Department of Human Resources Family Investment Administration

Transcription

1 Audit Report Department of Human Resources Family Investment Administration June 2001

2 This report and any related follow-up correspondence are available to the public and may be obtained by contacting the Office of Legislative Audits or the Department of Legislative Services Office of the Executive Director, 90 State Circle, Annapolis, Maryland. Alternate formats may be requested by contacting the Office of Legislative Audits at , (Voice), or (Toll-Free Voice), or (Fax), and (Maryland Relay Service). Audit reports can also be viewed or downloaded from the Internet. Our web site address is

3 June 1, 2001 Delegate Samuel I. Rosenberg, Co-Chair, Joint Audit Committee Senator Nathaniel J. McFadden, Co-Chair, Joint Audit Committee Members of Joint Audit Committee Annapolis, Maryland Ladies and Gentlemen: We have audited the Department of Human Resources - Family Investment Administration for the period beginning July 1, 1997 and ending September 12, Our audit disclosed certain deficiencies with regard to the Administration s quality assurance procedures for assistance programs. For example, the Administration had not fully implemented a payment accuracy measurement system required by State law to ensure the propriety of eligibility determinations for cash assistance programs and the accuracy of the related benefit payments. Such payments totaled $117 million during fiscal year Because of excessive food stamp payment error rates dating back to Federal fiscal year 1994, the State incurred Federal sanctions of $8.3 million. In addition, monitoring of local departments of social services efforts to implement local quality assurance procedures needs improvement. Similar findings with respect to the Administration s quality assurance procedures were also included in our preceding audit report. Our audit also disclosed that the Administration had not fully restricted employee access to critical public assistance automated data files. Also, payment monitoring for two significant contracts was not comprehensive. Finally, for one program that assists certain disabled customers when applying for Federal benefits, the Administration did not ensure that all eligible customers were helped. Since customers who secure Federal benefits no longer qualify for State cash assistance, the State receives the benefit of the related savings. Furthermore, for those who were determined eligible, the Administration did not ensure that the local departments of social services were recovering the State s share of retroactive Federal benefits. Respectfully submitted, Bruce A. Myers, CPA Legislative Auditor

4 (This Page Intentionally Left Blank) 2

5 Table of Contents Executive Summary 5 Background Information 7 Agency Responsibilities 7 Current Status of Findings From Preceding Audit Report 7 Findings and Recommendations 9 Quality Assurance Procedures for Assistance Programs Finding #1 - Food Stamp Payment Errors Were Not Reduced, Resulting 9 in Federal Sanctions Of Over $8 Million Finding #2 - Payment Accuracy System for Temporary Cash Assistance 10 Was Not Fully Implemented As Required by State Law Finding #3 - Results of Local Departments Comprehensive Program 11 Review Systems Were Not Adequately Monitored Finding #4 - Computer Match Results Identifying Potential Unreported 13 Customer Income Were Not Resolved in a Timely Manner Temporary Cash Assistance Work Participation * Finding #5 Approximately Month Customers Were Not 14 Participating in a Work Activity as Required Contract Monitoring Finding #6 - Contractor Charges For Welfare Research Services Were 16 Not Adequately Verified Computer Access Finding #7 - System Access Capabilities for the Clients Automated 17 Resources and Eligibility System Were Not Adequately Restricted Disability Entitlement Advocacy Program (DEAP) Finding #8 - Inadequate Monitoring Resulted in Excess Vendor 18 Advances Finding #9 - Procedures Were Not Adequate to Ensure All Qualified 19 Customers Were Referred to the Program Finding #10 - Local Departments Recovery of State s Share of 20 Retroactive Federal Benefits Was Not Adequately Monitored * Denotes item repeated in full or part from preceding audit. 3

6 Audit Scope, Objectives and Methodology 23 Agency Response Appendix 4

7 Executive Summary Legislative Audit Report on the Department of Human Resources Family Investment Administration June 2001 The Family Investment Administration oversees a number of public assistance programs that are administered Statewide by the 24 local departments of social services. These programs include the Temporary Cash Assistance program (TCA), which is funded by both General and Federal funds, and the Federal Food Stamp program, which is entirely Federally funded. According to Department financial reports, assistance payments totaled $353 million during fiscal year 2000, of which $68 was from General funds. Quality assurance procedures for assistance programs were not fully implemented and local department implementation was not sufficiently monitored. Because of excessive food stamp payment error rates, the State incurred Federal sanctions in excess of $8 million since Federal fiscal year To satisfy these sanctions, the Administration must spend State funds, within specified time periods, on specific initiatives aimed at reducing food stamp errors. Furthermore, a state-mandated payment accuracy measurement system, intended to reduce cash assistance payment errors, was not fully implemented. In addition, the Administration did not exercise sufficient monitoring over a statewide case review program designed to reduce case manager errors. The Administration should fully implement comprehensive quality assurance procedures that include monitoring at the local department level in an effort to reduce payment errors and improve the accuracy of public assistance eligibility determinations. Approximately 540 temporary cash assistance (TCA) customers who had received benefits for 24 months were not participating in a work activity as required by law. The Administration, in conjunction with the local departments, should ensure that all 24-month TCA customers are participating in a required work activity or that appropriate action (such as termination of benefits) is taken for customers who fail to comply with the work requirement. 5

