This is the accessible text file for GAO report number GAO-11-724 
entitled 'Disability Insurance: SSA Can Improve Efforts to Detect, 
Prevent, and Recover Overpayments' which was released on July 27, 2011. 

This text file was formatted by the U.S. Government Accountability 
Office (GAO) to be accessible to users with visual impairments, as 
part of a longer term project to improve GAO products' accessibility. 
Every attempt has been made to maintain the structural and data 
integrity of the original printed product. Accessibility features, 
such as text descriptions of tables, consecutively numbered footnotes 
placed at the end of the file, and the text of agency comment letters, 
are provided but may not exactly duplicate the presentation or format 
of the printed version. The portable document format (PDF) file is an 
exact electronic replica of the printed version. We welcome your 
feedback. Please E-mail your comments regarding the contents or 
accessibility features of this document to Webmaster@gao.gov. 

This is a work of the U.S. government and is not subject to copyright 
protection in the United States. It may be reproduced and distributed 
in its entirety without further permission from GAO. Because this work 
may contain copyrighted images or other material, permission from the 
copyright holder may be necessary if you wish to reproduce this 
material separately. 

United States Government Accountability Office: 
GAO: 

Report to the Subcommittee on Social Security, Committee on Ways and 
Means, House of Representatives: 

July 2011: 

Disability Insurance: 

SSA Can Improve Efforts to Detect, Prevent, and Recover Overpayments: 

GAO-11-724: 

GAO Highlights: 

Highlights of GAO-11-724, a report to the Subcommittee on Social 
Security, Committee on Ways and Means, House of Representatives. 

Why GAO Did This Study: 

The Social Security Administration’s (SSA) Disability Insurance (DI) 
program paid almost $123 billion in benefits in fiscal year 2010 to 
more than 10 million workers and dependents. The program is poised to 
grow further as the baby boom generation ages. GAO examined (1) what 
is known about the extent to which SSA makes overpayments to, and 
recovers overpayments from, DI beneficiaries who exceed program 
earnings guidelines, and (2) potential DI program vulnerabilities that 
may contribute to overpayments to beneficiaries who have returned to 
work. To answer these questions, GAO reviewed work continuing 
disability review (work CDR) policies and procedures, interviewed SSA 
headquarters and processing center officials, visited 4 of 8 
processing centers, and reviewed a random nongeneralizable sample of 
60 CDR case files across those 4 centers (15 from each). 

What GAO Found: 

Disability Insurance overpayments detected by SSA increased from about 
$860 million in fiscal year 2001 to about $1.4 billion in fiscal year 
2010. SSA estimates about 72 percent of all projected DI overpayments 
were work-related during fiscal years 2005 through 2009. While the 
agency collected, or recovered, $839 million in overpayments in fiscal 
year 2010, monies still owed by beneficiaries grew by $225 million 
that same year, and cumulative DI overpayment debt reached $5.4 
billion. SSA does not have agencywide performance goals for debt 
collection—for example, the percent of outstanding debt collected 
annually. And while SSA has a policy for full repayment within 3 
years, 19 of the 60 work CDR cases GAO reviewed had repayment plans 
exceeding 3 years. SSA officials said that lengthy repayment plans are 
often the result of an individual’s limited income, but SSA does not 
review or approve repayment plans which exceed agency policy. During 
the course of the review, GAO also found a limitation in SSA’s 
Recovery of Overpayments, Accounting and Reporting (ROAR) system. Used 
to track overpayments and collections, ROAR does not reflect debt due 
SSA past year 2049, so the total balance due the program is unknown 
and likely larger than the agency is reporting. SSA officials 
acknowledged this issue, but are unable to determine the extent of the 
problem at this time. They told GAO they have a work group which will 
recommend action to correct the problem. But until this issue is 
addressed, SSA officials said that the agency can only track and 
report on overpayments scheduled to be repaid through 2049. The amount 
owed after that year is unreflected in current totals even as it 
annually increases. SSA officials reported that the agency has ongoing 
initiatives to enhance debt collection. 

SSA has numerous policies and processes in place to perform work CDRs, 
though two key weaknesses have hindered SSA’s ability to identify and 
review beneficiary earnings which affect eligibility for DI benefits. 
First, SSA lacks timely earnings data on beneficiaries who return to 
work. In 49 of the 60 CDR cases GAO reviewed, there was no evidence in 
the file that the beneficiary reported his or her earnings, as 
required by program guidelines. To identify unreported work and 
earnings, SSA primarily relies on data matching with the Internal 
Revenue Service (IRS), then sends these matches to staff for a work 
CDR. However, the IRS data may be more than a year old when received 
by SSA, and SSA says it is not cost-effective to gain access to and 
use other sources of earnings information, such as the National 
Directory of New Hires database. In addition, GAO found that cases may 
wait up to 15 additional months before SSA staff begin work on the 
CDRs. Second, SSA lacks formal, agencywide performance goals for work 
CDRs. While it targets 270 days to complete a case, actual processing 
time ranged from 82 to 992 days (with a median of 396 days) in the 60 
cases GAO reviewed, and overpayments which accrued as a result topped 
$1 million total. SSA officials reported several initiatives to more 
effectively prioritize work CDR cases—for example, those with the 
largest potential overpayment amounts—but these efforts are in the 
early stages, and GAO could not yet assess their effectiveness as part 
of this review. 

What GAO Recommends: 

GAO recommends that SSA develop and adopt agencywide performance goals 
for recovering DI overpayments and processing work CDRs, require 
supervisory review of certain repayment plans, address a system 
limitation which precludes an accurate record of debt owed SSA, and 
explore options for obtaining more timely earnings information. SSA 
agreed with four of five recommendations. It disagreed with the need 
for supervisory review of repayment plans while acknowledging the need 
for more related guidance to its staff. 

View [hyperlink, http://www.gao.gov/products/GAO-11-724] or key 
components. For more information, contact Daniel Bertoni at (202) 512-
7215 or bertonid@gao.gov. 

[End of section] 

Contents: 

Letter: 

Background: 

Most DI Overpayments Are Work-Related, and Their Recovery Can Take 
Decades: 

Lack of Timely Earnings Data and Inconsistent Processing of Work CDRs 
Allow Overpayments to Accrue: 

Conclusions: 

Recommendations: 

Agency Comments and Our Evaluation: 

Appendix I: Enforcement Cumulative Totals and Distribution for IRS 
Enforcement Matches Conducted in 2008-2010 (Data for Earnings Years 
2007 through 2009): 

Appendix II: Financial Data on DI Overpayments and Overpayment Debt: 

Appendix III: Comments from the Social Security Administration: 

Appendix IV: GAO Contact and Staff Acknowledgments: 

Related GAO Products: 

Tables: 

Table 1: Enforcement Cumulative Totals and Distribution for Earnings 
Year 2009 Data: 

Table 2: Enforcement Cumulative Totals and Distribution for Earnings 
Year 2008 Data: 

Table 3: Enforcement Cumulative Totals and Distribution for Earnings 
Year 2007 Data: 

Figures: 

Figure 1: SSA's Enforcement Work CDR Process: 

Figure 2: SSA's Debt Recovery Process: 

Figure 3: DI Overpayment Debt Annually Detected by SSA, Fiscal Years 
2001-2010: 

Figure 4: Cumulative DI Overpayment Debt, Fiscal Years 2001-2010: 

Figure 5: Projected Years Needed for Repayment in Full for 42 Cases 
with Initial Plans: 

Figure 6: Age of Earnings Data Provided to SSA by IRS Earnings Alerts, 
of 60 Cases Reviewed: 

Figure 7: Time Pending Development after IRS Alert Delivered, of 60 
Cases Reviewed: 

Figure 8: Number of Months Overpayments Accrued as CDR Development Was 
Pending, of 60 Cases Reviewed: 

Figure 9: Variance in Total Case Processing Time across Four 
Processing Centers Visited, of 60 Cases Reviewed: 

Abbreviations: 

AERO: Automated Earnings Reappraisal Operations: 

CDR: continuing disability review: 

DI: Disability Insurance: 

enforcement operation: Continuing Disability Review Enforcement 
Operation: 

EPE: extended period of eligibility: 

IRS: Internal Revenue Service: 

MBR: Master Beneficiary Record: 

NDNH: National Directory of New Hires: 

OMB: Office of Management and Budget: 

POMS: Program Operations Manual System: 

ROAR: Recovery of Overpayments, Accounting and Reporting System: 

SSA: Social Security Administration: 

SGA: substantial gainful activity: 

SSI: Supplemental Security Income: 

Treasury: Department of the Treasury: 

TWP: trial work period: 

[End of section] 

United States Government Accountability Office: 
Washington, DC 20548: 

July 27, 2011: 

The Honorable Sam Johnson: 
Chairman: 
The Honorable Xavier Becerra: 
Ranking Member: 
Subcommittee on Social Security: 
Committee on Ways and Means: 
House of Representatives: 

The Social Security Administration's (SSA) Disability Insurance (DI) 
program is the nation's largest cash assistance program for workers 
with disabilities. The program provides cash benefits to workers who 
are blind or disabled and contributed to the DI Trust Fund as workers. 
In fiscal year 2010, the DI program paid about $123 billion in 
benefits to more than 10 million workers with disabilities and their 
dependents. The program has grown substantially in recent years and is 
poised to grow further as the baby-boom generation ages. Most 
importantly, the long-term solvency of the DI Trust Fund is currently 
jeopardized, and the fund is projected to be exhausted in 2018. 
[Footnote 1] 

SSA guidelines allow DI beneficiaries to work and earn up to $1,000 
per month for a limited period of time without affecting their 
benefits--a level of earnings called substantial gainful activity 
(SGA).[Footnote 2] After completing a 9-month "trial work period," 
beneficiaries who earn more than SGA are generally no longer entitled 
to benefits, and may be overpaid if SSA does not stop their benefits 
in a timely manner.[Footnote 3] To verify an individual's ongoing 
eligibility for DI benefits, SSA is required to periodically evaluate 
the beneficiary's impairments to determine whether the beneficiary 
remains entitled to disability benefits. This evaluation process is 
called a continuing disability review (CDR).[Footnote 4] If the review 
establishes that the beneficiary has substantial earnings, SSA will 
investigate the work activity to determine if the disability continues 
or ceases. The reviews of beneficiary earnings are referred to as work 
CDRs. These reviews typically involve SSA staff querying centralized 
agency data systems to identify earnings, sending forms to a 
beneficiary requesting information about work activity and earnings 
that may affect eligibility for DI benefits, contacting employers to 
verify earnings amounts, and assessing other factors such as employer 
subsidies and impairment-related work expenses. Most work CDRs are 
generated by SSA's Continuing Disability Review Enforcement Operation 
(enforcement operation).[Footnote 5] Enforcement operation staff use 
periodic computer matches between SSA's administrative data and 
Internal Revenue Service (IRS) earnings data to identify earnings, 
then send these cases to one of eight regional processing centers for 
review by SSA staff. 

If SSA does not obtain timely and accurate earnings information, or 
fails to act expeditiously to cease benefits to those no longer 
eligible, overpayments can accrue over several years and become very 
large--adding up to tens of thousands of dollars. Overpayments 
adversely affect program integrity, but can also create economic 
hardship for beneficiaries who have to repay them once the overpayment 
is detected.[Footnote 6] In addition, the prospect of having to repay 
an overpayment may be a disincentive for some beneficiaries to return 
to work, which runs counter to SSA's goal of helping beneficiaries 
become self-sufficient.[Footnote 7] 

Given the importance of identifying and collecting overpayments to DI 
beneficiaries with earnings in a timely manner, the Chairman and 
Ranking Member of the Subcommittee on Social Security, House Committee 
on Ways and Means, asked us to review SSA's enforcement work CDRs. We 
answered the following questions: (1) What is known about the extent 
to which SSA makes work-related overpayments to, and recovers 
overpayments from, DI beneficiaries? (2) What are SSA's policies and 
procedures for performing enforcement work CDRs, including potential 
DI program vulnerabilities that may contribute to work-related 
overpayments? 

