Nicholas Pacifico
Author
Improper Payment Recovery Processes, Necessary (Yet Neglected) Triage
Introduction
Improper payments is one of the largest causes of lost revenue for the United States. In past reports, the Project On Government Oversight has explored what improper payments are and why identifying them is so problematic.[1] Recovering improper payments is equally challenging. While preventing them is the ultimate goal, effective recovery efforts are needed to curb losses until that goal can be met.
This report initially set out to identify, examine, and develop best practices for the improper payment recovery processes of inspectors general (IGs); however, this turned out to be problematic as many IGs play little to no role in their agency’s recovery processes. To better illuminate the recovery process, we outline the legal framework for improper payment recovery, discuss the history of recovery efforts so far, and propose areas where IGs can become more effective in facilitating the recovery of improper payments.
Background
The Improper Payments Information Act of 2002 (IPIA) touches on recovering improper payments, but is more focused on outlining and identifying the problem than on trying to fix it.[2] The amendments to IPIA, the Improper Payments Elimination and Recovery Act of 2010 (IPERA) and the Improper Payments Elimination and Recovery Improvement Act of 2012 (IPERIA), establish procedures and protocols for the recovery of improper payments and are implemented through guidance from the Office of Management and Budget (OMB).[3]
OMB guidance requires that agencies have a cost-effective program of internal controls to prevent, detect, and recover overpayments. These programs may include policies and activities such as prepayment reviews, a requirement that all relevant documents be made available before making payment, and performance of post-award audits. OMB notes that effective internal controls may include:
- Payment recapture auditing techniques such as data matching with Federal, State, and local databases; and
- Data mining and predictive modeling to identify improper payments.[4]
While agencies do report non-audit recoveries, OMB has seemingly doubled down on “payment recapture audits” (PRAs), which have become the sole focus of OMB’s guidance to agencies regarding improper payment recovery.[5]
Payment Recapture Audits
OMB defines a Payment Recapture Audit (PRA) as “a review and analysis of an agency’s or program’s accounting and financial records, supporting documentation, and other pertinent information supporting its payments, that is specifically designed to identify overpayments.”[6] In other words, the agency specifically looks over its records for instances where it paid too much money. PRA’s were originally required only for agencies that entered into contracts with a total value in excess of $500 million in a fiscal year, but IPERA lowered that threshold to $1 million.[7] It is important to note that PRAs are not audits as defined and covered by Generally Accepted Government Auditing Standards, and thus are not required to meet those auditing standards.[8] Rather, they are solely an agency management function and responsibility, with agency heads responsible for determining the payment recapture activities that are expected to yield the most cost-effective results.[9]
OMB holds that PRAs should be, and only need be, undertaken if they are most likely going to be cost-effective. A cost-effective PRA is one in which the recaptured amounts exceed the costs associated with implementing and overseeing the program, such as staff time—in other words, one that recovers more money than it spends.[10] Some specific factors that agencies may consider when determining the cost-effectiveness of a PRA include whether:
- Recovery is legally allowed,
- The overpaid entity has the non-federal funds or resources needed to repay any overpayments,
- The overpaid entity has likely grounds to contest an overpayment finding, or
- Techniques or software exist to efficiently identify overpayments and diminish the need for costly manual review.
If an agency determines that a PRA would not be cost-effective, it must notify its IG and the OMB of its decision and the analysis used in arriving at that decision.[11] An agency must also report in its yearly financial report all programs and activities for which it determined a PRA would not be cost-effective, and the reasons and analysis it used to make such determinations.[12] OMB has the power to override an agency’s decision and require the agency to follow through with a PRA.[13]
OMB guidance allows PRAs to be conducted by:
- an agency’s own employees,
- other departments or agencies on an agency’s behalf,
- government contractors,
- other non-Federal entities who spend Federal funds (including state, local, non-profit, or other federal award recipients),[14] or
- any combination of the above.[15]
Agencies can utilize any of the above entities to audit their documents and recover any identified improper payments, and they can provide funds to external entities if necessary to assist with these efforts.[16] Any instances of potential fraud discovered through PRAs are to be reported immediately to the appropriate parties depending on the specific agency’s policies, such as to the agency’s IG office or the Department of Justice.[17]
Once the PRA has collected any recoveries, the funds are distributed in a variety of ways depending on the type of account the money was spent from.