8 Contractor costs related to a $3.5 million cost reimbursement welfare research contract were not verified for propriety. Contractor costs should be verified through comparison with supporting documentation. Access to critical automated assistance files was not adequately restricted. Access to critical files should be restricted to employees requiring such access to perform their duties. Certain deficiencies were noted in the Disability Entitlement Advocacy Program (DEAP), which assists certain disabled public assistance customers in applying for Federal SSI and SSDI. Customers who secure Federal benefits no longer qualify for State cash assistance, and therefore the State receives the benefit of the related savings. We also noted that the contractor who administered the Program was advanced funds in excess of program expenses and such excess funds were not sufficiently collateralized. In addition, the Administration did not ensure that local departments obtained Federal reimbursement for State benefits paid to customers while awaiting Federal benefits. Adequate monitoring procedures should be established to ensure that funds paid to the contractor are reasonable in relation to actual Program costs. In addition, all Federal funds due to the State should be recovered. 6

9 Agency Responsibilities Background Information The Family Investment Administration oversees a number of public assistance programs that are administered Statewide by the 24 local departments of social services. The Administration s policy development and program oversight primarily address the Temporary Cash Assistance program (TCA), which is funded by both General and Federal funds, and the Federal Food Stamp program, which is entirely Federally funded. The TCA program was established as a result of the Federal and State legislation that restructured Maryland s welfare system from an entitlement program to a program that promotes individual responsibility and self-sufficiency. According to the Administration s statistical reports, as of June 30, 2000, the Department s TCA caseload was approximately 76,000 recipients, representing a decrease of 66 percent since January According to Department financial reports, assistance payments totaled $353 million during fiscal year 2000 (General funds - $68 million; Federal funds - $263 million; special funds - $22 million). The payments were primarily for the Temporary Cash Assistance program ($117 million) and the Federal Food Stamp Program ($200 million). Current Status of Findings From Preceding Audit Report We reviewed the current status of the 10 fiscal/compliance items related to the Administration and contained in our preceding audit report on the Department of Human Resources dated December 15, The scope of that audit report included the: Office of the Secretary and related units, Family Investment Administration, Social Services Administration and Child Care Administration, all of which are now subject to separate audits and reporting. We determined that the Administration satisfactorily addressed 6 of these findings. We have repeated the remaining 4 findings, one of which is separated into two comments in this report. 7

10 (This Page Intentionally Left Blank) 8

11 Findings and Recommendations Quality Assurance Procedures for Assistance Programs Finding #1 Food stamp payment error rates continued to exceed Federal target rates. Resulting Federal sanctions totaled $8.3 million during Federal fiscal years 1994 through Analysis The Administration s payment error rates for the Food Stamp Program have exceeded allowable rates (referred to as the Federal target rate) established by the United States Department of Agriculture (USDA) for Federal fiscal years 1994 through 1999 (the most recent available). The allowable error rate is based upon the nationwide average. We noted that in Federal fiscal years 1998 and 1999, Maryland had the fourth and seventh highest error rates in the nation, respectively. Because the Food Stamp Program is entirely Federally funded, errors that cause program overpayments result in a direct loss to the Federal government. Therefore, the USDA imposes fiscal sanctions on States that exceed the allowable error rate. The Administration s failure to achieve the allowable error rate established by the USDA was commented upon in our two preceding audits. Federal Fiscal Year State Food Stamp Error Rate 11.24% 12.09% 11.26% 12.83% 15.40% 13.62% Federal Target Rate (national 10.32% 9.72% 9.22% 9.88% 10.69% 9.88% average) Variance Favorable (Unfavorable) (.92%) (2.37%) (2.04%) (2.95%) (4.71%) (3.74%) In addition to the approximately $1 million the State paid to the Federal government as commented upon in our preceding audit report, the USDA sanctioned the State approximately $8.3 million because of the State s failure to meet the Federal target rates during these years. In order to satisfy these sanctions, the Administration entered into various reinvestment agreements with the USDA that, rather than requiring payment to the Federal government, allowed the Administration to use $8.3 million in State funds to finance specific initiatives starting in State fiscal year 1998 to reduce food stamp errors. As of September 9