In conducting our review we analyzed DI overpayment debt collection 
and enforcement work CDR performance data, policies and procedures for 
conducting work CDRs, prior reports by SSA and its Office of the 
Inspector General, external research studies, relevant federal laws 
and regulations, and our prior reviews of the program. In addition, we 
interviewed SSA officials responsible for DI overpayment debt 
collection policy and operations, as well as enforcement work CDR 
processing management, policy, and oversight. We randomly selected 15 
cases from each of four processing centers we visited (Baltimore, 
Maryland; Chicago, Illinois; Kansas City, Missouri; and Queens, New 
York), which were closed in fiscal year 2009 with an overpayment. 
Together, the selected processing centers received almost 80 percent 
of SSA's enforcement alerts referred for work CDRs in fiscal year 
2009. We reviewed each of these 60 randomly selected cases to 
determine whether the case had been administered in accordance with 
SSA program guidelines for processing of work CDRs. We used random 
selection procedures to help ensure that we drew a wide range of cases 
for our review; however, the results from our randomly selected sample 
of cases cannot be generalized to the population of all work CDR cases 
due to our limited sample sizes. Finally, we conducted in-depth 
interviews with SSA management and line staff at headquarters and four 
of SSA's eight processing centers. In our site visits and interviews, 
we examined work CDR and overpayment debt recovery procedures; 
elicited management and staff views on the effectiveness of the work 
CDR processes in detecting and preventing earnings-related 
overpayments; and discussed the processing of selected work CDR cases, 
including debt recovery cases, and potential improvements to existing 
policies, processes, and systems. We assessed the reliability of all 
databases used in our review, primarily SSA's Disability Control File, 
Master Beneficiary Record (MBR), and Recovery of Overpayment, 
Accounting, and Reporting (ROAR) system. While we identified a ROAR 
system limitation that is discussed in the report, we found the 
databases to be sufficiently reliable for the purposes of this report. 

We conducted this performance audit from March 2010 to July 2011 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. 

Background: 

The DI program was established in 1956 to provide monthly cash 
benefits to individuals unable to work because of severe long-term 
disability.[Footnote 8] In fiscal year 2010, the program's average 
monthly benefit was about $922.[Footnote 9] To be eligible for 
benefits, workers with disabilities must have a specified number of 
recent work credits under Social Security when they acquired a 
disability.[Footnote 10] Individuals may also be able to qualify based 
on the work record of a deceased or retired parent with a disability 
or a deceased spouse.[Footnote 11] Benefits are financed by payroll 
taxes paid into the DI Trust Fund by covered workers and their 
employers and are based on a worker's earnings history. To meet the 
definition of disability under the DI program, an individual must have 
a medically determinable physical or mental impairment that (1) has 
lasted or is expected to last at least 1 year or to result in death 
and (2) prevents the individual from engaging in SGA.[Footnote 12] 
Individuals are engaged in SGA if they have earnings that average more 
than $1,000 per month in calendar year 2010, after applying any work 
incentives.[Footnote 13] Program guidelines direct DI beneficiaries to 
report their earnings to SSA in a timely manner to ensure they are 
still eligible for benefits and do not incur an overpayment. 

SSA has several programs that are designed to assist beneficiaries in 
returning to work. For example, the Ticket to Work and Work Incentives 
Improvement Act of 1999[Footnote 14] provided for the establishment of 
the Ticket to Work and Self-Sufficiency Program to provide eligible DI 
and Supplemental Security Income (SSI) beneficiaries with employment 
services, vocational rehabilitation services, or other support 
services to help them obtain and retain employment and reduce their 
dependency on benefits. SSA provides each eligible beneficiary with a 
ticket to obtain services from SSA-approved public or private 
providers, referred to as employment networks, or from traditional 
state vocational rehabilitation agencies.[Footnote 15] 

SSA conducts work CDRs to determine if beneficiaries are still 
eligible or are working above the SGA level.[Footnote 16] Work CDRs 
are event-driven and initiated when there is an indication of work 
activity. While work CDRs can be prompted by several events, most are 
generated by SSA's enforcement operation. This process involves 
periodic data matches between SSA's MBR database and IRS earnings 
data. The enforcement operation generates alerts for cases that exceed 
specified earnings thresholds,[Footnote 17] which are then forwarded 
to one of eight processing centers for additional development by 
staff.[Footnote 18] In fiscal year 2010, the enforcement operation 
identified approximately 2 million records of which more than 531,000 
were sent to SSA's processing centers and field offices for review. 
[Footnote 19] Appendix I provides detailed information on the results 
of the enforcement operation for fiscal years 2008 to 2010, which used 
IRS earnings data from 2007 to 2009 (enforcement matches are typically 
conducted using the prior year's earnings data). 

Work CDRs can also be triggered by other events, such as beneficiaries 
reporting their earnings to SSA. For example, SSA requires 
beneficiaries to undergo periodic medical reviews called medical 
continuing disability reviews, or medical CDRs, to assess whether they 
continue to have a disabling impairment.[Footnote 20] During such 
reviews, the disability examiner sometimes discovers evidence that a 
beneficiary is working and forwards the case to an SSA field office or 
processing center for earnings or work development. Third-party 
reports from state vocational rehabilitation agencies, federal 
agencies, or anonymous individuals may also trigger a work CDR. 
Finally, some DI beneficiaries report their earnings to SSA as 
directed under program guidelines by visiting an SSA field office or 
calling the agency's 800 number. 

While most work CDRs are initially sent to processing centers as a 
result of action by SSA's enforcement operation, some of these cases 
are later referred to one of SSA's more than 1,300 field offices. 
Field offices can be asked to assist processing centers in the 
development of cases when obtaining local information about 
beneficiaries or their employers would expedite case processing. Field 
offices also tend to be the focal points for work CDRs generated by 
medical CDRs, third-party reporting, and beneficiary self-reporting. 
Work CDRs entail processes that can be both labor-intensive and time-
consuming. For each case, SSA staff must review electronic case files 
in SSA's eWork[Footnote 21] and associated data systems, conduct 
interviews, and contact beneficiaries and their employers to verify 
earnings and any applicable work incentives, such as subsidies or 
impairment-related work expenses.[Footnote 22] After the initial 
review of cases indicating a cessation of benefits, a "disability 
processing specialist" or "disability examiner" determines whether 
benefits should be discontinued and an overpayment assessed.[Footnote 
23] (Figure 1 provides an overview of the work CDR process.) 

Figure 1: SSA's Enforcement Work CDR Process: 

[Refer to PDF for image: illustration] 

1) IRS data matching in February, May, and August. 

Do the data meet screening criteria? 
Yes: go to #3; 
No: go to #2. 

2) No action taken and no contact made with beneficiary. 

3) Remaining matches distributed by Social Security number and age to 
one of eight processing centers. 

Does processing center screening identify work issues? 
Yes: go to #4; 
No: go to #2. 

4) Work continuing disability review (CDR) assigned to SSA staff. 

5) Earnings development: As needed, SSA staff will: 
* Check SSA records; 
* Send forms to beneficiary; 
* Send forms to employer(s); 
* Check Work Number; 
* Check National Directory of New Hires; 
* Contact bank for up to date beneficiary address. 

Is the work determination outcome ‘cessation’? 
Yes: go to #7; 
No (and beneficiary not yet contacted): go to #2. 
No (and beneficiary has been contacted): go to #6. 

6) Beneficiary notified that benefits will continue. 

7) Processing center sends a proposed cessation decision to 
beneficiary: 
Not contested: go to #8; 
Contested: go to #9. 

8) Final cessation notice informs beneficiary of overpayment. 
Not contested: go to #10; 
Contested: go to #9. 

9) New evidence is considered. 
Decision upheld or modified: go to #10; 
Decision reversed: go to #6. 

10) Overpayment debt recovery begins. 

Source: GAO analysis of SSA procedural guidance. 

[End of figure] 

When a DI work-related overpayment is identified, the beneficiary is 
notified of the overpayment and may request reconsideration or waiver 
of that overpayment.[Footnote 24] SSA may grant a waiver request if 
the agency finds the beneficiary was not at fault and recovery or 
adjustment would either defeat the purpose of the program or be 
against equity and good conscience, as defined by SSA.[Footnote 25] If 
SSA denies a reconsideration or waiver request, full repayment is 
requested. If the beneficiary is receiving DI or certain other SSA 
benefits, SSA may withhold partial payment of these benefits to 
recover the debt.[Footnote 26] However, if no SSA benefits are being 
received, or if the beneficiary asserts that the proposed withholding 
amount is too large, the agency generally requests repayment over 12 
to 36 months. SSA policy requires a minimum monthly payment of $10 
dollars. SSA may also attempt to recover payments due from the 
individual's estate or subsequent survivor's benefits. (Figure 2 
provides an overview of SSA's debt recovery system.) The agency uses 
the ROAR system to track DI overpayments and collections. 

Figure 2: SSA's Debt Recovery Process: 

[Refer to PDF for image: illustration] 

Disability Insurance overpayment enters Debt Management System: 

Debtor may request a “reconsideration” of the overpayment itself 
within 60 days or a waiver of all or part of the debt at any time. 

SSA's interaction with beneficiary: 

1) Will debtor work with SSA to repay the overpayment?
No (or ignored): go to #7; 
Yes: go to #2. 

2) Will debtor repay in full now? 
Yes: debt collected; 
No: go to #4. 

3) Will debtor repay in full over the next 12 months? 
Yes: go to #6; 
No: go to #4. 

4) Will debtor repay in full over the next 36 months? 
Yes: go to #6; 
No: go to #5. 

5) Will debtor repay at least $10 a month over a longer period? 
Yes: go to #6; 
No: go to #7. 

6) Debt collected. If payments stop for 45 days, with no installment 
agreement in the last 30 days, SSA may refer the debt for external 
collection. Collection efforts don’t begin, however, until the debtor 
is notified and given a 60-day due process period.[A] 

7) If SSA cannot reach an agreement with the debtor, or if the debtor 
becomes delinquent, the remaining debt is referred for external 
collection, which can include: 
* tax refund offset; 
* administrative offset; 
* federal salary offset; 
* administrative wage garnishment; 
* referral to credit bureaus. 

Source: GAO analysis of SSA policies. 

[End of figure] 

When a debtor is no longer receiving benefits, SSA can also recover 
debt through several external collection tools. When a beneficiary is 
not repaying as agreed in the repayment plan, SSA terminates its 
collection activity and, after a due process period, the debt is 
referred for external collection. 

External debt collection tools include: 

* tax refund offset, which withholds or reduces federal tax refunds; 

* federal salary offset, which withholds or reduces wages and payments 
to a federal employee; 

* administrative offset (against other than SSA benefits), which 
withholds or reduces federal payments other than tax refunds or salary; 

* administrative wage garnishment, which garnishes[Footnote 27] wages 
and payments paid by private employers or state and local governments; 
and: 

* credit bureau referral, which refers delinquent accounts to credit 
bureaus.[Footnote 28] 

Once a debt is referred to external collection, the debtor remains 
subject to offsets until the debt is repaid in full or other 
resolution of the case. 

Most DI Overpayments Are Work-Related, and Their Recovery Can Take 
Decades: 

Nearly Three-Quarters of All DI Overpayments Are to Beneficiaries Who 
Have Returned to Work: 

Medical and work-related overpayments in the DI program detected by 
SSA grew from about $860 million in fiscal year 2001 to about $1.4 
billion in fiscal year 2010,[Footnote 29] and though the true extent 
of overpayments due to earnings is currently unknown, our review 
suggests that most of them are related to beneficiaries who work above 
SGA while receiving benefits. (Figure 3 shows the increase in 
overpayments detected by SSA in recent years.) SSA officials estimate 
that from fiscal years 2005 through 2009, about 72 percent[Footnote 
30] of all projected DI overpayments were work-related, meaning the 
overpayments were to beneficiaries who returned to work and were no 
longer eligible. This figure is higher than cited in past years by 
SSA. SSA officials attribute the increase in the percentage of 
overpayments that are work-related during this period to improved 
detection by its enforcement operation, and to changes in how the 
agency estimates the overpayment numbers. Agency officials also 
explained that approximately half of the increase in overpayment 
dollars during this period may be due to the increase in DI program 
benefit levels. 

Figure 3: DI Overpayment Debt Annually Detected by SSA, Fiscal Years 
2001-2010: 

[Refer to PDF for image: vertical bar graph] 

Fiscal year: 2001; 
DI Overpayment Debt: $859.3 million. 

Fiscal year: 2002; 
DI Overpayment Debt: $886.6 million. 

Fiscal year: 2003; 
DI Overpayment Debt: $989.6 million. 

