Welfare Reform: Effects of Reduced Reimbursements on the Summer Food
Service Program (Letter Report, 11/10/98, GAO/RCED-99-20).
Pursuant to a congressional request, GAO reviewed the Department of
Agriculture's (USDA) Summer Food Service Program and the impacts the
Personal Responsibility and Work Opportunity Reconciliation Act of 1996
had on the program, focusing on the: (1) number and characteristics of
sponsors participating in and dropping out of the program before and
after the decrease in the reimbursements; (2) number of children and
meals served by the program before and after the reduction; and (3)
changes sponsors made to their programs as a result of the reduced
reimbursements.
GAO noted that: (1) the reduction in meal reimbursements that resulted
from the 1996 Welfare Reform Act had a minimal impact on the number and
characteristics of the sponsors of the Summer Food Service Program; (2)
since the reduction went into effect in 1997, the number of sponsors
participating in the program increased by 8 percent overall, from 3,753
to 4,046; (3) the characteristics of the sponsors providing meals
remained about the same between 1996 and 1997; (4) in both years, a
relatively small percentage of sponsors--5 percent--served most of the
children in the program; (5) most sponsors continued to be schools and
camps; (6) in terms of the sponsors that dropped out after the welfare
reform changes, fewer than 10 percent left the program in each of the 2
years since the reduced reimbursements have been in effect; (7) however,
only a small percentage of these dropouts left because of the reduction,
according to officials in the 50 states; (8) after the reduced
reimbursements were mandated, USDA and some states took actions to
maintain the level of participation of sponsors and children; (9) the
number of children and meals served in fiscal year (FY) 1997 was greater
than in previous years; (10) the total number of children participating
in the program increased by over 2 percent after the reimbursements were
reduced, to almost 2.3 million, in 1997; (11) the number of meals served
rose by 2 percent, to over 128 million, despite a new restriction on the
number of meals for which some sponsors could receive reimbursements;
(12) state officials identified a small number of sponsors that left the
program because of the reduced reimbursements; (13) over 820 children
lost access to the program in FY 1997 because sponsors that had served
them in 1996 did not participate in 1997 as a result of the reduction,
and no other sponsor was available; (14) at least 780 children lost
access in FY 1998; (15) other children may have lost access when
continuing sponsors reduced the number of sites they operated because of
the rate reduction; (16) in response to the reduced meal reimbursements,
some sponsors reported making changes to their program operations; (17)
even with the changes, more sponsors reported that their operating costs
exceeded the amount they received in federal reimbursements in FY 1997
than in 1996; and (18) the limited impact on the number of sponsors,
children, and meals served that has been observed to date is due in part
to sponsors' continuing to contribute funds to offset the decreased
reimbursements.
--------------------------- Indexing Terms -----------------------------
REPORTNUM: RCED-99-20
TITLE: Welfare Reform: Effects of Reduced Reimbursements on the
Summer Food Service Program
DATE: 11/10/98
SUBJECT: Children
Food programs for children
State-administered programs
Intergovernmental fiscal relations
Federal grants
Welfare benefits
Federal/state relations
Program evaluation
Subsidies
Statistical data
IDENTIFIER: Summer Food Service Program for Children
******************************************************************
** This file contains an ASCII representation of the text of a **
** GAO report. Delineations within the text indicating chapter **
** titles, headings, and bullets are preserved. Major **
** divisions and subdivisions of the text, such as Chapters, **
** Sections, and Appendixes, are identified by double and **
** single lines. The numbers on the right end of these lines **
** indicate the position of each of the subsections in the **
** document outline. These numbers do NOT correspond with the **
** page numbers of the printed 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. **
** **
** A printed copy of this report may be obtained from the GAO **
** Document Distribution Center. For further details, please **
** send an e-mail message to: **
** **
** **
** **
** with the message 'info' in the body. **
******************************************************************
Cover
================================================================ COVER
Report to the Chairman, Committee on Education and the Workforce,
House of Representatives
November 1998
WELFARE REFORM - EFFECTS OF
REDUCED REIMBURSEMENTS ON THE
SUMMER FOOD SERVICE PROGRAM
GAO/RCED-99-20
Welfare Reform
(150285)
[email protected]. Department of Agriculture
Letter
=============================================================== LETTER
B-281248
November 10, 1998
The Honorable William F. Goodling
Chairman, Committee on Education
and the Workforce
House of Representatives
Dear Mr. Chairman:
The U.S. Department of Agriculture's (USDA) Summer Food Service
Program provides free meals to needy children when school is not in
session. In fiscal year 1997, over 3,800 sponsors--such as schools
and local government agencies--provided these meals to over 2 million
children at a federal cost of about $258 million. The Personal
Responsibility and Work Opportunity Reconciliation Act of 1996 (known
as the Welfare Reform Act) made major changes to this program,
effective in January 1997. It reduced the federal reimbursements
that sponsors receive for meals served and eliminated the federal
reimbursement for a fourth daily meal provided in summer camps and in
programs that primarily serve migrant children.\1
Because of questions raised about the impact that the reductions in
the federal reimbursements might have had on access to the program,
you asked us to report on (1) the number and characteristics of
sponsors participating in and dropping out of the program before and
after the decrease in the reimbursements, (2) the number of children
and meals served by the program before and after the reduction, and
(3) the changes sponsors made to their programs as a result of the
reduced reimbursements.
Our report is based on our analyses of the data provided to us by the
50 states about the program's sponsors. Through two mail surveys, we
collected detailed information on the sponsors for fiscal years 1996
and 1997 and on sponsors' participation in 1998.\2 We also conducted
a telephone survey with officials for the Summer Food Service Program
in the 50 states and the District of Columbia. In addition, to
obtain their views on the effect the changes had on the program, we
visited 20 sponsors in six states as well as advocacy organizations.
Appendix I discusses our scope and methodology in more detail.
--------------------
\1 On Mar. 10, 1998, we testified on our preliminary findings on
these issues. See Welfare Reform: Effects of Changes Made to the
Summer Food Service Program (GAO/T-RCED-98-120).