12 30, 2000, expenditures totaling $4.8 million still needed to be reinvested within specified time periods, with the last reinvestment of $1 million to be completed by September 30, Furthermore, our review of the reinvestment agreements and related initiatives disclosed that the Administration did not ensure that accurate reinvestment expenditure reports were submitted in a timely manner. For example, the report due to the USDA by July 30, 2000 was still in draft form as of January 5, In addition, two initiatives totaling $1.1 million, proposed in the Administration s August 2000 plan, were denied by the USDA in October 2000 and, as of December 31, 2000, the Administration had not identified other initiatives in which to reinvest these State funds. Since each reinvestment agreement provides for funds to be reinvested over a period of several fiscal years, it is imperative that the Administration monitor the reinvestment expenditures. Funds not reinvested by the deadlines stipulated in the reinvestment agreements are to be repaid to the USDA. Recommendation #1 We again recommend that the Administration take immediate action to reduce the Food Stamp Program error rate to avoid future liabilities by monitoring the results of its initiatives to reduce payment errors and taking prompt remedial action when the results do not meet expectations. We also recommend that the Administration comply with all terms of the reinvestment agreements, and take appropriate action to ensure the timely submission of accurate expenditure reports. Finding #2 Payment accuracy measurement system for the temporary cash assistance program had not been fully implemented as required by State law. Analysis The Administration had not fully implemented a payment accuracy measurement system for its Temporary Cash Assistance (TCA) program, as required by State law. Chapter 486, Laws of Maryland, 1999, effective July 1, 1999, required the Administration to develop a comprehensive process to systematically analyze cash assistance payment errors, formulate strategies to reduce errors, and monitor the implementation of those strategies. In October 1999, the Administration implemented the Temporary Cash Assistance Quality Control System statewide, developed the first year s sampling plan, and began sampling cases for review at the local departments. Specifically, we noted the following: 10

13 Based on the Administration s statistical sampling plan, 660 cases should have been reviewed for the 12-month period ending September 2000; however, only 352 cases were reviewed. The Administration advised us that due to a lack of personnel, it was unable to complete the review of 660 cases. Following the sampling plan would ensure that the resultant accuracy rate is representative of the TCA case population. As of February 2, 2001, the Administration had not fully compiled the results of its payment accuracy reviews. Specifically, the Administration had compiled the results for only 268 of the 352 aforementioned cases reviewed during the period from October 1999 through September Based on the compilation of the 268 cases, the Administration calculated a preliminary payment accuracy rate of 90.7 percent. However, this rate could be affected by the results of the remaining 84 cases, and its statistical validity is questionable since the planned sample of 660 cases was not completed. During fiscal year 2000, payments for the TCA program totaled approximately $117 million. A similar comment regarding the need to establish a comprehensive process to analyze cash assistance payment errors was commented upon in our preceding audit report. Recommendation #2 We again recommend that the Administration fully implement a comprehensive process to analyze cash assistance payment errors, formulate strategies to reduce errors, and monitor the implementation of these strategies. In addition, we recommend that the Administration ensure that it follows its sampling methodology so that its payment accuracy rate is representative of the TCA case population. We also again recommend that the Administration compile the results of its reviews in a timely manner. Finding #3 The Administration did not adequately monitor the results of the Comprehensive Program Review System or ensure that local departments implemented appropriate corrective action based on system findings. Analysis The Administration did not adequately monitor the results of the Comprehensive Program Review System (CPRS) to ensure that local departments were adhering to system requirements or implementing appropriate strategies to improve performance of case managers. The CPRS is a process established by the Administration that enables the local departments to identify strategies and 11

14 processes for improving case manager performance by reducing or eliminating case manager errors. The cornerstone of the system is the case review process in which the work of individual case managers is periodically sampled by their supervisors to determine the correctness of their decisions and the propriety of related cash assistance and food stamp client benefits awarded. Our review of the CPRS disclosed instances of inadequate monitoring by the Administration. Case review results, which are summarized by all of the local departments and forwarded to the Administration for monitoring purposes, were not submitted by two of the larger local departments in the required format. As a result, the Administration could not readily analyze the findings and identify areas for corrective action for these two local departments. In addition, although the Administration required corrective action plans to address both cash assistance and food stamp errors, we noted that nine local departments had either not submitted corrective action plans or the plans submitted did not identify the actions to be taken to reduce the types of errors disclosed by the case reviews. Twelve local departments did not include all required components in their individualized CPRS implementation plans. For example, one plan did not contain information sufficient to quantify the number of case reviews required to be conducted, preventing effective monitoring by the Administration. The Administration did not ensure that the local departments took timely action to correct food stamp and cash assistance payment errors (meaning an improper payment was made) identified during the case review process. For example, during one month one local department reported over 250 errors that were unresolved in excess of the 21 days allowed by the Administration for correction. Failure to take timely corrective action could lead to a continuance of payment errors and have a potential adverse impact on the Federal food stamp error rate (see Finding #1). The Administration did not issue the reports of its periodic on-site reviews of the local departments CPRS compliance within 30 days of completion as required by its policy. As of November 3, 2000, the reports for 5 local departments were still in draft form, even though the reviews had been conducted from 3 to 8 months prior. Similar concerns pertaining to the quality assurance efforts of the local departments were commented upon in our two preceding audit reports. 12