Recoveries from non-expired discretionary funds must be returned to the fund from which they were appropriated or used to pay PRA contractors; they cannot be used for any other purposes.[18] Overpayments from mandatory, trust, or special fund accounts must be returned to their respective accounts with no exceptions.[19]
Recoveries from expired discretionary funds are distributed at the discretion of the relevant agency heads, although amounts allocated to the financial management improvement program, the appropriation or fund from which the overpayment was made, or inspector general activities are subject to maximum limits under IPERA.[20] For example, inspectors general can only be allocated up to 5 percent of any recoveries.[21]
At the end of each fiscal year (FY), agencies are required to report the results of their recovery efforts in their Agency Financial Reports (AFRs) and Performance and Accountability Reports (PARs). IPERIA requires agencies to specifically report on their recovery audit contract programs, and provide specific information on amounts of payments recovered by audit contractors.
This information has been reported for less than a decade, but it provides valuable insights into the effectiveness of OMB guidance on the recovery process.
The Numbers
Methodology and Caveats
The quality of PRA results is tied to the accuracy of an agency’s accounting, financial, and other supporting information. As discussed in our previous reports, a lack of transparency and accuracy of these systems is a major concern when it comes to examining the improper payment data reported by agencies, and it is no less a concern when thinking about agency recovery efforts. Effective discussions about how to improve improper payment identification, prevention, and recovery hinge on the government’s ability to enhance this transparency and accuracy, without which any analysis or comparison of recovery numbers should be inherently suspect, including those in this report.
The numbers reported below are aggregated from AFRs and PARs for FYs 2008 through 2015 for all agencies that reported improper payments.[22] We chose fiscal year 2008 as the start of the date range to allow for a sense of how recovery efforts were before IPERA implemented more established recovery processes in 2010, and FY 2015 as the end of the range because it is the most recent year for which data has been released. In instances where estimates, identifications, or recoveries for a given year were updated in future AFRs or PARs, POGO used the most recently released data in its calculations. This is under the assumption that the most recent information will be the most accurate.
Additionally, this report makes reference to general “recovery rates” and “effective recovery rates” (ERR). “Recovery rates” refers to the amount of improper payments recovered by an agency (or agencies) divided by the amount it identified for recovery; “effective recovery rates” is the amount of improper payments recovered by an agency (or agencies) divided by the total amount of improper payments it is estimated to have made.
There are issues with this approach to effective recovery rates, however. First, improper payments estimates include both overpayments and underpayments, while PRAs focus on overpayments. To make things even more complicated, PRAs are expected to fix any underpayments they find in the course of their audits, but there is no standardized methodology for doing so or for reporting on such actions. For simplicity, we made no adjustments to our effective recovery rate calculations, but we think it is important to note these qualifications. Historically, overpayments have accounted for 88 percent of improper payments, with a rate of 92 percent for FY 2015.[23]
With this information in mind, the historical trends of improper payment recoveries provide interesting insights into how the law and OMB guidance has affected these processes.
The Actual Data
There were over $893 billion in improper payments between FYs 2008 and 2015, which is in line with estimates of over $1 trillion since 2003. Only around $203 billion was identified for recovery during that time, with only about half of that (just over $103 billion) actually recovered.
Prior to IPERA in 2010, recovery information was not required and the threshold to trigger a PRA was substantially higher. Implementation of IPERA seems to have resulted in increased recovery efforts, with a jump in identifications and recoveries in 2011 and a continued increase in both through 2013. Surprisingly, identifications and recoveries declined in 2014, then stagnated through 2015, while estimates continued to rise from a small dip in 2012.
PRAs have helped identify many more improper payments as compared to non-PRA efforts—$121 billion versus about $82 billion—even with the PRA threshold being significantly higher for three out of the eight years. The recovery rate for PRAs is much less than that of the non-PRA efforts—31.13 percent versus 79.88 percent, respectively—but PRAs seem to have substantially increased total recovery dollars per year, ranging from $345 million in 2008 to almost $20 billion in 2015.
That being said, the effective recovery rate still lags far behind where it needs to be, with only 14.35 percent of estimated improper payments recovered in FY 2015. And again, that is assuming the estimates are correct. Various sources have told POGO that the current estimates are most likely significantly lower than the true number of improper payments, which would only cause this effective recovery rate to fall.
It seems clear that recovery efforts need to improve, and quickly.