\2 This information does not include the District of Columbia, Puerto
Rico, and the U.S. Virgin Islands.
RESULTS IN BRIEF
------------------------------------------------------------ Letter :1
The reduction in meal reimbursements that resulted from the 1996
Welfare Reform Act had a minimal impact on the number and
characteristics of the sponsors of the Summer Food Service Program.
Since the reduction went into effect in 1997, the number of sponsors
participating in the program increased by 8 percent overall, from
3,753 to 4,046. Additionally, the characteristics of the sponsors
providing meals remained about the same between 1996, the year before
the subsidies were reduced, and 1997. In both years, a relatively
small percentage of sponsors--5 percent--served most of the children
in the program. Most sponsors continued to be schools and camps. In
terms of the sponsors that dropped out after the welfare reform
changes, fewer than 10 percent left the program in each of the 2
years since the reduced reimbursements have been in effect. However,
only a small percentage of these dropouts left because of the
reduction, according to officials in the 50 states. Nevertheless,
after the reduced reimbursements were mandated, USDA and some states
took actions, such as expanding outreach efforts, to maintain the
level of participation of sponsors and children. These efforts may
have mitigated the impact of the rate reduction.
Despite the reduced reimbursements, the number of children and meals
served in fiscal year 1997 was greater than in previous years. The
total number of children participating in the program increased by
over 2 percent after the reimbursements were reduced, to almost 2.3
million, in 1997. The number of meals served rose by 2 percent, to
over 128 million, despite a new restriction on the number of meals
for which some sponsors could receive reimbursements. However, state
officials identified a small number of sponsors that left the program
because of the reduced reimbursements, which meant that some of the
children whom the sponsors had served lost access to the program.
Over 820 children lost access to the program in fiscal year 1997
because the sponsors that had served them in 1996 did not participate
in 1997 as result of the reduction, and no other sponsor was
available. Similarly, at least 780 children lost access in fiscal
year 1998. Other children may have lost access when continuing
sponsors reduced the number of sites they operated because of the
rate reduction.
In response to the reduced meal reimbursements, some sponsors
reported making changes to their program operations. These changes
included substituting less expensive foods for those previously
served, such as replacing fresh fruit with fruit juices; reducing
staff's wages; and reducing the number of sites at which meals are
served. Even with these changes, according to the information
provided by the 50 states, more sponsors reported that their
operating costs exceeded the amount they received in federal
reimbursements in fiscal year 1997 than in 1996. The limited impact
on the number of sponsors, children, and meals served that has been
observed to date is due in part to sponsors' continuing to contribute
funds to offset the decreased reimbursements.
BACKGROUND
------------------------------------------------------------ Letter :2
The Summer Food Service Program is a federal entitlement program that
provides funds for program sponsors to serve free nutritious meals to
children in low-income areas when school is not in session. It is
administered by USDA's Food and Nutrition Service, which provides
money to state agencies to operate the program and to reimburse local
eligible sponsors for meals served to children at designated
locations.\3 Eligible sponsoring organizations include (1) public or
private nonprofit schools; (2) units of local, municipal, county, or
state governments, such as county or city recreation programs; (3)
private nonprofit groups, such as Boys and Girls Clubs or churches;
(4) residential camps; and (5) National Youth Sports Programs.\4 In
fiscal year 1997, sponsors served over 128 million meals at a total
federal cost of about $258 million.
Local program sponsors can qualify to be reimbursed for the free
meals served to all children 18 or younger by operating a site in an
eligible area. Eligible areas are those in which at least 50 percent
of the local children are from households with an income at or below
the eligibility level for free and reduced-price school meals--185
percent of the federal poverty guidelines ($29,693 for a family of
four in the summer of 1997).\5
Sponsors can also qualify for reimbursements for free meals served to
all children at sites not located in eligible areas if at least 50
percent of the children enrolled at such sites are eligible for free
or reduced-price school meals. Finally, camps may be reimbursed only
for meals that are served to children who have been individually
determined to be eligible because of their household income.
The meals that sponsors provide must meet the program's nutritional
requirements. Most sponsors can receive federal subsidies for only
two meals per child per day.\6
However, camps and programs primarily serving migrant children can
receive subsidies for up to three meals each day for each child.
This three-meal allowance is a change in the program made by the
Welfare Reform Act. Previously, these sponsors could receive
reimbursements for up to four meals per day.
Sponsors are reimbursed in two different categories for their costs
of preparing and serving free meals. One category covers the
administrative costs incurred in the management of the Summer Food
Service Program, such as office expenses, the support staff's
salaries, insurance, and some financial management costs. The second
category covers the operating costs for purchasing, preparing,
transporting, and serving the food; supporting program activities,
paying salaries for the staff supervising the children; and providing
transportation in rural areas. Sponsors must maintain records to
document all costs and the number of meals they claim for
reimbursements.
Sponsors' reimbursements are based on the lesser of (1) the number of
approved meals served multiplied by the established rate for each
type of meal or (2) the actual costs reported. The reimbursement
rates for both administrative and operating costs are set by law and
adjusted each year to reflect changes in the Consumer Price Index.
Sponsors with costs that exceed the federal reimbursements must find
other sources of funding to pay their additional costs. Some
sponsors anticipate that program costs will exceed the federal
reimbursements, and these sponsors offset their excess costs with
other sources of funds. Other sponsors inadvertently exceed the
federal reimbursements by small amounts because it is difficult to
budget the program exactly. Because sponsors can claim
reimbursements only for meals that are served to eligible children,
their costs would exceed reimbursements if they have low attendance
or prepare too many meals that cannot be claimed for reimbursements.
The Welfare Reform Act reduced the operating subsidies for meals and
snacks served under the Summer Food Service Program, effective for
the 1997 summer. It did not reduce subsidies for administrative
costs.\7 Table 1 highlights the changes in the reimbursement rates
for meals since 1996. While the 1997 and 1998 rates reflect
increases to account for inflation, they are still lower than the
rates established for 1996.