15 Recommendation #3 We again recommend that the Administration effectively monitor the local departments compliance with the Comprehensive Program Review System requirements. We also again recommend that the Administration obtain and review reports on the results of the local department case reviews, and monitor and assess the success of the related strategies for improving performance. In addition, we recommend that the Administration issue results of its on-site reviews within the required 30-day period. Finally, we recommend that the Administration ensure that the local departments take appropriate action to correct errors, including the recovery of any overpayments. Finding #4 Computer match results disclosing potential unreported customer income were not resolved in a timely manner, and system problems have prevented certain unreported customer income from being identified for further followup. Analysis The Administration did not ensure the timely resolution of the results of various computer matches performed to independently verify the completeness and accuracy of information (for example, wage earnings) provided by the recipients of cash assistance and food stamps benefits. Various matches between the Administration s Clients Automated Resources and Eligibility System (CARES) and third parties, such as the statewide new hire registry maintained by the State s Department of Labor, Licensing and Regulation, are routinely performed and the results are forwarded to the local departments for follow-up and appropriate action. Our review of the results of these matches disclosed that local department followup was not always timely despite evidence of Administration monitoring. For example, based on Administration records, as of August 31, 2000, there were 542 new hire registry matches that had not been investigated and resolved for over 20 days, and 242 of these matches had been outstanding for over 80 days. The Administration s policy requires the local departments to resolve the new hire registry matches, which indicate that a customer has begun a new job, within 20 days. The timely resolution of these matches is crucial since unresolved matches may result in assistance overpayments. In addition, as a result of system problems with the new hire registry match, customers who already have earned income reported are not identified for further follow-up if they obtain subsequent employment. 13

16 It should be noted that Administration believes that the failure to detect and resolve the existence of earned income, such as would result from the new hire registry match, is the leading cause of the State s high food stamp error rate (see Finding #1) as well as for cash assistance payment errors. A similar comment regarding the need to investigate and resolve the results of computer matches was commented upon in our two preceding audit reports. Recommendation #4 We again recommend that the Administration and the local departments take immediate action to resolve the match results on a timely basis. We also recommend that the Administration, in conjunction with the Department of Labor, Licensing and Regulation, take corrective action to ensure that all new hire registry matches are identified. Temporary Cash Assistance (TCA) Work Participation Finding #5 Approximately 540 TCA customers who have received benefits for 24 months were not engaged in a work activity during fiscal year 2000 as required by law. Analysis Administration records indicate that for the fiscal year ending June 30, 2000 approximately 540 TCA customers, on average, were not engaged in a required work activity. This represents approximately seven percent of the 24-month TCA customers required to participate in a work activity. Federal and State law requires that customers who have received TCA benefits for 24 months be engaged in a work activity, and that a customer s failure to participate can lead to termination of benefits. Because of the significant decline in the State s TCA caseload, this condition did not have an impact on the State s receipt of Federal grant funds. Nonetheless, the Administration needs to continue its efforts so that the desired results of welfare reform (moving people from the welfare rolls to selfsufficiency) are maximized. Our review of Administration reports disclosed that for fiscal year 2000, the local departments reported that on average, 93 percent of the approximately 7,400 customers statewide who had received assistance for a 24-month period were engaged in a work-related activity. The remaining 7 percent (540 customers) were not in a work activity or otherwise exempted. The reports also disclosed that 14