Fiscal year: 2004; 
DI Overpayment Debt: $927.2 million. 

Fiscal year: 2005; 
DI Overpayment Debt: $1.353 billion. 

Fiscal year: 2006; 
DI Overpayment Debt: $1.617 billion. 

Fiscal year: 2007; 
DI Overpayment Debt: $1.325 billion. 

Fiscal year: 2008; 
DI Overpayment Debt: $1.384 billion. 

Fiscal year: 2009; 
DI Overpayment Debt: $1.399 billion. 

Fiscal year: 2010; 
DI Overpayment Debt: $1.39 billion. 

Source: SSA. 

Notes: The DI overpayment debt for fiscal years 2001-2010 is 
understated in this figure due to a ROAR system limitation. SSA 
officials are aware of the system limitation and are working to 
address the issue. 

[End of figure] 

Debt figures are not adjusted for inflation. ASSA officials attributed 
the jump in overpayment detections in fiscal years 2005 and 2006 to 
the agency's implementation of eWork in 2004. The eWork system is an 
automated program used to develop, control, and adjudicate work CDRs. 
According to SSA officials, eWork helped the agency to process work 
CDRs faster and more efficiently, resulting in an increase in the 
number of work CDRs processed and an increase in the level of 
overpayments detected in those years. 

We found that detected overpayments could be even larger than SSA's 
data reflect because some overpayments have been accidentally removed 
from SSA records due to manual processing errors.[Footnote 31] In our 
review of 60 work CDR cases, we found two manual processing errors 
which resulted in overpayments totaling $53,097 being removed from 
agency records. In one case, staff entered a code to correct an 
overpayment amount but instead deleted the overpayment entirely. As a 
result of our detection, SSA officials reentered the overpayment debts 
into the system and indicated they would proceed with debtor 
notification and recovery. Because the results of our case review are 
not generalizable, the incidence of such occurrences is currently 
unknown and thus the potential impact on total DI overpayments owed by 
ineligible beneficiaries is not clear. SSA officials said that they do 
not have a mechanism for detecting, or a process of supervisory review 
to catch, such processing errors. 

In the 60 cases we reviewed, we also found that individual overpayment 
amounts varied widely, ranging from $1,126 to $53,436, with a median 
of $16,917 per individual. The size of individual overpayments can 
also be affected by SSA's Automated Earnings Reappraisal Operations 
(AERO), which is a computer program that periodically screens a 
beneficiary's earnings record for changes, and uses that information 
to adjust, as needed, the monthly benefit amount.[Footnote 32] Of the 
60 cases, we found 54 in which AERO increases in benefit amount 
occurred even while SSA conducted a work CDR to determine if the 
beneficiary was still eligible for benefits at all. In addition, an 
individual overpayment can result in additional overpayments to 
dependents (family members who receive benefits). In 8 of the 60 
cases, additional overpayments occurred because the primary 
beneficiary had dependents. For example, in one case, the overpayments 
to two dependent children added $6,580 to the primary beneficiary's 
existing overpayment debt of $17,102, for a total of $23,682 owed. 

A beneficiary's total DI overpayment debt can also increase because of 
multiple periods of employment. DI beneficiaries may reenter and leave 
the workforce based on their ability to perform SGA. As a result, a 
beneficiary could be subject to multiple periods of DI overpayments if 
he or she does not report increased earnings to SSA in a timely 
manner, as regulations instruct. In 49 of the 60 cases we randomly 
selected for review, there was no indication in the file that the 
individual had reported his or her earnings to SSA, and in 15 of the 
60, SSA had detected two or more separate periods of earnings which 
resulted in overpayments. In one of these cases, the ineligible 
beneficiary owed SSA for multiple overpayments totaling $69,976. 

SSA Lacks Agencywide Performance Goals for DI Debt Recovery and Debt 
Continues to Accrue: 

SSA does not currently have formal, agencywide performance goals for 
debt recovery. Specifically, the agency does not have goals for the 
percentage of DI overpayment debt recovered within the 36-month time 
frame as required by its own policy. Under the Government Performance 
Results Act of 1993, federal agencies are required to establish 
performance goals to define level of performance and establish 
performance indicators to be used in measuring relevant outputs, 
service levels, and outcomes for each program activity.[Footnote 33] 
Although SSA's policy manual, the Program Operations Manual System 
(POMS), requires staff to ask for full repayment within 36 
months,[Footnote 34] the agency has not made this time frame a 
performance goal. SSA officials said they are currently working to 
develop debt recovery goals. In the meantime, without agencywide 
performance goals for debt recovery, SSA cannot adequately assess its 
performance or fully leverage and target its resources to recover 
overpayments from ineligible beneficiaries and reduce the total owed 
to SSA. Despite an increase in DI debt collections--$340 million to 
$839 million from fiscal year 2001 through fiscal year 2010--
outstanding DI debt grew from $2.5 billion to $5.4 billion during this 
time, including a $225 million increase in fiscal year 2010.[Footnote 
35] (Figure 4 shows the growth in cumulative overpayment debt in 
recent years.) 

Figure 4: Cumulative DI Overpayment Debt, Fiscal Years 2001-2010: 

[Refer to PDF for image: line graph] 

Cumulative DI overpayment debt (Outstanding debt balance at the end of 
the fiscal year): 

Fiscal year: 2001; 
Cumulative DI Overpayment Debt: $2.457 billion. 

Fiscal year: 2002; 
Cumulative DI Overpayment Debt: $2.687 billion. 

Fiscal year: 2003; 
Cumulative DI Overpayment Debt: $2.961 billion. 

Fiscal year: 2004; 
Cumulative DI Overpayment Debt: $3.161 billion. 

Fiscal year: 2005; 
Cumulative DI Overpayment Debt: $3.755 billion. 

Fiscal year: 2006; 
Cumulative DI Overpayment Debt: $4.336 billion. 

Fiscal year: 2007; 
Cumulative DI Overpayment Debt: $4.699 billion. 

Fiscal year: 2008; 
Cumulative DI Overpayment Debt: $4.982 billion. 

Fiscal year: 2009; 
Cumulative DI Overpayment Debt: $5.188 billion. 

Fiscal year: 2010; 
Cumulative DI Overpayment Debt: $5.413 billion. 

Source: SSA. 

Note: The cumulative DI overpayment debt for fiscal years 2001 to 2010 
is understated in this figure due to a limitation in one of SSA's data 
systems. SSA officials are aware of the system limitation and are 
working to address the issue. 

[End of figure] 

Cumulative overpayment debt is comprised of existing debt carried 
forward from prior years, new debt, and reestablished debts (debts 
reactivated for collection due to re-entitlement or other event). 
Write-offs, including waivers and terminated collections, are not 
included in cumulative overpayment debt. Write-offs generally 
represent money the agency will never recover because it has either 
waived an overpayment or has terminated collection activities on the 
overpayment.[Footnote 36] Write-offs in the DI program totaled about 
$4 billion over the 10-year period, including about $460 million in 
fiscal year 2010. (See appendix II for more specific information on 
overpayment debt in fiscal years 2001 to 2010.) In the 60 cases we 
reviewed, 20 contained a waiver request from the beneficiary. SSA 
approved 2 of these requests for a total of $36,413 waived, and denied 
14 requests. Four other waiver requests were pending a decision at the 
time of our review. SSA may waive an overpayment if the agency 
determines that an individual is not at fault for the overpayment and 
recovery would defeat the purpose of the program or be against equity 
and good conscience. 

Most overpayment debt is collected by SSA through partial benefit 
withholding or the withholding of future DI benefits for which a 
beneficiary is still eligible. SSA attributes 77 percent of the 
approximately $839 million of debt collected in fiscal year 2010 to 
withholding of DI benefits. The amount withheld from benefits to 
recoup previous overpayments may be negotiated with the debtor and 
based on a monthly amount the debtor can afford. The remainder of 
overpayment debt is collected in a variety of ways, including payments 
by the debtor and return of uncashed DI benefit checks; withholding of 
other SSA benefits, such as SSI;[Footnote 37] or through external 
collection including federal salary offset, administrative offset 
(other than against SSA benefits), tax refund offset, and 
administrative wage garnishment. SSA estimates that only about 11 
percent of collections is through external means. In our 60 cases, 5 
were referred for external collection at the time of our review, for a 
total owed of $79,950, but just $2,478 had been recovered through 
these methods.[Footnote 38] 

SSA Policy Does Not Require Supervisory Review of Repayment Plans, 
although Some Last Decades: 

SSA does not require supervisory review of repayment plans prior to 
approval, including those in which repayment periods exceed the 
recommended 36 months. The agency reported that in fiscal year 2010, 
the average time to collect a DI overpayment debt in full was 48 
months. However, in our review of 60 cases, we found that SSA agreed 
to some initial repayment plans which will take many decades. We 
analyzed the initial repayment plans established for individuals in 
these cases and found[Footnote 39] 42 of the 60 had a repayment plan 
in place, with a median repayment time for all 42 of approximately 34 
months. While SSA's POMS require that staff should seek full repayment 
within 36 months, SSA officials reported that no supervisory approval 
is needed to exceed the 36 months. Of the 42 cases with a repayment 
plan, 19 had initial plans requiring more than 36 months for repayment 
in full and 7 of these required 20 years or more. Repayment time 
frames for the 42 cases ranged from less than 1 year to nearly 223 
years for a case with a 60-year-old debtor who was paying $10 a month 
on $26,715 owed. (Figure 5 shows the years that it would take SSA to 
recover each overpayment based on the initial repayment plan 
established.) SSA officials told us they are often unable to increase 
monthly repayment amounts and thus shorten repayment time frames 
because of a debtor's limited income. For instance, in a case we 
reviewed with an initial repayment plan of 148 years for $44,465 in 
overpayments owed to SSA, SSA records show the individual earned less 
than $100 in 2010.[Footnote 40] As this case illustrates, SSA's policy 
allowing for extended repayment plans means that some of these plans 
will likely extend beyond the beneficiary's anticipated retirement age 
and total lifespan, thereby reducing the prospect of ultimate 
collection. 

Figure 5: Projected Years Needed for Repayment in Full for 42 Cases 
with Initial Plans: 

[Refer to PDF for image: plotted point graph] 

Repayment goal is 3 years. 

Size of debt: $1,126; 
Number of years to repay: 0.1. 

Size of debt: $3,842; 
Number of years to repay: 0.2. 

Size of debt: $2,620; 
Number of years to repay: 0.2. 

Size of debt: $3,218; 
Number of years to repay: 0.2. 

Size of debt: $3,299; 
Number of years to repay: 0.2. 

Size of debt: $8,005; 
Number of years to repay: 0.4. 

Size of debt: $2,677; 
Number of years to repay: 0.5. 

Size of debt: $7,372; 
Number of years to repay: 0.5. 

Size of debt: $4,116; 
Number of years to repay: 0.5. 

Size of debt: $6,660; 
Number of years to repay: 0.6. 

Size of debt: $10,957; 
Number of years to repay: 0.7. 

Size of debt: $20,641; 
Number of years to repay: 1. 

Size of debt: $9,976; 
Number of years to repay: 1. 

Size of debt: $3,202; 
Number of years to repay: 1. 

Size of debt: $11,320; 
Number of years to repay: 1.1. 

Size of debt: $4,898; 
Number of years to repay: 1.2. 

Size of debt: $11,935; 
Number of years to repay: 1.2. 

Size of debt: $15,911; 
Number of years to repay: 1.2. 

Size of debt: $17,010; 
Number of years to repay: 1.8. 

Size of debt: $29,089; 
Number of years to repay: 2.4. 

Size of debt: $47,606; 
Number of years to repay: 2.7. 

Size of debt: $38,734; 
Number of years to repay: 2.9. 

Size of debt: $16,904; 
Number of years to repay: 3. 

Size of debt: $5,134; 
Number of years to repay: 3.3. 

Size of debt: $1,933; 
Number of years to repay: 3.9. 

Size of debt: $3,096; 
Number of years to repay: 5.2. 

Size of debt: $12,482; 
Number of years to repay: 5.2. 

Size of debt: $42,301; 
Number of years to repay: 6.6. 

Size of debt: $8,442; 
Number of years to repay: 7.9. 

Size of debt: $19,843; 
Number of years to repay: 8.3. 

Size of debt: $5,062; 
Number of years to repay: 8.4. 

