oversight

Supplemental Security Income: SSA Could Enhance Its Ability to Detect Residency Violations

Published by the Government Accountability Office on 2003-07-29.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

             United States General Accounting Office

GAO          Report to the Chairman, Subcommittee
             on Human Resources, Committee on
             Ways and Means, House of
             Representatives

July 2003
             SUPPLEMENTAL
             SECURITY INCOME

             SSA Could Enhance
             Its Ability to Detect
             Residency Violations




GAO-03-724
                                               July 2003


                                               SUPPLEMENTAL SECURITY INCOME

                                               SSA Could Enhance Its Ability To Detect
Highlights of GAO-03-724, a report to the      Residency Violations
Chairman, Subcommittee on Human
Resources, Committee on Ways and
Means, House of Representatives




The Supplemental Security Income               Overpayments resulting from residency violations totaled about $118 million
(SSI) program paid about $35                   between 1997 and 2001. However, this figure, which represents only
billion in benefits to about 6.8               violations detected by SSA, likely understates the true level of the problem.
million recipients in 2002. In                 Additionally, the extent of violations appears to vary by geographic region,
recent years, the Social Security              with overpayments being more prevalent in several large metropolitan areas.
Administration (SSA) has identified
                                               GAO found that 54 percent of all overpayments detected by SSA during this
a general increase in the amount of
annual overpayments made to (1)                period occurred in just 15 counties. In addition, GAO found that recipients
individuals who are found to have              born outside the United States accounted for at least 87 percent of all
violated program residency                     residency overpayments.
requirements or (2) recipients who
leave the United States and live               SSA’s ability to detect and deter residency violations is impeded by three
outside the country for more than              kinds of weaknesses. First, the agency relies heavily on self-reported
30 consecutive days without                    information from recipients to determine domestic residency, often without
informing SSA. This problem has                independently verifying such information. Second, SSA makes insufficient
caused concern among both                      use of existing tools to detect violations, such as its “risk analysis” system,
program administrators and policy              redeterminations, and home visits. Finally, the agency has not adequately
makers. As such, GAO was asked
                                               pursued independent sources of information from other federal agencies or
to determine what is known about
the extent to which SSI benefits are           private organizations to detect nonresidency of SSI recipients. GAO
improperly paid to individuals who             recognizes that the SSI program is complex to administer, and residency
are not present in the United States           requirements are particularly difficult to enforce because they can
and to identify any weaknesses in              necessitate time-consuming, labor-intensive verification checks, such as
SSA’s processes and policies that              home visits. However, SSA has not employed a systematic, comprehensive
impede the agency’s ability to                 approach to this problem that would allow the agency to use its available
detect and deter residency                     systems and procedures more efficiently and reduce the program’s exposure
violations.                                    to additional violations.
                                               Top 15 Counties for SSI Residency Overpayments (1997-2001)


GAO is making recommendations
to the Commissioner of Social
Security that will allow the agency                           California
to make optimal use of existing                               San Francisco
tools and new data sources to                                 Alameda
better detect potential residency                             San Mateo                                      New York
                                                              Santa Clara                                    Queens
violators. Social Security generally                          Los Angeles
                                                                                                  Illinois                New
                                                                                                  Cook       New York
agreed with GAO’s                                             Riverside                                      Kings        Jersey
recommendations, but identified                               Orange                                         Bronx        Passaic
some challenges to their                                      San Diego
implementation.



                                                                                                                        Florida
                                                                                                                        Dade
www.gao.gov/cgi-bin/getrpt?GAO-03-724.
To view the full report, including the scope
                                               Source: GAO.
and methodology, click on the link above.
For more information, contact Robert E.
Robertson (202) 512-7215 or
RobertsonR@gao.gov.
Contents


Letter                                                                                             1
                       Results in Brief                                                            3
                       Background                                                                  4
                       SSA Detected Overpayments of $118 Million for Residency
                         Violations over 5 Years, but More May Go Undetected                       6
                       Reliance on Self-Reported Information and Other Vulnerabilities
                         Impede SSA’s Ability to Detect and Deter Violations                     14
                       Conclusions                                                               23
                       Recommendations                                                           24
                       Agency Comments and Our Evaluation                                        25

Appendix I             Scope and Methodology                                                      27



Appendix II            Comments from the Social Security Administration                           31
                       GAO Comments                                                              36

Appendix III           GAO Contacts and Staff Acknowledgments                                     37
                       GAO Contacts                                                              37
                       Staff Acknowledgments                                                     37

Related GAO Products                                                                              38



Tables
                       Table 1: Variables That Indicate a Recipient May Potentially Be Out
                                of the United States                                             29
                       Table 2: Odds Ratios for the Variables in Our Logistical Regression
                                Analysis                                                         30


Figures
                       Figure 1: Top 15 Counties for SSI Residency Overpayments,
                                1997-2001                                                          9
                       Figure 2: Top 15 Countries of Origin for SSI Residency
                                Overpayments, 1997-2001                                          12




                       Page i                                GAO-03-724 Supplemental Security Income
Abbreviations

ATM               automated teller machines
CMS               Centers for Medicare and Medicaid Services
DHS               Department of Homeland Security
INS               Immigration and Naturalization Service
OIG               Office of Inspector General
SSA               Social Security Administration
SSI               Supplemental Security Income




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. It may contain copyrighted graphics, images or other materials.
Permission from the copyright holder may be necessary should you wish to reproduce
copyrighted materials separately from GAO’s product.




Page ii                                        GAO-03-724 Supplemental Security Income
United States General Accounting Office
Washington, DC 20548




                                   July 29, 2003

                                   The Honorable Wally Herger
                                   Chairman, Subcommittee on Human Resources
                                   Committee on Ways and Means
                                   House of Representatives

                                   Dear Mr. Chairman:

                                   The Supplemental Security Income (SSI) program is the nation’s largest
                                   cash assistance program for the poor. The program paid about $35 billion
                                   in benefits to about 6.8 million aged, blind, and disabled recipients in
                                   2002.1 Most SSI recipients are also eligible to receive medical benefits
                                   under the Medicaid program.2 Benefit eligibility and payment amounts for
                                   the SSI population are determined by complex and often difficult to verify
                                   factors such as an individual’s living arrangements, including whether an
                                   individual resides in the United States. Individual circumstances such as
                                   permanent residence may change frequently, requiring staff to regularly
                                   reassess recipients’ eligibility for benefits. Thus, the SSI program tends to
                                   be difficult, time-consuming, and labor-intensive to administer. The
                                   program’s complexity has also made it susceptible to overpayments.3 In
                                   recent years, the Social Security Administration (SSA) has identified a
                                   general increase in the amount of annual overpayments made to (1)
                                   individuals who are found to have violated program residency
                                   requirements or (2) recipients who leave the United States4 and live
                                   outside the country for more than 30 consecutive days without informing




                                   1
                                    This figure includes $30 billion in federal funds and $4.7 billion in state funded
                                   supplemental funds.
                                   2
                                     In a 1999 report, we estimated that the combined federal cost for SSI and Medicaid
                                   benefits is $122,000 per recipient over the next 10 years. U.S. General Accounting Office,
                                   Supplemental Security Income: Additional Actions Needed to Reduce Program
                                   Vulnerability to Fraud and Abuse, GAO/HEHS-99-151 (Washington, D.C.: Sept. 15, 1999).
                                   3
                                    In 2001, outstanding SSI debt and newly detected overpayments for the year totaled
                                   $4.7 billion.
                                   4
                                    For SSI eligibility purposes, the United States is defined as one of the 50 states, the District
                                   of Columbia, and the Northern Mariana Islands. Residents of Puerto Rico are not eligible
                                   for SSI.