17 six of the local departments reported a participation rate below 90 percent (including two local departments with participation rates of approximately 74 percent). In addition, our limited test disclosed that accurate customer participation data was not always reported to the Federal government as required. Specifically, Federal law also requires periodic reporting of actual participation rates, for which the Administration uses monthly statistical sampling. This sampling process requires that each local department verify and report the active participation of selected customers in a work activity. The Administration is also required to report the actual number of hours worked by those customers for the month sampled. However, we noted that the reported activity for TCA customers from one local department did not always agree to available documentation (such as time and attendance records), meaning that the local department could not substantiate the customers reported active participation in a work activity. In our performance audit report on the participation of temporary cash assistance customers in work activities, which was issued in December 1999, we commented upon a similar situation of less than 100 percent participation of customers in work activities. Also, in our preceding fiscal/compliance report we commented upon the lack of accuracy in the reporting of work activity to the Federal government. Recommendation #5 We recommend that the Administration, in conjunction with the local departments, ensure that all 24-month TCA customers are actively engaged in a work activity as required or that appropriate action (for example, terminate benefits when appropriate) is taken as provided for by law for those customers who fail to comply with the work activity requirements. Also, we recommend that the Administration ensure that the local departments maintain adequate documentation to substantiate TCA customers participation in all work activities, as required. 15

18 Contract Monitoring Finding #6 The Administration did not adequately monitor a welfare research and reporting contract entered into with a State university to ensure that payments were in accordance with the terms of the contract. Analysis The Administration had not taken action to verify the propriety of various contractual payments, including contractor salary and certain other direct costs. The Administration entered into a 22-month contract from September 1, 1999 through June 30, 2001, valued at approximately $3.45 million, with a State university to provide certain welfare consulting services. Payments were to be based on the actual time and materials expended by the contractor, and the contract included a fee schedule listing employees and their related hourly rates. The Administration had made payments totaling $1.6 million as of December 30, Since this contract was awarded as an inter-agency agreement, under State Procurement Regulations it was not subject to competitive procurement. Our review disclosed the Administration did not verify the propriety of direct labor charges and certain allocated direct expenses (such as rent and utilities), which comprised the vast majority of the total contract costs. Specifically, there was no verification of any allocated direct expenses billed and paid for even though these expenses represented more than 35 percent of the total project cost. Furthermore, billed rates for direct labor did not agree to the fee schedule in the contract and, in a number of cases, employees listed on the invoices were not included in the fee schedule. For example, on one monthly invoice, one employee s time was billed at 13 different hourly rates ranging from $23 to $46, all greater than the rate included in the contract ($22). A comparison of rates for the same invoice for all employees identified in the contract disclosed an overall 10 percent overpayment (approximately $8,100 based upon a total payroll-related costs of $84,000). Recommendation #6 We recommend that the Administration institute appropriate procedures to ensure the propriety of contractor costs through verification with supporting documentation (for example, employee time sheets, contractor expenditure records). In addition, we recommend that the Administration promptly identify and recover any overpayments. 16

19 Computer Access Finding #7 System access capabilities for the Clients Automated Resources and Eligibility System were not adequately restricted. Analysis The Administration did not fully utilize the security features of the Clients Automated Resource and Eligibility System (CARES) to restrict user access and prevent unauthorized transactions. The Administration uses CARES to record and authorize cash assistance and food stamp benefits to eligible customers. During fiscal year 2000, the Administration paid cash assistance and food stamp benefits totaling $117 million and $200 million, respectively. Our review disclosed that 45 Administration employees were assigned a CARES access code that allowed unrestricted access to all local department public assistance data (for example, to change benefit amounts). We were advised by the Administration that some of these employees require this access to perform their job function. Nevertheless, the Department of Human Resources Office of Information Management Security Handbook states that such access should only be given to a very limited number of staff. Recommendation #7 We recommend that the Administration reassess such access and restrict critical system capabilities to employees whose job responsibilities require such access. Disability Entitlement and Advocacy Program (DEAP) Background In June 1997, the Administration entered into a three-year $8.5 million contract with a vendor to administer the Disability Entitlement Advocacy Program (DEAP). Since entering into the contract, several modifications and extensions occurred, which increased the contract s total value to approximately $16.6 million and extended the contract period through April 30, As of September 30, 2000, the Administration had paid the vendor $14 million under this contract. 17