Size of debt: $7,392; 
Number of years to repay: 12.3. 

Size of debt: $15,429; 
Number of years to repay: 12.9. 

Size of debt: $17,509; 
Number of years to repay: 14.6. 

Size of debt: $22,830; 
Number of years to repay: 19. 

Size of debt: $22,582; 
Number of years to repay: 25.1. 

Size of debt: $16,938; 
Number of years to repay: 28.2. 

Size of debt: $21,539; 
Number of years to repay: 30. 

Size of debt: $52,749; 
Number of years to repay: 30. 

Size of debt: $10,314; 
Number of years to repay: 30. 

Size of debt: $44,465; 
Number of years to repay: 30. 

Size of debt: $26,715; 
Number of years to repay: 30. 

Source: GAO analysis of SSA data. 

[End of figure] 

In the course of analyzing repayment plans, we also identified a 
system limitation that could further impede SSA's ability to collect 
overpayment debt in future years. More specifically, we found that the 
ROAR system cannot capture and track overpayment debt scheduled to be 
collected beyond the year 2049. This ROAR system limitation stems from 
a program modification used to address the change of the century (Y2K) 
computer issue, and which extended the debt recovery date in the ROAR 
system from "1999" to "2049." Under existing SSA policies and 
procedures,[Footnote 41] SSA staff manually remove from the ROAR 
system the portion of any debt that cannot be collected before the 
year 2050, and create a reminder in the system to recover that balance 
beginning in the year 2050. However, because this is a manual process, 
the intended recovery action could be potentially missed by staff 
responsible for processing the cases in the future. As a result, the 
overpayment debt on the agency's books, and reported to the Department 
of the Treasury (Treasury) for the federal government's consolidated 
financial statements, is likely understated to an unknown extent. 
Equally important, we found that unless this problem is corrected, 
more overpayments will likely continue to be underreported as the 
years progress, and the impact on the DI Trust Fund may become more 
pronounced as time goes on. Overall, we found that 3 of the 60 cases 
reviewed had a total of $43,285 in overpayments removed from ROAR 
system records because collection of these payments will occur after 
the year 2049. However, because the results of our case review are not 
generalizable, we could not determine how many additional disability 
overpayment cases detected by SSA fell into this category. Since 
bringing this issue to their attention, SSA officials told us that the 
agency has begun to study this ROAR system limitation and an agency 
working group will recommend a course of action to correct the problem. 

SSA Initiatives Target the Recovery of Delinquent Debt: 

SSA officials reported that the agency is planning certain initiatives 
that could improve the recovery of overpayment debt. First, SSA 
officials told us that the agency is in the process of amending 
administrative offset and tax refund offset regulations to change the 
existing requirement regarding referrals of delinquent debt to 
Treasury. This change would allow SSA to refer debts that are 
delinquent for 10 years or more to Treasury for collection--something 
that it does not currently do. Second, Treasury has launched a pilot 
program providing for reciprocal offset of federal and state payments 
to individuals who are in debt to either federal or state programs. 
This program would allow a federal agency such as SSA to, for example, 
offset a beneficiary's state income tax refund to collect certain 
delinquent federal debts (such as DI program overpayments), while 
allowing for a reciprocal agreement with individual states. Four 
states--Kentucky, Maryland, New Jersey, and New York--are currently 
participating in the Treasury pilot program, and Treasury reports that 
it plans to expand the program to all states.[Footnote 42] On March 2, 
2011, SSA officials published proposed amendments to regulations that 
would allow SSA to utilize these collection tools.[Footnote 43] 

Lack of Timely Earnings Data and Inconsistent Processing of Work CDRs 
Allow Overpayments to Accrue: 

Dated IRS Earnings Data and Failure of Beneficiaries to Report 
Earnings Increase Work-Related Overpayments: 

SSA conducts periodic computer matches with wage data from the IRS to 
independently verify beneficiaries' earnings. However, earnings data 
provided through the IRS match are often more than a year old when SSA 
staff begin the work CDR prompted by the match. Managers and staff at 
the four processing centers we visited cited this delay as a major 
obstacle to limiting the occurrence and size of overpayments. Our work 
shows that this has delayed processing of work CDRs. In the 60 cases 
we reviewed, the earnings data were already between 6 and 26 months 
old by the time they were available to SSA staff for performing work 
CDRs. (Figure 6 shows how old earnings data were for the 60 cases.) 

Figure 6: Age of Earnings Data Provided to SSA by IRS Earnings Alerts, 
of 60 Cases Reviewed: 

[Refer to PDF for image: vertical bar graph] 

Age of earnings data: 0-12 months; 
Number of cases: 3. 

Age of earnings data: 12-18 months; 
Number of cases: 29. 

Age of earnings data: 18-24 months; 
Number of cases: 24. 

Age of earnings data: 24 or more months; 
Number of cases: 4. 

Source: GAO analysis of SSA data. 

[End of figure] 

While DI beneficiaries are responsible for notifying SSA when they 
return to work as a condition of receiving benefits, they sometimes 
fail to make such notifications. Our review of 60 cases found no 
indication in 49 that the individual had reported their work and 
earnings to SSA as instructed by regulation. In the other 11 cases, 
beneficiaries had reported returning to work, including the name of 
their employer and the amount of their wages, at some point. Yet 6 of 
these cases resulted in about $78,000 in total overpayments, even 
though these beneficiaries reported returning to work more than a year 
prior to initiation of the work CDR. In the remaining 5 cases, the 
beneficiary reported working only after the CDR was initiated. 

Earnings data from IRS or from beneficiaries may age further once 
received by SSA because no specific program guidelines exist that 
require immediate action on results of the IRS match, and consequently 
staff do not always begin a work CDR immediately. From the date of the 
initial IRS alert to the date staff begin work on the CDR, cases that 
have been identified by the IRS match and selected for a work CDR (but 
for which no action has yet been taken to process the work CDR) are 
categorized by SSA as "pending development." In the 60 cases we 
reviewed, the median time cases were pending development was 205 days, 
or about 7 months, and ranged from 2 to 466 days, or more than 15 
months.[Footnote 44] For example, in the 466-day case, the IRS alert 
came to SSA in September 2007, when earnings (for 2006) were already 
15 months old, then aged an additional 15 months until SSA staff began 
developing the work CDR. SSA officials could not explain what caused 
the delay in initiating development of this case or of several others 
we reviewed. (Figure 7 shows the number of days that the 60 cases were 
pending development.) 

Figure 7: Time Pending Development after IRS Alert Delivered, of 60 
Cases Reviewed: 

[Refer to PDF for image: vertical bar graph] 

Elapsed time: 0-100 days; 
Number of cases: 12. 

Elapsed time: 100-300 days; 
Number of cases: 36. 

Elapsed time: 300-500 days; 
Number of cases: 12. 

Source: GAO analysis of SSA data. 

[End of figure] 

The delays that occur when staff do not act promptly to begin a work 
CDR, in combination with the initial delays in receiving beneficiary 
earnings data (either from the IRS enforcement operation or 
beneficiaries' failure to self-report earnings), can result in 
multiple DI overpayments which may continue to accrue for extended 
periods of time before they are addressed. For example, in the 60 
cases we reviewed, delays which occurred after IRS alerts were 
delivered to SSA resulted in individual beneficiaries being overpaid 
for up to 38 months.[Footnote 45] Most received fewer than 12 months 
of overpayments, but 19 of the cases received 18 or more months of 
overpayments. According to an SSA official, staff shortages and the 
need to focus resources on competing workloads, such as initial DI 
claims and medical CDRs, are among the factors delaying development of 
work CDRs once earnings information is received. (Figure 8 shows the 
amount of time that overpayments accrued pending development.) 

Figure 8: Number of Months Overpayments Accrued as CDR Development Was 
Pending, of 60 Cases Reviewed: 

[Refer to PDF for image: vertical bare graph] 

Months of overpayments: 0-6; 
Number of cases: 18. 

Months of overpayments: 6-12; 
Number of cases: 14. 

Months of overpayments: 12-18; 
Number of cases: 9. 

Months of overpayments: 18-24; 
Number of cases: 9. 

Months of overpayments: 24 or more; 
Number of cases: 10. 

Source: GAO analysis of SSA data. 

[End of figure] 

In 2004, we recommended that SSA seek to use large scale batch matches 
with an alternative database of earnings, the National Directory of 
New Hires (NDNH), which was originally established to help states 
locate noncustodial parents for child support payments. The NDNH could 
provide SSA with quarterly wage information on employees within 4 
months of the end of a calendar quarter.[Footnote 46] Several federal 
programs and agencies currently use the NDNH to verify program 
eligibility, detect and prevent potential fraud or abuse, and collect 
overpayments. In 2009, SSA conducted a cost-effectiveness study on use 
of the NDNH, which concluded that a match would generate a large 
number of alerts needing development that were not of high quality. 
They also said the study found a return on investment of only about 
$1.40 in savings for each $1 spent. However, based on the information 
provided, we believe that some of the assumptions used in the analysis 
were overly pessimistic. For example, though SSA initially estimated 
savings of nearly $8 for every dollar spent by using the NDNH, the 
agency subsequently reduced this estimate by concluding that much of 
the anticipated savings from this match would ultimately be captured 
by the current enforcement operation. This assumption does not appear 
to fully take into account the fact that initial use of the NDNH would 
likely identify many potential overpayments earlier in the process, 
thus reducing the duration and size of those overpayments compared 
with the current enforcement operation. Thus, it is likely that the 
actual savings that result could in fact be higher. The agency's 
actual experience with the NDNH in its SSI program suggests its 
application in the DI program may be more cost-effective than 
indicated by SSA's analysis. According to SSA officials, the NDNH 
match with the SSI program has resulted in an estimated $200 million 
in savings per year for the program.[Footnote 47] Furthermore, even if 
the savings resulting from use of the NDNH for identifying DI 
beneficiaries' earnings is only $1.40 for every $1 spent, as estimated 
by SSA, this still represents a 40 percent rate of return. 

In a 2010 report, we also recommended that SSA evaluate the 
feasibility of incorporating the twice-annual AERO process in the 
agency's work CDR process.[Footnote 48] SSA does not use AERO to 
identify beneficiaries who work. SSA officials told us they are in the 
early stages of evaluating this use and, as part of this effort, they 
said they may include an AERO alert as one of several screening 
factors to identify cases at high risk for DI overpayments. Agency 
officials also told us that they plan to match cases that have both an 
AERO alert and an IRS alert and delay any AERO-awarded changes to 
benefit amounts until the CDR has been completed to potentially reduce 
overpayments. 

SSA Lacks Agencywide Performance Goals and a Consistent Approach for 
Processing Work CDRs: 

Similar to our findings about the lack of agencywide performance goals 
for overpayment debt recovery, we also found that SSA does not have 
agencywide performance goals or a consistent approach for processing 
work CDRs across its processing centers. Specifically, the agency does 
not have performance goals for the number of days taken to completely 
process a work CDR. While SSA has established an agencywide goal for 
processing a certain number of medical CDRs in a fiscal year, and 
includes this goal in the agency's annual performance plan, SSA 
officials told us they have not established similar goals for work 
CDRs. Instead, they have established "targets" for the processing 
centers. For example, in fiscal year 2011, SSA set targets for the 
completion of 95 percent of IRS alerts on earnings generated in 2008 
or earlier by September 24, 2010, and for processing centers to 
complete development of at least 99 percent of cases within 270 days 
by September 30, 2011.[Footnote 49] Overall, we found that while SSA's 
policies establish steps for work CDR processing to be followed across 
all processing centers, processing times across the four centers we 
visited varied widely once development was initiated. More 
specifically, we found that processing times for the 60 cases we 
reviewed ranged from 82 to 992 days (with a median of 396 
days)[Footnote 50] and resulted in combined overpayments totaling more 
than $1 million. We also found the median processing time for the 
cases we reviewed from three centers ranged from 307 to 397 days, and 
the median processing time at the fourth center, which processes about 
50 percent of all work CDRs, was 626 days. (Figure 9 shows the 
variance in processing time across the four processing centers.) 

Figure 9: Variance in Total Case Processing Time across Four 
Processing Centers Visited, of 60 Cases Reviewed: 

[Refer to PDF for image: illustration] 

Processing center A: 
Minimum: 254; 
Median: 397; 
Maximum: 552. 

Processing center B: 
Minimum: 205; 
Median: 307; 
Maximum: 633. 

