Issue Date July 31, 2008 Audit Report Number 2008-FW-1011 TO: Patricia A. Knight Director, Recovery and Prevention Corps, PQCC FROM: Gerald R. Kirkland Regional Inspector General for Audit, Fort Worth Region, 6AGA SUBJECT: The Dallas Housing Authority, Dallas, Texas, Mismanaged Its Housing Choice Voucher Program HIGHLIGHTS What We Audited and Why As part of our strategic plan objective to assist the U. S. Department of Housing and Urban Development’s (HUD) efforts to reduce rental assistance overpayments, we audited the Dallas Housing Authority’s (Authority) Housing Choice Voucher program (voucher program). Our objective was to determine whether the Authority properly administered its overall voucher program. What We Found The Authority acknowledged its longstanding weaknesses, as previously reported by the Office of Inspector General (OIG), HUD, and prior independent public accountants, and made a commitment to improve its operations. Although the Authority reorganized its leased housing department and made other changes, it failed to correct systemic weaknesses and continued to mismanage its voucher program. Further, analysis of Authority data disclosed that it spent almost $20 million in questionable costs in 2006 and 2007. This amount included payments for clients that it did not report to HUD; payments for clients after they left its voucher program; duplicate payments to landlords; and payments for clients who, based on their reported Social Security numbers, were deceased. In addition, the Authority’s data showed that it backdated 22 and 45 percent of the examinations it reported to HUD in 2006 and 2007, respectively. What We Recommend We recommend that the Director of HUD’s Recovery and Prevention Corps1 (Director) require the Authority to establish and implement: policies and procedures to address its systemic weaknesses, an effective quality control process, and an effective accountability process. We also recommend that the Director require the Authority to support or repay nearly $20 million. In addition, we recommend that the Director incorporate the recommendations in this report with the management decision and corrective actions for recommendation 1D of OIG audit report 2008-FW-1006. For each recommendation without a management decision, please respond and provide status reports in accordance with HUD Handbook 2000.06, REV-3. Please furnish us copies of any correspondence or directives issued because of the audit. Auditee’s Response We provided a draft report to the Authority on June 24, 2008, with comments due by July 14, 2008. We held an exit conference on July 9, 2008. On July 10, 2008, the Authority requested an extension to respond by July 21, 2008. In its July 21, 2008 response, the Authority stated it believed its leased housing department had been in transition and was steadily making progress. It generally agreed with our recommendations. The Authority’s response along with our evaluation is included in appendix B of this report. 1 As of April 2008, HUD’s Recovery and Prevention Corps assumed responsibility for servicing the Authority. 2 TABLE OF CONTENTS Background and Objectives 4 Results of Audit Finding 1: The Authority Mismanaged Its Voucher Program 5 Finding 2: The Authority’s Mismanagement Caused Nearly $20 Million in 17 Questionable Costs Scope and Methodology 24 Internal Controls 26 Followup on Prior Audits 28 Appendixes A. Schedule of Questioned Costs and Funds to Be Put to Better Use 29 B. Auditee Comments and OIG’s Evaluation 30 3 BACKGROUND AND OBJECTIVES In 1938, the Dallas City Council established the Dallas Housing Authority (Authority) to provide housing to low-income persons. A five-member board of commissioners (board)2 governs the Authority. The board appoints a president and chief executive officer to administer the operations of the Authority. The Authority’s main office is located at 3939 North Hampton Road, Dallas, Texas 75212. The Authority administers the U.S. Department of Housing and Urban Development’s (HUD) Housing Choice Voucher program (voucher program). Under the voucher program, HUD pays rental subsidies so that eligible families can afford decent, safe, and sanitary housing. The Authority administers more than 17,000 vouchers annually pursuant to an annual contributions contract with HUD. In 2006 and 2007, HUD provided the Authority with almost $279 million in funds for housing assistance payments for its voucher program. HUD also provided the Authority nearly $19 million during the period to administer the voucher program. This is the third and final report on the Authority’s voucher program. The first audit focused on the portability features of the Authority’s voucher program.3 We reported that the Authority mismanaged its portable vouchers; collected $3.7 million from HUD based on inaccurate, unreliable, and altered records; and violated HUD requirements. The second audit concluded that Authority management failed to implement internal controls over the financial management of its voucher program4 and did not exercise sound management practices. As a result, the Authority had unreliable financial data and incorrect fund balances and could not assure HUD that it spent program funds in accordance with requirements. Also, the Authority certified to HUD that it expended about $32 million less in program funds than it received in 2005 and 2006. Our objective for this audit was to determine whether the Authority properly administered its overall voucher program. 2 The mayor of Dallas appoints board members. 3 Audit report number 2008-FW-1003, entitled “The Dallas Housing Authority, Dallas, Texas, Mismanaged Its Portable Vouchers,” issued December 5, 2007. 4 Audit report number 2008-FW-1006, entitled “The Dallas Housing Authority, Dallas, Texas, Management Failed to Implement Internal Controls over Its Housing Choice Voucher Program,” issued March 20, 2008. 4 RESULTS OF AUDIT Finding 1: The Authority Mismanaged Its Voucher Program The Authority established a cumbersome and ineffective process to administer and operate its voucher program. Although it reorganized its leased housing department and made other changes, it failed to correct systemic weaknesses. For example, the Authority continued to operate in a paper-intensive environment and used methods incompatible with HUD’s reporting requirements. Further, Authority management failed to instill accountability and provide employees with the tools necessary to operate an effective and efficient voucher program. The Authority’s lack of controls resulted in unacceptable error rates and did not ensure the detection and resolution of errors. Because the Authority established a system that was devoid of accountability and inadequately equipped its employees, its records were in disarray, and it effectively encouraged employees to circumvent requirements. The Authority must implement effective policies and procedures to ensure that it administers its voucher program efficiently and in compliance with requirements, thereby assuring that it will better administer its approximately $133 million annual voucher program. The Authority Established a Cumbersome and Ineffective Process Authority management established a cumbersome and paper-intensive process for obtaining and reporting family information to HUD. The process failed to provide efficient and effective administration of its voucher program. For example, rather than using the HUD-mandated format to collect information about families,5 the Authority used its own method, which differed materially from the required format. Further, Authority management revoked case managers’ data entry privileges, requiring them to use an unproductive and redundant manual process to calculate family income, rent, and assistance payment amounts.6 This process directly increased the likelihood and occurrence of errors. Authority management assigned its case managers an excessive caseload without regard to effective case management or providing the necessary tools to operate an adequate voucher program. Generally, case managers 5 Form HUD-50058, Family Report. 6 While the Authority’s case managers had insufficient access to its computer system for voucher processing, other employees had full and free access to make unsupervised changes to its financial systems. 