Methodology Using HUD’s data systems, we initially identified 224,930 Development of Scope loans, totaling over $25.5 billion in original mortgage amounts, that Wells Fargo either originated or sponsored nationwide, and had a loan closing date from January 1, 2001 through December 31, 2002. The following chart depicts the adjustments made to the initial universe of 224,930 loans identified for testing, and a narrative explanation follows the chart: Number Original of Loans Mortgage Amounts Originated or Sponsored by Wells Fargo from 1/1/01 through 12/31/02 224,930 $25,526,058,486 New Construction Loans 457 $57,612,457 Submitted Before First Payment Due Date 12 $1,435,718 Submitted within 66 Days After Closing 130,037 $14,395,250,180 Home Equity Conversion Loans 1,069 $0 Transferred Prior to Submission 1,487 $168,609,492 Submitted but Not Endorsed 307 $35,388,611 Loans Subject to Late Request for Endorsement Requirements 91,561 $10,867,762,028 Payment Histories Not Provided – Bond Loans 1,159 $109,165,076 Payment Histories Not Provided – New FHA Case Numbers 107 $13,092,377 Payment Histories Not Provided – Other 18 $2,354,270 Loans Tested 90,277 $10,743,150,305 Page 39 2004-KC-1003 Table of Contents Appendix C Of the 224,930 loans in the initial universe, we removed 457 new construction loans and 12 loans submitted for endorsement before the first payment due date because these loans were not subject to the 60-day submission requirement. We further limited our universe to only those loans received by HUD more than 66 days after the loan closed. While HUD requires mortgagees to submit loans for endorsement within 60 days of the loan closing, we allowed six additional days to ensure that we conservatively selected loans for further testing. We allowed six extra days because HUD’s mailroom and endorsement contractors have a total of three business days to process each loan and because any submission may be delayed in the mail for up to three days over a weekend. As a result, for our testing purposes, we considered only those loans submitted more than 66 days after closing as late requests for endorsement. After removing the 130,037 loans submitted within 66 days of closing, 94,424 loans remained as late requests for endorsement. In evaluating the 94,424 loans, we determined that 1,069 of the loans were Home Equity Conversion Mortgages. These loans are not subject to HUD’s direct endorsement procedures; therefore, we removed the loans from the sample. We also identified 1,487 loans in which Wells Fargo had transferred the loan servicing to another lender/servicer before submission for endorsement; therefore, we also removed these loans from our sample. Finally, we identified 307 loans that Wells Fargo submitted for endorsement; however, HUD did not endorse these loans, totaling $35,388,611. Therefore, we removed these loans as well. Wells Fargo could not provide payment histories for 1,284 Loan Payment Histories loans, totaling $124,611,723 in original mortgage Not Available for Testing amounts, that it originated or sponsored during our audit period. Wells Fargo told us that it sold 1,159 of these loans, totaling $109,165,076 to housing agencies as bond- related mortgages. Wells Fargo originated and closed the loans, but transferred the loans to housing agencies for 2004-KC-1003 Page 40 Table of Contents Appendix C servicing. Therefore, Wells Fargo had no automated payment histories for loans sold to housing agencies. Wells Fargo told us that when it submits a late request for endorsement for a loan that was sold to a housing agency, it contacts the servicer of the loan and obtains a hardcopy loan payment history to include in the loan file submitted to HUD for endorsement. We reviewed the official HUD case for a sample of 10 of the 1,159 bond-related loans and confirmed that the 10 loans were related to bond transactions. Wells Fargo had included the hardcopy payment history in the HUD case file submitted for insurance endorsement in 9 of 10 cases. Wells Fargo did not include a late request for endorsement