                                   Page 1                                            GAO-03-724 Supplemental Security Income
SSA. This problem, which we refer to as a “residency violation,” has
caused concern among both program administrators and policy makers.5

Because of your interest in the SSI program’s potential vulnerability to
residency violations, you asked us to (1) determine what is known about
the extent to which SSI benefits are improperly paid to individuals who
are not present in the United States and (2) identify any weaknesses in
SSA’s processes and policies that impede the agency’s ability to detect and
deter residency violations. To answer these questions, we reviewed SSI
performance data, prior reports by SSA and its Office of Inspector General
(OIG), and our prior work on the SSI program. We analyzed SSI payment
data over a 5-year period between 1997 and 2001 and examined cases in
which SSA identified recipients who were residing outside the country. In
addition, we developed and tested a statistical model to help predict
whether certain SSI recipients were more likely than others to have
residency violations. Finally, we conducted in-depth interviews with more
than 150 management and line staff from SSA’s headquarters; its regional
offices in Atlanta, Chicago, Dallas, New York, and San Francisco; 17 field
offices; as well as officials from other federal agencies, including the
former Immigration and Naturalization Service (INS)6 and the Centers for
Medicare and Medicaid Services. During our meetings, we documented
management and staff views on how extensive residency violations are in
the SSI program; the effectiveness of current procedures and processes for
detecting and preventing residency violations; and potential improvements
to existing program processes, policies, and systems. We performed our
work from September 2002 through May 2003 in accordance with
generally accepted government auditing standards.




5
 For purposes of this report, we do not distinguish between two related statutory
requirements for SSI eligibility—presence in the United States and United States residence.
The presence requirement states that a recipient is not eligible for SSI if the recipient is
outside the country for 30 consecutive days or more. However, a recipient may retain
eligibility if they are temporarily outside the country for less than 30 consecutive days. The
residency requirement states that an individual must establish a dwelling in the United
States with the intent to continue to live in the country. The rules for determining whether
an individual meets the physical presence requirement are outlined at 20 C.F.R. § 416.215
and § 416.1327. The rules for determining whether an individual meets the residency
requirement are outlined at 20 C.F.R. § 416.202(b), § 416.1329, and § 416.1603.
6
The INS is currently being divided into three separate bureaus within the Department of
Homeland Security.




Page 2                                          GAO-03-724 Supplemental Security Income
                   SSA detected overpayments of $118 million for residency violations
Results in Brief   between 1997 and 2001, but interviews with OIG and agency officials
                   suggest that the agency detects only a portion of the violations that occur
                   each year, at least in some parts of the country. Prior studies and special
                   projects by SSA and its OIG show that residency violations are a pervasive
                   problem in some field offices. Additionally, the extent of violations
                   appears to vary by geographic region, with overpayments being more
                   prevalent in several large metropolitan areas. In particular, we found that
                   54 percent of all overpayments detected by SSA during this period
                   occurred in just 15 counties in five states—California, Florida, Illinois,
                   New Jersey, and New York. Finally, we found that recipients born outside
                   the United States accounted for at least 87 percent of all residency
                   overpayments.

                   Three kinds of weaknesses impede SSA’s ability to detect and deter
                   residency violations: The agency (1) relies heavily on self-reported
                   information from recipients to verify domestic residency; (2) makes
                   insufficient use of its existing tools to detect violations; and (3) has not
                   pursued independent sources of information to detect nonresidency of SSI
                   recipients. First, in asking SSI recipients about their current residence,
                   field staff often rely on recipients’ own assertions and may accept only
                   minimal documentation from them, such as rent receipts and statements
                   from neighbors or clergy. Recipients who wish to misreport their
                   residency can manipulate such documents. Second, the agency makes
                   limited use of tools at its disposal to detect possible violators. For
                   example, while SSA routinely employs a risk analysis system to identify
                   SSI recipients who are more likely to incur overpayments, it does not use
                   this tool to specifically consider and target potential residency violators. A
                   model we developed and tested to predict residency violations–that
                   includes characteristics such as prior violations, use of post office boxes,
                   and birth outside the United States–suggests that SSA could, in fact, make
                   better use of its risk analysis system to detect and prevent residency
                   violations. Additionally, while we found that home visits are being utilized
                   as part of redetermination reviews in some of the offices we visited and
                   are cost-effective, SSA has not systematically implemented this tool in
                   other offices. Other tools the agency has made only limited use of are
                   monetary penalties and administrative sanctions, such as loss of benefits.
                   Several field staff told us they rarely employ either, because monetary
                   penalties are too small to deter potential violators, and administrative
                   sanctions involve a time-consuming and cumbersome process. Finally,
                   SSA has not adequately pursued the use of independent, third-party data
                   such as recipient bank account information to help detect residency
                   violations. Although SSA is currently working with an independent


                   Page 3                                  GAO-03-724 Supplemental Security Income
             contractor to obtain access to SSI recipients’ financial data, the agency
             plans to use the information only to verify their financial resources. It does
             not plan to use the information to detect those who may be living and
             making financial transactions outside the United States for extended
             periods of time.

             Detecting residency violations in a cost-effective manner presents
             particularly difficult challenges. However, we believe that there are a
             number of opportunities that SSA can take advantage of, and accordingly,
             we are recommending that the Commissioner of Social Security direct the
             agency to make optimal use of existing tools while exploring new data
             sources to better detect potential residency violators.

             Social Security generally agreed with our recommendations, but also
             identified some challenges to their implementation.


             SSI provides financial assistance to people who are age 65 or older, blind
Background   or disabled, and who have limited income and resources. The program
             provides individuals with monthly cash payments to meet basic needs for
             food, clothing, and shelter. In 2002, about 6.8 million recipients were paid
             about $35 billion in SSI benefits.

             Individuals may apply for SSI benefits at any of about 1,300 SSA field
             offices. During the initial interview, SSA staff solicit information on
             applicants’ financial situation and the disability being claimed. Applicants
             are required to report any information that may affect their eligibility for
             benefits, such as income, resources, and their living arrangements
             (including current residence). Similarly, once individuals receive SSI
             benefits, they are required to report changes in their address or residence
             to SSA in a timely manner. The Social Security Act (Section 1614
             (a)(1)(B)(i)) requires that an individual be a resident of the United States
             to be eligible for SSI payments. SSA guidelines define a resident of the
             United States as a person who has established a dwelling in the United
             States with the intent to live in the country. Section 1611(f) of the act also
             stipulates that no individual is eligible for SSI payments for any month
             during all of which the individual is outside the United States. Further, an
             individual who is outside the United States for 30 consecutive days cannot
             be eligible for SSI benefits until he or she has been back in the United
             States for 30 days. Recipients who fail to establish residency in accordance
             with SSI program guidelines, or do not report absences of 30 consecutive
             days or more, may be subject to overpayments, monetary penalties, and
             administrative sanctions such as suspension of benefits. Similarly, SSI


             Page 4                                  GAO-03-724 Supplemental Security Income
recipients who become ineligible for SSI benefits because they violate SSI
residency guidelines may also be ineligible to receive Medicaid benefits.

To a significant extent, SSA depends on program applicants and current
recipients to accurately report important eligibility information. To verify
this information, SSA may use computer matches to compare SSI records
against recipient information in records of third parties such as other
federal agencies. SSA also periodically conducts “redetermination”
reviews to verify important eligibility factors such as income and
resources to determine whether recipients remain eligible for benefits
after the initial assessment. Recipients are reviewed at least every 6 years,
but reviews may be conducted more frequently if SSA determines that
changes in eligibility are likely. To determine which recipients should
receive more frequent reviews, SSA uses a risk analysis system7 to identify
recipients who may be more likely to incur overpayments. Those identified
as “high-risk” generally have a redetermination conducted at least annually
by SSA field office staff who contact the recipient in person or by phone,
while lesser-risk redetermination reviews (such as those designated “low-
risk”) may only be conducted once every several years by mail. In
addition, SSA uses “limited issue” redetermination reviews to review a
specific factor that may affect a recipient’s eligibility, such as income or
current residence. These reviews tend to be less time-consuming and labor
intensive for field staff to perform than “full” redetermination reviews,
which often require an examination of numerous eligibility factors.

In recent years, detected overpayments from residency violations
increased from about $13.7 million in 1995 to about $22 million in
2001, reaching a high of almost $27 million in calendar year 2000.8 In
addition, the number of individuals SSA detected receiving SSI benefits
while outside the United States increased from about 44,000 recipients in
1997 to almost 49,000 recipients in 2001.




7
 We use the term “risk analysis system” to describe SSA’s statistical redetermination
profiling system used to identify SSI recipients who are more likely to incur overpayments.
8
 SSA reported about $13.7 and $15.5 million in residency overpayments for 1995 and 1996,
respectively. However, the 1995 and 1996 total dollars of residency overpayments were
reported on a fiscal year basis. SSA reported overpayment data for 1997 to 2001 on a
calendar year basis for our report. While a direct comparison cannot be made because of
differences in reporting periods, the data show that residency overpayments have been
increasing over time.