Processing center C: 
Minimum: 82; 
Median: 314; 
Maximum: 454. 

Processing center D: 
Minimum: 476; 
Median: 626; 
Maximum: 992. 

Source: GAO analysis of SSA data. 

[End of figure] 

Recent SSA Initiatives Are Aimed at Identifying Cases at Greatest Risk 
of Overpayments: 

Within the last year, SSA has started work on several new initiatives 
to identify work CDR enforcement alerts from its IRS match that 
correspond to those cases which are more likely to result in large 
overpayments. First, the agency has started prioritizing IRS alerts 
with reported earnings that are greater than or equal to 12 times the 
current SGA level ($1,000 per month in calendar year 2011). Second, in 
response to a prior GAO recommendation, SSA is testing a "predictive 
model" intended to identify those cases with the highest probability 
of having large overpayments and prioritizing them for review by 
processing center staff. Using existing screening criteria, the model 
will assign cases a numeric score based on the beneficiary's age, type 
of disability, benefit amount, earnings, time on disability rolls, and 
number of denials for eligibility, among others. The highest-scoring 
alerts would be processed first because SSA believes they are most 
likely to result in cessation of DI benefit payments and thus be 
incurring overpayments. This initiative is currently being piloted in 
three processing centers. 

In a third initiative, SSA is piloting a comprehensive review of how 
all types of work CDRs are conducted using a sample of 1,000 recently 
completed CDRs. This review is intended to identify any aspects in the 
initial process that did not appear to be completed correctly. Because 
work CDRs are the most common type of CDR the agency encounters, SSA 
anticipates they will be a majority of the sample. Agency officials 
shared some of the review's early findings with us, including 
inadequate documentation of case development, inconsistencies and 
vague language in SSA's operating procedures, and errors in 
calculating SGA, which may contribute to overpayments. SSA's Office of 
Quality and Performance expects to issue its study results in fall 
2011. Fourth, SSA staff are working to update and streamline work 
issue procedures regarding the initiation, follow-up time frames, and 
overall completion of work CDRs for processing center personnel. The 
agency anticipates releasing the updated POMS that will incorporate 
these new procedures in October 2011. 

Finally, SSA has revised the forms filed by beneficiaries to document 
work activity and to make SGA determinations by SSA staff. According 
to agency officials, the forms should be easier for beneficiaries to 
understand and complete, and should improve compliance with 
beneficiary reporting requirements. As part of this initiative, SSA 
officials told us that they have also reduced the number of follow-up 
requests for these forms before staff are allowed to make a decision 
concerning a beneficiary's continued eligibility for DI benefits. 
According to agency officials, the Office of Management and Budget 
approved both of the revised work activity reports. In addition, SSA 
staff are hopeful that the revised form will result in more timely 
work CDR processing and reduced overpayments. While these initiatives 
represent promising steps, it is too early to assess what impact they 
may have on the prevalence and size of DI overpayments. 

Conclusions: 

Our review shows that while SSA has some initiatives to improve how it 
performs work CDRs, existing policies and processes impede its ability 
to more effectively detect, prevent, and recover overpayments. In 
particular, the existence of repayment plans that extend out past a 
debtor's full retirement age or projected lifespan means that it is 
unlikely those debts will be repaid in full. In addition, the lack of 
both agencywide performance goals that specify debt collection time 
frames and policies that specify the review of individual debt 
repayment plans to ensure that they are meeting these goals hinder the 
agency's debt collection efforts. Compounding this problem, the 
inability of the ROAR system to process debt recovery actions past the 
year 2049 jeopardizes the agency's ability to accurately track and 
report on the recovery of all overpayments. In addition, SSA's 
continued reliance on outdated earnings information to identify 
beneficiaries who may no longer be eligible for benefits means that 
overpayments, including some large overpayments, are probably 
inevitable. SSA's dependence on such data also means that staff 
responsible for performing work CDRs are often dealing with cases that 
are already aged when they receive them, and thus more difficult and 
time-consuming to develop. As a result, SSA is often in a "pay and 
chase" mode that further stretches limited staff and budgetary 
resources. Absent more timely sources of earnings data to inform the 
work CDR process, this problem is likely to persist. Additionally, 
other weaknesses such as the lack of agencywide performance goals for 
processing centers and staff likely contribute to delays in processing 
this workload. These delays, in turn, can become a factor in the large 
size of some overpayments. Without performance goals specifying the 
time that cases should be pending development, or the number of days 
to completely process a work CDR, improvements to the agency's 
operations in this area are made less likely. We recognize that 
ensuring the integrity of the DI program while also performing other 
work, such as processing medical CDRs and helping beneficiaries return 
to work, is a challenge for SSA. However, the continuing weaknesses we 
identified in SSA's existing work CDR processes and policies, as well 
as the mounting overpayment debt and its impact on the financial 
health of the DI Trust Fund, require sustained management attention 
and a more proactive stance by the agency. 

Recommendations: 

To enhance SSA's ability to recover debt and to improve the detection 
and, where possible, prevention of overpayments in the DI program, we 
recommend that the Commissioner of Social Security: 

1. develop and adopt agencywide performance goals, for the recovery of 
DI overpayment debt, such as the percent of outstanding debt collected 
annually. 

2. require supervisory review and approval of repayment plans which 
exceed SSA's target of 36 months. 

3. correct the ROAR 2049 system limitation so that debt scheduled for 
collection after 2049 is included in the system and available for SSA 
management, analysis, and reporting. 

4. explore options for obtaining more timely earnings information for 
DI program beneficiaries who may be working, and thus are more likely 
to incur overpayments. This would include developing data sharing 
agreements to access pertinent earnings-related databases, such as the 
NDNH. 

5. develop and adopt additional formal agencywide performance goals 
for work CDRs to measure the time that cases are pending development, 
and the number of days taken to process a work CDR. 

Agency Comments and Our Evaluation: 

We obtained written comments on a draft of this report from the 
Commissioner of the Social Security Administration. The comments are 
reproduced in appendix III. SSA also provided additional technical 
comments, which have been incorporated in the report as appropriate. 
SSA agreed with four of five recommendations we made to the 
Commissioner to strengthen SSA's processes and management controls 
over the detection, prevention, and recovery of DI overpayments. 

SSA agreed with our first recommendation to develop and adopt 
additional agencywide performance goals for the recovery of DI 
overpayments, such as the percent of outstanding debt collected 
annually. The agency noted that the Office of Management and Budget 
(OMB) guidance for the Improper Payments Elimination and Reduction Act 
of 2010 requires agencies to establish targets that drive annual 
performance based on percentage of recovery (i.e., proportion of 
overpayment dollars identified which are recovered in a fiscal year). 
The agency said it is working with OMB to develop appropriate debt 
recovery goals. We agree this is a positive step, as would be 
including that targeted percentage in its annual agencywide 
performance goals. 

SSA disagreed with our second recommendation to require supervisory 
review and approval of DI repayment plans which exceed SSA's targeted 
36 months. In particular, the agency noted that it negotiates these 
plans based on a beneficiary's income, ability to pay, and personal 
circumstances. SSA said this recommendation would not improve 
collection rates, and an additional approval step is unnecessary. 
However, the agency said it will issue guidance to its employees 
reminding them to make every effort to recover debts within 36 months. 
We agree that issuing guidance to staff to remind them about repayment 
time frames is a positive step, but based on recent experiences, we 
continue to believe supervisory review and approval of plans that 
exceed 36 months is needed to strengthen SSA's controls over the 
process and decrease the incidence of repayment plans that extend past 
a debtor's full retirement age or projected lifespan. 

SSA agreed with our third recommendation that SSA correct the ROAR 
2049 system limitation so debt scheduled for collection after 2049 is 
included in the system and available for current SSA management, 
analysis, and reporting. SSA noted the agency is considering solutions 
for this limitation. 

SSA agreed with our fourth recommendation to explore options for 
obtaining more timely earnings information for DI program 
beneficiaries who may be working, and thus more likely to incur 
overpayments. These options include data sharing agreements to access 
pertinent earnings-related databases such as the NDNH. The agency 
noted it has plans in this area. Specifically, SSA said while it had 
previously determined it would not be cost-effective to use the NDNH 
for this purpose, it will reevaluate that decision. SSA also said it 
will continue to explore other options for obtaining timely earnings 
information from other sources. In addition, the agency said it has 
developed, based on a previous GAO recommendation, a computer match of 
Social Security employee payroll records and DI and SSI benefit rolls. 
We agree this internal SSA computer match should help to obtain more 
timely earnings information on DI beneficiaries who may be working. We 
also encourage SSA to develop data-sharing agreements to leverage 
other databases such as NDNH which allow for regular batch-file 
matches to verify beneficiaries' earnings. 

SSA agreed with our fifth recommendation to develop and adopt 
additional formal, agencywide performance goals for the time work CDR 
cases are pending development and the number of days taken to process 
them. The agency commented that it already has internal targets for 
overall work CDR processing times. SSA said it will consider GAO's 
recommendation as it works on its agency performance goals. We agree 
the agency's internal targets for work CDR processing are a positive 
step. However, we continue to believe that formal, agencywide work CDR 
performance goals are needed to assist SSA in assessing its 
performance and fully leveraging its resources in this area. The 
agency should consider measuring the time cases are pending 
development, and the number of days taken to process a work CDR, 
against agencywide annual performance goals it develops. 

We are sending copies of this report to the Commissioner of Social 
Security, appropriate congressional committees, and other interested 
parties. The report is also available at no charge on GAO's Web site 
at [hyperlink, http://www.gao.gov]. 

If you or your staffs have any questions about this report, please 
contact me at (202) 512-7215 or bertonid@gao.gov. Contact points for 
our Offices of Congressional Relations and Public Affairs may be found 
on the last page of this report. Key contributors to this report are 
listed in appendix IV. 

Signed by: 

Daniel Bertoni: 
Director, Education, Workforce, and Income Security Issues: 

[End of section] 

Appendix I: Enforcement Cumulative Totals and Distribution for IRS 
Enforcement Matches Conducted in 2008-2010 (Data for Earnings Years 
2007 through 2009): 

Table 1: Enforcement Cumulative Totals and Distribution for Earnings 
Year 2009 Data: 

2009 run 1: 06/23/2010; 
Number of enforcement records: 1,640,631; 
Alerted to processing centers 1-8: 
Count: 422,586;
Percent: 25.76%; 
Alerted to field offices: 
Count: 9,878; 
Percent: 0.60%; 
Not alerted[A]: 
Count: 1,208,167; 
Percent: 73.64%; 
Total: 
Count: 1,640,631; 
Percent: 100%; 
Processing Center 1: 
Count: 30,039; 
Percent: 7.11%; 
Processing Center 2: 
Count: 25,599; 
Percent: 6.06%; 
Processing Center 3: 
Count: 37,087; 
Percent: 8.78%; 
Processing Center 4: 
Count: 32,074; 
Percent: 7.59%; 
Processing Center 5: 
Count: 31,150; 
Percent: 7.37%; 
Processing Center 6: 
Count: 37,262; 
Percent: 8.82%; 
Processing Center 7: 
Count: 228,956; 
Percent: 54.18%; 
Processing Center 8: 
Count: 419; 
Percent: 0.10%; 
Total: 
Count: 422,586; 
Percent: 100%. 

2009 run 2: 09/08/2010; 
Number of enforcement records: 423,159; 
Alerted to processing centers 1-8: 
Count: 97,730; 
Percent: 23.09%; 
Alerted to field offices: 
Count: 1,289; 
Percent: 0.30%; 
Not alerted[A]: 
Count: 324,140; 
Percent: 76.60%; 
Total: 
Count: 423,159; 
Percent: 100%. 
Processing Center 1: 
Count: 6,476; 
Percent: 6.63%; 
Processing Center 2: 
Count: 7,031; 
Percent: 7.19%; 
Processing Center 3: 
Count: 10,395; 
Percent: 10.64%; 
Processing Center 4: 
Count: 9,177; 
Percent: 9.39%; 
Processing Center 5: 
Count: 8,924; 
Percent: 9.13%; 
Processing Center 6: 
Count: 11,541; 
Percent: 11.81%; 
Processing Center 7: 
Count: 44,062; 
Percent: 45.08%; 
Processing Center 8: 
Count: 124; 
Percent: 0.13%; 
Total: 
Count: 97,730; 
Percent: 100%. 