Potential for Improvement
There is a lot of work to be done to improve recovery efforts. While PRAs are the current focus of OMB guidance, non-PRA recoveries still return a significant amount of money. Enhancing both identification methods is key to attaining a better grasp on unrecovered improper payments. Both IGs and agencies have a role to play in these enhancements: IGs need to become more involved in the recovery process, and agencies need to get better at sharing data with their IGs and each other.
IGs can be used more effectively
The law and OMB guidance make little mention of IGs playing a role in improper payment recovery other than specifying that IGs are to receive reports on PRA cost-effectiveness determinations and sometimes in cases of fraud. Yet IGs have skills and expertise that make them ideal for performing PRAs. IGs live and breathe audits and investigations, and generally have a high return on investment (ROI) to boot. Some of the highest reporters of improper payments have IGs with the highest ROIs—for example, Health and Human Services’ and the Social Security Administration’s IGs had ROIs of $19.38 and $43.60 respectively between 2010 and 2014.[24] IGs should be, at a minimum, main contenders for executing PRAs where reasonable. As always, the IGs’ independence should be a top consideration when making such decisions. That said, requiring that IGs be at the top of the list when agencies contemplate PRAs would be a step in the right direction, although it might require reworking IPERA and increasing incentives for IGs to participate in this process.
IGs are also missing opportunities to comment and provide constructive feedback on recovery processes. IGs are required to receive reports whenever their agency decides a PRA would not be cost effective. It seems that it would be in the purview of the IG to analyze and report on such decisions and provide suggestions for the future. But, aside from legal compliance issues, there are very few instances where IGs chose to make such efforts. While it varies between agencies, this appears to be an issue resulting from a lack of resources and incentives to go beyond what the law requires.
One avenue to get IGs more involved is to increase the maximum amount of recoveries that can be allotted to IG offices. As stated above, the current maximum is 5 percent of recoveries, and most agencies do not take advantage of that. Whether this is because IGs are generally not involved in the process or because agencies do not want them to be is unknown. However, IGs definitely will not become more involved in this process if not required to or if there is no incentive to do so.
If it turns out incentives or resources are not the main issue, Congress could always specify the roles they want IGs to play in the process via legislation. IG offices we have talked to seem interested in pursuing improper payment recoveries if they could get the authority and resources to do so.
IGs are also running into barriers when trying to access additional data to make better decisions in their non-PRA efforts. This is discussed more in depth in our second report, but it bears mentioning now.
Recovery information should be aggregated by OMB
The lack of progress in improper payment recoveries is partly due to inadequate consolidation of information from various agencies. Agencies are required to submit reports to OMB detailing their improper payment recovery efforts, including what they did to recover overpayments and how much they recovered. Some agencies are better at this than others, and could have processes and best practices that other agencies could learn from. Yet there is no easy way for agencies to review each other’s work. If an agency were interested in figuring out how to better its own process by learning from good examples, they’d first have to figure out which agencies would provide those examples and then gain access to the improper payment identification and recovery information. It would be a time- and labor-intensive exercise. There does appear to be a relatively simple solution, however.
Agencies are currently required to send all of their improper payment data, including recovery information, to OMB. OMB then publishes a subset of this data, none of it related to the reported recovery efforts, on their PaymentAccuracy.gov website.[25] Since OMB is already required to receive all that data, it could compile and report the recovery data while creating its annual report of improper payment information for PaymentAccuracy.gov. Even if it doesn’t report it publicly, it could send a report to each agency with improper payments about the current state of recovery efforts on a government-wide level. Given that OMB is beginning its revamp of PaymentAccuracy.gov, this seems like a great time to work the additional information into their reporting process.[26] This would not be a significant change given OMBs access to and familiarity with the data, so there doesn’t appear to be a reason to not take such a common-sense, low-cost action.
Conclusion
While the recovery process is improving, much can still be done. Concerns over the accuracy and quality of information reported continues to inhibit any ability to effectively discuss next steps for the improper payment system. Moreover, recovery processes will not improve until agencies and IGs can compare their results and processes with one another to determine best practices. Ultimately, at a minimum, IGs need to become more involved in the recovery process, and OMB needs to revamp its reporting by including recovery data on its PaymentAccuracy.gov platform.
Appendix A
Agency Financial Report Links
Corporation for National Community Service
Department of Health and Human Services
Department of Homeland Security
Department of Housing and Urban Development
Department of Veterans Affairs
Environmental Protection Agency
Federal Communications Commission
General Services Administration
Office of Personnel Management
Social Security Administration
Appendix B
Sources: AFRs and PARs for FYs 2008 through 2015 for all agencies with improper payments.
Appendix C
Improper Payment Totals by Year for FYs 2008 through 2015, Sorted by Year
Year | Total IP Estimated ($) | Total IP Identified ($) | Total IP Recovered ($) | Identification Rate | Recovery Rate | Effective Recovery Rate |
2015 | 137,008,362,000 | 42,979,233,700 | 19,654,398,800 | 31.37% | 45.73% | 14.35% |
2014 | 126,323,192,000 | 42,078,482,233 | 19,933,960,383 | 33.31% | 47.37% | 15.78% |
2013 | 107,275,582,000 | 44,479,544,722 | 22,508,831,795 | 41.46% | 50.60% | 20.98% |
2012 | 106,856,833,000 | 40,868,052,328 | 19,275,505,969 | 38.25% | 47.17% | 18.04% |
2011 | 114,670,751,947 | 25,645,608,364 | 16,868,117,501 | 22.36% | 65.77% | 14.71% |
2010 | 127,358,286,000 | 6,289,073,020 | 4,171,237,553 | 4.94% | 66.33% | 3.28% |
2009 | 97,952,432,000 | 444,015,999 | 357,607,631 | 0.45% | 80.54% | 0.37% |
2008 | 76,106,299,000 | 406,765,260 | 345,012,018 | 0.53% | 84.82% | 0.45% |
Total | 893,551,737,947 | 203,190,775,626 | 103,114,671,650 | 22.74% | 50.75% | 11.54% |
Sources: AFRs and PARs for FYs 2008 through 2015 for all agencies with improper payments.
Appendix D
Improper Payment Totals for FYs 2008 through 2015 by Agency,
Sorted by Effective Recovery Rate
Department | Total IP Estimated ($) | Total IP Identified ($) | Total IP Recovered ($) | Identification Rate | Recovery Rate | Effective Recovery Rate |
GSA | 117,580,000 | 319,995,810 | 216,119,733 | 272.15% | 67.54% | 183.81% |
RRB | 588,790,000 | 447,100,000 | 387,000,000 | 75.94% | 86.56% | 65.73% |
OPM | 3,896,550,000 | 2,401,580,000 | 2,180,390,000 | 61.63% | 90.79% | 55.96% |
EPA | 197,690,000 | 139,864,405 | 94,516,663 | 70.75% | 67.58% | 47.81% |
DoD | 8,035,160,000 | 2,904,747,940 | 3,095,310,185 | 36.15% | 106.56% | 38.52% |
SSA | 67,453,010,000a | 96,659,921,000 | 19,756,814,000 | 143.30% | 20.44% | 29.29% |
HUD | 8,781,580,000 | 5,205,779,488 | 1,905,791,653 | 59.28% | 36.61% | 21.70% |
FCC | 395,292,000 | 424,305,000 | 72,069,052 | 107.34% | 16.99% | 18.23% |
HHS | 534,028,290,000 | 79,553,864,237 | 66,102,038,862 | 14.90% | 83.09% | 12.38% |
DOL | 72,489,050,000 | 12,214,801,271 | 6,779,185,120 | 16.85% | 55.50% | 9.35% |
USDA | 43,815,120,000 | 1,707,147,000 | 1,732,686,000 | 3.90% | 101.50% | 3.95% |
VA | 16,172,982,000 | 329,504,800 | 315,772,800 | 2.04% | 95.83% | 1.95% |
DHS | 2,178,620,000 | 38,435,000 | 31,909,000 | 1.76% | 83.02% | 1.46% |
Education | 10,959,355,000 | 450,790,000 | 132,390,000 | 4.11% | 29.37% | 1.21% |
Transportation | 3,891,213,000 | 33,473,735 | 25,597,087 | 0.86% | 76.47% | 0.66% |
SBA | 3,624,550,000 | 11,393,985 | 2,940,229 | 0.31% | 25.81% | 0.08% |
Treasury | 116,900,000,000 | 83,690,094 | 80,055,950 | 0.07% | 95.66% | 0.07% |
CNCS | 26,905,947 | 0 | 0 | 0.00% | N/A | 0.00% |
Energy | 0 | 109,460,000 | 102,940,000 | N/A | 94.04% | N/A |
DOJ | 0 | 94,536,861 | 77,236,316 | N/A | 81.70% | N/A |
DOC | 0 | 55,588,000 | 23,580,000 | N/A | 42.42% | N/A |
Interior | 0 | 4,797,000 | 329,000 | N/A | 6.86% | N/A |
Total | 893,551,737,947 | 203,190,775,626 | 103,114,671,650 | 22.74% | 50.75% | 11.54% |
Sources: AFRs and PARs for FYs 2008 through 2015 for all agencies with improper payments.
Note: Rates listed as N/A are not available because they would require dividing by zero.
a SSA improper payment estimates for FY 2015 used in the estimates section are from Office of Management and Budget, “PaymentAccuracy – Supplemental Security Income (SSI).” https://paymentaccuracy.gov/tracked/supplemental-security-income-ssi-2015 (Downloaded September 30, 2016); Office of Management and Budget, “PaymentAccuracy – Retirement, Survivors, and Disability Insurance (RSDI).” https://paymentaccuracy.gov/tracked/retirement-survivors-and-disability-insurance-rsdi-2015 (Downloaded September 30, 2016)
Appendix E
Improper Payment Identifications and Recoveries by PRAs and Non-PRA efforts for
FYs 2008 through 2015, Sorted by PRA Identifications
Department | PRA Identifications ($) | PRA Recoveries ($) | Non-PRA Identifications ($) | Non-PRA Recoveries ($) | |
SSA | 96,652,751,000 | 19,749,944,000 | 7,170,000 | 6,870,000 | |
DOL | 12,207,610,000 | 6,778,720,000 | 7,191,271 | 465,120 | |
HHS | 10,682,334,237 | 9,924,219,862 | 68,871,530,000 | 56,177,819,000 | |
DoD | 982,807,940 | 799,750,185 | 1,921,940,000 | 2,295,560,000 | |
VA | 206,713,800 | 173,351,800 | 122,791,000 | 142,421,000 | |
FCC | 154,304,000 | 17,518,552 | 270,001,000 | 54,550,500 | |
GSA | 146,368,829 | 74,922,303 | 173,626,981 | 141,197,430 | |
HUD | 123,210,000 | 49,349,000 | 5,082,569,488 | 1,856,442,653 | |
Energy | 88,590,000 | 82,630,000 | 20,870,000 | 20,310,000 | |
DOJ | 81,648,861 | 70,151,316 | 12,888,000 | 7,085,000 | |
EPA | 48,438,112 | 38,576,770 | 91,426,293 | 55,939,893 | |
Treasury | 36,844,300 | 34,290,172 | 46,845,794 | 45,765,778 | |
Transportation | 7,594,717 | 6,330,890 | 25,879,018 | 19,266,197 | |
DHS | 5,644,000 | 1,380,000 | 32,791,000 | 30,529,000 | |
USDA | 3,737,000 | 3,594,000 | 1,703,410,000 | 1,729,092,000 | |
SBA | 2,080,000 | 2,080,000 | 9,313,985 | 860,229 | |
Interior | 255,000 | 232,000 | 4,542,000 | 97,000 | |
DOC | 10,000 | 0 | 55,578,000 | 23,580,000 | |
CNCS | 0 | 0 | 0 | 0 | |
Education | 0 | 0 | 450,790,000 | 132,390,000 | |
OPM | 0 | 0 | 2,401,580,000 | 2,180,390,000 | |
RRB | 0 | 0 | 447,100,000 | 387,000,000 | |
Total | 121,430,941,796 | 37,807,040,850 | 81,759,833,830 | 65,307,630,800 |
Sources: AFRs and PARs for FY
Endnotes
[1] This is the Project On Government Oversight’s third report examining the realm of improper payments. Our first report introduced the basics of this topic, while our second report examined data matching restrictions preventing inspectors general and agencies from sharing digital information and inhibiting improper payment identification, prevention, and recovery. Project On Government Oversight, Federal Improper Payments Are Significant, Costing Taxpayers Billions, July 12, 2016. Project On Government Oversight, Research on Data Matching Barriers, October 15, 2016.
[2] Pub. L. 107-300.