20 This Program assists certain disabled public assistance customers in applying for Federal Supplemental Security Income (SSI) and Social Security Disability Income (SSDI). By helping to successfully secure Federal benefits, State funds are saved which would otherwise be spent to maintain customers in the various assistance programs. These assistance programs include Transitional Emergency Medical and Housing Assistance (TEMHA), Public Assistance to Adults (PAA), Temporary Cash Assistance (TCA) and Foster Care. In addition, if Federal benefits are secured, the State is to receive reimbursement for the State s share of customer benefits paid out during the related application period. Since the inception of the contract, the vendor s status reports indicate that, as of September 2000, it has received 110,221 customer referrals and has obtained Federal benefits for 32,542 customers. Finding #8 The Administration did not adequately monitor contract costs resulting in funds totaling $1.9 million being advanced to the vendor beyond actual expenditures. Analysis The Administration did not adequately monitor the contract for the administration of the State s Disability Entitlement Advocacy Program (DEAP) to ensure that the vendor was advanced reasonable amounts for services rendered. Although the vendor deposits funds advanced by the Administration into a money market account, and remits the related interest earnings to the State, the vendor s June 30, 2000 audited financial statements noted that the majority of the $1.9 million in funds on deposit were at risk since they were in excess of Federal Deposit Insurance limits and not otherwise collateralized. Furthermore, although the vendor had remitted to the Administration over $140,000 in interest earned on the excess funds that it had on deposit during fiscal years 1998 to 2000, these funds were not returned to the State s General Fund as required by State law. Instead, the Administration used these funds to reduce operating expenditures, which effectively increased the Administration s appropriations. This situation was caused by the Administration s policy of paying the vendor based on monthly budgeted expenditures ($458,000 during fiscal year 2000), rather than on budgeted expenditures adjusted for prior months actual expenditures (which averaged $320,000 during fiscal year 2000) as stipulated in the contract. Furthermore, the Administration did not receive reports of actual expenditures from the vendor as required by the contract. For example, in April 2000 based on budgeted expenditures, the vendor requested, and later received, a $2.3 million payment to cover expenses for the period from February 2000 through 18

21 June The vendor s audited financial statements reported that, as of June 30, 2000 however, $1.9 million in advanced funds remained on deposit for future Program services and costs. Furthermore, this amount remained even after, at the Administration s request, the vendor returned $1.5 million in excess funds to the State on June 5, Recommendation #8 We recommend that the Administration obtain monthly reports of actual expenditures from the DEAP vendor and make vendor payments in the manner required by the contract. We also recommend that the Administration revert any interest earnings to the State s General Fund. Finding #9 Adequate procedures were not in place to ensure that disabled TCA customers were referred to DEAP by the local departments as required. Analysis The Administration lacked assurance that TCA customers with 12-month or longer disabilities were referred to the DEAP by the local departments as required. Specifically, based on our review of Administration records, as of September 2000, there were 1,390 customers identified on CARES that, based on eligibility criteria, should have been referred to DEAP; however, vendor records only documented 632 referrals at that time. Even when CARES indicated that referrals had actually occurred there was a lack of assurance that the vendor had received them. A CARES report identified 84 customers that had been referred to DEAP by four local departments in September 2000, however, vendor records only identified 37 customer referrals from these local departments. These non-referrals could represent actual lost funds to the State since the eventual shift of financial burden to the Federal government has been delayed. According to Administration records, the average annual cost for a TCA customer is about $1500. Recommendation #9 We recommend that the Administration establish procedures to ensure that eligible TCA customers are referred to DEAP as required. This process should also include verification that customers referred by the local departments are actually received by the vendor. 19

22 Finding #10 The Administration did not ensure that the local departments were obtaining reimbursement for Federal retroactive assistance payments. Analysis The Administration did not ensure that the local departments were obtaining reimbursement for State funded benefits paid to disabled TCA customers while awaiting Federal benefits (known as interim assistance reimbursement or IAR). Once the customer is approved for SSI or SSDI, the Federal government will issue to the appropriate local department a retroactive payment for the Federal assistance payments beginning with the date of application. The State is permitted to recover from this payment, the amount of benefits it paid to the customer during that period. This calculation to determine the State s share is performed by either the Federal government or the State depending upon the amount of the payment. When performed by the State, it is the Administration s policy that if the local department does not complete the calculation within 10 days of receipt of the check, then the entire amount of Federal assistance must be returned to the customer. Although Federal guidelines establish a goal of 10 days to perform this calculation, if the goal is not met they do not require the State to return the entire amount to the customer. The Administration advised us that it was not monitoring the local departments to ensure that the IAR s were being properly calculated and collected. In April 2001 we contacted three of the four largest local departments and were advised by representatives from two local departments that they were not recovering IAR s for TCA customers because they were unaware of the requirement to do so. The third local department advised that it was recovering IAR s for TCA customers, but appeared to be using an inappropriate methodology for calculating the IAR. Because of the complexity of the IAR calculation, the Administration submitted a CARES modification request to the Department of Human Resources Office of Information Management in June 1999 to have the automated system calculate the IAR for TCA customers. However, this modification had not been made as of January 5, Based on certain estimates provided by the DEAP vendor in its technical proposal, we calculated that as of September 2000, the potential State reimbursement from the IAR could have totaled in excess of $500,000 since November

23 Recommendation #10 We recommend that the Administration take immediate action to ensure that all local departments are properly calculating the IAR for applicable TCA customers, and that all funds due the State are recovered. We also recommend that the Administration discontinue the practice of returning the State s share of retroactive Federal assistance payments to customers. 21

24 (This Page Intentionally Left Blank) 22

25 Audit Scope, Objectives and Methodology We have audited the Department of Human Resources - Family Investment Administration for the period beginning July 1, 1997 and ending September 12, The audit was conducted in accordance with generally accepted government auditing standards. As prescribed by the State Government Article, Section of the Annotated Code of Maryland, the objectives of this audit were to examine the Administration s financial transactions, records and internal control, and to evaluate its compliance with applicable State laws, rules and regulations. We also determined the current status of the findings contained in our preceding audit report. In planning and conducting our audit, we focused on the major financial related areas of operations based on assessments of materiality and risk. Our audit procedures included inquiries of appropriate personnel, inspection of documents and records, and observation of the Administration s operations. We also tested transactions and performed other auditing procedures that we considered necessary to achieve our objectives. Our audit did not include certain support services provided by the Department of Human Resources - Office of the Secretary. These support services (such as processing of disbursements for operating expenses, maintenance of accounting records, and related fiscal functions) are included within the scope of our audits of the Office of the Secretary. We did not audit the Administration s Federal financial assistance programs for compliance with Federal laws and regulations because the State of Maryland engages an independent accounting firm to annually audit such programs administered by State agencies. The Administration s management is responsible for establishing and maintaining effective internal control. Internal control is a process designed to provide reasonable assurance that objectives pertaining to the reliability of financial records, effectiveness and efficiency of operations including safeguarding of assets, and compliance with applicable laws, rules and regulations are achieved. 23

26 Because of inherent limitations in internal control, errors or fraud may nevertheless occur and not be detected. Also, projections of any evaluation of internal control to future periods are subject to the risk that conditions may change or compliance with policies and procedures may deteriorate. Our reports are designed to assist the Maryland General Assembly in exercising its legislative oversight function and to provide constructive recommendations for improving State operations. As a result, our reports generally do not address activities we reviewed that are functioning properly. This report includes findings related to conditions that we consider to be significant deficiencies in the design or operation of internal control that could adversely affect the Administration s ability to maintain reliable financial records, operate effectively and efficiently and/or comply with applicable laws, rules and regulations. Our report also includes findings regarding significant instances of noncompliance with applicable laws, rules, or regulations. The Department of Human Resources response, on behalf of the Administration, to our findings and recommendations is included as an appendix to this report. As prescribed in the State Government Article, Section of the Annotated Code of Maryland, we will advise the Department regarding the results of our review of its response. 24

27

28 Findings and Recommendations Quality Assurance Procedures for Assistance Programs Finding #1 Food stamp payment error rates continued to exceed Federal target rates. Resulting Federal sanctions totaled $8.3 million during Federal fiscal years 1994 through Recommendation #1 We again recommend that the Administration take immediate action to reduce the Food Stamp Program error rate to avoid future liabilities by monitoring the results of its initiatives to reduce payment errors and taking prompt remedial action when the results do not meet expectations. We also recommend that the Administration comply with all terms of the reinvestment agreements, and take appropriate action to ensure the timely submission of accurate expenditure reports. Departmental Response Error Rate Reduction We agree and appropriate actions are being taken. The Administration is actively working to reduce the Food Stamp error rate. The error rate for FFY 2000 is 11.06%. This is a reduction of 2.56 percentage points over our FFY 1999 error rate. We will review the results of the initiatives currently in place and make adjustments in areas where the results are not effective. Expenditure Monitoring We have established a fiscal chart of accounts for each reinvestment strategy in each spending plan. We anticipate this will provide for improved monitoring and a reduced number of apparent discrepancies between the expenditures provided on the Administration's reports and those reported by the Department's Office of the Secretary. We will maintain documentation of any required adjustments made. A separate chart of accounts should also improve the timeliness of preparing reports. We have also dedicated a staff position to track reinvestment expenditures. Additional staff have been assigned to track local department monthly activity and expenditure reports to ensure they are submitted regularly and timely and to prepare status reports.