2009 run 3: 
Number of enforcement records: 0; 
Alerted to processing centers 1-8: 
Count: 0; 
Percent: 0%; 
Alerted to field offices: 
Count: 0; 
Percent: 0%; 
Not alerted[A]: 
Count: 0; 
Percent: 0%; 
Total: 
Count: 0; 
Percent: N/A; 
Processing Center 1: 
Count: 0; 
Percent: 0%; 
Processing Center 2: 
Count: 0; 
Percent: 0%; 
Processing Center 3: 
Count: 0; 
Percent: 0%; 
Processing Center 4: 
Count: 0; 
Percent: 0%; 
Processing Center 5: 
Count: 0; 
Percent: 0%; 
Processing Center 6: 
Count: 0; 
Percent: 0%; 
Processing Center 7: 
Count: 0; 
Percent: 0%; 
Processing Center 8: 
Count: 0; 
Percent: 0%; 
Total: 
Count: 0; 
Percent: N/A; 

2009 totals: 
Number of enforcement records: 2,063,790; 
Alerted to processing centers 1-8: 
Count: 520,316; 
Percent: 25.21%; 
Alerted to field offices: 
Count: 11,167; 
Percent: 0.54%; 
Not alerted[A]: 
Count: 1,532,307; 
Percent: 74.25%; 
Total: 
Count: 2,063,790; 
Percent: 100%; 
Processing Center 1: 
Count: 36,515; 
Percent: 7.02%; 
Processing Center 2: 
Count: 32,630; 
Percent: 6.27%; 
Processing Center 3: 
Count: 47,482; 
Percent: 9.13%; 
Processing Center 4: 
Count: 41,251; 
Percent: 7.93%; 
Processing Center 5: 
Count: 40,074; 
Percent: 7.70%; 
Processing Center 6: 
Count: 48,803; 
Percent: 9.38%; 
Processing Center 7: 
Count: 273,108; 
Percent: 52.47%; 
Processing Center 8: 
Count: 543; 
Percent: 0.10%; 
Total: 
Count: 520,316; 
Percent: 100%. 

Source: SSA provided DI earnings enforcement data. 

[A] After the initial match, selection criteria are applied to 
identify those records with earnings that SSA did not previously know 
about, and are most likely to affect DI benefit payments. The records 
that are not selected for review are not sent to the processing 
centers or field offices for review. 

[End of table] 

Table 2: Enforcement Cumulative Totals and Distribution for Earnings 
Year 2008 Data: 

2008 run 1a: 08/5/2009; 
Number of enforcement records: 1,775,131; 
Alerted to processing centers 1-8: 
Count: 469,051;
Percent: 26.42%; 
Alerted to field offices: 
Count: 15,208; 
Percent: 0.86%; 
Not alerted[A]: 
Count: 1,290,872; 
Percent: 72.72%; 
Total: 
Count: 1,775,131; 
Percent: 100%; 
Processing Center 1: 
Count: 34,412; 
Percent: 7.34%; 
Processing Center 2: 
Count: 30,162; 
Percent: 6.43%; 
Processing Center 3: 
Count: 42,360; 
Percent: 9.03%; 
Processing Center 4: 
Count: 36,505; 
Percent: 7.78%; 
Processing Center 5: 
Count: 35,450; 
Percent: 7.56%; 
Processing Center 6: 
Count: 42,778; 
Percent: 9.12%; 
Processing Center 7: 
Count: 246,903; 
Percent: 52.64%; 
Processing Center 8: 
Count: 481; 
Percent: 0.10%; 
Total: 
Count: 469,051; 
Percent: 100%. 

2008 run 1b: 08/12/2009; 
Number of enforcement records: 5; 
Alerted to processing centers 1-8: 
Count: 5; 
Percent: 100%; 
Alerted to field offices: 
Count: 0; 
Percent: 0%; 
Not alerted[A]: 
Count: 0; 
Percent: 0%; 
Total: 
Count: 5; 
Percent: 100%. 
Processing Center 1: 
Count: 0; 
Percent: 0%; 
Processing Center 2: 
Count: 0; 
Percent: 0%; 
Processing Center 3: 
Count: 0; 
Percent: 0%; 
Processing Center 4: 
Count: 0; 
Percent: 0%; 
Processing Center 5: 
Count: 0; 
Percent: 0%; 
Processing Center 6: 
Count: 0; 
Percent: 0%; 
Processing Center 7: 
Count: 5; 
Percent: 100%; 
Processing Center 8: 
Count: 0; 
Percent: 0%; 
Total: 
Count: 5; 
Percent: 100%. 

2008 run 2: 03/24/2009; 
Number of enforcement records: 483,226; 
Alerted to processing centers 1-8: 
Count: 72,443; 
Percent: 14.99%; 
Alerted to field offices: 
Count: 1,117; 
Percent: 0.23%; 
Not alerted[A]: 
Count: 409,666; 
Percent: 84.78%; 
Total: 
Count: 483,226; 
Percent: 100%; 
Processing Center 1: 
Count: 4,548; 
Percent: 6.28%; 
Processing Center 2: 
Count: 4,669; 
Percent: 6.45%; 
Processing Center 3: 
Count: 7,749; 
Percent: 10.70%; 
Processing Center 4: 
Count: 6,612; 
Percent: 9.13%; 
Processing Center 5: 
Count: 5,888; 
Percent: 8.13%; 
Processing Center 6: 
Count: 7,794; 
Percent: 10.75%; 
Processing Center 7: 
Count: 35,090; 
Percent: 48.43%; 
Processing Center 8: 
Count: 93; 
Percent: 0.13%; 
Total: 
Count: 72,443; 
Percent: 100%. 

2008 totals: 
Number of enforcement records: 2,258,362; 
Alerted to processing centers 1-8: 
Count: 541,499; 
Percent: 23.97%; 
Alerted to field offices: 
Count: 16,325; 
Percent: 0.72%; 
Not alerted[A]: 
Count: 1,700,538; 
Percent: 75.29%; 
Total: 
Count: 2,258,362; 
Percent: 100%; 
Processing Center 1: 
Count: 38,960; 
Percent: 7.19%; 
Processing Center 2: 
Count: 34,831; 
Percent: 6.43%; 
Processing Center 3: 
Count: 50,109; 
Percent: 9.25%; 
Processing Center 4: 
Count: 43,117; 
Percent: 7.96%; 
Processing Center 5: 
Count: 41,338; 
Percent: 7.63%; 
Processing Center 6: 
Count: 50.572; 
Percent: 9.34%; 
Processing Center 7: 
Count: 281,998; 
Percent: 52.08%; 
Processing Center 8: 
Count: 574; 
Percent: 0.11%; 
Total: 
Count: 541,499; 
Percent: 100%. 

Source: SSA provided DI earnings enforcement data. 

[A] After the initial match, selection criteria are applied to 
identify those records with earnings that SSA did not previously know 
about, and are most likely to affect DI benefit payments. The records 
that are not selected for review are not sent to the processing 
centers or field offices for review. 

[End of table] 

Table 3: Enforcement Cumulative Totals and Distribution for Earnings 
Year 2007 Data: 

2007 run 1: 7/23/2008; 
Number of enforcement records: 1,397,272; 
Alerted to processing centers 1-8: 
Count: 394,405;
Percent: 28.23%; 
Alerted to field offices: 
Count: 20,746; 
Percent: 1.48%; 
Not alerted[A]: 
Count: 982,121; 
Percent: 70.29%; 
Total: 
Count: 1,397,272; 
Percent: 100%; 
Processing Center 1: 
Count: 26,221; 
Percent: 6.65%; 
Processing Center 2: 
Count: 22,753; 
Percent: 5.77%; 
Processing Center 3: 
Count: 31,782; 
Percent: 8.06%; 
Processing Center 4: 
Count: 28,399; 
Percent: 7.20%; 
Processing Center 5: 
Count: 25,252; 
Percent: 6.40%; 
Processing Center 6: 
Count: 30,832; 
Percent: 7.82%; 
Processing Center 7: 
Count: 228,731; 
Percent: 57.99%; 
Processing Center 8: 
Count: 435; 
Percent: 0.11%; 
Total: 
Count: 394,405; 
Percent: 100%. 

2007 run 2: 10/1/2008; 
Number of enforcement records: 690,823; 
Alerted to processing centers 1-8: 
Count: 136,293; 
Percent: 19.73%; 
Alerted to field offices: 
Count: 4,359; 
Percent: 0.63%; 
Not alerted[A]: 
Count: 550,171; 
Percent: 79.64%; 
Total: 
Count: 690,823; 
Percent: 100%. 
Processing Center 1: 
Count: 9,235; 
Percent: 6.78%; 
Processing Center 2: 
Count: 8,439; 
Percent: 6.19%; 
Processing Center 3: 
Count: 11.956; 
Percent: 8.77%; 
Processing Center 4: 
Count: 10,952; 
Percent: 8.04%; 
Processing Center 5: 
Count: 11,634; 
Percent: 8.54%; 
Processing Center 6: 
Count: 13.647; 
Percent: 10.01%; 
Processing Center 7: 
Count: 70,263; 
Percent: 51.55%; 
Processing Center 8: 
Count: 175; 
Percent: 0.13%; 
Total: 
Count: 136,301; 
Percent: 100%. 

2008 run 3a: 4/1/2009; 
Number of enforcement records: 110,474; 
Alerted to processing centers 1-8: 
Count: 72; 
Percent: 0.07%; 
Alerted to field offices: 
Count: 14; 
Percent: 0.01%; 
Not alerted[A]: 
Count: 110,388; 
Percent: 99.92%; 
Total: 
Count: 110,474; 
Percent: 100%; 
Processing Center 1: 
Count: 23; 
Percent: 31.94%; 
Processing Center 2: 
Count: 4; 
Percent: 5.56%; 
Processing Center 3: 
Count: 0; 
Percent: 0%; 
Processing Center 4: 
Count: 0; 
Percent: 0%; 
Processing Center 5: 
Count: 0; 
Percent: 0%; 
Processing Center 6: 
Count: 3; 
Percent: 4.17%; 
Processing Center 7: 
Count: 42; 
Percent: 58.33%; 
Processing Center 8: 
Count: 0; 
Percent: 0%; 
Total: 
Count: 72; 
Percent: 100%. 

2008 run 3b: 5/19/2009; 
Number of enforcement records: 110,388; 
Alerted to processing centers 1-8: 
Count: 13,841; 
Percent: 12.54%; 
Alerted to field offices: 
Count: 506; 
Percent: 0.46%; 
Not alerted[A]: 
Count: 96,041; 
Percent: 87.00%; 
Total: 
Count: 110,388; 
Percent: 100%; 
Processing Center 1: 
Count: 663; 
Percent: 4.79%; 
Processing Center 2: 
Count: 763; 
Percent: 5.51%; 
Processing Center 3: 
Count: 1.156; 
Percent: 8.35%; 
Processing Center 4: 
Count: 922; 
Percent: 6.66%; 
Processing Center 5: 
Count: 967; 
Percent: 6.99%; 
Processing Center 6: 
Count: 1,261; 
Percent: 9.11%; 
Processing Center 7: 
Count: 8,097; 
Percent: 58.50%; 
Processing Center 8: 
Count: 12; 
Percent: 0.09%; 
Total: 
Count: 13,841; 
Percent: 100%. 

2007 totals: 
Number of enforcement records: 2,308,957; 
Alerted to processing centers 1-8: 
Count: 544,611; 
Percent: 23.59%; 
Alerted to field offices: 
Count: 25,625; 
Percent: 1.11%; 
Not alerted[A]: 
Count: 1,738,721; 
Percent: 75.30%; 
Total: 
Count: 2,308,957; 
Percent: 100%; 
Processing Center 1: 
Count: 36,142; 
Percent: 6.64%; 
Processing Center 2: 
Count: 31,959; 
Percent: 5.87%; 
Processing Center 3: 
Count: 44,894; 
Percent: 8.24%; 
Processing Center 4: 
Count: 40.273; 
Percent: 7.39%; 
Processing Center 5: 
Count: 37,853; 
Percent: 6.95%; 
Processing Center 6: 
Count: 45,743; 
Percent: 8.40%; 
Processing Center 7: 
Count: 307,133; 
Percent: 56.39%; 
Processing Center 8: 
Count: 622; 
Percent: 0.11%; 
Total: 
Count: 544,619; 
Percent: 100%. 