[3] Pub. L. 111-204; Pub. L. 112-248; Memorandum from Shaun Donovan, Director of the Office of Management and Budget, to the Heads of Executive Departments and Agencies, regarding “Appendix C to Circular No. A-123, Requirements for Effective Estimation and Remediation of Improper Payments,” October 20, 2014. (Downloaded March 23, 2016) (Hereinafter Requirements for Effective Estimation and Remediation of Improper Payments)
[4] Requirements for Effective Estimation and Remediation of Improper Payments, PDF p. 33, Report p. 30.
[5] Requirements for Effective Estimation and Remediation of Improper Payments, PDF pp. 31-41, Report pp. 28-38.
[6] Requirements for Effective Estimation and Remediation of Improper Payments, PDF p. 32, Report p. 29.
[7] Requirements for Effective Estimation and Remediation of Improper Payments, PDF p. 31, Report p. 28.
[8] Requirements for Effective Estimation and Remediation of Improper Payments, PDF p. 32, Report p. 29.
[9] Requirements for Effective Estimation and Remediation of Improper Payments, PDF p. 32, Report p. 29.
[10] Requirements for Effective Estimation and Remediation of Improper Payments, PDF p. 32, Report p. 29.
[11] Requirements for Effective Estimation and Remediation of Improper Payments, PDF p. 35, Report p. 32.
[12] Requirements for Effective Estimation and Remediation of Improper Payments, PDF p. 32, Report p. 29.
[13] Requirements for Effective Estimation and Remediation of Improper Payments, PDF p. 32, Report p. 29.
[14] Non-Federal entities is used here as defined in the Uniform Guidance, 2 CFR Subpart A, Section 200.69.
[15] Requirements for Effective Estimation and Remediation of Improper Payments, PDF p. 36, Report p. 33.
[16] Requirements for Effective Estimation and Remediation of Improper Payments, PDF p. 36, Report p. 33.
[17] Requirements for Effective Estimation and Remediation of Improper Payments, PDF p. 34, Report p. 31.
[18] Requirements for Effective Estimation and Remediation of Improper Payments, PDF p. 38, Report p. 35.
[19] Mandatory funds are expenditures that are provided for and required by a law, excluding appropriation acts. Mandatory fund accounts hold and disburse mandatory funds. Discretionary funds are expenditures that are provided for and controlled by appropriation acts. Discretionary fund accounts hold and disburse discretionary funds. Special fund accounts are fund accounts created and designated for a specific purpose by a law, such as acting as a receipt for specific expenditures. Government Accountability Office, A Glossary of Terms Used in the Federal Budget Process, September 2005. (Downloaded July 16, 2016); Requirements for Effective Estimation and Remediation of Improper Payments, PDF p. 39, Report p. 36.
[20] Pub. L. 111-204, Section 2(h)(3)(D); 124 STAT. 2229-2230; Congressional Research Service, Improper Payments and Recovery Audits: Legislation, Implementation, and Analysis, October 18, 2013, PDF p. 8, Report p. 5. (Downloaded September 27, 2016)
[21] Pub. L. 111-204, Section 2(h)(3)(D); 124 STAT. 2230.
[22] It is important to note that the improper payment “estimates” most commonly referred to by the media are not the same as the improper payment “identifications” for PRAs. Agencies develop improper payment estimates by evaluating a selection of payments in a program or activity to determine if the payments were improper, and then extrapolate those results to determine the program’s or activity’s annual improper payment amount and rate. PRA improper payment identifications are not statistical extrapolations, but instead are specific payments found as a result of a targeted examinations of specific types of high-risk payments that the agency determines are most likely to have improper payments and able to be cost-effectively recaptured. Requirements for Effective Estimation and Remediation of Improper Payments, PDF p. 41, Report p. 38.
[23] Project On Government Oversight, Federal Improper Payments Are Significant, Costing Taxpayers Billions, July 12, 2016.
[24] Brookings Center for Effective Public Management, Sometimes cutting budgets raise deficits: The curious case of inspectors’ general return on investment, April 2015, PDF p. 6, Report p. 6. (Downloaded August 1, 2016)
[25] Office of Management and Budget, PaymentAccuracy.
[26] House Committee on Oversight, Subcommittee on Government Operations, Examining Billion Dollar Waste Through Improper Payments, September 22, 2016. (Downloaded September 30, 2016)
Nicholas Pacifico 
Oversight in your inbox
Sent Saturdays