Page 5                                         GAO-03-724 Supplemental Security Income
                               SSA detected overpayments of $118 million for residency violations
SSA Detected                   between 1997 and 2001, but interviews with OIG and SSA officials suggest
Overpayments of                that the agency detects only a portion of the violations that occur each
                               year, at least in some parts of the country. Special initiatives of limited
$118 Million for               duration conducted by SSA and its OIG have uncovered additional
Residency Violations           residency overpayments. According to our own analysis of SSA’s data,
                               residency overpayments appear to vary by geographic region, with the
over 5 Years, but More         majority of overpayments having been detected in several large
May Go Undetected              metropolitan areas. Finally, we determined that most of the overpayments
                               detected during this period were attributable to recipients who were born
                               outside the United States.

Residency Violations May       SSA detected an average of about 46,000 recipient residency violations
Be More Prevalent than         annually between 1997 and 2001, resulting in $118 million in
SSA Currently Detects          overpayments. While SSA’s data show that less than 1 percent of all SSI
                               recipients violate residency requirements annually, SSA field staff and OIG
                               officials suggest that the problem may be more prevalent. For example,
                               staff and managers from field offices in several SSA regions estimated that
                               over 40 percent of all recipients are in violation of residency requirements
                               at some point. A small number of staff told us that as much as 90 percent
                               of recipients served by their office may be involved in such violations.
                               Other staff said that while the problem is more pervasive than SSA
                               currently detects, it is difficult to estimate the true extent of the problem
                               because the agency relies heavily on recipients to self-report absences
                               from the country. In addition, a number of OIG officials we spoke with
                               told us that residency violations are significantly higher than SSA currently
                               detects; one official familiar with this problem estimated that as much as
                               70 percent of SSI recipients in some areas close to the southern border of
                               the United States improperly receive benefits outside the country.

                               Over the past few years, SSA and its OIG have initiated a number of
                               projects that uncovered additional residency violations and overpayments.
                               Although there is no empirical data to determine the true level of
                               residency violations nationwide, these studies have estimated that
                               residency violations in certain regions of the country may represent as
                               much as 26 percent of SSI cases in those areas. These initiatives, which
                               were limited in duration and were performed within specific geographic
                               areas, include the following:

                           •   A 1997 SSA and OIG joint study of SSI residency used home visits in
                               southern California to identify potential residency violations. The study
                               concluded that about 25 percent of SSI recipients in 1 field office were
                               living outside of the country. The study also determined that 47 percent of


                               Page 6                                 GAO-03-724 Supplemental Security Income
    SSI recipients from this field office could not be located at their reported
    residence, an indication that they may be violating residency requirements.

•   A 1998 OIG eligibility study in El Paso, Texas, found that about 26 percent
    of recipients investigated were violating residency requirements. This
    project identified a total of about $3 million in residency overpayments.

•   In 1998 and 1999, joint SSA/OIG studies examined 32,641 recipients in New
    York and California who had not used their Medicaid benefits for at least
    1 year.9 Using redetermination reviews, these studies found that
    1,281 recipients (about 4 percent) were living outside the United States.10

•   A 2002 SSA residency verification project in 5 South Florida field offices
    used a targeted sample of 750 noncitizen recipients that uncovered a total
    of over $107,000 in additional residency overpayments. Staff performed
    special redetermination reviews in which recipients were required to
    produce a valid passport as proof of continuing residency in the United
    States and found 46 recipients (6 percent) violating residency
    requirements.

•   A 2002 OIG address verification project in New York uncovered
    205 recipients violating SSI residency requirements resulting in a total of
    about $262,000 in overpayments. These recipients were found to be
    receiving both SSI and Title II Social Security benefits at addresses in two
    different countries. The study also found that SSA’s automated controls
    and special projects did not identify SSI recipients who had their benefits
    direct deposited into banks in Puerto Rico or the U.S. Virgin Islands.

    While the results of these projects suggest that the problem is more
    pervasive than the 1 percent of recipients that SSA’s data show, SSA has
    not systematically implemented similar projects in other areas that might
    benefit from these efforts.




    9
     The rationale for targeting these cases was that financially needy individuals who were
    aged or disabled are likely to use Medicaid services on a regular basis. Thus, SSI recipients
    who have not used Medicaid for long periods of time may have left the United States or
    died.
    10
     These studies considered the effect of only one potential indicator on residency
    violations—Medicaid nonutilization.




    Page 7                                          GAO-03-724 Supplemental Security Income
Many Violations Are   Our analysis of SSA’s data also shows that overpayments due to residency
Geographically        violations are more prevalent in a number of large metropolitan areas. For
Concentrated          example, overpayments from violations detected in Los Angeles County,
                      California, represented 10.5 percent of the nation’s SSI residency
                      overpayments between 1997 and 2001. Overall, our analysis indicates that
                      just 15 counties in five states—California, Florida, Illinois, New Jersey, and
                      New York—accounted for 54 percent of all residency overpayments
                      detected by SSA during this period. (See fig. 1.) In addition to Los Angeles
                      County, other counties with a significant percentage of SSI residency
                      overpayments include Queens County, New York (5.2 percent); New York
                      County, New York (5.0 percent); Kings County, New York (4.8 percent);
                      San Diego County, California (4.1 percent); and Bronx County, New York
                      (3.5 percent). Moreover, of approximately 3,000 counties in the United
                      States, 50 accounted for 77 percent of all residency overpayments detected
                      by SSA during this time.




                      Page 8                                  GAO-03-724 Supplemental Security Income
Figure 1: Top 15 Counties for SSI Residency Overpayments, 1997-2001



                                                                                 Illinois      New Jersey    New York
                                                                                 Cook          Passaic       Queens
                                                                                                             New York
                                                                                                             Kings
                                                                                                             Bronx




               California
               San Francisco
               Alameda
               San Mateo
               Santa Clara
               Los Angeles
               Riverside                                                                                        Florida
               Orange                                                                                           Dade
                                                                                 Overpayments
               San Diego                                 County                  (millions of dollars)
                                                         Los Angeles, Calif.     $ 12.4
                                                         Queens, N.Y.            $ 6.2
                                                         New York, N.Y.          $ 5.9
                                                         Kings, N.Y.             $ 5.7
                                                         San Diego, Calif.       $ 4.9
                                                         Bronx, N.Y.             $ 4.2
                                                         Orange, Calif.          $ 3.9
                                                         Santa Clara, Calif.     $ 3.9
                                                         Dade, Fla.              $ 3.3
                                                         Cook, Ill.              $ 2.9
                                                         San Francisco, Calif.   $ 2.7
                                                         Alameda, Calif.         $ 2.5
                                                         Passaic, N.J.           $ 2.2
                                                         San Mateo, Calif.       $ 1.6
                                                         Riverside, Calif.       $ 1.5

Source: GAO.




                                       Page 9                                               GAO-03-724 Supplemental Security Income
Most Overpayments Were      SSA’s data also show that individuals born outside the United States
Made to Recipients Born     accounted for at least 87 percent of all SSI residency overpayments
outside the United States   between 1997 and 2001.11 Residency overpayments were most common
                            among recipients who were born in Latin America, the Caribbean, and
                            South/Southeast Asia, but included other areas as well, such as the Middle
                            East. (See fig. 2.) Recipients from the Philippines accounted for the
                            greatest amount of residency violations or $24 million of all SSI residency
                            overpayments during this period. SSA data also show that recipients from
                            just 14 countries and one United States territory accounted for about
                            73 percent of all residency overpayments during this period. In addition to
                            the Philippines (20.2 percent), these include the Dominican Republic
                            (12.3 percent), Mexico (7.6 percent), Puerto Rico (7.5 percent), India
                            (7.1 percent), and Iran (3.4 percent). Moreover, a prior study by SSA’s OIG
                            found that SSI residency violations are more prevalent among recipients
                            born outside the United States than for native-born citizens. Specifically,
                            the OIG found that more than 20 percent of the recipients born outside the
                            United States had periods of ineligibility because of absences from the
                            United States, compared with 0.2 percent for native-born recipients.




                            11
                              The percentage of total residency overpayments attributed to recipients born outside of
                            the United States may be higher than 87 percent because SSA could not identify a specific
                            country of birth for recipients that represent about $10 million in SSI overpayments.




