This is the accessible text file for GAO report number GAO-14-262 
entitled 'School-Meals Programs: USDA Has Enhanced Controls, but 
Additional Verification Could Help Ensure Legitimate Program Access' 
which was released on June 16, 2014. 

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

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

United States Government Accountability Office: 
GAO: 

Report to Congressional Requesters: 

May 2014: 

School-Meals Programs: 

USDA Has Enhanced Controls, but Additional Verification Could Help 
Ensure Legitimate Program Access: 

GAO-14-262: 

GAO Highlights: 

Highlights of GAO-14-262, a report to congressional requesters. 

Why GAO Did This Study: 

In fiscal year 2012, over 31.6 million children participated in USDA's 
National School Lunch Program (NSLP) at a cost of about $11.6 billion. 
In fiscal year 2013, USDA estimated NSLP certification errors of more 
than 8 percent, or $996 million. GAO was asked to review possible 
beneficiary fraud within the program. 

This report assesses (1) steps taken to help identify and prevent 
ineligible beneficiaries from receiving benefits in school-meal 
programs and (2) what opportunities exist to strengthen USDA's 
oversight of the school-meals programs. 

GAO reviewed NSLP policies, interviewed program officials, and 
randomly selected a nongeneralizable sample that included 25 of 7.7 
million approved household applications from 25 of 1,520 school 
districts in the Dallas, Texas, and Washington, D.C., regions. GAO 
performed limited eligibility testing using civilian federal-employee 
payroll data from 2010 through 2013 due to the unavailability of other 
data sources containing nonfederal employee income. GAO also conducted 
interviews with households. Ineligible households were referred to the 
Inspector General. 

What GAO Found: 

The U.S. Department of Agriculture (USDA) has taken several steps to 
implement or enhance controls to identify and prevent ineligible 
beneficiaries from receiving school-meals benefits. For example: 

* USDA worked with Congress to develop legislation to automatically 
enroll students who receive Supplemental Nutritional Assistance 
Program benefits for free school meals; this program has a more-
detailed certification process than the school-meals program. 

* Starting in the 2013-2014 school year, USDA increased the frequency 
with which state agencies complete administrative reviews of school 
districts from every 5 years to every 3 years. As part of this 
process, state agencies review applications to determine if 
eligibility determinations were correctly made. 

* In 2012, USDA issued guidance to clarify that school districts have 
the authority to verify approved applications for school-district 
employees when information indicates that the applicant misrepresented 
his or her income. 

GAO identified opportunities to strengthen oversight of the school-
meals programs while ensuring legitimate access, such as the following: 

* Exploring the feasibility of computer matching external income data, 
such as state payroll data, with participant information to identify 
households whose income exceeds eligibility thresholds for 
verification could help identify ineligible participants. Currently, 
school districts verify a sample of approved applications deemed 
“error-prone”-—statutorily defined as those with reported income 
within $1,200 of the annual income levels specified in program-
eligibility guidelines—-to determine whether the household is 
receiving the correct level of benefits (referred to as standard 
verification in this report). In a nongeneralizable review of 25 
approved applications, GAO found that 9 of 19 households that self-
reported household income and size information were ineligible and 
only 2 could have been subject to standard verification. 

* Verifying a sample of categorically eligible applications could help 
identify ineligible households. Currently, school-meal applicants who 
indicate categorical eligibility (by participating in certain public-
assistance programs or meeting an approved designation, such as foster 
children) are eligible for free meals and are generally not subject to 
standard verification. In a nongeneralizable review of 25 approved 
applications, 6 households indicated categorical eligibility, 2 of 
which were ineligible, and another may have been eligible for reduced-
price meals instead of free school meals. 

Figure: Results of GAO's Analysis of a Nongeneralizable Sample of 25 
Approved Household Applications from the 2010-2011 School Year: 

[Refer to PDF for image: illustration] 

1. 7.7 million approved household applications nationwide; 

2. 25 approved household applications; 

3. 6 household applications indicated categorical eligibility; 
4. 19 household applications indicated eligibility based on household 
size and income; 

5. 1 household application (from #3) was not eligible for free meals, 
but may have been eligible for reduced-price meals; 
6. 2 household applications (from #3) were not eligible for free or 
reduced-price meal benefits; 
7. 9 household applications (from #4) were not eligible for free or 
reduced-price meal benefits; 

11 household applications (from #6 and 7) were not eligible for free 
or reduced-price meal benefits; 
2 household applications (from #7) could have been subject to standard 
verification. 

Source: GAO analysis of school-meals household applications. 

[End of figure] 

What GAO Recommends: 

Among other things, GAO recommends that the Secretary of Agriculture 
develop a pilot program to explore the feasibility of using computer 
matching to identify households with income that exceeds program-
eligibility thresholds for verification, and explore the feasibility 
of verifying a sample of categorically eligible households. USDA 
generally agreed with the recommendations. 

View [hyperlink, http://www.gao.gov/products/GAO-14-262]. For more 
information, contact Steve Lord at (202) 512-6722 or lords@gao.gov. 

[End of section] 

Contents: 

Letter: 

Background: 

USDA Has Taken Steps to Help Identify and Prevent Ineligible 
Participants from Receiving Benefits: 

USDA Could Explore Options to Enhance the Verification Process to 
Further Strengthen Integrity While Ensuring Legitimate Access: 

Conclusions: 

Recommendations for Executive Action: 

Agency Comments: 

Appendix I: Objective, Scope, and Methodology: 

Appendix II: Additional Benefits Based on Eligibility in the School-
Meals Programs: 

Appendix III: Responsibilities of the Food and Nutrition Service, 
State Agencies, and School Districts: 

Appendix IV: Sample School-Meals Household Application: 

Appendix V: Comparison between Federal and Private-Sector Employees: 

Appendix VI: Comments from the U.S. Department of Agriculture Food and 
Nutrition Service: 

Appendix VII: GAO Contacts and Staff Acknowledgments: 

Tables: 

Table 1: Annual Income Eligibility Guidelines for a Family of Four: 

Table 2: Estimated Improper Payment Rates in the National School Lunch 
Program (NSLP) and School Breakfast Program (SBP) for Fiscal Year 2013: 

Figures: 

Figure 1: School Meals Application Highlighting Categorical 
Eligibility Requirements: 

Figure 2: Household-Application and Direct-Certification Methods for 
Eligibility Determinations in the School-Meals Programs: 

Figure 3: Standard Verification Thresholds for a Four-Member Household 
during the 2010-2011 School Year: 

Figure 4: Results of GAO's Analysis of Approved Applications: Limited 
Number of Ineligible Households Met Requirements for Standard 
Verification: 

Abbreviations: 

APEC: Access, Participation, Eligibility, and Certification: 

CBO: Congressional Budget Office: 

CNR: Child Nutrition and WIC Reauthorization Act of 2004: 

FDPIR: Food Distribution Program on Indian Reservations: 

FNS: Food and Nutrition Service: 

IPIA: Improper Payments Information Act of 2002: 

NSLA: Richard B. Russell National School Lunch Act: 

NSLP: National School Lunch Program: 

OIG: Office of the Inspector General: 

OMB: Office of Management and Budget: 

SBP: School Breakfast Program: 

SNAP: Supplemental Nutrition Assistance Program: 

TANF: Temporary Assistance for Needy Families: 

USDA: United States Department of Agriculture: 

USPS: U.S. Postal Service: 

[End of section] 

United States Government Accountability Office: 
GAO:
441 G St. N.W. 
Washington, DC 20548: 

May 15, 2014: 

Congressional Requesters: 

A well-balanced and nutritional diet for school children is essential 
for their overall health and well-being, and helps promote academic 
achievement. With children spending a considerable amount of their day 
at school, meals served during the school day play an important role 
in providing such a diet. In 1946, the Richard B. Russell National 
School Lunch Act (NSLA)--established the National School Lunch Program 
(NSLP),[Footnote 1] and in 1966 the Child Nutrition Act established 
the School Breakfast Program (SBP).[Footnote 2] These programs provide 
reimbursement to school districts and independent schools based on the 
number of meals served that meet certain federal requirements. Any 
child may purchase a meal through the NSLP or SBP; however, depending 
on household income, children may be eligible for reduced-price or 
free meals. The NSLP and SBP have the same eligibility requirements--
families with incomes at or below 130 percent of the federal poverty 
level are eligible for free meals, and those with incomes between 130 
and 185 percent of the federal poverty level are eligible for reduced-
price meals. In fiscal year 2012, over 31.6 million children 
participated in the NSLP each school day at a program cost of about 
$11.6 billion. Of the NSLP participants, 21.4 million--or about 68 
percent--received their lunches free or at a reduced price. In fiscal 
year 2012, approximately 12.9 million children participated in the SBP 
each school day, at a program cost of about $3.3 billion. Of those, 
over 10.8 million--or about 84 percent--received their breakfasts free 
or at a reduced price. 

These programs--administered by the United States Department of 
Agriculture's (USDA) Food and Nutrition Service (FNS)--operate in over 
89,000 schools and institutions.[Footnote 3] All schools participating 
in the NSLP or SBP must make free and reduced-price meals available to 
eligible children. Individual student or household eligibility for 
free or reduced-price meals is determined by school-district review of 
applications submitted by households or through a process referred to 
as "direct certification." Under direct certification, state agencies 
provide school districts with a list of students whose households 
receive certain public-assistance benefits, such as through the 
Supplemental Nutrition Assistance Program (SNAP), and school districts 
confer eligibility for free school meals to these students. Thus, 
participants are automatically certified for school-meals benefits 
without having to fill out a separate school-meals application. 
Students receiving certain public-assistance benefits or meeting an 
approved designation, such as if they are homeless or foster children, 
are categorically eligible for free-meal benefits. Categorically 
eligible students can be certified into the school-meals program 
either through an application or direct certification. The Office of 
Management and Budget (OMB) has designated the NSLP as 1 of 13 federal 
"high-error" programs due to its large estimated improper payments--
approximately $1.8 billion in fiscal year 2013.[Footnote 4] According 
to OMB guidance, an improper payment is any payment that should not 
have been made or that was made in an incorrect amount (including 
overpayments and underpayments) under statutory, contractual, 
administrative, or other legally applicable requirements.[Footnote 5] 

In July 2012, we reported that some children were inappropriately 
certified for free school meals because their households were 
considered categorically eligible for SNAP even though they were not 
receiving SNAP benefits.[Footnote 6] NSLA, as amended by the Child 
Nutrition and WIC Reauthorization Act of 2004 (CNR), requires 
participants in SNAP to be directly certified for school-meals 
benefits without a separate school-meals application.[Footnote 7] We 
recommended that USDA provide guidance clarifying that children in 
households receiving no SNAP benefits should not be directly certified 
to receive free school meals. USDA agreed with our recommendation and 
has taken steps to address it. In September 2009, we also reported 
that states and school food authorities[Footnote 8] conduct program-
integrity reviews, but that gaps in federal requirements for these 
reviews limit the reviews' effectiveness at identifying meal counting 
and claiming errors.[Footnote 9] We recommended that USDA modify the 
requirements for state and school food authority reviews and improve 
guidance on data collection; USDA agreed with and implemented our 
recommendations. 