5 serviced 750 to 800 clients each,7 versus a typical caseload of 300 to 350 at other housing authorities. Further, although the case managers should have been the Authority’s program experts and the best equipped to gather relevant family information to report to HUD, Authority management stripped them of accountability and responsibility by fragmenting the voucher administration process. Those employees who should have had program knowledge were not responsible for the actions that were ultimately reported to HUD. Rather, employees who had limited program knowledge performed these tasks and sometimes incorrectly revised the information they entered. Although the Authority used its fragmented approach to justify higher caseloads, it did not result in a functional operation. The Authority Operated in a Paper-Intensive Environment The Authority’s manual process required its case managers to print an internally created document, cross out prior information, hand-write changes to the family information, and manually record any new calculations, such as adjusted family income or a new housing assistance payment amount. The Authority did not have written guidance covering its process. However, based on interviews with Authority staff and reviews of client files, the process included Routing the internal document to information services data entry clerks for input into the Authority’s computer system. Routing the internal document to finance data entry clerks if the action required a change in payments to landlords or clients. Either routing the internal document to the file room, sometimes taking months to make it into client files, or back to case managers for review. Hand-writing or photocopying duplicate copies of the internal document, which were sent to various staff throughout the leased housing department. As a result, staff sometimes failed to process the documents or processed them multiple times (see finding 2). Circulating the document among staff for discussion and resolution if information on the internal document was unclear, causing unnecessary delays. As described above, the Authority’s manual process included multiple staff handling documents multiple times, which fragmented responsibility for the actions. Management was responsible for establishing an internal control environment that ensured effective accomplishment of goals and that held staff accountable. An effective system should minimize the handling of documents to lessen redundancy, improve efficiency, and 7 Although individual caseloads varied, this equates to about two and one-half to three hours per client annually. This amount of time appeared insufficient for case managers to complete required tasks. 6 assign accountability. Management failed to implement an effective system. The Authority Used an Incompatible Reporting Method The Authority used an internal document incompatible with HUD’s required reporting form that HUD uses to manage the voucher program. As shown in table 1, even the most commonly-used codes were incompatible. Codes on the Relationship and Authority’s action type codes Form HUD-50058 codes internal form H Head Granddaughter S Spouse Other relation K Co-head Niece F Foster child/foster adult Daughter Y Other youth under 18 Uncle E Full-time student 18+ Son L Live-in aide Brother A Other adult Head of household 0 None New application 1 New admission New move-in 2 Annual reexamination Interim exam 3 Interim reexamination Annual reexam 4 Portability move-in Special 5 Portability move-out Section 8 relocation 6 End participation Contract renegotiation Table 1: HUD required the Authority to categorize the relation or role of each household member using eight codes, while the Authority’s internal form allowed selection from 21 codes. HUD required reporting of the family’s type of action using 15 codes, while the Authority’s internal form only allowed selection from seven codes. To compensate for deviating from HUD requirements, the Authority programmed its computer system to convert its internal document into codes that HUD’s system would accept. These incompatible codes had been ingrained in the Authority’s processes for years. This practice could cause confusion and errors if the Authority implements a process that complies with HUD requirements. 7 The Authority Reorganized and Adopted an “Establishment of Standards Policy” In December 2007, the Authority reorganized its leased housing department by merging its fragmented functions from three departments into one.8 Although the Authority changed the physical location of employees and reassigned some managers, the changes did not address its systemic weaknesses. After its reorganization, the leased housing department continued to operate in an inefficient, ineffective, manually driven, and paper intensive environment. The Authority must begin to properly process its vouchers and implement an effective quality control process that includes a mechanism for evaluating the accuracy of the information it reports to HUD. In response to prior OIG audits, the Authority acknowledged that it did not properly administer its voucher program and reported that on September 20, 2007, its board adopted an “establishment of standards policy.” Since then, the Authority has approved or distributed various policies and procedures relevant to the voucher program as shown in table 2. Board Distribution Title approval 9 date date Administration Plan for the Section 8 Housing Choice Nov. 20, 2007 Feb. 11, 2008 Voucher Housing Choice Voucher Portability Procedure Mar. 4, 2008 Disaster Housing Assistance Payment (DHAP) Sept. 20, 2007 Oct. 2, 2007 Operating Plan Verification Procedures and Documents for all Assisted Oct. 2, 2007 Housing Programs Administered and/or Managed by the Housing Authority of the City of Dallas and the Dallas Housing Authority Table 2: Policies and procedures distributed after September 20, 2007. The Authority’s verification procedure was meant to provide guidance to Authority employees in collecting and verifying the necessary documentation during the application, eligibility, admission, interim, and annual reexamination processes. The guidance, which was comprised of more than 100 pages including 33 exhibits, was ineffective because it did not have a table of contents or index. The Authority should ensure that the 8 As reported in audit report number 2008-FW-1003, employees in the Authority’s leased housing, finance, and information services departments each had roles in processing vouchers, and they all reported to different managers. 9 The Authority only required policies to be presented to its board of commissioners for approval. Procedures could be approved by its compliance department. 8 guidance it provides to its employees is organized and clear and can easily be used as reference material. Statistical Sampling Confirmed Continuing Need to Address Systemic Weaknesses The Authority acknowledged its longstanding weaknesses and made a commitment to improve its operations. In late 2007, it adopted some new policies and procedures. This audit was designed to evaluate the effectiveness of the Authority’s new policies and procedures and assess its current conditions. We reviewed basic attributes for a statistical sample of 18 reexaminations effective between December 2007 and March 2008 (see Scope and Methodology). Table 3 summarizes the sample results. Projected Sample error rate error at 90% Attribute test Yes No rate confidence10 Identity/eligibility documented for each 13 5 28% 14% family member? Verification of income documented? 3 15 83% 67% Income correctly calculated? 4 14 78% 60% Disability and dependent allowances 14 4 10% appropriate? Voucher size appropriate based on 15 3 17% 6% family composition? Correct payment standard used? 15 3 17% 6% Correct utility allowance used? 12 6 33% 19% Rent to owner supported/reasonable? 12 6 33% 19% Application of lower of payment 18 0 0% 0% standard or gross rent correct? All income and rent calculations 3 15 83% 67% accurate? Housing assistance payment to owner 12 6 33% 19% and/or utility reimbursement match sample amount? Table 3: The audit universe failed the attribute test if an error was found. However, by continuing to test the sample until all selected items were reviewed, auditors could estimate the actual error rate and its precision. 