Table 1
Reimbursement Rates, per Meal Served,
1996-98
1997 rates
----------------------
Prior to
welfare
Meal 1996 rates reform\a Actual\b 1998 rates
---------------------- ---------- ---------- ---------- ----------
Breakfast $1.2075 $1.24 $1.16 $1.19
Lunch/supper 2.1675 2.23 2.02 2.08
Snack 0.57 0.58 0.47 0.48
----------------------------------------------------------------------
\a These are the rates that would have been in effect if the Welfare
Reform Act had not been enacted. Rates would have been higher than
1996 rates because of indexing for inflation.
\b The Welfare Reform Act set the rates at $1.13 for breakfast, $1.97
for lunch/supper, and $0.46 for snacks. However, the 1997 rates were
slightly higher than the rates established in the law because they
were indexed for inflation.
--------------------
\3 In 1997, USDA operated the program in the place of state agencies
in four states--Georgia, Michigan, New York, and Virginia.
Throughout this report, the term "state officials" includes the USDA
officials who manage the program in these four states.
\4 The National Youth Sports Program offers sports and educational
activities to disadvantaged children who are 10 to 16 years old. The
program is typically at colleges or universities and is associated
with the National Collegiate Athletic Association.
\5 These levels are slightly higher in Alaska and Hawaii.
\6 A meal can be breakfast, lunch, supper, or a snack.
\7 The 1998 administrative rate ranges from 4.75 cents to 21.75 cents
per meal and varies by type of meal; sponsor (e.g., rural); and
preparation (e.g., sponsor-prepared meal).
NUMBER AND CHARACTERISTICS OF
SPONSORS PARTICIPATING IN AND
DROPPING OUT OF THE PROGRAM
SINCE THE REIMBURSEMENTS WERE
REDUCED
------------------------------------------------------------ Letter :3
Since the reduced reimbursements went into effect in 1997, the number
of sponsors participating in the program has increased by 8 percent
overall. Although there were some changes in sponsors from one year
to the next, the characteristics of these sponsors--by the type of
organization, number of sites operated, and number of meals and
children served--remained about the same. Of those sponsors leaving
the program, fewer than 10 percent left in each of the 2 years, and
of these sponsors, few left because of the rate reduction, according
to state officials. USDA and some states took actions following the
rate reduction aimed at enrolling new sponsors and/or encouraging
current sponsors to expand their programs.
NUMBER OF SPONSORS INCREASED
OVERALL AFTER THE
REIMBURSEMENTS WERE REDUCED
---------------------------------------------------------- Letter :3.1
Since the reduced reimbursements have gone into effect, the total
number of sponsors has increased by about 8 percent. According to
the data provided by state officials, the total number of sponsors
participating in the program rose from 3,753 in fiscal year 1996 to
3,875 in 1997, and to 4,046 in 1998. While some sponsors left the
program, the number of sponsors joining it exceeded the number
leaving.
This overall increase in the number of sponsors occurred after
virtually no increase from 1995 through 1996, according to USDA's
data. However, this overall increase is lower than the increases
that occurred from 1991 through 1994, when the number of sponsors
grew each year by about 8 to 10 percent. According to USDA
officials, no growth occurred from 1995 to 1996 because of
uncertainty stemming from the public policy debates over the future
of the Summer Food Service Program and other child nutrition
programs. Figure 1 shows the rate of change in the number of
sponsors each year from 1991 through 1998.
Figure 1: Rate of Change in
Sponsors' Participation,
1991-98
(See figure in printed
edition.)
Notes: Rates of change for 1991 to 1996 are based on USDA's data on
the number of sponsors operating in July. Rates of change for 1997
and 1998 are based on state-provided data on the total number of
sponsors, including those that did not operate in July. USDA's data
for 1997 show an increase in the number of sponsors--4.8
percent--that is higher than that shown by the states' data, which is
3.3 percent.
From 1982 through 1989, most private nonprofit sponsors were not
allowed to participate in the program. The growth in the number of
sponsors beginning in 1991 is partially due to private nonprofit
sponsors joining the program.
Source: GAO's analysis of data provided by USDA and the states.
CHARACTERISTICS OF SPONSORS
WERE SIMILAR BEFORE AND
AFTER RATE REDUCTION
---------------------------------------------------------- Letter :3.2
The types of organizations serving as program sponsors were similar
in fiscal years 1996 and 1997. In both years, schools represented
the largest group of sponsors (about 45 percent), followed by camps
(19 percent), government agencies (about 17 percent), private
nonprofit organizations (about 16 percent), and National Youth Sports
Programs (3 percent). School sponsors can serve meals to children at
school buildings as well as at other types of locations, such as
parks and churches.
Similarly, the rate reduction did not have a significant effect on
the number of sites operated by sponsors. For example, in both
fiscal years, half of the sponsors (51 percent) operated 1 site, and
6 percent operated 25 or more sites. In addition, between fiscal
years 1996 and 1997, the total number of sites increased by 5
percent, from 29,220 to 30,587.
As tables 2 and 3 show, the 1997 reduction in meal reimbursements did
not appear to cause sponsors to change the size of their programs.
There was little change in the size of sponsors--in terms of the
number of meals and children served--between fiscal years 1996 and
1997. However, the number of meals and children served by individual
sponsors varied greatly in each year.\8 For example, one of the
smallest nonprofit sponsors served 92 meals in 1997, while the
largest--the New York City Board of Education--served over 8.8
million meals.\9 In both years, a small percentage of the sponsors
provided most of the meals. Only 5 percent of the sponsors served 62
percent of children participating in July of each year, 1996 and
1997. Similarly, 5 percent of the sponsors served 58 percent and 59
percent of all meals in 1996 and 1997, respectively.
Table 2
Size of Sponsors' Programs in Terms of
Children Served, 1996 and 1997
Average number of children Percent of Percent of
served daily sponsors in 1996\a sponsors in 1997\b
------------------------------ ------------------ ------------------
75 or fewer 25.9 25.7
76 to 299 34.0 32.9
300 to 999 20.9 20.4
1,000 or more 11.9 11.9
No July program 7.4 9.0
----------------------------------------------------------------------
\a Does not include data from New Jersey and Arkansas.