letter or hardcopy payment history in one case file even though the loan should have been submitted as a late request for endorsement. Wells Fargo identified 107 of the 1,284 loans, totaling $13,092,377, as loans that the FHA case number had changed after the loan was originated, and Wells Fargo had not provided the automated payment history because it had not identified the new FHA case number when providing the automated data for our testing. After further research of these 107 loans, Wells Fargo told us that its records showed that 53 of the loans were paid in full, 35 were not endorsed with FHA insurance, 17 were current at endorsement, and the remaining 2 were not current at endorsement, but became current soon after. Wells Fargo did not provide documentation supporting these conclusions for 106 of the loans; therefore, we were not able to further evaluate these loans for proper late request for endorsement. Wells Fargo provided a hardcopy payment history for the remaining loan; however, this loan was never considered an improperly endorsed loan. Wells Fargo was unable to provide automated payment histories for the remaining 18 of the 1,284 loans. As such, we were not able to test these loans, totaling $2,354,270 in mortgage amounts, by using automated testing procedures. Wells Fargo provided hardcopy payment histories for 7 of the 18 loans, but was unable to provide the remaining 11 payment histories. We reviewed the seven hardcopy payment histories and determined that three loans were current at the time of Page 41 2004-KC-1003 Table of Contents Appendix C endorsement, and two were paid in full and no longer FHA insured. Therefore, we did not include these five loans as part of the improperly endorsed loans. The payment history for remaining two loans indicated that the loan was paid in full; however, HUD’s systems show the loan as FHA insured. We did not remove these two loans from the questioned results, relying on the insurance status in HUD’s systems. For the remaining 11 loans for which we did not receive a payment history, Wells Fargo told us that it has no record of having any responsibility for these 11 loans, as either the originator or sponsor, even though HUD’s systems identified Wells Fargo as the sponsor of these loans. We did not include these loans in our questioned results; however, HUD needs to work with Wells Fargo to resolve the sponsorship status of these 11 loans. We tested the remaining 90,277 loans, with original mortgage amounts totaling $10,743,150,305 for improper late requests for endorsement. To test the loans for proper submission, we derived a Methodology of Testing submission date from dates in HUD’s systems. We considered the submission date to be the date HUD began reviewing the loan for insurance endorsement; however, if HUD rejected the loan and returned it to Wells Fargo for correction of deficiencies, we used the date Wells Fargo resubmitted the loan to HUD for review. If HUD’s data did not contain the date Wells Fargo resubmitted the loan for endorsement, we used the endorsement date as the submission date. Our tests also required the use of the loan closing dates to identify those loans submitted to HUD more than 66 days after the loan closed. We compared the closing dates provided by Wells Fargo to those in HUD’s Single Family Data Warehouse and determined that over 26 percent of Wells Fargo’s closing dates differed from those in HUD’s system. We discussed the closing dates with Wells Fargo information systems staff and were told that the closing date in Wells Fargo’s data was not consistent, and did not always reflect the actual date the loan closed. Therefore, we relied on the closing dates in HUD’s systems. 