10 While the actual sample error rate is the best estimate, the projected error rate is the minimum error rate one can expect to find in the population. 9 Systemic Weaknesses Still Existed The Authority showed some improvement over previous reports by HUD and its compliance department in the following areas: Assigning the appropriate bedroom size, Using the correct payment standard, and Correctly applying the lower of the payment standard or the gross rent. Although it showed modest improvement, the sample evaluation demonstrated the Authority had not made sufficient progress toward improving its operations and its new procedures were largely unimplemented. The Authority failed 10 of the 11 attributes tested with 17 of the 18 sampled files failing at least two attributes. The sample results demonstrated continued poor compliance with basic program regulations. Some of the errors identified in the sample included The Authority’s case managers did not consistently follow HUD’s verification of income hierarchy recommendations and did not document the files showing how they resolved income discrepancies. The following are examples of the lack of action: o In 2 of 18 statistically selected files, clients reported that they had no income, apparently to avoid paying their portion of rent. While the case managers included income identified in HUD’s income verification system in the rent calculation, they did not properly document the client files or properly address the client’s failure to report income. The Authority required case managers to take steps to terminate the client from the program for failing to report income. For these two clients (11 percent of the sample), the case managers disregarded the Authority’s policy. o On May 14, 2007, a client reported that she was laid off from work on May 4, 2007; thus, the Authority recalculated and reduced her rent. However, the client started another job at a higher rate of pay on June l, 2007. The client did not report the new employment or income to the Authority within 30 days as required. At her annual reexamination appointment on November 13, 2007, the client reported the information, but the case manager did not address the client’s failure to do so in a timely manner or require the client to repay the Authority. As a result, the Authority allowed the client to pay less rent than required while it paid the difference. 10 o A client did not report income on the Authority-required personal declarations questionnaire, dated November 20, 2007. However, HUD’s income verification system showed that the client had income from three employers. A third-party verification provided by one employer showed that the client’s employment ended on August 1, 2007. However, HUD’s income verification system showed that the client started other jobs on May 11 and June 11, 2007. While the Authority completed a request for information related to unemployment benefits, there was no evidence that the Authority sent the request or that it received a response. There was no evidence that the case manager performed further inquiries regarding the two other jobs. In three instances, case managers did not consider the clients’ medical expenses or disabilities although the client files contained information indicating that they should. In these cases, the case managers did not grant the allowance or deduction; thus, the clients paid excess monthly rent amounts. In its response to HUD’s September 2006 Rental Integrity Monitoring review, the Authority pledged to follow a specific procedure it designed to ensure that it properly calculated utility allowances. Although the Authority correctly calculated 12 of 18 utility allowances, it did not follow its established, yet unnecessary, procedure for any of them. Case managers acknowledged that essential tasks went undone and generally claimed they did not have time to complete the work because of their large caseloads. They reported client file maintenance, interim reexaminations, and client followup as the tasks most overlooked. For example, on May 13, 2008, a client visited the Authority to complain about his situation and a lack of customer service. The client also met with auditors and explained that the Authority had stopped paying housing assistance for his family11 and provided no explanation to him or his landlord. The client was disabled and faced eviction without understanding why the Authority had made such a decision. The client reported several unreturned phone calls to his case manager. In addition, the Authority failed to respond to several attempts from auditors to get an update on the client’s situation. 11 The Authority issued the six-member family a two-bedroom voucher instead of the required three- bedroom voucher and paid the reduced housing assistance. It also appeared that the Authority disregarded essential family information in making its decision to reduce the family’s housing assistance. 11 The Authority claimed that its manual process was a form of effective quality control. However, separating data entry from decision making and requiring re-performance of the related calculations on the Authority’s internal form did not disclose case manager errors identified in the sample. An effective quality control program should include review of the actual work in comparison to established standards. The Authority Did Not Properly Maintain Client Files The Authority’s system for maintaining client files was disorganized and unmanaged. Client files should aide case managers in processing vouchers, serving as a repository for essential family information. The Authority’s client files hindered the process because they were in disarray. Staff routinely put loose documents haphazardly into the client files without regard to date, duplication of documents, assurance that related documents stayed together, and whether the files had become unmanageable due to the volume of documents. Further, staff did not always file documents in a timely manner. As a result, the Authority experienced difficulty in locating documents that should have been in client files. The review of the 18 client files found that More than 14 months after the Authority reported implementing a standardized filing system, none of the 18 statistically selected files met the Authority’s standard. Further, there was no evidence that the Authority had reviewed the process to ensure that the standards had been implemented as intended. Rather than ensuring that they filed current documents together, staff routinely separated and haphazardly put them among documents that were several years old. Case managers often kept documents at their desks rather than having them placed in the client files. One case manager reported keeping documents for up to three months. Another case manager reported seeing 20 to 25 clients per day. Collecting documents from that many clients without an effective process for ensuring that it properly filed them contributed to the Authority’s inefficient and ineffective operations. Some client files did not contain essential documents while some files contained duplicate documents. For example, one file had only a driver’s license as identification, while other files had multiple copies of the same birth certificates and Social Security cards. In one instance, the file contained birth certificates for a different family. 