\b Does not include data from New Jersey.
Source: GAO's analysis of states' data.
Table 3
Size of Sponsors' Programs in Terms of
Meals Served, Fiscal Years 1996 and 1997
Characteristic Fiscal year 1996 Fiscal year 1997
------------------------------ ------------------ ------------------
Percent of sponsors serving
----------------------------------------------------------------------
Fewer than 3,500 meals per 25.4% 26.7%
year
3,500 to 9,999 meals per year 32.2% 31.3%
10,000 to 29,999 meals per 24.9% 24.9%
year
30,000 or more meals per year 17.5% 17.1%
Median number of total meals 7,832 7,746
served annually
Minimum number of meals served 62 92
annually by a sponsor
Maximum number of meals served 6,673,337 8,843,294
annually by a sponsor
Total meals served during the 120,539,635 125,082,928
year
----------------------------------------------------------------------
Source: GAO's analysis of states' data.
--------------------
\8 USDA generally reports the number of children served on the basis
of average daily attendance in July because the highest number of
children are typically served in July. While we requested the states
to provide average daily attendance in July for each sponsor, a few
states provided us with a different estimate for participation. (See
app. I.) New Jersey did not provide attendance data for 1996 and
1997, and Arkansas did not provide data for 1996. Our analyses of
the number of children served is based on state-provided data,
excluding New Jersey and Arkansas, as appropriate.
\9 Private nonprofit sponsors tended to operate smaller programs for
a number of reasons. For example, subsection 1761 (a) (7) of Title
42, United States Code, limited private nonprofit sponsors to serving
no more than 2,500 children per day at 5 urban sites or 20 rural
sites. However, these limits on private nonprofit sponsors were
changed by legislation signed by the President on October 31, 1998,
which (1) limits the number of sites to 25 without any distinction
between rural and other areas and (2) eliminates the 2,500 children
ceiling.
FEW SPONSORS LEFT BECAUSE OF
THE REDUCED REIMBURSEMENTS
---------------------------------------------------------- Letter :3.3
After the first year's experience with the reduced reimbursements,
both USDA and some state officials expressed concern that more
sponsors would drop out of the program in the future than did in 1997
because of the reduced rates. USDA officials believed that some
sponsors chose to continue participating in the program in 1997 to
test their ability to manage the program financially with the reduced
reimbursements and to determine whether they would continue in 1998.
However, an increase in dropouts did not occur in 1998.
According to the states' information, a total of 675 sponsors have
stopped participating in the Summer Food Service Program since the
reimbursements were reduced. Of those that participated in fiscal
year 1996, 370, or 9.9 percent, did not return in 1997; and 305, or
7.9 percent, of those that participated in 1997 did not return in
1998.\10
According to state officials, few of the sponsors that dropped out of
the program left specifically because of the reduced reimbursements.
State officials identified the reduced reimbursements as the major
reason for dropping out for only 37 (5.5 percent) of the 675 sponsors
that left in fiscal years 1997 and 1998. For an additional 182
sponsors, state officials were unable to identify the major reason
for dropping out. For some of these sponsors, the reduction may have
been the reason they dropped out. The remaining 456 sponsors left
for a variety of other reasons, such as the loss of personnel or
lower participation than anticipated. Table 4 shows the number and
percentage of sponsors leaving the program for various reasons since
the rate reduction.
Table 4
Reasons Sponsors Dropped Out of the
Program Since the Reduced Reimbursements
Total dropped out
Dropped out between Dropped out between since reduced
Reason 1996 and 1997 1997 and 1998 reimbursements
----------------- ---------------------- ---------------------- ----------------------
Number of Number of Number of
sponsors Percent sponsors Percent sponsors Percent
----------------- ---------- ---------- ---------- ---------- ---------- ----------
All reasons 370 100.0 305 100.0 675 100.0
Because of the 24 6.5 13 4.3 37 5.5
reduced rates
Because of other 218 58.9 238 78.0 456 67.6
reasons such as
Low participation 17 4.6 45 14.8 62 9.2
Personnel loss or 28 7.6 34 11.1 62 9.2
change
Primary activity 23 6.2 29 9.5 52 7.7
changed or
canceled
Costs exceeded 17 4.6 24 7.9 41 6.1
reimbursements
Poor management 11 3.0 16 5.2 27 4.0
Disruption 9 2.4 17 5.6 26 3.9
because of
construction
Paperwork burden 8 2.2 12 3.9 20 3.0
Replaced by 8 2.2 13 4.3 21 3.1
another meal
program
Other\a 97 26.2 48 15.7 145 21.5
Reason unknown 128 34.6 54 17.7 182 27.0
-----------------------------------------------------------------------------------------
\a Other reasons include sponsors that merged with other sponsors in
the program, could not locate a site, or moved to another state.
Source: GAO's analysis of states' data.
--------------------
\10 We did not compare these dropout rates with the rates of prior
years because USDA does not collect information on dropouts, and we
did not obtain this information from the states.
SMALL SPONSORS AND PRIVATE
NONPROFIT SPONSORS WERE MORE
LIKELY TO LEAVE THE PROGRAM
---------------------------------------------------------- Letter :3.4
According to our nationwide data, small sponsors were more likely to
drop out of the Summer Food Service Program than large sponsors, and
private nonprofit sponsors were more likely to drop out than other
types of sponsors. This was also true for the sponsors that state
officials identified as leaving the program because of the reduced
reimbursements. For example, the sponsors that left the program in
fiscal year 1997 because of the decrease had served an average of 207
children in July 1996, while the sponsors that remained in the
program in 1997 had served an average of 809 children in July 1996.
(App. II provides information on the dropout rates for various types
of sponsors.)
USDA AND SOME STATES
INITIATED ACTIONS AIMED AT
MITIGATING THE POTENTIAL
IMPACT OF THE REDUCED
REIMBURSEMENTS ON THE NUMBER
OF SPONSORS
---------------------------------------------------------- Letter :3.5
Some states and USDA took actions following the reduced
reimbursements for meals served to enroll new sponsors and/or
encourage current sponsors to expand their programs. These actions
may have mitigated the impact of the rate reduction. For example,
officials from several states said that they expanded outreach
efforts to enroll more new sponsors. Some states' Summer Food
Service Programs--such as those in Minnesota, New York, Vermont, and
Washington--received state funds to offset the loss of federal funds
in fiscal years 1997 and 1998.
In addition, in 1998, USDA began an initiative to encourage sponsors'
participation by conducting outreach and reducing the administrative
burden on sponsors. To encourage participation in 1998, USDA
provided program information to the American School Food Service
Journal, the National Conference of Mayors, and the National
Recreation and Park Associations for distribution to schools, local
governments, and private nonprofit organizations. USDA also provided
materials to the states for their own outreach efforts. Furthermore,
the Department issued 13 policy memorandums to the states revising
and clarifying policies to improve the program's operations and
reduce its administrative burden.
Despite the actions by USDA and the states to encourage new
organizations to participate, one state official reported that the
reduced reimbursement rates discouraged some organizations from
entering the program.
NUMBER OF CHILDREN AND MEALS
SERVED SINCE THE REDUCED
REIMBURSEMENTS
------------------------------------------------------------ Letter :4
The total number of children participating in the program continued
to increase in 1997 after the reduced reimbursements, as it had in
all but one of the previous 6 years. However, some of the children
served by sponsors in 1996 lost access to the program in 1997 when
those sponsors left the program. The number of meals served also
increased in 1997, as in the prior 6 years, even though welfare
reform limited the number of meals that some sponsors could claim for
reimbursements.
NUMBER OF CHILDREN
PARTICIPATING CONTINUED TO
INCREASE AFTER THE REDUCED
REIMBURSEMENTS
---------------------------------------------------------- Letter :4.1
In the year following the implementation of the reduced rates, the
number of children participating in the program increased overall, as
it did in all but one of the prior 6 years. According to the latest
available USDA data, the average number of children served daily in
July for the entire program increased by 2.3 percent, from 2,215,625
in 1996 to 2,266,319 in 1997.\11 Figure 2 shows the rate of change in
children's participation from 1991 to 1997.
Figure 2: Rate of Change in
the Average Number of Children
Served Daily, 1991-97
(See figure in printed
edition.)
Note: According to a USDA official, the reduction in the number of
children served in 1995 was due in part to active and potential
sponsors' anticipation of major program changes, including a possible
reduction in the meal reimbursement rates.
Source: GAO's analysis of USDA's data.
When we used the states' data to determine which types of sponsors
contributed to the growth in the number of children served after the
rate decrease in 1997, we found that existing sponsors had expanded
their programs while new sponsors in 1997 were smaller than the
sponsors that left the program after 1996. Sponsors that
participated in both fiscal years 1996 and 1997 served 8 percent more
children, on average, in 1997 than in 1996. On the other hand, new
1997 sponsors did not serve as many children, on average, as sponsors
that left the program after 1996. Specifically, new 1997 sponsors
served 19 percent fewer children, on average, in 1997, than were
served in 1996 by sponsors that left the program.
--------------------
\11 USDA's data for the 1998 program were not available at the time
of our review. Our survey data show that approximately 340 sponsors
that participated in 1997 were not active in July. These sponsors
accounted for approximately 2 percent of the total meals served in
1997.
SOME CHILDREN LOST ACCESS TO
THE PROGRAM
---------------------------------------------------------- Letter :4.2
Despite the overall increase in the number of children served between
1996 and 1997, some children lost access to program benefits because
the sponsors serving them dropped out, and the children were not
served by other sponsors. These fluctuations in sponsors'
participation and in children's access to the program occur even in
years when the program's rules and reimbursements do not change,
according to USDA officials.
The 370 sponsors that participated in 1996 and did not return in 1997
served approximately 118,224 children per day in July 1996.
According to our analysis of the information provided by state
officials, at least 17,238, or about 15 percent, of these children
lost access to the program in 1997. Sponsors that dropped out of the
program specifically because of the reduction had served
approximately 4,559 children in July 1996. At least 823, or 18
percent, of these children did not have access to the program in
1997. These 823 children represented about .03 percent of all
children who participated in the program in 1996. (See app. II.)
Other children may have lost access when continuing sponsors reduced
the number of sites they operated because of the rate reduction.\12
In fiscal year 1998, following the loss of 305 sponsors, at least
17,983 children, or 31 percent of the 58,562 children that had been
served by these sponsors, lost access to the program. Of the 2,926
children served by sponsors that left the program specifically
because of the reduction, at least 780 (27 percent) did not have
access to the program through another sponsor in 1998. These
children represent about .03 percent of all children who participated
in the program in 1997. Appendix III provides more detailed
information on our estimates of the number of children who had been
served by sponsors that dropped out of the program and who were
served by other sponsors the following year.
--------------------
\12 437 sponsors operated fewer sites in 1997 than in 1996. We did
not obtain the reasons these sites were closed or the number of
children who lost access to the program as a result.
THE NUMBER OF MEALS SERVED
INCREASED SLIGHTLY
---------------------------------------------------------- Letter :4.3
The number of meals served by sponsors increased by over 2
percent--from over 125 million meals in fiscal year 1996 to over 128
million meals in 1997--the year after the reimbursements were
reduced.\13 The number of meals had increased during each of the
previous 6 years reviewed.
The increase in meals served in fiscal year 1997 would probably have
been larger without the welfare reform change that affected camps and
sponsors that primarily serve migrant children. As mentioned
earlier, the 1996 act decreased the number of meals that these groups
could claim for reimbursements from four to three. According to many
state officials, if these sponsors had previously submitted four
meals for reimbursements--breakfast, lunch, supper, and a snack--they
did not submit the snack for reimbursement in 1997 because it has the
lowest reimbursement rate. Our analysis of USDA's data supports this
conclusion. From 1996 to 1997, the number of meals for which
sponsors requested reimbursements increased by only 2.4 percent.
However, excluding snacks, the number of meals for which sponsors
requested reimbursements increased by 4.7 percent. Figure 3 compares
the rate of change for total meals and meals not including snacks for
fiscal years 1991 to 1997.
Figure 3: Rate of Change in
All Meals and Meals, Excluding
Snacks, Fiscal Years 1991-97
(See figure in printed
edition.)
Note: The number of meals served, excluding snacks, increased at a
slower rate in 1995 than in any other year. According to a USDA
official, this occurred because some sponsors anticipated that major
changes in the program might include greatly reduced reimbursement
rates.
Source: GAO's analysis of USDA's data.
--------------------
\13 1997 is the latest year for which USDA has data on program meals.
SOME SPONSORS CHANGED THEIR
PROGRAMS BECAUSE OF THE REDUCED
REIMBURSEMENTS, AND SOME PAID
MORE OF THE PROGRAM COSTS
THEMSELVES
------------------------------------------------------------ Letter :5
Because sponsors received lower federal reimbursements for their
Summer Food Service Programs in 1997 and 1998, some adjusted their
programs to lower their costs, according to many officials
interviewed in our nationwide survey and the sponsors we visited.
Frequently reported changes were to lower meal, labor, and/or
location costs. Nevertheless, more sponsors reported operating costs
that exceeded their reimbursements in 1997 than in 1996.
SOME SPONSORS MADE PROGRAM
CHANGES TO LOWER COSTS
---------------------------------------------------------- Letter :5.1
In our telephone survey, officials in 24 states reported that there
were program changes resulting from the loss of federal funds that
were not related to the number of sponsors and children participating
in the program. Some sponsors also told us that they made changes to
mitigate the effects of the reduction. Following are frequently
mentioned changes and examples of how these changes were implemented:
-- Meal changes. The Maine program director reported that sponsors
in his state had to select their food more cautiously, favoring
less costly items. Wisconsin officials said sponsors provided
more prepackaged juices in place of fruit and vegetables to
decrease labor and food costs. In Hawaii, where the Department
of Education prepares the food for sponsors to distribute, an
official said that while the Department did not change the
entree or fruit/vegetable servings, it did serve smaller bread
and dessert portions to save money. The Georgia director said
that sponsors served less fresh fruit and did not offer
additional foods, such as desserts and chips, as often as in the
past.
-- Staff adjustments. Several sponsors we visited reported
decreasing their labor costs in 1997 and/or 1998. For example,
a school sponsor in South Carolina decreased its workers'
salaries from $7.25 per hour to $6.25 per hour. A school in
Texas did not pay cafeteria personnel the cost-of-living
adjustments that they had received during the school year. A
school in Oregon hired fewer, less experienced staff and reduced
their hours.
-- Fewer meal sites. According to Pennsylvania officials, sponsors
had to close or consolidate sites that were too expensive to
operate. The small sites that served only 15 to 20 children
could not afford to provide meals at the reduced rate.
California officials also reported that some sponsors closed
sites they could no longer afford to operate. For example, one
school sponsor closed three of its six sites because it could no
longer afford the labor costs. One of the largest sponsors in
Texas, a school district, closed almost all the sites that
served fewer than 50 children a day in 1997, and almost all the
sites that served fewer than 100 children a day in 1998.
However, the total number of children served by the sponsor has
increased since 1996.
MORE SPONSORS REPORT COSTS
HIGHER THAN THE
REIMBURSEMENTS IN 1997 THAN
BEFORE THE REDUCTION
---------------------------------------------------------- Letter :5.2
More sponsors submitted operating costs that exceeded their federal
reimbursements in fiscal year 1997 than in 1996. This means that
more sponsors are covering some of the cost of operating the program
with other funds. In 1996, prior to the rate decrease, 30 percent
(1,084) of sponsors were reimbursed for all their reported program
costs because these costs were equal to or lower than the maximum
reimbursements (total meals multiplied by the reimbursement rates).
The remaining 70 percent, or 2,565 sponsors, reported costs exceeding
their federal reimbursements. In 1997, fewer sponsors--25 percent
(966)--were reimbursed for their reported costs, while more
sponsors--75 percent (2,831)--reported costs that exceeded their
reimbursements. These totals include sponsors that dropped out of
the program after 1996 and new sponsors in 1997.\14 The limited
impact on the number of sponsors, children, and meals served that has
been observed to date is due in part to sponsors' continuing to
contribute funds to offset the decreased reimbursements. Appendix IV
provides more details on the type of sponsors that reported operating
costs in excess of their reimbursements.
--------------------
\14 These totals do not include 104 sponsors in 1996 and 78 sponsors
in 1997 because the state-provided data show that these sponsors
received more in reimbursements than the costs they reported.
AGENCY COMMENTS
------------------------------------------------------------ Letter :6
We provided USDA with copies of a draft of this report for review and
comment. We met with Food and Nutrition Service officials, including
the Branch Chief for Program Analysis and Monitoring Branch, Child
Nutrition Division, who generally agreed with the report's findings
and provided us wih a number of technical comments that we
incorporated into the report as appropriate.
---------------------------------------------------------- Letter :6.1
We are sending copies of this report to appropriate congressional
committees, interested Members of Congress, the Secretary of
Agriculture, and other interested parties. We will also make copies
available upon request.
If you have any questions about this report, please call me at (202)
512-5138. Major contributors to this report are listed in appendix
V.
Sincerely yours,
Lawrence J. Dyckman
Director, Food and
Agriculture Issues
OBJECTIVES, SCOPE AND METHODOLOGY
=========================================================== Appendix I
Because of questions raised about the impact that the reduction in
the meal reimbursement rate might have had on the Summer Food Service
Program, the Chairman of the House Committee on Education and the
Workforce asked us to report on (1) the number and characteristics of
sponsors participating in and dropping out of the program before and
after the decrease in reimbursements, (2) the number of children and
meals served by the program before and after the reduction, and (3)
the changes sponsors made to their program as a result of the reduced
reimbursements.
To address these objectives, we conducted three nationwide surveys
with officials from the 50 states.\1 First, we conducted a telephone
survey to gather information and views on the impact of the rate
reduction on sponsors, children, and meals.
Second, we sent a mail survey to officials in the 50 states to
collect detailed fiscal year 1996 and 1997 data on their sponsors.
The survey covered the number of meals, sites, July's average daily
attendance, and reported operating costs and reimbursements. State
officials were also asked to identify (1) sponsors that dropped out
of the program after 1996, (2) the major reason for dropping out, (3)
the number of sites picked up by another sponsor, and (4) the portion
of the children served by a new sponsor.
For states that did not identify the sponsors that dropped out, we
identified these sponsors by using financial and other data provided
by the states. Two states did not provide average daily attendance
for both years, and several provided different measures for
attendance.\2 In March 1998, we testified on the preliminary results
of these surveys.\3
Third, we sent a mail survey to the 50 states to collect their 1998
data. This third survey was aimed primarily at collecting
information on the (1) number of sponsors that dropped out of the
program after 1997, (2) reasons they dropped out, and (3) number of
children who lost access to the program. To estimate the number of
children who lost access to the program, we assigned percentages to
the various response categories. For example, if state officials
reported that "all or almost all" of participants were picked up by
another sponsor, then we estimated that 95 percent of the
participants had program access through another sponsor and 5 percent
lost access to the program.\4
We combined the 1996, 1997, and 1998 data provided by the states to
provide a picture of welfare reform to date. For example, we
determined the percentage of sponsors that reported costs higher than
their reimbursements prior to and after the rate reductions.
Although we did not verify the accuracy of all the data provided by
the states for each of the over 4,000 sponsors, we did conduct a
variety of tests to verify the data's internal consistency. For
example, we ensured that for sponsors categorized as dropping out
after fiscal year 1996 there were no 1997 reimbursement data. We
also identified any sponsors for which the data provided were outside
of expected ranges. Where possible, we contacted the states to
correct obvious errors. For a small percentage of sponsors, the
state-provided information suggests the sponsors may have been
overpaid. For example, some sponsors received reimbursements in
excess of the costs they reported. A listing of these sponsors and
their data will be provided to USDA for further review. We also
compared our state data on the number of meals served in 1996 and
1997 with USDA's data on the number of meals served for 1996 and
1997. Our total was generally within 5 percent of USDA's total.\5
To determine the changes sponsors made to their program because of
the reduced rates, we selected 20 sponsors to study in greater
detail. These sponsors--in California, Illinois, Oregon,
Pennsylvania, South Carolina, and Texas--were selected to represent a
variety of sponsor sizes and types (e.g., school, nonprofit, and
government) in both rural and urban areas. We conducted in-depth,
in-person interviews with these sponsors to determine the changes
they had made to their programs in response to rate reductions. We
observed the operation of the program--including meals being prepared
by vendors, transported to sites, and eaten by children at 17 sites.
In addition to conducting these in-depth studies and three surveys,
we interviewed officials from USDA's Food and Nutrition Service, the
American School Food Service Association, and the Food Research and
Action Center. We also reviewed related legislation, USDA
documents--such as program guidance and budget data--and studies
conducted by associations and states.
We conducted our review from October 1997 through October 1998 in
accordance with generally accepted government auditing standards.
--------------------
\1 The District of Columbia participated in the telephone survey but
did not provide sufficient information requested in the other surveys
so that it could be included in our analysis.
\2 New Jersey did not provide any data for average daily attendance
in July for 1996 or 1997. Arkansas did not provide average daily
attendance in July for 1997. Delaware (1996 only), Florida, and
Oregon provided average daily attendance for the year. Mississippi
provided attendance for June; Colorado provided average daily
participation, which is different from average daily attendance.
Illinois provided estimates of average daily participation; Texas
provided a participation figure reflecting the sponsors' week of
highest attendance.
\3 Welfare Reform: Effects of Changes Made to the Summer Food
Service Program, (GAO/T-RCED-98-120, Mar. 10, 1998).
\4 If "more than half" was reported as picked up by another sponsor,
we estimated that 75 percent had program access and 25 percent did
not. Our estimates for other categories' access were (1) "about
half"--50 percent had access and 50 percent did not; (2) "less than
half"--25 percent had access and 75 percent did not; and (3) "few, if
any"--5 percent had access and 95 percent did not.
\5 For the following states, our data for total meals were more than
5 percent different from USDA's data for 1 or both years: Alaska,
Hawaii, North Dakota, New Hampshire, Oklahoma, and Vermont.
INFORMATION ON THE DROPOUT RATES
FOR VARIOUS TYPES OF SPONSORS IN
FISCAL YEARS 1996 AND 1997
========================================================== Appendix II
This appendix provides information for 1996 and 1997 on the total
number of sponsors and the percent of sponsors that dropped out of
the program in terms of the (1) number of children served, (2) number
of meals served, and (3) type of sponsor.
Table II.1
Number and Percent of Sponsors That
Dropped Out of the Program by Number of
Children Served in July, Fiscal Years
1996 and 1997
Fiscal year 1996 Fiscal year 1997
---------------------------------- ----------------------------------
Average
number
of
children Percent of 1996 Percent of 1997
served sponsors that sponsors that
daily in All 1996 dropped out in All 1997 dropped out in
July sponsors 1997 sponsors 1998
-------- ---------------- ---------------- ---------------- ----------------
None-- 259 15 341 12
no July
program
1 -75 909 16 971 12
76 -299 1,192 9 1,242 8
300 - 732 6 770 5
999
1,000 or 418 5 449 2
more
--------------------------------------------------------------------------------
Note: 1996 and 1997 sponsors in New Jersey and 1996 sponsors in
Arkansas are excluded from this table because data were not provided.