29 Finding #2 Payment accuracy measurement system for the temporary cash assistance program had not been fully implemented as required by State law. Recommendation #2 We again recommend that the Administration fully implement a comprehensive process to analyze cash assistance payment errors, formulate strategies to reduce errors, and monitor the implementation of these strategies. In addition, we recommend that the Administration ensure that it follows its sampling methodology so that its payment accuracy rate is representative of the TCA case population. We also again recommend that the Administration compile the results of its reviews in a timely manner. Departmental Response We agree. The development of strategies to analyze and reduce cash payment errors and to ensure that errors identified by Quality Control are corrected was only partially implemented due to the fact that no positions were available to complete this function until October In the interim, existing staff did a random sample of cases to follow-up on identified errors. All three (3) TCA Monitor positions were filled between late November and mid December Staff have since completed random monitoring of the sample in accordance with the plan. A Standard Operating Procedure (SOP) is in the process of being developed. The TCA Monitors have also begun to develop the comprehensive process of analyzing cash assistance payment errors by developing a database for monitoring such cases. We have developed strategies to reduce errors and monitor the implementation of those strategies, and we are working with the local departments via telephone, mail and conducting site visits when necessary. As noted in the report, less than 660 cases were completed in the FFY 2000 sample due to lack of sufficient staff. Two additional PINS for Quality Control Analysts became available in December Those PINS have now been filled, and the new staff will begin work May 30, We expect that these staff will be trained and proficient by September Four additional reviewer positions are available, but they are contractual positions. Recruitment is underway to fill these positions, but response to the recruitment notice has been minimal. These positions were converted to permanent positions effective July We expect to be able to recruit successfully to fill these positions. These staff should be on board by September 2001 and trained and proficient by January With this additional staff, we will be able to complete a statistically representative statewide sample for the FFY 2002 sample year. 2

30 The results of the FFY 2000 TCA sample were compiled on February 13, 2001, and produced a 91.17% payment accuracy rate for the year. For the first three months of the current sample year (October to December 2000) the payment accuracy rate was 92.1%. Currently, the reports are compiled quarterly due to the limited sample size. These are also prepared after the food stamp reports are completed to enable us to meet the federal timelines required by that program. Federal food stamp regulations require that all food stamp reviews be completed by the 95 th day following the end of the sample month (ex: October reviews must be completed by February 3). States can be assessed additional federal penalties by USDA for failure to meet those deadlines. Therefore, we anticipate continuing to give priority to meeting the federally imposed food stamp requirements in completing reviews and reports. Finding #3 The Administration did not adequately monitor the results of the Comprehensive Program Review System or ensure that local departments implemented appropriate corrective action based on system findings. Recommendation #3 We again recommend that the Administration effectively monitor the local departments compliance with the Comprehensive Program Review System requirements. We also again recommend that the Administration obtain and review reports on the results of the local department case reviews, and monitor and assess the success of the related strategies for improving performance. In addition, we recommend that the Administration issue results of its on-site reviews within the required 30-day period. Finally, we recommend that the Administration ensure that the local departments take appropriate action to correct errors, including the recovery of any overpayments. Departmental Response Please note, Montgomery and Anne Arundel Counties are submitting the results of their reviews in an approved format under their Local CPRS Plan. The results of their target reviews, as approved in their Local Plan are not currently adaptable to the central office s CPRS database for analysis. We have requested that departments on Local CPRS Plans submit updated Local CPRS Plans for approval. Approval will require: A procedure to regularly analyze the results of the local department's target error element reviews to identify areas where corrective action is needed or to submit their review findings to the central office on the standard 102 review form for analysis by the central office, and 3

31 A procedure to regularly review other error data sources (such as Quality Control error data) to assess the effectiveness of the established target error element and to modify the target error element when appropriate. Additionally, the central office has requested all local departments to submit updated Corrective Action Plans (CAP). Approval will require, for each strategy, identification of the following: The type of error that the corrective action is planned to address, and The source of data used to identify the error problem. Local departments with Food Stamp error rates above the national tolerance level will be required to update their Corrective Action Plans annually. All local departments cited for deficiencies in their local Standard Operation Procedures (SOP) that were submitted as required by AT #00-07 (revised) have been requested to amend their SOP to address the deficient areas. The central office experienced significant difficulties in implementing the Secondlevel Review monitoring requirements of the revised CPRS due to an unanticipated staff vacancy rate in the central office Bureau. This resulted in unacceptable delays in producing reports within the required thirty (30) days. Recent recruitment efforts have succeeded in filling most of the vacancies. We have eliminated the CPRS reports backlog. There is currently one (1) report that is ten (10) days behind schedule. We anticipate producing ongoing Second-level CPRS Reports timely. We have requested local departments to establish a procedure to ensure that errors identified by CPRS reviews are corrected within the required twenty-one (21) days and that overpayments identified by CPRS reviews are processed for collection promptly. In addition, the central office will review the effectiveness of the local department procedure concurrent with performing the Second-level CPRS reviews in local departments. The central office will also follow-up to ensure that all errors identified by the central office s Second-level Reviews are corrected and that a random sample of deficiencies identified are corrected. Finding #4 Computer match results disclosing potential unreported customer income were not resolved in a timely manner, and system problems have prevented certain unreported customer income from being identified for further followup. Recommendation #4 We again recommend that the Administration and the local departments take immediate action to resolve the match results on a timely basis. We also 4