12 In many instances, client files contained original landlord vendor-number requests that should have been in landlord files, not client files. The Authority also acknowledged that it had duplicate landlord files. These errors may have contributed to the Internal Revenue Service (IRS) rejecting 784 Forms 1099-Miscellaneous12 due to mismatched Social Security/taxpayer identification numbers. As in many previous instances, Authority managers from different departments claimed that other departments were responsible for correcting the errors. However, the Authority was unable to provide documentation showing that it corrected the errors or that it correctly reported the information to the IRS. All of the case managers interviewed acknowledged that the sampled client files fairly represented the conditions and agreed they were poorly maintained. However, they blamed their workload and other staff for the poor condition of the files. Authority management must accept responsibility for its client files and develop a filing system to ensure that client information is well maintained and organized in a logical manner that permits ready access to information and lends itself to review and verification. The Authority Did Not Provide Case Managers with Necessary Tools The Authority expected case mangers to manage extremely high caseloads in a manual environment, with poorly maintained client files, while they did not have responsibility for the information ultimately reported to HUD. Further, until recently, Authority management discouraged case managers from having client files with them when they interviewed clients, which led to duplication, errors, omissions, and a lack of accountability. If the case managers had the client files at the time of the interviews, they could file documents as they collected them. Further, if the client files were conducive to routine review, case managers would know in advance what decision-impacting information was already available and could avoid having clients continually provide documents already in the files. All of the case managers interviewed complained of being overwhelmed by their workload. They also expressed concern over the amount of staff turnover, which often led to increases in their already unmanageable caseloads. Instead of providing manageable caseloads that included complete responsibility for voucher processing and insisting on 12 The Authority had to report annual landlord payments of $600 or more to the IRS using Form 1099- Miscellaneous. 13 accountability, the Authority burdened its case managers with unnecessary and unproductive manual processes. The Authority’s ineffective systems, which contained duplicative efforts, led to sloppiness, lacked accountability, exacerbated the conditions, and reduced effectiveness and efficiency. The Authority Recently Hired New Program Managers The Authority recently hired new managers for its leased housing department, which operates the voucher program. Rather than engaging in the previous Authority practices of attributing deficiencies to computer systems or claiming that the deficiencies resulted because of the Authority’s large voucher program, these managers seemed to recognize the importance of operating of an effective and efficient program. The Authority Should Use Its Resources The Authority should use its compliance department to augment its quality control process. In 2006, the Authority’s compliance department conducted three separate reviews of the Authority’s client files and reported systemic weaknesses similar to those reported by others. Those reviews disclosed error rates as high as 91 percent.13 In addition, the Authority had access to HUD systems that contained useful information. Had the Authority taken actions based on the findings and recommendations made by its compliance department and the information available in HUD systems, conditions identified by OIG and other audits and reviews might have been mitigated. In February 2008, the Authority’s compliance department concluded a review of the Authority’s Section Eight Management Assessment Program (SEMAP) submission. This effort appeared to be an earnest attempt at self-assessment. It also demonstrated the potential benefit of having an organizationally independent body within the Authority. However, to better ensure independence from Authority management, the Authority’s board should consider requiring the compliance officer to dual report directly to the board and the president/chief executive officer.14 13 HUD’s September 2006 rental integrity monitoring review reported an 89 percent error rate. 14 The vice president of compliance reported to the Authority’s general counsel. 14 The Authority’s Lack of Controls Has Been an Ongoing Weakness The Authority was repeatedly informed of its internal control weaknesses. Its independent audit reports15 for fiscal years 2004 and 2005 contained the following finding: “The system of internal control as designed and maintained by DHA [the Authority] appears to be inadequate and not operating effectively to reasonably ensure DHA’s compliance with Federal laws, regulations, and program compliance requirements.” In addition, the Authority received a number of reviews and correspondence from HUD16 and its compliance department citing systemic weaknesses related to its voucher program. Despite the Authority having been informed of its lack of internal controls and systemic weaknesses, management has not corrected them. This was further reported in our previous reports issued in December 2007 and March 2008. The Authority must address its systemic weaknesses including designing and implementing an internal control process that provides assurance that it processes vouchers effectively and efficiently and accurately reports performance to managers and HUD. The process should be written and include quality control procedures to measure performance. Conclusion The Authority failed to adequately manage its voucher program and its leased housing department. As reported many times over the last six years, the Authority's lack of internal controls negatively affected its voucher program. Although it made modest changes, the Authority continued to be unable to assure HUD that it administered its program in compliance with requirements. It must address its systemic weaknesses identified in this report, our two previous reports, and other reports and make substantive changes to its processes to ensure that it adequately administers its voucher program in accordance with HUD requirements. As part of this effort, the Authority should use its available resources, such as its compliance department and information available from HUD systems, to improve its operations. By insisting on accountability from its staff and making necessary improvements to its fragmented, inefficient, 15 Office of Management and Budget Circular A-133 audits performed by KPMG International. 16 Rental Integrity Monitoring reviews dated November 2002, November 2003, and September 2006. 15 and ineffective voucher program, the Authority would better administer the $133 million that HUD provides it annually. Recommendations We recommend that the Director of HUD’s Recovery and Prevention Corps 1A. In addition to its actions to address recommendations in previous OIG reports, require the Authority to establish and implement policies and procedures to address its systemic weaknesses and eliminate redundant, ineffective, and unnecessary procedures to include implementing an effective quality control process. 1B. Require the Authority to implement an effective accountability process that includes acting on reported instances of programmatic noncompliance and ensuring dual reporting by the compliance officer to the board and the president/chief executive officer. 1C. Incorporate the recommendations in this report in your proposed management decision and corrective actions for Recommendation 1D17 of OIG audit report 2008-FW-1006. 17 See page 28. 