                            Page 10                                       GAO-03-724 Supplemental Security Income
Page 11   GAO-03-724 Supplemental Security Income
Figure 2: Top 15 Countries of Origin for SSI Residency Overpayments, 1997-2001




                                                                                                 Mexico
                                                                                                 Haiti
                                                                                                 Dominican Republic
                                                                                                 Puerto Ricoa
                                                                                                 El Salvador
                                                                                                 Columbia
                                                                                                 Peru




                                                      Overpayments
                            Country of origin         (millions of dollars)
                            Dominican Republic        $ 14.5
                            Mexico                    $ 9.0
                            Puerto Rico               $ 5.9
                            Haiti                     $ 2.5
                            Columbia                  $ 1.6
                            El Salvador               $ 1.3
                            Peru                      $ 1.2




Source: GAO.
                                         a
                                             Puerto Rico is a United States territory.




                                         Page 12                                         GAO-03-724 Supplemental Security Income
                                                        Korea
                                                        China
                                                        Taiwan
                                                        Vietnam
                                                        Philippines




                               India
                               Pakistan
                               Iran




                      Overpayments
 Country of origin   (millions of dollars)
 Philippines         $ 24.0
 India               $ 8.4
 Iran                $ 4.1
 Vietnam             $ 3.5
 China               $ 3.4
 Korea               $ 2.0
 Pakistan            $ 1.3
 Taiwan              $ 1.3




Page 13                                      GAO-03-724 Supplemental Security Income
                              SSA’s ability to detect and deter residency violations is impeded by three
Reliance on Self-             kinds of weaknesses: dependence on self-reported information by clients,
Reported Information          insufficient use of existing compliance tools, and a failure to pursue
                              independent data sources for its verifications. First, the agency relies
and Other                     heavily on self-reported information from recipients to determine
Vulnerabilities Impede        domestic residency, often without independently verifying such
                              information. Second, to detect and deter residency violations, SSA makes
SSA’s Ability to Detect       insufficient use of its existing tools for program integrity, such as its risk
and Deter Violations          analysis system to screen for high-risk cases. To test the feasibility of
                              using the agency’s system to screen for residency violations, we developed
                              and tested, with some success, a statistical model of factors that may be
                              associated with such violations. SSA has also not made optimal use of
                              redetermination reviews, home visits, monetary penalties, and
                              administrative sanctions. Finally, the agency has not employed the use of
                              independent data sources from other federal agencies or private
                              organizations to detect nonresidency of SSI recipients.


SSA Relies Heavily on Self-   SSA relies heavily on self-reported information, such as documents and
Reported Information That     statements from recipients to establish proof of U.S. residency. According
Can Be Manipulated            to SSA and OIG officials, however, this practice increases the SSI
                              program’s vulnerability to residency fraud and abuse. Our prior work has
                              shown that about 77 percent12 of all payment errors in the SSI program
                              were attributable to recipients who do not comply with reporting
                              requirements.13 In our current review, about half of SSA field staff we
                              interviewed reported that they rely on recipients to self-report important
                              information with respect to travel outside the United States. SSI program
                              guidelines generally direct SSA staff to accept recipients’ assertions
                              concerning residency unless they have reason to question the accuracy of
                              their statements. SSA staff also have discretion with respect to the types of
                              documents they can accept to confirm that a recipient resides at a given
                              address. For example, program guidelines direct field staff to obtain a
                              combination of two or more documents as proof of initial residency.
                              Acceptable documentation includes such things as rent receipts, utility
                              bills, driver licenses, pay stubs, or mail addressed to the recipient. If SSA
                              field staff have reason to believe that a recipient has been outside the


                              12
                                   This figure represents data from fiscal years 1991 through 1995.
                              13
                               U.S. General Accounting Office, Supplemental Security Income: Action Needed on Long-
                              Standing Problems Affecting Program Integrity, GAO/HEHS-98-158 (Washington, D.C.:
                              Sept. 14, 1998).




                              Page 14                                            GAO-03-724 Supplemental Security Income
country for more than 30 days, or information in SSA records conflicts
with recipients’ statements, they may request additional documentation
such as an airline ticket, passport (or similar evidence that establishes
date of entry into the United States), or a signed statement from one or
more U.S. residents, such as neighbors, clergy, or others who may have
knowledge of the individual’s whereabouts. In general, program guidelines
do not require field staff to perform any additional verification steps to
establish recipients’ residency during initial or post-entitlement eligibility
reviews.14

We were also told that some of the documents accepted by SSA as proof of
residence are subject to manipulation or forgery. For example, staff in 1
field office noted that documents such as rent receipts can be purchased
from a local drugstore and easily forged. Other field staff said that
statements from neighbors could be falsified or manipulated to support
assertions that an individual has not traveled outside the country. Field
staff also reported that recipients may use multiple passports in order to
conceal extended stays outside the country. For example, staff in two SSA
regions we visited said that SSI recipients sometimes use a foreign
passport to exit and reenter the country while maintaining a separate,
“clean” U.S. passport for evidence of continuing residency. One field office
staff reported that some recipients have even paid foreign customs
officials not to stamp their passport to conceal evidence that they were
outside the country.

Given the agency’s heavy reliance on self-reported information, we found
that SSA field staff often relied on their personal experience, judgment,
and ad hoc interviewing procedures to detect potential residency
violations. In particular, SSA field staff look for inconsistencies in
recipient’s statements or their inability to answer simple questions about
where they live. For example, recipients may be asked about the names of
people living in their household, or basic facts about their neighborhood
such as the location of a well-known landmark. Field staff may also ask
whether a recipient owns property outside the United States. Questionable



14
 SSI program guidance allows field staff to use home visits in selected circumstances, such
as in response to a report from a third party that a recipient is outside the United States. In
addition, home visits may be employed if a recipient fails to provide information requested
by SSA staff, or if a recipient does not respond to letters and/or telephone calls from staff
asking them to appear at the local office. However, program guidelines give field office
managers discretion in determining when to use home visits and allow them to take into
consideration factors such as the safety of staff who perform such visits.




Page 15                                          GAO-03-724 Supplemental Security Income
                        or inconsistent answers to such questions may result in requests to
                        provide additional documentation. However, the ability of staff to
                        effectively identify violators often depends on the experience and
                        persistence of individual staff. This is particularly true in the case of
                        recipients who are “coached” by advocacy groups or others to provide
                        false information to SSA in order to obtain or retain SSI benefits. For
                        example, one official told us that they recently identified an SSI applicant
                        who apparently was coached to provide false information to SSA
                        regarding her residence in the United States. Upon further questioning, the
                        recipient admitted that she was not residing continuously in the United
                        States and had misreported her residency to SSA in order to obtain SSI
                        benefits. In addition, field staff who were familiar with this problem told
                        us that they sometimes look for suspicious documents that recipients may
                        reveal during the course of an office visit, such as a foreign driver license
                        or foreign voter registration card. One manager in a field office close to the
                        Mexican border also noted that some field staff check address listings in
                        Mexican telephone books or check the parking lot for cars with foreign
                        license plates.

                        Our review also found that the procedures for documenting recipients’
                        residency vary widely among the offices we visited. In particular, the
                        number and types of evidentiary documents requested by staff differed
                        across the field offices we visited. While staff in several offices reported
                        that they often request only the most basic documentation required by SSI
                        program guidelines, staff in other offices told us that they routinely ask for
                        additional documentation for recipients. For example, some field staff we
                        interviewed noted that they ask recipients to provide a second passport or
                        other documents (such as travel documents from foreign consular offices)
                        to determine whether the individual has been outside the country for more
                        than 30 days. While these steps are not required by SSI program
                        guidelines, some field staff reported that they have been effective in
                        identifying potential violators and deterring future violations. SSA staff
                        reported a number of reasons for different documentation requirements
                        such as variance in individual office policies, personal preferences based
                        on experience, time pressures to complete cases, and the inability to
                        effectively verify supplied documentation.


SSA Does Not Fully      SSA has not made optimal use of several tools that could be used to detect
Exploit Its Tools for   residency violations. These include its “risk analysis system” for screening
Detecting Program       cases more likely to result in overpayments, its “redetermination reviews”
                        of recipients’ eligibility, and home visits to verify recipients’ whereabouts.
Violations              Given its current focus on other types of program violations such as


                        Page 16                                 GAO-03-724 Supplemental Security Income
excess income or resources, some staff told us that SSA’s risk analysis
system is not entirely effective at identifying residency violators. SSA has
used statistical risk analysis techniques for many years in the SSI program
to identify recipients who are more likely to be overpaid. Since SSA lacks
adequate staff resources to conduct an annual redetermination for every
recipient, it routinely screens for and targets those participants who are
most likely to have a change in their eligibility status or benefit amount.15
Despite the proven effectiveness of its risk analysis system to help the
agency detect cases with highest potential for overpayments, SSA has not
used this tool to specifically identify residency violations. In fact, a
number of field staff told us that, in their experience, residency violations
frequently occur among SSI recipients who are designated as medium or
low risk for payment errors by SSA’s system. The agency may not discover
these violations for several years (if at all) unless it detects a change in a
recipient’s circumstances that causes the individual to be designated as
high risk.

To determine whether it is possible for SSA to target potential residency
violators more effectively using its existing systems, we developed and
tested a statistical model of factors possibly associated with residency
violations.16 Based on our field work and prior SSA and OIG studies, we
selected the following factors for testing: recipients born outside the
United States, prior residency violations, payments made to post office
boxes, direct deposit payments, and lack of response to agency inquiries
or recipients with unknown whereabouts.17 Using this model as a screen,
we identified all recipients who were currently in violation of residency



15
  SSA’s risk analysis system incorporates about 48 different characteristics—or variables—
to help the agency determine which recipients will be selected for annual redetermination
reviews. Recipients identified as being at higher risk for overpayments are designated as
high error profile cases and may be subject to more frequent reviews that entail personal
contact with SSA field office staff. Those recipients identified as being less likely to incur
an overpayment are designated as medium or low error profile and may only receive a
redetermination conducted by mail rather than in person. Some low error profile cases are
only examined once every 6 years.
16
 The variables used in our model are not an exhaustive list of potential variables that SSA
could use in its risk analysis system. They represent just a few of the characteristics that
were frequently cited by prior reviews as well as SSA and OIG staff as potentially good
predictors of residency violations.
17
 Another variable frequently cited as a potential indicator of residency violations—
Medicaid nonutilization—was not included in our model because SSA does not currently
have automated data on Medicaid nonusage by SSI recipients. However, SSA is negotiating
access to such data with the Centers for Medicare and Medicaid Services.




Page 17                                         GAO-03-724 Supplemental Security Income
requirements as of April 2003.18 We found that recipients born outside the
United States—noncitizens as well as naturalized citizens—were more
than 40 times as likely to be violating residency requirements than were
native-born recipients. Similarly, recipients with prior residency violations
were about 10 times as likely to be current violators compared with
recipients who have no prior violations. We also found that recipients who
used post office boxes were somewhat more likely to be receiving benefits
outside the country than those without post office boxes. Some of the
other factors we considered, however, such as recipients who use direct
deposit, and lack of response of agency inquiries, were less likely to be
residency violators. Given the potential usefulness of this limited modeling
demonstration, it may be possible for SSA to expand and refine its risk
analysis system to better target potential violators.

Beyond the targeting problems we identified with SSA’s risk analysis
system, we found that the agency was not using redeterminations as
efficiently as it could despite the fact that SSA’s data and our prior reviews
have documented their effectiveness for verifying recipients’ eligibility.19 In
particular, home visits are not used frequently enough during
redetermination reviews according to staff in a number of offices we
visited. Although a number of field staff who use home visits reported that
it is a highly effective tool for verifying recipients’ residency, SSA and OIG
officials told us that this technique is not currently employed in some
offices that could benefit from the practice. For example, while a number
of field offices in two SSA regions we visited routinely use home visits,
other offices in the same geographic area rarely use this tool. SSA officials
and field office staff told us that a number of factors account for the
variation in how frequently this technique is used. These include a lack of
adequate staff resources in some offices, differences in the priorities of
field office managers, and differences in how individual staff view the
seriousness of the residency problem.




18
 SSI recipients with residency violations were compared against recipients with no
violations.
19
 SSA data show that, in 1998, refining the case selection methodology increased estimated
overpayment benefits—amounts detected and future amounts prevented—by $99 million
over the prior year. SSA officials have estimated that conducting substantially more
redetermination reviews would yield hundreds of millions of dollars in additional
overpayment benefits annually. U.S. General Accounting Office, Supplemental Security
Income: Progress Made in Detecting and Recovering Overpayments, but Management
Attention Should Continue, GAO-02-849 (Washington, D.C.: Sept. 6, 2002).




Page 18                                       GAO-03-724 Supplemental Security Income
Those field offices that do carry out home visits as part of their
redetermination procedures have found them effective. About half of the
field offices we visited (9 of 17) routinely employ home visits at least some
of the time to verify whether recipients actually live at the address they
report to SSA. For example, SSA’s regional office in Dallas, Texas,
currently contracts with a private investigation firm to conduct residency
home visits. Using these investigators, field offices within the region
performed 4,200 home visits that uncovered at least $2.1 million in
additional overpayments between October 1997 and January 2003.
According to SSA data, this project achieved a benefit-to-cost ratio of
almost 8 to 1. Similarly, the California Department of Health Services has
worked cooperatively with SSA field offices in the San Diego area by
conducting residency home visits. Because Medicaid eligibility is often
directly tied to SSI eligibility, identifying residency violations may save
funds from both programs. Between October 2000 and September
2002, state Medicaid investigators identified about 1,600 SSI recipients
with residency violations. SSI staff in participating offices refer recipients
suspected of violating residency guidelines to the Medicaid investigators,
who subsequently perform unannounced home visits to establish the
recipient’s residency. Both SSA and state officials we interviewed told us
that this project has been effective at identifying residency violations that
might not have been detected by the agency using standard verification
procedures. For example, in April of 2002, state investigators discovered
an SSI recipient who was using a residence in southern California as a
mailing address. The investigators determined that this recipient had been
residing in Tijuana, Mexico, for at least 8 years. Similarly, in June
2002, state investigators found an SSI recipient using a post office box in
southern California as a mailing address. Upon further examination, the
investigators determined that the recipient had been living in San Felipe,
Mexico, since 1982. In addition, in July 2002, state investigators identified
two SSI recipients who improperly received SSI benefits while residing in
Tijuana, Mexico, between August 1999 and April 2002. SSA estimates that
these recipients were overpaid more than $40,000 during this time. Finally,
because the state provides this service to SSA free of charge, it is highly
cost-effective.

In terms of deterring future violations, we found that monetary penalties
and administrative sanctions are rarely, if ever, used in the offices we
visited.20 For example, about 72 percent of the field staff we interviewed


20
 Prior GAO reports indicate that monetary penalties and administrative sanctions may be
underutilized in the SSI program. GAO-02-849.



Page 19                                      GAO-03-724 Supplemental Security Income
said that penalties or sanctions are not used in their offices, or are only
used occasionally. National data on SSA’s use of monetary penalties and
administrative sanctions also suggest that these tools are not routinely
utilized for recipients who fail to report important information that can
affect their eligibility, including absences from the country. For example,
in a recent report, we estimated that at most about 3,500 recipients were
penalized for reporting failures in fiscal year 2001.21 Under the law, SSA
may impose monetary penalties on recipients who do not file timely
reports about factors or events that can affect their benefits. A penalty
causes a reduction in 1 month’s benefits. Penalty amounts are $25 for a
first occurrence, $50 for a second occurrence, and $100 for the third and
subsequent occurrences. The penalties are meant to encourage recipients
to file accurate and timely information so SSA can adjust its records to
correctly pay benefits. However, a large number of staff we interviewed
noted that monetary penalties are too low to be an effective deterrent
against future residency violations. In addition, the Foster Care
Independence Act of 1999 (Pub. L. No. 106-169) gave SSA authority to
impose administrative sanctions on persons who misrepresent material
facts that they know, or should have known, were false or misleading. In
these circumstances, SSA may suspend benefits for 6 months for the initial
violation, 12 months for the second violation, and 24 months for
subsequent violations. Despite having this authority, we found that
administrative sanctions such as benefit suspension are rarely if ever used
by field staff for residency violators. Consistent with the results of our
field work, a prior review shows that administrative sanctions were only
imposed in 21 cases nationwide as of January 2002.22 A substantial number
of staff told us that they rarely use this tool because the process for
imposing administrative sanctions is often time-consuming and
cumbersome. In addition, some staff reported that SSA management does
not encourage the use of penalties or sanctions to deter residency
violations. In response to recommendations we made in a recent report,
SSA is currently evaluating its policies for imposing monetary penalties
and administrative sanctions.23




21
     See GAO-02-849.
22
     Ibid.
23
     Ibid.




Page 20                               GAO-03-724 Supplemental Security Income
SSA Has Not Actively       While SSA uses third-party information to verify certain aspects of
Pursued Third-Party Data   recipients’ eligibility such as income, we found that the agency lacks
Sources to Detect          adequate outside data sources to detect potential residency violations.24
                           SSA is planning to conduct periodic computer matches with immigration
Potential Violators        databases to identify noncitizen SSI recipients who voluntarily report their
                           planned absences from the country25 or are deported from the United
                           States. The agency currently receives periodic paper reports from
                           immigration officials on noncitizens who have current and planned
                           absences from the United States and sends them to the appropriate SSA
                           field offices for follow up. However, these procedures are only effective
                           for recipients who voluntarily report their absence to immigration
                           officials. Thus, SSA will remain limited in its ability to independently verify
                           the residency of SSI recipients who deliberately seek to conceal extended
                           periods outside the country. Over half of the SSA managers and field staff
                           we interviewed told us that access to automated immigration data would
                           help them to more accurately verify recipients’ residency.

                           Despite this limitation, SSA has not adequately explored the potential for
                           obtaining access to emerging data sources such as an entry-exit system
                           being developed by the Department of Justice and the Department of
                           Homeland Security (DHS).26 This system is being implemented as a
                           mechanism to monitor all major ports of entry/exit in the United States,
                           including land crossings, seaports, and airports. Once operational, this
                           system will allow authorized federal agencies to collect, maintain, and
                           share information on selected individuals who enter and exit the United
                           States to ensure border security, among other purposes. Our work
                           suggests that this system may also provide information that could help
                           SSA determine when noncitizen SSI recipients exit the country for
                           extended periods of time. We acknowledge that such databases could have
                           limitations that affect their accuracy and completeness—especially given




                           24
                            For example, SSA routinely uses information from the Department of Health and Human
                           Service’s National Directory of New Hires to verify SSI recipients’ income.
                           25
                            SSA expects to save approximately $28 million annually by implementing these matches
                           and suspending benefits for recipients who are identified.
                           26
                            A new system called the United States Visitor and Immigrant Status Indication
                           Technology system is currently being developed by DHS and will incorporate existing
                           entry-exit databases.




                           Page 21                                      GAO-03-724 Supplemental Security Income
problems that our prior work has identified with some immigration data.27
Thus, SSA would likely have to determine the reliability and cost-
effectiveness of accessing such data before negotiating data sharing
agreements and using the information for detecting potential residency
violations.

SSA has also not fully utilized its authority to obtain independent data
from other sources, such as financial institutions, as a tool for detecting
potential residency violations. The Foster Care Independence Act of
1999 granted SSA new authority to verify recipients’ financial accounts. To
implement this authority, SSA issued proposed regulations on its new
processes for accessing financial data in May 2002.28 These regulations,
which were still in draft at the time of our report, may permit SSA to
obtain a variety of financial records from banks, credit card companies,
and other financial institutions, including those operating branches and
automated teller machines (ATM) outside the United States. However,
according to SSA officials, the agency only intends to use this information
to verify recipients’ bank account balances as a way of verifying their
financial resources. SSA does not currently plan to use financial institution
data more broadly to detect potential residency violations.

Given the relatively narrow scope of SSA’s proposed use of financial data,
the agency may be unnecessarily limiting its ability to detect residency
violations. In particular, SSA may be missing potentially helpful sources of
information such as data on recipients who conduct banking transactions
outside the United States using ATMs. As noted previously, a large
proportion of the residency overpayments SSA detected between 1997 and
2001 were tied to recipients who originated in various countries in Latin
America and South/Southeast Asia. However, SSA currently has no way to
identify recipients who withdraw SSI benefits from ATMs outside the
United States. Information we obtained from a national financial data
vendor indicates that it is now possible for authorized users to obtain
detailed information on individuals’ financial transactions from a large
number of national and international institutions. Such data sources



27
 U.S. General Accounting Office, Illegal Immigration: INS Overstay Estimation Methods
Need Improvement,GAO/PEMD-95-20 (Washington, D.C.: Sept. 26, 1995) and Immigration
Statistics: Information Gaps, Quality Issues Limit Utility of Federal Data to
Policymakers, GAO/GGD-98-164 (Washington, D.C.: July 31, 1998).
28
 See Access to Information Held by Financial Institutions, 67 Fed. Reg. 22021 (to be
codified at 20 C.F.R. pt. 416) (proposed May 2, 2002).




Page 22                                        GAO-03-724 Supplemental Security Income
              include basic information such as bank account balances from banks in
              the United States, as well as more sophisticated information such as ATM
              activity that is transmitted on the international telecommunications
              networks. Our review suggests that such data could provide SSA with a
              potentially powerful tool to identify residency violations. For example,
              SSA may be able to obtain data for recipients whose SSI benefits are direct
              deposited into a U.S. bank and then withdrawn from ATMs outside the
              country for extended periods of time. However, SSA does not currently
              plan to obtain access to direct deposit data from financial institutions or
              ATM networks that operate in other countries.


              SSA has made progress in addressing residency violations in recent years,
Conclusions   especially through the special initiatives it has undertaken. However, many
              of these initiatives have been short-lived and limited to a small number of
              field offices. Thus, the agency’s approach to this problem has been
              generally ad hoc and restricted in scope. As a result, our review suggests
              that SSA identifies only a portion of the violations and resulting
              overpayments that occur each year. We recognize that the SSI program is
              complex to administer and residency requirements are particularly
              difficult to enforce because they can necessitate time-consuming, labor-
              intensive verification checks, such as home visits. However, SSA has not
              employed a systematic, comprehensive approach to this problem that
              would allow the agency to use its available systems and procedures more
              efficiently and reduce the program’s exposure to additional violations. In
              particular, SSA has not reengineered its current systems and processes to
              make better use of limited budgetary and staff resources. For example, our
              review shows that minor modifications of its risk analysis system could
              help the agency identify recipients who are most likely to violate residency
              requirements. Without such modifications, however, it will be difficult for
              the agency to effectively target its redetermination reviews and
              incorporate home visits in a cost-effective manner.

              Additionally, SSA has not made sufficient use of independent, third-party
              data sources to help verify recipients’ residency despite having
              successfully used such tools to verify other aspects of recipient eligibility,
              such as their income and other financial resources. SSA could improve SSI
              program integrity and reduce residency overpayments by exploring more
              creative use of technology, including the use of financial institution data to
              detect recipients who use ATMs for bank transactions outside the United
              States for extended periods of time. The agency may also benefit from
              pursuing other emerging data sources such as entry/exit systems being
              developed by the Departments of Justice and Homeland Security, although


              Page 23                                 GAO-03-724 Supplemental Security Income
                  prior problems we have identified with immigration data may require SSA
                  to determine the reliability of such databases.

                  Ultimately, failure to implement a more strategic approach to this problem
                  and to reengineer its existing processes may compromise the agency’s
                  ability to use its limited resources in the most cost-effective manner.
                  Moreover, failure to make optimal use of existing tools and access
                  emerging data sources could leave SSI and other programs, such as
                  Medicaid, vulnerable to continuing residency violations and additional
                  overpayments.


                  In order to further strengthen and increase SSA’s ability to detect SSI
Recommendations   residency violations and reduce resulting overpayments due to recipient
                  absences from the United States, we recommend that the Commissioner of
                  Social Security take the following actions:

                  •   Consider reengineering the agency’s risk analysis system to more
                      specifically target potential residency violators. The list of potentially
                      high-risk characteristics we have developed and tested could provide a
                      starting point for such refinements. To accomplish this, SSA may wish
                      to test the idea on a one-time basis using methods the agency deems
                      appropriate to assess its effectiveness.

                  •   Consider expanding the use of unannounced home visits in some areas
                      as a way of verifying the residency of recipients whom the agency
                      identifies as potentially being at high risk for violations. To ensure that
                      only cases with a high potential for success are selected, any potential
                      profile of high-risk recipients that SSA develops could be a primary
                      source of referrals. To maximize limited staff resources, SSA should
                      apply a strategic approach to this problem, recognizing that violations
                      are not equally prevalent in all areas of the country.

                  •   Study the feasibility of expanding the type of information SSA obtains
                      from financial institutions as authorized by The Foster Care
                      Independence Act of 1999 (Pub. L. No. 106-169). Additional information
                      to help identify violators could include bank and ATM withdrawal
                      records to help identify SSI recipients who may be accessing their SSI
                      benefits outside the United States for extended periods of time.

                  •   Investigate the potential for obtaining access to emerging third-party
                      data sources such as entry/exit databases being developed by DHS to
                      help field staff more accurately verify whether SSI recipients are
                      violating program regulations.


                  Page 24                                  GAO-03-724 Supplemental Security Income
                     We provided a draft of this report to SSA and DHS for review and
Agency Comments      comment.29 SSA generally agreed with our recommendations, but noted
and Our Evaluation   some challenges to their implementation.

                     While agreeing with each of our recommendations, SSA supplied
                     additional information concerning its ability to implement these
                     recommendations. With regard to our first recommendation to reengineer
                     its risk analysis system to more specifically target potential residency
                     violators, SSA agreed with the intent of the recommendation but
                     expressed concern that including individuals born outside the United
                     States as one risk factor could be considered discriminatory. As noted in
                     our report, we suggest that this factor could be included as one of several
                     different factors that the agency could use to refine its risk analysis
                     system. We believe that such an approach could be implemented in a non-
                     discriminatory manner and would help SSA use its limited resources more
                     efficiently. Moreover, as discussed in the report, our analysis suggests that
                     this factor is a potentially powerful indicator of possible residency
                     violations.

                     With regard to our second recommendation to expand the use of
                     unannounced home visits, SSA agreed that home visits are a useful tool for
                     verifying SSI recipients’ residency, but noted that costs and employee
                     safety must be considered. We agree that these are important issues that
                     SSA must consider as it studies how home visits can be used most
                     effectively. Further, as we discuss in the report, some states have
                     successfully used state personnel and private investigators to perform
                     home visits. We also note that at least one state has found the use of
                     private investigators to be cost-effective. Thus, we believe that SSA could
                     look more closely at the experience of these states to identify potential
                     best practices for conducting home visits.

                     SSA also agreed with our third recommendation that the agency study the
                     feasibility of expanding the type of information it obtains from financial
                     institutions. SSA noted some potential legal and technical issues that will
                     require further study by the agency. For example, SSA noted that financial
                     records may not be an accurate basis for identifying recipients who are
                     outside the country for more than 30 days. While we agree that
                     definitively determining whether a recipient is outside the country for
                     30 consecutive days or more presents a challenge for the agency, we only


                     29
                          DHS did not provide formal comments.




                     Page 25                                     GAO-03-724 Supplemental Security Income
suggest that SSA could use financial institution data as a potential
indicator of residency violations.

Finally, with regard to our fourth recommendation, SSA agreed that there
may be potential benefits to accessing external data sources to help verify
recipients’ residency. The agency indicated that it will explore the
potential feasibility of using such data sources as part of its SSI Corrective
Action Plan.

SSA’s formal comments appear in appendix II. SSA also provided
additional technical comments that we have incorporated in the report, as
appropriate.


We are sending copies of this report to the House and Senate Committees
with oversight responsibility for the Social Security Administration. We
will also make copies available to other parties upon request. In addition,
the report will be available at no charge on GAO’s Web site at
http//:www.gao.gov. If you have any questions concerning this report,
please contact me at (202) 512-7215.

Sincerely yours,




Robert E. Robertson
Director, Education, Workforce,
 and Income Security Issues




Page 26                                 GAO-03-724 Supplemental Security Income
             Appendix I: Scope and Methodology
Appendix I: Scope and Methodology


             This appendix provides additional details about our analysis of the
             Supplemental Security Income program’s (SSI) residency violations,
             including potential weaknesses in the Social Security Administration’s
             (SSA) policies and procedures. To meet the objectives of this review, we
             reviewed prior and ongoing projects by SSA and its Office of Inspector
             General (OIG), conducted independent audit work, and reviewed our prior
             work on the SSI program. We also reviewed SSA’s policies and SSI
             program guidelines concerning eligibility determinations and procedures
             for detecting potential residency violations. In addition, we analyzed SSI
             payment data between 1997 and 2001 and examined studies in which SSA
             or its OIG identified recipients who were residing outside the country. We
             reviewed our past work on the SSI program to evaluate the current use of
             such tools as administrative sanctions and monetary penalties. Finally, we
             interviewed SSA and OIG officials at its headquarters in Baltimore,
             Maryland, and key regional and field managers and staff responsible for
             administering and monitoring the SSI program.

             We conducted independent audit work in five states (California, Florida,
             Michigan, New York, and Texas) to identify common residency violation
             characteristics and examine SSA’s processes to identify residency
             violations. We selected locations for field visits based on the following
             criteria: (1) geographic dispersion, (2) states previously included in a SSA
             or OIG special initiative, (3) states with large numbers of SSI recipients,
             (4) states with large dollars of SSI expenditures, and (5) states with large
             numbers of noncitizen SSI recipients. These states represented about
             72 percent of the total noncitizen population potentially eligible for SSI
             benefits, about 41 percent of the total SSI recipients, and 45 percent of
             total SSI benefits paid in the United States. In total, we visited 17 field
             offices and interviewed 112 SSA field office managers and line staff
             responsible for the SSI program. We visited field offices more prone to
             having recipients with residency violations, especially offices near border
             crossings either by land, sea, or air. Where appropriate, we also visited
             offices that were involved with a prior or ongoing SSA or OIG special
             initiative to detect residency violations.

             During our meetings with SSA and OIG officials, we documented
             management and staff views on how extensive residency violations are in
             the SSI program; the effectiveness of current procedures and processes for
             detecting and preventing residency violations; and potential improvements
             to existing program processes, policies, and systems. We also interviewed
             certain state officials knowledgeable about or involved with SSI residency
             verifications. In addition, we interviewed officials from other federal
             agencies, including the former Immigration and Naturalization Service,


             Page 27                                GAO-03-724 Supplemental Security Income
Appendix I: Scope and Methodology




and the Centers for Medicare and Medicaid Services (CMS) to determine
how these agencies could assist SSA in verifying recipients’ residency. We
also interviewed officials from a national financial data vendor to obtain
information on currently available financial data. We conducted our work
from September 2002 through May 2003 in accordance with generally
accepted government auditing standards.

As part of our study, we developed and tested a logistic regression model
to help predict whether certain SSI recipients were more likely than others
to have residency violations. The factors we used in our model were
recipients who (1) were born outside of the United States, (2) have their
SSI benefits direct deposited into bank accounts, (3) use post office boxes
to receive their mail, (4) have had a prior residency violation, or (5) could
not be located by SSA for an extended period.1 In deciding which variables
to include in our regression analysis, we chose variables that were most
frequently reported to us by SSA and OIG staff during our fieldwork.
Additional potentially useful variables were also reported. (See table 1 for
a more comprehensive list of the variables cited.) This is a partial listing of
the factors reported to us; it does not include all responses.2




1
 An additional variable cited frequently by staff as a potentially effective tool to identify
residency violations—Medicaid nonutilization information—was not included in our model
because SSA does not currently have automated data on Medicaid nonusage by SSI
recipients. However, SSA is currently negotiating access to such data with the CMS.
2
 In a prior study, SSA’s OIG identified several factors as potential indicators that a recipient
may be violating the SSI residency requirements. These factors include: (1) age when a
recipient applies for SSI, (2) sex, (3) disability status, (4) citizenship, (5) use of commercial
mailboxes, (6) excess resources or income, (7) questionable addresses, (8) failure to
provide sufficient evidence to SSA staff, (9) lack of an in-person benefit review for an
extended period of time, (10) direct deposit of benefits, (11) prior address changes,
(12) residential address that differs from a recipients’ mailing address, and (13) living in
another person’s household. The OIG noted that more than half of residency violators
exhibited four or more of these characteristics.




Page 28                                           GAO-03-724 Supplemental Security Income
Appendix I: Scope and Methodology




Table 1: Variables That Indicate a Recipient May Potentially Be Out of the United
States

 Variable
 Recipients with prior residency violations
 Recipients who use direct deposit to receive their benefits
 Recipients receiving mail at a post office box or commercial mail drop
 Recipients who were born outside of the United States
 Recipients who do not utilize their Medicaid benefits for an extended period of time
 Recipients who SSA cannot locate or contact
 Recipients reporting different residential and mailing addresses
 Recipients living with family or friends without their own residency
 Aged recipients
 Multiple recipients using the same physical mailing address
 Recipients with immediate family in another country
 Recipients with excess resources or income
Source: GAO analysis.

Note: Factors in table 1 include all responses where at least 10 percent of the total staff interviewed
cited that a specific variable would be a valid predicator that a recipient may be out of the country and
thus violating SSI residency requirements.


Our model resulted in estimates of the relative likelihood (odd ratios) of a
current residency violation depending on the absence or presence of the
included five variables (see table 2). If there is no significant difference
between the presence and the absence of one of the variables, with
respect to a current residency violation, the odds would be approximately
equal, and the ratio of the odds would be close to 1.00. The more the odds
ratio differs from 1.00 in either direction, the larger the effect it represents.
For example, if there were very little difference between those
beneficiaries who did and did not use direct deposit, with respect to a
current (as of Apr. 2003) residency violation, the odds ratio for direct
deposit would be close to 1.00.




Page 29                                               GAO-03-724 Supplemental Security Income
Appendix I: Scope and Methodology




Table 2: Odds Ratios for the Variables in Our Logistical Regression Analysis

    All of the following odds ratios are statistically significant
                                                 Odds
    Variable                                      ratio      Interpretation
    Recipients who were                          44.19       Recipients who were born outside of the
    born outside of the                                      United States are approximately 44 times
    United States.                                           more likely to be current residency violators.
    Recipients who did use                         0.83      Recipients who did use direct deposit to
    direct deposit to receive                                receive their benefits are 1.2 times less likely
                                                                                               a
    their benefits.                                          to be current residency violators.
    Recipients receiving mail                      1.77      Recipients receiving mail at a post office box
    at a post office box.                                    are 1.8 times more likely to be current
                                                             residency violators.
    Recipients with prior                          9.80      Recipients with prior residence violations are
    residence violations.                                    about 10 times more likely to be current
                                                             residency violators.
    Recipients who SSA                             0.06      Recipients who SSA cannot locate or contact
    cannot locate or contact.                                are 17 times less likely to be current
                                                             residency violators.b
Source: GAO’s analysis of SSA data on residency violation.
a
 For odds rations that fall between 0 and 1, the reciprocal of the odds ratio describes the odds of
being less likely. Thus, the odds ratio of 0.83 is interpreted as 1 divided by 0.83, which is
approximately 1.2 times less likely.
b
Similarly, the reciprocal of 0.06 is approximately 17.




Page 30                                                         GAO-03-724 Supplemental Security Income
                            Appendix II: Comments from the Social
Appendix II: Comments from the Social
                            Security Administration



Security Administration

Note: GAO comments
supplementing those in
the report text appear at
the end of this appendix.




                            Page 31                                 GAO-03-724 Supplemental Security Income
                 Appendix II: Comments from the Social
                 Security Administration




See comment 1.




                 Page 32                                 GAO-03-724 Supplemental Security Income
Appendix II: Comments from the Social
Security Administration




Page 33                                 GAO-03-724 Supplemental Security Income
                 Appendix II: Comments from the Social
                 Security Administration




See comment 2.




                 Page 34                                 GAO-03-724 Supplemental Security Income
Appendix II: Comments from the Social
Security Administration




Page 35                                 GAO-03-724 Supplemental Security Income
               Appendix II: Comments from the Social
               Security Administration




               The following are GAO’s comments on SSA’s letter dated June 27, 2003.
GAO Comments
               1. Based on our analysis, we continue to believe that this factor—
                  recipients born outside the United States—is a good indicator of a
                  potential residency violation. Use of this factor may help SSA further
                  refine its risk analysis system (see report page 18).

               2. Our report does not state that direct deposit is a good indicator of a
                  residency violation. Rather, we discuss the potential use of financial
                  institution data such as recipients’ banking transactions outside the
                  United States using automated teller machines, which is currently
                  unavailable to SSA (see report page 22).




               Page 36                                 GAO-03-724 Supplemental Security Income
                  Appendix III: GAO Contacts and Staff
Appendix III: GAO Contacts and Staff
                  Acknowledgments



Acknowledgments

                  Daniel Bertoni (202) 512-5988
GAO Contacts      Jeremy D. Cox (202) 512-5717


                  In addition to those named above, Jeff Bernstein, Sue Bernstein, Kriti
Staff             Bhandari, Salvatore F. Sorbello, Vanessa Taylor, Wendy Turenne, and
Acknowledgments   Shana Wallace made important contributions to this report.




                  Page 37                                GAO-03-724 Supplemental Security Income
             Related GAO Products
Related GAO Products


             Supplemental Security Income: Progress Made in Detecting and
             Recovering Overpayments, but Management Attention Should Continue.
             GAO-02-849. Washington, D.C.: September 16, 2002.

             Supplemental Security Income: Status of Efforts to Improve
             Overpayment Detection and Recovery. GAO-02-962T. Washington, D.C.:
             July 25, 2002.

             Social Security Administration: Agency Must Position Itself Now to Meet
             Challenges. GAO-02-289T. Washington, D.C.: May 2, 2002.

             Social Security Administration: Status of Achieving Key Outcomes and
             Addressing Major Management Challenges. GAO-01-778. Washington,
             D.C.: June 15, 2001.

             High Risk Series: An Update. GAO-01-263. Washington, D.C.: January
             2001.

             Major Management Challenges and Program Risks: Social Security
             Administration. GAO-01-261. Washington, D.C.: January 2001.

             Supplemental Security Income: Additional Actions Needed to Reduce
             Program Vulnerability to Fraud and Abuse. GAO/HEHS-99-151.
             Washington, D.C.: September 15, 1999.

             Supplemental Security Income: Long–Standing Issues Require More
             Active Management and Program Oversight. GAO/T-HEHS-99-51.
             Washington, D.C.: February 3, 1999.

             Major Management Challenges and Program Risks: Social Security
             Administration. GAO/OCG-99-20. Washington, D.C.: January 1, 1999.

             Supplemental Security Income: Action Needed on Long-Standing
             Problems Affecting Program Integrity. GAO/HEHS-98-158. Washington,
             D.C.: September 14, 1998.

             Supplemental Security Income: Opportunities Exist for Improving
             Payment Accuracy. GAO/HEHS-98-75. Washington, D.C.: March 27, 1998.

             High Risk Program: Information on Selected High-Risk Areas. GAO/HR-
             97-30. Washington, D.C.: May 16, 1997.




             Page 38                             GAO-03-724 Supplemental Security Income
           Related GAO Products




           High Risk Series: An Overview. GAO/HR-97-1. Washington, D.C.: February
           1997.




(130186)
           Page 39                             GAO-03-724 Supplemental Security Income
                         The General Accounting Office, the audit, evaluation and investigative arm of
GAO’s Mission            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.


                         The fastest and easiest way to obtain copies of GAO documents at no cost is
Obtaining Copies of      through the Internet. GAO’s Web site (www.gao.gov) contains abstracts and full-
GAO Reports and          text files of current reports and testimony and an expanding archive of older
                         products. The Web site features a search engine to help you locate documents
Testimony                using key words and phrases. You can print these documents in their entirety,
                         including charts and other graphics.
                         Each day, GAO issues a list of newly released reports, testimony, and
                         correspondence. GAO posts this list, known as “Today’s Reports,” on its Web site
                         daily. The list contains links to the full-text document files. To have GAO e-mail
                         this list to you every afternoon, go to www.gao.gov and select “Subscribe to daily
                         E-mail alert for newly released products” under the GAO Reports heading.


Order by Mail or Phone   The first copy of each printed report is free. Additional copies are $2 each. A
                         check or money order should be made out to the Superintendent of Documents.
                         GAO also accepts VISA and Mastercard. Orders for 100 or more copies mailed to a
                         single address are discounted 25 percent. Orders should be sent to:
                         U.S. General Accounting Office
                         441 G Street NW, Room LM
                         Washington, D.C. 20548
                         To order by Phone:     Voice:    (202) 512-6000
                                                TDD:      (202) 512-2537
                                                Fax:      (202) 512-6061


                         Contact:
To Report Fraud,
                         Web site: www.gao.gov/fraudnet/fraudnet.htm
Waste, and Abuse in      E-mail: fraudnet@gao.gov
Federal Programs         Automated answering system: (800) 424-5454 or (202) 512-7470


                         Jeff Nelligan, Managing Director, NelliganJ@gao.gov (202) 512-4800
Public Affairs           U.S. General Accounting Office, 441 G Street NW, Room 7149
                         Washington, D.C. 20548