Source: SSA provided DI earnings enforcement data. 

[A] After the initial match, selection criteria are applied to 
identify those records with earnings that SSA did not previously know 
about, and are most likely to affect DI benefit payments. The records 
that are not selected for review are not sent to the processing 
centers or field offices for review. 

[End of table] 

[End of section] 

Appendix II: Financial Data on DI Overpayments and Overpayment Debt: 

Fiscal year: 2001; 
Cumulative DI debt balance at the beginning of fiscal year: $2.165 
billion; 
Plus: All new DI debts detected, re-established debts, and 
adjustments: $892.8 million; 
Minus: Collections: $340.1 million; 
Minus: Write-offs[A]: $261.4 million; 
Equals: Cumulative DI debt balance at the end of fiscal year: $2.457 
billion. 

Fiscal year: 2002; 
Cumulative DI debt balance at the beginning of fiscal year: $2.457 
billion; 
Plus: All new DI debts detected, re-established debts, and 
adjustments: $918.2 million; 
Minus: Collections: $366.9 million; 
Minus: Write-offs[A]: $320.8 million; 
Equals: Cumulative DI debt balance at the end of fiscal year: $2.687 
billion. 

Fiscal year: 2003; 
Cumulative DI debt balance at the beginning of fiscal year: $2.687 
billion; 
Plus: All new DI debts detected, re-established debts, and 
adjustments: $1.029 billion; 
Minus: Collections: $430.9 million; 
Minus: Write-offs[A]: $324.5 million; 
Equals: Cumulative DI debt balance at the end of fiscal year: $2.961 
billion. 

Fiscal year: 2004; 
Cumulative DI debt balance at the beginning of fiscal year: $2.961 
billion; 
Plus: All new DI debts detected, re-established debts, and 
adjustments: $990.9 million; 
Minus: Collections: $431.2 million; 
Minus: Write-offs[A]: $359.4 million; 
Equals: Cumulative DI debt balance at the end of fiscal year: $3.161 
billion. 

Fiscal year: 2005; 
Cumulative DI debt balance at the beginning of fiscal year: $3.161 
billion; 
Plus: All new DI debts detected, re-established debts, and 
adjustments: $1.435 billion; 
Minus: Collections: $480.1 million; 
Minus: Write-offs[A]: $361.3 million; 
Equals: Cumulative DI debt balance at the end of fiscal year: $3.755 
billion. 

Fiscal year: 2006; 
Cumulative DI debt balance at the beginning of fiscal year: $3.755 
billion; 
Plus: All new DI debts detected, re-established debts, and 
adjustments: $1.706 billion; 
Minus: Collections: $629.4 million; 
Minus: Write-offs[A]: $495.8 million; 
Equals: Cumulative DI debt balance at the end of fiscal year: $4.336 
billion. 

Fiscal year: 2007; 
Cumulative DI debt balance at the beginning of fiscal year: $4.336 
billion; 
Plus: All new DI debts detected, re-established debts, and 
adjustments: $1.513 billion; 
Minus: Collections: $685.6 million; 
Minus: Write-offs[A]: $463.3 million; 
Equals: Cumulative DI debt balance at the end of fiscal year: $4.700 
billion. 

Fiscal year: 2008; 
Cumulative DI debt balance at the beginning of fiscal year: $4.700 
billion; 
Plus: All new DI debts detected, re-established debts, and 
adjustments: $1.573 billion; 
Minus: Collections: $787.0 million; 
Minus: Write-offs[A]: $504.0 million; 
Equals: Cumulative DI debt balance at the end of fiscal year: $4.982 
billion. 

Fiscal year: 2009; 
Cumulative DI debt balance at the beginning of fiscal year: $4.982 
billion; 
Plus: All new DI debts detected, re-established debts, and 
adjustments: $1.534 billion; 
Minus: Collections: $839.8 million; 
Minus: Write-offs[A]: $488.3 million; 
Equals: Cumulative DI debt balance at the end of fiscal year: $5.188 
billion. 

Fiscal year: 2010; 
Cumulative DI debt balance at the beginning of fiscal year: $5.188 
billion; 
Plus: All new DI debts detected, re-established debts, and 
adjustments: $1.524 billion; 
Minus: Collections: $838.9 million; 
Minus: Write-offs[A]: $460.1 million; 
Equals: Cumulative DI debt balance at the end of fiscal year: $5.413. 

Source: SSA: 

[A] Write-offs consist of waivers and terminations of collection. A 
waiver is a permanent removal of debt from an individual's record and 
from SSA's balance sheet. These are generally approved when an 
individual is both not responsible for the overpayment and incapable 
of repaying the debt. A termination of debt collection means that SSA 
is giving up efforts to collect the debt internally, but may refer the 
debt for external collection. 

[End of table] 

[End of section] 

Appendix III: Comments from the Social Security Administration: 

Social Security: 
Office of the Commissioner: 
Social Security Administration: 
Baltimore, MD 21235-0001: 

July 15, 2011: 

Mr. Daniel Bertoni, Director: 
Education, Workforce, and Income Security Issues: 
United States Government Accountability Office: 
441 G. Street, NW: 
Washington, D.C. 20548: 

Dear Mr. Bertoni, 

Thank you for the opportunity to review the draft report, "Disability 
Insurance: SSA Can Improve Efforts to Detect, Prevent, and Recover 
Overpayments" (GAO-11-724). Our response is enclosed. 

If you have any questions, please contact me at (410) 965-0520. Your 
staff may contact Frances Cord, Director, Audit Management and Liaison 
Staff at (410) 966-5787. 

Sincerely, 

Signed by: 

Dean S. Landis: 
Deputy Chief of Staff: 

Enclosure: 

[End of letter] 

SSA Comments On The Government Accountability Office (GAO), Draft 
Report, "Disability Insurance: SSA Can Improve Efforts To Detect, 
Prevent, And Recover Overpayments" GA0-11-724: 

General Comments: 

We use many tools to assist us in collecting overpayments — benefit 
withholding, repayment arrangements, and external collection agencies. 
We pursue other options to prevent overpayments from occurring. For 
example, in January 2011 we revised Publication I405-10095, "Working 
While Disabled-How Can We Help" to provide specific instructions to 
Supplemental Security Income (SSI) and Social Security Disability 
Insurance (SSDI) recipients on how and when to report earnings or 
changes in work activity. We also published a one-page chart at 
[hyperlink, http://www.socialsecurity.gov/disability/] that provides 
guidance to SSDI beneficiaries on how to report information on income 
and work activity. We published a similar chart for SSI beneficiaries. 

We are also pursuing legislative changes. In the fiscal year 2012 
President's budget submission, we included a legislative proposal for 
a Work Incentives Simplification Pilot (WISP). The legislation, if 
passed, will allow us to replace complex rules with a clear, simple, 
unified process that is easier to understand and administer. It would 
reduce improper payments, and allow us to redirect resources to other 
areas. 

Responses To The Recommendations: 

Recommendation 1: 

Develop and adopt additional agency-wide performance goals, for the 
recovery of Disability Insurance (DI) overpayment debt, such as the 
percent of outstanding debt collected annually. 

Response: 

We agree. The Office of Management and Budget's (OMB) guidance for the 
Improper Payments Elimination and Reduction Act of 2010 requires 
agencies to establish targets that drive annual performance. Targets 
must be based on percentage-of-recovery — i.e., amount of improper 
overpayments recovered divided by the amount of improper overpayments 
identified. We are working with OMB to develop appropriate recovery 
goals. 

Recommendation 2: 

Require supervisory review and approval of repayment plans which 
exceed SSA's target of 36 months. 

Response: 

We disagree. Your recommendation would not improve collection rates. 
We negotiate repayment arrangements based on beneficiaries' income, 
ability to repay, and other circumstances. An additional approval 
process is unnecessary. However, we will issue guidance to our 
employees reminding them to make every effort to recover debts within 
36 months. 

Recommendation 3: 

Correct the ROAR 2049 system limitation so that debt scheduled for 
collection after 2049 is included in the system and available for SSA 
management, analysis, and reporting. 

Response: 

We agree. We are currently considering solutions to address this 
limitation. 

Recommendation 4: 

Explore options for obtaining more timely earnings information for Di 
program beneficiaries who may be working, and thus more likely to 
incur overpayments. This would include developing data sharing 
agreements to access pertinent earnings related databases, such as the
National Directory of New Hires. 

Response: 

We agree and have already taken actions in this area. You had a 
similar recommendation in your June 2010 report, "Cases of Federal 
Employees and Transportation Drivers and Owners Who Fraudulently 
and/or Improperly Received SSA Disability Payments." After you issued 
that report, we developed a computer match that compares Social 
Security employee payroll records to DI and SSI benefit rolls.
While we determined previously that it would not be cost effective to 
utilize the National Directory of New Hires for similar purposes, we 
will reevaluate that decision. We will continue to explore other 
options for obtaining timely earnings information from other sources. 

Recommendation 5: 

Develop and adopt additional formal agency-wide performance goals for 
work CDRs to measure the time that cases are pending development, and 
the number of days taken to process a work CDR. 

Response: 

We agree. We have already established internal targets for overall 
work CDR processing times. We will consider your recommendation as we 
work to establish our high priority performance goals. 

[End of section] 

Appendix IV GAO Contact and Staff Acknowledgments: 

GAO Contact: 

Daniel Bertoni, (202) 512-7215 or bertonid@gao.gov: 

Staff Acknowledgments: 

Jeremy Cox (Assistant Director) and Arthur T. Merriam Jr. (Analyst-in- 
Charge) managed all aspects of the assignment. Angela Jacobs, Joel 
Marus, and Katharine Kairys made significant contributions to this 
report, in all aspects of the work. In addition, Vanessa Taylor and 
Walter Vance provided technical support; Craig Winslow and Sheila 
McCoy provided legal support; Susan Aschoff assisted with the 
development of the message and report; David Forgosh, Monika Gomez, 
Cady Panetta, and Nyree Ryder Tee contributed to quality assurance for 
this product, and James Bennett provided graphics support. 

[End of section] 

Related GAO Products: 

Status of Fiscal Year 2010 Federal Improper Payments Reporting. 
[hyperlink, http://www.gao.gov/products/GAO-11-443R]. Washington, 
D.C.: March 25, 2011. 

Social Security Disability: Ticket to Work Participation Has Improved, 
but Additional Oversight Needed. [hyperlink, 
http://www.gao.gov/products/GAO-11-324]. Washington, D.C.: May 2011. 

High-Risk Series: An Update. [hyperlink, 
http://www.gao.gov/products/GAO-11-278]. Washington, D.C.: February 
16, 2011. 

Social Security Administration: Cases of Federal Employees and 
Transportation Drivers and Owners Who Fraudulently and/or Improperly 
Received SSA Benefits. [hyperlink, 
http://www.gao.gov/products/GAO-10-444]. Washington, D.C.: June 2010. 

Social Security Administration: Cases of Federal Employees and 
Transportation Drivers and Owners Who Fraudulently and/or Improperly 
Received SSA Benefits. [hyperlink, 
http://www.gao.gov/products/GAO-10-949T]. Washington, D.C.: August 4, 
2010. 

Disability Insurance: SSA Should Strengthen Its Efforts to Detect and 
Prevent Overpayments. [hyperlink, 
http://www.gao.gov/products/GAO-04-929]. Washington, D.C.: September 
10, 2004. 

[End of section] 

Footnotes: 

[1] See SSA and Medicare Boards of Trustees, Status of the Social 
Security and Medicare Programs: A Summary of the 2010 Annual Reports, 
[hyperlink, http://www.socialsecurity.gov/oact/trsum/index.html] 
(accessed October 2010). 

[2] 20 C.F.R. §§ 404.1571 and 404.1574(b)(2) (2011). The SGA level was 
$1,000 per month in 2010 for beneficiaries with disabilities and 
$1,640 per month for blind beneficiaries. While the SGA amount for 
statutorily blind individuals has remained the same for fiscal years 
2009 through 2011, the SGA amount for non-blind individuals in 2009 
was $980, and for 2010 and 2011 $1,000, respectively. See 75 Fed.Reg. 
65696 (October 26, 2010); 74 Fed.Reg. 55614 (October 28, 2009); and 73 
Fed.Reg. 54651 (October 30, 2008). 

