School Meal Programs: Competitive Foods Are Widely Available and 
Generate Substantial Revenues for Schools (08-AUG-05,		 
GAO-05-563).							 
                                                                 
Recent increases in child obesity have sparked concerns about	 
competitive foods--foods sold to students at school that are not 
part of federally reimbursable school meals. The nutritional	 
value of these foods is largely unregulated, and students can	 
often purchase these foods in addition to or instead of school	 
meals. In our April 2004 report on competitive foods		 
(GAO-04-673), we reported that several states had enacted	 
competitive food policies that were more restrictive than federal
regulations. However, these policies differed widely in the type 
and extent of restrictions. In addition, it was unclear how and  
to what extent states were monitoring compliance with these	 
policies. GAO was also asked to provide a national picture of	 
competitive foods in schools, as well as strategies that	 
districts and schools themselves are taking to limit the	 
availability of less nutritious competitive foods. This report	 
provides information from two nationally representative surveys  
about the prevalence of competitive foods in schools, competitive
foods restrictions and groups involved in their sale, and the	 
amounts and uses of revenue generated from the sale of		 
competitive foods. It also provides information about strategies 
schools have used to limit the availability of less nutritious	 
competitive foods, based on visits to a total of six school	 
districts in California, Connecticut, Mississippi, Missouri, and 
South Carolina. 						 
-------------------------Indexing Terms------------------------- 
REPORTNUM:   GAO-05-563 					        
    ACCNO:   A32277						        
  TITLE:     School Meal Programs: Competitive Foods Are Widely       
Available and Generate Substantial Revenues for Schools 	 
     DATE:   08/08/2005 
  SUBJECT:   Competition					 
	     Data collection					 
	     Elementary schools 				 
	     Food programs for children 			 
	     Food services					 
	     Nutrition research 				 
	     Nutrition surveys					 
	     School districts					 
	     Secondary schools					 
	     California 					 
	     Connecticut					 
	     Mississippi					 
	     Missouri						 
	     National School Breakfast Program			 
	     National School Lunch Program			 
	     South Carolina					 

******************************************************************
** This file contains an ASCII representation of the text of a  **
** GAO Product.                                                 **
**                                                              **
** No attempt has been made to display graphic images, although **
** figure captions are reproduced.  Tables are included, but    **
** may not resemble those in the printed version.               **
**                                                              **
** Please see the PDF (Portable Document Format) file, when     **
** available, for a complete electronic file of the printed     **
** document's contents.                                         **
**                                                              **
******************************************************************
GAO-05-563

United States Government Accountability Office

GAO

                       Report to Congressional Requesters

August 2005

SCHOOL MEAL PROGRAMS

  Competitive Foods Are Widely Available and Generate Substantial Revenues for
                                    Schools

GAO-05-563

[IMG]

August 2005

SCHOOL MEAL PROGRAMS

Competitive Foods Are Widely Available and Generate Substantial Revenues for
Schools

What GAO Found

Nearly 9 out of 10 schools sold competitive foods to students in school
year 2003-2004, and the availability of competitive foods sold in middle
schools and through a la carte lines has increased over the last 5 years.
Schools often sold these foods in or near the cafeteria and during lunch,
and the competitive foods available ranged from nutritious items such as
fruit and milk to less nutritious items such as soda and candy. High and
middle schools were more likely to sell competitive foods than elementary
schools.

Many different people made decisions about competitive food sales, but no
one person commonly had responsibility for all sales in a school. In a
majority of schools, district officials made competitive food policies,
while school food authority directors and principals made decisions about
specific sales. Other groups, such as student clubs and booster groups,
also made competitive food decisions through their direct involvement in
sales.

Many schools, particularly high schools and middle schools, generated
substantial revenues through competitive food sales in 2003-2004.
Specifically, the nearly 30 percent of high schools generating the most
revenue from these sales raised more than $125,000 per school. Food
services, responsible for providing federal school meals, generally spent
the revenue they generated through a la carte sales on food service
operations. Other school groups often used revenues for student
activities.

The six school districts visited all recently took steps to substitute
healthy items for less nutritious competitive foods. In each district,
committed individuals took actions to initiate and lead change while also
involving those affected. However, districts faced several barriers to
change, including opposition due to concerns about revenue losses. In the
districts visited, the effects of changes on revenues were often unclear
because of limited data.

A Majority of Schools Sell Competitive Foods to Students through Vending
Machines

Source: GAO.

                 United States Government Accountability Office

Contents

Letter

Results in Brief

Background

Almost All Schools Sold Competitive Foods in 2003-2004, and Middle School
Availability Has Increased over the Last 5 Years

Many People Made Decisions about Competitive Food Sales, but No One Person
Commonly Had Responsibility over All Sales in a School

Many Schools Raised a Substantial Amount of Revenue through Competitive
Food Sales and Used It to Support Food Service Operations and Student
Activities

School Districts We Visited Substituted Healthy Competitive Foods for Less
Nutritious Items while Overcoming Obstacles to Change, and the Effects on
Revenue Were Unclear

Concluding Observations

Agency Comments

                                       1

                                      3 5

12

21

26

34 45 46

Appendix I Scope and Methodology

Appendix II GAO Contacts and Staff Acknowledgments

Related GAO Products

Tables

Table 1: Select USDA Initiatives That Promote Child Nutrition and Address
the Competitive Food Environment 11 Table 2: Estimated Percentage of
Schools of Different Levels with Each Competitive Food Venue in 2003-2004
14 Table 3: Types of Competitive Foods Often or Always Available through
Each Venue in Schools, by Nutrition Category 17

Table 4: Types of Competitive Foods Often or Always Available through Any
Venue in Schools, by School Level and Nutrition Category 19

Table 5: District, School, and Community Groups Involved in the Process of
Changing Competitive Foods, by District Visited 37

Table 6: Sampling Error Calculations for Questions in Which the

Error Exceeded 15 Percent 51 Table 7: SFA Director Survey: Population and
Sample by School

Level 54 Table 8: Principal Survey: Population and Sample by School Level
54

Figures

Figure 1: Categories and Sources of Competitive Foods in Schools 8

Figure 2: States That Have Made Efforts to Restrict Competitive Foods in
Schools beyond USDA Regulations, as of April 2005 9

Figure 3: Estimated Percentage of Schools Selling Competitive Foods
through Each Type of Venue in 2003-2004 13

Figure 4: Estimated Percentage of Schools with a Written Competitive Food
Policy in 2003-2004 Enacted by Districts and Schools 22

Figure 5: Number of Groups Directly Involved in Competitive Food Sales in
2003-2004, Reported as an Estimated Percentage of Schools with Sales 24

Figure 6: Groups Most Frequently Involved in Various Competitive Food
Venues Commonly Available in High Schools 25

Figure 7: Estimated Percentage of Schools Generating Different Minimum
Amounts of Total Competitive Food Revenue in 2003-2004, by School Level 28

Figure 8: Estimated Percentage of High Schools in Which Food Services
Generated Different Amounts of Revenue through a la Carte Sales in
2003-2004 30

Figure 9: Estimated Percentage of High Schools in Which Other School
Groups Generated Different Amounts of Revenue through Exclusive Beverage
Contracts in 2003-2004 32

Figure 10: Estimated Percentage of Schools Using Competitive Food Revenue,
Excluding Food Service Revenue, for Various Purposes in 2003-2004 33

Abbreviations

CCD Common Core of Data
CDC Centers for Disease Control and Prevention
FMNV foods of minimal nutritional value
FNS Food and Nutrition Service
HHS Department of Health and Human Services
NSLP National School Lunch Program
SFA school food authority
USDA U.S. Department of Agriculture
WIC Special Supplemental Nutrition Program for Women,

Infants, and Children

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. However, 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 Washington, DC 20548

August 8, 2005

The Honorable Tom Harkin
Ranking Member
Committee on Agriculture, Nutrition, and Forestry
United States Senate

The Honorable Patrick J. Leahy
Ranking Member
Subcommittee on Research, Nutrition, and General Legislation
Committee on Agriculture, Nutrition, and Forestry
United States Senate

The Honorable George Miller
Ranking Member
Committee on Education and the Workforce
House of Representatives

The Honorable Lynn Woolsey
Ranking Member
Subcommittee on Education Reform
Committee on Education and the Workforce
House of Representatives

Increasing child obesity rates have recently focused attention on
children's
health and nutrition and have raised concerns about foods available in
schools that compete nutritionally and financially with federally
regulated
school meal programs. The number of children who are overweight has
more than doubled, and the number of adolescents who are overweight
has more than tripled since 1980, according to the U.S. Centers for
Disease
Control and Prevention (CDC). These changes are related, in part, to poor
nutrition. According to U.S. Department of Agriculture (USDA) data, more
than 60 percent of young people eat too much fat and less than 20 percent
of the recommended daily servings of fruits and vegetables. In addition to
having negative health outcomes, children with poor nutrition may have a
harder time concentrating and succeeding in school than other children.
The Surgeon General's 2001 call to action identified schools as one of the
key settings for public health strategies to address child nutrition.
Since
children spend a large portion of their day in school, providing them with
healthful food options throughout the school day can be an important step
toward good child nutrition.

The key school meal programs, the National School Lunch Program and the
School Breakfast Program, provide millions of children with nutritious
meals each school day. USDA's Food and Nutrition Service administers these
programs through local school food authorities (SFA) and subsidizes the
meals served in local schools as long as meals meet certain nutritional
guidelines. However, other foods not provided through these programs,
typically referred to as competitive foods, are often available to
children at school. Competitive food sales can take place at a variety of
venues in schools, including vending machines, school stores, and a la
carte lines in the cafeteria, through which the SFA sells individually
priced food and beverage items. Federal restrictions concerned with the
nutritional value of competitive foods are limited. Specifically, federal
regulations require that one segment of competitive foods, defined as
foods of minimal nutritional value, not be sold to students during the
breakfast and lunch periods in food service areas.

In recent years, federal, state, and local governments have increasingly
focused on the role that competitive foods play in children's diets. In
our April 2004 report on competitive foods,1 we reported that increasing
numbers of state legislatures have enacted and proposed legislation to
restrict the availability of competitive foods in schools. In addition,
school districts and schools themselves are taking steps to limit the
availability of competitive foods.

Because of your interest in further understanding issues related to
competitive foods in schools, you asked us to answer the following
questions: (1) How prevalent is the sale of competitive foods in schools
across the country, and has this prevalence changed over time? (2) Who
makes decisions about competitive food sales in schools? (3) What amount
of revenue is generated from the sale of competitive foods, and for what
purposes is the revenue used? (4) What strategies have schools used to
limit the availability of less nutritious competitive foods, what
obstacles did they face, and how have these strategies affected sales
revenue?

To answer your first three questions, we obtained information through two
Web surveys, one of school principals and the other of district-level SFA
directors. To conduct our surveys, we selected a stratified random sample

1 See GAO, School Meal Programs: Competitive Foods Are Available in Many
Schools; Actions Taken to Restrict Them Differ by State and Locality,
GAO-04-673 (Washington, D.C.: April 23, 2004).

from the 80,000 public schools nationwide that participate in the National
School Lunch Program, which allowed us to provide national estimates based
on school level. The surveys were administered between October 19, 2004,
and February 11, 2005, with 65 percent of principals and 70 percent of SFA
directors responding.2 The surveys asked respondents about conditions in
their schools during specific school years, primarily 2003-2004, and
therefore, all years cited refer to school years. In addition, all
estimates presented from the surveys have margins of error of plus or
minus 15 percent or less, unless otherwise noted. To answer the fourth
question, we conducted site visits to 6 school districts in California
(Oakland), Connecticut (New Haven), Mississippi (McComb), Missouri
(Independence and Fort Osage), and South Carolina (Richland One),
including visits to a total of 10 schools. Our site visit localities were
selected from a group of approximately 100 districts and schools
recognized as making efforts to limit access to less nutritious
competitive foods. The 6 districts visited were also selected because it
appeared that they used different strategies to restrict competitive
foods, and when viewed as a group, they provided variation across
characteristics such as geographic location, district size, and
socioeconomic status. See appendix I for detailed information on our
surveys, sampling strategy, and site visits.

