OFFICE OF
THE INSPECTOR GENERAL

SOCIAL SECURITY ADMINISTRATION

EFFECTIVENESS OF
THE SINGLE SELECT
EDIT ROUTINE

September 2007

A-03-07-17065

AUDIT REPORT

Mission

By conducting independent and objective audits, evaluations and investigations, we inspire public confidence in the integrity and security of SSA's programs and operations and protect them against fraud, waste and abuse. We provide timely, useful and reliable information and advice to Administration officials, Congress and the public.

Authority

The Inspector General Act created independent audit and investigative units, called the Office of Inspector General (OIG). The mission of the OIG, as spelled out in the Act, is to:

Conduct and supervise independent and objective audits and investigations relating to agency programs and operations.
Promote economy, effectiveness, and efficiency within the agency.
Prevent and detect fraud, waste, and abuse in agency programs and operations.
Review and make recommendations regarding existing and proposed legislation and regulations relating to agency programs and operations.
Keep the agency head and the Congress fully and currently informed of problems in agency programs and operations.

To ensure objectivity, the IG Act empowers the IG with:

Independence to determine what reviews to perform.
Access to all information necessary for the reviews.
Authority to publish findings and recommendations based on the reviews.

Vision

We strive for continual improvement in SSA's programs, operations and management by proactively seeking new ways to prevent and deter fraud, waste and abuse. We commit to integrity and excellence by supporting an environment that provides a valuable public service while encouraging employee development and retention and fostering diversity and innovation.

MEMORANDUM

Date: September 28, 2007

To: The Commissioner

From: Inspector General

Subject: Effectiveness of the Single Select Edit Routine (A-03-07-17065)

OBJECTIVE

Our objective was to assess the effectiveness of the Single Select edit routine in resolving name/Social Security number (SSN) mismatches on reported earnings and posting these earnings to the correct earnings records.

BACKGROUND

As part of the Annual Wage Reporting process, the Social Security Administration (SSA) validates the names and SSNs on wage reports submitted by employers against SSA's Numident file-the repository for all issued SSNs. Earnings items that contain a name/SSN combination matching the Numident file are posted to the Master Earnings File (MEF). Earnings items that cannot be matched to SSA's records are posted to the Earnings Suspense File (ESF)-a file of unmatched items. As of October 2006, the ESF had accumulated about $586 billion in wages and 264 million wage items for Tax Years (TY) 1937 through 2004. In TY 2004 alone, the ESF grew by $66 billion in wages and 9.5 million wage items.

The Single Select edit routine is one of the Agency's SSN/name matching routines used to resolve unmatched name and SSN data submitted to SSA in employers' wage reports (see Appendix C for a list of matching routines and related correspondence). The Single Select edit assumes the individual's reported last name is correct, but some mistake has been made with the SSN. Single Select creates up to 89 possible variations of the SSN and matches them against the Numident. If one SSN/name match is found, the earnings are posted to the validated SSN. If no exact match is found on the last name, the Single Select edit is extended further to search for a match on the individual's entire name. If the SSN/name is validated through this new Extended Single Select edit routine, the earnings are posted to that validated SSN. If the extended search fails to validate the name and SSN, the record is marked as invalid and is posted to the ESF.

SSA uses the ESF Reinstates File to record earnings items posted as a result of the Single Select edit, as well as other edits, to maintain a record of the change from the reported name/SSN to the corrected name/SSN. The Reinstates File was not designed to maintain a record on every reinstatement, but instead maintain sufficient information on each unique correction so that this information can be used if the same error is reported in future TYs by the same employer.

RESULTS OF REVIEW

Our review found that SSA's Single Select edit routine has been an effective edit for resolving name/SSN mismatches on reported FICA earnings. The edit has corrected an estimated 46.6 million items associated with approximately $228 billion in earnings since TY 1937. Moreover, we reviewed a sample of postings and determined that the earnings items posted to the MEF were consistent with the identity and work history of the located earners. While we found that the Extended Single Select corrections were not being properly identified in SSA's ESF Reinstates File at the time of our audit, SSA has since corrected this problem. Finally, we found anomalies related to non-FICA earnings, though we determined the problems related to employer reporting issues and were not associated with the Single Select edit process. The non-FICA earnings issues we identified related to five employers who reported excessive amounts for Health Savings Accounts (HSA) and deferred compensation.

SINGLE SELECT CORRECTIONS

We estimate that the Single Select process has reinstated an estimated 46.6 million items representing approximately $228 billion in earnings between TYs 1937 and 2005. When compared to the other edit processes, Single Select has resolved the most mismatched names/SSNs on reported earnings. While we found that the Extended Single Select corrections were not being properly identified in SSA's ESF Reinstates File at the time of our audit, SSA has since corrected this problem.

Estimated Corrections

In our review of 5 percent of the MEF, we found that 2.3 million items associated with $11.4 billion in earnings were posted to the MEF via the Single Select process for TYs 1937 to 2005 (see Table 1). These Single Select postings related to the original Single Select edit routine as well as the more recent Extended Single Select edit. In terms of the entire MEF, we estimate that approximately 46.6 million items associated with $228 billion in earnings had been reinstated to the MEF during this period under either the Single Select or Extended Single Select routine.

Table 1: Single Select FICA Postings for TYs 1937 to 2005
(as of October 2006)
Edit Type
Posted Earnings Items
(1 Segment) Estimated Posted Earnings Items
(20 Segments)

Posted Earnings (1 Segment)
Estimated
Posted Earnings (20 Segments)
Single Select 2,328,100 46,562,000 $11.4 billion $228 billion
Extended Single Select 4,300 86,000 $20.7 million $414 million
Total 2,332,400 46,648,000 $11.4 billion $228.4 billion
Note: Numbers rounded to the nearest billion.

About 99.8 percent of the earnings items were posted as result of the original Single Select edit routine. The remaining 0.2 percent of the earnings items were posted as a result of the Extended Single Select edit routine.

Reinstates File

A review of the ESF Reinstates File for TY 2005 indicated the Single Select edit routine corrected more items than any other edit process, or approximately 65 percent of the earnings items in this file (see Appendix C for a description of all the processes). In Figure 1, we provide a breakout of the content of the ESF Reinstates File for TY 2005.

Figure 1: Contents of the Earnings Suspense File
Reinstates File (TY 2005)

The next most successful process is Prior Reinstate with 28 percent of the Reinstates File. This edit is used when subsequent employer wage reports contain an unmatched SSN/name fitting the same characteristics as a previously corrected wage item from the same employer's wage report. Hence, Prior Reinstate items are basically a "ripple effect" from prior corrections. Earnings items posted as a result of Decentralized Correspondence (DECOR) represented 6 percent of the file. The remaining 1 percent of the items in the file was posted for various other reasons.

Prior Reinstate Items

To better understand the Prior Reinstate items, we reviewed 50 randomly selected Prior Reinstate items from the TY 2005 Reinstates File to determine the originating edit routine. We found that approximately 78 percent of TY 2005 Prior Reinstate postings were due to initial Single Select postings (see Table 2 for our results). Hence, Single Select has a significant "ripple effect" on later employer wage reports, which further increases the effectiveness of this edit.
Table 2: Originating Edit for 50 Prior Reinstate Items (TY 2005)
Original Reinstatement Process Items Percentage
Single Select 39 78
Decentralized Correspondence 3 6
Reinstatement from Internal Revenue Service 4 8
Other 4 8
Total Prior Reinstate Items 50 100%
Note: See Appendix C for a description of all processes.

Extended Single Select Reinstatements

In our analysis of the ESF Reinstates File data, we determined that the Extended Single Select corrections were not being recorded in this file. We reviewed a random sample of 10 Extended Single Select postings from the MEF and traced them back to the Reinstates File. In all 10 cases, we found no evidence under the employers' records that these earnings items had been corrected via this edit. While the lack of this data in the Reinstates File did not impact the individual earner, the corrected item was not available for future corrections under the Prior Reinstate edit. An identical error on a future earnings item would still be corrected via the Extended Single Select edit, but this is inconsistent with other edits and the role of the Prior Reinstates edit.

When we spoke to SSA staff, they concurred that the ESF Reinstates File should have recorded the correction though the operating procedures did not include a requirement to do so. The Earnings Branch Manager said the Agency resolved the problem in time for the TY 2006 processing cycle.

ACCURACY OF POSTINGS

We did not find any questionable postings in our review of a sample of Single Select and Extended Single Select corrections related to FICA postings to the MEF. We tested 50 randomly selected Single Select/Extended Single Select earnings items corrections from a population of 68,741 corrections processed during Calendar Year (CY) 2005, representing approximately $383 million in corrected earnings.

To determine whether the information in SSA's records matched the information on the applicable Wage and Tax Statements (Form W-2), we traced the reported information on the W-2 associated with the posted earnings (that is, reported SSN, individual's name, FICA wages and total compensation) to SSA's MEF, Numident, and ESF Reinstates File records. We found that the FICA earnings and total compensation data shown in SSA's records matched the data on the W-2s for all 50 sample items. We also found that the 50 earnings items posted to the MEF were consistent with the identity and work history of the located earners.

ACCURACY OF NON-FICA EARNINGS

In our review of the non-FICA earnings reinstated under Single Select, we found some questionable trends among the earnings being reinstated. The problems related to employers reporting excessive amounts rather than the Single Select process itself.

We performed a limited review of non-FICA earnings corrected as a result of Single Select to determine whether these postings contained any anomalies, such as excessive or duplicate amounts that could represent a questionable posting. We limited our review to transactions processed in CY 2005. We found six non-FICA earnings records reinstated under Single Select processed during this period that exceeded $1 million and did not appear to relate to legitimate earnings activity. These six earnings records, reported by four different employers, represented approximately $96.6 million in non FICA earnings. These six postings contained identical amounts of $16,085,074 reported on the associated W-2s as HSA amounts.

In the six cases, the non-FICA earnings reported by employers and recorded by SSA in the individuals' MEF records were inconsistent with the individuals' earnings histories. For example, 1 of the $16 million HSA amounts posted in TY 2004 related to a 20-year-old earning small amounts whose work history started in TY 2002. He earned $607 in TY 2002, showed no earnings for TYs 2003 through 2004, and then earned $1,805 in TY 2005 and $7,227 in TY 2006. It is unlikely the employer, a restaurant owner, intended to report anything for this individual, since the HSA amount was reported for TY 2004 even though the individual worked only once for the employer in TY 2002, earning $68.
We also found a seventh deferred earnings transaction processed during CY 2002 that contained a related error. In the seventh case, we found $80.4 million related to deferred compensation for TY 2001 was recorded for a 24 year-old who earned a total of $1,516 at the employer in question during TYs 2001 and 2004. In reviewing the TY 2001 W-2 for this individual, we found that that the employer reported $16,085,074 in deferred compensation in five different categories for this individual, amounting to a total of about $80.4 million.

A further review of the seven non-FICA earnings items reported by the five employers for the TYs in question indicates that similar errors were reported and recorded within SSA's MEF for all reported wages. If we assume the same errors occurred throughout the entire wage reports, we estimate the overstatement on SSA's MEF could be as high as $64.4 billion (see Table 3 and Appendix D for additional details).

Table 3: Estimate of Overstated Non-FICA Earnings

Employer
Tax Year in Question
Amount in Question Total Wage Reports for Tax Year in Question Potential Overstatement in Master Earnings Filea
Employer 1 2004 $16,085,074 256 $4.1 billion
Employer 2 2004 $16,085,074 315 5.1 billion
Employer 3 2004 $16,085,074 781 12.3 billion
Employer 4 2004 $16,085,074 313 5.0 billion

Employer 5 2001 $80,425,368 471 37.9 billionb
Totals $144,765,664 2,136 $64.4 billion
Note a: Represents actual amounts reported by employers to SSA as HSA funds.
Note b: This is an estimated amount of deferred earnings based on the number of reported wage items since SSA's systems could not provide this detail for TY 2001.

We shared our findings with SSA systems staff and they noted that they were aware of a vendor software problem during TY 2004 that led to employers over-reporting amounts related to HSAs. The staff stated that they sought resolution to the problem by attempting to get employers to discontinue use of the flawed wage reporting software products and to submit corrected reports for previous wage reporting errors. In addition, the staff noted that the TY 2004 cases caused SSA to implement several improvements to provide better protection against a future occurrence, including the following:

In TY 2006, SSA's process required submitters of wage reports to provide a software vendor ID code. This will help SSA identify the vendor, and to confirm cases where vendor software is suspected of having a specific flaw.

In TY 2007, SSA plans to add new edits that cause an electronic wage report to be investigated when (1) the average money field on a wage report exceeds the
$1 million threshold and (2) a specific number of W-2s also exceed this threshold. In such cases, the Office of Central Operations will contact the submitters of the wage reports and ask the parties to confirm the accuracy of the amounts.

As of June 2007, none of the overstated non-FICA amounts had been corrected. SSA staff stated that both the employers and the IRS were notified of the TY 2004 error. Further, SSA staff noted that the overstated HSA amounts represent more of an employer reporting issue then a program issue since these amounts would not count as income or resources for the purpose of calculating SSA benefits.

CONCLUSION AND RECOMMENDATIONS

We found that SSA's Single Select edit routine has been an effective edit for resolving name/SSN mismatches on reported FICA earnings. While we found that the Extended Single Select corrections were not being properly identified in SSA's ESF Reinstates File at the time of our audit, SSA has since corrected this problem. In addition, as part of our audit we found a number of anomalies related to non-FICA earnings that need management's attention.

To improve the accuracy of non-FICA earnings records, we recommend SSA:

1. Review the non-FICA earnings cases identified to determine what additional actions SSA can take to ensure (a) the employers are aware of these reporting errors, (b) the earlier amounts reported in error are corrected, and (c) other similar employer problems occurring during TY 2004 are corrected. These additional actions could include additional coordination with the Internal Revenue Service.

2. Ensure the planned edits for TY 2007 are established and tested to initiate an investigation of when (1) the average money field on a wage report exceeds the $1 million threshold and (2) a specific number of W-2s also exceed this threshold and, as appropriate, contact employers and/or wage report submitters to confirm the accuracy of these amounts.

AGENCY COMMENTS
SSA agreed with our recommendations. The Agency's comments are included in Appendix E.

Patrick P. O'Carroll, Jr.

Appendices
APPENDIX A - Acronyms
APPENDIX B - Scope and Methodology
APPENDIX C - Edit Routines and Correspondence
APPENDIX D - Reported Earnings for Five Employers
APPENDIX E - Agency Comments
APPENDIX F - OIG Contacts and Staff Acknowledgments

Appendix A
Acronyms
CY Calendar Year
DECOR Decentralized Correspondence
DoB Date of Birth
EAD Earnings After Death
EDCOR Educational Correspondence
ESF Earnings Suspense File
FICA Federal Insurance Contributions Act
HSA Health Savings Account
ICOR Item Correction
IRS Internal Revenue Service
MEF Master Earnings File
NH Numberholder
OIG Office of the Inspector General
SSA Social Security Administration
SSN Social Security Number
TY Tax Year
WBDOC Wilkes Barre Data Operations Center
YCER Young Children's Earnings Record
Forms
Form W-2 Wage and Tax Statement

Appendix B
Scope and Methodology

To accomplish our objective, we:

Reviewed prior Social Security Administration (SSA) Office of the Inspector General audit reports.

Reviewed polices and procedures regarding the Single Select and Extended Single Select edit routines.

Reviewed management reports on Single Select and Extended Single Select edit statistics.

Gained an understanding of SSA's Single Select and Extended Single Select edit processes.

Obtained a Master Earning File (MEF) data extract of Single Select and Extended Single Select postings for segment 12 of the MEF. We analyzed this file to determine whether FICA earnings and total compensation amounts were accurately posted to SSA's wage reporting system. We also extracted, summarized, and tested non-FICA earnings and deferred compensation amounts posted as a result of these edit routines.

Obtained an Earning Suspense File (ESF) Reinstates File as of October 2006. Using this information, we determined the number of items reinstated from the ESF to the MEF and the reason for the reinstatements.

Randomly selected 50 earnings items corrected by Single Select and Extended Single Select from a total of 68,741 corrections processed in Calendar Year 2005 for 1 segment of the MEF. We compared the information on the Wage and Tax Statement (Form W-2) reported to SSA, to the name and Social Security number (SSN) recorded within SSA's Numident, the name and SSN on the ESF Reinstates File, and the earnings history of the SSN owner recorded on SSA's MEF.

Randomly selected 50 earnings items corrected by the Prior Reinstate edit from a total of 526,998 corrections posted to the ESF Reinstates File for Tax Year 2005. We then reviewed prior correction activity in both the Reinstates File and MEF to determine the original edit process that led to this later correction.

We determined the MEF data extract used in our review was sufficiently reliable for us to meet our objective. The entity responsible for the Single Select and Extended Single Select edit routines is the Earnings Records Maintenance Branch under the Deputy Commissioner for Systems. Our work was conducted in Philadelphia, Pennsylvania, from December 2006 to June 2007. We conducted our audit in accordance with generally accepted government auditing standards.

Appendix C
Edit Routines and Correspondence
As part of the Annual Wage Reporting process, the Social Security Administration (SSA) uses a variety of manual and automated matching routines to resolve name and Social Security number (SSN) combinations that do not match SSA's records. In addition, SSA sends correspondence to employees and employers in an effort to obtain corrected information.

Edit Processes

SSA uses a variety of editing routines and other processes to correct and post many of the wages items with name/SSN mismatches or other problems, both before and after they go into the Earnings Suspense File (ESF). The following is a description of the various edits.

Overnight Validation: This process identifies name/SSN mismatches on the reporting year's paper Wage and Tax Statement (Form W-2) before routine edits are performed. Any processed name/SSN mismatches go through additional edits and those items not corrected are highlighted in a return electronic transmission from the National Computer Center to the Wilkes-Barre Data Operations Center (WBDOC). WBDOC technicians look at the image of the original Form W-2 and re-key any incorrect items. They also input the employee address, if available, to all returned items-whether corrected or not-for later Decentralized Correspondence (DECOR) and FERRET activities (see below).

Prior Reinstate: This process checks the ESF Reinstate File for instances where the same SSN/Name reporting error associated with the same employer had previously occurred and was corrected. If a match is found, the earnings item is posted to the earner's record associated with the earlier correction.

Single Select: This operation assumes the reported name is correct and the SSN is wrong. Many errors are caused when the name is correct, but there is a transposition error in the SSN. The operation creates "ghost" records from combinations of numbers in the reported SSN with the reported last name. The system then screens these records against their related Numident records-the Numident file is SSA's database of all valid SSNs-creating up to 89 possible variations of the SSN. If one and only one Numident record matches the reported name, the item is reinstated.

ST/Single Select Combination: If no exact match is found on the last name under Single Select, the edit process is extended further to search for a match on the individual's entire name. If the SSN/name is validated through this edit routine, the earnings are posted to that validated SSN. If the extended search fails to validate the name and SSN, the earnings item is posted to the ESF.

Operation 30: This process identifies ESF items with valid SSNs and connects SSA's Numident records with the ESF item. It assumes the SSN is correct, but the name is wrong. Technical staff performs a sight comparison to review reported data against online SSA records and make judgments to accept wage items for Master Earnings File (MEF) posting or send the data back to the ESF.

SWEEP: SWEEP is an electronic operation that periodically uses SSA's latest system enhancements and validation rules, including the more than 20 routine edits used on incoming wages, to remove items from the ESF and reinstate them to wage earners' MEF records.

GAP SWEEP: GAP SWEEP scans earnings records for valid SSNs in the ESF and assesses whether yearly gaps in earnings exist in the MEF record and might be linked to similar earnings in the ESF.

Item Correction (ICOR): This process allows SSA staff to correct the earnings record of an individual through a system called ICOR. The system is a computerized process for adjusting an individual's earnings record thereby helping SSA establish and maintain an accurate and complete MEF. This system allows SSA employees to add, change, move, or delete an individual's earnings overnight via online interactive screens. ICOR is a paperless system-with proofs and rationale recorded electronically after an initial inspection by an SSA employee(s).

Reinstatements from the IRS: While SSA is attempting to resolve mismatched names and SSNs within the ESF, the Internal Revenue Service (IRS) is performing a similar process. The IRS provides SSA a file containing resolved mismatches so that SSA can use this information to locate the owners of suspended items in the ESF.

SSA/IRS Earnings Reconciliation Process: The SSA/IRS earnings reconciliation process compares employee wage data submitted to the IRS to wage data submitted to SSA. Employers, their representatives, third parties and agents submit wage data to both agencies. When more wages are reported to the IRS than to SSA, SSA examines these cases and attempts to resolve any difference without contacting the employer. When this effort is unsuccessful, SSA sends a notice and questionnaire to the employer, requesting information to resolve the case. If SSA does not receive a response within 45 days, the employer is sent a second notice. When no response is received after the second notice, the IRS is responsible for contacting the employer and may impose penalties, if necessary.

FERRET: FERRET is a periodic electronic operation that uses SSA and IRS records to make reinstatements from the ESF to wage earners' records. WBDOC creates the FERRET file-basically an address file-from the nonresponder DECOR file. The FERRET file is processed against IRS tax returns for matches to address data. Names or parts of names are then propagated from the IRS file to create possible reinstatements. These possible matches are then screened using SSA's SSN validation process.

Additional Edits: The Agency has also modified its automated processes to improve the identification of numberholders (NH) related to items in the ESF. SSA stated the new processes use information stored on the earnings and benefits records whereas previous internal edits only used the names and SSNs related to the suspended wages.

Correspondence to Employers and Employees

SSA sends out millions of letters to employers and employees each year requesting additional information to correct suspended wage items. The four main letters sent to employers and employees are (1) DECOR, (2) Educational Correspondence (EDCOR), (3) Earnings After Death (EAD), and (4) Young Children's Earnings Record (YCER).

DECOR: When wage items are posted to the ESF, SSA's system generates notices to employees and employers. The main purpose of DECOR notices is to query employees and employers to resolve SSN and/or name discrepancies. While these notices are usually mailed to employees, letters are mailed to an employer if there is no address for the employee. SSA reviews the returned DECOR notices, validates the information provided, and if appropriate, removes the wage item from the ESF for posting to an individual's MEF record. If individuals do not respond to DECOR notices, their information goes through the FERRET operation (described above).

EDCOR: When SSA processes a Form W-2 report with a name and/or SSN that does not match SSA's records, it sends a notice to the employer. These notices state that SSA received wage items that could not be validated and list up to 500 SSNs in an attachment, but do not provide names. SSA requests that employers file corrected Form(s) W-2 to correct the error(s). The notices sent to employers also specify that mismatches do not imply that incorrect information was intentionally provided and that the letter is not a basis, in and of itself, for an employer to take any adverse action against an employee.

SSA is currently sending notices to all employers who submitted more than 10 Forms W-2 that SSA could not process, and the mismatched forms represent more than 0.5 percent of the total Forms W-2 reported to SSA.

EAD: SSA also has processes in place to detect unusual earnings reports-such as instances where earnings relate to someone recorded as deceased on SSA's records. Under the EAD process, when a date of death is present on the Numident, all earning items reported for TYs after the year of death are placed in the ESF. The earnings are also transmitted to an EAD investigative file so that notices can be printed and mailed to employers and/or earners. SSA sends EAD notices to employers and employees. Employer responses are returned to SSA for processing. If the employer states the individual was working for them, SSA sends a notice to the employee requesting that he or she visit a field office to correct his or her earnings information. At the field office, staff interviews the individual and verifies his or her identification. If the evidence appears valid, SSA personnel reinstate the wages to the proper MEF account. If the employer states the wage earner is deceased, SSA informs the employer to refund the employee's share of the Social Security taxes to the employee's estate or next of kin, and the relevant wages will remain in the ESF. We reviewed the EAD process in a prior audit.

YCER: Another unusual earnings pattern monitored by SSA relates to young earners. Under the YCER process, SSA checks the Date of Birth (DoB) for the SSN on each earnings report. If a DoB indicates that the NH of the SSN is a child under age 7, the earnings will be placed into the ESF. When the Form W-2 reporting process is complete, a YCER investigate file is generated to determine whether the earnings belong to the reported SSN; i.e. a child under age 7. SSA sends YCER notices to employers and employees. Employer responses are returned to SSA for processing. If the employer states the NH's SSN, name and DoB agree with SSA's records, the wages are reinstated to the NH. If the employer states the NH's name and SSN are the same as SSA's records, but the DoB is different, a form is sent to the NH advising him or her to contact the local SSA office to correct the discrepancy. If the employer states the name and/or SSN is different from SSA's records, the information is further researched. If the employer does not return the form or states that the NH of the SSN did not work for them, a letter is sent to the NH of the SSN asking him or her to contact the local SSA field office. We reviewed the YCER process in a prior audit.

Appendix D
Reported Earnings for Five Employers
Seven wage item errors reported by five employers were identified during our review. These seven errors related to excessive non-Federal Insurance Contributions Act (FICA) earnings of approximately $144.7 million posted to the Social Security Administration's (SSA) Master Earnings File (MEF). If we assume the same employer-related errors occurred throughout the entire wage reports of the five employers associated with these seven earnings items, we estimate the overstatement on SSA's MEF could be as high as $64.4 billion. Approximately $26.5 billion of this amount relates to Health Savings Accounts (HSA) and the remaining $37.9 billion relates to deferred compensation. SSA staff noted that the overstated HSA amounts represent more of an employer reporting issue then a program issue since these amounts would not count as income or resources for the purpose of calculating SSA benefits. Table D-1 shows the information reported by the five employers.

Table D-1: Estimate of Overstated Non-FICA Earnings in the Master Earnings File

Employer
Tax Year in Question
Amount Posted to Master Earnings Filea
Total Wage Reports for Tax Year in Question
Medicare Earnings
Reported by Employerb Health Savings Account (HSA)/ Deferred Income Reported by Employerc

Estimated Overstatement in the Master Earnings File
Employer 1 2004 $16,085,074 256 $625,386 $4.1 billion $4.1 billion
Employer 2 2004 $16,085,074 315 $1,139,782 5.1 billion 5.1 billion
Employer 3 2004 $16,085,074 781 $6,876,202 12.3 billion 12.3 billion
Employer 4 2004 $16,085,074 313 $689,725 5.0 billion 5.0 billion
Employer 5 2001 $80,425,368 471 $1,479,346 NAd 37.9 billion
Totals $144,765,664 2,136 $10,810,441 $26.5 billion $64.4 billione
Note a: Represents actual amounts reported by employers on the Wage and Tax Statement (Form W-2) with the exception of Employer 5. In this case, the employer reported $16,085,074 in 5 different deferred income fields, amounting to $80,425,368 per W-2.
Note b: We used Medicare wages since these wages represent total annual compensation rather than the portion of earnings subject to the FICA tax. For instance, someone earning $100,000 in Tax Year 2004 would show $100,000 in Medicare wages and only $87,900 in FICA wages.
Note c: Shows actual amounts reported by employers. Employers 1-4 relate to HSA reported amounts, whereas Employer 5 relates to deferred income.
Note d: SSA's systems could not provide this detail for TY 2001. However, a review of other earnings items associated with this employer indicated that $80,425,368 was consistently reported as deferred income.
Note e: Approximately $26.5 billion of this amount relates to HSAs reported by Employers 1-4 and the remaining $37.9 billion relates to deferred compensation reported by Employer 5.

Appendix E
Agency Comments

MEMORANDUM

Date: September 21, 2007

To: Patrick P. O'Carroll, Jr.
Inspector General

From: Larry W. Dye

Subject: Office of the Inspector General (OIG) Draft Report, "Effectiveness of the Single Select Edit Routine" (A-03-07-17065)--INFORMATION

We appreciate OIG's efforts in conducting this review. Our comments on the recommendations are attached.

Please let me know if we can be of further assistance. Staff inquiries may be directed to Ms. Candace Skurnik, Director, Audit Management and Liaison Staff, at (410) 965-4636.

COMMENTS ON THE OFFICE OF THE INSPECTOR GENERAL'S (OIG) DRAFT REPORT, "EFFECTIVENESS OF THE SINGLE SELECT EDIT ROUTINE" (A-03-07-17065)

Thank you for the opportunity to review and provide comments on this draft report. We would like to stress that the Health Savings Account employer reporting errors found during this review do not affect the way we calculate benefits and do not have any impact on the trust funds.

Our comments on the draft recommendations are as follows.

Recommendation 1

Review the non-Federal Insurance Compensation Act (FICA) earnings cases identified to determine what additional actions SSA can take to ensure: a) the employers are aware of these reporting errors; b) the earlier amounts reported in error are corrected; and c) other similar employer problems occurring during tax year (TY) 2004 are corrected. These additional actions could include additional coordination with the Internal Revenue Service (IRS).

Comment

We agree. We will obtain from OIG the specific employer information for those cases identified during this review. We will contact those employers and encourage them to correct their reporting errors by submitting a Form W2-c (Corrected Wage and Tax Statement). Once we obtain the corrected wage reports (Form W2-c), we will post the corrections to the earnings records and transmit the corrected information to IRS through our normal process. Therefore, no additional contact with IRS is necessary, as they are already involved in the corrected wage reporting process. As for the review of similar employer reporting problems that may have occurred during TY 2004, we do not currently have the available resources to expend on such a task. However, we do have a wage reporting reconciliation process in place with IRS to ensure that employer wage reported totals are in agreement with the reported tax information submitted to IRS. This reconciliation process helps to ensure that reported covered wage earnings are properly accounted for and posted to individual earnings records. As noted in OIG's report, we did take actions to provide better protection against such wage reporting software occurrence in the future. At the time of the discovery of the employer reporting error, we attempted to have the employers discontinue the use of the flawed wage reporting software products and to submit correct wage reports (Forms W2-c).

Recommendation 2

Ensure the planned edits for TY 2007 are established and tested to initiate an investigation of when: a) the average money field on a wage report exceeds the $1 million threshold; and b) a specific number of W-2s also exceed this threshold and, as appropriate, contact employers and/or wage report submitters to confirm the accuracy of these amounts.

Comment

We agree. We have already taken steps to place the edits into the TY 2007 software for the Annual Wage Reporting process. Currently, the TY 2007 software is in the process of moving into the Validation Stage of the System Life Cycle. Implementation is on target for
January 2008.

Appendix F
OIG Contacts and Staff Acknowledgments
OIG Contacts
Walter E. Bayer, Director, Philadelphia Audit Division, (215) 597-4080
Cylinda McCloud-Keal, Audit Manager, Philadelphia Audit Office, (215) 597-0572
Acknowledgments
In addition to those named above:
Brenda Williams, Auditor-in-Charge
Rich Devers, Information Technology Specialist

For additional copies of this report, please visit our web site at www.socialsecurity.gov/oig or contact the Office of the Inspector General's Public Affairs Specialist at (410) 965-3218. Refer to Common Identification Number A-03-07-17065.

Overview of the Office of the Inspector General
The Office of the Inspector General (OIG) is comprised of our Office of Investigations (OI), Office of Audit (OA), Office of the Chief Counsel to the Inspector General (OCCIG), and Office of Resource Management (ORM). To ensure compliance with policies and procedures, internal controls, and professional standards, we also have a comprehensive Professional Responsibility and Quality Assurance program.

Office of Audit
OA conducts and/or supervises financial and performance audits of the Social Security Administration's (SSA) programs and operations and makes recommendations to ensure program objectives are achieved effectively and efficiently. Financial audits assess whether SSA's financial statements fairly present SSA's financial position, results of operations, and cash flow. Performance audits review the economy, efficiency, and effectiveness of SSA's programs and operations. OA also conducts short-term management and program evaluations and projects on issues of concern to SSA, Congress, and the general public.

Office of Investigations
OI conducts and coordinates investigative activity related to fraud, waste, abuse, and mismanagement in SSA programs and operations. This includes wrongdoing by applicants, beneficiaries, contractors, third parties, or SSA employees performing their official duties. This office serves as OIG liaison to the Department of Justice on all matters relating to the investigations of SSA programs and personnel. OI also conducts joint investigations with other Federal, State, and local law enforcement agencies.

Office of the Chief Counsel to the Inspector General
OCCIG provides independent legal advice and counsel to the IG on various matters, including statutes, regulations, legislation, and policy directives. OCCIG also advises the IG on investigative procedures and techniques, as well as on legal implications and conclusions to be drawn from audit and investigative material. Finally, OCCIG administers the Civil Monetary Penalty program.

Office of Resource Management
ORM supports OIG by providing information resource management and systems security. ORM also coordinates OIG's budget, procurement, telecommunications, facilities, and human resources. In addition, ORM is the focal point for OIG's strategic planning function and the development and implementation of performance measures required by the Government Performance and Results Act of 1993.