Given the cost of the USDA school-meals programs and the high 
estimated improper-payment rates, you asked us to review possible 
beneficiary fraud occurring within the programs. Specifically, we 
assessed (1) what steps, if any, has USDA taken to help identify and 
prevent ineligible beneficiaries from receiving benefits in the school-
meals programs, and (2) what opportunities, if any, exist to 
strengthen USDA's oversight of the school-meals programs? We also 
report case-study examples of households that may have improperly 
received program benefits. Because of limited salary and income data 
available for all U.S. households, our case-study examples are limited 
to civilian executive-branch employees and United States Postal 
Service (USPS) employees.[Footnote 10] 

To determine any steps USDA has taken to identify and prevent 
ineligible beneficiaries from receiving benefits and what 
opportunities, if any, exist to strengthen USDA's oversight of the 
school-meals programs, we reviewed FNS policies and regulations, as 
well as the laws authorizing the school-meals programs. We also 
reviewed prior USDA reports and other studies concerning the programs. 
Further, we interviewed FNS officials to discuss the agency's 
administration of the programs. We selected the Washington, D.C., and 
Dallas, Texas metropolitan regions--areas with different federal-
employee concentrations--for further work.[Footnote 11] We conducted 
interviews of officials responsible for administration and oversight 
of the program at the city level in Washington, D.C., and at the state 
level in Maryland, Texas, and Virginia, as well as at 25 school 
districts located in these regions. We initially obtained data from 28 
school districts for our review--14 located in the Dallas, Texas 
metropolitan region and 14 in the Washington, D.C., metropolitan 
region, which includes Washington, D.C., and its Maryland and Virginia 
suburbs.[Footnote 12] In the Dallas, Texas metropolitan region we 
selected school districts with student enrollment over 10,000 
students. We selected all school districts in the Washington, D.C., 
metropolitan region excluding 56 charter school districts in 
Washington, D.C. However, we did not use data from 3 school districts--
1 located in the Dallas, Texas metropolitan region and 2 located in 
the Washington, D.C., metropolitan region--because the data were not 
reliable for our purposes. During the 2010-2011 school year, there 
were 57 school districts in Washington, D.C.; 49 in Maryland; 1,260 in 
Texas; and 154 in Virginia for a total of 1,520. This selection is not 
representative of all states, school districts, or school-meal 
participants. We assessed controls related to the identification and 
prevention of ineligible beneficiaries in accordance with internal 
control standards.[Footnote 13] 

To further identify opportunities, if any, that exist to strengthen 
USDA's oversight of the school-meals programs, we tested controls that 
are designed to identify and prevent ineligible school-meal 
beneficiaries. To do this, we selected a nongeneralizable sample of 48 
households participating in the NSLP for further review and 
investigation.[Footnote 14] To select the sample, we matched school-
meals eligibility data for the 2010-2011 school year from the 25 
school districts to executive-branch civilian federal-employee payroll 
data for approximately 2.5 million individuals to identify households 
that school districts deemed eligible for free or reduced-price school 
meals.[Footnote 15] We used civilian federal-employee payroll data 
obtained in 2010 through 2013 due to the unavailability of other data 
sources containing salary information for nonfederal employees. 
According to a January 2012 Congressional Budget Office Report, in 
2010, 1.7 percent of the U.S. workforce was made up of federal 
civilian employees--approximately 2.3 million compared to 111 million 
that were employed by the private sector and 20 million employed by 
state and local governments.[Footnote 16] The results of our work were 
not designed to be generalizable and do not include private-sector 
employees. 

A household member earning income does not preclude children in the 
household from being eligible for school-meals benefits. The purpose 
of our work was to select households for review and to determine 
whether any of the selected households were potentially ineligible for 
benefits based upon their pay and USDA eligibility guidelines. To 
assess the reliability of the school-meals eligibility and payroll 
data, we reviewed relevant documentation including NSLP applications, 
interviewed knowledgeable agency officials, and examined the data for 
errors and inconsistencies. We concluded that the school-meals 
eligibility data and payroll data were sufficiently reliable for the 
purposes of this report. 

In selecting our sample, we narrowed the civilian federal-employee 
payroll data to those with income during the July 2010 to December 
2010 period--to coincide with the start of the school year, when most 
school-meals eligibility determinations are made. We then matched the 
school-district beneficiary data to the federal-employee payroll data 
using the Social Security number, address, and name fields, to the 
extent they were available.[Footnote 17] On the basis of these 
matches, we randomly selected up to two households in each of the 25 
school districts for an in-depth review, for a total of 48 cases. 
[Footnote 18] Specifically, for each of the school districts, we 
reviewed one household that submitted an application that was used for 
benefit determination (25 cases), as well as one household that was 
directly certified (23 cases). For the 25 applicant cases, we reviewed 
available applications from 2010-2011 and salary information during 
this period, as well as any applications these households submitted 
during the 2011-2012 and 2012-2013 school years. For the 23 cases in 
which the household was directly certified into the school-meals 
programs without submitting an application due to the household's 
participation in SNAP or other assistance programs, we reviewed 2010-
2011 salary information and obtained and reviewed the application the 
household submitted for the assistance program that allowed direct 
certification. 

Our investigators interviewed the federal employees associated with 
the 48 households in our nongeneralizable sample to obtain additional 
information about their applications, households, and employment. The 
specific findings from the selected cases cannot be generalized to 
other, or all, school-meals beneficiaries or federal-employee 
households that received school-meals benefits. Because our data were 
limited to federal-employee households in 25 school districts that 
were randomly selected on a nongeneralizable basis, the results of our 
cases cannot be generalized to a larger population of school-meals 
participants or to the entire federal workforce. We plan to refer any 
potentially ineligible households to USDA and their school districts 
for appropriate action as warranted. 

We conducted this performance audit from February 2012 to May 2014 in 
accordance with generally accepted government auditing 
standards.[Footnote 19] Those standards require that we plan and 
perform the audit to obtain sufficient, appropriate evidence to 
provide a reasonable basis for our findings and conclusions based on 
our audit objectives. We believe that the evidence obtained provides a 
reasonable basis for our findings and conclusions based on our audit 
objectives. We conducted our related investigative work from June 2013 
to May 2014 in accordance with standards prescribed by the Council of 
the Inspectors General on Integrity and Efficiency. For a more-
detailed description of our scope and methodology, see appendix I. 

Background: 

School-Meals Programs Administration: 

Within USDA, FNS has overall responsibility for overseeing the school-
meals programs, which includes promulgating regulations to implement 
authorizing legislation setting nationwide eligibility criteria and 
issuing guidance. School-meals programs are administered at the state 
level by a designated state agency that issues policy guidance and 
other instructions to school districts providing the meals to ensure 
awareness of federal and state requirements. School districts are 
responsible for completing application, certification, and 
verification activities for the school-meals programs, and for 
providing children with nutritionally balanced meals each school day. 
The designated state agency conducts periodic reviews of the school 
districts to determine whether the program requirements are being met. 
Schools and households that participate in free or reduced-price meal 
programs may be eligible for additional federal and state benefits. 
Appendix II discusses those benefits. A graphic depicting the 
responsibilities of FNS, state agencies, and school districts can be 
found in appendix III. 

Eligibility Determinations and Verification: 

Children from families with incomes at or below 130 percent of the 
federal poverty level are eligible for free meals. Those with incomes 
between 130 percent and 185 percent of the federal poverty level are 
eligible for reduced-price meals. Income is any money received on a 
recurring basis--including, but not limited to, gross earnings from 
work, welfare, child support, alimony, retirement, and disability 
benefits--unless specifically excluded by statute.[Footnote 20] Table 
1 below shows the annual income-eligibility guidelines in effect for a 
family of four during the 2010-2011 through the 2013-2014 school 
years.[Footnote 21] 

Table 1: Annual Income Eligibility Guidelines for a Family of Four: 

School year: 2010-2011; 
Maximum household income for a family of four to qualify for free-meal 
benefits[A]: $28,665; 
Maximum household income for a family of four to qualify for reduced-
price-meal benefits[A]: $40,793. 

School year: 2011-2012; 
Maximum household income for a family of four to qualify for free-meal 
benefits[A]: $29,055; 
Maximum household income for a family of four to qualify for reduced-
price-meal benefits[A]: $41,348. 

School year: 2012-2013; 
Maximum household income for a family of four to qualify for free-meal 
benefits[A]: $29,965; 
Maximum household income for a family of four to qualify for reduced-
price-meal benefits[A]: $42,643. 

School year: 2013-2014; 
Maximum household income for a family of four to qualify for free-meal 
benefits[A]: $30,615; 
Maximum household income for a family of four to qualify for reduced-
price-meal benefits[A]: $43,568. 

Source: U.S. Department of Agriculture (USDA). 

[A] Excludes Alaska and Hawaii. 

[End of table] 

Children from families with incomes over 185 percent of the federal 
poverty level pay full price, though their meals are still subsidized 
to some extent.[Footnote 22] 

In addition, students who are in households receiving benefits under 
certain public-assistance programs--specifically, SNAP, Temporary 
Assistance for Needy Families (TANF), or Food Distribution Program on 
Indian Reservations (FDPIR)--or meet certain approved designations 
(such as students who are designated as homeless, runaway, or migrant; 
a foster child) are eligible for free school meals regardless of 
income. 

Certification: 

School districts certify students into the school-meals programs using 
one of two methods--either through (1) a household application 
identifying household information such as income and household size, 
or information on participation in public-assistance programs or (2) 
direct certification.[Footnote 23] During the 2012-2013 school year, 
11.7 million students were certified for free or reduced-price meals 
through a household application, and 12.3 million students were 
directly certified.[Footnote 24] Once a child is certified into the 
school-meals program, the eligibility determination is in effect for 
the entire year; households are not required to inform the school if 
wages rise above the income-eligibility guidelines during the school 
year. 

* Household application. Under the household-application method, a 
household submits an application provided by the school district. 
Since a household application is used for all members in the 
household, a single application can list multiple students. Schools 
send school-meals applications home at the beginning of each school 
year, but household applicants may apply at any time during the course 
of the year. Online applications are also available in some school 
districts. The applicant lists all sources of household income, the 
frequency with which it is received, and the names of all household 
members, among other information. One adult from the household signs 
the application, certifying that the information provided is correct. 
No supporting documentation--such as tax returns or pay stubs--is 
required at the time of application. In accordance with USDA guidance, 
school districts are not to conduct any actions to verify the 
information on the application during the certification process; they 
must accept the applications at face value and determine eligibility 
based on the information voluntarily disclosed in the application. 
[Footnote 25] In addition, students who are in households receiving 
benefits under certain public-assistance programs, including SNAP or 
TANF, or meet an approved designation--regardless of income--are 
categorically eligible for free school meals. For example, students 
who are designated as (1) homeless, runaway, or migrant; (2) a foster 
child; or (3) enrolled in a federally funded Head Start Program are 
categorically eligible for free meals. These households must state the 
reason for their categorical eligibility on the application along with 
any applicable public-assistance identification numbers. FNS officials 
told us that school district officials have a responsibility to verify 
homeless, runaway, and migrant applications as part of the application 
approval process. Figure 1 below is an example of a school-meals 
application and information required to be deemed categorically 
eligible. Appendix IV provides a sample school-meals household 
application. The school district reviews data on the application, such 
as household size, income, or participation in an approved public-
assistance program or other approved designation, and makes an 
eligibility determination. Starting with the 2011-2012 school year, 
applicants are required to provide the last four digits of their 
Social Security numbers rather than the entire nine-digit number. 
[Footnote 26] 

Figure 1: School Meals Application Highlighting Categorical 
Eligibility Requirements: 

[Refer to PDF for image: illustration] 

Applicants that indicate categorical eligibility do not need to 
provide income or household size information: 

Part 2. Benefits—-If any member of your household receives SNAP, 
FDPIR, or TANF, provide the Eligibility Determination Number (EDG) for 
the person who receives benefits and skip to Part 5. If no one 
receives these benefits, skip to Part 3. 

EDG: 

Part 3. If any child you are applying for is homeless, migrant, or a 
runaway, check the appropriate box and call your child's school. 
Homeless: 
Migrant: 
Runaway: 

Source: Excerpt from school-meals application provided by a school 
district. 

[End of figure] 

* Direct Certification. Children in households that receive certain 
public-assistance benefits--SNAP, FDPIR, or TANF--are automatically 
eligible for free school meals through "direct certification." Under 
the direct-certification method, school districts certify children who 
are members of households receiving public assistance as eligible for 
free school meals based on information provided by the state or local 
agency administering those programs. Starting in the 2008-2009 school 
year, school districts were required to directly certify SNAP 
households into the school-meals programs.[Footnote 27] A student or 
household that meets an approved designation--such as homeless or 
foster children--can also be directly certified into the school-meals 
programs without having to complete a household application. 

Figure 2 below describes the household-application and direct-
certification methods that households use to become certified for free 
or reduced-price meals. 

Figure 2: Household-Application and Direct-Certification Methods for 
Eligibility Determinations in the School-Meals Programs: 

[Refer to PDF for image: illustration] 

Household Application: 

Household submits school-meals application to school. 

School district reviews the information: 
The school district reviews the application-—analyzing information 
such as the household size, income, participation in an approved 
public assistance program and makes an eligibility determination. 