We conducted our work from May 2004 through July 2005 in accordance with
generally accepted government auditing standards.

Almost all schools sold competitive foods to students in school year
20032004, and over the last 5 years, the availability of competitive foods
has increased both in middle schools and in a la carte lines in many
schools. We estimate that nearly 9 out of 10 schools offered competitive
foods through one or more of the following venues in 2003-2004: a la carte
cafeteria lines, vending machines, and school stores. While competitive
foods were commonly sold in schools of all levels, high schools and middle
schools were more likely to sell these foods than elementary schools. For
example, vending machines were available to students in almost all high
schools and middle schools but in less than half of elementary schools.
Schools often sold competitive foods in or near the cafeteria and during
lunchtime, allowing students to purchase these foods

2 These percentages reflect those principals and SFA directors who
actually received the Web surveys. We were unable to contact a subset of
principals and SFA directors selected in our sample of 656 schools. See
appendix I for detailed information on response rates for each survey.

Results in Brief

as their lunch or to supplement their lunch. The competitive foods
available ranged from nutritious items such as fruit and milk to less
nutritious items such as soda and candy, with nutritious foods more
frequently available through a la carte lines than through vending
machines or school stores. Between 1998-1999 and 2003-2004, the
availability of competitive foods increased in middle schools, and the
volume and variety of a la carte foods sold increased in many schools.

Many people, including district and school officials as well as members of
groups involved in sales at schools, made decisions about competitive
foods, but no one person commonly had responsibility for all competitive
food sales at the school level. The decisions ranged from broad policies
about the school nutrition environment to decisions about which foods to
sell at a specific venue or event. According to school principals, an
estimated 60 percent of schools had written policies in place in school
year 2003-2004 that restricted competitive food sales to students, and in
a majority of those schools the policies were set at the district level,
often by superintendents and school boards. Regarding competitive food
sales in schools, district SFA directors were commonly involved in policy
decisions related to a la carte sales, while school principals often had
final approval over other competitive food sales, such as items sold
through vending machines. In addition to SFA directors and school
principals, many other groups such as teachers, student clubs,
parent-teacher associations, and booster groups were involved in selling
competitive foods in schools. These groups therefore often made decisions
concerning the types of food to sell to students and when to make such
food available. The number and variety of groups involved in these sales
typically increased as the school level increased.

Many schools raised a substantial amount of revenue through competitive
food sales in school year 2003-2004 and used this revenue to support food
service operations and student activities. High schools and middle schools
generally raised more revenue from competitive food sales than elementary
schools, reflecting the greater availability of competitive foods in high
and middle schools. According to our survey, the nearly 30 percent of high
schools generating the most revenue from competitive food sales raised
more than $125,000 per school in 2003-2004. Across all competitive food
sales, food services generated more revenue than other school groups,
largely through a la carte sales, and they generally used this revenue to
support overall food service operations. Other school groups commonly used
their revenues to support student activities, and the most frequent uses
were student field trips, school assemblies and programs, and athletic
equipment and facilities.

The six school districts we visited all recently took steps to substitute
healthy competitive foods for less nutritious fare while overcoming
obstacles to these changes, and the effects of these changes on sales
revenues were often unclear because of limited data. Specifically, many of
the schools we visited increased the availability of healthy items,
including low-fat and low-sugar foods and beverages, while they decreased
the availability of less nutritious foods, such as deep-fried French
fries, candy, and soda. Further, although different districts used
different approaches and achieved different outcomes, district and school
officials identified several factors that consistently facilitated change
and several that hindered it. For example, in all of the districts we
visited, motivated individuals took action to initiate and lead the
process of change while obtaining support from those affected in the
district, schools, and community. However, districts noted that they also
faced many barriers to implementing changes, such as opposition due to
concerns about potential revenue losses. Regarding the effect of changes
on sales revenues, none of the districts we visited had clear and reliable
data concerning the impact of competitive food changes on sales revenues.
From the limited data that were available, it appeared that changes had
varied effects on revenues across districts. Related to this, while a few
districts anticipated and planned for the effects of changes on sales
revenues, most had not. Consequently, many officials expressed strong
concerns about potential revenue losses because competitive food revenues
have provided them with a valued source of funding.

Background 	Competitive foods in schools are those foods sold to students
during the school day that are not part of the federal meal programs.
These federal programs, the National School Lunch Program (NSLP) and the
School Breakfast Program, subsidize public school meals and regulate their
nutritional content. Competitive foods, however, are only minimally
regulated at the federal level. They are typically sold a la carte in the
cafeteria, and through vending machines and school stores.

NSLP and School Breakfast Program

The two largest federal school meal programs, the NSLP and the School
Breakfast Program, aim to address problems of hunger, food insecurity, and
poor nutrition by providing nutritious meals to children in schools. The
NSLP, established in 1946, provides nutritionally balanced low-cost or
free lunches in participating schools to more than 28 million children
each school day, as well as reimbursement for snacks served to those
through age 18 in after-school educational and enrichment programs.
Similarly, the School Breakfast Program, permanently established in 1975,
provides free

or reduced price breakfasts to more than 8 million schoolchildren daily.3
At the federal level, these programs are administered by USDA's Food and
Nutrition Service (FNS). As part of its strategic goal to improve the
nation's nutrition and health, the department has laid out plans to
increase access to, and utilization of, these school meal programs.

In fiscal year 2004, the federal government spent over $8 billion on the
NSLP and the School Breakfast Program. FNS provides reimbursement in the
form of cash subsidies and donated commodities based on the number of
lunches and breakfasts served that meet certain federal requirements. The
meals must adhere to the Dietary Guidelines for Americans, which include
limits on total fat and saturated fat and call for diets moderate in
sodium.4 The meals must also meet standards for the recommended daily
allowances of calories, as well as nutrients such as protein, calcium,
iron, and vitamins A and C. Compliance with the standards is determined by
averaging the nutritional content of the meals offered over a school week.
USDA reimburses states, usually through the state departments of
education, which in turn reimburse local SFAs that operate the programs in
one or more schools.

SFAs function as the governing entities responsible for the local
administration of the federal meals programs. They are often, but not
always, responsible for school meals in an entire school district. SFAs
have some flexibility in operating their school meal programs. For
example, they may operate the programs themselves or contract with food
service management companies to perform functions such as planning and
preparing menus and selecting and buying food. All or some food
preparation may occur at on-site school kitchens or at central kitchens,
which then distribute food to satellite schools. In addition, SFAs may
select among different menu-planning approaches to comply with the federal
nutritional requirements.

SFAs receive a significant portion of their funding from federal
reimbursements that are based on the number of meals served to students

3 These data are based on the fiscal year 2004 average daily participation
in the NSLP and the School Breakfast Program, according to the FNS Program
Information Report for December 2004 from USDA.

4 These requirements for the federal meal programs were established by
Congress in 1994 through the passage of the Healthy Meals for Healthy
Americans Act, Pub. L. No. 103-448, S: 106 (1994).

in their schools. In addition, SFAs also receive some funding from states
for program operations, and they may generate revenues by selling
competitive foods or by offering fee-based catering services. Further,
SFAs are permitted to combine costs and revenues for reimbursable meals
and nonreimbursable offerings, such as competitive foods, as long as they
maintain their nonprofit status. Therefore, if revenues from reimbursable
meals are less than the costs of producing these meals, SFAs may use
competitive food revenues to support the cost of reimbursable meals.
Likewise, if revenues from reimbursable meals are more than the costs of
producing these meals, SFAs may use these funds to support competitive
food sales.

Minimal Federal Restriction of Competitive Foods

Competitive foods are those foods sold in schools, during the school day,
that are not part of the federal school meal programs-that is, they
compete with the nutritionally regulated school meal programs. These foods
can range from candy and soda to pizza and popcorn to apples and milk and
are typically available in cafeteria a la carte lines, vending machines,
and school stores.

Unlike federally subsidized school meals, the sale and nutritional content
of competitive foods are largely unregulated by the federal government.
Federal regulations prohibit the sale of certain competitive foods, known
as foods of minimal nutritional value (FMNV),5 during meal periods in
school cafeterias and other food service areas. FMNV, as defined by USDA,
include soda, chewing gum, and hard candy, for example (see fig. 1). Other
than this restriction, federal regulations do not prohibit or limit the
sale of any other competitive foods anywhere on school grounds at any
time.6 In contrast, from 1980 to 1983, federal regulations prohibited the
sale of FMNV anywhere in the school from the beginning of the school day
until the last meal period. In National Soft Drink Ass'n v. Block, 721 F.
2d 1348 (D.C. Cir. 1983), the Court of Appeals for the District of
Columbia overturned this regulation and construed a 1977 amendment to the
Child Nutrition Act as allowing USDA to regulate the sale of competitive
foods only in food service areas during meal periods. Following this
decision,

5 FMNV are defined in regulations for the NSLP (7 C.F.R. S: 210.11) and
listed in appendix B of those regulations. USDA has the authority to
change the definition of FMNV and also has established procedures to amend
the list of these foods.

6 According to regulations, all income from the sale of competitive foods
in the food service area must accrue to the nonprofit food service
provider, the school, or an organization approved by the school.

USDA amended its regulation to limit the prohibition of these foods to
food service areas during meal periods.

Figure 1: Categories and Sources of Competitive Foods in Schools

Source: GAO, clip art source: Art Explosion.

According to federal regulations, states and SFAs may impose further
restrictions on all foods sold at any time throughout their schools. As of
April 2005, 28 states have made efforts to restrict the sale of
competitive foods beyond USDA regulations (see fig. 2). Five state
policies do not restrict particular food items, but instead typically
address the competitive food environment more broadly. For example, some
of these states have created committees to develop policies concerning
competitive foods in schools or have encouraged schools to find ways to
improve their competitive food environments. The remaining 23 of these
state policies place some form of specific restrictions on competitive
foods, though they differ in the type and extent of restrictions. 7 The
majority of these policies

restrict some, but not all, competitive foods and restrict foods only at

7 See GAO-04-673 for more information on the type and extent of
restrictions implemented by state competitive food policies in place as of
March 2004.

Recent Federal Initiatives to Promote Better Nutrition in Schools

The federal government has an interest in improving child nutrition in
order to promote the health and wellness of the nation's children.
Moreover, the current child obesity trend poses public health risks
because of the relationship of obesity to serious illnesses, such as type
2 diabetes and hypertension. These illnesses can result in substantial
longterm costs to society. In response, USDA has recently developed
initiatives to support school efforts to provide a healthy nutrition
environment, including competitive food sales. Beginning in 1995, USDA
introduced the School Meals Initiative for Healthy Children in an effort
to improve the nutritional quality of meals served through the NSLP and
the School Breakfast Program. That same year, in order to assist with
implementation of the School Meals Initiative, USDA launched Team
Nutrition to focus on schools and promote the nutritional health of the
nation's children. Team Nutrition provides schools with nutrition
education materials for children and families, technical assistance
materials for school food services, and materials to build school and
community support for healthy eating and physical activity.9 Since 1995,
USDA has also created additional resources to help schools improve student
nutrition, address competitive foods, and foster long-term health,
sometimes in collaboration with other federal agencies (see table 1).10

9 For more information on federally funded nutrition education programs,
including Team Nutrition, see GAO, Nutrition Education: USDA Provides
Services through Multiple Programs, but Stronger Linkages among Efforts
Are Needed, GAO-04-528 (Washington, D.C.: April 27, 2004).

10 In addition, USDA published its School Nutrition Dietary Assessment
Study II in 2001, providing information on the nutritional quality of
meals served in public schools that participate in the NSLP and the School
Breakfast Program. This study found that students in school year 1998-1999
had access to a variety of breakfast and lunch options other than the
federal meal programs.