16 Finding 2: The Authority’s Mismanagement Caused Nearly $20 Million in Questionable Costs Analysis of the Authority’s historical data showed that in 2006 and 2007 it spent more than $11 million on behalf of 872 clients that it did not report to HUD, nearly $8 million for 2,305 clients after it terminated them from its voucher program, more than $250,000 for 376 duplicate landlord payments, and more than $167,000 on behalf of 45 clients whose Social Security numbers identified them as deceased. In addition, the Authority’s data showed that it backdated 22 percent and 45 percent of the examinations it reported to HUD in 2006 and 2007, respectively. This condition occurred because the Authority mismanaged its voucher program by not implementing an adequate internal control system. As a result, the Authority did not use the funds in accordance with requirements. The Authority could better manage its voucher program if it made use of its available resources. The Authority Paid $11.18 Million for Clients It Did Not Report to HUD Analysis of the Authority’s data showed that it paid $11.18 million in housing assistance on behalf of 872 clients (about 5 percent of the clients assisted) in 2006 and 2007 without reporting the family information to HUD.18 HUD relies on public housing agencies to submit accurate, complete, and timely data to administer, monitor, and report on the management of the national voucher program. HUD also uses this information to justify its budget requests to Congress. Underreporting family information to HUD impacts HUD’s ability to administer the voucher program at national and local levels. HUD required the Authority to submit 100 percent of its family records and would subject the Authority to sanctions if it failed to maintain at least a 95 percent reporting rate.19 In addition, HUD used the family information the Authority submitted to verify the Authority’s self-certification under SEMAP. If the Authority’s data become insufficient to verify its certification, HUD will assign a zero rating for five key areas. The Authority’s December 31, 2007 SEMAP rating was 69 percent (100 out of 145 possible points) and included 30 18 Based on the Authority’s submissions to HUD with effective dates between January 1, 2004, and May 3, 2008. 19 Notices PIH (Public and Indian Housing) 2005-17, 2006-24, and 2007-29. 17 points for the five areas verified using data it submitted. The Authority must ensure that it submits complete and accurate family information to HUD as required or risk losing 30 points, which would result in HUD’s designating it as a troubled housing agency.20 As reported in previous OIG audits, the Authority’s internal controls were inadequate to ensure it only made authorized payments. Therefore, the Authority could not assure HUD these unreported payments were appropriate and eligible. The Authority should review its records for the 872 unreported clients to determine whether its $11.18 million in unsupported payments went to qualified landlords on behalf of eligible families. It should repay HUD for any ineligible amounts and submit the necessary family information to HUD. The Authority Paid Almost $8 Million in Assistance for Clients It Terminated from Its Voucher Program The Authority’s data showed that it paid almost $8 million in assistance for 2,305 clients after the termination effective dates it reported to HUD. It made 17,075 ineligible or unsupported payments for terminated clients in 2006 and 2007. In some cases, the Authority’s fragmented procedures for processing client information may have resulted in termination information not being processed; thus, payments to landlords and families continued. In other cases, the Authority may have legitimately backdated the effective dates of the terminations based on relevant information it discovered later, such as client fraud or death. In these cases, the Authority should have recovered any overpayments from the landlord and the family. However, it did not consistently recover overpayments. In cases in which the landlord had multiple clients in the voucher program, the Authority sometimes deducted overpayments from subsequent landlord checks. If the landlord did not have other clients in the voucher program, the Authority did not attempt to collect the overpaid amounts. Because the data the Authority provided for audit did not contain reliable deduction information, the data could not be tested to determine whether the Authority recovered ineligible overpayments from landlords when it backdated terminations. The Authority should review its records to determine whether it recovered the overpayments it made on behalf of clients it terminated from its 20 If the Authority lost 30 points in its SEMAP score because it did not report all records to Multifamily Tenants Characteristics System (MTCS) and Public and Indian Housing Information Center (PIC), its score would decrease from 69 percent (standard rating) to 48 percent (troubled). (24 CFR (Code of Federal Regulations) 985.103). 18 voucher program. If not, it should recover and repay to HUD the ineligible portion of almost $8 million in unsupported payments. Further, it should establish and implement procedures to ensure that it processes termination information for all clients who leave its voucher program and to ensure that it recovers all overpayments from landlords and families. The Authority Paid More Than $250,000 in Duplicate Assistance The Authority’s data showed that it paid duplicate assistance totaling $253,833 for 59 clients. The analysis identified 376 duplicate payments, some of which were ongoing, by looking for payments for more than one contract number21 for the same client on the same date. Based on descriptions by staff, the Authority’s internal process when a client changed units involved preparing and processing two separate manual forms: one to terminate the contract for the existing unit and one to begin the contract for the new unit. However, as in other operational areas, the Authority did not provide staff written guidance as to how to process the changes. The duplicate payments likely resulted from the inadequacy and fragmentation in this process that allowed the Authority to process new contracts without terminating expired contracts. The Authority should determine which of the duplicate payments were ineligible, recover the funds from the landlords, and repay HUD accordingly. On several occasions during the audit, we requested that the Authority provide its most recent payment information for analysis to help it identify and stop ongoing duplicate payments. The Authority was unresponsive, electing not to take advantage of this opportunity to cease its ineligible payments for expired contracts. To better manage its voucher program, the Authority should use available information to avoid unnecessary duplication of assistance. The Authority Paid Assistance for Deceased Clients The Authority paid assistance for 235 clients between January 2006 and December 2007 who used Social Security numbers that indicated they were deceased as of April 2007.22 Because the resource used to identify 21 The Authority identified the program and unit under contract through a housing assistance payment (HAP) number that was supposed to be unique to the client and unit. 22 As discussed in the Scope and Methodology section, because its data had limited reliability related to client identity, the Authority will need to verify whether the clients were deceased at the time of payment. 19 the possibly deceased clients did not contain the date of death, it was not feasible to determine whether the Authority made payments for the entire period. However, it made 322 payments totaling $140,975 on behalf of 44 of the 235 possibly deceased clients after April 2007. In a separate example, a family’s head of household died in March 2003. The family did not inform the Authority of the death and repeatedly ignored the Authority’s attempts to schedule eligibility reexaminations until the Authority notified the family in May 2006 that it was terminating assistance. During this time, a case manager processed an interim reexamination, effective October 2003, and changed the family’s income and rent amounts. There was no documentation in the file supporting that the family supplied the information, suggesting that the case manager processed the action without conducting the required reexamination. In addition, in January 2006, the Authority fired another case manager assigned to the family for falsely certifying client data.23 The client file contained an income verification report printed in November 2005 indicating that the head of household was deceased, but the case manager failed to act on the information. The Authority submitted records to HUD showing that it had terminated assistance for the family effective October 1, 2003. It did not explain why the effective date was six months after the head of household’s death. In July and August 2006, the Authority recovered a total of $33,579 in overpayments to the landlord; however, the Authority’s calculation of the recovered funds was incorrect. Further, on August 3, 2006, the Authority paid the landlord $24,177 on behalf of the family and on September 1, 2006, made additional payments to the landlord and the family that it could not support. The Authority’s records indicated that it made net payments of $24,462 to the landlord and $1,950 to the family on behalf of the deceased head of household after her death. The Authority should recover and repay the $26,41224 paid on behalf of the family after the head of household’s death. The Authority had access to a HUD system that contained reports that identified deceased clients. In addition, HUD provided multiple training sessions on effective use of its system. The Authority should use this management tool to monitor its program and assist it in identifying deceased clients to avoid making ineligible payments. 