Household size and income: 
* Students whose applications report household incomes above 185 
percent of the federal poverty level: Not eligible for reduced-price 
or free meals; 
* Students whose applications report household incomes above 130 
percent but no more than 185 percent of the federal poverty level: 
Eligible for reduced-price meals. 
* Students whose applications report household income of no more than 
130 percent of the federal poverty level: Eligible for free meals. 

Categorical eligibility: 
* Students whose application indicates categorical eligibility: 
Eligible for free meals. 

Direct Certification: 

Household completes public assistance application or meets an approved 
designation, such as homeless or foster child[A]; 

Supplemental Nutrition Assistance Program (SNAP), Temporary Assistance 
for Needy Families (TANF), Food Distribution Program on Indian 
Reservations (FDPIR) application submitted to state or local agency; 

State or local agency sends list of students to school district: 
School district obtains a list of children whose household receives 
SNAP, TANF, FDPIR, or who meet an approved designation, such as 
homeless or foster child[A] from the state or local agency 
administering those programs and automatically enrolls the students 
for free meals. 

Source: GAO analysis of U.S. Department of Agriculture (USDA) 
information. Art Explosion (images). 

[A] Students who meet an approved designation--(1) homeless, runaway, 
or migrant; (2) foster child; or (3) enrolled in a federally funded 
Head Start Program--are categorically eligible for free school meals. 

[End of figure] 

Verification: 

After school districts certify household eligibility for school-meals 
program benefits, they must annually verify a sample of household 
applications approved for free or reduced-price school-meals benefits 
to determine whether the household has been certified to receive the 
correct level of benefits--we refer to this process as "standard 
verification."[Footnote 28] As dictated by statute, school districts 
are required to verify a random sample of applicants. The sample size 
is equal to the lesser of 3 percent of approved applications, selected 
from error-prone applications, or 3,000 error-prone applications 
unless an alternative sample size is used.[Footnote 29] For the 
purposes of standard verification, the NSLA defines error-prone 
applications as certified applications with monthly income within $100 
of--or with annual income within $1,200 of--the income eligibility 
limits for free or reduced-price meals. Households that indicate 
categorical eligibility on an application and households that enter 
the program through direct certification are generally not subject to 
the standard verification process.[Footnote 30] 

Further, as described in USDA's eligibility manual for school meals, 
school districts are obligated to verify additional applications if 
they deem them to be questionable, which is referred to as for-cause 
verification.[Footnote 31] Verification--whether standard or for-
cause--is conducted only for those beneficiaries receiving benefits 
through the household-application process; directly certified 
households are not subject to verification. Households selected for 
verification (standard or for-cause) must submit supporting 
documentation--such as pay stubs, benefit award letters from state 
agencies for benefits such as Social Security, or supplemental 
security income, or support payment decrees from courts--to the school 
district, or be removed from the program. The school district reviews 
the information, determines whether the household's free or reduced-
price status is correct, and makes corrections, as necessary. 

Improper Payment Rates Associated with the School Meals Programs: 

The Improper Payments Information Act of 2002, as amended,[Footnote 
32] requires agencies to identify, measure, prevent, and report their 
improper payment amounts and to develop and implement improper payment 
reduction plans, among other things. For fiscal year 2013, USDA 
reported that the NSLP and SBP had estimated improper payment rates of 
approximately 15.7 percent and 25.3 percent, respectively--equating to 
about $1.8 billion and $831 million.[Footnote 33] Table 2 below shows 
the portion of improper payments in the NSLP and SBP attributable to 
certification errors versus counting and claiming errors, as reported 
for fiscal year 2013. 

Table 2: Estimated Improper Payment Rates in the National School Lunch 
Program (NSLP) and School Breakfast Program (SBP) for Fiscal Year 2013: 

Type of error: Certification; 
NSLP: 
Amount of improper payments: $996 million; 
Percentage of program payments deemed improper: 8.8%; 
SBP: 
Amount of improper payments: $312 million; 
Percentage of program payments deemed improper: 9.5%. 

Type of error: Counting/claiming; 
NSLP: 
Amount of improper payments: $778 million; 
Percentage of program payments deemed improper: 6.9; 
SBP: 
Amount of improper payments: $519 million; 
Percentage of program payments deemed improper: 15.8. 

Type of error: Total; 
NSLP: 
Amount of improper payments: $1.774 billion; 
Percentage of program payments deemed improper: 15.7%; 
SBP: 
Amount of improper payments: $831 million; 
Percentage of program payments deemed improper: 25.3%. 

Source: U.S. Department of Agriculture (USDA). 

Note: The United States Department of Agriculture Office of the 
Inspector General (OIG) reported that NSLP and SBP improper payment 
estimates may be unreliable because they were based on the 2005-2006 
school year and confidence levels could not be provided for subsequent 
years. Estimated improper payments in the NSLP and SBP are due to 
certification errors, such as incorrect eligibility determinations, as 
well as noncertification errors, such as meal-counting and meal-
claiming errors. 

[End of table] 

In March 2013, USDA's Office of the Inspector General (OIG) reported 
that FNS officials were continuing to work with states to address the 
high estimated improper-payments rate. However, in the same report, 
the USDA OIG noted that FNS baselines for improper payments may be 
unreliable because they were based on the 2005-2006 school year and 
confidence levels could not be provided for subsequent years.[Footnote 
34] To update data used to determine FNS current improper-payment 
estimates, FNS hired a contractor to conduct a study for the 2012-2013 
school year. In 2012, FNS officials told us that they believe this 
study will better reflect current improper-payment rates for the NSLP 
and SBP, and in April 2014 stated that they expect to release the 
results of the study in early 2015. 

USDA Has Taken Steps to Help Identify and Prevent Ineligible 
Participants from Receiving Benefits: 

USDA Has Increased the Use of Direct Certification: 

To address the high improper-payment rates in the school-meals 
programs, among other actions, USDA worked with Congress to develop 
the Child Nutrition and WIC Reauthorization Act of 2004 
(CNR).[Footnote 35] CNR required school districts to directly certify 
students that receive SNAP benefits for free meals in all school 
districts by the 2008-2009 school year. USDA officials told us that 
they are emphasizing the use of direct certification, because, in 
their opinion, it helps prevent certification errors without 
compromising access. School-meals programs and SNAP have similar 
income-eligibility limits.[Footnote 36] Further, the application 
process for SNAP is more detailed than the certification process under 
the NSLP. Direct certification has reduced the administrative burden 
on SNAP households, as they do not need to submit a separate school-
meals application. It also reduces the number of applications school 
districts must review. In commenting on a draft of this report, USDA 
reiterated that school districts do not have access to SNAP 
eligibility documents, are not required to review household SNAP 
applications, and therefore accept the SNAP eligibility determination 
at face value. FNS officials also told us that once school districts 
receive confirmation that a household is eligible for direct 
certification, they are not required to determine whether the 
household is still eligible throughout the year. 

To test the effectiveness of direct certification in identifying and 
preventing ineligible participants from receiving benefits, we 
reviewed a nongeneralizable sample of 23 households who were directly 
certified for free-meal benefits and found two cases where the 
household appeared ineligible for SNAP benefits, and therefore may 
have been inappropriately directly certified into the school-meals 
programs, as described below. Because these households were directly 
certified for school-meals benefits, the school district would not be 
aware of the SNAP error unless notified by the appropriate state 
agency. 

* One household received SNAP benefits from October 2009 to October 
2010. However, one household member started employment in March 2010 
and, based upon his biweekly pay of approximately $3,300, his 
household of four members would no longer have qualified for SNAP 
benefits. The SNAP Notice of Food Benefit Extension sent to the 
household dated March 7, 2010, required notification of changes to job 
status and rate within 10 days beginning in May 2010. Based on the 
change of wages in March 2010, this household would have not have 
remained eligible for SNAP benefits and thus would not have been 
eligible for direct certification for free school meals during the 
2010-2011 school year. 

* One household's SNAP application incorrectly omitted a member of the 
household who earned income and provided financial support. Had the 
SNAP application included this household member's income, the 
household would not have qualified for SNAP benefits; therefore, this 
household should not have been directly certified for free school 
meals during the 2010-2011 school year. 

We will include these instances in our referrals to USDA and the state 
agency administering SNAP for appropriate action, as warranted. 

Since passage of the CNR, the number of school districts directly 
certifying SNAP-participant children has continued to increase. For 
example, during the 2008-2009 school year, 78 percent of school 
districts directly certified students, and by the 2012-2013 school 
year, this percentage had grown to 91 percent of school districts, 
bringing the estimated percentage of SNAP-participant children 
directly certified for free school meals to 89 percent.[Footnote 37] 

USDA is also conducting demonstration projects in selected states and 
school districts to explore the feasibility of directly certifying 
children that participate in the Medicaid program.[Footnote 38] During 
the demonstration projects, eligible children will be directly 
certified for free school meals based on a review of income and 
participation information received from Medicaid agencies through 
automated data-matching processes, with no household-application 
requirement. Five states participated in the studies during the 2012-
2013 school year, six participated during the 2013-2014 school year, 
and more are expected to participate during the 2014-2015 school year. 
[Footnote 39] 

USDA Has Increased the Frequency of State Agency Reviews: 

USDA requires administering state agencies to conduct regular, on-site 
reviews--referred to as administrative reviews--to evaluate school 
districts that participate in the school-meals programs.[Footnote 40] 
The Healthy, Hunger-Free Kids Act of 2010 increased the frequency of 
these reviews from every 5 years to every 3 years.[Footnote 41] 
Starting with the 2013-2014 school year, state agencies are required 
to conduct administrative reviews at least once during a 3-year review 
cycle, with no more than 4 years between the reviews. During this 
process, state agencies are to determine whether free, reduced-price, 
and paid lunches were properly provided to eligible students; and that 
meals are counted, recorded, consolidated, and reported through a 
system that consistently yields correct claims.[Footnote 42] 

As part of this process, state agencies are to conduct on-site reviews 
of school districts to help ensure that applications are complete and 
that the correct eligibility determinations were made based on 
applicant information. In reviewing eligibility determinations, the 
state agency may elect to review documentation for all students 
certified for free or reduced-price or a statistically valid sample. 
Once the names of students subject to review have been identified, the 
state agency reviews the household application or direct certification 
to determine whether or not the certification decision was correct; 
supporting documents--such as payroll records or benefit award letters-
-are not obtained during the administrative review process. In 
addition, during the on-site review, state agency officials are to 
observe the meal service--to determine whether the meals claimed meet 
the federal requirements for a reimbursable lunch, including nutrition 
and portion requirements--as well as the process of counting and 
recording meals. School districts that have administrative review 
findings are to submit a corrective-action plan to the state agency, 
and the state agency is to follow up to determine whether the issue 
has been resolved. USDA regulations require all state agencies to 
report the results of administrative reviews to FNS by March 1 of each 
school year.[Footnote 43] FNS officials told us that as part of their 
oversight of state agencies, they confirm that agencies have completed 
the administrative reviews. 

We reviewed administrative review reports from the 25 school districts 
we selected that were completed between February 2008 and December 
2012. Administrative review reports from 11 school districts cited 
some incorrect eligibility determinations. Incorrect eligibility 
determinations ranged from 1 to 15 per district--based on the stated 
information on the application. The number of incorrect determinations 
found in each school district was small compared to the number of 
applications reviewed, which ranged from 687 to 8,398. As required, 
these 11 school districts submitted a corrective-action plan to the 
state addressing how they would ensure that all meal-benefit 
applications are reviewed and certified based on eligibility 
guidelines. The state agency determines if the school district's 
corrective action satisfactorily resolves the problem; the state 
agency cannot close the review until all identified issues have been 
addressed. The administrative review reports from the remaining 14 
school districts in our sample did not cite any incorrect eligibility 
determinations. 

In commenting on a draft of this report, USDA told us that it makes 
grant funds available annually to states to fund the performance of 
additional administrative reviews, oversight, and training for school 
districts with a high level or risk of administrative errors. USDA 
stated that since fiscal year 2005, $4 million has been set aside 
annually for these grants. According to USDA, from fiscal year 2005 to 
2013, FNS awarded 60 grants totaling $26.5 million. 