 Table 1: Select USDA Initiatives That Promote Child Nutrition and Address the
              Competitive Food Environment Initiative Description

Changing the Scene-Improving the School Nutrition Toolkit that focuses on
improving the school nutrition environment and

Environment (2000) 	serves as a guide to local action, developed with
input from 16 education, nutrition, and health organizations

Fruits and Vegetables Galore (2004) 	Toolkit for school food service
professionals that contains tips on promoting fruits and vegetables to
children

HealthierUS School Challenge (2004) 	Effort to recognize schools that have
met higher standards for nutrition and physical activity than those
required by the federal government, and to encourage other schools to
achieve such results

Making it Happen! School Nutrition Success Stories Report that shares
stories from 32 schools and school districts that have

(2005) 	made innovative changes to improve the nutritional quality of all
foods and beverages sold on school campuses, including competitive foods,
developed in collaboration with CDC and supported by the Department of
Education

Source: GAO.

The Department of Health and Human Services (HHS) has also focused on the
school nutrition environment through various initiatives by CDC. These
include

o  	an eight-component coordinated health model for schools that includes
school nutrition services as one component, and

o  	a School Health Index designed to help schools assess their
environments and improve the effectiveness of their health and safety
policies and programs.11

In addition to these efforts to support a healthy school nutrition
environment, the National Academy of Sciences' Institute of Medicine
recently released a broad-based report on preventing childhood obesity.
Among other things, the institute recommended that the current federal
funding structure of school meals and the policies and practices of
selling competitive foods in schools be examined for improvements that
would encourage students to consume nutritious foods and beverages, and
that nutritional standards be developed and implemented for all
competitive foods sold or served in schools.

11 CDC also reported in its 2000 School Health Policies and Programs Study
that competitive foods were widely available in schools.

On the Horizon: School Wellness Policies and Research on Nutritional
Standards

Almost All Schools Sold Competitive Foods in 2003-2004, and Middle School
Availability Has Increased over the Last 5 Years

The Child Nutrition and WIC Reauthorization Act of 2004 requires school
districts that participate in the federal meal programs to establish local
wellness policies by the first day of the 2006-2007 school year. 12
Congress added this requirement, in part, in order to promote nutrition
and address child obesity by encouraging localities to provide healthy
school environments. These policies must include nutrition guidelines for
all foods available on each school campus during the school day and goals
for nutrition education and physical activity, as well as establish a plan
for measuring implementation of the local wellness policy. Further, the
local wellness policies must be developed in collaboration with the
community-including a combination of school officials, parents, students,
and the public. The act also requires that USDA, HHS-through CDC-and the
Department of Education provide technical assistance to districts
regarding wellness policies. In addition, in order to assist schools in
setting appropriate nutrition standards for foods available in schools,
Congress-through the Conference Report of the Consolidated Appropriations
Act of 2005-provided $1 million to the Institute of Medicine. With these
funds, the institute will conduct a study and provide recommendations
regarding appropriate nutritional standards for the availability, sale,
content, and consumption of all foods at school, with a particular
emphasis on competitive foods.

Nearly 9 out of 10 schools sold competitive foods to students in
2003-2004, and over the last 5 years, the availability of competitive
foods has increased both in middle schools and in a la carte lines in many
schools. While competitive foods were commonly available in all school
levels, students in high schools and middle schools had greater access to
these foods than students in elementary schools. The competitive foods
available ranged from nutritious items such as fruit and milk to less
nutritious items such as soda and candy. Between 1998-1999 and 20032004,
the availability of competitive foods increased in middle schools, and the
volume and variety of a la carte foods sold increased in many schools.

12 WIC is the acronym commonly used to refer to the Special Supplemental
Nutrition Program for Women, Infants, and Children.

Competitive Foods Were We estimate that almost 90 percent of schools sold
competitive foods to Available in Almost All students in 2003-2004 through
one or more of the following venues: a la Schools and Were More carte
lines, vending machines, and school stores.13 Considering each type

of venue individually, a majority of schools sold foods through a la
carteCommon in High Schools lines and vending machines, as shown in figure
3.and Middle Schools

Figure 3: Estimated Percentage of Schools Selling Competitive Foods
through Each Type of Venue in 2003-2004

                          Percentage of schools 100 90

                                       80

                                       75

                                       70

                                       60

                                       50

                                       40

                                       30

                                       20

                                      10 0

teA la car

dingmachines Sc

nVe

                                   hoolstores

                                  Source: GAO.

High schools and middle schools were more likely to sell competitive foods
than elementary schools. Table 2 shows the percentage of elementary,
middle, and high schools selling competitive foods through each type of
venue and through one or more venues.

13 We asked survey respondents questions about "school stores and/or snack
bars." Throughout this report, we will use the term "school stores" to
refer to both school stores and snack bars.

Table 2: Estimated Percentage of Schools of Different Levels with Each
Competitive Food Venue in 2003-2004

                                              Elementary   Middle        High 
                                               schools     schools    schools 
                                  A la carte           67        88 
                            Vending machines           46        87 
                               School stores           15        25 
             One or more of the above venues           83        97 

Source: GAO.

In addition to the competitive food venues regularly available in schools,
students in some schools also were able to purchase competitive foods
through on-campus fund-raisers. For example, more than 4 out of 10 schools
allowed fund-raising-such as seasonal candy sales or short-term sales of
baked goods raising revenues for school organizations-through the sale of
foods to students during the school day in 2003-2004. Such fund-raisers
were permitted in two-thirds of high schools and less than 40 percent of
middle and elementary schools.

While federal regulations restrict access to FMNV-a subset of competitive
foods-in food service areas during meal periods, many types of competitive
foods are allowed to be sold in these locations at meal times. According
to our survey, competitive foods sold in 2003-2004 through a la carte
lines, vending machines, and school stores were frequently available for
purchase in or near school cafeterias and during lunch.14

o  	A la carte items were available to students in the cafeterias of
schools that offered them and were available to students during lunch in
94 percent of those schools.

o  	One-half of schools with vending machines had machines in or near the
cafeteria, and one-third of schools with vending machines had machines
that were available to students during lunch.

14 A la carte foods, vending machines, and school stores were also
available in some schools during other periods of the school day. In
addition, vending machines and school stores were available in other
locations in some schools, such as outside school buildings.

o  	Nearly half of schools with stores had such stores in or near the
cafeteria, and about one-third of schools with stores sold competitive
foods through these stores during lunch.

Although schools that sold competitive foods through a la carte lines or
school stores often sold these foods in just one physical location,
schools with vending machines typically had multiple machines available
throughout the school, ranging from 1 to 25 machines. For example, the
quarter of high schools with the most vending machines had 10 or more
machines, the top quarter of middle schools had 7 or more machines, and
the top quarter of elementary schools had 3 or more machines.15 Schools
generally had more beverage vending machines than snack vending machines.

Further, in many schools, particularly high schools, beverages sold in
vending machines or elsewhere in the school were provided through an
exclusive beverage contract-a contract granting a company exclusive rights
to sell beverages to students in that school. In addition to covering
vending machine sales, these contracts may require schools to provide
beverages through the contracted company in other venues, such as school
stores or athletic event concessions. Nearly half of all schools in
2003-2004 had an exclusive beverage contract. In over a third of schools
with exclusive beverage contracts, the contracts covered 5 years or more,
with some covering at least 10 years.16 Nearly 75 percent of high schools,
65 percent of middle schools, and 30 percent of elementary schools had
exclusive beverage contracts.

Types of Competitive Competitive foods available through a la carte lines,
vending machines, Foods Ranged from and school stores ranged from
nutritious items, such as vegetables and Nutritious to Less salad, to less
nutritious items, such as soda and candy. Nutritious foods

were more frequently available through a la carte lines than
throughNutritious, with High and vending machines and school stores. For
example, as shown in table 3, weMiddle Schools Selling a estimate that
many of the types of foods commonly available through a la Wider Variety
of Items carte sales were nutritious foods and beverages, such as fruit
and milk.

15 The elementary school estimate has a margin of error that exceeds plus
or minus 15 percent. See table 6 in appendix I for more information.

16 While contracts could be negotiated by the school district, the school,
the school food service, or a combination of those groups, over half of
schools with exclusive beverage contracts had a contract that was
negotiated with the school district.

However, types of less nutritious items, such as sweet baked goods and
salty snacks not low in fat, were also available through a la carte lines
in at least one-third of schools. Furthermore, many of the types of foods
commonly available through vending machines and school stores were less
nutritious ones, such as soda and salty snacks.17

17 While the federal government prohibits the sale of soda and certain
candy in cafeterias and food service areas during mealtimes, these foods
can be sold in other locations and during other periods of the school day.
Other less nutritious items, such as sweet baked goods and salty snacks,
can be sold in any school location during any period of the day.

Table 3: Types of Competitive Foods Often or Always Available through Each
Venue in Schools, by Nutrition Category

                                    A la carte Vending machines School stores 
                              Water                             
                   Milk, 1% or skim                             
                  Milk, whole or 2%                             
                         100% juice                             
                              Fruit                             
            Vegetables and/or salad                             
                             Yogurt                             
               Less than 100% juice                             
                      Sports drinks                             
               Low-fat salty snacks                             
          Low-fat sweet baked goods                             
            Low-fat frozen desserts                             
                         Sandwiches                             
                              Pizza                             
                   Fried vegetables                             
               Frozen desserts (not                             
                           low-fat)                             
         Salty snacks (not low-fat)                             
         Sweet baked goods (not                                 
         low-fat)                                               
                              Candy                             
                               Soda                             

Nutritious

Neither clearly nutritious nor less nutritious

Less nutritious

Item is estimated to be available in approximately half or more schools
with the venue

Item is estimated to be available in approximately one-third or more
schools with the venue

Source: GAO.

Note: The nutrition categories, as signified by the shading, are general
descriptions of the foods in each category. GAO created these nutrition
categories to generally reflect the Dietary Guidelines for Americans,
recognizing that they apply to many but not all foods of each
type-nutritional content can vary depending on the ingredients and the
methods used to prepare foods. Four of the estimates in this figure have
margins of error that exceed plus or minus 15 percent. See table 6 in
appendix I for more information.

In addition, our data suggest that students may have had increased access
to more types of competitive foods as they progressed from elementary
school to middle school and high school. While nutritious foods were
commonly available in schools of each level, students in high schools and
middle schools had access to a greater variety of types of less nutritious
foods than students in elementary schools. For example, salty snacks,
sweet baked goods, soda, and candy were available in at least one-third of
high schools and middle schools with competitive foods but in less than
one-third of such elementary schools, as shown in table 4.18

18 However, less nutritious foods were available in some elementary
schools. For example, frozen desserts not low in fat were available in
nearly a quarter of elementary schools with competitive foods.

Table 4: Types of Competitive Foods Often or Always Available through Any
Venue in Schools, by School Level and Nutrition Category

                                                         Middle  High schools 
                                      Elementary schools schools 
                                Water                            
                     Milk, 1% or skim                            
                    Milk, whole or 2%                            
                           100% juice                            
                                Fruit                            
              Vegetables and/or salad                            
                               Yogurt                            
                 Less than 100% juice                            
                        Sports drinks                            
                 Low-fat salty snacks                            
            Low-fat sweet baked goods                            
              Low-fat frozen desserts                            
                           Sandwiches                            
                                Pizza                            
                     Fried vegetables                            
                 Frozen desserts (not                            
                             low-fat)                            
           Salty snacks (not low-fat)                            
         Sweet baked goods (not                                  
         low-fat)                                                
                                Candy                            
                                 Soda                            

Nutritious

Neither clearly nutritious nor less nutritious

Less nutritious

Item is estimated to be available in approximately half or more schools
with any venue

Item is estimated to be available in approximately one-third or more
schools with any venue

Source: GAO.

Note: The nutrition categories, as signified by the shading, are general
descriptions of the foods in each category. GAO created these nutrition
categories to generally reflect the Dietary Guidelines for Americans,
recognizing that they apply to many but not all foods of each
type-nutritional content can vary depending on the ingredients and the
methods used to prepare foods.