23 The Authority found that although the case manager indicated that she had interviewed and conducted annual reexaminations for 40 clients, she had not. 24 $24,462 + $1,950 = $26,412. 20 The Authority Backdated Reexaminations File reviews showed backdated reexaminations The review of 18 statistically selected client files showed that the Authority backdated the effective dates of its annual reexaminations as evidenced by the supporting documentation and processing stamps on the Authority’s internal form being dated after the effective date. For example, in one case, the Authority did not process the paperwork for a change with an effective date of December 1, 2007, until February 19, 2008. In such situations, the case managers wrote notes to the data entry clerks asking that the change in subsidy amount take effect at a future date and that the client not be penalized. When asked about such notes, employees explained that the Authority is rated monthly on late reexaminations so they had to be dated on the due date, regardless of when they were processed. The Authority’s practice was to provide a 30-day notice to clients if the client’s portion of the rent increased. Neither HUD nor the Authority25 had a formal policy requiring a 30-day notice to the client if the client’s portion of the rent increased. In cases in which the subsidy increased and the client portion decreased, the Authority processed manual checks to pay the back rent and reimburse the client for the period between the effective date and the date the Authority completed the reexamination and processed the changes. Analysis revealed extent of backdated reexaminations in 2006 and 2007 Analysis of the Authority’s data showed that it submitted records to HUD for 11,922 and 12,780 annual reexaminations with effective dates in 2006 and 2007, respectively. In 2006, the Authority did not pay the housing assistance amount submitted to HUD on the effective date of the reexamination for 22 percent of reexaminations. In 2007, the number more than doubled to 45 percent of the reexaminations. The delay in changing the payment amount ranged from one to 17 months. This delay is a strong indication that the Authority routinely backdated its annual reexaminations to conceal their lateness. Backdating reexaminations was futile Late reexaminations could negatively affect the Authority’s performance rating. Its SEMAP score would be reduced if more than 5 percent of its annual reexaminations were more than two months overdue. Since HUD based timeliness on overdue reexaminations the Authority had not submitted rather than the timeliness of submitted reexaminations, this 25 The Authority’s administrative plan dated November 20, 2007, added this provision. 21 practice would not help the Authority’s score in this area. Further, by engaging in these activities, the Authority reduced the amount of time available to conduct its annual reexaminations in a timely manner during the next annual reexamination cycle. For example, if the Authority backdated a reexamination by three months, it would only have nine months to perform the subsequent annual reexamination. This intensified the Authority’s already heavy caseload. The Authority Experienced Difficulty in Providing Basic Programmatic Information In all three OIG audits, the Authority experienced great difficulty in providing basic information. In addition to the weaknesses already reported, the Authority could not provide sufficient evidence that it properly admitted clients from its waiting list. The Authority’s administrative plan required selection from the waiting list in chronological order by date and time of application. In the reports the Authority provided, the date and time applicants entered the waiting list were inconsistent from one report to the next. The Authority should ensure that it accurately reflects and tracks entrance and selection from its waiting list. If the Authority could not provide simple and reliable programmatic documentation for audit purposes, it was doubtful that it could retrieve and analyze it for purposes of efficiently operating its program(s). The Authority must use its available tools to operate its program efficiently, effectively, and in accordance with HUD requirements. Conclusion Analysis of the Authority’s data disclosed almost $20 million in ineligible and unsupported payments during 2006 and 2007. Although many have reported on the Authority’s systemic weaknesses, its own data revealed these inappropriate payments that would not be identified through traditional means. The Authority must use the many tools it has available to identify and address its systemic weaknesses, including analysis of its own data and existing HUD reports. Unless it does so, it will continue to mismanage its voucher program. 22 Recommendations We recommend that the Director of HUD’s Recovery and Prevention Corps 2A. Require the Authority to establish and implement policies and procedures to ensure that it submits complete, accurate, and timely information to HUD as required. The Authority should also use the data and resources it has available and stop payments when warranted. 2B. Require the Authority to support or repay HUD $11,181,396 in unsupported payments for clients it did not report to HUD. 2C. Require the Authority to support or repay HUD $7,981,640 in unsupported payments for clients it terminated from its voucher program. 2D. Require the Authority to support or repay HUD $253,833 in ineligible duplicate payments to landlords. 2E. Require the Authority to support or repay HUD $167,387 in unsupported payments for possibly deceased clients. 23 SCOPE AND METHODOLOGY Our objective was to determine whether the Authority properly administered its overall voucher program. To accomplish our objective, we • Reviewed relevant criteria, • Interviewed HUD and Authority management and staff regarding the Authority’s operations, • Reviewed relevant Authority personnel files, • Reviewed the Authority’s compliance department audit reports, • Reviewed a statistical sample of client files, and • Performed analytical procedures on the Authority’s data related to voucher payments and client eligibility reexaminations. We conducted the audit in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. Our audit generally covered the period January 2006 through March 2008. We expanded the review period as necessary to accomplish our objective. We performed audit fieldwork at the Authority’s administrative offices in Dallas, Texas, from November 2007 through May 2008. Revised Audit Approach Our initial audit approach involved reviewing a large statistical sample of client files supporting housing assistance payments made during calendar years 2006 and 2007. However, HUD and other external auditors previously performed work covering that period and identified areas of concern, which the Authority pledged to correct. Rather than duplicate the work of others, we agreed to assess the Authority’s most recent operations with respect to new policies and procedures for the voucher program as well as its reorganization of staff performing data entry for the program. We revised the audit approach to select a statistical sample of client files for review from a list of client eligibility reexaminations with effective dates between December 1, 2007, and March 31, 2008. This approach resulted in a smaller sample of client files for review and allowed us to assess whether the Authority’s management changes were