USDA Has Issued Guidance on the Use of For-Cause Verification: 

As discussed earlier in this report, school districts are obligated to 
verify the eligibility of applicants whose application information is 
deemed questionable under the "for-cause" verification process. 
Examples of relevant recent cases include the following: 

* The Chicago Board of Education OIG reported that in fiscal year 
2012, a cohort of highly paid and high-level Chicago Public Schools 
administrators falsified information on school-meals applications and 
the office noted the possibility of system-wide school-meals fraud. 
[Footnote 44] Specifically, the report cited 21 principals, assistant 
principals, and recently promoted assistant principals who understated 
their own income or falsified the number of household members, 
including leaving themselves off the applications. 

* In July 2013, the State of New Jersey Office of the State 
Comptroller issued a report on fraudulent school-lunch program 
applications filed by public employees.[Footnote 45] The report, 
reviewing a sample of schools that received more than $1 million in 
reimbursements for school lunches in the 2010-11 school year, found a 
number of public employees who materially underreported their 
household income on school-lunch applications, including 101 public 
employees (elected school-board members and school-district employees 
among them) who provided materially false information. Further, 
according to the July 2013 report, numerous applicants substantially 
underreported the income of household members and many failed to list 
income-generating household members on their applications. 

In February 2012, USDA distributed guidance to state administrators to 
clarify that school districts have the authority to review approved 
applications for free or reduced-price meals for school-district 
employees when known or available information indicates school-
district employees may have misrepresented their incomes on their 
applications. However, this for-cause verification should be used 
selectively and not to verify the household income of all school 
district employees whose children are certified for free or reduced-
price meals. Under the guidance, school districts can identify 
children of school-district employees and use salary information 
available to them to identify questionable applications and then 
conduct for-cause verification on the questionable applications, if 
necessary. In August 2012, USDA also updated its school-meals 
eligibility manual--used by school districts to determine and verify 
eligibility--with this guidance. Our analysis of this guidance is 
presented below. 

USDA Could Explore Options to Enhance the Verification Process to 
Further Strengthen Integrity While Ensuring Legitimate Access: 

Guidance to Clarify Use of For-Cause Verification Could Enhance 
Oversight: 

As discussed earlier in this report, USDA regulations require that 
school districts conduct for-cause verification of all questionable 
applications.[Footnote 46] Officials from 11 of the 25 school 
districts told us during our interviews that they conduct for-cause 
verification. These officials provided examples of how they would 
identify suspicious applications, such as when a household submits a 
modified application--changing income or the household members--after 
being denied or when different households include identical public-
assistance benefit numbers (e.g., if different households provide 
identical SNAP numbers). However, officials from 9 of the 25 school 
districts told us that they did not conduct any for-cause 
verification. For example, one school-district official explained that 
the school district accepts applications at face value. An official 
from another school district said that his district does not conduct 
for-cause verification and added that he is not sure how to identify 
questionable applications. Additionally, officials from 5 of the 25 
school districts told us they only conduct for-cause verification if 
someone (such as a member of the public or a state agency) informs 
them of the need to do so on a household. Although not generalizable, 
responses from these school districts provide insights about if and 
under what conditions for-cause verifications are conducted. 

USDA officials stated that school districts have the obligation to 
conduct for-cause verification if they suspect inaccurate information, 
but added that staff may be hesitant to perform it because of the 
potential work burden it may create. USDA officials also told us that 
some school districts may be reluctant to conduct for-cause 
verification because of concerns about appearing to target certain 
groups of people. In April 2013, USDA issued a memorandum stating that 
effective for the 2013-2014 school year, all school districts must 
specifically report the total number of applications that were 
verified for cause. Prior to this, USDA did not collect any 
information on applications that have undergone for-cause 
verification. USDA officials told us that they will use the 
information to determine the frequency with which school districts 
conduct for-cause verification. This information is to be provided to 
USDA in April 2014; however, since this is the first year the 
information is being collected, it may take school districts 
additional time to finalize the reports. While school districts are to 
report the number of applications verified for cause, the outcomes of 
those verifications will be grouped with the outcomes of applications 
that have undergone standard verification. As a result, USDA plans to 
review the results to determine the frequency with which school 
districts conduct for-cause verification but will not have information 
on specific outcomes, which it may need to assess the effectiveness of 
for-cause verifications and to determine what actions, if any, are 
needed to improve program integrity. 

During our review of 25 households that applied for and received 
school-meals benefits, we identified one household that reapplied for 
school-meals benefits during the 2011-2012 school year less than 2 
weeks after being denied benefits for not meeting the eligibility 
requirements. The new application removed a source of income--child 
support--and the household was approved for reduced-price meals. 
[Footnote 47] When we interviewed the applicant, she said that she 
could not remember if she received child support payments at the time 
she resubmitted the application. This household also applied for 
school-meal benefits during the 2012-2013 school year. The application 
did not indicate child-support payments, and the household was 
subsequently approved for reduced-price meals. This household was not 
subjected to for-cause verification by the school district even though 
households resubmitting an application with less income a short time 
after being denied benefits could be a red flag to indicate that for-
cause verification should be conducted.[Footnote 48] 

While USDA has issued guidance specific to school-district employees 
and instructs school districts to verify questionable applications in 
its school-meals eligibility manual, we found that the guidance does 
not provide possible indicators or describe scenarios that could 
assist school districts in identifying questionable applications. 
Standards for Internal Control in the Federal Government call for 
agencies to design control activities to ensure management's 
directives are carried out.[Footnote 49] Reviewing the data gathered 
on for-cause verification for the 2013-2014 school year could help 
USDA determine if data on the outcome of for-cause verifications 
should be reported separately from standard verification results. 
Further, as noted above, evaluating this data could help USDA 
determine whether additional guidance would be beneficial to assist 
school districts in identifying applications that should be subject to 
for-cause verification. Such guidance could include criteria and 
examples of possible indicators of questionable or ineligible 
applications. 

Income-Based Verification Could Enhance Oversight: 

USDA's standard verification process--the terms of which are 
statutorily defined--makes it difficult to detect all households that 
misreport their income and that are ineligible for program benefits. 
It could also result in the removal of eligible beneficiaries, as 
households that do not respond to the verification notice are removed 
from the program. Electronically matching household-application 
information to other data sources--such as state income databases or 
public-assistance databases--could hold promise in identifying high-
income households for validation while not disrupting program benefits 
to eligible households. 

Standard Verification Limited to Small Portion of Beneficiaries: 

As described earlier in this report, with the exception of for-cause 
verification, standard verification is generally limited to approved 
applications considered "error-prone."[Footnote 50] Error-prone is 
statutorily defined as approved applications where stated income is 
within $100 of the monthly or $1,200 of the annual applicable income-
eligibility guideline. Households with reported incomes that are more 
than $1,200 above or below the free-meals eligibility threshold and 
more than $1,200 below the reduced-price threshold would generally not 
be subject to this verification process.[Footnote 51] Figure 3 shows 
the income thresholds of applicants that would and would not be 
considered error-prone for a four-person household during the 2010-
2011 school year. 

Figure 3: Standard Verification Thresholds for a Four-Member Household 
during the 2010-2011 School Year: 

[Refer to PDF for image: illustration] 

Annual household income at time of application for the 2010-2011 
school year: 

185% of the federal poverty level: $40,793; 
Subject to standard verification: $39,593 to $40,793; 
Eligible for reduced-price meal benefits: 
Annual incomes between 130% and 185% of the federal poverty level: 
$28,665 (130% of the federal poverty level); 
Subject to standard verification: $27,465 to $29,865; 
Eligible for free-meal benefits: 
Annual incomes at or below 130% of the federal poverty level. 

Source: GAO analysis of legislation and U.S. Department of Agriculture 
data. 

Note: Data are from the Richard B. Russell National School Lunch Act 
(NSLA) and the United States Department of Agriculture (USDA) income-
eligibility guidelines for the 2010-2011 school year. Only approved 
applications are subject to standard verification. 

[End of figure] 

In addition to the nongeneralizable sample of 23 households receiving 
school-meal benefits through direct certification discussed in the 
previous section, we reviewed a nongeneralizable sample of 25 
households receiving school-meals benefits through an approved 
application. Nineteen household applications were certified based upon 
their stated income and household size.[Footnote 52] Of these, we 
determined that 9 were not eligible for free or reduced-price-meal 
benefits they were receiving because their income exceeded eligibility 
guidelines. Further, 2 of these 9 households stated annualized incomes 
that were within $1,200 of the eligibility guidelines. These two 
households could have been subject to standard verification had they 
been selected as part of the sample by the district; however, they 
were not selected or verified. The remaining 7 of 9 households stated 
annualized incomes that did not fall within $1,200 of the eligibility 
guidelines and thus would not have been subject to standard 
verification. Figure 4 shows the results of our review. 

Figure 4: Results of GAO's Analysis of Approved Applications: Limited 
Number of Ineligible Households Met Requirements for Standard 
Verification: 

[Refer to PDF for image: illustration] 

25 approved household applications: 

19 household applications: indicated eligibility based on household 
size and income; 
* Of the 19: 9 household applications were not eligible for free or 
reduced-price meals; and; 
* Of the 9: 2 household applications could have been subject to 
standard verification. 

6 household applications: indicated categorical eligibility; 
* Of the 6: 2 household applications were not eligible for free or 
reduced-price meals; and; 
* 1 household application was not eligible for free meals, but may 
have been eligible for reduced-price meals. 

A combination of 11 household applications from the two categories 
were not eligible for free or reduced-price meals. 

Source: GAO analysis of school-meals household applications. 

Note: There are two ways children may be classified as categorically 
eligible: (1) through participation in certain public-assistance 
programs, such as Supplemental Nutrition Assistance Program (SNAP) or 
Temporary Assistance for Needy Families (TANF), or (2) through meeting 
an approved designation, such as homeless or foster child. 

[End of figure] 

Of the 19 households shown above that indicated eligibility based on 
self-reported household size and income, we determined that 9 were not 
eligible for free or reduced-price-meal benefits they were receiving 
because their known income exceeded eligibility guidelines. For 
example, one household we reviewed submitted a school-meals 
application for the 2010-2011 school year seeking school-meals 
benefits for two children. The household stated an annual income of 
approximately $26,000 per year, and the school district appropriately 
certified the household to receive reduced-price-meal benefits based 
on the information on the application. However, through review of the 
payroll records, we determined that the adult applicant's income at 
the time of the application was approximately $52,000--making the 
household ineligible for benefits. This household also applied for and 
received reduced-meal benefits for the 2011-2012 and 2012-2013 school 
years by understating its income. Its 2012-2013 annualized income was 
understated by about $45,000. Because the income stated on the 
application during these school years was not within $1,200 per year 
of the income-eligibility requirements, the application was not deemed 
error-prone and was not subject to standard verification. Had this 
application been subjected to verification, a valid pay stub would 
have indicated the household was ineligible. We interviewed the adult 
applicant as part of our investigation, and the applicant admitted to 
underestimating her income. 

Another household in our sample submitted a school-meals application 
for the 2010-2011 school year--stating an income that equated to 
approximately $32,500 annually and a household size of five members--
and was approved for free-meal benefits. However, at the time of the 
application, the household's annualized income was at least $60,000, 
making the household ineligible for free or reduced-price meals. The 
household application stated an annualized income that put it within 
the error-prone range; however, it was not among the 3 percent sample 
of error-prone applications selected for verification.[Footnote 53] 
This household applied for school-meals benefits for the 2011-2012 
school year--stating an annualized income that equates to 
approximately $39,600--and was approved for reduced-price meals. 
However, based on our review of payroll information, household income 
was at least $73,000 during 2011--a difference of about $33,000--
making this household ineligible for free or reduced-price meals. When 
interviewed, the applicant said that her children completed the 
application and that she signed it.[Footnote 54] 

In another instance, a household submitted a school-meals application 
for the 2010-2011 school year--stating an annualized income that 
equates to approximately $19,200 and a household size of four--and was 
approved for free school-meals benefits. This application omitted a 
parent living in the household and earning annualized income of 
approximately $55,000. Had the wage-earner and his income been 
included, this household would not have qualified for free or reduced-
price meals. This household applied for and was approved for free 
school-meals for the 2011-2012 and 2012-2013 school years. Again, 
these applications omitted the parent and his wages--which amounted to 
approximately $62,000 during 2011 and $64,000 during 2012. Had his 
income been included, the household would not have qualified for free 
or reduced-price meals. When interviewed, the parent said that he was 
not aware that his children had been receiving free school-meals 
benefits. Because the stated income on the application was outside the 
error-prone range, and the school district only verified error-prone 
applications during these school years, this household would not have 
been subject to standard verification. Individuals with knowledge of 
the program-eligibility guidelines could understate their income to 
avoid scrutiny, as this would likely prevent the application from 
being reviewed under standard verification, although for-cause 
verification could identify the understatement.[Footnote 55] 

For fiscal year 2013, USDA reported NSLP and SBP certification errors 
of approximately 8.8 percent and 9.5 percent as part of its improper 
payment estimation.[Footnote 56] As explained previously, USDA OIG 
noted that these estimates may be unreliable because they were based 
on the 2005-2006 school year and confidence levels could not be 
provided for subsequent years. FNS has hired a contractor to conduct a 
revised study for the 2012-2013 school year, which is expected to be 
complete in November 2014.[Footnote 57] 

Current Standard Verification Process Could Remove Eligible 
Beneficiaries: 

Once a household application has been certified as eligible to receive 
benefits, and if it is selected for verification, school districts 
obtain supporting documentation from the applicant--such as pay stubs 
or benefit-award letters--in order to evaluate whether or not the 
household's free or reduced-price status is correct.[Footnote 58] 
However, because the verification process relies on responses from 
applicants, it could lead to eligible children being removed from the 
program if the applicant does not respond to the school district's 
verification request. USDA told us that during the 2012-2013 school 
year, school districts verified approximately 203,200 applications. 
[Footnote 59] Of these, 43.5 percent were receiving the correct level 
of benefits, and approximately 23.6 percent had their level of 
benefits adjusted to properly reflect their eligibility based on 
verification. Of the applications selected for verification, 32.8 
percent did not respond and were excluded from receiving free or 
reduced-price school meals. 

A household failing to respond to a request for verification may not 
mean the household was ineligible for the initial benefits 
received.[Footnote 60] USDA officials also told us that households may 
not respond because the adult applicant may not have time to provide 
supporting documentation, may not understand the instructions, or may 
be hesitant to provide income information, though the household is 
still eligible to receive the benefits. Further, a study commissioned 
by the USDA to examine outcomes of the verification process during the 
fall of 2002 found that approximately half of the households that did 
not respond to the verification request were eligible for free or 
reduced-price meals.[Footnote 61] 

Electronic Matching Could Help to Identify Potentially Ineligible 
Applications: 

As described above, standard verification is generally limited to 
approved applications where stated income is within $1,200 of the 
annual applicable income-eligibility guideline amount. Applications 
with stated income outside of these thresholds would generally not be 
subject to standard verification. However, our review of a 
nongeneralizable sample of 25 households found 9 applications that 
were ineligible for benefits, 7 of which would have been excluded from 
standard verification. 

Standards for Internal Control in the Federal Government indicate that 
internal controls should include control activities and risk 
assessments.[Footnote 62] These, among other controls, should be 
effective and efficient in enforcing program requirements. Independent 
verification is a key detection and monitoring component of an 
agency's fraud-prevention framework and is a fraud-control best 
practice.[Footnote 63] One method to identify potentially ineligible 
applicants and effectively enforce program-eligibility requirements is 
through the independent verification of income information with an 
external source, such as state payroll data. States or school 
districts, through data matching, could identify households that have 
income greater than the eligibility limits for further follow-up. This 
risk-based approach would allow school districts to focus on 
potentially ineligible families, while not interrupting program access 
to other participants. 

While electronic verification could yield positive results, there are 
some potential limitations. For example, state income databases may 
not contain all sources of household income--such as child-support 
payments or income earned by individuals who do not have a Social 
Security number. Additionally, it may not be cost-effective or 
possible for school districts to use external data when conducting 
verification. Thus, states may be better positioned to complete this 
matching and to report findings to specific school districts. 

A study commissioned by USDA to explore the feasibility of computer 
matching in the NSLP during the 2004-2005 school year cited 
limitations to having school districts directly verify income 
information with state agencies.[Footnote 64] For example, because 
income data are reported for individuals, not households, school 
districts would need to obtain Social Security numbers for all income 
earners in the household in order to verify household income. The 
study also found that computer-matching results can be inaccurate and 
that income discrepancies between the state database and the household 
application would require follow-up with the household that is similar 
to the existing verification process. However, technology and data-
matching software and techniques have improved significantly in the 
last decade and could hold promise in efficiently identifying only 
potentially ineligible households for further follow-up while not 
removing program beneficiaries whose incomes are within the 
eligibility guidelines. 

Electronic verification of a sample of applicants (beyond those that 
are statutorily defined as error-prone) through computer matching by 
school districts or state agencies with other sources of information--
such as state income databases or public-assistance databases--could 
help effectively identify potentially ineligible applicants. However, 
it is not clear whether such a process is cost-effective. Thus, 
developing a pilot to explore the feasibility of implementing a cost-
effective mechanism to conduct electronic verification at the state or 
school-district level could help inform the extent to which this 
alternative is feasible. Because standard verification is dictated by 
statute, if the results of the pilot show promise in identifying 
ineligible beneficiaries, developing a legislative proposal to expand 
the verification process to include independent electronic 
verification for a sample of all school-meals applications could help 
USDA identify and prevent ineligible beneficiaries in the school-meals 
program. 

Expanded Verification of Applications That Indicate Categorical 
Eligibility Could Enhance Program Integrity: 

We found that ineligible households may be receiving free school-meals 
benefits by submitting applications that falsely state that a 
household member is categorically eligible for the program due to 
participating in certain public-assistance programs--such as SNAP or 
TANF--or meeting an approved designation--such as foster child or 
homeless. Of the 25 household applications we reviewed, 6 were 
approved for free school-meals benefits based on categorical 
eligibility, and 3 of these were potentially ineligible for the 
benefit.[Footnote 65] Figure 3 illustrates these results. 
Specifically, we found the following: 

* One household applied for benefits during the 2010-2011 school year--
providing a public-assistance benefit number--and was approved for 
free-meal benefits. However, when we verified the information with the 
state, we learned that the number was for medical-assistance benefits--
a program that is not included in categorical eligibility for the 
school-meals programs. When interviewed, the parent said that he could 
not remember if the benefits they received were SNAP or medical 
assistance. On the basis of our review of payroll records, this 
household's annualized income of at least $59,000 during 2010 would 
not have qualified the household for free or reduced-price-meal 
benefits. This household applied for school-meals benefits during the 
2011-2012 and 2012-2013 school years, again indicating the same public-
assistance benefit number--and was approved for free-meal benefits. 

* Another household applied for benefits during the 2010-2011 school 
year--providing a public-assistance benefit number--and was approved 
for free-meal benefits. When interviewed, the parent said that the 
household received SNAP benefits. However, when we verified the 
information with the state, officials told us the household was not 
receiving public-assistance benefits at the time of application. 
[Footnote 66] 

* In a 2010-2011 school year application, one household indicated that 
the student was a foster child; however, when we interviewed the 
applicant, she told us that she has never had foster children. This 
household was not eligible for free meals, but may have been eligible 
for reduced-price meals.[Footnote 67] A school-district official told 
us that this household was directly certified for free-meal benefits 
during the 2011-2012 and 2012-2013 school years. 

Because applications that indicate categorical eligibility are 
generally not subject to standard verification, these ineligible 
households would likely not be identified unless they were selected 
for for-cause verification or as part of the administrative review 
process, even though they contained inaccurate information. These 
cases underscore the potential benefits that could be realized by 
verifying beneficiaries with categorical eligibility. We will refer 
these potentially ineligible households to USDA and their school 
district for appropriate action as warranted. Furthermore, the 
administrative review report for one district we reviewed noted that 
categorical-eligibility determinations were not always correct, 
including migrant, homeless, runaway, Head Start, and Even Start 
programs. 

USDA's eligibility manual states that school districts should be aware 
of the characteristics of a valid SNAP or TANF number and are allowed 
to verify this information with the appropriate agency. However, these 
numbers can vary in terms of length. An official from one state told 
us that because the length of SNAP and TANF numbers varies, it is 
difficult to determine whether a number is valid simply by looking at 
it. A household could also provide an old case number--which appears 
valid--and the school district would not know that the household is 
not receiving public-assistance benefits unless the school district 
verifies the information with the appropriate state agency. 

Standards for Internal Control in the Federal Government state that 
control activities should be effective and efficient in enforcing 
program requirements and help in detecting errors and fraud. Since 
applications that indicate categorical eligibility are generally not 
selected for standard verification, there is limited oversight over 
these beneficiaries. 

Individuals with knowledge of the program-eligibility guidelines could 
indicate categorical eligibility to avoid scrutiny, as this would 
prevent the application from ever being verified unless the school-
district official certifying the household had specific knowledge that 
the information was not accurate. Verifying a sample of applications 
that indicate categorical eligibility could assist in identifying 
ineligible households that are receiving benefits and help improve 
program integrity. For example, USDA could have school districts 
select a sample of applications indicating categorical eligibility and 
verify the information with the appropriate agency. With the increase 
in the number of school districts that directly certify SNAP-
participant children, school districts may already have mechanisms to 
match students with SNAP data provided by the state agency. 
Alternatively, USDA could consider having the state agency perform 
this verification as part of its periodic administrative review of the 
school district. 

Conclusions: 

OMB's designation of the school-meals program as a "high-error" 
program with significant estimated improper payments makes it 
important that internal controls and oversight for the school-meals 
programs be strengthened while simultaneously ensuring that students 
who qualify for benefits are not adversely affected. USDA has taken 
steps to strengthen controls and to increase access to eligible 
individuals by working with Congress, school districts, and other 
public-assistance programs to find new ways to provide benefits to 
those requiring assistance. However, the cases we identified in which 
households received school-meals benefits that they were not eligible 
for highlight the deficiencies with current controls and the need for 
additional corrective actions. Evaluating the data collected on 
completed for-cause verifications for the 2013-2014 school year could 
help USDA determine whether specific data on for-cause verification 
outcomes should be reported separately from standard verification 
results and whether additional guidance for conducting for-cause 
verification--including criteria and examples of possible indicators 
of questionable or ineligible applications--would be beneficial. 
Moreover, a cost-effective mechanism to electronically verify 
applicant information with income or other data sources such as public-
sector wage records could help enhance the current verification 
process and strengthen program integrity. While challenges may exist 
in verifying beneficiary income through computer matching, 9 years 
have passed since USDA conducted a pilot to determine the feasibility 
of electronic verification. The cost of the school-meals programs, 
continued high improper payments, and advances in technology support 
the need to revisit the feasibility of conducting computer matching in 
the school-meals programs to enhance current verification efforts. If 
appropriate, developing a legislative proposal to expand the 
statutorily defined verification process to include additional 
independent electronic verification for a sample of all school-meals 
applications could help USDA identify and prevent ineligible 
applicants from participating in the school-meals program. In 
addition, verifying a sample of applications that indicate categorical 
eligibility could assist in identifying ineligible households that are 
receiving benefits and help improve program integrity. 

Recommendations for Executive Action: 

To improve integrity and oversight of the school-meals programs, we 
recommend that the Secretary of Agriculture take the following four 
actions: 

* Evaluate the data collected on for-cause verifications for the 2013-
2014 school year to determine if for-cause verification outcomes 
should be reported separately, and if appropriate, develop and 
disseminate additional guidance for conducting for-cause verification 
that includes criteria for identifying possible indicators of 
questionable or ineligible applications. 

* Develop and assess a pilot program to explore the feasibility of 
computer matching school meal participants with other sources of 
household income, such as state income databases, to identify 
potentially ineligible households--those with income exceeding program-
eligibility thresholds--for verification. 

* If the pilot program shows promise in identifying ineligible 
households, develop a legislative proposal to expand the statutorily-
defined verification process to include this independent electronic 
verification for a sample of all school-meals applications. 

* Explore the feasibility of verifying the eligibility of a sample of 
applications that indicate categorical eligibility for program 
benefits and are thus not subject to standard verification. 

Agency Comments: 

We provided a draft of this report to USDA for its review and comment. 
Written comments from the Administrator for FNS are reprinted in 
appendix VI. In its written comments, FNS indicated that it has long 
recognized the importance of addressing improper payments and program 
integrity problems to meet the mission of its programs and that it 
will carefully consider our specific recommendations as it moves 
forward in its efforts to improve integrity in the school-meals 
programs. The letter also describes several steps FNS is taking to 
strengthen program integrity, many of which are highlighted in this 
report. An e-mail dated May 2, 2014, from the FNS GAO 
Liaison/Coordinator stated that FNS generally agreed with our 
recommendations. FNS also provided technical comments, which we 
incorporated as appropriate. In its technical comments, FNS outlined 
potential challenges in implementing computer matching of school meal 
participants with data on other sources of income, such as challenges 
in working with state data or with incomplete Social Security numbers, 
and the potential costs of verifying income data and following up with 
households. We noted potential challenges in our report and 
acknowledge them in our recommendation that USDA develop and assess a 
pilot program to explore the feasibility of this process. We believe 
that the continued high improper payments rate and advances in 
technology support the need to conduct this pilot and, if it shows 
promise, to develop a legislative proposal to expand its use. 

As agreed with your offices, unless you publicly announce the contents 
of this report earlier, we plan no further distribution until 30 days 
from the report date. At that time, we will send copies of this report 
to interested congressional committees, the Secretary of Agriculture, 
and other interested parties. In addition, the report will be 
available at no charge on the GAO website at [hyperlink, 
http://www.gao.gov]. 

If you or your staff have questions about this report, please contact 
me at (202) 512-6722 or lords@gao.gov. Contact points for our Offices 
of Congressional Relations and Public Affairs may be found on the last 
page of this report. GAO staff who made major contributions to this 
report are listed in appendix VII. 

Signed by: 

Stephen M. Lord: 
Managing Director, Forensic Audits and Investigative Service: 

List of Requesters: 

The Honorable Thomas R. Carper: 
Chairman: 
The Honorable Tom Coburn, M.D. 
Ranking Member: 
Committee on Homeland Security and Governmental Affairs: 
United States Senate: 

The Honorable Claire McCaskill: 
Chairman: 
Subcommittee on Financial and Contracting Oversight: 
Committee on Homeland Security and Governmental Affairs: 
United States Senate: 

The Honorable Debbie Stabenow: 
Chairwoman: 
The Honorable Thad Cochran: 
Ranking Member: 
Committee on Agriculture, Nutrition, and Forestry: 
United States Senate: 

The Honorable John Kline: 
Chairman: 
Committee on Education and the Workforce: 
House of Representatives: 

The Honorable Todd Rokita: 
Chairman: 
Subcommittee on Early Childhood, Elementary, and Secondary Education: 
Committee on Education and the Workforce: 
House of Representatives: 

[End of section] 

Appendix I: Objective, Scope, and Methodology: 

This report assesses (1) what steps, if any, has the U.S. Department 
of Agriculture (USDA) taken to help identify and prevent ineligible 
beneficiaries from receiving benefits in school-meals programs, and 
(2) what opportunities, if any, exist to strengthen USDA's oversight 
of the school-meals programs? We also report case-study examples of 
households that may have improperly received program benefits. Because 
of limited salary and income data available for all U.S. households, 
our case-study examples are limited to civilian executive-branch 
employees and United States Postal Service (USPS) employees.[Footnote 
68] 

To determine any steps USDA has taken to identify and prevent 
ineligible beneficiaries from receiving benefits and what 
opportunities, if any, exist to strengthen USDA's oversight of the 
school-meals programs, we reviewed USDA's program policies and 
regulations, as well as legislation relating to the school-meals 
programs. We also reviewed prior USDA reports and other studies 
concerning the programs. Further, we interviewed officials from USDA's 
Food and Nutrition Service to discuss the agency's administration of 
the programs. We selected the Washington, D.C., and Dallas, Texas 
metropolitan regions--areas with different federal-employee 
concentrations--for further work.[Footnote 69] We conducted interviews 
of officials responsible for administration and oversight of the 
program at the city level in Washington, D.C., and the state level in 
Maryland, Texas, and Virginia, as well as at 25 school districts 
located in these regions. We initially obtained data from 28 school 
districts for our review--14 located in the Dallas, Texas, 
metropolitan region and 14 in the Washington, D.C., metropolitan 
region, which includes Washington, D.C., and its Maryland and Virginia 
suburbs.[Footnote 70] In the Dallas, Texas, metropolitan region we 
selected school districts with student enrollment over 10,000 
students. We selected all school districts in the Washington, D.C., 
metropolitan region, excluding 56 charter school districts in 
Washington, D.C. However, we did not use data from 3 school districts--
1 located in the Dallas, Texas metropolitan region and 2 located in 
the Washington, D.C., metropolitan region--because the data were not 
reliable for our purposes. During the 2010-2011 school year there were 
57 school districts in Washington, D.C.; 49 in Maryland; 1,260 in 
Texas; and 154 in Virginia for a total of 1,520. This selection is not 
representative of all states, school districts, or school-meal 
participants. We assessed controls related to the identification and 
prevention of ineligible beneficiaries in accordance with internal 
control standards.[Footnote 71] 

To further identify opportunities, if any, that exist to strengthen 
USDA's oversight of the school-meals programs we tested controls that 
are designed to identify and prevent ineligible school-meals 
beneficiaries. To do this, we selected a nongeneralizable sample of 48 
households participating in the National School Lunch Program (NSLP) 
for further review and investigation. To select the sample, we matched 
school-meals eligibility data for the 2010-2011 school year from the 
25 school districts to civilian executive-branch federal-employee 
payroll data. The 2010-2011 school year was the last year in which the 
school-meals applications requested that the adult applicant provide 
his or her complete Social Security number. While we do not expect 
federal employees to be any more or less likely to commit fraud than 
employees in other sectors, we completed case-study work based on the 
availability of centralized salary, address, Social Security number, 
and employment data for federal employees--these data were used to 
identify participants in NSLP, regardless of income. The results of 
our work cannot be generalized to all participants because it does not 
include private-sector employees. Additional information comparing 
federal and private-sector employee wages can be found in appendix V. 

We began by examining databases containing students deemed eligible 
for free or reduced-price school meals for the 2010-2011 school year 
from the 25 school districts. These data generally contained 
personally identifiable information for the child and an adult 
household member, as well as household income and size. The data also 
contained information about whether a household was directly certified 
into the program or approved through a household application. We also 
obtained civilian federal-employee payroll data for approximately 2.5 
million individuals from five federal-payroll processors. These data 
contained personally identifiable information for the federal 
employee, as well as wages by pay period for some or all of calendar 
year 2010. We used federal-employee payroll data to develop case 
studies due to the unavailability of other data sources containing 
salary information for nonfederal employees. To assess the reliability 
of the school-meals eligibility and payroll data, we reviewed relevant 
documentation, interviewed knowledgeable agency officials, and 
examined the data for obvious errors and inconsistencies. We concluded 
that the school-meals eligibility data and payroll data were 
sufficiently reliable for purposes of this report. 

Next, we narrowed the civilian federal-employee payroll data to those 
with income during the July 2010 to December 2010 period--to coincide 
with the start of the school year when most school-meals eligibility 
determinations are made. We matched the school district and federal 
payroll data using the Social Security number of the adult household 
member, an address key composed of the address and zip code, and name 
fields, to the extent they were available. Our matches included 
households that, based on income, appeared both eligible and 
ineligible to participate in the school-meals programs. A household 
member earning income does not preclude children in the household from 
being eligible for school-meals benefits. From our matches, we 
generated randomly sorted lists of free and reduced-price school-meals 
participants who submitted an application and randomly sorted lists of 
students who were directly certified for free school meals in each of 
the 25 school districts. 

We then randomly selected up to two households in each of the 25 
school districts for an in-depth review, for a total of 48 cases. 
Specifically, for each of the school districts, we reviewed one 
household that submitted an application that was used for benefit 
determination (25 cases), as well as one household that was directly 
certified (23 cases).[Footnote 72] Two of the 25 school districts did 
not have any directly certified students who matched with the payroll 
data. We applied a minimum threshold of $6,000 to the amount of 
federal salary earned during July 2010 to December 2010 in order to 
identify active employees for our sample. In the event an adult 
applicant was deceased or could not be located, we selected the next 
participant from the randomly sorted list. The specific findings from 
the selected cases cannot be generalized to other, or all, school-
meals beneficiaries or federal-employee households that received 
school-meals benefits. Because our data were limited to federal-
employee households in 25 school districts that were selected on a 
nonrandom basis, the results of our cases cannot be generalized to a 
larger population of school-meals participants or to the entire 
federal workforce. 

Once we identified the sample, we contacted the school districts and 
the states in our sample to obtain supporting documentation. For the 
25 households that submitted a school-meals application, we requested 
and reviewed the available applications from the 2010-2011 school year 
to see what the applicant listed as his or her household income and 
household size. We used the school-meals income-eligibility guidelines 
to determine whether school districts correctly determined eligibility 
based upon the information stated on the application. We also reviewed 
school-meals applications from the 2011-2012 and 2012-2013 school 
years, if submitted. For the 23 directly certified households, we 
obtained and reviewed the public-assistance application associated 
with the household from state agencies in the District of Columbia, 
Maryland, Texas, and Virginia to see what the applicant listed as the 
household income and composition. 

We then reviewed the payroll records of the applicant or other 
household member to obtain information on their actual minimum income 
during the period the application was signed and to determine whether 
the federal employee's income stated on the school-meals or public-
assistance application was accurate. If the applicant's income, along 
with the income of other household members listed on the application, 
exceeded the eligibility guideline based on the number of household 
members stated on the application, we considered these households to 
be potentially ineligible for school-meals benefits.[Footnote 73] 

To conduct our investigative work, we interviewed individuals from the 
48 households in our nongeneralizable sample--23 households that were 
directly certified and 25 that applied for benefits. We interviewed 
these individuals to determine whether the information entered on the 
applications was accurate, to confirm their income, and to determine 
the composition of their households. Investigators also conducted a 
review of the associated payroll records and school-meals application 
or public-assistance application to inform the interviews. 

We conducted this performance audit from February 2012 to May 2014 in 
accordance with generally accepted government auditing 
standards.[Footnote 74] Those standards require that we plan and 
perform the audit to obtain sufficient, appropriate evidence to 
provide a reasonable basis for our findings and conclusions based on 
our audit objectives. We believe that the evidence obtained provides a 
reasonable basis for our findings and conclusions based on our audit 
objectives. We conducted our related investigative work from June 2013 
to May 2014 in accordance with standards prescribed by the Council of 
the Inspectors General on Integrity and Efficiency. 

[End of section] 

Appendix II: Additional Benefits Based on Eligibility in the School-
Meals Programs: 

Schools and households that enroll students in free or reduced-price 
school meals may be eligible for additional federal and state 
benefits. For example, the U.S. Department of Education, through Title 
I of the Elementary and Secondary Education Act, provided 
approximately $13.8 billion in funds to schools with high 
concentrations of low-income families during fiscal year 2013. The 
distribution of federal Title I funds within schools and school 
districts can be based, in part, on the number of students eligible 
for free and reduced-price meals. In addition, separate state funding 
for schools can also be tied to the percentage of students eligible 
for free or reduced-price school meals. Further, the Universal Service 
Program for Schools and Libraries, also known as the E-rate program--
created by the Telecommunications Act of 1996--provides schools with 
discounts on eligible telecommunications services, Internet access, 
and internal connections (such as network wiring).[Footnote 75] 
Discounts can range from 20 to 90 percent, and the primary measure for 
determining the discount is the percentage of students eligible under 
the National School Lunch Program (NSLP) for free or reduced-price 
meals. 

Households can also receive additional benefits by participating in 
the school-meals programs. For example, the Lifeline Program, 
administered by the Universal Service Administration Company on behalf 
of the Federal Communications Commission, provides phone service 
discounts for households that participate in the NSLP or other 
qualifying assistance programs. We reviewed guidance from one school 
district that provides other benefits to households qualifying for 
free or reduced-price meals, including textbook assistance, and 
waiving college application fees and athletic fees. 

[End of section] 

Appendix III: Responsibilities of the Food and Nutrition Service, 
State Agencies, and School Districts: 

[Refer to PDF for image: illustration] 

USDA Food and Nutrition Service (FNS): 
* Has overall responsibility for school-meals programs; 
* Issues guidance to states and school districts. 
* Provides state agencies overseeing the school-meals programs with 
technical assistance; 
* Conducts an evaluation of the state agency's administration of the 
school-meals programs; 
* Reimburses state agencies for meals served in the school districts. 

State Agencies Overseeing School-Meals Programs: 
* Conduct administrative reviews of the school districts; 
* Provide school districts with technical assistance; 
* Secure any needed corrective actions to get school districts in 
compliance with school-meals program requirements; 
* Assess fiscal (financial) action on school districts, if necessary; 
* Reimburse school districts for meals served. 

School Districts: 
* Make eligibility determinations; 
* Conduct verification activities; 
* Provide nutritionally balanced school meals. 

Source: GAO analysis of U.S. Department of Agriculture (USDA) 
information. 

[End of section] 

Appendix IV: Sample School-Meals Household Application: 

[Refer to PDF for image: Sample School-Meals Household Application] 

[End of section] 

Appendix V: Comparison between Federal and Private-Sector Employees: 

According to a January 2012 Congressional Budget Office (CBO) Report, 
in 2010, 1.7 percent of the U.S. workforce was made up of federal 
civilian employees--approximately 2.3 million compared to 111 million 
that were employed by the private sector and 20 million employed by 
state and local governments.[Footnote 76] Another 800,000 are employed 
by government enterprises that typically pay for employee compensation 
through the sale of services rather than through tax revenue, the 
United States Postal Service (USPS) being the largest such employer. 
Further, federal employees can be both part-time and intermittent, 
such as census enumerators, whose jobs last from 2 to 8 weeks. 

According to CBO's study, employees of the federal government have 
varying levels of educational attainment. For example, about 50 
percent of federal employees have either a high-school diploma or less 
or, at most, some college, compared to about 70 percent of the private-
sector workforce. In the federal government, about 50 percent of 
employees have a bachelor's degree or higher education attainment, 
while in the private sector about 30 percent of the workforce has 
comparable education. The wages of federal versus private-sector 
employees vary when considering level of education attained. In 
addition, CBO found that federal civilian workers with no more than a 
high-school education earned about 21 percent more, on average, than 
similar workers in the private sector, while those with some college 
earned 15 percent more, on average than similar workers in the private 
sector. Employees whose highest level of education is a bachelor's 
degree earned roughly the same hourly wage, on average, in the federal 
government as in the private sector. Federal workers with a doctorate 
or professional degree earned 23 percent less per hour, on average 
than similar workers in the private sector. 

While this CBO report does provide a point of comparison between 
civilian federal workers and the private sector, the data analyzed do 
not mirror the federal-employee population used in this report. For 
example, the CBO report does not include workers in government 
enterprises, including USPS, seasonal, or part-time civilian federal 
employees, while our analysis does. Therefore, while the results of 
the CBO report are presented for informational purposes, they should 
not be used to draw conclusions about civilian federal-employee and 
USPS pay used in our analysis. 

[End of section] 

Appendix VI: Comments from the U.S. Department of Agriculture Food and 
Nutrition Service: 

United States Department of Agriculture: 
Food and Nutrition Service: 
3101 Park Center Drive: 
Alexandria, VA 22302-1500: 

April 29, 2014: 

Mr. Stephen M. Lord: 
Managing Director: 
Forensic Audits and Investigative Service: 
U.S. Government Accountability Office: 
441 G Street NW: 
Washington, D.C. 20548: 

Dear Mr. Lord: 

I am writing in response to the report, School-Meals Programs: USDA 
Has Enhanced Controls, but Additional Verification Could Help Ensure 
Legitimate Program Access (GAO-l4-262), provided in draft form to our 
office in April 2014. 

The Department of Agriculture's (USDA's) Food and Nutrition Service 
(FNS) has long recognized that addressing improper payments and other 
program integrity problems are integral to meeting the ongoing mission 
of the school meals and other nutrition assistance programs. Erroneous 
payments draw scarce program resources away from the children who need 
them the most. Equally as important, these programs are ultimately not 
sustainable without continued public confidence that their benefits go 
to those who qualifY for them, are used appropriately, and achieve the 
purposes for which they are intended. 

As you note in your report, Congress structured the school meal 
eligibility determination process, including limits on verification 
activities, to carefully balance access to program benefits for 
eligible children and program integrity. Within the framework of the 
current statutory structure, USDA has taken numerous steps to 
strengthen program integrity. Many of these are supported by the 
Healthy, Hunger-Free Kids Act of201O. Some of our significant efforts 
include: 

* Improving application-based eligibility determinations by clarifying 
school district ability to increase verification of applications "for 
cause," implementing an improved verification data collection process, 
and publishing a final rule in February 2014 to require high-risk 
school districts to conduct a second review of applications for 
accuracy, prior to providing free or reduced price benefits. 

* Supporting improved direct certification through implementation of 
direct certification benchmarks and State continuous improvement plans 
under the Healthy, Hunger-Free Kids Act, and improving accuracy of 
direct certification reporting and rate calculation processes. These 
steps have resulted in a substantial improvement in the rate of 
children directly certified. In School Year 2012-13, 89 percent of 
children participating in the Supplemental Nutrition Assistance 
Program were directly certified for free school meals, up from 69 
percent in School Year 2007-08. 

* Improving oversight of school districts by implementing a three-year 
review cycle (instead of five-year), and revamping the State 
administrative review process to develop a more effective, unified 
process for conducting reviews. The new process streamlines 
administrative reviews by incorporating new risk-based approaches and 
tools, targeting areas of potential risk for noncompliance, and 
utilizing resources on those targeted areas. 

* Providing ongoing funding to support improved oversight and data 
analysis by States, through annual grants to States for administrative 
reviews and training. 

Several additional actions that will support improved program 
integrity are underway, including a proposed rule that provides States 
the authority to impose fines for egregious or ongoing program 
compliance issues. We expect the proposed rule to be published later 
this year. We are also currently working with the Office of Personnel 
Management's Innovation Lab on an initiative to improve the free and 
reduced price application, and will implement changes to the 
application as appropriate. 

Thank you for the opportunity to respond. FNS will carefully consider 
the specific recommendations in your report as we move forward in our 
efforts to improve integrity in the School Meal Programs. 

Sincerely, 

Audrey Rowe: 
Administrator: 

[End of section] 

Appendix VII: GAO Contacts and Staff Acknowledgments: 

GAO Contacts: 

Stephen M. Lord, (202) 512-6722 or lords@gao.gov: 

Staff Acknowledgments: 

In addition to the contact named above, Heather Dunahoo (Assistant 
Director), John W. Cooney, Heather Cowles, Ranya Elias, Erika Lang, 
Kathryn Larin, Maria McMullen, Dan Meyer, Linda Miller, Sandra Moore, 
Robert Ridley, and Daniel Silva made key contributions to this report. 

[End of section] 

Footnotes: 

[1] 42 U.S.C. § 1751 et seq. 

[2] 42 U.S.C. § 1771 et seq. The SBP started as a pilot project in 
1966 and was made permanent in 1975. 

[3] Not all schools that participate in the NSLP participate in the 
SBP. During the 2013-2014 school year, the NSLP operated in over 
100,000 schools and institutions, and the SBP operated in over 89,000 
schools and institutions. 

[4] High-error programs are those programs that reported roughly $750 
million or more in improper payments in a given year, did not report 
an error amount in the current reporting year but previously reported 
an error amount over the threshold, or have not yet established a 
program error rate and have measured components that were above the 
threshold. Office of Management and Budget, Issuance of Part III to 
OMB Circular A-123, Appendix C, Memorandum M-10-13 (Washington, D.C.: 
Mar. 22, 2010). USDA estimates that approximately $996 million of its 
fiscal year 2013 improper payments represents certification errors and 
approximately $778 million represents school district counting and 
claiming errors. The SBP had approximately $831 million in improper 
payments in fiscal year 2013. 

[5] OMB, Issuance of Part III to OMB Circular A-123, Appendix C. 
Improper payment estimates reported by federal agencies are not 
intended to be an estimate of fraud in federal agencies' programs and 
activities. This guidance also instructs agencies to report as 
improper payments any payments for which insufficient or no 
documentation was found. 

[6] GAO, Supplemental Nutrition Assistance Program: Improved Oversight 
of State Eligibility Expansions Needed, [hyperlink, 
http://www.gao.gov/products/GAO-12-670] (Washington, D.C.: July 26, 
2012). 

[7] Pub. L. No. 108-265 § 104 (June 30, 2004). Some states designate 
households as categorically eligible for SNAP if they receive 
Temporary Assistance for Needy Families (TANF) benefits, even if those 
benefits are nonmonetary. Nonmonetary TANF benefits include, but are 
not limited to, an informational brochure, employment assistance, or 
childcare assistance. Households designated as categorically eligible 
for SNAP must go through a separate benefit determination process and 
not all are found eligible for SNAP benefits. Children in households 
that do not receive a SNAP benefit do not automatically qualify for 
free or reduced-price school meals. 

[8] For the purpose of the school-meals programs, the term "school 
food authority" is more-commonly used for purposes of operating the 
school meals program, such as when discussing agreements or 
nutritional requirements. A "local educational agency" is a public 
authority that is recognized in a state as an administrative agency 
for its public elementary or secondary schools. In this report, we use 
the term school district or district to refer to school food 
authorities and local educational agencies. 

[9] GAO, School Meal Programs: Improved Reviews, Federal Guidance, and 
Data Collection Needed to Address Counting and Claiming Errors, 
[hyperlink, http://www.gao.gov/products/GAO-09-814] (Washington, D.C.: 
Sept. 9, 2009). 

[10] Throughout this report, we use the term "federal employees" to 
refer to both civilian executive-branch employees and USPS employees. 

[11] The Washington, D.C.,and Dallas, Texas, metropolitan regions 
ranked 1st and 18th, respectively, among the 50 metropolitan regions 
with the largest number of executive-branch federal employees during 
fiscal year 2012. The Washington, D.C., region includes Washington, 
D.C.; Maryland; and Virginia. 

[12] There were approximately 18,500 school districts in the United 
States during the 2010-2011 school year. 

[13] GAO, Internal Control: Standards for Internal Control in the 
Federal Government, [hyperlink, 
http://www.gao.gov/products/GAO/AIMD-00-21.3.1] (Washington, D.C.: 
November 1999). 

[14] USDA reported that approximately 7.7 million household 
applications were approved for free or reduced-price meals during the 
2010-2011 school year. 

[15] The 2010-2011 school year was the last year in which the school-
meals applications requested that the adult applicant provide his or 
her complete Social Security number. 

[16] Congressional Budget Office, Comparing the Compensation of 
Federal and Private-Sector Employees (Washington, D.C.: January 2012). 

[17] We applied a minimum threshold of $6,000 to the amount of federal 
salary earned during July 2010 to December 2010 in order to identify 
active employees for our nongeneralizable sample. Individuals with 
incomes of less than $6,000 were not part of the sample. 

[18] Two selected school districts did not have any directly certified 
households that matched with federal-employee payroll data. 

[19] The time frame required for our review was a result of various 
factors, including time required to complete data reliability 
assessments and data standardization of multiple databases. 

[20] Income not to be counted in the determination of a household's 
eligibility includes, but is not limited to, the value of benefits 
under SNAP or Food Distribution Program on Indian Reservations 
(FDPIR), student financial assistance benefits, and loans. 78 Fed. 
Reg. 17628 (Mar. 22, 2013). 

[21] The income-eligibility guidelines are in effect from July 1 to 
June 30 of each year. The income-eligibility guideline amounts in 
Alaska and Hawaii are higher. 

[22] Local school districts set their own prices for full-price (paid) 
meals, but must operate their meal services as nonprofit programs. For 
July 1, 2013, through June 30, 2014, the basic cash reimbursement rate 
for lunch ranged from $2.93 to a maximum allowable of $3.16 for free 
meals, $2.53 to $2.76 for reduced-price meals, and $0.28 to $0.42 for 
paid meals. The cash reimbursement rate for breakfast ranged from 
$1.58 to a maximum allowable of $1.89 for free meals, $1.28 to $1.59 
for reduced-price meals, and $0.28 for paid meals. Higher NSLP and SBP 
reimbursement rates are in effect for Alaska and Hawaii. 

[23] 42 U.S.C. § 1758. Categorical eligibility can be conferred 
through either a household application or direct certification. 

[24] A household application is used for all members in the household; 
therefore, an application can contain multiple students. While 24 
million students were deemed eligible for free or reduced-price meals, 
according to USDA officials, not all students who were deemed eligible 
participate. 

[25] 7 C.F.R. § 245.6(c)(4) and USDA Eligibility Manual. In commenting 
on a draft of this report, USDA explained that beginning on July 1, 
2014, school districts that demonstrate a high level of or high risk 
for certification, verification, and other administrative errors will 
be required to ensure that the initial eligibility determination for 
each application is reviewed for accuracy prior to notifying a 
household of its eligibility status 

[26] 42 U.S.C. § 1758(d). 

[27] The Child Nutrition and WIC Reauthorization Act of 2004 (CNR) 
required all states to establish a system of direct certification of 
school-age SNAP participants by the 2008-2009 school year. The 
requirement applies only to children participating in SNAP; however, 
states and school districts may also directly certify children from 
TANF and FDPIR households, as well as children meeting an approved 
designation. 

[28] 42 U.S.C. § 1758(b)(2)(D). 

[29] If there are not an adequate number of error-prone applications 
to draw a 3 percent sample, school districts are to conduct standard 
verification on all error-prone applications and select the remaining 
required applications from all certified applications, including 
applications indicating categorical eligibility. School districts may 
qualify to use one of two alternate sample sizes based upon the 
nonresponse rate for the preceding school year. The sample size under 
alternate one is equal to the lesser of: 3 percent of all applications 
or 3,000 applications selected at random. The sample size under 
alternate two is equal to the lesser of: 1,000 of all applications 
plus the lesser of (a) 500 applications that provide a case number in 
lieu of income information or (b) ½ of 1 percent of applications 
approved that provide a case number in lieu of income information. 
USDA guidance also states that school districts must not verify more 
or less than the standard sample size or alternative sample size. 

[30] Households that indicate categorical eligibility could be subject 
to verification if a school district did not have enough error-prone 
applications to meet the 3 percent sampling requirement and drew 
additional applications from all those certified for standard 
verification. If selected, the validity of the case number stated on 
the application would be verified. 

[31] United States Department of Agriculture, Food and Nutrition 
Service, Eligibility Manual for School Meals: Determining and 
Verifying Eligibility (Washington, D.C.: January 2008). This guidance 
has been included in all subsequent manuals. 

[32] Pub. L. No. 107-300 (Nov. 26, 2002). 

[33] United States Department of Agriculture, Agency Financial Report, 
Fiscal Year 2013 (Washington, D.C.: December 2013). 

[34] USDA commissioned a November 2007 NSLP/SBP Access, Participation, 
Eligibility, and Certification (APEC) Study to estimate certification 
and noncertification errors in the NSLP and SBP during the 2005-2006 
School Year. The study developed a series of models that USDA uses to 
estimate program improper payments annually. 

[35] Pub. L. No. 108-265 (June 30, 2004). 

[36] To qualify for SNAP benefits, households must meet certain income 
and resource tests. In general, a household's gross monthly income 
cannot exceed 130 percent of the federal poverty level. 

[37] States that fail to achieve a direct-certification rate of at 
least 95 percent by the 2013-2014 school year are required to 
implement direct-certification improvement plans. According to a USDA 
report, states cited the inability of direct-certification improvement 
measures to account for children who receive SNAP benefits but who are 
not enrolled in schools and thus not eligible for direct 
certification. These students include home-schooled children, school 
dropouts, and some homeless and migrant children. Another commonly 
cited challenge was incorporating nonpublic schools more efficiently 
into the direct-certification process. U.S. Department of Agriculture, 
Food and Nutrition Service, Office of Policy Support, Direct 
Certification in the National School Lunch Program: State 
Implementation Progress, School Year 2012-2013, Special Nutrition 
Programs Report No. CN-13-DC, prepared by Mathematica Policy Research, 
Inc. (Alexandria, Va.: November 2013). 

[38] In all states, the combined income eligibility limit for Medicaid 
exceeds the NSLP income eligibility limit of 130 percent of the 
federal poverty guideline. 

[39] USDA, Food and Nutrition Service, National School Lunch Program 
and School Breakfast Program: Request for Applications for 
Participation in Demonstration Projects to Evaluate Direct 
Certification with Medicaid (Alexandria, Va.: November 2012). 

[40] 7 C.F.R. § 210.18, Prior to the 2013-2014 school year, these 
reviews were referred to as Coordinated Review Efforts. In commenting 
on a draft of this report, USDA clarified that administrative reviews 
include off-site procedures where state agencies evaluate the school 
district's system for making eligibility determinations including 
direct certification. 

[41] Pub. L. No. 111-296, § 207 (Dec. 13, 2010). 

[42] Additional changes are being made to the administrative review 
process, including focusing the application review process at the 
school district level instead of the school level. 

[43] 7 C.F.R. § 210.18 (n). 

[44] Chicago Board of Education, Office of the Inspector General, 
Annual Report 2012 (Chicago, Ill.: Dec. 28, 2012). 

[45] State of New Jersey Office of the State Controller, Investigative 
Report: Fraudulent School Lunch Program Applications Filed by Public 
Employees (July 17, 2013). 

[46] 7 C.F.R. § 245.6a(c)(7). 

[47] Income to be reported on the application includes: gross earnings 
from work, welfare, child support, alimony, retirement, and disability 
benefits. 

[48] We did not confirm if the applicant received child support 
payments at the time of application. An official from this school 
district told us that this school district does not conduct for-cause 
verification. 

[49] [hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1]. 

[50] As dictated by statute, school districts are required to verify a 
sample of applicants. The sample size is equal to the lesser of 3 
percent of approved applications, selected from error-prone 
applications, or 3,000 error-prone applications unless an alternative 
sample size is used. If there are not an adequate number of error-
prone applications to draw a 3 percent sample, school districts are to 
conduct standard verification on all error-prone applications and 
select the remaining required applications from all certified 
applications, including applications indicating categorical 
eligibility. All applications can be subject to for-cause verification. 

[51] Households with reported incomes more than $1,200 above the 
reduced-price threshold should not be certified for free or reduced-
price meals and therefore not subject to verification. 

[52] Six of the 25 approved household applications we reviewed 
indicated categorical eligibility. The results of these cases are 
discussed later in this report. 

[53] The annual eligibility guideline for a household of five during 
the 2010-2011 school year was $33,527 for free meals and $47,712 for 
reduced-price meals; the error-prone range is $46,512 to $47,712 and 
$32,327 to $34,727. 

[54] This household did not submit a school-meals application for the 
2012-2013 school year. 

[55] None of the applicants we interviewed told us they deliberately 
understated their income to avoid verification. 

[56] United States Department of Agriculture, Agency Financial Report, 
Fiscal Year 2013 (Washington, D.C.: December 2013). 

[57] USDA commissioned a November 2007 NSLP/SBP Access, Participation, 
Eligibility, and Certification (APEC) Study to estimate certification 
and noncertification errors in the NSLP and SBP during the 2005-2006 
school year. The study developed a series of models that USDA uses to 
estimate program improper payments annually. 

[58] According to USDA guidance, school districts are not to conduct 
any actions to verify the information on the application during the 
certification process; they must accept the applications at face value 
and determine eligibility based on the information voluntarily 
disclosed in the application. In 2004, USDA issued the results of a 
pilot study to determine the effects of requiring documentation to 
households applying for benefits and reported that it had the adverse 
effect of limiting access to students eligible to receive school-meals 
benefits. Mathematica Policy Research, Inc., Evaluation of the 
National School Lunch Program Application/Verification Pilot Projects, 
vol. 1: "Impacts on Deterrence, Barriers, and Accuracy" (Princeton, 
N.J.: February 2004). This report was prepared for USDA. 

[59] Total application verification rates may be greater than 3 
percent because school districts are required to verify all 
questionable applications through for-cause verification even if that 
would increase the sample size above the 3 percent maximum. 

[60] School districts are required to follow up with households that 
do not respond to the initial request for verification. Households 
that do not respond to the verification request may reapply--with 
documentation--for school-meals benefits at any time. 

[61] United States Department of Agriculture, Food and Nutrition 
Service, Case Study of National School Lunch Program Verification 
Outcomes in Large Metropolitan School Districts, report no. CN-04-AV3 
(April 2004). Districts in this study were, on average, about 10 times 
larger than the average public school district, both in the number of 
schools they operate and their total student enrollment. Therefore, 
the districts in this study were not selected to be representative of 
all regular public school districts nationwide. 

[62] [hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1]. 

[63] The fraud-prevention framework was developed on the basis of 
findings from a GAO audit of Hurricanes Katrina and Rita relief 
efforts, and the internal control standards from the Standards for 
Internal Control in the Federal Government. For more detail, see GAO, 
Hurricanes Katrina and Rita Disaster Relief: Prevention Is the Key to 
Minimizing Fraud, Waste, and Abuse in Recovery Efforts, [hyperlink, 
http://www.gao.gov/products/GAO-07-418T] (Washington, D.C.: Jan. 29, 
2007); and [hyperlink, http://www.gao.gov/products/GAO/AIMD-00-21.3.1]. 

[64] United States Department of Agriculture, Data Matching in the 
School Lunch Program: 2005, vol. 1: Final Report, report no. CN-06-DM 
(Alexandria, VA.: February 2007). 

[65] One of the six categorically eligible households in our sample 
included a SNAP number for a person sharing the same address as the 
child. Program regulations allow students whose household income 
exceeds the school-meals eligibility guidelines to qualify for free 
school-meals benefits if they share an address with someone receiving 
public-assistance benefits. Under USDA guidance, categorical 
eligibility for free meals is extended to all children in a household 
if any member of a household--whether responsible for the child or not-
-receives SNAP, TANF, or FDPIR. In this instance, the student's 
guardian earned income that exceeded the limits for free or reduced-
price school meals, but was residing with a relative who, as a 
separate household, received public assistance. Because the student 
shared an address with someone receiving SNAP, the student was 
certified for free school meals in accordance with program rules. 

[66] This household submitted an application in September 2010, but 
based on information from the state, public-assistance benefits did 
not start until November 2010. The household may have qualified for 
free meals based on household size and income at the time of 
application; however, this information was not part of the application. 

[67] We cannot conclusively determine whether this household was 
eligible for reduced-price meals because we do not have access to 
nonfederal payroll records and other sources of income. 

[68] Throughout this report, we use the term "federal employees" to 
refer to both civilian executive-branch employees and USPS employees. 

[69] The Washington, D.C and Dallas, Texas metropolitan regions ranked 
1ST and 18TH, respectively, among the 50 metropolitan regions with the 
largest executive-branch federal employees during fiscal year 2012. 
The Washington, D.C. metropolitan region includes Washington, D.C., 
Maryland, and Virginia. 

[70] There were approximately 18,500 school districts in the United 
States during the 2010-2011 school year. 

[71] GAO, Internal Control: Standards for Internal Control in the 
Federal Government, [hyperlink, 
http://www.gao.gov/products/GAO/AIMD-00-21.3.1]. (Washington, D.C.: 
November 1999). 

[72] USDA reported that approximately 7.7 million household 
applications were approved for free or reduced-price meals during the 
2010-2011 school year. 

[73] Because we do not have access to all sources of income--such as 
child support or income earned by non-federal-employee household 
members, we cannot conclusively determine whether a household was 
eligible for school-meals benefits. 

[74] The timeframe required for our review was a result of various 
factors, including time required to complete data reliability 
assessments and data standardization for multiple databases. 

[75] 47 U.S.C. § 254. 

[76] Congressional Budget Office, Comparing the Compensation of 
Federal and Private-Sector Employees, (Washington, D.C.: January 2012). 

[End of section] 

GAO's Mission: 

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

Obtaining Copies of GAO Reports and Testimony: 

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

Order by Phone: 

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

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

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

Connect with GAO: 

Connect with GAO on facebook, flickr, twitter, and YouTube.
Subscribe to our RSS Feeds or E mail Updates. Listen to our Podcasts.
Visit GAO on the web at [hyperlink, http://www.gao.gov]. 

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

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

Congressional Relations: 

Katherine Siggerud, Managing Director, siggerudk@gao.gov: 
(202) 512-4400: 
U.S. Government Accountability Office: 
441 G Street NW, Room 7125: 
Washington, DC 20548. 

Public Affairs: 

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

[End of document]