Over the Last 5 Years, the Availability of Competitive Foods Sold in
Middle Schools and through a la Carte Lines Increased

According to our survey, the availability of competitive food venues in
middle schools increased during the period between 1998-1999 and 20032004.

o  	The percentage of middle schools offering competitive foods through a
la carte lines, vending machines, or school stores increased from 83 to 97
percent during this time period.19

o  	The percentage of middle schools with exclusive beverage contracts
increased between 1998-1999 and 2003-2004.20

o  	The number of vending machines per school increased between 19981999
and 2003-2004 in more than one-third of middle schools that had vending
machines.21 In addition, the number of vending machines per

school increased in more than half of high schools that had vending

machines.

In addition, the availability of a la carte items, particularly the volume
sold and the variety available for purchase, increased between 1998-1999
and 2003-2004 in many schools.

o  	The volume of a la carte items sold-that is, the overall amount of all
a la carte items sold-increased in more than two-thirds of high schools,
more than half of middle schools, and nearly one-third of elementary
schools that had a la carte sales.22

o  	The variety-that is, the number of different types-of a la carte items
available to students increased in about one-half of all schools that had
a la carte sales.

19 For this analysis, we compared the percentage of middle schools that
had any competitive food venues between school years 1998-1999 and
2003-2004 with the percentage that had any venues in 2003-2004.

20 According to our survey, the percentage of middle schools with
exclusive beverage contracts increased to 65 percent in 2003-2004 from 26
percent in 1998-1999. An additional 31 percent of middle school principals
were unsure if their school had an exclusive beverage contract in
1998-1999.

21 This estimate has a margin of error that exceeds plus or minus 15
percent. See table 6 in appendix I for more information.

22 The elementary school estimate has a margin of error that exceeds plus
or minus 15 percent. See table 6 in appendix I for more information.

According to SFA directors, reasons for the increases in a la carte volume
and variety between 1998-1999 and 2003-2004 included responding to student
demand, providing more nutritious foods, making foods more appealing to
students, and generating additional revenue for the food service.

Aside from increases in the availability of competitive foods sold in
middle school and through a la carte lines, we did not find that the
availability of competitive foods in schools changed considerably during
the period between 1998-1999 and 2003-2004. For example, according to our
survey, there have not been considerable changes in the percentage of high
schools or elementary schools offering competitive foods through a la
carte lines, vending machines, or school stores during this time period.
Further, in a majority of all schools with school stores or vending
machines, the number of different types of food offered for sale through
these venues stayed the same.

Many people, including district and school officials as well as members of
groups selling foods in schools, made decisions about competitive food
sales, but no one person consistently had responsibility for all
competitive food sales at the school level. The decisions ranged from
broad policies about the school nutrition environment to decisions about
which foods to sell at a specific venue or event. In 2003-2004, a majority
of schools had policies in place that restricted competitive food sales to
students, and these policies were often set at the district level by
superintendents and school boards. Regarding the actual selection and sale
of competitive foods in schools, SFA directors were commonly involved in
decisions related to a la carte sales, while principals often had final
approval over other competitive food sales. In addition, many different
groups were directly involved in selling competitive foods in schools in
2003-2004, and these groups could make decisions about which foods to sell
and when to make them available. The number and variety of groups involved
in these sales typically increased as the school level increased.

Many People Made Decisions about Competitive Food Sales, but No One Person
Commonly Had Responsibility over All Sales in a School

District and School According to principals, an estimated 60 percent of
schools had written Officials Made Decisions policies in place that
restricted competitive foods accessible to students in about Competitive
Food 2003-2004, and most often, districts enacted those policies (see fig.
4).23 In Policies contrast, 40 percent of schools had no such policies.

Figure 4: Estimated Percentage of Schools with a Written Competitive Food
Policy in 2003-2004 Enacted by Districts and Schools

2%

Policy enacted by school

10%

Policy enacted by district and school

10%

Do not know who enacted policy

78%

Policy enacted by district

                                  Source: GAO.

In addition to superintendents and school boards that were involved in
making these competitive food policy decisions for a district, both SFA
directors and school principals commonly made policy decisions about
actual food sales at the school level, resulting in no one person having
responsibility over all sales. Specifically, district SFA directors often
made ongoing decisions about policies affecting the school nutrition
environment. For example, SFA directors provided many of the foods
available to students through their administration of the federal meal
programs and typically decided which foods to serve through school a la
carte sales. In addition, SFA directors were also often concerned with

23 We did not collect information on the type and extent of restrictions
placed on competitive foods by these policies or on the enforcement of
these policies.

other competitive food sales in the schools, such as those through vending
machines, school stores, and fund-raising sales. According to our survey,
84 percent of SFA directors in 2003-2004 considered addressing the
competitive food environment in schools to be part of their
responsibilities. More than three-quarters of those directors considered
it a priority.

Moreover, principals also made decisions about competitive food policies
in their schools. For example, as shown in figure 4, aside from the more
than three-quarters of schools with competitive food policies developed by
their districts, school principals enacted policies in conjunction with
their districts in an additional 10 percent of schools and enacted their
own policies in another 2 percent of schools with policies. Regarding
operational decisions, principals in more than half of schools with
competitive food sales reported in our survey that they provided final
approval over the foods and beverages sold through vending machines,
school stores, and fund-raisers in their schools. Also, similar to SFA
directors, principals in a majority of schools reported that they
considered addressing the competitive food environment one of their
responsibilities.

Many Different Groups Were Directly Involved in Deciding What to Sell and
Selling Competitive Foods

In addition to the district and school officials involved in decisions
related to competitive food policy, myriad individuals and groups were
directly involved in the sale of competitive foods. These groups could
make decisions about which specific foods to sell to students and when to
conduct sales. During 2003-2004, groups such as students, parent-teacher
associations, and booster groups-in addition to SFA directors and school
principals-were most commonly involved in sales, according to our survey.

The number of groups involved in sales typically increased as the school
level increased. For example, three or more different groups were much
more likely to be involved in competitive food sales in high schools than
in middle and elementary schools in 2003-2004 (see fig. 5).

Figure 5: Number of Groups Directly Involved in Competitive Food Sales in
20032004, Reported as an Estimated Percentage of Schools with Sales

Percentage of schools

100

90

                                       83

80

70

60

50

40

30

20

10

0

Elementary Middle schools High schools schools

One group

Two groups

Three or more groups

Source: GAO.

Note: The estimates for elementary schools and one estimate for middle
schools in this figure have margins of error that exceed plus or minus 15
percent. See table 6 in appendix I for more information.

The groups directly involved in sales varied by school level. In
elementary schools, the SFA/school food service and student
associations/clubs were commonly involved in sales.24 In middle schools,
in addition to these groups, school officials/administrators were most
commonly involved.25 In high schools, where the greatest number of
competitive food venues was typically available, these three groups and a
variety of others were directly involved in sales. Consequently, within a
high school that has a number of competitive food sales occurring
simultaneously, a student at lunchtime

24 The estimates for SFA/school food service and student
associations/clubs have margins of error that exceed plus or minus 15
percent. See table 6 in appendix I for more information.

25 The estimate for school officials/administrators has a margin of error
that exceeds plus or minus 15 percent. See table 6 in appendix I for more
information.

could be faced with many different food options sold by a variety of
different groups through several venues (see fig. 6).

Figure 6: Groups Most Frequently Involved in Various Competitive Food
Venues Commonly Available in High Schools

Source: GAO analysis.

Note: The competitive food venues shown were estimated to be available in
a majority of high schools, according to our survey. The groups listed
with these venues were estimated to be directly involved in competitive
food sales through the specified venue in at least 25 percent of high
schools.

Many Schools Raised a Substantial Amount of Revenue through Competitive
Food Sales and Used It to Support Food Service Operations and Student
Activities

Many schools generated substantial revenue through competitive food sales
in 2003-2004, often using this revenue to support food service operations
and student activities. High schools and middle schools generally raised
more revenue from competitive food sales than elementary schools,
reflecting the greater availability of competitive foods in high and
middle schools. Across all competitive food sales, food services generated
more revenue than other school groups, largely through a la carte sales.
Some food service directors said they relied on this revenue to support
overall food service operations, while other school groups primarily used
their competitive food revenues to fund student activities.

Total Competitive Food Revenue Varied by School Level From More than
$125,000 in Some High Schools to More than $5,000 in Some Elementary
Schools

Many schools generated a substantial amount of revenue through competitive
food sales in 2003-2004.26 Total revenue generated through competitive
food venues varied by school level, reflecting, among other things, the
greater availability of competitive foods in high schools and middle
schools than in elementary schools.27 In particular, we estimate that
about 30 percent of all high schools generated more than $125,000 per
school through competitive food sales in 2003-2004, while about 30 percent
of all elementary schools generated more than $5,000 per school through
these sales (see fig. 7). These estimates of total competitive food
revenue are conservative, as they are based on the sum of the minimums of
specified revenue ranges.28 Therefore, many schools likely generated more
total revenue from competitive food sales than our analysis reflects.

26 Throughout this report, revenue for each type of competitive food venue
includes all revenue generated through competitive food sales. We did not
ask survey respondents for information on profits retained after covering
expenses.

27 While the number of students in each school likely affects the amount
of revenue generated through competitive foods, our data do not allow us
to determine the effect of school size on revenue.

28 Total revenue reflects the combined minimums of revenue ranges reported
by schools for a la carte lines, vending machines, school stores, and
exclusive beverage contracts. To conduct this analysis, we used matched
survey responses, which combined the principal and SFA director's
responses for each specific school. We defined the minimum for each venue
as the lower bound of the revenue range selected by the respondent, and we
then summed the minimum revenues across all venues for each school. See
appendix I for a description of this analysis.

Figure 7: Estimated Percentage of Schools Generating Different Minimum
Amounts of Total Competitive Food Revenue in 2003-2004, by School Level

Elementary schools

                    Minimum revenue per school (in dollars)

Over 125,000 100,001 -125,000 75,001-100,000 50,001-75,000

25,001-50,000 5,001-25,000 1,001-5,000 0-1,000

                                       52

0 10 203040 50

                             Percentage of schools

Middle schools

                    Minimum revenue per school (in dollars)

Over 125,000 100,001-125,000 75,001-100,000 50,001-75,000

25,001-50,000 5,001-25,000 1,001-5,000 0-1,000

                                     21 21

0 10 203040 50

                             Percentage of schools

High schools

Minimum revenue per school (in dollars) Over 125,000 29 100,001-125,000
75,001-100,000

                 50,001-75,000 25,001-50,000 5,001-25,000 1,001-5,000 0-1,000

0 10 20 30 40 5060 Percentage of schools

Source: GAO.

Note: See appendix I for more information on this analysis, which used
data obtained from the matched responses. One of the estimates in the high
school figure has a margin of error that exceeds plus or minus 15 percent.
See table 6 in appendix I for more information.

Food Services Generated More Revenue through These Sales than Other School
Groups, with Food Services Using Revenue for Their Operations and Other
School Groups Using Revenue for Student Activities

Food Services

Across all competitive food sales, food services generated more revenue
than other school groups, such as school administrators, student
associations, and booster groups.29 Specifically, food services generated
a greater amount of revenue through a la carte sales than through any
other type of competitive food sale. Other school groups raised a greater
amount of revenue through exclusive beverage contracts than through any
other type of competitive food sale.30 In addition to raising varying
amounts of competitive food revenues through different types of sales,
food services and other school groups generally used their revenues for
different purposes.

The revenue food services generated through a la carte sales was
substantial in many schools. For example, we estimate that 40 percent of
high school food services and nearly a quarter of middle school food
services selling competitive foods through a la carte lines generated more
than $50,000 per school through these sales in 2003-2004. Furthermore,
food services in 20 percent of high schools selling a la carte items
generated more than $125,000 per school through a la carte sales, as shown
in figure 8.

29 For the purposes of this discussion, revenue generated by food services
refers to revenue that was raised by both the school and district food
services.

30 Many schools also held fund-raisers to generate revenue for activities
and programs, but this revenue is difficult to measure because of the
involvement of numerous groups, and it is not included in our analysis.