[3] 20 C.F.R. § 404.1592 (2011). 

[4] 20 C.F.R. § 404.1589 - 404.1590 (2011). 

[5] Work CDRs also can originate from other sources, including work 
reports from beneficiaries or third parties, such as state vocational 
rehabilitation agencies. SSA also conducts medical CDRs, which focus 
on beneficiaries' medical improvements, but may also identify 
earnings. For purposes of this report, when we refer to work CDRs, we 
are referring to work CDRs performed by SSA's enforcement operation. 

[6] 42 U.S.C. § 404(a). 

[7] SSA administers the Ticket to Work program, to provide eligible DI 
beneficiaries with employment services, vocational rehabilitation 
services, or other support services to help them obtain and retain 
employment and reduce their dependence on benefits. See GAO, Ticket to 
Work Program: Participation Has Increased, but Additional Oversight 
Needed, [hyperlink, http://www.gao.gov/products/GAO-11-324] 
(Washington, D.C.: May 6, 2011). 

[8] Social Security Amendments of 1956, Pub. L. No. 84-880, § 103, 70 
Stat. 807, 815-24. 

[9] Average benefit amount to workers with disabilities, their 
spouses, and dependents, as of the end of September 2010. 

[10] 42 U.S.C. § 423(c). 

[11] 42 U.S.C. § 402. 

[12] 42 U.S.C. § 423(d). A blind individual must be prevented by 
blindness from engaging in SGA requiring skills or abilities like 
those previously engaged in regularly for a substantial period of time. 

[13] SGA for blind beneficiaries was $1,640 per month in 2010. 

[14] Pub. L. No. 106-170, § 101, 113 Stat. 1860, 1863-73 (codified as 
amended at 42 U.S.C. § 1320b-19). 

[15] Administered by the Department of Education since 1973, the 
Vocational Rehabilitation program provides funds to states to offer 
employment services ranging from treatment of impairments to job 
counseling and placement. 

[16] 20 C.F.R. § 404.1589 (2011). We use the term "work CDRs" to 
describe "full" work CDRs in which a case is fully developed and staff 
fill out specific forms to receive work credit for completing a work 
CDR, as well as instances in which SSA staff perform limited 
development of beneficiary earnings because they determine that a full 
work CDR is not necessary (an activity that SSA refers to as a "work 
CDR action"). SSA also conducts medical CDRs to periodically assess 
beneficiaries' continuing medical eligibility for benefits. 

[17] SSA generally uses six times the monthly SGA amount, or $6,000 in 
2010, as the annual earnings cutoff. Beneficiaries whose annual 
earnings are $6,000 or less are likely to keep their DI benefits 
because their monthly earnings are expected to be below program 
earning limits. 

[18] About half of the cases are sent to the processing center in 
SSA's Office of Disability Operations in Baltimore, Maryland. The 
office is responsible for handling beneficiaries who are less than 54 
years of age and live in the United States. The remaining cases are 
sent to one of the remaining seven processing centers. 

[19] The remaining records did not meet SSA's criteria for conducting 
an enforcement work CDR. 

[20] 20 C.F.R. § 404.1589 (2011). SSA contracts with state Disability 
Determination Services that are responsible for assessing whether an 
individual has a disability (a medical CDR). 

[21] In 2004, SSA implemented the eWork system, which is the primary 
system for processing work CDR cases in headquarters and field 
locations. 

[22] SSA guidelines allow DI beneficiaries to work for a limited time 
without affecting their benefits. The 9-month trial work period (TWP) 
allows beneficiaries to test their ability to work. SSA pays benefits 
during this period, no matter how high the earnings. After completing 
the TWP, the extended period of eligibility (EPE) begins. During the 
EPE, beneficiaries can receive benefits for any months their earnings 
are below a level of earnings called SGA. Generally, SSA considers 
earnings of more than $1,000 a month ($1,640 a month for blind 
beneficiaries) in 2011 to be SGA. SSA pays benefits to beneficiaries 
for the first month of SGA in the EPE and the next 2 months, and any 
month in which earnings are below SGA during the first 36 months of 
the EPE, the re-entitlement period. Eligibility for benefits will not 
terminate until the beneficiary has earnings above SGA after the 36- 
month re-entitlement period of the EPE. 

[23] "Earnings reviewers" in the processing centers are generally 
responsible for initial analysis of a beneficiary's earnings; however, 
only disability processing specialists have the authority to cease 
benefits. In SSA's field offices, the claims representatives are 
responsible for the duties performed by both the disability processing 
specialist and the earnings reviewer. 

[24] A beneficiary requests reconsideration when he or she disputes 
the occurrence of the overpayment itself, 20 C.F.R. § 404.907 (2011), 
and requests a waiver when asserting he or she is both not responsible 
for the overpayment and incapable of repaying the debt, 20 C.F.R. § 
404.506 (2011). A waiver permanently terminates collection of a debt 
and removes the debt from SSA's balance sheet. 

[25] 20 C.F.R. § 404.509 (2011). 

[26] 20 C.F.R. § 404.530 (2011). 

[27] Garnishment is a process whereby a debtor's property (in this 
case, wages) is turned over to the debtor's creditor. In using 
administrative wage garnishment, SSA orders an employer to withhold 
amounts each payday from an employee who owes a debt to the agency, 
and the employer remits those amounts to SSA. 

[28] Credit bureau referral does not directly result in increased 
collections, but acts as a disincentive to individuals who decline to 
establish a repayment plan with SSA. The agency sends out notices to 
individuals indicating that a failure to establish a repayment plan 
will result in the referral of the debt information to credit bureaus. 
20 C.F.R. § 422.306(b) (2011). A poor credit score from a credit 
bureau can result in greater difficulty in borrowing money on 
favorable terms and other negative consequences for the debtor. 

[29] DI benefits paid by the program increased from about $58 billion 
to nearly $123 billion from fiscal year 2001 through fiscal year 2010. 
Reported overpayments do not include amounts removed from the record 
due to systems limitations, discussed later in this report. 

[30] Percentage applies to projected overpayment dollars, not 
incidents of overpayments. The Office of Quality Performance reviews a 
sample of work CDR cases each year to project total DI overpayments 
for the year as well as the prevalence of types of errors resulting in 
those overpayments. GAO previously reported that the estimated share 
of work-related, or SGA-related, overpayments between 1999 and 2003 
was about 31 percent based on available data from SSA at that time. 
GAO, Disability Insurance: SSA Should Strengthen Its Efforts to Detect 
and Prevent Overpayments, GAO-04-929 (Washington, D.C.: Sept. 10, 
2004). 

[31] This could affect all overpayment records, not just work-related 
overpayment records. 

[32] AERO checks, or recomputes, earnings by a beneficiary to ensure 
they are properly credited for that individual in determining the 
monthly disability benefit amount, including DI benefits. If an 
increase in benefit amount is due, AERO processes a benefit change and 
notifies the beneficiary. If no increase is due, AERO does not send a 
notice. AERO is run twice for each earnings year, usually in late 
October and late March. 

[33] Pub. L. No. 103-62, § 4(b), 107 Stat. 285, 287 (codified at 31 
U.S.C. § 1115(a)). 

[34] See SSA POMS GN 02210.214(C) and GN 02210.030(B). 

[35] The stated amounts for DI overpayment debt do not include 
interest or penalties. 

[36] A waiver is a permanent removal of debt from an individual's 
record and from SSA's balance sheet and cannot be collected in the 
future. Some waiver decisions are beyond the agency's control, such as 
those attributable to bankruptcy and administrative law judge 
decisions instructing the agency to waive overpayments. A termination 
of debt collection means that SSA is giving up efforts to collect the 
debt internally but will refer the debt for external collection, if 
the debt meets specific criteria for the applicable external 
collection tools. If the debtor subsequently receives DI or other 
benefits, SSA will resume its own collection efforts by benefit 
withholding for any outstanding debt, or balance of a debt, not 
previously waived. 

[37] There is no limit on the withholding from old age and survivors 
(retiree) monthly benefits. Withholding from SSI is limited to the 
lesser of the amount of the benefit or 10 percent of the beneficiary's 
monthly income. 42 U.S.C. § 1320b-17(b). 

[38] Case file data were pulled between September 2010 and November 
2010, or for roughly the first quarter of fiscal year 2011. Most of 
these criteria were established for and by the Department of the 
Treasury, which administers external collection, per the Debt 
Collection Improvement Act of 1996, Pub. L. No. 104-134. § 31001, 110 
Stat. 1321, 1321-358 - 1321-380. Treasury determines the criteria for 
collection from debts referred from creditor agencies; therefore, 
amounts recovered by this means are beyond SSA's control. 

[39] In the 60 cases we reviewed, we analyzed the number of years it 
would take to repay the initial work-related overpayment debt, less 
any initial waivers or one-time payments, given the first recurring 
monthly payment established in the ROAR record. 

[40] DI does not have a cap on program benefit withholding to recover 
DI overpayments. Instead, debt specialists set withholding amounts on 
a case-by-case basis. 

[41] Effective as of April 2007. 

[42] As of May 2011, Treasury reported that Louisiana, West Virginia, 
Wisconsin, and the District of Columbia had passed the necessary 
legislation to participate in the state repayment program and were 
formulating implementation plans. 

[43] Recovery of Delinquent Debts--Treasury Offset Program 
Enhancements, 76 Fed. Reg. 11402. 

[44] Among the four processing centers we visited, the median time 
spans from alert to beginning work on the CDR was 157, 165, 199, and 
214 days. 

[45] This is illustrative of how long overpayments occurred even 
before the case was flagged for review by the enforcement operation. 

[46] The NDNH contains quarterly state wage information, which is more 
recent than the annual wage information that SSA obtains through its 
current IRS data match. SSA currently uses the NDNH to periodically 
monitor the earnings of SSI recipients. 

[47] Social Security Administration, Office of the Inspector General, 
Supplemental Security Income Overpayments, A-01-04-24022 (2004). 

[48] GAO, Social Security Administration: Cases of Federal Employees 
and Transportation Drivers and Owners Who Fraudulently and/or 
Improperly Received SSA Disability Payments, [hyperlink, 
http://www.gao.gov/products/GAO-10-444] (Washington, D.C.: June 25, 
2010). 

[49] SSA begins measuring this target from the time staff begin work 
on developing the case through the time the cessation decision was 
made. For Processing Center 7 in Baltimore, the target for completing 
development of cases is at least 99 percent within 365 days by 
September 30, 2011. 

[50] We measured processing time from the time the IRS alert was 
generated through the time the cessation decision was made. 

[End of section] 

GAO's Mission: 

The Government Accountability Office, the audit, evaluation and 
investigative arm of Congress, exists to support Congress in meeting 
its constitutional responsibilities and to help improve the performance 
and accountability of the federal government for the American people. 
GAO examines the use of public funds; evaluates federal programs and 
policies; and provides analyses, recommendations, and other assistance 
to help Congress make informed oversight, policy, and funding 
decisions. GAO's commitment to good government is reflected in its core 
values of accountability, integrity, and reliability. 

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through GAO's Web site [hyperlink, http://www.gao.gov]. Each 
weekday, GAO posts newly released reports, testimony, and 
correspondence on its Web site. To have GAO e-mail you a list of newly 
posted products every afternoon, go to [hyperlink, http://www.gao.gov] 
and select "E-mail Updates." 

Order by Phone: 

The price of each GAO publication reflects GAO’s actual cost of
production and distribution and depends on the number of pages in the
publication and whether the publication is printed in color or black and
white. Pricing and ordering information is posted on GAO’s Web site, 
[hyperlink, http://www.gao.gov/ordering.htm]. 

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or
TDD (202) 512-2537. 

Orders may be paid for using American Express, Discover Card,
MasterCard, Visa, check, or money order. Call for additional 
information. 

To Report Fraud, Waste, and Abuse in Federal Programs: 

Contact: 

Web site: [hyperlink, http://www.gao.gov/fraudnet/fraudnet.htm]: 
E-mail: fraudnet@gao.gov: 
Automated answering system: (800) 424-5454 or (202) 512-7470: 

Congressional Relations: 

Ralph Dawn, Managing Director, dawnr@gao.gov: 
(202) 512-4400: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7125: 
Washington, D.C. 20548: 

Public Affairs: 

Chuck Young, Managing Director, youngc1@gao.gov: 
(202) 512-4800: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7149: 
Washington, D.C. 20548: