This is the accessible text file for GAO report number GAO-03-292
entitled 'Purchase Cards: Control Weaknesses Leave the Air Force 
Vulnerable to Fraud, Waste, and Abuse' which was released on December 
20, 2002. 

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

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

United States General Accounting Office: 
GAO: 

Report to Congressional Requesters: 

December 2002: 

Purchase Cards: 

Control Weaknesses Leave the Air Force Vulnerable to Fraud, Waste, and 
Abuse: 

GAO-03-292: 

GAO Highlights: 

Highlights of GAO-03-292, a report to Congressional requesters. 

Why GAO Did This Study: 

In July 2001 and March 2002, GAO testified on significant breakdowns in 
internal controls over purchase card transactions at two Navy sites 
that resulted in fraud, waste, and abuse. As a result, the Congress 
asked GAO to audit purchase card controls at DOD. This report focuses 
on Air Force purchase card controls and addresses whether the overall 
management control environment and key internal controls were effective 
in preventing potentially fraudulent, improper, and abusive purchase 
card transactions. 

What GAO Found: 

Weaknesses in the overall control environment and breakdowns in key
controls relied on to manage the purchase card program leave the Air 
Force vulnerable to fraud, waste, and abuse. Major contributors to the 
weak control environment included excessive numbers of purchase cards, 
with about one purchase card for every seven employees, approving 
official span of control that far exceeded DOD guidelines, and credit 
limits that were 12 to 20 times higher than actual spending. 

Of the five key control activities tested, the Air Force had 
significant control breakdowns in at least three of them—(1) receiving 
of goods and services by someone other than the card holder, (2) 
cardholder reconciliation, and (3) approving official review of the 
cardholder’s reconciled statements. The highest failure rates—69 to 87 
percent—at the four locations tested related to approving official 
review—viewed by DOD as the first line of defense against misuse of the 
purchase card. 

As shown in the table, the control breakdowns resulted in purchases that
were potentially fraudulent, improper, and abusive or questionable. GAO
also identified potentially fraudulent transactions for which supporting
documentation was not available to show the quantity and type of items
purchased. Air Force officials could not recall the purpose of these
transactions. 

Table: Examples of Potentially Fraudulent, Improper, and Abusive or 
Questionable Transactions: 

Types of items purchased: Down payment on a $10,000 sapphire ring; 
Examples of vendors: E-Z Pawn; 
Amount: $2,443. 

Types of items purchased: Suitcases, garment and flight bags,
briefcases; 
Examples of vendors: El Portal, 1-800 Luggage, Patagonia, Franklin 
Covey; 
Amount: $23,760. 

Types of items purchased: Clothes for parachutists, pilots, and others
Examples of vendors: REI, LL Bean, Old Navy, Nordstrom; 
Amount: $23,602. 

Types of items purchased: 2 reclining rocking chairs with full lumbar
support and vibrator-massage features; 
Examples of vendors: LA-Z-Boy Furniture; 
Amount: $1,935. 

Types of items purchased: Tractor rentals; 
Examples of vendors: Crown Ford; Ford Motor; 
Amount: $52,500. 

Types of items purchased: Dinner party and show for visiting general,
including $800 for alcohol; 
Examples of vendors: Treasure Island Hotel and Casino; 
Amount: $2,141. 

[End of table] 

In addition, GAO identified (1) improper transactions related to 
weaknesses in controls relied on to prevent splitting purchases into 
multiple transactions to circumvent micropurchase and cardholder 
transaction limits and (2) the failure to use mandated sources of 
supply. Finally, GAO found that cardholders who abused or improperly 
used the purchase card were not subject to strong disciplinary action 
or consequences. 

The Air Force has taken a number of steps to improve control over the
purchase card program. For example, it implemented automated controls
during fiscal year 2002 to help monitor approving official span of 
control, credit limits, and cardholder reconciliation and approving 
official review of monthly statements. If effectively implemented, 
these controls should help strengthen the overall Air Force purchase 
card control environment as well as controls over statement 
reconciliation and approval. 

What GAO Recommends: 

GAO makes several recommendations to DOD and the Air Force, including 
the following: 

* Reduce the number of purchase card accounts. 

* Minimize credit limits. 

* Reduce approving official span of control consistent with DOD 
guidelines. 

* Establish specific training courses for cardholders, approving 
officials, and agency program coordinators tailored to the specific 
responsibilities associated with their roles. 

* Hold cardholders and approving officials accountable for misuse of the
purchase card. 

DOD and Air Force officials concurred or partially concurred on our 
recommendations and indicated that some actions have been already been 
initiated. 

[hyperlink, http://www.gao.gov/cgi-bin/getrpt?GAO-03-292]: 

To view the full report, including the scope and methodology, click on 
the link above. For more information, contact Gregory Kutz, (202) 512-
9505. 

[End of section] 

Contents: 

Letter: 

Results in Brief: 

Weaknesses in Overall Control Environment: 

Tests of Key Control Activities: 

Potentially Fraudulent, Improper, and Abusive or Questionable
Transactions: 

Management Improvements: 

Conclusions: 

Recommendations for Executive Action: 

Agency Comments and Our Evaluation: 

Appendixes: 

Appendix I: Objectives, Scope, and Methodology: 

Appendix II: Overview of the Air Force Purchase Card Process: 

Appendix III: Examples of Air Force Purchase Card Fraud Cases: 

Appendix IV: GAO Contacts and Staff Acknowledgments: 

Tables: 

Table 1: Ratio of Cardholder Accounts to Approving Officials, August 
2002: 

Table 2: Fiscal Year 2001 Historical Purchases vs. Credit Limits for 
Selected Air Force Case Study Locations: 

Table 3: Lack of Documented Appointment Letters Delegation of 
Purchasing Authority, and Training for Cardholders and Approving 
Officials: 

Table 4: Installation Program Coordinator Span of Control as of 
September 30, 2002: 

Table 5: Internal Control Activity Statistical Testing Results: 

Table 6: Property Items Not Recorded in Property Records: 

Table 7: Potentially Fraudulent Air Force Purchase Card Transactions: 

Table 8: Improper Air Force Purchase Card Transactions: 

Table 9: Purchases from Other Than Required Sources of Supply: 

Table 10: Improper Uses of the Purchase Card and Convenience Checks: 

Table 11: Abusive or Questionable Air Force Purchase Card Transactions: 

Table 12: Installations Audited and Associated Major Commands: 

Table 13: Estimated Results of Statistical Tests of Fiscal Year 2001 
Purchase Card Transactions for Cardholder and Approving Official 
Appointments: 

Table 14: Estimated Results of Statistical Tests of Fiscal Year 2001
Purchase Card Transactions for Cardholder and Approving Official 
Initial Training: 

Table 15: Estimated Results of Statistical Tests of Fiscal Year 2001
Purchase Card Transactions for Cardholder Delegations of Purchasing 
Authority: 

Table 16: Estimated Results of Statistical Tests of Fiscal Year 2001
Purchase Card Transactions for Advance Purchase Authorization: 

Table 17: Estimated Results of Statistical Tests of Fiscal Year 2001
Purchase Card Transactions for Independent Receipt and Acceptance: 

Table 18: Estimated Results of Statistical Tests of Fiscal Year 2001
Purchase Card Transactions for Cardholder Reconciliations: 

Table 19: Estimated Results of Statistical Tests of Fiscal Year 2001
Purchase Card Transactions for Approving Official Review: 

Table 20: Estimated Results of Statistical Tests of Fiscal Year 2001
Purchase Card Transactions for Supporting Invoice or Receipt: 

Table 21: Number and Value of Air Force Fiscal Year 2001 Purchase Card 
Transactions: 

Figures: 

Figure 1: Air Force Purchase Card Program Management Structure: 

Figure 2: Air Force Purchase Card Process: 

[End of section] 

United States General Accounting Office: 
Washington, D.C. 20548: 

December 20, 2002: 

The Honorable Charles E. Grassley: 
Ranking Minority Member: 
Committee on Finance: 
United States Senate: 

The Honorable Stephen Horn: 
Chairman: 
The Honorable Janice D. Schakowsky: 
Ranking Minority Member: 
Subcommittee on Government Efficiency, Financial Management and 
Intergovernmental Relations: 
Committee on Government Reform: 
House of Representatives: 

The Department of Defense (DOD) is promoting departmentwide use of
government purchase cards for obtaining goods and services. DOD 
reported that during fiscal year 2001, about 230,000 cardholders used
purchase cards to make about 10.7 million transactions for goods and
services costing over $6.1 billion. Within these amounts, the Air Force
reported that it used about 80,000 cardholder accounts to make about
3 million purchase card transactions for goods and services costing 
about $1.4 billion. Purchase card transactions include acquisitions at 
or below the $2,500 micropurchase threshold and payments on contracts. 
The use of purchase cards has increased dramatically as agencies have 
sought to eliminate the lengthy process and paperwork associated with 
making small purchases. The benefits of using purchase cards versus 
traditional contracting and payment processes are lower transaction 
processing costs and less red tape for both the government and the 
vendor community. We support the use of a well-controlled purchase card 
program to streamline the government’s acquisition processes. However, 
it is important that agencies have adequate internal control procedures 
to protect the government from fraud, waste, and abuse. 

In July 2001 and March 2002, we testified on significant breakdowns in
internal control over purchase card transactions at two Navy sites in 
San Diego, California. [Footnote 1] As a result of our initial audit of 
purchase card controls at the two Navy sites and continuing concern 
about fraud, waste, and abuse in DOD’s purchase card program, you 
requested that we expand our audits of purchase card controls. Our 
audit of Army [Footnote 2] purchase card controls identified a weak 
internal control environment; ineffective implementation of key control 
activities; and potentially fraudulent, improper, and abusive 
purchases. Our broader audit of Navy [Footnote 3] purchase card 
controls identified continuing control weaknesses and discussed actions 
underway by the Navy to resolve these weaknesses. This report focuses 
on Air Force purchase card controls. 

The objective of our audit of the Air Force purchase card program was to
assess the effectiveness of internal control over purchase card use and
payment of purchase card transactions during fiscal year 2001.
Specifically, we addressed whether (1) the Air Force’s overall control
environment and management of the purchase card program were
effective, (2) the Air Force’s key internal control activities operated
effectively and provided reasonable assurance that purchase cards were
used appropriately, and (3) indications existed of potentially 
fraudulent, [Footnote 4] improper, and abusive or questionable 
transactions. We also identified management actions taken by DOD and 
the Air Force during fiscal year 2002 to improve purchase card 
controls. 

We identified four major Air Force commands that accounted for about
69 percent of total purchase card charges and 65 percent of total
transactions for fiscal year 2001 and selected one location from each 
of the four commands based on the magnitude of purchase card 
transactions and payments for case study analysis. For each of the case 
study locations, we tested a statistical sample of fiscal year 2001 
purchase card transactions and conducted other audit work to evaluate 
the design and implementation of key internal control procedures and 
activities. The results of our audit of the transactions constituting 
the statistical samples can only be projected to the individual 
installations where we performed the testing and cannot be used to 
project the adequacy of control activities at the command level or the 
Air Force as a whole. However, the cumulative results of our work offer 
significant perspective on the adequacy of the design and 
implementation of purchase card controls across the Air Force. 

We also looked for indications of potentially fraudulent, improper, and
abusive or questionable purchases as part of our tests of statistical 
samples of transactions and through analysis of non-representative 
selections of transactions using data mining. Our data mining focused 
on identifying transactions with vendors that were likely to sell 
unauthorized or personal use items for both our case study locations 
and Air Force-wide. Because of the large number of transactions that 
met these criteria, we did not look at all potential abuses of the 
purchase card. Our work was not designed to identify, and we cannot 
determine, the extent of potentially fraudulent, improper, and abusive 
or questionable purchases. See appendix I for details on our scope and 
methodology. 

We performed our audit work from January through mid-November 2002 in
accordance with U.S. generally accepted government auditing standards,
and we performed our investigative work in accordance with standards
prescribed by the President’s Council on Integrity and Efficiency. We
received oral comments on a draft of this report from DOD and Air Force
purchase card officials on December 13, 2002. We addressed the comments
in the “Agency Comments and Our Evaluation” section. 

Results in Brief: 

Our audit of Air Force controls over fiscal year 2001 purchase card 
activity identified control environment weaknesses and breakdowns in key
controls that leave the Air Force vulnerable to fraud, waste, and abuse.
However, we found that the Air Force was aware of many of these
problems and had, in several cases, initiated actions to resolve them. 
Major contributors to the weak control environment included excessive 
numbers of purchase cards and approving official span of control, and 
credit limits that were 12 to 20 times higher than actual spending. As 
of September 2002, the Air Force had about 77,000 purchase cards—about 
1 purchase card for every 7 employees compared to the Navy’s ratio of 1 
purchase card for every 31 employees. 

We also found that although the Air Force had issued standard, Air 
Force-wide purchase card operating procedures, [Footnote 5] employees 
responsible for carrying out purchase card program activities 
frequently did not follow them. For example, many of the problems we 
identified related to failure to follow purchase card guidelines in the 
Federal Acquisition Regulation and DOD and Air Force policies and 
procedures, including guidelines on micropurchases and mandated sources 
of supply. Further, we found that misuse of the purchase card was not 
always subject to strong disciplinary action or consequences, even 
though Air Force operating procedures require installation purchase 
card program coordinators to take appropriate action to document 
violations and preclude their reoccurrence. 

Of the five key control activities we tested, we found that all four 
Air Force locations had significant control breakdowns in at least 
three of them—(1) receiving of goods and services by someone other than 
the cardholder, (2) cardholder reconciliation, and (3) approving 
official review of cardholders’ monthly reconciled statements. Our 
statistical test results at the four case study locations showed that 
failure rates for these controls ranged from 21 to 87 percent, with the 
highest failure rates—69 to 87 percent—relating to approving official 
review of reconciled cardholder statements. The high failure rate for 
approving official review is of particular concern because the Air 
Force uses a “pay and confirm” policy, which is inconsistent with 
governmentwide [Footnote 6] and DOD guidelines, [Footnote 7] on 
reconciliation and payment of purchase card bills. Both Treasury 
guidelines and DOD’s Purchase Card Reengineering Implementation 
Memorandum #3 require that purchase card statements be reconciled and
forwarded for payment in a timely manner and allow pay and confirm only
in instances where purchased items have not been received before
payment is due on the monthly purchase card bills. In contrast, Air 
Force purchase card policy permits cardholder statements to be 
reconciled and approved after payment has been made. While 
conscientious post payment reconciliation and approval processes may 
provide adequate control, the lack of documented evidence of post 
payment reconciliation and approval and the undetected, potentially 
fraudulent transactions identified in our work underscore concerns 
about noncompliance with DOD and Treasury guidelines. In July 2002, Air 
Force management asked its purchase card contractor, U.S. Bank, to 
“shut down” (suspend from use [Footnote8]) over 4,000 unreconciled 
cardholder accounts until the reconciliations were completed and the 
approving officials had reviewed and approved them. Accounts that had 
not been reconciled as of the end of August 2002 were canceled. 

Further, our audit identified potentially fraudulent, improper, and 
abusive or questionable purchase card transactions. Some of the 
potentially fraudulent transactions we identified appear to be due to 
compromised accounts related to nationwide credit card fraud. [Footnote 
9] While Air Force cardholders identified some of these potentially 
fraudulent transactions and took steps to dispute them with U.S. Bank, 
other potentially fraudulent purchases were not disputed, even though 
the cardholders, approving officials, or program coordinators were 
aware of them. Potentially fraudulent transactions that were not 
disputed included a $2,443 down payment at a pawn shop on a $10,000 
sapphire ring and several purchases totaling $3,232 at San Diego area 
stores, such as Old Navy, Target, K-Mart, and a Ross Store. We also 
identified potentially fraudulent transactions for which no supporting 
documentation was available to show the quantity and type of items 
purchased. Air Force officials could not recall the purpose of these 
transactions. 

In addition, we identified numerous examples of improper and abusive
purchase card transactions. Improper transactions included purchases of
food, clothing, luggage, briefcases, and personal items, such as 
sunglasses. We also identified improper transactions related to 
weaknesses in controls relied on to prevent splitting purchases into 
multiple transactions to circumvent micropurchase and cardholder 
transaction limits and the failure to use mandated sources of supply. 
One such purchase we identified involved a fiscal 2001 year-end 
purchase of about $100,000 in helmets from an installation supply 
store. The purchase was split into four separate transactions to stay 
within the cardholder’s $25,000 transaction limit. We also determined 
that although the purchase was made to prevent unused funds from 
expiring, all the helmets were not actually needed, and the cardholder 
left the items in the store until requirements could be better 
identified. On October 1 and 2, 2001, the cardholder obtained credits 
for the unneeded items and used the credits to purchase other items. The
subsequent credits and reuse of the funds in early October 2001, in 
effect, converted fiscal year 2001 appropriations to fiscal year 2002 
budget authority—a violation of appropriation law. 

In addition to improper transactions, we found numerous examples of
purchase card transactions that involved waste and abuse. One of the
abusive transactions we identified related to a Nellis Air Force Base 
(AFB) dinner party to entertain a visiting Joint Forces General. Air 
Force policy permits installation commanders to use representation 
funds to conduct entertainment on a modest basis. However, we 
determined that the cost of the dinner party, which totaled $2,141—over 
$100 per person—was excessive. Further, the total cost of the dinner 
included about $800 for alcohol—over $40 per person. 

We also found instances of questionable purchases and wasteful spending
related to computer equipment. For example, we found that a Travis AFB
unit purchased computer equipment costing $14,128 at the beginning of
fiscal year 2001, and, shortly thereafter, decided to convert to Dell
computers. As a result, within 1 year of their purchase, these items, 
as well as a number of other computers, were sent to the Defense 
Reutilization Marketing Service as excess property. We also found that 
an Edwards AFB unit purchased several computers at the end of fiscal 
year 2001 in anticipation of hiring during the first quarter of fiscal 
year 2002. Hiring was delayed and we found that about half the computer 
equipment was still stored in boxes 8 months later, raising questions 
about whether fiscal year 2001 funds should have been used to purchase 
these items. Appropriated funds are available only to meet legitimate 
needs of the agency during the fiscal year for which the funds were 
appropriated. 

The Air Force has taken a number of steps to improve controls over the
purchase card program, particularly with respect to the overall purchase
card program management control environment. For example, the Air
Force worked with its contractor, U.S. Bank, to implement automated
controls during fiscal year 2002 to help monitor approving official 
span of control, credit limits, and cardholder reconciliation and 
approving official review of monthly purchase card statements. The U.S. 
Bank controls permit purchase card accounts to be automatically 
suspended when specified control requirements, such as cardholder 
reconciliations and approving official span of control, are not 
followed. If effectively implemented, these controls should help 
strengthen the overall Air Force purchase card control environment as 
well as controls over purchase card statement reconciliation and 
approval. In addition, in response to our DOD purchase card audits, the 
Congress recently enacted the DOD fiscal year 2003 appropriation and 
authorization acts, which contain requirements for DOD to develop 
guidelines on disciplinary actions for employees who abuse or 
fraudulently use the purchase card. The DOD authorization act also 
includes a number of provisions for purchase card management
improvements. 

This report contains recommendations to the Air Force to further improve
the overall control environment for its purchase card program; to 
strengthen key internal control activities; and to increase attention to
preventing potentially fraudulent, improper, and abusive and 
questionable transactions. We also recommend that the DOD task force 
assess the DOD-wide applicability of both the recommendations addressed 
to the Air Force and the strengths in the Air Force purchase card 
program that we identified, such as the use of automated controls. In 
oral comments on a draft of this report, DOD’s Purchase Card Joint 
Program Management Office and Air Force purchase card officials 
concurred on 29 of our 39 recommendations and partially concurred with 
9 recommendations. At the time we finalized our work, DOD had not 
provided a response to our remaining recommendation that the Charge 
Card Task Force assess the recommendations in this report and 
incorporate them to the extent applicable into its future 
recommendations to improve purchase card policies and procedures 
throughout DOD. 

For the nine recommendations involving partial concurrences, the issues
raised by DOD and Air Force officials generally related to details such 
as the terminology or responsible party included in our recommendations.
However, in each case, the officials suggested actions that, if 
effectively implemented, would address the intent of our 
recommendations. A detailed discussion of the DOD and Air Force 
comments is presented in the “Agency Comments and our Evaluation” 
section of this report. 

Background: 

The Air Force purchase card program is part of the governmentwide
Commercial Purchase Card Program established to streamline federal
agency acquisition processes by providing a low-cost, efficient vehicle 
for obtaining goods and services directly from vendors. DOD has mandated
the use of the purchase card for all purchases at or below $2,500 and 
has authorized the use of the card to pay for specified larger 
purchases. Purchases over the $2,500 micropurchase threshold must be in 
accordance with simplified acquisition guidelines in the Federal 
Acquisition Regulation. The purchase card should normally not be used 
for cash advances; travel-related purchases; rentals or leases of land 
or buildings; utility services; or hazardous/dangerous items, such as 
explosives, munitions, toxins, and firearms. 

The Air Force purchase card program operates under a task order 
associated with the General Services Administration’s governmentwide
purchase card contract, as do the purchase card programs of all federal
agencies. Government acquisition laws and regulations, such as Federal
Acquisition Regulation, Part 13, “Simplified Acquisition Procedures,”
establish the criteria for using purchase cards to place orders and make
payments. The Air Force issued Instruction 64-117, Governmentwide
Purchase Card Program, dated December 6, 2000, to establish
responsibilities and procedures and provide administrative guidance for 
its government purchase card operations. The Air Force Instruction 
contains standardized Air Force guidance that has been implemented Air 
Forcewide. 

The Assistant Secretary of the Army for Acquisition, Logistics, and
Technology in coordination with the Under Secretary of Defense
(Comptroller), has overall responsibility for DOD’s purchase card 
program. The DOD Purchase Card Joint Program Management Office, in the 
office of the Assistant Secretary of the Army for Acquisition, 
Logistics, and Technology [Footnote 10] is responsible for overseeing 
DOD’s program. The Air Force agency program coordinator, within the 
headquarters acquisition office has oversight over the Air Force 
purchase card program. However, the primary management responsibility 
for the purchase card program lies with the Assistant Secretary of the 
Air Force for Acquisition, the Deputy Assistant Secretary for 
Contracting, and the contracting offices in the major Air Force 
commands and local installations. Figure 1 shows the Air Force purchase 
card program management hierarchy as it was during our audit. For the 
major commands, the figure shows the number of installation program 
coordinators within the command. For the four installations we audited, 
the figure shows the number of approving officials and cardholders at 
each installation. 

Figure 1: Air Force Purchase Card Program Management Structure: 

[See PDF for image] 

This figure is an illustration of the Air Force Purchase Card Program 
Management Structure, as follows: 

* Assistant Secretary of the Air Force for Acquisition, Logistics, and 
Technology: 

* DOD Purchase Card Program Management Office: 

Both of the above connect to: 

* Air Force Agency Program Coordinator: 

The coordinator is connected to the following: 

* Air Combat Command; Command Program Coordinator: 
- Edwards AFB, CA; 
- Approving officials - 122; 
- Cardholders - 405. 

* Air Mobility Command; Command Program Coordinator: 
- Travis AFB, CA; 
- Approving officials - 131; 
- Cardholders - 486. 

* Air Force Materiel Command; Command Program Coordinator: 
- Nellis AFB, NV; 
- Approving officials - 99; 
- Cardholders - 542. 

Air Force Education and Training Command; Command Program Coordinator: 
- Lackland AFB, TX; 
- Approving officials - 113; 
- Cardholders - 569. 

Source: GAO analysis of Air Force purchase card program organization. 

[End of figure] 
 
At each Air Force installation, a certifying officer in the installation
Financial Services Office is responsible for certifying the monthly 
purchase card statements for payment. This is done within a few days of 
receipt of the monthly statement. Accuracy of the monthly statements is 
confirmed after payment has been made. Personnel in three 
positions—program coordinator, approving official, [Footnote 11] and 
cardholder—are collectively responsible for providing reasonable 
assurance that purchase card transactions are appropriate and meet a 
valid government need. The installation program coordinator, typically 
a full-time position under the director of the contracting squadron, is 
responsible for the day-to-day management, administration, and 
oversight of the program. Installation program coordinators issue and 
cancel purchase cards, train cardholders and approving officials, and 
coordinate with other Air Force units and U.S. Bank—the Air Force’s 
purchase card-issuing bank. Approving officials, who are typically 
responsible for more than one cardholder, are to review cardholders’ 
transactions and the cardholders’ monthly, reconciled purchase card 
statements and approve the statements as already paid or ensure that 
any invalid transactions are disputed and credited, either by the 
merchant or the bank. In accordance with Air Force Instruction 64-117,
cardholders are required to reconcile their monthly purchase card
statements and approving officials are required to review and approve 
the reconciled statements within 15 days after receipt of the 
statements at the installation but not later than the 15th day of the 
month following the statement date. Approving officials receive a 
monthly, consolidated statement that covers their cardholders’ monthly 
purchase card statements. Appendix II provides additional details on 
the Air Force purchase card program. 

Weaknesses in Overall Control Environment: 

[Management and employees should establish and maintain an environment 
throughout the organization that sets a positive and supportive 
attitude toward internal control and conscientious management. A 
positive control environment is the foundation for all other standards. 
It provides discipline and structure as well as the climate which 
influences the quality of internal control. GAO’s Standards for 
Internal Control in the Federal Government (GAO/AIMD-00-21.3.1, 
November 1999)] 

We found overall control environment weaknesses at the four case study
locations we audited as well as indications of similar weaknesses in 
our Air Force-wide analysis that contributed to breakdowns in key 
control activities and potentially fraudulent, improper, and abusive 
purchase card transactions. For example, we encountered numerous 
instances where supporting documentation was not available because 
installations destroyed purchase card records on a rolling 1-year basis 
due to faulty records retention guidance in the Air Force purchase card 
Instruction that did not comply with federal guidelines. We also found 
weaknesses in the areas of (1) the number of cardholders and accounts, 
(2) approving official span of control, (3) credit limits compared to 
historical spending, (4) documentation of cardholder and approving 
official training, (5) implementation of audit and internal review 
recommendations, and (6) accountability and disciplinary action. 
Further, given the magnitude of the purchase card program at the 
installations we audited, we found the human capital infrastructure for 
program monitoring and oversight to be inadequate. As discussed in a 
later section of this report, these control environment weaknesses have 
contributed to fraudulent, improper, and abusive purchase card 
activity. 

We also found some positive aspects of the Air Force purchase card 
program, such as Air Force-wide purchase card operating procedures and
aggressive Air Force Audit Agency reviews of installation purchase card
programs since 1996. The importance of the role of management in
establishing a positive internal control environment cannot be 
overstated. GAO’s Standards for Internal Control in the Federal 
Government, [Footnote 12] discusses management’s key role in 
demonstrating and maintaining an organization’s integrity and ethical 
values, especially in setting and maintaining the organization’s 
ethical tone, providing guidance for proper behavior, and removing 
temptations for unethical behavior. 

Failure to Comply with Federal Records Retention Requirements: 

During our audit of Air Force fiscal year 2001 purchase card activity, 
we encountered numerous instances where supporting documentation was
not available because installations destroyed purchase card records on a
rolling 1-year basis due to faulty records retention guidance in the Air
Force purchase card Instruction. Federal records retention requirements
in the Federal Acquisition Regulation, [Footnote 13] the General 
Records Schedule [Footnote 14] established by the National Archives and 
Records Administration (NARA), and DOD’s Financial Management 
Regulation [Footnote 15] require records supporting program and 
contracting activity to be retained for 3 years and records supporting 
financial transactions to be retained for 6 years and 3 months. 
However, the records retention guidance in Air Force Instruction 64-117 
called for documentation received and generated by the cardholder, such 
as vendor invoices, sales receipts, and shipping reports, and 
cardholder logs to be maintained for only 1 year after the final 
payment. In addition, 36 C.F.R. 1228.30, which covers disposition of 
federal records, requires agencies to establish a schedule for 
disposing of their records and obtain approval of their records 
disposition schedules from the NARA. We saw no evidence that the Air 
Force developed a records disposition schedule related for purchase 
card records. 

While missing records affected all aspects of purchase card controls, 
the greatest impact was on cardholder and approving official 
appointments and training and cardholder delegations of purchasing 
authority. Further, we found that cardholders at Wilford Hall Medical 
Center, located at Lackland Air Force Base (AFB), did not adhere to the 
Air Force requirement to retain purchase card transaction records for 
even the 1-year period. As discussed in the next section of this report 
on tests of key controls, 52 of the 152 transactions in our Lackland 
AFB sample were for Wilford Hall purchase card activity, and required 
supporting documentation was not available for 23 of these 52 
transactions. 

The Air Force Instruction also called for documentation generated by the
installation program coordinator and approving officials, such as 
records of training, delegations of authority, and surveillances, to be 
retained only as long as the cardholder and approving official are 
performing that function. Air Force headquarters officials told us that 
the Air Force Instruction, which is currently being revised, would 
include corrections to the records retention guidelines. 

Proliferation of Cardholders and Accounts Result in Unreasonable Span of
Control: 

While the number of Air Force purchase cardholders peaked at about 
80,000 cardholder accounts in September 2001, the overall number of
cardholders from October 2000 through September 2002 has remained
about the same. As of September 2002, the Air Force reported that it had
about 77,000 purchase card accounts—translating to about 1 purchase card
for every 7 employees. In contrast, the Navy had reduced the number of 
its purchase cardholders from about 52,000 to about 23,000 and only 
about 1 of every 31 employees was a purchase cardholder. 

We determined that the Air Force did not have specific policies 
governing the number of cards to be issued or criteria for identifying 
employees eligible for the privilege of cardholder status. Purchase 
cards were given out on the basis of a request from an individual 
employee’s unit commander. The request was then forwarded to the 
installation’s purchase card Agency program coordinator, who approved 
the request and began the process for obtaining a new card from U.S. 
Bank. 

In addition to an excessive number of cardholders, we found that some 
Air Force cardholders had as many as 10 government purchase cards. The 
Air Force permits cardholders to have numerous purchase card accounts to
facilitate accounting for purchase card expenditures related to 
different funding sources—appropriated and nonappropriated funds—as 
well as different accounting lines within the same fund source. 
Assigning multiple purchase cards is not an appropriate method of 
accounting for purchase card transactions. Further, this practice 
places substantial credit risk in the hands of one individual. 

Some Approving Officials’ Span of Control over Purchase Card Accounts
Exceeds DOD Guidelines: 

In response to concerns about approving official span of control raised
during our initial Navy purchase card work, the Director of DOD’s 
Purchase Card Joint Program Management Office issued a memorandum
on July 5, 2001, that called for no more than five to seven cardholders 
per approving official. [Footnote 16] During fiscal year 2002, the Air 
Force established goals for reducing the number of cardholder accounts 
assigned to approving officials. Our analysis of span of control ratios 
at the four installations we audited disclosed that when looking at 
average ratios, the Air Force has adhered to the DOD span of control 
guidelines. However, as shown in table 1, these averages have masked 
the wide range of ratios across each Air Force installation—some of 
which exceeded the DOD guidelines. As of August 2002, we found that the 
four installations we audited had from 22 to 32 approving officials 
that were responsible for more than 7 cardholder accounts, including 
four approving officials at two installations—Travis AFB and Edwards 
AFB—that were responsible for more than 20 cardholder accounts. 

Table 1: Ratio of Cardholder Accounts to Approving Officials, August 
2002: 

Air Force location: Edwards; 
Average ratio of cardholder accounts to approving officials: 4.3 to 1; 
Number of approving officials: 122; 
Number of cardholder open accounts[A]: 530; 
Percentage and number of approving officials with over seven cardholder 
monthly account statements to review: 18% (22); 
Highest ratio of cardholder accounts to approving officials: 29. 

Air Force location: Lackland; 
Average ratio of cardholder accounts to approving officials: 5.9 to 1; 
Number of approving officials: 113; 
Number of cardholder open accounts[A]: 662; 
Percentage and number of approving officials with over seven cardholder 
monthly account statements to review: 27% (30); 
Highest ratio of cardholder accounts to approving officials: 20. 

Air Force location: Nellis; 
Average ratio of cardholder accounts to approving officials: 6.4 to 1; 
Number of approving officials: 99; 
Number of cardholder open accounts[A]: 638; 
Percentage and number of approving officials with over seven cardholder 
monthly account statements to review: 32% (32); 
Highest ratio of cardholder accounts to approving officials: 19. 

Air Force location: Travis; 
Average ratio of cardholder accounts to approving officials: 5.4 to 1; 
Number of approving officials: 131; 
Number of cardholder open accounts[A]: 702; 
Percentage and number of approving officials with over seven cardholder 
monthly account statements to review: 20% (26); 
Highest ratio of cardholder accounts to approving officials: 31. 

Source: GAO analysis of U.S. Bank data provided by the Air Force. 

[A] Includes multiple accounts for individual cardholders. 

[End of table] 

As shown in table 1, the percentage of approving officials whose span of
control exceeded DOD guidelines ranged from 18 percent to 32 percent at
the four Air Force locations we audited. 

Credit Limits Exceed Procurement Needs: 

Our analysis of purchase card spending compared with credit limits 
showed that monthly credit limits for the four case study locations far
exceeded their actual monthly spending. Limiting credit available to
cardholders is a key factor in managing the purchase card program and in
minimizing the government’s financial exposure. On August 13, 2001,
DOD’s Director of Defense Procurement sent a memorandum to the 
directors of all defense agencies stating that supervisors should set
reasonable limits based on what each person needs to buy as part of his 
or her job and that every cardholder does not need to have the maximum
transaction or monthly credit limit. 

Air Force officials told us that their credit limits appear excessive 
because these limits include credit limits of primary and alternate, or 
backup, approving officials and cardholders. The officials explained 
that they assign primary and alternate approving officials and 
cardholders to local units to ensure that purchases can continue when 
primary approving officials or cardholders are on leave, assigned to 
temporary duty locations, or deployed. The officials also told us that 
alternate approving officials and cardholders are given the same credit 
limits as the primary approving officials and cardholders. The Director 
of DOD’s Purchase Card Joint Program Management Office told us that DOD 
guidelines suggest that credit limits for inactive accounts, including 
alternate accounts, should be reduced to $1. However, Air Force 
officials told us that they do not deactivate alternate accounts or 
reduce credit limits during periods when alternate purchasing authority 
is not needed due to frequent schedule changes and the administrative 
burden associated with turning accounts on and off. As shown in table 
2, total financial exposure as measured in terms of purchase card 
credit limits substantially exceeded historical purchase card spending. 

Table 2: Fiscal Year 2001 Historical Purchases vs. Credit Limits for 
Selected Air Force Case Study Locations (Dollars in millions): 

Historical measures: Total fiscal year 2001 purchases; 
Edwards: $23.8; 
Lackland: $36.2; 
Nellis: $27.7; 
Travis: $24.7. 

Historical measures: Average monthly purchases; 
Edwards: $2.0; 
Lackland: $3.0; 
Nellis: $2.3; 
Travis: $2.7. 

Historical measures: Average monthly credit limit[A]; 
Edwards: $24.1; 
Lackland: $36.7; 
Nellis: $42.3; 
Travis: $55.4[B]. 

Historical measures: Ratio of credit limit to average fiscal year 2001 
monthly purchases; 
Edwards: 12 to 1; 
Lackland: 12.2 to 1; 
Nellis: 18.4 to 1; 
Travis: 20.5 to 1. 

Source: GAO analysis of U.S. Bank data provided by the Air Force. 

[A] We used approving officials’ credit limits to calculate average 
monthly credit limits. 

[B] Travis AFB did not retain monthly reports for fiscal year 2001; 
therefore, we used the Travis AFB October 2001 monthly report as an 
indication of the average credit limit. 

Air Force officials told us that to better control cardholder spending 
limits, they worked with U.S. Bank to develop an automated control that 
will tie quarterly authorizations of budget authority to cardholder 
credit limits. According to Air Force officials, this control was 
implemented during fiscal year 2002. 

Lack of Documentation of Cardholder and Approving Official Training and
Authority: 

[Effective management of an organization’s workforce—its human 
capital—is essential to achieving results and an important part of 
internal control. Training should be aimed at developing employee skill 
levels to meet changing organizational needs. GAO’s Standards for 
Internal Control in the Federal Government (GAO/AIMD-00-21.3.1, 
November 1999)]. 

Documentation of cardholder and approving official training and 
authority varied widely across the four locations we audited. For 
example, we found that two of our four case study locations had 
maintained documentation on cardholder and approving official 
appointments, cardholder delegations of purchasing authority, and 
cardholder and approving official training. However, at Edwards AFB and 
Travis AFB, our test results indicated that 12 percent [Footnote 17] 
and 51 percent, [Footnote 18] respectively, of their fiscal year 2001 
purchase card transactions were made by cardholders and/or approving 
officials who had no documentation showing they had received initial
training. Further, none of the four case study locations we audited had
adequate documentation to show that both cardholder and approving
official training was current for the transactions in our sample. 

Air Force Instruction 64-117 requires commanders or chiefs to prepare a
Letter of Appointment designating the proposed cardholder and approving
official and identifying their name, rank, duty title, telephone 
number, and e-mail address; the types of purchases to be made; and 
funds to be used to pay for the purchase card purchases. The 
installation program coordinator is to coordinate single and monthly 
purchase limits with the designated billing official and forward 
documentation to the bank to set up the purchase card account. While 
the Instruction provides for denial of appointments, it does not 
provide criteria or qualifications for who may be a cardholder or 
approving official. 

The Instruction also requires that prior to establishing a purchase card
account and issuing a purchase card, all prospective cardholders and
approving officials must receive training on requirements in the Federal
Acquisition Regulation and Air Force purchase card and acquisition
policies and procedures. Once initial training is received, the 
Instruction requires all cardholders to receive supplemental training 
in the form of annual refresher training and it implies that approving 
officials should have refresher training. [Footnote 19] The Air Force 
Instruction requires installation program coordinators to maintain 
documentation of approving official and cardholder purchase card 
appointments and training and cardholder delegations of purchasing 
authority as long as these individuals are serving in those capacities. 

Contracting officials at Travis AFB explained that in accordance with 
Air Force Instruction 64-117, they did not retain records of 
appointments, training, and delegations of authority for approving 
officials and purchase cardholders who were no longer performing those 
duties or were no longer at Travis AFB. We also found that Travis AFB 
did not maintain central files to document cardholder and approving 
official appointments and cardholder delegations of purchasing 
authority, and documentation of training was incomplete or could not be 
located by individual units at that installation. In addition, Edwards 
AFB contracting officials told us that they did not retain cardholder 
and approving official appointment letters. Table 3 shows the results 
of our statistical tests for documentation of appointments, delegations 
of purchasing authority, and training for cardholders and approving 
officials associated with the transactions in our sample. 

Table 3: Lack of Documented Appointment Letters, Delegation of 
Purchasing Authority, and Training for Cardholders and Approving 
Officials: 

Key control: Cardholder and approving official appointment letters; 
Percentage of breakdowns in key purchase card controls[A], Edwards: 82; 
Percentage of breakdowns in key purchase card controls[A], Lackland: 0; 
Percentage of breakdowns in key purchase card controls[A], Nellis: 0; 
Percentage of breakdowns in key purchase card controls[A], Travis: 97. 

Key control: Initial training; 
Percentage of breakdowns in key purchase card controls[A], Edwards: 12; 
Percentage of breakdowns in key purchase card controls[A], Lackland: 0; 
Percentage of breakdowns in key purchase card controls[A], Nellis: 0; 
Percentage of breakdowns in key purchase card controls[A], Travis: 51. 

Key control: Cardholder delegations of purchasing authority; 
Percentage of breakdowns in key purchase card controls[A], Edwards: 13; 
Percentage of breakdowns in key purchase card controls[A], Lackland: 0; 
Percentage of breakdowns in key purchase card controls[A], Nellis: 1; 
Percentage of breakdowns in key purchase card controls[A], Travis: 84. 

[A] The projections represent point estimates for the populations based 
on our statistical samples rounded to the nearest percentage point. The 
intervals are two-sided 95-percent confidence intervals. 

Source: GAO testing and statistical analysis of Air Force purchase card 
transaction files. 

[End of table] 

Although Edwards AFB and Travis AFB officials asserted that their
cardholders and approving officials had received required purchase card
training, we found numerous improper purchase card transactions related
to failure to follow federal guidelines on micropurchases and mandated
sources of supply that indicate that cardholders and approving officials
either had not been trained, training was inadequate, or cardholders 
were not following guidelines addressed in the training classes. For 
example, we found that cardholders at Edwards AFB were unaware that 
they are required to obtain competitive price quotes from three 
different vendors when their purchases exceed the $2,500 micropurchase 
threshold. We also found that Travis AFB cardholders frequently were 
not following these guidelines. Further, we found that the Edwards AFB 
installation program coordinator routinely waived micropurchase and 
cardholder transaction limits to permit cardholders to purchase goods 
and services at higher amounts. We also found that Edwards AFB 
cardholders frequently did not follow federal guidelines on mandated 
sources of supply and Travis AFB cardholders did not notify property 
book officers when they used the purchase card to buy computer 
equipment and other pilferable property. 

In addition, we were unable to determine whether cardholder and
approving official training was current due to missing documentation and
the use of E-mail notices, bulletins, and newsletters used to update
purchase card training at three case study locations. Travis AFB lacked
documented evidence that most of its cardholders and approving officials
had received annual refresher training and the other three case study
locations could not document informal training that was provided outside
of classrooms. During fiscal year 2002, due to concerns about whether
approving officials and cardholders had received adequate refresher
training, Edwards AFB and Nellis AFB began using classroom training and
sign-in sheets to document annual purchase card refresher training for 
both approving officials and cardholders. 

Significant Audit Activity, but Corrective Actions Not Always Taken: 

[Monitoring of internal control should include policies and procedures 
for ensuring that the findings of audits and other reviews are promptly 
resolved. GAO’s Standards for Internal Control in the Federal 
Government (GAO/AIMD-00-21.3.1 November 1999)]. 

We found that the four case study locations conducted annual purchase
card internal reviews, called surveillances, and the Air Force Audit 
Agency performed aggressive reviews; [Footnote 20] however, we also 
identified significant repeat findings indicating a lack of commitment 
to adhere to purchase card regulations and DOD and Air Force policies 
and procedures. In addition, at one of the four case study locations we 
audited—Edwards AFB—internal reviewers did not disclose all identified 
problems in the reports on surveillance results. The Air Force Audit 
Agency reported [Footnote 21] similar findings in its reviews of fiscal 
year 2000 installation-level purchase card activity. The Air Force 
auditors concluded that without enforcement of purchase card 
requirements, such as administratively disciplining the offenders or 
revoking purchase card privileges, purchase card discrepancies will 
likely continue. The Air Force Audit Agency reported that inaction by 
contracting officials reduced the effectiveness of other purchase 
controls, increased transaction errors, and undermined the Air Force 
purchase card goals. 

As stated in GAO’s Standards for Internal Control in the Federal
Government, monitoring of internal control should include policies and
procedures for ensuring that the findings of audits and other reviews 
are promptly resolved. Managers are to (1) promptly evaluate findings 
from audits and other reviews, including those showing deficiencies, and
recommendations reported by auditors and others who evaluate agency
operations, (2) determine proper actions in response to findings and
recommendations from audits and reviews, and (3) complete, within
established time frames, all actions that correct or otherwise resolve 
the matters brought to management’s attention. 

Internal Surveillances: 

The Air Force purchase card Instruction, which was issued in December
2000, requires the installation purchase card program coordinator to
perform a surveillance (internal review) of each approving official’s 
billing account [Footnote 22] and 25 percent of the cardholders’ 
accounts at least every 12 months. The Air Force Instruction includes 
an optional surveillance checklist that covers key purchase card 
requirements, including most of the control activities covered in our 
audit. Our review of the surveillances associated with the approving 
officials and cardholders responsible for the transactions in our case 
study samples showed that many of the problems identified by 
installation internal reviewers were consistent with the problems we 
found in our audit. For example, these surveillance results identified 
failure to adhere to Air Force guidance on $2,500 micropurchase limits, 
purchases from required sources, failure of billing officials to review
their cardholders’ accounts, prohibited purchases, and failure to 
maintain purchase logs. 

Further, we determined that Edwards AFB intentionally did not address
problems related to noncompliance with micropurchase requirements and
mandated sources of supply nor did it recommend corrective actions in 
its reports on surveillance results. For example, 40 of the 44 Edwards 
AFB surveillance reports that we reviewed stated “no discrepancies were
noted,” even though documentation supporting this conclusion for 11
surveillance reports indicated that problems, such as splitting 
purchases to avoid obtaining competitive prices when purchases exceeded 
the $2,500 micropurchase threshold and failure to use mandated vendors, 
were identified. Further, while use of the surveillance checklist is 
optional, we found that several of the checklists provided as 
documented support for the surveillances that we reviewed were blank. 
As a result, we were unable to determine whether a review had been 
performed, but was not documented, or if no review had been performed 
as a basis for the conclusions in the associated surveillance letters. 
When we discussed our concerns with the Director of Contracting at 
Edwards AFB, the Director told us that internal reviewers do not 
address splitting purchases to avoid micropurchase requirements and 
failure to use mandated sources of supply in their surveillance reports 
because they consider these requirements to be “procedural matters.” 
However, the guidelines on micropurchases and mandated sources of 
supply are prescribed in law and the Federal Acquisition Regulation and 
are not merely procedural. It is important that surveillance reports 
identify noncompliance with requirements in law and regulations, which 
are designed to prevent improper purchases. Air Force headquarters 
officials agreed with our position. 

In addition, we found that installation program coordinators at two of 
our four case study locations had issued their surveillance reports to 
approving officials rather than unit commanders. In contrast, Nellis 
AFB surveillance letters were addressed to unit commanders and were 
signed by the Contracting Director. Further, the Nellis AFB Contracting 
Director wrote personal notes on the face of the surveillance letters 
commending unit commanders for good surveillance results and noting 
areas that must be improved when the surveillances had identified 
failure to follow purchase card guidelines. This positive “tone at the 
top” served to hold unit commanders accountable for effective 
implementation of their purchase card programs. 

Air Force Audit Agency Audits: 

In its August 2002 report on the results of its audit of fiscal year 
2000 purchase card transactions at 46 Air Force installations, Air 
Force Audit Agency auditors concluded that overall, Air Force guidance 
established adequate purchase card controls and oversight procedures. 
At the same time, Air Force auditors found that installation purchase 
card program coordinators and approving officials did not adhere to 
this guidance in executing their surveillance responsibilities. For 
example, consistent with our audit, Air Force auditors found continuing 
problems with (1) advance approval for purchases of computer equipment, 
(2) splitting purchases into multiple transactions to circumvent 
micropurchase and cardholder single transaction limits, (3) 
accountability for pilferable property items purchased with a purchase 
card, and (4) purchase card statement reconciliations and approving 
official review. Further, according to Air Force auditors, of the 
nearly $150 million in transactions evaluated, cardholders acquired 
supplies and services totaling approximately $25.4 million using 
purchase methods specifically disallowed under established policy and 
guidance. 

Specifically, Air Force auditors reported the following findings 
related to their audits of purchase card activity during fiscal year 
2000 at Edwards, Lackland, and Travis Air Force bases and purchase 
activity during fiscal year 2001 at Nellis AFB. 

* At Edwards AFB, Air Force auditors had repeat findings [Footnote 23] 
of split purchases and lack of required advance purchase authorizations 
for purchases of computer equipment. The auditors also found that 
cardholders were not familiar with micropurchase requirements. 

* At Lackland AFB, Air Force auditors reported [Footnote 24] that the 
issues of improper reconciliation procedures to validate purchases and 
charges and ineffective surveillance of cardholders and billing 
officials were repeat conditions reported in a prior audit report. 

* Air Force auditors reported [Footnote 25] a repeat finding that 
Travis AFB did not always maintain property accountability for items 
costing over $500. In addition, the auditors reported that cardholders 
did not always obtain required advance authorizations for purchases of 
computer and communication equipment. 

* Air Force Audit Agency auditors reported [Footnote 26] that the 99th 
Wing at Nellis AFB effectively controlled purchase card purchases, 
noting that purchases were properly approved, recorded in cardholder 
logs, and that approving officials reviewed cardholder records monthly. 
However, the auditors found that some units improperly purchased 
bottled water and that opportunity existed for unit commanders to 
ensure greater visibility over unit purchases by reviewing automated 
records in U.S. Bank’s database. 

Weaknesses in Accountability and Lack of Disciplinary Actions: 

[Management plays a key role in demonstrating and maintaining an 
organization’s integrity and ethical values, especially in setting and 
maintaining the organization’s ethical tone, providing guidance for 
proper behavior, removing temptations for unethical behavior, and 
providing discipline when appropriate. GAO’s Standards for Internal 
Control in the Federal Government (GAO/AIMD-00-21.3.1 November 1999)]. 

During our work, we noted that misuse of the purchase card was not
always subject to strong disciplinary action or consequences, even 
though Air Force Instruction 64-117 requires installation purchase card 
program coordinators to take appropriate action to document violations 
and preclude their reoccurrence. The Instruction also requires approving
officials to document cardholder violations and forward the information 
to the program coordinator. The Instruction states that action(s) taken
should be commensurate with the violation(s) and gives examples of
actions, such as suspending the cardholder or approving official 
account, requiring remedial training, requiring restitution for any 
unauthorized purchases, and permanently revoking purchase card 
privileges. Holding individuals responsible for proper program 
execution is an integral part of a strong control environment. 

Our review of the results of annual purchase card surveillances 
performed all four case study locations determined that the 
surveillances found violations of Federal Acquisition Regulation 
guidelines on micropurchases and requirements to use mandated sources 
of supply. Recommended disciplinary actions for violations of these 
federal regulations were limited to requiring the offending approving 
officials and cardholders to take remedial training. For repeat 
offenders, the surveillance reports generally recommended that the 
cardholder’s and/or approving official’s purchase card account(s) be 
suspended until remedial training was completed. In only a few cases, 
had surveillance reports recommended canceling the accounts and 
revoking cardholder privileges, even for repeat offenders. As 
previously discussed, Edwards AFB surveillances did not include 
findings on these issues and, therefore, Edwards AFB did not take 
disciplinary action for failure to follow federal guidelines on 
micropurchases and using mandated sources of supply. 

Further, although three Edwards AFB surveillance reports identified
improper use of the purchase card to buy food for employees, pay for 
party supplies, and purchase invitations for a change of command 
ceremony, the reports did not recommend disciplinary action, and we saw 
no evidence that the cardholder or the benefiting individuals were 
required to pay for the unauthorized purchases. According to the 
Edwards AFB Contracting Director, cardholders have been counseled and 
in some cases referrals were made to unit commanders to take 
appropriate disciplinary action for improper use of the purchase card. 
However, Edwards AFB officials provided no documentation that any 
disciplinary actions were taken. 

The Air Force Audit Agency reported that inaction by contracting 
officials reduced the effectiveness of other purchase controls, 
increased transaction errors, and undermined the Air Force purchase 
card goals. Air Force auditors concluded that without enforcement of 
purchase card requirements, such as administratively disciplining the 
offenders or revoking purchase card privileges, purchase card 
discrepancies will likely continue. We agree with the Air Force Audit 
Agency’s conclusions. 

In response to our DOD purchase card audits, the Congress has recently
addressed the question of discipline for those who misuse the government
purchase card. Section 8149(c) of the Department of Defense 
Appropriations Act, 2003 (fiscal year 2003 appropriations act), 
[Footnote 27] and section 1007(a) of the Bob Stump National Defense 
Authorization Act for Fiscal Year 2003 (fiscal year 2003 authorization 
act) [Footnote 28] provide that the Secretary of Defense establish 
guidelines and procedures for disciplinary actions to be taken against 
department personnel for improper, fraudulent, or abusive use of 
government purchase cards. 

Inadequate Infrastructure for Program Monitoring and Oversight: 

[Management should ensure that skill needs are continually assessed and 
that the organization is able to obtain a workforce that has the 
required skills that match those necessary to achieve organizational 
goals….As a part of its human capital planning, management should also 
consider how best to retain valuable employees, plan for their eventual 
succession, and ensure continuity of needed skills and abilities. GAO’s
Standards for Internal Control in the Federal Government (GAO/AIMD-00-
21.3.1, November 1999)]. 

Ineffective oversight of the purchase card program also contributed to
weaknesses in the overall environment. Installation program coordinators
are established as the pivotal officials in managing and overseeing the
purchase card program. The coordinators at the installations we audited 
had little training on what they should be doing to oversee the program 
and limited time to carry out oversight activities that are called for 
in Air Force Instruction 64-117, such as reviewing U.S. Bank exception 
reports and following up on suspicious transaction activity as well as 
conducting annual surveillances. Effective oversight activities also 
would include other management reviews and evaluations to assess risks 
associated with the number of credit card accounts, credit limits, and 
approving official span of control and to assess the effectiveness of 
controls, identify systemic weaknesses, and determine the extent of 
potentially fraudulent, improper, and abusive or questionable 
purchases. 

We also found that the program coordinators did not have the grade 
level or organizational authority—“clout”—to routinely deal with unit
commanders, approving officials, and resource managers across the
installation, who may significantly outrank them, to enforce purchase 
card guidelines and controls. Program coordinators have the primary
responsibility for purchase card program management and significant
control over procurement activities carried out by a large number of
individuals. For example, during fiscal year 2001, the Nellis AFB 
program coordinator who was a GS-9, similar in grade to a master 
sergeant, had responsibility for over 55,000 purchase card transactions 
totaling over $27 million involving 638 cardholder accounts and 99 
approving officials. 

Installation contracting officials told us that program coordinator 
staffing had not kept pace with the growth in the purchase card 
program. GSA data show that Air Force purchase card activity has grown 
from 2.8 million transactions totaling $1.3 billion at the beginning of 
fiscal year 2000 to 3.2 million transactions totaling about $1.4 
billion at the end of fiscal year 2001. During fiscal year 2002, 
Lackland AFB increased the number of staff assigned to the program 
coordinator’s office from three staff assigned in fiscal year 2001 to a 
total of six staff to provide better oversight of Wilford Hall’s 
purchase card program as well as to monitor purchase card use by units 
transferred to Lackland AFB when Kelly AFB closed. The other three 
installations we audited had not increased their program coordinator
staffing. As of September 2002, Edwards AFB had three staff assigned,
Nellis AFB had two staff, and Travis AFB had one and a half staff to 
oversee their purchase card programs. Further, at Travis AFB, where 
military employees have held the program coordinator position, there 
have been four different individuals assigned to the program 
coordinator position since October 2000 due to high turnover associated 
with military positions. As a result, Travis AFB program coordinators 
spend much of their time learning versus overseeing the purchase card 
process. Table 4 shows the grade level, staffing, and span of control 
data for installation purchase card program coordinators at our four 
case study locations as of August 2002. 

Table 4: Installation Program Coordinator Span of Control as of 
September 30, 2002: 

Air Force installation: Edwards AFB; 
Grade level of installation program coordinator: GS-12; 
Staff in program coordinator’s office: 3; 
Number of approving officials[A]: 122; 
Number of cardholder accounts[A]: 530; 
Number of transactions: 39,600; 
Value of transactions (in millions): $21.7. 

Air Force installation: Lackland AFB' 
Grade level of installation program coordinator: GS-11; 
Staff in program coordinator’s office: 6; 
Number of approving officials[A]: 113; 
Number of cardholder accounts[A]: 662; 
Number of transactions: 93,446; 
Value of transactions (in millions): $43.9. 

Air Force installation: Nellis AFB; 
Grade level of installation program coordinator: GS-9; 
Staff in program coordinator’s office: 2; 
Number of approving officials[A]: 99; 
Number of cardholder accounts[A]: 638; 
Number of transactions: 52,591; 
Value of transactions (in millions): $30.6. 

Air Force installation: Travis AFB; 
Grade level of installation program coordinator: 1st lieutenant; 
Staff in program coordinator’s office: 1.5; 
Number of approving officials[A]: 131; 
Number of cardholder accounts[A]: 703; 
Number of transactions: 56,682; 
Value of transactions (in millions): $30.3. 

Source: GAO analysis of U.S. Bank data. 

[A] Data presented are as of the end of August 2002. 

[End of table] 

In September 2002, the Travis AFB Contracting Director told us that he 
had requested approval to hire a full-time GS-9 staff member to assist 
the program coordinator. In early November, the Deputy Contracting 
Director told us that they had received approval hire a GS-11 civilian 
employee as the installation purchase card program coordinator as well 
as the GS-9 assistant, beginning in January 2003. 

Given the risks associated with ineffective purchase card program
management identified in our purchase card work, it is imperative that
agencies have sufficient numbers of qualified, experienced purchase card
program coordinator staff to help oversee their purchase card programs
and that their grade levels be commensurate with their 
responsibilities. 

Tests of Key Control Activities: 

[Internal control activities help ensure that management’s directives 
are carried out. The control activities should be effective and 
efficient in accomplishing the agency’s control objectives. GAO’s 
Standards for Internal Control in the Federal Government (GAO/AIMD-
00-21.3.1, November 1999)]. 

Our tests of statistical samples of transactions at four Air Force
installations found weaknesses in key purchase card control activities. 
For example, of the five key control activities we tested, we found 
that all four Air Force locations had significant control breakdowns in 
at least three of them. However, on a positive note, our tests of 
easily pilferable property items included in our statistical sample 
transactions showed that two of the four Air Force installations we 
audited were able to account for all the property items that we 
selected for testing. 

Control activities occur at all levels and functions of an agency. They
include a wide range of diverse activities such as approvals, 
authorizations, verifications, reconciliations, performance reviews, 
and the production of records and documentation. For the Air Force 
purchase card program, we tested those control activities that we 
considered to be key in creating a system to provide reasonable 
assurance that transactions are correct and proper throughout the 
procurement process. The key control activities and techniques we 
tested include (1) advance approval of purchases, (2) independent 
receiving and acceptance of goods and services, (3) cardholder 
reconciliation of monthly statements, (4) independent review by an 
approving official of the cardholder’s reconciled statements and 
supporting documentation within the Air Force-prescribed time frame, 
and (5) cardholders obtaining receipts and maintaining invoices that
support their purchases and provide the basis for reconciling cardholder
statements. Table 5 summarizes the results of our statistical testing.
Appendix I includes the specific criteria that we used to conclude on 
the effectiveness of these controls. 

Table 5: Internal Control Activity Statistical Testing Results: 

Air Force installation: Edwards; 
Percentage of breakdowns in key purchase card controls[A], Advance 
authorization: 6; 
Percentage of breakdowns in key purchase card controls[A], Independent
receiving: 68; 
Percentage of breakdowns in key purchase card controls[A], Cardholder
reconciliations: 22; 
Percentage of breakdowns in key purchase card controls[A], Approving
official review: 70; 
Percentage of breakdowns in key purchase card controls[A], Supporting
invoice/receipt: 9. 

Air Force installation: Lackland; 
Percentage of breakdowns in key purchase card controls[A], Advance 
authorization: 12; 
Percentage of breakdowns in key purchase card controls[A], Independent
receiving: 61; 
Percentage of breakdowns in key purchase card controls[A], Cardholder
reconciliations: 26; 
Percentage of breakdowns in key purchase card controls[A], Approving
official review: 87; 
Percentage of breakdowns in key purchase card controls[A], Supporting
invoice/receipt: 30. 

Air Force installation: Nellis; 
Percentage of breakdowns in key purchase card controls[A], Advance 
authorization: 4; 
Percentage of breakdowns in key purchase card controls[A], Independent
receiving: 53; 
Percentage of breakdowns in key purchase card controls[A], Cardholder
reconciliations: 37; 
Percentage of breakdowns in key purchase card controls[A], Approving
official review: 69; 
Percentage of breakdowns in key purchase card controls[A], Supporting
invoice/receipt: 0. 

Air Force installation: Travis; 
Percentage of breakdowns in key purchase card controls[A], Advance 
authorization: 2; 
Percentage of breakdowns in key purchase card controls[A], Independent
receiving: 56; 
Percentage of breakdowns in key purchase card controls[A], Cardholder
reconciliations: 21; 
Percentage of breakdowns in key purchase card controls[A], Approving
official review: 73; 
Percentage of breakdowns in key purchase card controls[A], Supporting
invoice/receipt: 8. 

[A] The numbers represent point estimates for the population based on 
our sampling tests rounded to the nearest percentage point. The 
confidence intervals for our sample estimates are presented in appendix 
I of this report. 

Source: GAO analysis. 

[End of table] 

The results in table 5 include transactions for which supporting 
documentation was destroyed due to improper records retention guidance
in the Air Force Instruction. 

Advance Authorization of Purchases: 

Our test work showed that, for the most part, the four case study 
locations we tested had documentation of required advance authorization 
with estimated failure rates for required advance authorization of 
purchases ranging from 2 percent to 12 percent. The segregation of 
duties between officials who authorize a purchase and the cardholder 
who makes the purchase helps reduce the risk of fraud, waste, and abuse 
in the purchase card program. The Air Force purchase card program 
Instruction 64-117 requires cardholders to obtain authorization from 
the specified controlling/servicing organization on base for certain 
purchases, including purchases of computer and communication equipment, 
video equipment, medical items, and hazardous materials, before making 
the purchase. The Air Force installations we audited documented advance 
authorizations on forms established specifically for that purpose. 

Independent Receiving and Acceptance: 

The requirement for documentation of independent receiving and 
acceptance by someone other than the cardholder is not specifically
addressed in DOD policy or Air Force purchase card program Instruction
64-117. We believe that independent documentation of receipt of items
purchased by a cardholder is a basic internal control activity that 
provides additional assurance to the government that purchased items 
are not acquired for personal use and that they come into the 
possession of the government. Based on our statistical testing, we 
estimated that the failure rate for independent documentation of 
receipt and acceptance—receiving of goods and services by someone other 
than the cardholder—ranged from 53 percent to 68 percent at the four 
Air Force locations we tested. The types of items in our sampled 
transactions that lacked independent evidence of receipt and acceptance 
included computer software and memory cards, a fax machine, a cassette 
recorder, camera film, hardware, supplies, and tools. These items were 
purchased at stores such as Homebase, Staples, and Radio Shack. Because 
the Air Force purchases items for valid, government purposes from 
stores that are widely used by consumers to acquire items for personal 
use, verification of receipt of goods and services by an individual 
other than the cardholder is necessary to reduce the risk of fraudulent 
transactions. 

Cardholder Reconciliation and Approving Official Review: 

Cardholder reconciliation is a key control activity for detecting 
invalid transactions, including billing errors and unauthorized 
purchases. However, based on our statistical testing, we estimated that 
cardholders at the four case study locations lacked documented evidence 
of timely reconciliations of monthly purchase card statements from 21 
percent to 37 percent of the time. As evidence that purchase card 
statements were reconciled, we accepted check marks, notes, sequential 
numbering, and numbering systems that tied transactions on the 
statement to items on the cardholders’ purchase card logs. Independent 
approving official review of monthly, reconciled cardholder statements 
is a key control activity for segregation of duties, whereby no one 
individual has control over all aspects of a transaction. Based on our 
statistical testing, we estimated that approving officials at the four 
case study locations lacked documented evidence that they had reviewed 
monthly, reconciled purchase card statements, or had reviewed them 
within required time frames, from 69 percent to 87 percent of the time. 
Our statistical tests of approving official review considered only 
documented review of statements with evidence of reconciliation. The 
high failure rates are due, in part, to approving officials’ failure to 
date the reconciled monthly statements when they reviewed them. The 
failure rate also may be attributable to approving official duties 
falling into the category of “other duties as assigned” and the span of
control issues discussed earlier. 

The high failure rate for approving official review is of particular 
concern because the Air Force uses a “pay and confirm” policy, which is
inconsistent with governmentwide and DOD guidelines on reconciliation
and payment of purchase card bills. In a letter dated April 30, 2002, 
DOD informed us that its reengineering memorandums and other
pronouncements are in compliance with 10 U.S.C. 2784, which requires the
Secretary of Defense to issue regulations that require, among other 
things, reconciliation of purchase card statements to receipts before 
the statements are forwarded to the disbursing office. Both section 
4535 of volume 1 of the Treasury Financial Manual and DOD’s Purchase 
Card Reengineering Implementation Memorandum #3 (change 1, June 30, 
1998) require that purchase card statements be reconciled and forwarded 
for payment in a timely manner and allow “pay and confirm” only with 
respect to verification of government receipt of the purchased items or 
services. 

In contrast, Air Force purchase card policy permits cardholder 
statements to be reconciled and approved after payment has been made. 
While a conscientious postpayment reconciliation and approval process 
may provide reasonable control, the lack of documented evidence of 
postpayment reconciliation and approval and the undisputed, potentially
fraudulent transactions identified in our work underscore concerns about
noncompliance with the law. In July 2002, Air Force management asked its
contractor, U.S. Bank, to “shut down” (suspend from use [Footnote 29]) 
over 4,000 unreconciled, unapproved cardholder accounts until the 
reconciliations were completed and the approving officials had reviewed 
them. Accounts that had not been reconciled as of the end of August 
2002 were canceled. According to an Air Force headquarters official, 
these accounts would need to be manually reconciled because they are no 
longer active in U.S. Bank’s system. 

Under Air Force “pay and confirm” procedures, the installation Financial
Services Office designates a certifying officer to verify availability 
of funding and certify the monthly installation purchase card invoices 
for payment prior to receipt of the confirmation of reconciled 
statements from the approving official. Monthly invoices are to be paid 
in full and are not to be adjusted for disputed items. Instead, 
cardholders and approving officials are to resolve any irregularities 
through a separate dispute process. The Air Force Instruction requires 
that approving officials review and approve reconciled cardholder 
statements and submit the confirmed statements to the Financial 
Services Office within 15 days of receipt of the monthly statement, but 
no later than the 15th day of the following month—commonly referred to 
as the 15-day rule. The Financial Services Office files the confirmed 
statements with a copy of the previously certified statement. 

The pay and confirm process has yielded benefits, such as increased 
rebate earnings, and has almost eliminated late payment interest. 
However, without effective controls over cardholder reconciliation and 
approving official review, the pay and confirm process increases the 
risk that fraudulent, improper, and wasteful purchase card expenditures 
could occur and go undetected. DOD and Air Force officials told us that 
they were concerned about the lack of compliance with requirements for
purchase card statement reconciliation and approval. 

DOD and U.S. Bank officials told us that this control was implemented
upon receipt of the April 25, 2002, monthly purchase card statements. 
The officials told us that on July 1, 2002, approximately 4,000 
cardholder accounts were suspended and no further charges could be made 
to these accounts until they were reconciled, and reviewed and approved 
by the approving officials. The officials also told us that 
reconciliation of these accounts resulted in a high volume of disputed 
transactions as cardholders began to reconcile their statements and 
approving officials had to review and approve them, indicating that 
fraudulent or erroneous transactions may have occurred and had not been 
previously detected. 

On August 22, 2002, DOD’s Purchase Card Joint Program Management Office 
Director told us that 149 of the purchase card accounts that were 
suspended on July 1, 2002, had not been reviewed and confirmed by the
approving officials and, as a result, the accounts had not been 
reactivated. The DOD Director said that he planned to cancel these 
accounts because they apparently are not needed. However, the failure 
to reconcile these accounts raises questions about whether cardholders 
and/or approving officials may have made fraudulent or improper 
transactions for which they want to avoid scrutiny. Without 
reconciliation and independent review, DOD and the Air Force have no 
assurance that such purchase card activity did not involve fraudulent 
or improper transactions. 

Supporting Invoice or Receipt: 

As shown in table 5, three of the Air Force case study locations we
audited—Edwards, Nellis, and Travis Air Force bases—maintained the vast
majority of the receipts for items purchased with the government 
purchase card. For example, our statistical test results showed the 
estimated failure rates for this control activity ranged from 0 to 
about 9 percent across these three locations. Our statistical test 
results for Lackland AFB showed that this location had an estimated 30 
percent failure rate for this control activity. Of the 37 Lackland AFB 
transactions that were missing receipts, 23 related to Wilford Hall 
Medical Center transactions. Another three of the transactions in our 
Lackland AFB sample related to a security forces unit that transferred 
to another Air Force installation and did not retain purchase card 
documentation. In testing for evidence of a receipt, we accepted either 
the original or a copy of the invoice, sales slip, or other store 
receipt. 

GAO’s Internal Control Standards state, “all transactions and other
significant events need to be clearly documented, and the documentation
should be readily available for examination. All documentation and
records should be properly managed and maintained.” Without supporting
sales receipts or invoices, it is not possible to tell the quantity and 
type of items purchased or whether those items were for government 
business or of a personal nature. In such cases, a thorough 
investigation would be needed to determine whether a transaction was 
proper, or if it represented a potentially fraudulent, improper, or 
abusive transaction needing corrective action. Further, without a 
receipt, two other key control activities—independent receipt and 
acceptance and approving official review—become ineffective. 
Independent receiving cannot confirm that the purchased items were 
received and the approving official cannot review a cardholder 
statement reconciled with the supporting receipt. A near zero failure 
rate is a reasonable goal considering that receipts are easily obtained 
or replaced when inadvertently lost. 

Controls Over Accountable Property: 

As shown in table 6, three of the four installations we audited recorded
most of the items we selected for testing in their accountable property
records. In addition, the Air Force was able to locate and we confirmed
that all of the items that were not recorded at two installations were 
in the possession of the government. However, Air Force officials were 
unable to locate 4 of the 114 accountable property items we tested at 
Edwards AFB and 14 of the 70 accountable items we tested at Travis AFB, 
indicating that these items may have been lost or stolen. The property 
book officer at Travis AFB told us that cardholders do not always 
notify the property office and provide documentation of accountable 
items purchased with the government credit card—even though many of 
them are easily pilferable and desirable items. As previously 
discussed, Travis AFB’s failure to record in the installation’s 
property records, accountable property purchased using a government 
purchase card was also an Air Force Audit Agency repeat audit finding. 

Table 6: Property Items Not Recorded in Property Records: 

Air Force installation: Edwards; 
Property items selected for testing: 114; 
Items not on property books: 12; 
Items that could not be located: 4. 

Air Force installation: Lackland; 
Property items selected for testing: 35; 
Items not on property books: 1; 
Items that could not be located: 0. 

Air Force installation: Nellis; 
Property items selected for testing: 68; 
Items not on property books: 10; 
Items that could not be located: 0. 

Air Force installation: Travis; 
Property items selected for testing: 70; 
Items not on property books: 49; 
Items that could not be located: 14. 

Source: GAO nonrepresentative selection of property items included in 
statistical samples of Air Force purchase card transactions. 

[End of table] 

Items such as a digital camera, a laser printer, and computers and 
monitors were not included in base property records. The Edwards AFB 
property items that could not be located included two computer servers 
and two monitors costing a total of $11,258 that were ordered for other 
installations. The Travis AFB property items that could not be located 
included a digital camera costing $812 that, according to investigative 
records, was previously reported stolen from an employee’s office and 
eight computers costing $14,128 that were allegedly sent to the Defense 
Reutilization Marketing Service (DRMS) as excess items during the year 
they were purchased. Because serial numbers for the computers had not 
been recorded, we could not confirm that the eight computers we 
selected for testing were items that were sent to DRMS. A Travis AFB 
contracting official told us that these new computers should not have 
been sent to DRMS as excess property. The official told us that 
unneeded computer equipment is required to be turned in to the 
information technology unit for assignment to other installation units. 
Other missing items included a laptop computer and four computer 
monitors costing under $500. 

GAO’s internal control standards and DOD Instruction 5000.64, Defense
Property Accountability, require that accountable property be recorded 
in property records as it is acquired. The DOD Instruction refers to
accountable property as “controlled inventory items” and defines these
items as those designated as having characteristics that require them 
to be identified, accounted for, secured, segregated, or handled in a 
special manner to ensure their safekeeping and integrity. The 
Instruction defines pilferable items as those that have a ready resale 
value or application to personal possession and that are, therefore, 
especially subject to theft. However, the DOD Instruction does not 
include a list of items that fall into these categories. Accountable 
property generally includes high-cost property items and easily 
pilferable or sensitive items, such as computers and related equipment, 
cameras, cell phones, and power tools. 

Air Force Instruction 33-112, Computer Systems Management, requires
mandatory inclusion of computer items costing $500 or more in inventory
records. However, the Air Force does not have a policy for recording 
other types of easily pilferable or sensitive items in its property 
records. According to an Air Force headquarters acquisition official, 
decisions on how to control items, such as computers and related 
equipment, cameras, cell phones, and power tools, are left to the 
discretion of the installation commanders. As a result, there is no 
assurance that Air Force installations are following DOD policy. At the 
four installations we audited, we found that most computers were 
recorded in either central or unit-level property systems. However, 
cardholders at the installations we audited were not always aware that 
items such as digital cameras, fax machines, or computer items costing 
less than $500 meet the definition of controlled inventory items and/or 
pilferable items and thus should be recorded in their property records. 

One factor that may explain the positive Air Force test results at 
three of the four installations we audited is that these installations 
made greater use of centralized purchasing and receiving for computer 
equipment. As a result, contracting and information technology units 
controlled purchasing and receiving for these items and assured that 
the items were recorded in the property records when they were 
received. In contrast, at Travis AFB, nearly half of the property items 
tested were not recorded in property records and management could not 
locate many of these items. The use of central purchasing and receiving 
helps to mitigate against control breakdowns where cardholders do not 
take action to ensure that accountable items are recorded in property 
records. 

Potentially Fraudulent, Improper, and Abusive or Questionable 
Transactions: 

We identified numerous purchase card transactions at the four 
installations we audited and in our Air Force-wide data mining that 
were potentially fraudulent, improper, and abusive or questionable. 
Buying items with purchase cards without the requisite control 
environment and key control activities in place creates unnecessary 
risk of fraud and abusive and wasteful spending. Also, the lack of 
records previously discussed raises concerns about whether files could 
have been destroyed so that potentially fraudulent, improper, or 
abusive transactions were not documented and subjected to review. In 
addition, we saw a number of potentially fraudulent transactions at the 
case study locations we audited and in our Air Force-wide analysis that 
related to compromised accounts. We did not review all potentially 
fraudulent, improper, and abusive transactions identified in our work. 
As discussed in appendix I, our work was not designed to identify, and 
we cannot determine, the extent of potentially fraudulent, improper, 
and abusive or otherwise questionable transactions. 

Potentially Fraudulent Purchase Card Transactions: 

We identified transactions that Air Force officials acknowledged to be
fraudulent, as well as potentially fraudulent transactions for which no
supporting documentation was available, at all four installations we
audited, as well as in our Air Force-wide analysis. Some transactions
identified as potentially fraudulent resulted from compromised accounts 
in which a purchase card or account number was stolen and used by 
someone other than the cardholder to make unauthorized purchases. We 
found five potentially fraudulent transactions involving two purchase 
card accounts that were not disputed with the bank. Further, none of 
these transactions were referred to Air Force investigators until we 
questioned them. We considered potentially fraudulent purchases to 
include those made by cardholders that were unauthorized and intended 
for personal use. We also considered transactions for which there was 
no supporting documentation to be potentially fraudulent because, in 
the absence of supporting documentation, it is not possible to 
determine whether these transactions represented valid government 
purchases, fraudulent transactions that went undetected, or fraudulent 
transactions for which documentation was intentionally destroyed to 
cover up the fraud. In these instances, cardholders and approving 
officials were unable to tell us the types of items purchased or the 
purpose of the transactions. However, we determined that they had not 
disputed any of these transactions. Potentially fraudulent transactions 
can also involve vendors charging purchase cards for items that 
cardholders did not buy. Although collusion can circumvent what 
otherwise might be effective internal control activities, a robust 
system of guidance, internal control activities, and oversight can 
create a control environment that provides reasonable assurance of 
preventing or quickly detecting fraud, including collusion. 

Air Force and U.S. Bank officials told us that in July 2001, U.S. Bank
identified numerous fraudulent transactions due to compromised 
accounts. According to U.S. Bank officials, this was a widespread fraud
involving many credit card banks that was apparently related to a fraud
ring that used a computer to randomly generate credit card account 
numbers and/or counterfeit credit cards, which they then used. U.S. Bank
officials told us that the compromised accounts initially were believed 
to be associated with transactions in a few states, including 
California and Georgia. However, the fraud was subsequently determined 
to be a nationwide problem. According to Air Force and U.S. Bank 
officials, numerous Air Force purchase card accounts were canceled due 
to this fraud. Given the risk associated with such fraud, it is 
extremely important that cardholders reconcile their monthly statements 
in order to detect and dispute potentially fraudulent transactions. 
Table 7 illustrates the types of potentially fraudulent transactions 
that we identified. 

Table 7: Potentially Fraudulent Air Force Purchase Card Transactions: 

Air Force locations: Andrews; 
Vendor: E-Z Pawn; 
Total amount: $2,443. 

Air Force locations: Edwards; 
Vendor: BEF Corporation; 
Total amount: $19,000. 

Air Force locations: Travis; 
Vendor: Flowers Sent Today, Inc., Flowers Anytime
Total amount: $432. 

Air Force locations: Scott; 
Vendor: Ocean Drive Fashions, Inc., Rendezvous on the Beach, Donald 
Pliner Concept Store, Sunglass Hut, Watch World; 
Total amount: $2,401. 

Air Force locations: Patrick; 
Vendor: Citgo 7-Eleven, Publix, Chevron, Kash N Karry, Speed SM; 
Total amount: $249. 

Air Force locations: McConnell; 
Vendor: Ross Stores, Old Navy, K-Mart, Target; 
Total amount: $3,232. 

Air Force locations: Luke; 
Vendor: Nationwide Gourmet of AZ; 
Total amount: $100. 

Air Force locations: Lackland; 
Vendor: Oakley, Cabela’s, Rebel Hobbies, LACAF Golf Course, Franklin 
Covey, Handspring, Inc., That Fish Place, and Gargoyles; 
Total amount: $8,707. 

Air Force locations: Nellis; 
Vendor: Eb’s Stuff; 
Total amount: $185. 

Source: GAO analysis of selected Air Force fiscal year 2001 purchase 
card transactions. 

[End of table] 

Our Air Force-wide data mining and analysis of purchase card
documentation identified potentially fraudulent purchase card 
transactions at Andrews and McConnell Air Force bases that were never 
disputed with the bank or credited to the government. Our analysis of 
documentation related to the potentially fraudulent transactions that 
were not disputed disclosed the following facts. 

* The E-Z Pawn transaction was for a $2,443 down payment on a $10,000
sapphire ring. An October 4, 2001, agency program coordinator purchase 
card surveillance report indicated that the reviewer had questioned the 
lack of receipt for the E-Z Pawn transaction; however, the surveillance 
did not question the propriety of this transaction. No further action 
was taken, even though pawnshops are coded to a merchant category that 
is required to be blocked as a means of preventing fraudulent 
transactions from being processed. In response to our inquiry, the 
program coordinator told us that there is no evidence that either the 
cardholder or the approving official disputed the transaction. Also, we 
found no credit for this transaction in the Air Force purchase card 
database from U.S. Bank. According to the cardholder, at the time the 
potentially fraudulent transaction occurred, he did not have possession 
of the purchase card. The cardholder told us that his purchase card 
account was being closed due to outsourcing of his unit’s function, and 
he had turned over his purchase card to another individual. The 
cardholder stated that, as a result, he never received his monthly 
statement and thus did not perform a reconciliation or identify the 
potentially fraudulent transaction. We determined that the program 
coordinator took no further action to ensure that the potentially 
fraudulent transaction was disputed. 

The Air Force headquarters acquisition official who obtained the 
supporting documentation on these transactions for our review and
analysis told us that she referred these transactions to the Air Force
Office of Special Investigations and our investigators confirmed that
Air Force investigators had opened a case to investigate these frauds. 

* The four potentially fraudulent McConnell AFB transactions totaling
$3,232 were made on July 31, 2001. These potentially fraudulent
transactions included charges made at San Diego area stores, including
$690 at a Ross Store, $873 at Old Navy, $689 at K-Mart, and $980 at
Target. These charges were all made when the purchase card was in the
possession of the approving official while the cardholder was assigned
to Biloxi, Mississippi, for over 2 months for noncommissioned officer
training. Although the approving official stated that his review of the
cardholder’s August 2001 statement detected the potentially fraudulent
transactions, he did not dispute these transactions. The program
coordinator told us that disputing erroneous charges is covered
repeatedly in cardholder and approving official training. The program
coordinator also told us that she counseled the approving official
extensively on cardholder and approving official responsibilities. The
program coordinator told us that she informed the approving official
that she had reduced the credit limit on the account to $1 to avoid
further charges. She also said that she instructed the approving 
official that he should dispute these transactions in the absence of the
cardholder’s dispute. However, instead of disputing these transactions, 
the approving official waited for the cardholder to return from training
in mid-September and asked the cardholder to seek guidance from the
program coordinator. A circular discussion ensued about who would take 
action to dispute the potentially fraudulent transactions, with the 
approving official stating that the cardholder was to dispute these
transactions and the cardholder stating that he thought the program
coordinator would dispute the transactions. As a result, the 
transactions were never disputed and the program coordinator took no
further action. 

The Air Force headquarters acquisition official who obtained the
supporting documentation on these transactions for our review and
analysis told us that she referred these transactions to the Air Force
Office of Special Investigation. However, Air Force investigators told
us they did not initiate an investigation because they believed that the
transactions had been credited. Our review of the Air Force purchase
card database and discussions with the installation program coordinator 
determined that the transaction had not been credited and the potential 
fraud is unresolved. Therefore, we have referred this matter to our 
investigators for further investigation. 

In contrast to the failure to dispute the potentially fraudulent 
transactions discussed above, the following are examples of potentially 
fraudulent transactions that Air Force cardholders detected and 
disputed. 

* A $19,000 charge for a duplicator from BEF Corporation—an imaging
technology vendor in Allentown, Pennsylvania—related to a compromised 
purchase card account. Our investigators confirmed that the Edwards AFB 
cardholder called the bank and disputed the transaction before the item 
was delivered. As a result, neither the government nor the bank 
incurred a loss related to this compromised account. Our investigators 
confirmed that the cardholder had referred the transaction to Air Force 
investigators. 

* Three unauthorized Travis AFB purchases, including one transaction at
Flowers Sent Today, Inc., and two transactions at Flowers Anytime, 
totaling $432 were made by an inmate at a local county jail. Travis AFB
officials told us that they disputed these transactions because they 
were unauthorized and they believed that the account had been 
compromised. These potentially fraudulent transactions were 
subsequently investigated by U.S. Bank and the purchase card account
was canceled. 

* Five potentially fraudulent charges were made to a Scott AFB, 
Illinois, cardholder’s account on August 21 and August 22, 2001, 
totaling about $2,400. The charges were made at vendors in the Miami 
area, including charges at Ocean Drive Fashions, Inc., for $426, 
Rendezvous on the Beach for $224, Donald Pliner Concept Store for $520, 
Sunglass Hut for $586, and Watch World for $645. Our review of Air 
Force records showed that the cardholder disputed these transactions 
and a U.S. Bank investigation of the potentially fraudulent 
transactions was initiated on November 9, 2001. Our review of U.S. Bank 
records showed that the purchase card account was credited for the 
fraudulent transactions on June 6, 2002—nearly a year after the 
transactions were made. 

* A Patrick AFB cardholder identified unauthorized purchase card 
transactions totaling $249 that were made at Citgo 7-Eleven, Publix,
Chevron, Kash N Karry, and Speed SM by his wife between April 1 and 6,
2001. According to the unit commander, the cardholder’s purchase card
privileges were revoked, and the cardholder agreed to pay for his wife’s
unauthorized charges. However, as of the end of October 2002, we
determined that the cardholder had not reimbursed the government for
his wife’s unauthorized use of the purchase card. We suggested that the
employee should submit a check to the installation’s Financial Services
Office along with an explanation for the payment. 