Figure 8: Estimated Percentage of High Schools in Which Food Services
Generated Different Amounts of Revenue through a la Carte Sales in
2003-2004

                        Revenue per school (in dollars)

Over 125,000

100,001-125,000 75,001-100,000 50,001-75,000

25,001-50,000 5,001-25,000 1,001-5,000 0-1,000

                                       26

                             Percentage of schools

Source: GAO.

Note: Estimated percentages do not add to 100 because some SFA directors
were unsure how much revenue they generated through a la carte sales.

Food services typically used their substantial a la carte revenue to
support overall food service operations, supplementing revenue earned
through the sale of school meals. According to our survey of SFA
directors, in 2003-2004, food services in 40 percent of schools earned
less revenue than they spent, generating a loss, and food services in an
additional 20 percent of schools broke even. Food services in the
remaining 40 percent of schools generated more revenue than they spent in
2003-2004, yielding a gain.31 From the data we collected, it is not clear
what proportion of this

food service revenue came from federal reimbursement for meals served and
what proportion came from revenue generated through competitive food
sales.32 In our previous work, we found that food services in some

states had a small but increasing shortfall in total revenue compared with
expenses between school years 1996-1997 and 2000-2001 and that a la

31 In addition, food services in 20 percent of schools received
supplemental funds from the district or school in 2003-2004, while food
services in 21 percent of schools transferred funds to the district or
school in that year.

32 Further, because we did not collect information on total school food
service revenue, we were not able to compare food services' competitive
food revenue with their total revenue in 2003-2004 in order to determine
the effect of competitive food revenue on food service budgets.

                              Other School Groups

carte sales had become an increasingly important source for augmenting
total food service revenue.33 Moreover, some SFA directors told us in

survey comments for this study that they sold competitive foods in order
to maintain balanced budgets. For example, one food service director
commented that the food service would not be able to maintain a balanced
budget without the substantial revenue generated through a la carte sales.
Another commented that the district food service generated nearly half of
its revenue through competitive food sales.

Other school groups raised more revenue through exclusive beverage
contracts than through any other type of competitive food sales. In
particular, we estimate that school groups in nearly one-quarter of high
schools with exclusive beverage contracts generated more than $15,000 per
school through these contracts in 2003-2004, as shown in figure 9.
However, it is important to note that 15 percent of high school principals
did not know how much revenue exclusive beverage contracts generated for
their schools in 2003-2004.

33 GAO, School Meal Programs: Revenue and Expense Information from
Selected States, GAO-03-569 (Washington, D.C.: May 9, 2003). This report
analyzed revenue and expense data from six selected states.

Figure 9: Estimated Percentage of High Schools in Which Other School
Groups Generated Different Amounts of Revenue through Exclusive Beverage
Contracts in 2003-2004

Revenue per school (in dollars)

Over 15,000

10,001-15,000

5,001-10,000 23

1,001-5,000

0-1,000

0 5 10 1520 25

Percentage of schools

Source: GAO.

Note: Estimated percentages do not add to 100 because some school
principals were unsure how much revenue other school groups generated
through exclusive beverage contracts.

Exclusive beverage contracts also provided some schools with noncash
benefits-goods and services such as athletic scoreboards and in-kind
support of school events. In particular, nearly 30 percent of schools of
all levels selling competitive foods through exclusive beverage contracts
received noncash benefits. Nearly one-third of those schools received
athletic equipment, facilities, or uniforms, and a small number of schools
also received support for assemblies and programs, scholarships, and
personal items for students and school staff, such as cups and coolers.
The value of those items was sometimes considerable: in one-quarter of
those schools receiving noncash benefits through exclusive beverage
contracts in 2003-2004, the benefits were worth more than $5,000 per
school.34

School groups other than food services most commonly used their
competitive food revenues to support student activities such as field
trips and assemblies, as shown in figure 10. Similarly, groups in many
schools

34 This estimate has a margin of error that exceeds plus or minus 15
percent. See table 6 in appendix I for more information.

spent competitive food revenues on athletic equipment, facilities, or
uniforms. However, some used revenues to meet school needs such as
expenses associated with general school overhead or with textbooks and
school supplies.

Figure 10: Estimated Percentage of Schools Using Competitive Food Revenue,
Excluding Food Service Revenue, for Various Purposes in 2003-2004

Revenue uses

Student field trips 36

School assemblies and related programs

Athletic equipment, facilities, or uniforms

     General school overheadTextbooks and school suppliesLibrary suppliesArts
                 programs excluding band or chorusBand or choral equipment or
                  uniformsComputers and information technology equipmentStaff
                                              developmentPlayground equipment

Scholarships

Transportation

Nutrition education

Construction of new facilities, except athletic

Staff salaries and/or benefits

010 20 3040

Percentage of schools Source: GAO.

Note: General school overhead includes facilities and grounds maintenance.
Respondents in 22 percent of schools said they spent competitive food
revenues on other uses, such as student rewards and incentives.

School Districts We Visited Substituted Healthy Competitive Foods for Less
Nutritious Items while Overcoming Obstacles to Change, and the Effects on
Revenue Were Unclear

The six school districts we visited all recently took steps to substitute
healthy competitive foods for less nutritious items while overcoming
several obstacles to change, and in the end, the effects of these changes
on revenue were unclear. Although the districts we visited increased the
availability of healthy competitive foods and decreased less nutritious
items through differing approaches, perseverant and committed individuals
took actions in each district to initiate and lead the process of change
while also taking steps to involve and obtain support from those affected.
At the same time, officials noted that they faced several barriers to
making changes, including concerns about potential revenue losses, among
others. Concerning the effects of changes on sales revenues, none of the
districts we visited had sufficient data to examine these effects, and few
had planned for these effects before implementing changes. Regardless of
the limited data on revenue, many officials expressed strong concerns
about potential revenue losses largely because competitive food sales have
provided a source of flexible funding used for a wide variety of purposes.

School Districts We Visited Substituted Healthy Competitive Foods for Less
Nutritious Items

Though the six school districts we visited varied in terms of
socioeconomic status, student population size, and geographic location,
they were all able to take steps to restrict the availability of less
nutritious competitive foods in their schools. Further, rather than just
remove less nutritious competitive foods from their schools, these
districts continued to offer competitive foods to students by substituting
healthy fare for the items removed. In addition, these districts all took
steps beyond current federal and state competitive food regulations, and
while there were differences in the details of changes, most districts'
changes had similar characteristics. Specifically, most, if not all, of
the districts we visited

o  	made changes to competitive foods in all of the schools in their
districts, rather than changes only to certain schools, such as
elementary;35

o  	made changes to the availability of competitive foods throughout the
entire school day, rather than limiting availability only during certain
hours of the day;

35 We selected Fort Osage School District as one of our site visits
because of the changes made to competitive foods at Fire Prairie Middle
School, and all references to Fort Osage in this report reflect only the
changes made at Fire Prairie.

o  	made changes to the availability of competitive foods in each of the
venues through which they were sold, including fund-raising;36and,

o  	made simultaneous changes to foods served through school meal
programs.

Regarding districts' efforts to increase the availability of healthy
competitive foods, as they defined them, all of the districts we visited
recently took steps to make water and juice more available to students in
their schools. For example, districts and schools often replaced soda in
their vending machines with bottled water and juices with higher
concentrations of real fruit juice. Further, several districts also
replaced fried potato chips with baked potato chips, and packaged desserts
with granola bars or similar items. Several districts also increased the
variety of flavored milk available in schools to encourage milk
consumption.

All of the districts we visited also took steps to restrict less
nutritious items. The specific types of competitive foods restricted
varied by district, with two of the districts limiting the availability of
primarily soda and candy, and four districts limiting the availability of
competitive foods high in sugar and fat.37 In these districts, different
criteria were used by each district to define foods high in sugar and fat.
For example, in Fort Osage, all competitive foods must be low-fat (no
specified percentage) and cannot have sugar listed as the first
ingredient, while in Richland One, all competitive foods must contain less
than 40 percent sugar (or other sweeteners) and less than 8 grams of fat
per 1 ounce serving. In the other two districts, New Haven and
Independence, formulating specific nutrition criteria has been an ongoing
process.

In addition to making changes to the types of competitive foods sold in
schools, all of the districts we visited also made similar changes to
increase healthy foods available through school meals. For example, SFAs
in both Independence and Oakland recently removed deep fryers from school
cafeterias, and they now bake all foods. In Richland One, the SFA decided
to implement stricter standards for school meals than those

36 The only exception to this was Independence, where districtwide changes
had been made to only those competitive foods sold by the SFA through a la
carte lines and vending machines at the time of our visit.

37 Independence, one of the four districts restricting competitive foods
high in sugar and fat, restricted many, but not all, of these foods.
Further, as noted earlier, Independence made changes to only those
competitive foods sold by the SFA.

required by the federal government. Specifically, Richland One requires
school meal menus to meet the federal requirements for nutrients, fat, and
sugar intake on a daily basis, rather than averaging the nutritional value
of meal components over the course of a week. In addition, several SFA
directors stated that they would like to introduce a greater variety of
fresh fruits and vegetables into school meal menus. However, they also
expressed their opinion that the federal school meal reimbursement is
insufficient to provide these items more frequently.38

Individuals Who Initiated and Led Change Also Obtained the Support of
Multiple Groups

While the characteristics of the six districts we visited differed, as
well as the process of change in each, districts typically noted several
key components to their success. These key components included an
enthusiastic initiator of change, leadership by dynamic and committed
individuals, and support from groups directly affected by changes, such as
teachers, parents, and students. Overall, those involved in the process of
change agreed on the importance of improving student nutrition and health
and directing resources and energy toward achieving this goal. As shown in
table 5, the types of individuals and groups involved in the process of
change varied across the districts visited and often included individuals
from the district, school, and community.

38 In recognition of the importance of fruits and vegetables in children's
diets, Congress included the Fruit and Vegetable Pilot Program in the 2002
Farm Bill and expanded and made the program permanent in the 2004 Child
Nutrition and WIC Reauthorization Act. This program provides federal
grants to schools in eight states and on three Indian reservations to
provide free fruits and vegetables to students in order to improve student
nutrition and introduce healthy snack options.

Table 5: District, School, and Community Groups Involved in the Process of
Changing Competitive Foods, by District Visited

District-level involvement School-level involvement Community involvement
Source: GAO.

Note: In Independence, New Haven, Oakland, and McComb, some of the groups
indicated were involved in the process of changing competitive foods
through districtwide committees to address school nutrition and health
issues.

As shown in table 5, in the districts we visited, initiators of change
sometimes came from within the ranks of district staff and sometimes from
the community, while leaders of change often were district or school
staff. In some districts, one person or group acted as both the initiator
and leader of change, while in other cases, the initiator pushed the idea
of changing competitive foods forward and then the leader took over
implementation of the changes. For example, in several districts, the
superintendent or SFA director initiated and led changes to competitive
foods because of concerns about both student nutrition and competition
with the school food service. In another district, changes were initiated
and led largely by a middle school principal and a physical education
teacher. In contrast, in New Haven, a local pediatrician who was also the
district's medical adviser initiated change to competitive foods, and then
a committee of district staff, school staff, and community members took
over leadership of changes. In all of the districts we visited, a strong
leader helped formulate new policies, reached out to parties affected by
changes, and ensured that policies were implemented.

While competitive food changes were often initiated and led by a few
individuals, all six districts realized that changes would be successful
only with the involvement of a variety of people in the process in order
to ensure their support for the changes and help sustain changes. To
address this need, some districts, such as Oakland and New Haven, convened
formal committees to provide recommendations on school nutrition and
health issues. In other districts, leaders took steps to reach out
individually to those parties affected by changes, such as school
principals, teachers, and students, to obtain support before their
enactment. This support was also instrumental to sustaining changes. For
example, the McComb superintendent noted that successful implementation of
the district's coordinated school health program required both leadership
as well as substantial community involvement, including input from school
officials and teachers, parents, and health providers in the community.
The involvement of these groups in formulating changes helped ensure
continued commitment and support of the changes.

In addition to obtaining the support of a variety of groups before
implementing changes, many districts noted that obtaining student
acceptance of the changes was particularly important to the success of
change. To that end, some districts held nutrition and health fairs to
distribute related information to students and the broader community, and
others involved students in taste testing and voting on the foods they
preferred. For example, middle school students and parents in Fort Osage

taste-tested healthy competitive foods during parents' night and the
school's open house. In other districts, student feedback on policy
changes was sought by officials before their implementation, sometimes by
consulting the district's student advisory council.

Districts Faced Several Barriers to Changing Competitive Foods, Including
Schools' Concerns about Revenue Losses

While the districts we visited all noted several key components to
successful change, they also cited several barriers to implementing
changes to competitive foods that they had commonly faced. In particular,
officials in almost all of the districts visited cited opposition because
of concerns about future revenue losses as a barrier to changing the
availability of competitive foods. In these cases, school principals most
frequently expressed these concerns because competitive food revenues
often provided discretionary money that was otherwise unavailable to fund
a variety of projects and needs at the school level.

In addition, a lack of information on other districts' efforts to make
changes to competitive foods and a lack of nutritional guidelines for
these foods were also barriers to change in the districts we visited.
While a majority of the districts visited implemented their own changes to
competitive foods without knowledge of the steps taken by other districts
to make such changes, officials in Richland One and Oakland conducted
their own research on other state and district competitive food policies
in order to learn from the experiences of others. Related to this,
officials in Independence, Fort Osage, and New Haven noted that the lack
of agreedupon nutrition guidelines for competitive foods was an obstacle
to change because they had to independently develop their own nutrition
standards. From our observations during these visits, the absence of a
clear set of standards defining healthy and less nutritious foods can
create a problem for districts making changes to competitive foods as
continual debates and disagreements on such standards may slow the process
of policy formation, particularly when many groups are involved in the
process.

Several districts mentioned additional barriers to making changes, such as
determining the full extent of competitive food sales in schools and the
groups involved in sales. For example, the SFA director in Independence
noted that the district's delay in developing nutrition standards for
competitive foods was largely due to the difficulty of compiling a full
picture of all competitive food sales occurring in schools. Related to
this, given the number of groups that can be involved in these sales in
each school, and the lack of clear roles and responsibilities in this
environment, she noted that it has been difficult to determine the full
extent of groups involved in sales across the district. She emphasized
that it is important

for the groups that have a stake in the revenues generated by these sales
to be involved in discussions concerning competitive food policy in order
to obtain support for the policy and successful implementation of changes.

Some districts also noted that the need to continually monitor
implementation of competitive food changes at the school level can be an
obstacle to change. Officials in both Oakland and Richland One stated that
monitoring adherence to their competitive food policies at the school
level has been difficult but is necessary to effective implementation.
Even when school-level groups were involved in the process of developing
changes, monitoring policy implementation was difficult because of the
involvement of many groups in competitive food sales. In Richland One, the
SFA director noted that she often relies on observances by food service
staff working in the schools to ensure that the policy is being followed
by all groups selling competitive foods. Our own observations of the
school food environment during our visits to schools support the
conclusion that districts have difficulty monitoring the implementation of
policies, as we noticed a few deviations from district competitive food
policies in some schools. Even though school officials often seemed
devoted to the goal of improving the nutritional quality of foods
available in their schools, they typically faced many competing priorities
during the school day. Further, in many schools, it was unclear who was
responsible for ensuring that policies were effectively implemented.

Food service staff in Fort Osage and New Haven stated that the difficulty
of finding healthy foods that both meet district nutrition goals and
appeal to students was also a barrier to making changes to competitive
foods. They noted that some healthy foods students found appealing were
unavailable from vendors, while in other instances, the healthy foods
available were too expensive to sell to students. This sentiment was
echoed by students in several districts, as they expressed their opinions
that some of the newly introduced healthy foods, such as bottles of
flavored milk or juice, were too costly to purchase as part of their
lunch. In contrast to districts that mentioned problems obtaining new
healthy products from vendors, several districts noted that they were able
to work within their exclusive beverage contracts to obtain healthier
beverages to serve to students. However, at least one mentioned that the
major soft drink company with whom the district had a contract offered few
nonsoda options.

In addition, districts also faced the challenge of educating students
about healthy eating and encouraging students to change their behavior by
choosing healthy foods. In several of the districts, schools reported

providing students with nutrition information and education, sometimes
through classroom lessons, posters, and programs and activities to promote
healthy eating. During our visits, students we spoke with frequently
demonstrated their understanding of the importance of healthy eating, and
some noted that they would like to see the addition of certain healthy
foods to school offerings, such as a greater variety of fresh fruits and
vegetables. In addition, several middle school students in Fort Osage and
New Haven explained that after changes were made to competitive foods and
they began to learn more about good nutrition in their schools, they went
home and talked to their parents about these issues.39 However, experts
agree that it is more difficult to change behavior than to educate
individuals.40 Related to this, high school students in most of the
districts we visited mentioned that some students continue to purchase
less nutritious foods before school and after school from neighborhood
stores and restaurants. In addition, during our visits to schools, we
observed students eating a wide range of both healthy and unhealthy items
during lunch, with younger students being more likely to eat healthy foods
than high school students.41

As part of their role in helping districts develop wellness policies that
address, among other things, school nutrition, USDA has recently taken
several steps that may help districts overcome some of these barriers.42
In March 2005, USDA, in partnership with HHS/CDC and the Department of
Education, sent a letter to state superintendents, district
superintendents, and SFA directors describing the wellness policy
requirements. Through this letter, the agencies offered to provide
technical assistance to districts,

39 Related to this, students we met with made comments reflecting their
awareness that, in addition to schools, families and the broader community
play a key role in teaching children about good nutrition.

40 For more information on federally funded nutrition education programs
and efforts to increase healthy eating in schools, see GAO-04-528 and GAO,
School Lunch Program: Efforts Needed to Improve Nutrition and Encourage
Healthy Eating, GAO-03-506 (Washington, D.C.: May 9, 2003).

41 Specifically, we noticed that pizza appeared to be the most popular
item purchased for lunch by students in almost all of the schools we
visited. Burgers and fresh fruits and vegetables appeared to be the next
most popular items, as they were purchased and eaten by students during
lunch in almost half of the schools we visited. In the lunch periods we
observed, these foods were sometimes sold as competitive foods and were
sometimes served as components of the school lunch.

42 McComb used federal resources to assist its own process of changing
competitive foods. Changes in McComb were modeled on the CDC-developed
coordinated school health model.

and they also provided information on online resources available at the
federal level to help districts develop their policies. Specifically,
these agencies, in collaboration with several food and nutrition
organizations, have begun to compile resources that will provide districts
with information on state and local efforts to make changes to the school
health and nutrition environment, including examples of nutrition
standards used by states and localities that have already developed
competitive food policies.43 In addition to online resources, these three
agencies recently released Making it Happen! School Nutrition Success
Stories, a publication that describes local efforts to address the school
nutrition environment, in order to assist districts as they move forward
with their own changes.44

The Effects of Changes to Competitive Foods on Revenues Were Often Unclear
because of Limited Data; Nonetheless, Many Officials Expressed Concerns
about Revenue Losses

In the districts we visited, reliable data on how changes to competitive
food sales affected revenues were typically unavailable. Schools and
districts often did not maintain detailed revenue records to enable the
type of analyses needed to parcel out the direct effects of competitive
food changes on revenues. Nonetheless, most schools and districts were
able to provide partial data on revenue changes for specific venues, and
these limited data suggest that districts experienced mixed revenue
effects. Several schools we visited appear to have lost revenue from
competitive food sales after they made changes, while at least one may
have increased revenue. For example, after increasing the availability of
healthy a la carte foods and restricting less nutritious items, SFA
directors in both Independence and Richland One recorded decreases in a la
carte sales. In contrast, a middle school in McComb reported that after
removing soda from beverage vending machines and changing the policy
regarding the times sales were allowed, vending revenues increased.
However, these data did not account for other factors that may also have
affected revenues.

While the limited data available suggest that school districts experienced
mixed revenue effects after implementing competitive food changes, they
also illustrate the difficulty of tracking these effects. Because the

43 USDA stated that Action for Healthy Kids, the School Nutrition
Association, and the Food Research and Action Center are among the
organizations that have assisted with these efforts.

44 Food and Nutrition Service, USDA; CDC, HHS; and the U.S. Department of
Education. Making it Happen! School Nutrition Success Stories. Alexandria,
Va., January 2005.

competitive food environment is complex, sometimes involving many sales
and many groups, the effects of changes on revenues are often complicated
and may differ for each group involved in sales. Some groups may benefit
from changes, others may lose. For example, in Fort Osage, when the middle
school decided to remove all less nutritious competitive foods available
in the school and replace them with healthy items, it also simultaneously
decided to stop using outside vendors to supply its vending machines.
Subsequently, the district SFA took over operation of the middle school's
vending machines, and as a result, vending revenue began to accrue to the
SFA instead of school administrators. While the SFA director was unable to
compare the revenue before changes with that generated after changes, she
reported that the machines were selfsupporting. In this instance, because
of the changes made to competitive foods and their sale, school
administrators lost a source of revenue while the SFA gained one.

In addition to the challenge of understanding the revenue effects of
competitive food changes on different groups in schools, the relationship
between changes in a la carte sales and school meal participation adds
complexity. Although food service programs rely on reimbursement for
school meal participation as a primary source of funding, officials often
cite the importance of using additional revenue from a la carte sales to
balance their budgets. Therefore, officials take risks when they make
changes to the competitive foods available in schools, because changes may
affect revenues from these sales and they may also affect school meals
participation. In the districts we visited, competitive food changes were
often accompanied by increased school meals participation. In four
districts, federal reimbursements for meals subsequently increased,
benefiting the SFA, and in at least one instance, this increase more than
made up for food service losses in competitive food sales. While Richland
One reported losing approximately $300,000 in annual a la carte revenue
after implementing changes, school lunch participation and subsequent
federal reimbursements increased by approximately $400,000 in the same
year.

Despite the lack of conclusive data on revenue effects, district and
school officials often expressed strong concerns about potential revenue
losses. Because food services often operate on tight budgets and use
competitive food revenues to support their operations, they take the risk
of losing

important revenues when they make changes to these foods.45 In addition,
principals frequently stated that competitive food revenues are used at
the school level as discretionary funding, and they do not typically have
other sources of flexible funding available to use for the wide variety of
purposes toward which competitive food revenues are directed. Therefore,
when making changes to competitive foods, principals also risk losing what
is an oftentimes important source of funding. For example, in the schools
we visited, many principals reported using competitive food revenues for
student activities and classroom supplies, and some reported using these
revenues to support school dances and assist needy students.

Although data on revenue effects were limited and complicated by the
complex competitive food environment in the districts we visited, some
districts tried to lessen adverse revenue effects by the process through
which changes to competitive foods were implemented. Specifically, a few
districts and schools reported taking incremental steps to change
competitive foods in order to mitigate the severity of the effects on
revenue. For example, in one Independence high school, incremental changes
were being made to beverage vending machines to phase in juice and water
and phase out soda over a span of several years. The high school principal
reported that the school was conducting this change slowly in order to
avoid surprising students with sudden changes and to maintain revenue. In
addition, a few schools noted that their efforts to include students in
decisions about changing food offerings may have helped ensure that the
new foods would be accepted by students and mitigate the effects on
revenues. However, many of the districts we visited did not fully plan for
the effects on sales revenues when they were considering changes to
competitive food policy, and several recognized that efforts to do so
would have likely eased the implementation of policies. Moreover, some
principals reported that their schools were able to find ways to support
projects previously funded with competitive food sales after changes were
implemented. For example, in several districts, principals reported that
after restrictions on fund-raiser food sales were implemented, groups sold
nonfood items like wrapping paper and candles, and also raised funds by
providing services, such as car washes.

45 In the districts we visited, most SFA directors did not express
concerns about potential revenue losses resulting from changes to
competitive foods, possibly because they were often significantly involved
and invested in the process of making these changes.

Concluding Observations

In the publication Making it Happen! School Nutrition Success Stories, key
contacts in selected schools reported similar mixed revenue results from
their efforts to improve the school nutrition environment.46 Of the group
of schools that reported on revenue changes, some experienced increases in
revenue while others reported decreases or no change. These schools used
approaches similar to those in the schools we visited, such as replacing
less nutritious food with more nutritious choices, obtaining input from
the students, and using marketing to encourage students to make healthy
choices. In addition, several of the schools reported increases in school
meal participation.

Our nation's schools are uniquely positioned to positively influence the
eating habits of children, yet almost all schools sell readily available
foods that are largely unregulated by the federal government in terms of
nutritional content. While not all of these competitive foods are
unhealthy, many are. Although schools cannot be expected to solve the
current problems with child nutrition and growing obesity alone, many
states and districts have begun efforts to improve the nutritional
environments in their schools.

As districts across the country develop their required wellness policies
by school year 2006-2007, they will likely face decisions and challenges
similar to those of the districts we studied and may benefit from their
lessons learned. Although each district took a different approach, all of
them recognized the value of including those parties affected by the
changes, such as parents, teachers, and other community members, when
developing new policies. In addition, they recognized that students are
the ultimate consumers of competitive foods and took steps to consider
their opinions.

Because districts reported they typically lacked a source of recommended
nutrition standards for competitive foods in schools, officials were faced
with difficult decisions about the criteria they would use to determine
which foods were considered adequately nutritious to offer. The technical
assistance available from FNS, including multiple examples of nutrition
standards developed by other districts, as well as the Institute of

46 This publication contains self-reported information by key contacts
from 32 schools and districts nationwide.

Medicine's forthcoming recommendations on nutritional standards for foods
in schools, should help district efforts to address this issue.

In addition, given the multiple groups that rely on sales of competitive
foods for revenue, districts may choose to consider the possible revenue
effects of changes in food offerings as they develop and implement new
policies. Since competitive food revenues are often critical to food
service operations and provide principals with flexible funds relied on
for a multitude of discretionary purposes, making changes to competitive
foods entails risks for both groups. Districts we visited took varied
steps that may mitigate potential revenue changes, such as substituting
healthy foods for less healthy ones instead of removing all competitive
foods, asking students to taste and approve the more nutritious foods,
offering alternate means for fund-raising, or implementing change
gradually. Lack of support from the groups that use revenue from
competitive food sales can scuttle policy changes. Furthermore, the lack
of a single person responsible for the presence and sale of competitive
foods in schools complicates efforts to ensure that new policies will be
implemented as intended and maintained over time. Despite the complex food
environment in schools, new wellness policy requirements and USDA's
efforts to provide technical assistance to districts will provide an
opportunity for districts to plan and implement changes that recognize the
needs of the various groups and assign individuals with responsibility for
consistent and sustained implementation.

We provided a draft of this report to the U.S. Department of Agriculture
for review and comment. On June 17, 2005, FNS officials provided us with
their oral comments. The officials stated that they were in general
agreement with the findings as presented in the report and offered
technical comments that we have incorporated as appropriate. In addition,
the officials reiterated that the 1983 court decision in National Soft
Drink Ass'n v. Block is significant because they believe it severely
limits USDA's ability to restrict the sale of competitive foods. FNS
officials pointed out that prior to this ruling, USDA regulations
prohibited the sale of FMNV anywhere in the school from the beginning of
the school day until the last meal period. Following the decision, USDA
restricted the sale of FMNV only in food service areas during meal
periods. We agree that this ruling limited USDA's ability to regulate
competitive food sales as to time and place. However, we believe the
department has the authority to expand the definition of FMNV to include
additional foods with limited nutritional value. Doing so could further
limit the types of these foods available in the cafeteria during meal
times.

Agency Comments

Second, officials discussed what is known about the use of revenue from
competitive food sales and reimbursable meals compared to their costs. We
did not determine if revenues generated by competitive food sales were
sufficient to cover the actual cost of the competitive foods sold. The
officials stressed that the 1994 School Lunch and Breakfast Cost Study,
the last definitive study of cost and revenue in the NSLP and the School
Breakfast Program, found that regardless of size, most school food
authorities failed to generate enough revenue to cover the reported costs
of nonreimbursable food sales.47 The mean reported revenue-to-cost ratio
was 71 percent for the study period. The officials noted that this would
equate to a loss of 41 cents for every dollar received from the sale of
nonreimbursable foods. Further, this revenue-to-cost ratio did not include
all costs for school food service operations, such as uncharged labor
costs, indirect costs, and utilities. If these were included, the revenue
to cost ratio would generate even higher losses. FNS informed us that they
are in the process of contracting for a new school meal cost study.

We agree with FNS that our report focused on revenues generated by
competitive food sales and that we did not determine if revenues generated
by competitive food sales were sufficient to cover the actual cost of the
foods sold. However, our report and others have shown that the
availability of competitive foods, and particularly a la carte items, has
increased over time. In addition, the Cost Study's definition of
nonreimbursable meals included food sales such as adult meals and special
functions, as well as competitive foods, and therefore, it is unclear how
each of these types of sales contributed to the mean reported
revenue-to-cost ratio. Absent more current information on the actual costs
and revenues of providing competitive foods and reimbursable school meals,
we believe it is difficult to know whether the results of the 1994 study
are applicable today.

As agreed with your offices, unless you release the report's contents
earlier, we plan no further distribution of it until 30 days from its
issue date. We will send copies of this report to the Secretary of
Agriculture, appropriate congressional committees, and other interested
parties. In addition, the report will be available at no charge on GAO's
Web site at http://www.gao.gov.

47 Abt Associates, Inc. School Lunch and Breakfast Cost Study-Final
Report, a special report prepared at the request of USDA (Cambridge,
Mass.: October 1994).

If you or your staff have any questions about this report, please contact
me at (415) 904-2272 or [email protected]. Contact points for our Offices of
Congressional Relations and Public Affairs may be found on the last page
of this report. Key contributors to this report are listed in appendix II.

David D. Bellis Director, Education, Workforce, and Income Security Issues

                       Appendix I: Scope and Methodology

To obtain nationally representative information on competitive food
availability, policies, decision makers, groups involved in their sale,
and revenues generated by their sale, we conducted two Web-based surveys.
In addition, to gather information on strategies used by school districts
to restrict less nutritious competitive foods in their schools, we visited
six school districts. Further, to inform the design of our study, we spoke
with staff at the U.S. Department of Agriculture (USDA) as well as the
Centers for Disease Control and Prevention (CDC). We also interviewed
numerous researchers and organizations that have been involved with child
nutrition and school health environment issues in recent years.1

Surveys

To better understand competitive foods in the school environment, we
designed and administered two Web-based surveys. For a random sample of
schools, we administered one survey to each school's school food authority
(SFA) director and a second survey to each school's principal. We chose to
survey these officials because we believed they would be the most
knowledgeable sources on competitive food issues in schools. The surveys
were conducted between October 19, 2004, and February 11, 2005. We defined
competitive foods as all foods or beverages sold to students on school
grounds during the school day that are not part of federally reimbursable
school meals.

While neither survey asked questions about the full range of competitive
food issues, together the two surveys were designed to provide a broad
picture of the competitive food environment. Both Web surveys contained
school background, a la carte, (beverage and snack) vending machine, and
school store and snack bar sections. However, each survey had unique
sections as well. The SFA directors' survey included sections that asked
questions about the SFA, the school food service, and school meals
participation. The principals' survey included unique sections on school
and district policies for competitive foods, including fund-raising and
exclusive beverage contracts.

1 We spoke with staff of Mathematica Policy Research Inc., Abt Associates
Inc., Nutrition for the Future Inc., University of Minnesota-School of
Public Health, School Nutrition Association, the National Association of
Secondary School Principals, the National Association of Elementary School
Principals, Samuels & Associates, the Association of School Business
Officials International, the Council of Chief State School Officers, the
National Association of State Boards of Education, and the National School
Boards Association.

                       Appendix I: Scope and Methodology

Population

Sample Design and Errors

A majority of the survey questions asked both SFA directors and principals
to consider school year 2003-2004. To gain a sense of change for certain
competitive food issues, a few questions asked SFA directors and
principals to consider school year 1998-1999 alone, make comparisons
between these reporting periods (1998-1999 versus 2003-2004), or consider
change in specific competitive food issues over the entire time of these
reporting periods (from 1998-1999 to 2003-2004).

The target population consisted of all public schools in the 50 states and
the District of Columbia that participated in the National School Lunch
Program (NSLP) for the 2003-2004 school year. We used the Department of
Education's Common Core of Data (CCD) Public Elementary/Secondary School
preliminary file for the 2002-2003 school year as a basis for defining our
population. On the basis of our review of these data, we determined this
source to be adequate for the purposes of our work.

To define our sampling frame, we removed schools from the CCD that were
permanently or temporarily closed; not yet operational; special education,
vocational education, or alternative/other; run by the Department of
Defense or Bureau of Indian Affairs; or located in American Samoa, Guam,
Northern Marianas, Puerto Rico, or the Virgin Islands. From this analysis,
we obtained a sampling frame consisting of 85,569 regular public schools
in the 50 states and the District of Columbia. However, consistent
information specifically identifying a school's participation in the NSLP
was not available in the CCD.

The sample design for the Web surveys was a stratified random probability
sample of 656 schools that allows for estimates to be calculated for each
school level (elementary, middle, and high). We stratified by school
level, census region, and rural status, and we produced estimates by
school level. With this probability sample, each school in the population
had a known, nonzero probability of being selected. Each selected school
was subsequently weighted in the analysis to account statistically for all
the schools in the population, including those that were not selected.
Because each school was randomly chosen, some SFA directors had more than
one school under their responsibility selected for our study, and they
were therefore asked to complete a separate survey for each school.

Because we surveyed a sample of schools, our results are estimates of a
population of schools and thus are subject to sample errors that are
associated with samples of this size and type. Our confidence in the
precision of the results from this sample is expressed in 95 percent
confidence intervals, which are expected to include the actual results in
95

                       Appendix I: Scope and Methodology

percent of the samples of this type. We calculated confidence intervals
for this sample based on methods that are appropriate for a stratified
probability sample.

Through a telephone survey of the schools selected in our sample, we
determined the number of schools selected in our sample that participated
in the NSLP. We estimate that 80,245 (94 percent) schools in our
population participated in the NSLP. All estimates produced from the
sample and presented in this report are for the estimated target
population of 80,245 schools that participated in the NSLP. All percentage
and numerical estimates included in this report have margins of error of
plus or minus 15 percentage points or less, except for those shown in
table 6.