effective. In addition, we obtained and analyzed the Authority’s data containing voucher payment records between January 1, 2006, and December 31, 2007, to identify additional areas of concern that could not be identified by reviewing client files. Statistical Sampling We used discovery attribute sampling to identify the rate of error in the Authority’s reexaminations with effective dates between December 1, 2007, and March 31, 2008. Discovery sampling tests for compliance with policies, procedures, and practices of a function to determine the adequacy of internal controls or operational efficiency. It is not 24 designed to estimate questioned costs. Instead, poor compliance identified by a discovery sample will prompt recommendations to address weaknesses in the design or implementation of internal controls. We used EZ-Quant to calculate a sample size of 18 from the Authority’s list of 5,913 reexaminations based on a critical error rate of 10 percent and a maximum government risk of 15 percent. We randomly selected the 18 reexaminations and reviewed them for compliance with specific attributes as discussed in finding 1. An audit universe will fail the test for an attribute if an error is found. However, by continuing to test the sample until all selected items have been reviewed, auditors can estimate the actual error rate and its precision. We used EZ-Quant to project the lower limit of the pass/fail rates of the attributes tested in the sample to the overall population of reexaminations the Authority performed during the period with a confidence level of 90 percent. While the actual sample error rate is the best estimate, the projected error rate is the minimum error rate one can expect to find in the population. Data Reliability Assessments The Authority provided two electronic files containing reproductions of its voucher payment records for calendar years 2006 and 2007. We performed data reliability assessments of the data files by analyzing the information and comparing the records to the Authority’s printed check registers, its electronic bank statements, and the data it submitted to HUD. Although we identified some missing records from the data, we determined that the data were sufficiently reliable for the purposes of analyzing payments made to and on behalf of clients in its voucher program. Errors related to the missing information would result in an understatement of a problem and would not be material or misleading. The data files contained an unknown number of errors related to client identity. Comparison with the Authority’s live system data showed that when the Authority reissued a client number or contract number after a client left the program, the data files identified the previous client rather than the client assisted at the time of the payment. For this reason, conclusions based on analysis of client identity will require verification by the Authority. Lastly, the data were not sufficiently reliable with respect to deduction information because the data files did not attribute deductions from landlord payments to the appropriate client. This condition prevented using the data to determine whether the Authority recovered overpayments to landlords for specific clients. We also obtained data the Authority submitted to HUD’s PIC with effective dates between January 1, 2004, and May 8, 2008. We did not perform detailed testing on the reliability of the PIC data. Previous OIG assessments of PIC did not identify deficiencies that would impact the audit objectives. In addition, the Authority was directly responsible for the accuracy of the information it submitted to PIC. Therefore, analysis of the Authority’s payment data in comparison with what it submitted to PIC was appropriate and would not result in misleading conclusions. 25 INTERNAL CONTROLS Internal control is an integral component of an organization’s management that provides reasonable assurance that the following objectives are being achieved: • Effectiveness and efficiency of operations, • Reliability of financial reporting, • Compliance with applicable laws and regulations, and • Safeguarding resources. Internal controls relate to management’s plans, methods, and procedures used to meet its mission, goals, and objectives. Internal controls include the processes and procedures for planning, organizing, directing, and controlling program operations. They include the systems for measuring, reporting, and monitoring program performance. Relevant Internal Controls We determined the following internal controls were relevant to our audit objectives: • Program operations – Policies and procedures that management implemented to reasonably ensure that its program met its objectives. • Validity and reliability of data – Policies and procedures that management implemented to reasonably ensure that valid and reliable data were obtained, maintained, and fairly disclosed in reports. • Compliance with laws and regulations – Policies and procedures that management implemented to reasonably ensure that its resource use was consistent with laws and regulations. • Safeguarding resources – Policies and procedures that management implemented to reasonably ensure that its resources were safeguarded against waste, loss, and misuse. We assessed the relevant controls identified above. A significant weakness exists if management controls do not provide reasonable assurance that the process for planning, organizing, directing, and controlling program operations will meet the organization’s objectives. 26 Significant Weaknesses Based on our review, we believe the following items are significant weaknesses: • The Authority established cumbersome, unproductive internal processes and stripped its employees of responsibility. • The Authority did not use available tools to operate its voucher program efficiently or effectively. • The Authority failed to safeguard its resources against waste, loss, and misuse. Without these basic controls, it was unable to ensure compliance with HUD requirements (see findings 1 and 2). 27 FOLLOWUP ON PRIOR AUDITS Dallas Housing Authority, Dallas, Texas, Mismanaged Its Portable Vouchers 2008-FW-1003 OIG audit report 2008-FW-1003, issued December 5, 2007, reported that the Authority mismanaged its portable vouchers. The report contained four recommendations, all of which impacted our audit objectives and were considered in planning this audit. We concurred with HUD’s management decisions on all four of the recommendations on March 25, 2008. They remain open pending completion of corrective actions. Dallas Housing Authority, Dallas, Texas, Management Failed to Implement Internal Controls over Its Housing Choice Voucher Program 2008-FW-1006 OIG audit report 2008-FW-1006, issued March 20, 2008, reported that Authority management failed to implement internal controls over its voucher program. The report contained four recommendations, two of which impacted our audit objectives and were considered in planning this audit: 1C. Require the Authority to implement adequate internal controls over its financial management of its voucher program. At a minimum, the internal controls should address the weaknesses cited in this report. 1D. Take appropriate administrative sanctions, up to and including issuing a notice of default in accordance with section 15 of the annual contributions contract for the Rental Certificate and Rental Voucher programs. We concurred with HUD’s management decisions on recommendations 1B and 1C on June 11, 2008; recommendation 1A on June 20, 2008; and, recommendation 1D on July 8, 2008. They remain open pending completion of corrective actions. 28 APPENDIXES Appendix A SCHEDULE OF QUESTIONED COSTS AND FUNDS TO BE PUT TO BETTER USE Recommendation number Ineligible 1/ Unsupported 2/ 2B $11,181,396 2C $7,981,640 2D $253,833 2E $167,387 TOTALS $253,833 $19,330,423 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 those costs charged to a HUD-financed or HUD-insured program or activity when we cannot determine eligibility at the time of audit. Unsupported costs require a 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. 29 Appendix B AUDITEE COMMENTS AND OIG’S EVALUATION Comment 1 30 31 32 Comment 2 33 Comment 3 Comment 4 34 Comment 5 Comment 6 35 Comment 3 Comment 6 36 Comment 6 Comment 7 37 Comment 8 38 39 Comment 9 Comment 6 Comment 10 Comment 11 40 Comment 11 Comment 12 Comment 15 41 Comment 13 Comment 10 42 Comment 6 Comment 13 Comment 8 43 44 OIG Evaluation of Auditee Comments Comment 1 We acknowledge that the Authority’s voucher program has been in transition. However, it has been in transition throughout our three audits. The Authority has not provided a timeline or milestones for completing the transition. In part, because of the assertions by the Authority of significant improvement, we modified our audit approach to focus on more current activities. For reasons detailed in the findings, we disagree with the Authority’s conclusion that it operated its voucher program in an effective manner. Both HUD’s MTCS and SEMAP rely upon the Authority providing accurate data and self-certifications. As detailed in this audit, two previous OIG audits, and HUD reviews, the Authority could not support the information it supplied. For instance, the MTCS reporting rate relies upon information submitted by the Authority to the Voucher Management System. As reported previously, the Authority submitted inaccurate information to the Voucher Management System; therefore, the accuracy of the reporting rate by MTCS was questionable. The Authority has made some strides in improving its operations. However, the Authority still needs to make significant improvement and continue to work with HUD to have an effective and efficient voucher program. Comment 2 We modified the language in our conclusion. We did not audit the Authority’s inspection or maintenance of units nor its compliance with housing quality standards. The other accomplishments cited by the Authority in its response were based upon information that it provided to HUD. As stated in our reports and HUD reviews, the Authority did not have the controls and management in place to provide consistent, accurate, and reliable information. Comment 3 The Authority stated it reduced the caseload for the case managers from approximately 750-800 cases to 500. Decreasing the work load of the case workers by a third might improve the effectiveness of the Authority’s operations. However, it did not address its underlying problems. The Authority did not explain how reducing the caseload to 500 will ensure that it will meet HUD requirements and effectively serve its clients in the future. 45 Comment 4 The Authority acknowledged that it used an incompatible reporting method. We maintain our position that the Authority’s process was well outside industry standards and contributed to an ineffective and inefficient administration of its voucher program. Comment 5 Contrary to the Authority’s response, it assured us that it had started this reorganization in August 2007 with substantial completion by January 2008. At the Authority’s request, we modified our audit approach to focus on its activities from December 2007 through March 2008. The Authority agreed at the time of the audit that this would be a fair approach rather than us reviewing case files prior to August 2007 as originally planned. During our audit, Authority staff reported that many policies and procedures were in draft. We included in our report the policies and procedures relevant to the audit objectives and approved as of April 8, 2008. The Authority should continue to improve its operations by developing and implementing policies and procedures, streamlining operations, and addressing previous audits and reviews. Comment 6 We encourage the Authority to implement, monitor, and evaluate policies and procedures to improve the effectiveness and efficiency of its operations. The Authority should ensure its systems accurately measure and report its performance. Comment 7 We acknowledge steps the Authority has taken to improve its voucher program. However, as the evidence reviewed and analyzed during our audit showed, the Authority’s changes had not impacted its systemic weaknesses. Our audit work and that of others continued to identify weaknesses already reported by its compliance department with little or no evidence of correction or follow-up by management. We appreciate the Authority implementing the recommendation that the compliance officer report directly to the board. Comment 8 The Authority’s response cites its MTCS and self-certified SEMAP ratings as support that it operated an effective voucher program. Our audit tested the underlying information and effectiveness of its operations. The conclusions reached were based upon the evidence reviewed and analyzed during the audit. While we understand the Authority’s need to promote a positive image, the Authority must not avoid or deny problems to protect its reputation. The Authority’s response cites significant changes to its voucher program over the last year and promises further significant changes. 46 These changes confirm rather than contradict the conclusions in this report that the Authority mismanaged its voucher program and needs to take significant actions to correct its long-standing systemic weaknesses. Comment 9 As stated in recommendation 2B, the Authority must support the eligibility of its payments or repay HUD. The Authority must report accurate and reliable information to HUD. The Authority was not held to a higher standard. HUD required the Authority to submit 100 percent of its family records and may sanction the Authority if it submits fewer than 95 percent. The 872 clients that the Authority failed to report to HUD represented about 5 percent of the clients served by the Authority. We did not recommend sanctions; however, if HUD determines that the Authority can not correct its weaknesses and appropriately administer its voucher program, it may determine that sanctions are warranted. Comment 10 The Authority acknowledged that it made ineligible payments. The Authority’s procedures, recommendation 2A, should include routine analysis to identify potentially ineligible payments, and pursuit of remedies against anyone that it determines fraudulently received assistance. Comment 11 While the Authority believes that some of the amounts were immaterial, HUD required the Authority to expend funds on only eligible clients. The Authority did not have written policies during the audit period nor did it have a system to identify and correct these ineligible payments. Comment 12 We maintain our position and added an example in the report for clarification. The Authority’s incompatible software may have contributed to its problems. However, the client files reviewed contained documentary evidence of backdated reexaminations, none of which were attributable to software incompatibility. We applied analytical procedures to the Authority’s data to determine the extent of the problem. Comment 13 We maintain our position that the information requested was essential for the Authority to adequately manage its voucher program. The Authority attempted to minimize the condition by implying that it had the reports and information available, but it was not in the correct format. The facts did not support this assertion. For example, wait list reports printed on different days showed conflicting dates and times that applicants entered the wait list. Because of the inconsistency of this information, the Authority compromised the 47 integrity of its wait list. As another example, significant delays in providing basic information resulted because the Authority failed to archive its monthly check register reports. Therefore, the Authority had to recreate the check registers with considerable effort and decreased reliability as discuss in the scope and methodology section. The Authority never responded to other requests for information. Comment 14 Contrary to the Authority’s response, the loss of 30 points was not speculation. We included the possible reduction in the SEMAP rating to demonstrate the potential effect of the Authority providing insufficient information to HUD. 48
The Dallas Housing Authority, Dallas, Texas, Mismanaged Its Housing Choice Voucher Program
Published by the Department of Housing and Urban Development, Office of Inspector General on 2008-07-31.
Below is a raw (and likely hideous) rendition of the original report. (PDF)