2004-KC-1003 Page 42 Table of Contents Appendix D Improper Late Requests for Endorsement We provided HUD officials and Wells Fargo with spreadsheets of the loans improperly submitted to HUD as late requests for endorsement. Due to the volume, we have not included the detailed spreadsheets in this report, but can provide the spreadsheets upon request. The following table identifies the four categories of late requests for endorsement: Late Missed Payments Payments Gaps Other Totals Number of Loans 2,189 124 11 1 2,325 Original Mortgage Amount $249,340,103 $14,818,428 $1,152,431 $70,887 $265,381,849 Late Payments Loans with a transaction recorded after the month due. The spreadsheet lists the due dates of such transactions for each questioned loan. Missed Payments Loans with no payment history record (due date) for the month of submission. The spreadsheet provides payment records through the month of submission for each questioned loan. Gaps Loans with no payment history record (due date) for the months prior to the month of submission, but there was a due date for the month of submission. The spreadsheet provides payment records through the month of submission for each questioned loan. Other Loans for which Wells Fargo was unable to provide automated payment histories for testing, but provided hardcopy payment histories. No spreadsheets are available for these loans; however, hardcopy histories are available. Page 43 2004-KC-1003 Table of Contents Appendix D THIS PAGE LEFT BLANK INTENTIONALLY 2004-KC-1003 Page 44 Table of Contents Appendix E Loan Processing Deficiencies Chart Improper Lack of Lack of Inadequate Charges Liabilities Potential Improper Approval Support for Support for Qualifying Inadequate Not Derogatory Not Fraud Request to Method FHA Loan # Assets Income Ratios Documents Allowed Credit Reported Indicators Endorse Followed TOTALS 045-5590735 X 1 151-6228253 X 1 071-0911907 X X 2 091-3649290 X X 2 241-6632203 X X 2 161-1983235 X X X 3 231-0737528 X X 2 493-6802905 X 1 249-4457156 X X 2 251-2608849 X X 2 281-2811892 X X X X 4 491-7194198 X X X X 4 023-1027780 X X X X X X 6 048-2629355 X 1 094-4327314 X X 2 093-5264888 X X X X 4 093-5284768 X X X 3 091-3539496 X X 2 011-4889638 X X X 3 249-4520140 X X X 3 381-6226599 X X X X 4 251-2611609 X X X 3 137-1559390 X X X X X 5 251-2829840 X X X X X 5 043-6847650 X X 2 091-3387375 X 1 352-4457607 X 1 381-6283891 X X 2 022-1578743 X X 2 061-2477517 X 1 093-5041515 X X 2 023-0779672 X X X X 4 132-1527561 X X 2 151-6500980 X X X 3 161-1856642 X X 2 222-1550989 X X X 3 249-4485808 X 1 271-8771051 X X X X 4 Page 45 2004-KC-1003 Table of Contents Appendix E Improper Lack of Lack of Inadequate Charges Liabilities Potential Improper Approval Support for Support for Qualifying Inadequate Not Derogatory Not Fraud Request to Method FHA Loan # Assets Income Ratios Documents Allowed Credit Reported Indicators Endorse Followed TOTALS 381-6156610 X X X X 4 491-7182262 X 1 491-7350357 X X X X 4 105-0265801 X X X 3 061-2288048 X X 2 093-5112380 X X X X 4 105-0073677 X X 2 105-0502305 X X 2 271-8538244 X X X X X 5 093-4939095 X X 2 011-4990818 X 1 221-3290973 X X 2 241-6701224 X X X 3 281-2889244 X X X 3 042-7747772 X X 2 042-7791435 X X X X 4 043-6685141 X X X X 4 052-2091042 X X X X 4 411-3391760 X X X X 4 101-9923542 X X X 3 105-0040061 X X X X 4 105-0143601 X X X X X 5 331-1088693 X X X X 4 TOTALS 37 36 30 21 16 11 8 4 4 2 169 ***We provided HUD officials and Wells Fargo with the narrative case studies for each of the 61 questioned loans represented in the chart above. Due to the volume, we have not included the detailed narratives in this report, but can provide the narratives upon request. 2004-KC-1003 Page 46 Table of Contents Appendix F Auditee Comments Page 47 2004-KC-1003 Table of Contents Appendix F 2004-KC-1003 Page 48 Table of Contents Appendix F Page 49 2004-KC-1003 Table of Contents Appendix F 2004-KC-1003 Page 50 Table of Contents Appendix F Page 51 2004-KC-1003 Table of Contents Appendix F 2004-KC-1003 Page 52 Table of Contents Appendix F Page 53 2004-KC-1003 Table of Contents Appendix F 2004-KC-1003 Page 54 Table of Contents
Wells Fargo Home Mortgage Non-Supervised Direct Endorsement Lender, Des Moines, IA
Published by the Department of Housing and Urban Development, Office of Inspector General on 2004-07-16.
Below is a raw (and likely hideous) rendition of the original report. (PDF)