Table 6: Sampling Error Calculations for Questions in Which the Error
Exceeded 15 Percent

                 Page Question Estimate Lower bound Upper bound

15  Percentage of elementary schools with vending machines that         
                            had three or more                              
                vending machines in school year 2003-2004            29 15 47 
17 Percentage of schools with school stores in which salty snacks       
                            (not low-fat) were                             
      often or always available for students to purchase from these        
      stores in school year                                                
                                2003-2004                            45 29 61 
17    Percentage of schools with school stores in which sports          
                           drinks were often or                            
      always available for students to purchase from these stores in       
                            school year 2003-                              
                                   2004                              44 28 60 
17 Percentage of schools with school stores in which sweet baked        
      goods (not low-fat)                                                  
      were often or always available for students to purchase from         
      these stores in school                                               
                              year 2003-2004                         40 25 56 
17  Percentage of schools with school stores in which water was         
                             often or always                               
      available for students to purchase from these stores in school 47 32 62 
                              year 2003-2004                               
20  Percentage of middle schools with vending machines in which         
                           the total number of                             
       vending machines on school grounds increased between school         
                             years 1998-1999                               
                              and 2003-2004                          39 25 54 
20 Percentage of elementary schools with a la carte sales in            
      which the total volume of                                            
        a la carte items sold to students increased between school         
                           years 1998-1999 and                             
                                2003-2004                            31 19 47 
24    Percentage of elementary schools with one group directly          
                         involved in competitive                           
                   food sales in school year 2003-2004               36 19 56 
24   Percentage of elementary schools with two groups directly          
                         involved in competitive                           
                   food sales in school year 2003-2004               35 20 53 
24   Percentage of elementary schools with three or more groups         
                           directly involved in                            
             competitive food sales in school year 2003-2004         29 14 48 
24 Percentage of middle schools with two groups directly involved       
      in competitive food                                                  
                      sales in school year 2003-2004                 29 17 44 
24 Percentage of elementary schools with SFA/school food service        
      directly involved in                                                 
             competitive food sales in school year 2003-2004         62 45 77 

Appendix I: Scope and Methodology

                 Page Question Estimate Lower bound Upper bound

24 Percentage of elementary schools with student                        
      associations/clubs directly involved                                 
            in competitive food sales in school year 2003-2004       35 19 52 
24     Percentage of middle schools with school officials or            
                         administrators directly                           
                    involved in competitive food sales               35 21 51 
28    Percentage of high schools that generated total minimum           
                           combined revenue of                             
       $25,001 to $50,000 through competitive food sales in school   21 9  37 
                              year 2003-2004                               
32  Percentage of schools with exclusive beverage contracts that        
                             received noncash                              
        benefits through these contracts that were valued at over          
                          $5,000 in school year                            
                                2003-2004                            25 13 41 

                               Nonsampling Errors

Source: GAO.

We took steps to minimize nonsampling errors that are not accounted for
through statistical tests, like sampling errors. Nonsampling errors could
figure into any data collection effort and involve a range of issues that
could affect data quality, including variations in how respondents
interpret questions and their willingness to offer accurate responses.

In developing the Web surveys, we conducted several pretests of draft
instruments. We held pretest discussions of the principals' survey with
staff and members of the National Association of Elementary School
Principals and the National Association of Secondary School Principals. We
pretested the SFA survey with members of the American School Food Service
Association (now known as the School Nutrition Association). In addition,
both surveys were also pretested during a preliminary visit to the SFA and
an elementary school in Fairfax County, Virginia. All pretests were
conducted between July and September 2004.

For the survey pretests, we were generally interested in the clarity of
the questions and the flow and layout of the surveys. For example, we
wanted to ensure definitions used in the surveys were clear and known to
the respondents, categories provided in closed-ended questions were
complete and exclusive, and the ordering of the survey sections and the
questions within each section was appropriate. On the basis of our
pretests, the Web instruments underwent some slight revisions.

After the survey was closed, we also made comparisons between select items
from our competitive food Web-based survey data and other

                       Appendix I: Scope and Methodology

                                 Response Rates

national-level data sets.2 Our comparisons found our survey data were
reasonably consistent with the external sources. Of the basis of our
comparisons, we believe our survey data are sufficient for the purposes of
our work.

Using Web-based surveys also helped remove error in our data collection
effort. By allowing respondents to enter their responses directly into an
electronic instrument, this method automatically created a record for each
respondent in a data file and eliminated the need for and the errors (and
costs) associated with a manual data entry process. To further minimize
errors, programs used to analyze the survey data and make estimations were
independently verified to ensure the accuracy of this work.

For each school in our sample, we attempted to obtain valid e-mail
addresses for the principal and the SFA director. For the 656 schools in
our sample, we obtained valid e-mail addresses for 489 principals and 455
SFA directors. We administered the surveys to those groups, and we
received completed surveys from 70 percent of the SFA directors and 65
percent of the principals who received the surveys. The response rates for
our sample of 656 schools, including those officials we were unable to
contact, were 51 percent for both principals and SFA directors, excluding
the 26 non-NSLP schools. We received responses from both the SFA director
and the principal for the same school (matched responses) for 192 schools
(30 percent of schools that participated in the NSLP in our sample).
Tables 7 and 8 summarize the population and sample by school level for the
SFA director and principal surveys respectively.

2 We compared our Web-based survey data to data on competitive foods
reported by USDA in the School Nutrition and Dietary Assessment Study II
(1998-1999) and the School Meals Implementation Initiative- Third Year
Report (2002), and by CDC in the School Health Policies and Programs Study
(2000).

                       Appendix I: Scope and Methodology

      Table 7: SFA Director Survey: Population and Sample by School Level

                                                                         NSLP 
                Schools in Schools in Non-NSLP No valid    NSLP       schools 
                                                         schools   
School level population     sample  schools  e-mail   surveyed  responding 
    Elementary      51,997        188        6       55        127 
      Middle        15,737        188        3       45        140 
       High         14,979        188        6       48        134 
      Other          2,856         92       11       27         54 
      Total         85,569        656       26      175        455 

                                  Source: GAO.

        Table 8: Principal Survey: Population and Sample by School Level

                Schools in  Schools  Non-NSLP No valid    NSLP           NSLP 
                              in                        schools       schools 
School level population    sample  schools  e-mail   surveyed   responding 
    Elementary      51,997       188        6       39        143 
      Middle        15,737       188        3       39        146 
       High         14,979       188        6       41        141 
      Other          2,856        92       11       22         59 
      Total         85,569       656       26      143        489 

Source: GAO.

While the majority of our estimates are calculated based on survey
responses from either the SFA directors' survey or the principals' survey,
we used the matched responses (192 schools) to calculate the total
combined minimum revenue estimates. Specifically, the amount of revenue
earned from each competitive food venue within a school was reported by
SFA directors and principals in multiple ranges. For example, SFA
directors reported revenue generated by food services through competitive
food sales, such as a la carte sales, and principals reported revenue
generated by all other competitive food sales in the school. To estimate
the combined competitive food revenue for a school, we defined the minimum
for each of the school's venues as the lower bound of the revenue range
reported by SFA directors and principals. We then summed the minimum
revenue across all venues for each school.

Survey Nonresponse Issues	Another type of nonsampling error is nonresponse
or, in the case of our work, those SFA directors and principals from
schools in our sample who did not provide a complete survey. To increase
survey responses, after the Web surveys were initially deployed, we made
several follow-ups with

                       Appendix I: Scope and Methodology

nonrespondent SFA directors and principals via e-mail and phone to remind
them of their respective surveys.

After the surveys were closed, we analyzed each set of survey respondents
(SFA directors, principals, and the matched responses) to determine if
there were any differences between the responding schools, the
nonresponding schools, and the population.3 We performed this analysis for
four characteristics-total number of students enrolled, total number
eligible for free lunch, total number eligible for reduced price lunch,
and total number eligible for either free or reduced price lunch. We
determined whether sample-based estimates of these characteristics
compared favorably with the known population values, and we also tested
the differences of the estimates for survey respondents and
nonrespondents. We performed this analysis for all schools and separately
for each school level (high, middle, and elementary). For each set of
survey respondents, the population value for all of the characteristics we
examined fell within the 95 percent confidence intervals for the
estimates. We also determined that there were no significant differences
between estimates from the respondents and nonrespondents.

Additionally, we compared the distribution of several demographic
variables, including region, school level, and rural status, for survey
respondents and nonrespondents. On the basis of this analysis, we found no
significant distributional differences between respondents and
nonrespondents. Although the characteristics were selected because they
may be related to other school characteristics asked for on our surveys,
we do not know the extent to which the respondents reflect the population
characteristics for our specific survey questions. On the basis of both
sets of analyses, we chose to include the survey results in our report and
produce sample-based estimates to the population of schools that
participated in the NSLP.

Site Visits 	To gather information on local efforts to restrict the
availability of less nutritious competitive foods, we conducted site
visits to six districts between September 21 and December 9, 2004. The
districts visited included Independence School District (Independence,
Missouri), Fort Osage R-1 School District (Independence, Missouri), New
Haven Public

3 Nonresponding schools include both schools for which we were unable to
obtain valid email addresses and schools that received the survey but did
not respond.

Appendix I: Scope and Methodology

Schools (New Haven, Connecticut), Richland County School District One
(Columbia, South Carolina), Oakland Unified School District (Oakland,
California), and McComb School District (McComb, Mississippi). We selected
these districts from a list of approximately 100 districts and schools
recognized as making efforts to restrict access to less nutritious
competitive foods. This list was compiled by reviewing recently released
reports, studies, and articles that described local efforts to make
changes to competitive foods. The six districts visited were selected
because they used different strategies to restrict competitive foods, and
when viewed as a group, they provided variation across characteristics
such as geographic location, district size, and socioeconomic status.

During the site visits, we interviewed district officials, including the
superintendent and SFA director, as well as visited one or two schools
within each district. At the schools, we interviewed principals, food
service staff, and health and physical education teachers, as well as
others involved with the school food environment. From these interviews,
we gathered information on the district and school food environment,
strategies used to restrict competitive foods, individuals and groups
involved in implementing changes, facilitators and barriers to change,
revenues generated by competitive foods, ongoing efforts, students'
reactions to changes, and opinions on the school wellness policies
mandated in the 2004 Child Nutrition and WIC Reauthorization Act. In
addition to our interviews with district and school officials, we also met
with students to collect their opinions regarding nutrition and healthy
eating, competitive foods in schools, and school meals. Further, we
observed at least one lunch period in each school visited, in order to
better understand the school nutrition environment and the choices
students make at lunch.

Appendix II: GAO Contacts and Staff Acknowledgments

GAO Contacts

Staff Acknowledgments

David Bellis, Director, (415) 904-2272, [email protected]

Kay E. Brown, Assistant Director, and Rachel Frisk, Analyst in Charge,
managed this assignment and made significant contributions to all aspects
of this report. Marissa Jones, Avani Locke, Kevin Jackson, and Jim Ashley
also made significant contributions to this report. In addition, Daniel
Schwimer assisted in the legal analysis, Amber Edwards assisted in the
analysis of local efforts to restrict competitive foods across the
country, and Susan Bernstein assisted in the message and report
development.

Related GAO Products

Commercial Activities in Schools: Use of Student Data is Limited and
Additional Dissemination of Guidance Could Help Districts Develop
Policies. GAO-04-810. Washington, D.C.: August 20, 2004.

Nutrition Education: USDA Provides Services through Multiple Programs, but
Stronger Linkages among Efforts Are Needed. GAO-04-528. Washington, D.C.:
April 27, 2004.

School Meal Programs: Competitive Foods Are Available in Many Schools;
Actions Taken to Restrict Them Differ by State and Locality. GAO-04-673.
Washington, D.C.: April 23, 2004.

School Lunch Program: Efforts Needed to Improve Nutrition and Encourage
Healthy Eating. GAO-03-506. Washington, D.C.: May 9, 2003.

School Meal Programs: Revenue and Expense Information from Selected
States. GAO-03-569. Washington, D.C.: May 9, 2003.

GAO's Mission

Obtaining Copies of GAO Reports and Testimony

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.

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

Order by Mail or Phone 	The first copy of each printed report is free.
Additional copies are $2 each. A check or money order should be made out
to the Superintendent of Documents. GAO also accepts VISA and Mastercard.
Orders for 100 or more copies mailed to a single address are discounted 25
percent. Orders should be sent to:

U.S. Government Accountability Office 441 G Street NW, Room LM Washington,
D.C. 20548

To order by Phone: 	Voice: (202) 512-6000 TDD: (202) 512-2537 Fax: (202)
512-6061

To Report Fraud, Contact:

Waste, and Abuse in Web site: www.gao.gov/fraudnet/fraudnet.htm

E-mail: [email protected] Programs Automated answering system: (800)
424-5454 or (202) 512-7470

Gloria Jarmon, Managing Director, [email protected] (202)
512-4400Congressional U.S. Government Accountability Office, 441 G Street
NW, Room 7125 Relations Washington, D.C. 20548

Public Affairs 	Paul Anderson, Managing Director, [email protected] (202)
512-4800 U.S. Government Accountability Office, 441 G Street NW, Room 7149
Washington, D.C. 20548

                           PRINTED ON RECYCLED PAPER
*** End of document. ***