We also contacted the Air Force Office of Special Investigations to 
inquire about their purchase card fraud cases. Investigators told us 
that their investigative database did not contain codes that permit 
them to identify purchase card fraud cases. As a result, the 
investigators had to manually review all procurement-related cases to 
attempt to identify cases involving purchase card fraud. Appendix III 
summarizes some of the purchase card fraud cases that were investigated 
by the Air Force Office of Special Investigations. 

Improper Purchases and Improper Use of the Purchase Card: 

Besides potentially fraudulent activity, our work also identified 
numerous examples of transactions related to improper purchases, as 
well as improper use of the purchase card. Improper purchases are those
purchases that, although approved by Air Force officials and justified 
as intended for government use, are not permitted by law or regulation 
or DOD or Air Force policy. Improper use of the purchase card related 
to use of the card as an acquisition tool without a negotiated 
contract, use of the card by a nongovernment activity, and improper use 
of convenience checks [Footnote 30] associated with purchase card 
accounts. 

Improper Purchases: 

We identified the following three types of improper purchases. 

* Purchases that did not serve a legitimate government purpose. 

* Split purchases in which the cardholder circumvents the micropurchase
limit [Footnote 31] or other transaction limits. 

* Purchases from improper sources. Various federal laws and regulations
require procurement officials to acquire certain products from 
designated sources, such as Javits-Wagner-O’Day Act (JWOD) vendors. 
[Footnote 32] In addition, agencies are required to purchase furniture, 
if available, from Federal Prison Industries, Inc. (UNICOR), and DOD 
policy requires that printing services be obtained in-house through the 
Defense Automated Printing Service. 

We found several instances of purchases, such as clothing, luggage, and
food, in which cardholders improperly used their purchase cards, or they
purchased goods that were not authorized by law or regulation. The 
Federal Acquisition Regulation, 48 C.F.R. 13.301(a), provides that the 
governmentwide commercial purchase card may be used only for purchases 
that are otherwise authorized by law or regulation. We identified the 
improper Air Force transactions as part of our analysis of questionable 
vendor transactions at the four installations we audited and in our Air 
Force-wide data mining. Table 8 summarizes examples of the improper 
transactions we identified that do not serve a legitimate government 
purpose. 

Table 8: Improper Air Force Purchase Card Transactions: 

Types of items purchased: Clothing: Physical fitness clothing and fleece
jackets for drill instructors; 
Vendors: LL Bean, Oakley; 
Amount: $1,696. 

Types of items purchased: Clothing: Blazer and dresses for participants 
in Eubank Service Award Ceremony; 
Vendors: Filene’s Sports Coats, Dress Barn; 
Amount: $828. 

Types of items purchased: Clothing: Wool coats for a flight attendant; 
Vendors: Hecht’s; 
Amount: $380. 

Types of items purchased: Clothing: Clothes for parachutists, pilots, 
and others; 
Vendors: REI; 
Amount: $20,698. 

Types of items purchased: Luggage: Samsonite pullman suitcases; 
Vendors: El Portal Luggage; 
Amount: $5,500. 

Types of items purchased: Luggage: Garment bags; 
Vendors: SM Discount Luggage, 1-800 Luggage; 
Amount: $2,291. 

Types of items purchased: Luggage: Pathfinder pullmans for recruiters; 
Vendors: Foley’s; 
Amount: $620. 

Types of items purchased: Luggage: Duffel bags; 
Vendors: Patagonia; 
Amount: $685. 

Types of items purchased: Food/Water: Meals provided during planning 
meeting and unit luncheons; 
Vendors: Marriott Hotel, Rusty Pelican, Where Pigs Fly; 
Amount: $2,692. 

Types of items purchased: Food/Water: Bottled water; 
Vendors: Wishing Well Florist; 
Amount: $796. 

Types of items purchased: Other: Personal exercise equipment; 
Vendors: Target; 
Amount: $100. 

Types of items purchased: Other: Sunglasses; 
Vendors: Oakley; 
Amount: $540. 

Types of items purchased: Other: Briefcases and flight bags; 
Vendors: Franklin Covey; 
Amount: $14,664. 

Source: GAO analysis of Air Force fiscal year 2001 transactions and 
related documentation. 

[End of table] 

The following examples illustrate the types of purchases included in 
table 8. 

Clothing and sunglasses. We identified numerous purchases of clothing 
for military personnel that appeared to be personal preference items. We
determined that these purchases were improper based on our review of
DOD directives, [Footnote 33] Air Force policies, [Footnote 34] and 
discussions with Air Force headquarters officials. For example, at 
Lackland AFB, one of our test locations, we identified purchases of 
clothing for drill instructors, including physical fitness clothing 
items from LL Bean costing $816 and 16 fleece jackets from Oakley 
costing $880. We also identified numerous purchases of military 
clothing items from REI totaling $20,698, including paratrooper 
jumpsuits and cold weather pilot jackets. These clothing items are 
covered under the military pay clothing allowance funded in the 
Military Personnel, Air Force, appropriation and should not have been 
purchased with Operation and Maintenance appropriations using the 
purchase card. 

In addition, we found purchases of what we consider to be personal
clothing and accessory items that were authorized by installation 
officials. For example, Hanscom AFB, Massachusetts, authorized the 
purchase of a blue blazer from Filene’s costing $180 and eight dresses 
from Dress Barn costing $648 for civilian employees who participated in 
the Eubank Service Award Competition. [Footnote 35] Hickam AFB, in 
Hawaii, purchased two wool coats from Hecht’s in Waldorf, Maryland, 
costing $380 for a flight attendant assigned to the 65th Airlift 
Squadron. In addition, the Lackland AFB pararescue team purchased 12 
pairs of sunglasses from Oakley costing $540 and improperly justified 
them as meeting the requirement for free-fall paratrooper goggles. 
Oakley sunglasses do not qualify as paratrooper goggles. None of these 
items are authorized by Air Force policy. Therefore, we concluded that 
all of these purchases involved personal items, which employees should 
pay for from their own salaries. 

Luggage and briefcases. We identified numerous purchases of luggage
deemed necessary for employees who travel frequently, including 50
Samsonite suitcases costing $5,500 for members of the Thunderbirds team 
and garment bags costing $2,250 for the Bolling AFB band, and purchases
of two Pathfinder Pullman suitcases costing $620 for recruiters. In 
addition, our review of a limited selection of Franklin Covey 
transactions identified six purchases of leather briefcases costing 
$212 each (after a 20 percent discount) and purchases of 203 flight 
bags (similar to a briefcase) costing from about $50 to $86 each. 
Luggage and briefcases are considered personal items that should be 
paid for from employees’ salaries. 

Food and water for employees. We identified purchases of meals, bottled 
water, and payment for a unit luncheon. Without statutory authority,
appropriated funds may not be used to furnish meals to employees within
their normal duty stations. [Footnote 36] We identified improper 
charges for food provided to employees during an internal government 
meeting and unit luncheons. For example, Lackland AFB scheduled a 
strategic planning meeting for local employees of the Tri-Care 
organization at a Marriott Hotel in San Antonio, Texas. The total cost 
of the meeting was $1,052, including $538 for breakfast and lunch for 
18 individuals—3 consultants and 15 employees who were not on travel. 
In addition, appropriated funds may not be used to purchase bottled 
water for employees unless they are assigned to a duty station without 
potable water. Food and bottled water are considered personal items 
that employees should pay for from their own salaries. 

Split Purchases: 

Another category of improper transactions is a split purchase, which
occurs when a cardholder splits a transaction into more than one segment
to circumvent the requirement to obtain competitive prices for purchases
over the $2,500 micropurchase threshold or to avoid the other 
established credit limits. The Federal Acquisition Regulation and Air 
Force Instruction 64-117 prohibit these practices. Once items exceed 
the $2,500 threshold, they are to be purchased in accordance with 
simplified acquisition procedures, which are more stringent than those 
for micropurchases. Our analysis of purchases made at the four case 
study locations and our Air Force-wide data mining identified numerous 
split purchases. In addition, Air Force Audit Agency auditors 
identified split purchases as a continuing Air Force-wide problem. 

One split purchase we identified involved a violation of appropriations 
law. The purchase was made to avoid the expiration of unused fiscal 
year 2001 operation and maintenance appropriations. Our inquiries 
disclosed the following events surrounding this transaction. On 
September 30, 2001, at approximately 9:00 p.m. Eastern Standard Time, 
Air Combat Command headquarters at Langley AFB, Virginia, determined 
that $100,000 in fiscal year 2001 operation and maintenance 
appropriations would expire in 3 hours—at midnight—unless the funds 
could be obligated and spent quickly. At 6:00 p.m. Pacific Standard 
Time, the command contacted the USAF Weapons School, 57th Operations 
Support Squadron at Nellis AFB, Las Vegas, Nevada, with the direction 
to use the funds before they expired. At approximately 7:00 p.m., the 
Nellis AFB cardholder purchased 120 helmets at more than $800 each and 
several other items totaling just under $100,000 from the base supply 
store. The purchase was split into four separate transactions to stay 
within the cardholder’s single transaction limit of $25,000 per 
transaction. The items purchased were placed “on hold” and were not 
taken from the store. Over the next few days—October 1 and October 
2—the cardholder “returned” the items, exchanging them for other items 
totaling the same amounts. In an explanatory memorandum, the cardholder 
wrote that the unit had not previously identified its unfilled 
requirements and, therefore, did not have a list of items for purchase 
if “end-of-year” money was available. The subsequent credits and reuse 
of the funds in early October 2001, in effect, converted fiscal year 
2001 appropriations to fiscal year 2002 budget authority. 

Improper Sources: 

Another type of improper purchase occurs when cardholders do not buy
from a mandatory procurement source. Various federal laws and 
regulations, such as the Javits-Wagner-O’Day Act (JWOD), require
government cardholders to acquire certain products from designated
sources. The JWOD program is a mandatory source of supply for all
federal entities. It generates jobs and training for Americans who are 
blind or have other severe disabilities by requiring federal agencies 
to purchase supplies and services furnished by nonprofit agencies, such 
as the National Industries for the Blind and the National Institute for 
the Severely Handicapped. Most JWOD program supplies are small-value 
items such as office supplies, cleaning products, or medical/surgical 
supplies that nearly always fall into the micropurchase category. We 
noted that most cardholders at the four installations we audited made 
purchases from required sources and three of the four installations had 
JWOD stores on base. However, we found numerous Air Force-wide 
purchases from Franklin Covey for office supplies, such as calendars 
and day planners, which could have been purchased from JWOD vendors. We 
also found that Air Force cardholders charged $2,220 for 82 high-
quality pens from Franklin Covey costing from $16 to $60 each. Table 9 
summarizes transactions we identified that were made from other than 
required sources of supply. 

Table 9: Purchases from Other Than Required Sources of Supply: 

Source: GAO Analysis of Air Force fiscal year 2001 transactions and 
related documentation. 

Type of item purchased: Office supplies; 
Vendor: Franklin Covey; 
Total amount: $188,834. 

Type of item purchased: Printing; 
Vendor: Kinkos, Mayes Printing; 
Total amount: $336,680. 

[End of table] 

The failure to purchase designated items from JWOD vendors who support
the handicapped undermines public policy objectives to support programs
for the handicapped. For example, as discussed in our Army purchase card
report, [Footnote 37] the Director of Sales for the National Industries 
for the Blind told us that this program has experienced large decreases 
in sales over the past 2 years because cardholders were purchasing from 
commercial firms rather than buying the mandated products. Further, 
operating revenues of government service organizations, such as the 
Defense Automated Printing Service (DAPS), which is a required source 
of printing services for DOD agencies, and Federal Prison Industries, 
Inc. (UNICOR), which is a required source for furniture, are 
significantly reduced to the extent that cardholders do not use these 
sources for mandated printing and related services. 

Improper Use of Purchase Cards and Convenience Checks: 

In addition to the improper transactions discussed above, our analysis 
of Air Force-wide purchase card transactions identified the following 
three types of improper use of the purchase card. 

* Use of the purchase card as acquisition tool where a negotiated 
contract is required; 

* Purchase card use by a religious fund activity without authority in 
law, regulation, or DOD or Air Force policy, and; 

* Improper use of convenience checks billed to the purchase card
account. 

Table 10: Improper Uses of the Purchase Card and Convenience Checks: 

Improper use: Multiyear tractor rentals; 
Vendors: Crown Ford; 
Amount: $52,500. 

Improper use: Expenses of Chaplain’s Office volunteer; 
Vendors: Cheap Tickets, United Air Lines, Marriott Hotels, J.C. Penney, 
K-Mart, Wal-Mart, Sav-On Drugs, JoAnn Fabric and Crafts, Hobby Lobby; 
Amount: $6,609. 

Improper use: Convenience checks; 
Vendors: Payments over $2,500 and reimbursement of tuition expense to 
employees; 
Amount: $212,898. 

Source: GAO analysis of Air Force fiscal year 2001 purchase card 
transactions. 

[End of table] 

The following examples illustrate the types of purchases included in 
table 10. 

Month-to-Month Equipment Rental. We found that a cardholder at Whiteman 
AFB, Missouri, improperly used the purchase card as an acquisition 
vehicle without a negotiated contract. Specifically, the cardholder 
used the purchase card to rent tractors for use by the installation’s 
waste treatment facility at a cost of about $10,000 during fiscal year 
2001. Our analysis of records related to this lease showed that an 
initial month-to-month rental of a tractor covered 3 years beginning in
November 1997. In November 2000, the cardholder initiated monthly rental
of a new tractor for 2 years. According to the installation program
coordinator, the current cost to purchase the type of tractor that was
rented during fiscal years 2000 and 2001 would be in the $35,000 to 
$45,000 range; however, a lease versus purchase cost-benefit analysis 
was never performed. We determined that the cost of the two rentals, 
which together covered a 5-year period, totaled over $50,000. 

Unauthorized Use of the Card by Chaplain’s Religious Fund. Our analysis 
of fiscal year 2001 FE Warren AFB, Wyoming, purchase card transactions 
totaling over $6,600 by a Chaplain Office volunteer, who later became a
contractor, raised a number of questions about the proper use of the 
purchase card. Air Force purchase card policies state that commercial
purchase cards are provided to military members and federal civilian
employees to pay for official government purchases, and that only 
employees may be cardholders. According to the Chaplain Office 
officials, these transactions were for authorized purchases. They said 
they often use parishioners and volunteers to assist them in carrying 
out religious activities. However, Air Force purchase card policies 
state that commercial purchase cards are provided to military members 
and federal civilian employees to pay for official government 
purchases. Chaplain Office officials told us that because the Chaplain 
Religious Fund did not fall under Air Force purchase card authority as 
either an appropriated fund or a nonappropriated fund activity, they 
believed they could set up the government purchase card program for 
their office by working independently with the bank. 

We discussed our concerns about Chaplain Office authority to use the
government purchase card with Air Force attorneys and acquisition and
Chaplain Office officials. After reviewing Air Force policy, Air Force
attorneys advised us that the Chaplain’s Office did not have authority 
to use the government purchase card for Chaplain Religious Fund 
activities. 

Convenience Checks. We identified improper use of convenience checks 
related to payments in amounts over $2,500, payments for recurring 
services, and payments to vendors who accept purchase card payments.
Air Force Instruction 64-117 limits the use of convenience checks to
amounts of no more than $2,500 per check, prohibits the use of 
convenience checks for recurring services, and restricts convenience
check use to instances where vendors do not accept purchase cards.
Splitting amounts across more than one check to keep below the $2,500
limit also is prohibited. Another area of improper use of convenience
checks related to reimbursement of employees for tuition assistance. We
determined that reimbursement to employees is not a permitted use of
convenience checks. Further, because there is a 1.7 percent fee for 
using a convenience check, cost-benefit considerations are required 
when using convenience checks. 

Our analysis of fiscal year 2001 convenience check use determined that 
Air Force purchase cardholders who had convenience check authority had
issued 45 convenience checks totaling over $200,000 for amounts over
$2,500. We also found that a cardholder at Luke AFB, Arizona, improperly
used convenience checks for recurring monthly payments on a 2-year
automobile lease for authorized use by a military officer. It is common
knowledge that car dealerships accept credit cards. At one of our case
study locations—Travis AFB—we found that a cardholder had reimbursed
employees for $12,214 in tuition expenses. The cardholder wrote two
convenience checks—one check for $2,090 and another for $500—to
reimburse an employee for a total of $2,590 in tuition expenses. The
cardholder wrote two additional convenience checks—one check for
$6,125 and another for $3,500—to reimburse two other employees for
$9,614 in tuition expenses. The 1.7 percent fee on the first two checks 
was $44 and the fee on the second two checks was about 
$163—significantly more that the Defense Finance and Accounting Service 
fee of approximately $7 to process electronic payments. 

In its August 2002 purchase card report, the Air Force Audit Agency 
stated that its review found that cardholders issued convenience checks 
to pay salaries and wages totaling $512,378 for dieticians, nurses, and
administrative personnel, and to acquire recurring services, such as 
aircraft washing totaling $84,830, local area network support costing 
$110,495, equipment rentals costing $20,700, and lawn-care services 
costing $43,505. Air Force auditors determined that cardholders 
expended $2.6 million for recurring services and incurred unnecessary 
bank fees of $15,228 associated with the 1.7 percent service fee. 

Abusive or Questionable Use of the Purchase Card: 

We also identified abusive and questionable transactions at 
installations we audited and in our Air Force-wide data mining. We 
defined abusive transactions as those that were authorized, but the 
items purchased were at an excessive cost (e.g., “gold plated”) or for 
a questionable government need, or both. Abuse occurs when the conduct 
of a government organization, program, activity, or function falls 
short of societal expectations of prudent behavior. Often, improper 
purchases, such as those discussed in the previous section, are also 
abusive. For example, the purchases of personal clothing and luggage 
for employees were also abusive purchases because they were for a 
questionable government need. 

Questionable transactions are those that appear to be improper or 
abusive but for which there is insufficient documentation to conclude 
either. We deemed questionable those purchases for which there was not 
a reasonable and/or documented justification. Questionable purchases 
often do not easily fit within generic governmentwide guidelines on 
purchases that are acceptable for the purchase card program. They tend 
to raise questions about their reasonableness. Many, such as gym-
quality exercise equipment for fitness centers, are common Air 
Force—and DOD—purchases because the Air Force must provide more than 
merely a work environment for its soldiers. However, others involving 
excessive purchases of alcohol, payment for taxidermy services, and 
purchases of expensive leather computer cases discussed in this 
section, raise questions about whether they are appropriate purchases. 
Precisely because these types of purchases tend to raise questions and 
subject the Air Force to criticism, they require a higher level of 
advance purchase review and documentation than other purchases. 

When we examined these types of purchases, we usually did not find 
evidence of advance purchase justification. In attempting to justify
whether purchases were acceptable, improper, or abusive, program 
coordinators, approving officials, and cardholders often provided after-
the-fact rationales for the purchases. We believe that these types of
questionable purchases require scrutiny before the purchase, not after. 
The examples in table 11 illustrate our point. 

Table 11: Abusive or Questionable Air Force Purchase Card Transactions: 

Location: Nellis AFB; 
Type of item purchased: Entertainment - dinner party and
show for visiting general, including excessive purchases of alcohol
totaling $800; 
Vendor: Treasure Island Hotel and Casino; 
Total amount: $2,141. 

Location: Tyndall AFB; 
Type of item purchased: 2 reclining rocking chairs with vibrator 
massage feature; 
Vendor: LA-Z-Boy; 
Total amount: $1,935. 

Location: Air Force Academy, Colorado Springs; 
Type of item purchased: Mounted mule deer head; 
Vendor: Timberline Taxidermy; 
Total amount: $375. 

Location: Wright-Patterson AFB; 
Type of item purchased: Leather laptop case for Brigadier General; 
Vendor: Bentley’s Luggage; 
Total amount: $595. 

Location: Edwards AFB; 
Type of item purchased: Pictures for legal office; 
Vendor: SkyMall Airline Catalog; 
Total amount: $1,459. 

Location: Edwards AFB; 
Type of item purchased: Leather backpack for Dell laptop computer; 
Vendor: The Complement; 
Total amount: $224. 

Location: Edwards, Kelly, Travis and Peterson AFBs; 
Type of item purchased: Civilian clothes for military assistants; 
Vendor: Old Navy, Eddie Bauer, The Men’s Warehouse, Macy’s; 
Total amount: $2,016. 

Location: Elmendorf, Lackland, Peterson, and Travis AFBs; 
Type of item purchased: Costumes for Air Force regional band members; 
Vendor: Nordstrom’s, Macy’s, Purple Pansy, Gentlemen’s Choice, Foley’s, 
Men’s Warehouse, Dorothy Keck Dance; 
Total amount: $2,948. 

Source: GAO Analysis of Air Force fiscal year 2001 transactions and 
related documentation. 

[End of table] 

The following examples illustrate the problems associated with some of 
the transactions we identified as abusive or questionable. 

* At Nellis AFB, the purchase card was used to pay for dinner and a 
show for 18 people costing $2,141 at Treasure Island—a Las Vegas hotel 
and casino. The purpose of the event was to entertain the General of 
U.S. Joint Forces Command, who was visiting the base. Air Force 
Instruction 65-603, Official Representation Funds – Guidance and 
Procedures, permits the use of government funds for official 
entertainment. However, we determined that the nature of this 
entertainment did not meet certain Force guidelines related to 
requirements to conduct entertainment on a modest basis that is in the
interest of the taxpayer. For example, we determined that the cost of
the dinner party, which totaled $2,141, included about $800 for alcohol
for the 18 people who attended the event—over $40 per person. We
believe the excessive cost of alcohol purchased at this event falls 
short of societal expectations of prudent behavior and modest cost. 

* An Air Force Academy Natural Resource office in Colorado Springs,
Colorado, used the purchase card to pay Timberline Taxidermy $375 to
prepare a shoulder mount of a mule deer head. According to the
approving official, the deer was “road kill” that he found on the 
roadside and brought to the Natural Resources Office. The approving 
official then approved the purchase of the taxidermy service to prepare 
a stuffed shoulder mount of the deer. The deer head was hung on the wall
in the Natural Resources Office. The justification of the purchase of
taxidermy services provided to our auditors stated, "The mule deer
(Odocoileus hemionus) is the most common large mammal present on
US Air Force Academy grounds. The mount was created as an 
educational/interpretive tool and is on display in the USAFA Natural
Resources office. The mount can be removed from the office wall for
use in educational presentations to the base population (e.g., that only
males have antlers which are shed and re-grown each year.) The deer
died after being struck by a vehicle and was salvaged by USAFA Natural
Resources personnel." When our auditor asked the cardholder how often 
the deer head was removed from the office wall and used for educational 
purposes, the cardholder stated, "not much." The cardholder, the 
approving official, and two other Natural Resource employees occupy the 
office where the deer head currently hangs. 

* The Edwards AFB, 412th Test Wing/Electronic Warfare used fiscal 2001
year-end funds to purchase 21 computers and monitors at a total cost of
$47,372. An e-mail message dated September 13, 2001, within the 412th
Test Wing had subject lines stating “Last Minute Purchasing” and 
included one message line that stated, “… I gather they have quite a bit
of unspent credit card money and want to move out on it this week.”
The 21 computers and monitors were received on October 23, 2001. At
the time of our inspection of these items in June 2002, 11 of the 
computers and 9 of the monitors were located in a storage area and 
many of the items were still in the original shipping boxes. According 
to the Contracting Director, the computers were purchased to support
hiring of engineers and engineering contractors. However, recruiting
efforts were delayed because some applicants did not accept offered 
positions and further recruiting efforts were suspended during 
implementation of a new personnel system in the fall of 2001. The 
Contracting Director told us that the unit commander directed that the
computers be stored in their original boxes for safekeeping until 
recruiting was completed and the equipment could be assigned to 
engineering staff, which was accomplished in July or August 2002. The
fact that the computers were still in their original boxes at least 9
months after they were ordered raises questions about whether there
was a legitimate need for these items to be purchased with fiscal year
2001 appropriations. Appropriated funds are available only to meet
legitimate needs of the agency during the fiscal year for which the 
funds were appropriated. 

* Our testing of property items identified two Travis AFB transactions
dated October 4, 2000, for purchases of computer equipment totaling 
$14,128 that involved wasteful spending. For example, when we attempted 
to observe the items to confirm their existence, we were told that the 
unit had decided to convert to Dell computers. As a result, within 1 
year of their purchase, these items, as well as a number of other 
computers, were sent to the Defense Reutilization Marketing Service as
excess property. 

In addition, we questioned purchases of civilian clothing for military
assistants and costumes for regional band members. While DOD and the
Air Force have issued policy [Footnote 38] that permits the purchase of 
these types of clothing and designates them as “uniforms,” we believe 
this clothing represents personal preference attire and should be paid 
for by the employees. For example, in addition to standard issue 
uniform clothing items, military assistants are permitted to purchase 
slacks or skirts, shirts, and blazers to wear while serving as aides to 
general officers. In addition, we noted purchases including two 
tuxedos, six dresses, and earrings as costumes for members of regional 
Air Force bands. Under Air Force policy, regional bands are permitted 
to purchase tuxedos and evening gowns to be worn as costumes during 
performances. While the civilian clothing for military assistants and 
band costumes are considered government property, which may be reused 
as appropriate, they are not likely to be reissued to others. 
Therefore, we question whether taxpayer funds should be used to pay for 
these items. 

Management Improvements: 

DOD and Air Force managers told us that they initiated a number of 
actions during fiscal year 2002 to improve purchase card controls. 
These initiatives include use of automated U.S. Bank controls to (1) 
tie cardholder credit limits to allocations of budget authority, (2) 
automatically deactivate purchase card accounts where monthly 
statements have not been reconciled and reviewed by the approving 
officials within prescribed time frames, and (3) cancel purchase card 
accounts for approving officials with responsibility for excessive 
numbers of cardholders. According to Air Force officials, they plan to 
periodically lower the threshold for suspending purchase card accounts 
until the approving official’s span of control complies with DOD span 
of control guidelines. 

In addition, the DOD Comptroller appointed a Charge Card Task Force,
which issued its final report on June 27, 2002. The Task Force report
included a number of recommendations, including establishing a purchase
card concept of operations, accelerating the electronic certification 
and bill paying process, improving training materials, identifying best 
practices in areas such as span of control and purchase card management 
skill sets, and establishing more effective means of disciplining those 
who abuse the purchase cards. The recommendations address many of the 
concerns we identified in our Air Force work. 

In response to our DOD audits, the Congress has recently enacted 
amendments in section 1007(a) of DOD’s fiscal year 2003 authorization 
act that address requirements for (1) periodic reviews to be performed 
to determine whether each purchase card holder has a need for the 
purchase card, (2) periodic inspector general audits to identify 
potentially fraudulent, improper, and abusive uses of purchase cards, 
(3) appropriate training for cardholders and oversight officials, and 
(4) specific policies regarding the number of purchase cards issued by 
various organizations, authorized credit limits, and categories of 
employees eligible to be issued purchase cards. Further, DOD’s fiscal 
year 2003 appropriation and authorization acts each include 
requirements that the regulations issued by the Secretary of Defense 
provide for appropriate disciplinary actions or other punishment to be 
imposed in cases in which DOD employees violate purchase card 
regulations or are negligent or engage in misuse, abuse, or fraud with 
respect to a purchase card, including removal in appropriate cases. 

Conclusions: 

A well-controlled purchase card program is a valuable tool for 
streamlining the government’s acquisition processes. However, the 
problems we identified with missing receipts, lack of cardholder 
reconciliations and approving official review, and failure to follow 
requirements in laws, regulations, and DOD and Air Force policies and 
procedures resulted in control environment weaknesses that leave the 
Air Force vulnerable to fraud and improper use of the purchase card, as 
well as abuse and wasteful spending. Also, although Air Force 
management has been proactive in establishing improved controls, it has 
not ensured that installation-level program coordinators—the primary 
program management officials—have the tools to develop local control 
systems and adequate oversight activities. Further, installation 
contract officials have not consistently demonstrated the commitment to 
enforce established controls. Strengthening the control environment 
will require a renewed focus and attention and commitment to building a 
robust purchase card infrastructure. 

Recommendations for Executive Action: 

To strengthen the overall control environment and improve internal 
control over the Air Force purchase card program, we recommend that the
following actions be taken. 

Overall Program Management and Environment: 

We recommend that the Secretary of the Air Force direct the Assistant
Secretary of the Air Force for Acquisition and the Deputy Assistant
Secretary for Contracting to take the following actions. 

* Establish specific policies and strategies governing the number of
purchase cards to be issued with a focus on minimizing the number of
cardholders. 

* Direct all command and installation-level agency program coordinators
to review purchase card use with a view toward eliminating unneeded
purchase card accounts. 

* Eliminate purchase cards used to facilitate line item accounting. 

* Direct all agency program coordinators to review the number of 
cardholders who report to an approving official and make the changes
necessary so that approving officials do not have responsibility for
reviewing more cardholder accounts than allowed by Air Force and DOD 
policies. 

* Review existing credit limits and monthly spending and develop 
policies and strategies on credit limits provided to cardholders with a 
focus on minimizing specific cardholder spending authority and 
minimizing the federal government’s financial exposure. 

* Deactivate purchase card accounts of alternate cardholders and
approving officials when primary cardholders and approving officials
are available. 

* Establish specific training courses for cardholders, approving 
officials, and agency program coordinators tailored to the specific
responsibilities associated with each of those roles. 

* Require installation program coordinators to track and monitor 
corrective actions on purchase card audit and annual surveillance
findings and provide periodic status reports to their installation
contracting directors. 

* Develop and implement a program oversight system for program
coordinators that includes standard activities and analytical tools to 
be used in evaluating program results. 

* Require reports on annual surveillance results to include an 
assessment of control environment issues, including the ratio of 
cardholders to employees, ratio of approving officials to cardholder 
accounts, ratio of monthly credit limits to actual spending, and number 
of cardholders and approving officials requiring training. 

* Assess the adequacy of human capital resources devoted to the 
purchase card program, especially for oversight activities, at each
management level, and provide needed resources where appropriate. 

We also recommend that the Secretary of the Air Force direct the 
Assistant Secretary of the Air Force for Acquisition and the Deputy 
Assistant Secretary for Contracting to make to following revisions to 
Air Force Instruction 64-117, Air Force Government-wide Purchase Card 
Program. 

* Correct faulty records retention guidance by referring to specific
guidelines in the Federal Acquisition Regulation, National Archives and
Records Administration federal records retention guidelines, DOD’
Financial Management Regulation, and other federal guidelines as
appropriate. 

* Require purchase card program management and administrative records 
generated by installation program coordinators and approving officials, 
such as records of cardholder and approving official appointments and 
training, cardholder delegations of authority, and purchase card 
surveillances, to be retained for 3 years. 

* Stipulate, in the body of the Instruction, that approving officials 
are required to have annual purchase card refresher training. 

* Require that the surveillance checklist, which is included in an 
appendix to the Air Force Instruction, be used to guide and document 
surveillance results. 

* Require reports on the results of annual surveillances to be signed by
installation contracting directors to demonstrate management oversight
and “tone at the top.” 

* Require reports on surveillance results to be addressed to unit
commanders. 

* Require reports on surveillance results to include recommendations for
unit commander action, where approving officials and cardholders have 
failed to follow Air Force policy—particularly policy related to federal
regulations, such as micropurchase requirements and mandated sources of 
supply. 

Key Control Activities: 

To resolve noncompliance with requirements in law for proper 
certification of purchase card payments, we recommend that the 
Secretary of the Air Force take the following actions. 

* Direct the Assistant Secretary of the Air Force for Acquisition and 
the Deputy Assistant Secretary for Contracting to work with the Under
Secretary of Defense (Comptroller) to resolve inconsistencies between
DOD and Air Force policies and procedures for reconciling purchase
card statements prior to payment. 

* Develop a strategy for achieving Air Force compliance with 
requirements in the law that DOD purchase card policies and procedures 
require reconciliation of purchase card statements prior to payment. 

We recommend that the Secretary of the Air Force direct the Assistant
Secretary of the Air Force for Acquisition and the Deputy Assistant
Secretary for Contracting to revise Air Force Instruction 64-117 to 
provide cardholders, approving officials, and installation program 
coordinators with detailed instructions on the following specific 
control activities. 

* Establish appropriate criteria, including types of items and dollar
thresholds for documenting independent receiving and acceptance of
items obtained with a purchase card. 

* Establish specific procedures for documenting independent receiving,
such as requiring the approving official or supervisor to sign and date
the vendor invoice, sales receipt, or credit card receipt, or requiring 
the approving official to sign the cardholder’s monthly purchase log to
verify that items noted as having been received were actually received.
• Require cardholders to maintain documentation of timely and 
independent receiving and acceptance of items obtained with a purchase 
card. 

* Require reconciliation of monthly purchase card statements associated
with accounts that were “shut down” (suspended) in July 2002 due to 
lack of cardholder reconciliation and approving official review. 

* Verify that all potentially fraudulent and erroneous transactions that
have been detected are disputed and properly resolved. 

* Require timely cardholder notification to the property accountability
officer of pilferable property, such as fax machines, digital cameras, 
and palm pilots obtained with the purchase card. 

* Encourage installation contracting officers to consider the benefits 
of central purchasing and receiving and acceptance of computer 
equipment by installation information technology units to facilitate
recording computer equipment in accountable property records at the
time it is received. 

We also recommend that the Deputy Assistant Secretary for Contracting
revise Air Force Instruction 64-117 to define and list examples of 
sensitive and pilferable property purchased with a government purchase 
card, including cell phones, digital cameras, fax machines, palm 
pilots, and copiers and printers, and require prompt recording of these 
items in installation property systems. 

In addition, we recommend that the Assistant Secretary of the Air Force 
for Logistics establish policies and procedures for recording all 
pilferable and sensitive property, including digital cameras, palm 
pilots, and cell phones, in installation accountable property records. 
At a minimum, require installations to follow DOD policies and 
procedures on accountable property. 

Potentially Fraudulent, Improper, and Abusive and Questionable Purchase 
Card Activity: 

We recommend that the Assistant Secretary of the Air Force for Financial
Management (Comptroller) direct the Air Force Audit Agency and Air 
Force Office of Special Investigations to establish an Air Force-wide
database of known fraud cases by type of fraud, including purchase card
fraud, that can be used to identify systemic weaknesses and 
deficiencies in existing internal control and to develop and implement 
additional control activities, if warranted or justified. 

We recommend that the Assistant Secretary of the Air Force for 
Acquisition and the Deputy Assistant Secretary for Contracting take the 
following actions. 

* Establish an Air Force-wide database of known purchase card fraud 
cases by type of fraud, including vendor fraud and compromised 
accounts, that can be used to identify deficiencies in existing internal
control and implement additional control activities, if warranted. 

* Identify vendors with which the Air Force used purchase cards to make
frequent, recurring purchases, evaluate Air Force purchasing practices
with those vendors, and where appropriate, develop contracts with
those vendors to optimize Air Force purchasing power. 

* Review organizational use of the purchase card and revoke purchase
cards issued to organizations that do not have authority to participate 
in the governmentwide purchase card program. 

* Cancel convenience check privileges of cardholders who have continued 
to improperly use convenience checks. 

* Require accounting adjustments to be made to correct transactions that
were charged to the wrong appropriation account with respect to fiscal
year and purpose of the expenditures. 

* Establish appropriate, consistent Air Force-wide policy as a guide for
taking disciplinary actions with respect to cardholders and approving
officials who make or approve fraudulent, improper, or abusive purchase 
card transactions. 

* Require cardholders and/or approving officials to reimburse the
government for any unauthorized or erroneous purchase card transactions 
that were not disputed. 

* Require benefiting individuals to reimburse the government for the 
cost of any personal items that they requested or directed a cardholder 
to purchase for them. 

We also recommend that the Under Secretary of Defense (Comptroller) 
direct the Charge Card Task Force to assess the above recommendations,
as well as the strengths in the Air Force purchase card program that we
identified, and to the extent applicable, incorporate them into its 
future recommendations to improve purchase card policies and procedures
throughout DOD. 

Agency Comments and Our Evaluation: 

On December 13, 2002, DOD’s Purchase Card Joint Program Management
Office and the Air Force provided oral comments on a draft of this 
report. DOD and Air Force purchase card officials concurred on 29 of 
our 39 recommendations and partially concurred with the remaining 9
recommendations. At the time we finalized our work, DOD had not
provided a response to our remaining recommendation that the Charge
Card Task Force assess the recommendations in this report and
incorporate them to the extent applicable, into its future 
recommendations to improve purchase card policies and procedures 
throughout DOD. Of the 9 recommendations involving partial 
concurrences, the DOD and Air Force officials (1) agreed in substance 
with 5 of our recommendations, (2) noted that the Air Force Office of 
Special Investigations and DOD Inspector General have responsibility 
for actions on two of our recommendations related to establishing a 
database of fraud cases by type of fraud that can be used to identify 
systemic weaknesses and deficiencies in controls, and (3) indicated 
alternative actions have been initiated on the remaining 2 
recommendations. 

With regard to agreement on the substance of our recommendations, Air
Force officials stated that they would (1) suspend alternate accounts 
when primary cardholders and billing officials are available, (2) 
revise the Air Force purchase card Instruction to require reports on 
purchase card surveillance results to be signed by contracting squadron 
commander or chief of the contracting office, (3) require 
reconciliation of monthly purchase card statements associated with 
accounts that were “shut down” (suspended) in July 2002, (4) issue a 
policy letter to encourage installation Contracting Officers to 
consider the benefits of central purchasing and receiving and 
acceptance of computer equipment by installation Information Technology 
units, and (5) revise the Air Force purchase card Instruction to define 
and list examples of sensitive and pilferable property and establish 
clear accountability and/or visibility criteria. 

With regard to two recommendations related to establishing an Air 
Forcewide database of known fraud cases by type of fraud to identify 
and correct systemic weaknesses and deficiencies in existing internal 
control, Air Force officials stated that the Air Force Office of 
Special Investigations in conjunction with the other Defense Criminal 
Investigative Organizations now reports quarterly information on 
purchase card investigations to the DOD Inspector General. The 
officials told us that the DOD Inspector General has been directed to 
develop a centralized purchase card database on known fraud cases and 
audit results that can be used to identify potential deficiencies in 
existing internal controls. They said that the Air Force will evaluate 
the Air Force cases and audits to determine the effectiveness of 
existing internal controls and implement additional control activities, 
if warranted. 

Alternative Air Force actions relate to our recommendations that the
Deputy Assistant Secretary for Contracting (1) review organizational 
use of the purchase card and revoke purchase cards issued to 
organizations that do not have authority to participate in the 
governmentwide purchase card program and (2) establish appropriate, 
consistent Air Force-wide policy as a guide for taking disciplinary 
actions with respect to cardholders and approving officials who make or 
approve fraudulent, improper, or abusive purchase card transactions. 
With regard to action on the first recommendation, Air Force officials 
stated that the Chaplain Service has authority to issue its own 
policies and procedures, including purchase card authority. However, 
they stated that the Head of the Air Force Chaplain Office will 
recommend reinstatement of the Chaplain Funds in DOD Directive 1015.1, 
Establishment, Management, and Control of Nonappropriated Fund 
Instrumentalities, which is in the process of being updated to reflect 
current DOD and Air Force policies regarding the government purchase 
card. The Air Force also agreed to review organizational use of the 
purchase card and revoke purchase cards issued to organizations that do 
not have authority to participate in the governmentwide purchase card 
program. 

With regard to action on our recommendation to establish Air Force-wide
policy as a guide for taking disciplinary actions, Air Force officials 
noted the existing guidance in the Air Force purchase card Instruction, 
which is discussed earlier in this report. They also stated that the 
Deputy Assistant Secretary of the Air Force (Contracting) has issued a 
memorandum requiring that a summary of each purchase card fraud and 
each instance of repeated misuse of the purchase card be briefed 
quarterly by the contracting squadron commander to the installation 
commander along with the disciplinary action taken. The Congress 
recently enacted provisions in DOD’s appropriation and authorization 
acts that require the Secretary of Defense to establish guidelines and 
procedures for disciplinary actions to be taken against department 
personnel for improper, fraudulent, or abusive use of government 
purchase cards. 

In addition, the Air Force provided technical comments on our draft 
report stating that it disagreed with our position that civilian 
clothing for enlisted aides and costumes for military band members 
represented abusive or questionable transactions. The Air Force 
referred to its Enlisted Aide Handbook and Air Force Instruction 36-
2123 as authority for purchasing civilian attire, which is designated 
as a “uniform” for enlisted aides. Air Force officials pointed out that 
their handbook specifically states, “operation and maintenance funds 
are used when purchasing uniform items” and “local purchase is 
authorized and encouraged.” Air Force officials also stated that band 
costumes are authorized purchases in accordance with Air Force 
Instruction 35-101 and that band costumes may be reused, as 
appropriate. The Air Force’s position appears to be that any item 
defined in its policy as a uniform or band costume can be purchased 
using a purchase card and paid for with appropriated funds. We continue 
to believe that these clothing purchases are questionable because the 
Air Force did not adequately explain the circumstances of the 
purchases, such as the purpose of clothing and the vendor. The Air 
Force’s policy opens the door for abuse and its implementation merits 
close scrutiny. 

As agreed with your offices, unless you announce the contents of this
report earlier, we will not distribute this report until 30 days from 
its date. 

At that time, we will send copies to interested congressional 
committees; the Secretary of Defense; the Under Secretary of Defense 
for Acquisition, Technology, and Logistics; the Under Secretary of 
Defense (Comptroller); the Secretary of the Air Force; the Assistant 
Secretary of the Air Force for Acquisition; the Deputy Assistant 
Secretary of the Air Force for Contracting; the Assistant Secretary of 
the Air Force for Logistics; the Director of the Defense Finance and 
Accounting Service; and the Director of the Office of Management and 
Budget. We will make copies available to others upon request. The 
report also will be available free of charge on GAO’s Web page at 
[hyperlink, http://www.gao.gov]. 

Please contact Gregory D. Kutz at (202) 512-9505 or kutzg@gao.gov, John
Ryan at (202) 512-9587 or ryanj@gao.gov, or Gayle L. Fischer at (202) 
512-9577 or fischerg@gao.gov, if you or your staff have any questions
concerning this report. Major contributors to this report are 
acknowledged in appendix IV. 

Signed by: 

Gregory D. Kutz: 
Director: 
Financial Management and Assurance: 

Signed by: 

Robert J. Cramer: 
Office of Special Investigations: 
Managing Director: 

[End of section] 

Appendix I: Objectives, Scope, and Methodology: 

We audited the effectiveness of the Air Force’s internal controls and
payment of its fiscal year 2001 purchase card transactions. The Air 
Force’s purchase card program is the smallest of the three services, 
with fewer transactions and dollars spent than the Army or the Navy. We 
selected our four case study locations by identifying major commands 
with the largest purchase card sales volume and number of transactions. 
We selected major Air Force commands that accounted for about 69 
percent of total purchase card charges and 65 percent of total 
transactions for fiscal year 2001. We then selected one installation 
within each of the four commands based on the magnitude of purchase 
card activity (sales volume and number of transactions). We also 
considered the results of prior Air Force Audit Agency work. We 
selected the following Air Force installations for our case study work. 

Table 12: Installations Audited and Associated Major Commands: 

Installation and location: Edwards AFB, California; 
Major command:Air Force Materiel Command. 

Installation and location: Lackland AFB, Texas; 
Major command:Air Education and Training Command. 

Installation and location: Nellis AFB, Nevada; 
Major command:Air Combat Command. 

Installation and location: Travis AFB, California; 
Major command:Air Mobility Command. 

[End of table] 

At the four Air Force installations, we evaluated the policies and 
procedures used to guide the purchase card program, and we evaluated the
activities they engage in to oversee the program. We used a case study
approach to evaluate the local purchase card program, and our work 
consisted of three major segments—(1) an assessment of the overall 
control environment, including the adequacy of the Air Force’s policies 
and procedures, (2) an evaluation of the effectiveness of key internal 
control activities, and (3) a determination of whether evidence existed 
of potentially fraudulent, improper, and abusive or questionable 
transactions. Finally, we assessed management actions taken in fiscal 
year 2002 to improve purchase card controls. 

To assess the overall control environment, we used as our primary 
criteria applicable laws and regulations; our Standards for Internal 
Control in the Federal Government (GAO/AIMD-00-21.3.1, November 1999); 
and our Internal Control Standards: Internal Control Management and
Evaluation Tool (GAO-01-1008G, August 2001). To assess the management
control environment, we applied the fundamental concepts and standards 
in GAO’s Internal Control Standards to the practices followed by
management. 

To test the implementation of key control activities during fiscal year 
2001 at the four installations we audited, we obtained from DOD, U.S. 
Bank’s database of Air Force purchase card transactions from October 1, 
2000, through September 30, 2001. We did not verify the accuracy of 
U.S. Bank’s database. We selected stratified random probability samples 
of 150 to 152 purchase card transactions from the population of Air 
Force transactions [Footnote 39] for each case study location. With 
these statistically valid samples, each transaction in the four 
locations’ populations had a nonzero probability of being included, and 
that probability could be computed for any transaction. Within each 
installation we stratified the population of transactions by the dollar 
value of the transaction and by whether the transaction was likely to
be for a purchase of computers and related equipment. Each sample 
transaction for an installation was subsequently weighted in the 
analysis to account statistically for all the transactions in the 
population of that installation, including those that were not 
selected. 

For each transaction sampled, we tested whether key internal control
activities had been performed. For each control activity tested, we
projected an estimate of the percent of transactions for which the 
control activity was not performed, for each installation. Because we 
followed a probability procedure based on random selections of 
transactions, our sample for each installation is only one of a large 
number of samples that we might have drawn. Since each sample could 
have produced different estimates, we express our confidence in the 
precision of our particular samples’ results (that is, the sampling 
error) as 95 percent confidence intervals. These are intervals that 
would contain the actual population value for 95 percent of the samples 
we could have drawn. As a result, we are 95 percent confident that each 
of the confidence intervals in this report will include the true 
(unknown) values in the study populations. 

Although we projected the results of our samples to the populations of
transactions at the respective case study locations, the results cannot 
be projected to the population of Air Force transactions or 
installations as a whole. Tables 13 through 20 present (1) the results 
of our tests for one or more control attributes, (2) the point 
estimates of the failure rate for the attributes, (3) the two-sided 95 
percent confidence intervals for the failure rates for each attribute, 
(4) our assessments of the effectiveness of the controls, and (5) the 
relevant lower and upper bounds of a one-sided 95 percent confidence 
interval for the failure rate. All numbers in these tables are rounded 
to the nearest percentage point. 

We use one-sided confidence bounds to classify the effectiveness of a
control activity. If the one-sided upper bound does not exceed 5 
percent, then the control activity is effective. If the one-sided lower 
bound exceeds 10 percent, then the control is ineffective. Otherwise, 
we say that the control is partially effective. Partially effective 
controls may include those for which there is not enough evidence to 
assert either effectiveness or ineffectiveness. For example, if we were 
95 percent confident that the failure rate for a particular control is 
less than 3 percent, we would categorize that control activity as 
“effective” because 3 percent is less than the 5 percent standard. 
Similarly, if we were 95 percent confident that the failure rate for a 
particular control is greater than 72 percent, we would categorize that 
control as “ineffective” because 72 percent is greater than the 10 
percent standard. 

Table 13 shows the results of our tests of controls for documenting
cardholder and approving official appointments. Local commanders 
appoint cardholders and approving officials for their units and notify 
the installation program coordinator who then schedules these 
individuals for purchase card training. 

Table 13: Estimated Results of Statistical Tests of Fiscal Year 2001 
Purchase Card Transactions for Cardholder and Approving Official 
Appointments: 

Air Force installation: Edwards; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals):82% (71% to 90%);
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Ineffective; lower bound = 73%. 

Air Force installation: Lackland; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals):0% (0% to 4%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Effective; upper bound = 3%. 

Air Force installation: Nellis; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals):0% (0% to 4%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Effective; upper bound = 3%. 

Air Force installation: Travis; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals):97% (89% to 100%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Ineffective; lower bound =90%. 

Source: GAO analysis. 

[A] The numbers represent point estimates for the population based on 
our sampling tests. 

[End of table] 

Table 14 shows the results of our tests of controls for documenting 
initial training of cardholders and approving officials. Air Force 
Instruction 64-117 requires cardholders and approving officials to 
receive purchase card training before they can be assigned a purchase 
card account. 

Table 14: Estimated Results of Statistical Tests of Fiscal Year 2001 
Purchase Card Transactions for Cardholder and Approving Official 
Initial Training: 

Air Force installation: Edwards; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 12% (5% to 23%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Partially effective; lower bound = 6%, or 
upper bound = 21%. 

Air Force installation: Lackland; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 0% (0% to 4%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Effective; upper bound = 3%. 

Air Force installation: Nellis; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 0% (0% to 4%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Effective; upper bound = 3%. 

Air Force installation: Travis; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 51% (39% to 64%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Ineffective; lower bound = 41%. 

Source: GAO analysis. 

[A] The numbers represent point estimates for the population based on 
our sampling tests. 

[End of table] 

Table 15 shows the results of our tests of controls for documenting
cardholder delegations of purchasing authority. After cardholders 
complete purchase card training, installation program coordinators are 
to prepare a letter of delegation of purchasing authority indicating the
cardholder’s transaction level spending limit and monthly credit limit. 

Table 15: Estimated Results of Statistical Tests of Fiscal Year 2001 
Purchase Card Transactions for Cardholder Delegations of Purchasing 
Authority: 

Air Force installation: Edwards; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 13% (6% to 23%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Partially effective; lower bound = 7%, or 
upper bound = 21%. 

Air Force installation: Lackland; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 0% (0% to 2%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Effective; upper bound = 2%. 

Air Force installation: Nellis; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 1% (0% to 5%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Effective; upper bound = 4%. 

Air Force installation: Travis; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 84% (73% to 93%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Ineffective; lower bound = 75%. 

Source: GAO analysis. 

[A] The numbers represent point estimates for the population based on 
our sampling tests. 

[End of table] 

Table 16 presents the results of our tests for documentation of advance
purchase authorization. Air Force Instruction 64-117 requires advance
authorization for purchases of certain items, including computer and
communication equipment, video equipment, medical items, and hazardous
materials. Estimates for this table are based only on the sample
transactions for which advance authorization of purchases was required. 

Table 16: Estimated Results of Statistical Tests of Fiscal Year 2001 
Purchase Card Transactions for Advance Purchase Authorization: 

Air Force installation: Edwards; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 6% (1% to 16%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Partially effective; lower bound = 2%, or 
upper bound = 15%. 

Air Force installation: Lackland; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 12% (2% to 32%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Partially effective; lower bound = 3%, or 
upper bound = 29%. 

Air Force installation: Nellis; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 4% (1% to 11%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Partially effective; lower bound = 1%, or 
upper bound = 10%. 

Air Force installation: Travis; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 2% (0% to 8%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Partially effective; lower bound = 0%, or 
upper bound = 7%. 

Source: GAO analysis. 

[A] The numbers represent point estimates for the population based on 
our sampling tests. 

[End of table] 

Table 17 presents the results of our tests for documentation of 
independent receiving and acceptance, by someone other than the 
cardholder, of goods and services purchased with a government purchase 
card. This requirement is not specifically addressed in DOD policy or 
Air Force purchase card program Instruction 64-117. We believe that 
independent documentation of receipt of items purchased by a cardholder 
is a basic internal control activity that provides additional assurance 
to the government that purchased items are not acquired for personal 
use and that they come into the possession of the government. 

Table 17: Estimated Results of Statistical Tests of Fiscal Year 2001 
Purchase Card Transactions for Independent Receipt and Acceptance: 

Air Force installation: Edwards; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 68% (56% to 79%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Ineffective; lower bound = 58%. 

Air Force installation: Lackland; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 61% (48% to 73%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Ineffective; lower bound = 50%. 

Air Force installation: Nellis; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 53% (41% to 65%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Ineffective; lower bound = 43%. 

Air Force installation: Travis; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 56% (44% to 68%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Ineffective; lower bound = 46%. 

Source: GAO analysis. 

[A] The numbers represent point estimates for the population based on 
our sampling tests. 

[End of table] 

Table 18 presents the results of our tests for documentation of 
cardholder reconciliations. Cardholder reconciliations are key to 
identifying potentially fraudulent transactions resulting from 
compromised accounts, duplicate or improper vendor charges, and errors. 
As evidence that purchase card statements were reconciled, we accepted 
check marks, notes, sequential numbering, and numbering systems that 
tied transactions on the statement to items on the cardholders’ 
purchase card logs. 

Table 18: Estimated Results of Statistical Tests of Fiscal Year 2001 
Purchase Card Transactions for Cardholder Reconciliations: 

Air Force installation: Edwards; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 22% (13% to 34%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Ineffective; lower bound = 14%. 

Air Force installation: Lackland; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 26% (15% to 39%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Ineffective; lower bound = 17%. 

Air Force installation: Nellis; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 37% (25% to 51%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Ineffective; lower bound = 27%. 

Air Force installation: Travis; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 21% (12% to 33%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Ineffective; lower bound = 13%. 

Source: GAO analysis. 

[A] The numbers represent point estimates for the population based on 
our sampling tests. 

[End of table] 

Table 19 presents the results of our tests of timely approving official 
review of cardholders’ monthly, reconciled statements. Approving 
official review is a recognized control activity at all levels of the 
purchase card program, and the approving official review process has 
been described as the first line of defense against misuse of the card. 
DOD’s Purchase Card Joint Program Management Office and Air Force 
Instruction 64-117 recognize that approving official review of monthly 
purchase card statements is central to ensuring that purchase card 
transactions are appropriate. The Air Force Instruction requires that 
approving officials review and approve reconciled cardholder statements 
within 15 days of receipt of the monthly statement, but no later than 
the 15th day of the following month. 

Table 19: Estimated Results of Statistical Tests of Fiscal Year 2001 
Purchase Card Transactions for Approving Official Review: 

Air Force installation: Edwards; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 70% (58% to 81%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Ineffective; lower bound = 60%. 

Air Force installation: Lackland; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 87% (78% to 94%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Ineffective; lower bound = 80%. 

Air Force installation: Nellis; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 69% (56% to 80%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Ineffective; lower bound = 58%. 

Air Force installation: Travis; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 73% (60% to 84%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Ineffective; lower bound = 62%. 

Source: GAO analysis. 

[A] The numbers represent point estimates for the population based on 
our sampling tests. 

[End of table] 

Table 20 shows the results of our tests for documentation of supporting
invoices or receipts. GAO’s Internal Control Standards state, “all
transactions and other significant events need to be clearly documented,
and the documentation should be readily available for examination. All
documentation and records should be properly managed and maintained.”
Without a receipt, independent evidence of the description and quantity 
of what was purchased and the price paid is not available. In testing 
for evidence of a receipt, we accepted either the original or a copy of 
the invoice, sales slip, or other store receipt. 

Table 20: Estimated Results of Statistical Tests of Fiscal Year 2001 
Purchase Card Transactions for Supporting Invoice or Receipt: 

Air Force installation: Edwards; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 9% (3% to 19%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Partially effective; lower bound = 4% or 
upper bound = 18%. 

Air Force installation: Lackland; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 30% (19% to 44%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Ineffective; lower bound = 21%. 

Air Force installation: Nellis; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 0% (0% to 4%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Effective; upper bound = 3%. 

Air Force installation: Travis; 
Percentage of breakdowns in key purchase card controls[A] (and 2-sided 
95% confidence intervals): 8% (3% to 19%); 
Assessment of effectiveness of controls (and relevant bounds of 1-sided 
95% confidence intervals): Partially effective; lower bound = 3% or 
upper bound = 17%. 

Source: GAO analysis. 

[A] The numbers represent point estimates for the population based on 
our sampling tests. 

[End of table] 

We also tested nonrepresentative selections of accountable property 
items that were included in our sampled transactions. Because some
transactions were for property items that were physically located off 
base, we elected to perform our test work on property items that were 
assigned to the base. We tested whether these items had been recorded 
in the installation’s accountable property records, including unit-
level records, in a timely manner and whether the installation could 
demonstrate the item’s existence. We confirmed existence of the items 
we tested through physical observation. 

In addition to our audit of statistical samples of transactions at the 
four case study installations, we also used data mining techniques to 
identify other selected transactions at the four locations and 
throughout the Air Force’s fiscal year 2001 purchase card transactions 
to determine if indications exist of potentially fraudulent, improper, 
and abusive or questionable purchase card activity. Our data mining 
included identifying transactions with certain vendors were more likely 
to sell items that would be unauthorized or that would be personal 
items. We also based our selection on the nature, dollar amount, date, 
and other identifying characteristics of the transactions. Because of 
the large number of transactions that met these criteria, we did not 
look at all potential abuses of the purchase card. For a small number 
of these transactions at each of the four installations and from the 
Air Force-wide database, we requested limited documentation, usually 
the supporting invoice, that could provide additional indications as to 
whether the transactions were potentially fraudulent, improper, or 
abusive or questionable. If the additional documentation indicated that 
the transactions were likely proper and valid, we did not pursue 
further documentation. If the additional documentation was not 
provided, or if it indicated further issues related to the 
transactions, we obtained and analyzed additional documentation or
information about these transactions. While we identified some 
potentially fraudulent, improper, and abusive or questionable 
transactions, our work was not designed to identify, and we cannot 
determine, the extent of potentially fraudulent, improper, or abusive 
transactions. 

For those potentially fraudulent transactions that had been or were 
being investigated at the four audited installations, we discussed the 
cases with the investigators and/or obtained records and reports on the 
investigations. We also interviewed purchase card officials and Air 
Force criminal investigators to identify other Air Force purchase card 
fraud cases that had been or were being investigated. 

We did not audit the Defense Finance and Accounting Service’s purchase
card payment process. We also did not audit electronic data processing
controls used in processing purchase card transactions. The 
installations received hard copy paper monthly bills containing the 
charges for their purchases and used manual processes for much of the 
period we audited, which reduced the relevance of auditing electronic 
data processing controls. 

We briefed DOD managers, including officials in DOD’s Purchase Card 
Joint Program Management Office, major command purchase card program 
coordinators, and purchase card program officials at the installations 
we audited on the details of our audit, including our objectives, 
scope, and methodology and our findings. On November 20, we requested 
comments on a draft of this report. We obtained oral comments from DOD 
and Air Force purchase card officials on December 13, 2002, and have 
summarized those comments in the “Agency Comments and Our Evaluation” 
section of this report. We conducted our audit work from January 
through mid-November 2002 in accordance with U.S. generally accepted 
government auditing standards, and we performed our investigative work 
in accordance with standards prescribed by the President’s Council on 
Integrity and Efficiency. 

[End of section] 

Appendix II: Overview of the Air Force Purchase Card Process: 

The Air Force purchase card program is part of the governmentwide 
Commercial Purchase Card Program established to streamline federal
agency acquisition processes by providing a low-cost, efficient vehicle 
for obtaining goods and services directly from vendors. Under the 
General Services Administration’s blanket contract, the Air Force has 
contracted with U.S. Bank for its purchase card services. DOD reported 
that it used purchase cards to make about 10.7 million transactions for 
goods and services at a cost of over $6.1 billion. During this same 
period, the Air Force reported that it used government purchase cards 
to make about 3 million transactions at a cost of about $1.4 billion. 
This represents about 23 percent of DOD’s activity for fiscal year 
2001. Air Force purchase card transactions were made using about 80,000 
VISA cards issued to civilian and military employees. 

Table 21: Number and Value of Air Force Fiscal Year 2001 Purchase Card
Transactions: 

Major Air Force command: Air Education and Training Command; 
Number of transactions: 410,564
Cost of transactions (in millions): $173.1; 
Percent of total Air Force cost: 12.4. 

Major Air Force command: Air Force Materiel Command; 
Number of transactions: 597,212; 
Cost of transactions (in millions): $305.2; 
Percent of total Air Force cost: 21.9. 

Major Air Force command: Air Force Reserve Command; 
Number of transactions: 89,423; 
Cost of transactions (in millions): $28.1; 
Percent of total Air Force cost: 2.0. 

Major Air Force command: Air Force Space Command; 
Number of transactions: 158,799; 
Cost of transactions (in millions): $75.0; 
Percent of total Air Force cost: 5.4. 

Major Air Force command: Air Force Special Operations Command; 
Number of transactions: 52,822
Cost of transactions (in millions): $20.8; 
Percent of total Air Force cost: 1.5. 

Major Air Force command: Air Force Services Agency[A]; 
Number of transactions: 266,916; 
Cost of transactions (in millions): $61.5; 
Percent of total Air Force cost: 4.4. 

Major Air Force command: Air Combat Command; 
Number of transactions: 607,402; 
Cost of transactions (in millions): $282.3; 
Percent of total Air Force cost: 20.2. 

Major Air Force command: Air Mobility Command; 
Number of transactions: 452,927; 
Cost of transactions (in millions): $197.8; 
Percent of total Air Force cost: 14.2. 

Major Air Force command: Pacific Air Forces; 
Number of transactions: 188,722; 
Cost of transactions (in millions): $90.1; 
Percent of total Air Force cost: 6.4. 

Major Air Force command: Other major commands; 
Number of transactions: 336,186; 
Cost of transactions (in millions): $162.7; 
Percent of total Air Force cost: 11.6. 

Major Air Force command: Total, all major commands; 
Number of transactions: 3,160,973; 
Cost of transactions (in millions): $1,396.6; 
Percent of total Air Force cost: 100.0. 

Source: GAO analysis of Air Force purchase card program data. 

[A] The Air Force Services Agency is funded by nonappropriated sources. 

[End of table] 

DOD has mandated the use of the purchase card for all purchases at or 
below $2,500 and has authorized the use of the card to pay for specified
larger purchases. For example, the purchase card may be used to purchase
authorized supplies, equipment, and nonpersonal services up to the 
$2,500 micropurchase threshold. If authorized to make purchases above 
$2,500, cardholders not in contracting organizations are to use the 
government purchase card only to obtain items from prepriced contracts 
and other pricing agreements, such as the Federal Supply Schedule, 
blanket purchase agreements, and Indefinite Delivery/Indefinite 
Quantity contracts. Purchases over the $2,500 micropurchase threshold 
and up to the simplified acquisition threshold of $25,000 must be in 
accordance with streamlined acquisition guidelines in the Federal 
Acquisition Regulation (FAR). The purchase card should normally not be 
used for cash advances; travel-related purchases; rentals or leases of 
land or buildings; utility services; or hazardous/dangerous items, such 
as explosives, munitions, toxins, and firearms. 

Government Purchase Card Program Guidelines: 

The purchase card can be used for both micropurchases and payment of
other purchases. Although most cardholders have limits of $2,500, some
have limits of $25,000 or higher. The Federal Acquisition Regulation, 
Part 13, “Simplified Acquisition Procedures,” establishes criteria for 
using purchase cards to place orders and make payments. DOD has a
supplement to this regulation that contains sections on simplified
acquisition procedures. U.S. Treasury regulations govern purchase card
payment certification processing and disbursements. DOD’s Purchase
Card Joint Program Management Office, which is in the Office of the
Assistant Secretary of the Army for Acquisition Logistics and 
Technology, has issued departmentwide guidance related to the use of 
purchase cards. However, each service has its own policies and 
procedures governing the purchase card program. 

Air Force Purchase Card Acquisition and Payment Processes: 

The Air Force purchase card program operates under federal Air Force
guidance as the policy and procedural foundation for its purchase card
program. The Air Force headquarters Acquisition Office is responsible 
for the overall management of the Air Force’s purchase card program. The
Acquisition Office has published servicewide guidelines in Air Force
Instruction 64-117, Governmentwide Purchase Card Program, dated
December 6, 2000, to establish responsibilities and procedures and 
provide administrative guidance for its government purchase card 
operations. Under the Air Force instruction, each Air Force command’s 
head contracting officer authorizes agency purchase card program 
coordinators in local Air Force units to obtain purchase cards and 
establish credit limits. The program coordinators are responsible for 
administering the purchase card program within their designated span of 
control and serve as the communication link between Air Force units and 
the purchase card-issuing bank. The other key personnel in the purchase 
card program are the approving officials and the cardholders. They are 
responsible for implementing internal controls to ensure that 
transactions are appropriate. 

Purchase Card Process: 

Figure 2 illustrates the general design of the purchase card processes 
for the Air Force. The overall process begins with the cardholder 
ordering or purchasing goods or services. Each Air Force installation’s 
Financial Services Office certifies monthly bills for payment upon 
receipt. After certification, the Financial Services Offices notify the 
Defense Finance and Accounting Service that monthly purchase card 
statements are ready for payment. The process ends with cardholder 
reconciliation and approving official review and approval of monthly 
purchase card statements after the bills have been paid. Any invalid 
transactions identified during the reconciliation and review process 
are to be disputed first with the vendor, and if not resolved, a 
“Disputed Item” form is to be submitted to U.S. Bank for credit. 

Figure 2: Air Force Purchase Card Process: 

[See PDF for image] 

This figure is an illustration of the Air Force purchase card process, 
as follows: 

* Purchase cardholder orders/charges goods and services; 
- Items picked up (Vendor). 

* Goods and services received with or without documentation of
independent receiving; 
- Pilferable items to be recorded in accountable property records. 

* U.S. Bank reviews disputed charges and credits monthly statement or 
rejects the dispute; Monthly purchase card statements are received
from bank; 
- Installation Financial Services Office establishes obligation, 
certifies consolidated monthly purchase card statement and forwards to 
DFAS for payment to U.S. Bank within 3 to 5 days; 
- DFAS processes purchase card payments to U.S. Bank. 

* Cardholders are to reconcile underlying receipts/sales slips to 
monthly purchase card statements, identify any invalid charges, and 
prepare dispute forms. 
- Cardholders are to log items not received and follow up to (1) 
confirm receipt or (2) dispute the charge. 

* Approving official reviews cardholder support and approves reconciled
cardholder statement within 15 days, or no later than 15th day of the
following month. 

* Cardholder or approving official are to log disputed charges and send
forms to U.S. Bank for credit. 

Source: GAO analysis of Air Force purchase card program organization. 

[End of figure] 

Purchase cardholders are delegated limited contracting officer-ordering
responsibilities, but they do not negotiate or manage contracts. When a
supervisor requests that a staff member receive a purchase card, the
agency program coordinator is to first provide training on purchase 
card policies and procedures and then establish a credit limit and 
issue a purchase card to the staff member. After receiving training, 
cardholders are issued a purchase card, which bears their name and the 
account number that has been assigned to them. The cardholder is 
expected to safeguard the purchase card as if it were cash. Each 
cardholder has an established daily and monthly credit limit and is 
designated to make purchases at selected types of vendors. Cardholders 
use purchase cards to order goods and services for their units as well 
as their customers. Cardholders may pick up items ordered directly form 
the vendor or request that items be shipped directly to receiving 
locations or end users. 

The approving official is responsible for providing assurance that all
purchases made by the cardholders within his or her cognizance were
appropriate and that the charges are accurate. The approving official is
supposed to resolve all questionable purchases with the cardholder. In 
the event an unauthorized purchase is detected, the approving official 
is supposed to notify the agency program coordinator and other 
appropriate personnel within the command in accordance with the command
procedures. Under governmentwide guidelines, agencies are required to
first attempt to resolve invalid transactions with the vendors. 
Transactions that are not resolved with the vendors may be disputed 
with the U.S. Bank. 

The purchase card payment process begins with receipt of the monthly
purchase card billing statements from the bank. The Air Force uses a pay
and confirm process whereby the monthly purchase card statements 
received from U.S. Bank are certified as proper for payment by each
installation’s Financial Services Office within 3 to 5 business days and
forwarded to the Defense Finance and Accounting Service (DFAS) for
payment. Under guidelines in the Air Force purchase card Instruction,
cardholders are required to review and reconcile their monthly purchase
card statements within 5 days of receipt, and approving officials are
required to review cardholders’ monthly statements as reconciled and
dispute any invalid charges within 15 days of receipt, but no later 
than the 15th day of the following month. 

DFAS effectively serves as a payment processing service and relies on 
the Air Force Financial Services Office certification of the 
consolidated monthly bill for each installation as support to make the 
payment. The DFAS vendor payment system then makes a single payment to 
U.S. Bank by electronic funds transfer for each Air Force 
installation’s monthly purchase card expenditures. 

During the summer of 2001, the Air Force began implementing U.S. Bank’s
Customer Automation and Reporting Environment (CARE) system. CARE 
provides several automated purchase card management features, including
on-line cardholder purchase logs, transaction histories, and management
reporting and inquiry functions. CARE management reports identify 
managing accounts, approving officials and cardholders’ accounts, 
transaction history, rejected transactions, and the reasons for the
rejections, such as transactions in excess of the cardholder’s credit 
limit, potential split purchases, inactive accounts, and blocked 
merchant category codes. 

During fiscal year 2002, the Air Force implemented additional purchase
card management controls using U.S. Bank’s CARE system. These
enhanced controls include an automated link of cardholder credit limits 
to budgetary funding authorizations to help ensure that purchase card 
activity will not exceed available funds. Another control feature 
monitors approving official span of control over cardholder accounts to 
help ensure that installations are meeting DOD and Air Force goals for 
reducing and eliminating excessive approving official span of control. 
The enhanced controls also include automated tracking of cardholder 
review of individual transactions on their monthly purchase card 
statements and billing officials’ approval of those statements. 
Statements cannot be approved until the cardholders have physically 
“touched” (clicked on) each transaction on the computer screen to 
indicate that they have reviewed the transactions. CARE automatically 
shuts down the accounts of billing officials who have not approved 
their consolidated statements within 60 days. No charges can be 
processed against these accounts until they are reviewed/reconciled and 
approved. U.S. Bank has also shut down accounts that indicate potential 
fraud. For example, the bank has shut down Air Force purchase card 
accounts due to out of state transactions on weekends and other 
suspicious patterns of activity that indicate potentially compromised 
accounts. 

[End of section] 

Appendix III: Examples of Air Force Purchase Card Fraud Cases: 

The following examples illustrate the types of cases investigated by 
the Air Force Office of Special Investigations. 

Case #1: 

During May 2000, after a Nellis AFB approving official retired, the new
approving official’s review of a cardholder’s monthly statements 
detected questionable transactions for which no receipts were 
available. The new official notified contracting officials who 
contacted Air Force investigators. The cardholder, an E-4, senior 
airman, used her government purchase card to obtain between $5,000 and 
$20,000 in merchandise, which she then stole and sold, pawned, or left 
at her residence. When questioned by her supervisor, who was the 
approving official, the airman admitted that she stole the items she 
had purchased with the government purchase card. When confronted by 
investigators, the airman refused to identify specific items of 
equipment that she stole. For example, the airman only stated that she 
purchased items from Home Depot and a local hardware store. Nellis AFB 
contracting officials told us that the cardholder used the purchase 
card to buy fax machines, calling cards, cordless telephones, digital
cameras, chairs, and laser jet printers and sold them to pawnshops and 
at swap meets for personal gain. 

The day before being court-martialed, the airman paid back approximately
$7,100 to the government. The airman waived her Article 31 rights 
[footnote 40] and pleaded guilty to purchasing and pawning over $7,100 
worth of personal items on her purchase card between May 1, 1999, and 
May 1, 2000. The airman was convicted of larceny in a General Court-
Martial and sentenced on March 17, 2001, resulting in a reduction in 
grade to E-1, $14,768 in military pay forfeiture, 7 month’s 
confinement, and a Bad Conduct Discharge. This fraud was able to occur 
and continue because the first approving official apparently had not 
reviewed the cardholder’s monthly purchase card statements and, 
therefore, had not detected or questioned the fraudulent transactions. 

Case #2: 

On September 27, 2000, the purchase card program coordinator at Misawa
Air Base, Japan, notified Air Force investigators about possible 
government purchase card fraud. The program coordinator’s audit of a 
cardholder’s account had revealed numerous undocumented 
charges/purchases. Air Force investigators determined that the fraud 
was committed by an E-4, senior airman, in the Civil Engineering 
Squadron whose own purchase card had been revoked for misuse. The 
airman took advantage of a co-worker’s inexperience and limited English 
language capability to obtain and improperly use her purchase card. The 
cardholder was a Japanese citizen employed by the Air Force. 

In early October 2000, the cardholder gave investigators a signed, sworn
statement, in which she related that from approximately March through
September 2000, an E-4, senior airman in the Civil Engineering Squadron
had repeatedly used the cardholder’s government purchase card to pay 
bills and make purchases, often without the cardholder’s knowledge. 
While the cardholder was aware that some of the purchases were made at 
the squadron’s Self Help Store, she told Air Force investigators that 
she had no knowledge of the types of items purchased. The cardholder 
also stated that when she inquired as to the nature of the purchases, 
the senior airman told her that he would take care of purchases using 
the card because of her limited English language capabilities. No 
attempt was made to correct this misuse of the purchase card until the 
bank declined a large purchase of approximately $50,000 due to the high 
dollar amount. The declined transaction flagged the account, and the 
contracting squadron initiated an inquiry. 

Contracting squadron records showed that the senior airman had 
purchased approximately $10,000 of merchandise using the Japanese
cardholder’s account. Air Force investigators’ review of Self Help Store
records failed to identify what was purchased and/or if the items had 
ever been received at the store. As a result, investigators were unable 
to determine whether criminal abuse had occurred. However, because the
investigation did identify procedural violations, investigators 
referred this matter to the command for action. The airman subsequently 
was reassigned from the civil engineering squadron. The airman was able 
to use the purchase card for unauthorized transactions because the 
cardholder failed to maintain custody of the purchase card. 

Case #3: 

On August 14, 2001, investigators assigned to the 325th Security Forces
Squadron at Tyndall AFB, Florida, received an allegation that a WG-5 
[Footnote 41] maintenance employee was using his government purchase 
card to buy personal use items. According to a witness, the cardholder 
had bragged about using his government purchase card to purchase tools, 
a television set, and a computer for his personal use. The witness told 
the investigators that he had accompanied the cardholder to a local 
hardware and auto parts stores in Panama City, Florida, and had 
observed the cardholder using his government purchase card to buy tools 
and other items for his son’s vehicle. The cardholder subsequently gave 
one of the tools to the witness and told him to keep it for his 
personal use. Squadron investigators coordinated with a local hardware 
store and obtained security videotape as evidence. The video depicted 
the cardholder, who was present with the witness, purchasing a drill 
bit, which the cardholder subsequently gave to the witness. The 
cardholder also allegedly used his government purchase card to pay for 
major engine repairs to his son’s vehicle. 

The investigators’ preliminary review of the cardholder’s account 
disclosed several unauthorized charges for dental work totaling 
approximately $1,800 and charges for an automotive engine repair for 
$1,181. Numerous additional suspect charges were identified on the 
cardholder’s purchase card account. The investigation, which is 
ongoing, has identified an estimated $5,000 in fraudulent purchases. 
Coordination with the cardholder’s command disclosed indications of an 
almost total lack of oversight on the part of the approving official. 

[End of section] 

Appendix IV: GAO Contacts and Staff Acknowledgments: 

GAO Contacts: 

Gayle L. Fischer, (202) 512-9577: 
Harold D. Reich, (213) 830-1078: 

Acknowledgments: 

Staff making key contributions to this report include Bertram J. Berlin,
James D. Berry, Jr., Cindy Barnes-Brown, Francine DelVecchio, Carlos M.
Garcia, Kenneth M. Hill, Jeffrey A. Jacobson, Noel J. Lance, Richard A.
Larsen, James D. Moses, Jerrod J. O’Nelio, Mark F. Ramage, Kenneth H.
Roberts, Sidney H. Schwartz, and Gary R. Wiggins. 

[End of section] 

Footnotes: 

[1] U.S. General Accounting Office, Purchase Cards: Control Weaknesses 
Leave Two Navy Units Vulnerable to Fraud and Abuse, GAO-01-995T 
(Washington, D.C.: July 30, 2001) and Purchase Cards: Continued Control 
Weaknesses Leave Two Navy Units Vulnerable to Fraud and Abuse, GAO-02-
506T (Washington, D.C.: Mar. 13, 2002). 

[2] U.S. General Accounting Office, Purchase Cards: Control Weaknesses 
Leave Army Vulnerable to Fraud, Waste, and Abuse, GAO-02-732 
(Washington, D.C.: June 27, 2002) and Purchase Cards: Control 
Weaknesses Leave Army Vulnerable to Fraud, Waste, and Abuse, GAO-02-
844T (Washington, D.C.: July 17, 2002). 

[3] U.S. General Accounting Office, Purchase Cards: Navy is Vulnerable 
to Fraud and Abuse, but Is Taking Action to Resolve Control Weaknesses, 
GAO-03-154T (Washington, D.C.: Oct. 8, 2002) and Purchase Cards: Navy 
is Vulnerable to Fraud and Abuse, but Is Taking Action to Resolve 
Control Weaknesses, GAO-02-1041, Washington D.C.: Sept. 27, 2002). 

[4] For this report, we limit the use of the term “fraudulent” to 
describe those instances in which someone has been convicted, or 
punished, for fraudulent activity under the Uniform Code of Military 
Justice. In all other instances, we use the phrase “potentially 
fraudulent.” 

[5] Air Force Instruction, 64-117, Air Force-wide Purchase Card 
Program, issued December 6, 2000. 

[6] Section 4535 of volume 1 of the Treasury Financial Manual. 

[7] In a letter dated April 30, 2002, DOD informed us that its 
reengineering memorandums and other pronouncements comply with 10 
U.S.C. 2784, which requires the Secretary of Defense to issue 
regulations that require, among other things, reconciliation of 
purchase card statements to receipts before the statements are 
forwarded to the disbursing office. 

[8] Suspended purchase card accounts are placed in inactive status 
pending further decisions about their use, such as returning them to 
active status after a required action has occurred or canceling the 
accounts. 

[9] U.S. Bank officials told us that in July 2001, they became aware of 
fraudulent Air Force purchase card transactions due to compromised 
accounts. The compromised accounts were caused by a fraud ring that 
used computers to randomly generate purchase card accounts and/or make 
counterfeit credit cards that resulted in fraudulent charges to credit
cards issued nationwide by several banks. 

[10] DOD has assigned agencywide purchase card program management 
responsibility to the Assistant Secretary of the Army (Acquisition, 
Logistics, and Technology). 

[11] Approving officials are also referred to as billing officials or 
accountable officials. These terms are used interchangeably by the Air 
Force. 

[12] U.S. General Accounting Office, Standards for Internal Control in 
the Federal Government, GAO/AIMD-00-21.3.1 (Washington D.C.: November 
1999). 

[13] Federal Acquisition Regulation, Part 4, Section 4.805 states that 
contracts and related documents at or below the simplified acquisition 
threshold of $100,000 must be retained for 3 years. 

[14] General Records Schedule No. 6, “Accountable Officer’s Accounts 
Records.” 

[15] DOD Financial Management Regulation, Volume 1, Chapter 9, 
"Financial Records Retention." 

[16] Memorandum from Director, Purchase Card Joint Program Management 
Office to assistant secretaries of defense agencies. Subject: Internal 
and Management Controls – DOD Purchase Card Program (July 5, 2001). 

[17] The range of our confidence interval, at a 95-percent confidence 
level, was between 5 percent and 23 percent. 

[18] The range of our confidence interval, at a 95-percent confidence 
level, was between 39 percent and 64 percent. 

[19] While the Air Force Instruction does not require that approving 
officials have refresher training, the surveillance checklist, which is 
included in the Instruction, includes a question on whether approving 
officials have received refresher training, thus implying that this 
training is required. 

[20] The Air Force Audit Agency has performed 253 reviews of 
installation purchase card controls since fiscal year 1996. 

[21] Air Force Audit Agency, Air Force Purchase Card Program, Audit 
Report F2002-0006-C06400 (Aug. 6, 2002). 

[22] Approving official billing accounts (also referred to as managing 
accounts) are summary accounts that include the accounts of individual 
cardholders under the purview of the approving official. To approve the 
monthly billing account statement, the approving official is required 
to review the monthly, reconciled statements for each of the cardholder 
accounts that are included in the approving official’s monthly billing 
statement. 

[23] Air Force Audit Agency, Installation Report of Audit: Air Force 
Purchase Card Program, Air Force Flight Test Center, Edwards AFB CA, 
Audit Report F2002-0005-DD0000 (Washington, D.C.: Jan. 31, 2002). 

[24] Air Force Audit Agency, Installation Report of Audit: Air Force 
Governmentwide Purchase Card Program, 37th Training Wing, Lackland AFB 
TX, Audit Report WR002003 (Washington, D.C.: Oct. 17, 2001). 

[25] Air Force Audit Agency, Installation Report of Audit: Air Force 
Purchase Card Program, 60th Air Mobility Wing, Travis AFB CA, Audit 
Report F2002-0019-WM0000 (Washington, D.C.: Nov. 29, 2001). 

[26] Air Force Audit Agency, Commanders Audit Program, Report of Audit: 
Unit Purchase Controls for the International Merchants Purchase 
Authorization Card, 99th Air Base Wing, Nellis AFB NV, Audit Report 
WN001C02 (Washington, D.C.: Apr. 4, 2001). 

[27] Public Law 107-248, 116 Stat. 1519, 1572. 

[28] Public Law 107-314 (H. R. Rep. No. 107-702). 

[29] Suspended purchase card accounts are placed in inactive status 
pending further decisions about their use, such as returning them to 
active status after a required action has occurred or canceling the 
accounts. 

[30] Convenience checks are courtesy checks provided by the purchase 
card-issuing bank, which are charged to a related purchase card 
account. DOD and Air Force policy permit use of convenience checks 
within specified purposes and amounts. 

[31] Federal Acquisition Regulation guidelines prohibit splitting 
purchase requirements into more than one transaction to avoid the need 
to obtain competitive bids on purchases over the $2,500 micropurchase 
threshold. 

[32] The JWOD program is a mandatory source of supply for all federal 
entities. It generates jobs and training for Americans who are blind or 
have other severe disabilities by requiring federal agencies to 
purchase supplies and services furnished by nonprofit agencies, such as
the National Industries for the Blind and the National Institute for 
the Severely Handicapped. 

[33] DOD Directive, Armed Forces Clothing Monetary Allowance Policy, 
(Number 1338.5, Mar. 9, 1998), and DOD Instruction, Armed Forces 
Clothing Monetary Allowance Procedures, (Number 1338.18, Jan. 7, 1998). 

[34] Air Force Instruction 36-2903, Dress and Personal Appearance of 
Air Force Personnel (June 8, 1998) and AFI 36-3014, Clothing Allowances 
for Air Force Personnel (Sept. 1, 1998). 

[35] The Eubank Service Award was established in 1990 to recognize the 
best Air Force Services Program at small bases with 5,000 or fewer 
civilian and military employees. 

[36] 72 Comp. Gen. 178, 179 (1993); 65 Comp. Gen. 508, 509 (1986). 

[37] U.S. General Accounting Office, Purchase Cards: Control Weaknesses 
Leave Army Vulnerable to Fraud, Waste, and Abuse, GAO-02-732 
(Washington, D.C.: June 27, 2002). 

[38] DOD Directive, Armed Forces Clothing Monetary Allowance Policy, 
(Number 1338.5, March 9, 1998); DOD Instruction, Armed Forces Clothing 
Monetary Allowance Procedures, (Number 1338.18, January 7, 1998); Air 
Force Instruction 36-2123, Management of Enlisted Aides; Air Force 
Enlisted Aide Handbook; and AFI 35-101, Public Affairs Policies and 
Procedures, Chapter 8, “Community Relations.” 

[39] We only included positive transaction amounts in the population. 

[40] Article 31 of the Uniform Code of Military Justice prohibits 
compulsory self-incrimination. Thus, the individual waived her right to 
refuse to answer questions, which may have tended to incriminate her. 

[41] Wage Grade (WG) positions are paid at hourly rates and generally 
pertain to blue collar government positions, such as maintenance, 
printing, and landscaping. 

[End of section] 

GAO’s Mission: 

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

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through the Internet. GAO’s Web site [hyperlink, 
http://www.gao.gov] contains abstracts and full 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 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 
[hyperlink, http://www.gao.gov] and select “Subscribe to daily E-mail 
alert for newly released products” under the GAO Reports Order GAO 
Products 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: 

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

Public Affairs: 

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