Table II.2
Number and Percent of Sponsors That
Dropped Out of the Program by Number of
Meals Served, Fiscal Years 1996 and 1997
Fiscal year 1996 Fiscal year 1997
---------------------------------- ----------------------------------
Percent of 1996 Percent of 1997
Number sponsors that sponsors that
of meals All 1996 dropped out in All 1997 dropped out in
served sponsors 1997 sponsors 1998
-------- ---------------- ---------------- ---------------- ----------------
Fewer 952 17 1,035 15
than
3,500
meals
3,500 to 1,208 10 1,211 8
9,999
meals
10,000 933 6 964 3
to
29,999
meals
30,000 658 5 663 4
or more
meals
--------------------------------------------------------------------------------
Table II.3
Number of Sponsors and Percent of
Sponsors That Dropped Out of the Program
by Type of Sponsor, Fiscal Years 1996
and 1997
Fiscal year 1996 Fiscal year 1997
----------------- ---------------------------------- ----------------------------------
Percent of 1996 Percent of 1997
sponsors that sponsors that
All 1996 dropped out in All 1997 dropped out in
Sponsor type sponsors 1997 sponsors 1998
----------------- ---------------- ---------------- ---------------- ----------------
School 1,665 9 1,747 6
Government 670 10 662 7
Private nonprofit 572 19 600 16
Camp 728 5 743 6
National Youth 117 3 123 4
Sports Program
-----------------------------------------------------------------------------------------
THE NUMBER OF CHILDREN SERVED BY
SPONSORS THAT DROPPED OUT OF THE
PROGRAM
========================================================= Appendix III
This appendix provides information on the number of children served
by sponsors that dropped out of the program who did and did not have
access to the program through other sponsors.
Table III.1
Estimates of the Number of Children
Served by 1996 Sponsors That Did Not
Continue in 1997 Because of the Rate
Reductions and other Reasons and of the
Number of Those Children Who Did and Did
Not Have Access to the Program in 1997
Children served in 1996 by sponsors that did not continue in 1997
------------------------------------------------------------------------------
Had access to the
1997 program Did not have
through another access to the 1997 1997 access was
Total sponsor program unknown
------------------ ------------------ ------------------ ------------------
Reasons
sponsors
dropped
out of
program Number Percent Number Percent Number Percent Number Percent
--------- -------- -------- -------- -------- -------- -------- -------- --------
Rate 4,559 100 2,621 57 823 18 1,115 24
reductio
ns
Other 68,700 100 31,448 46 15,591 23 21,661 31
known
reasons\
a
Unknown 44,965 100 190 <1 823 2 43,952 98
All 118,224 100 34,258 29 17,238 15 66,728 56
reasons
-----------------------------------------------------------------------------------------
\a These other known reasons include low participation, personnel
loss, and construction.
Notes: Numbers may not add due to rounding.
For each sponsor that dropped out, state officials provided an
estimate of the portion of participants served by the sponsor who
were served by other sponsors in the next year. We assigned the
following percentages to the various response categories: (1) "all
or almost all"--95 percent had access and 5 percent did not; (2)
"more than half"--75 percent had access and 25 percent did not;
(3)"about half"--50 percent had access and 50 percent did not; (4)
"less than half"--25 percent had access and 75 percent did not; and
(5) "few, if any"--5 percent had access and 95 percent did not.
In 1996, 38 of the dropouts did not operate in July. An additional
19 sponsors were in New Jersey or Arkansas, which did not provide
data on the number of children served. These 57 sponsors served
336,744 meals in 1996, or 6 percent of all meals served by dropouts.
Two of the 24 sponsors that left because of the rate reduction did
not operate in July.
Table III.2
Estimates of the Number of Children
Served by 1997 Sponsors That Did Not
Continue in 1998 Because of the Rate
Reductions and Other Reasons and of the
Number of Those Children Who Did and Did
Not Have Access to the Program in 1998
Children served in 1997 by sponsors that did not continue in 1998
------------------------------------------------------------------------------
Had access to the
1998 program Did not have
through another access to the 1998 1998 access was
Total sponsor program unknown
------------------ ------------------ ------------------ ------------------
Re
as
on
s
sp
on
so
rs
dr
op
pe
d
ou
t
of
pr
og
ra
m Number Percent Number Percent Number Percent Number Percent
-- -------- -------- -------- -------- -------- -------- -------- --------
Ra 2,926 100 1,963 67 780 27 183 6
te
r
e
d
u
c
t
i
o
n
s
Ot 47,243 100 25,807 55 15,507 33 5,928 13
h
e
r
k
n
o
w
n
r
e
a
s
o
n
s
\
a
Un 8,393 100 4,183 50 1,695 20 2,515 30
k
n
o
w
n
Al 58,562 100 31,954 55 17,983 31 8,626 15
l
r
e
a
s
o
n
s
----------------------------------------------------------------------------------
\a These other reasons include low participation, personnel loss and
construction.
Notes: Numbers may not add due to rounding.
For each sponsor that dropped out, state officials provided an
estimate of the portion of participants served by the sponsor that
was served by other sponsors in the next year. We assigned the
following percentages to the various response categories: (1) "all
or almost all"-95 percent had access and 5 percent did not; (2) "more
than half"-75 percent had access and 25 percent did not; (3)"about
half"--50 percent had access and 50 percent did not; (4) "less than
half"--25 percent had access and 75 percent did not; and (5) "few, if
any"--5 percent had access and 95 percent did not.
Some programs did not operate in July and are not included in this
table. In 1997, 42 of the total drop-outs did not operate in July.
An additional four sponsors were in New Jersey, which did not provide
data on the number of children served. These 46 sponsors served
152,038 meals in 1997 or 5 percent of all meals served by drop-outs.
All sponsors that left due to the rate reduction operated in July.
SPONSORS THAT REPORTED PROGRAM
COSTS EXCEEDING THE MAXIMUM
FEDERAL REIMBURSEMENTS
========================================================== Appendix IV
This appendix provides information, for fiscal years 1996 and 1997,
on sponsors reporting costs that exceeded the maximum federal
reimbursements, by participation category and by type of sponsor.
Table IV.1
Sponsors Reporting Program Costs That
Exceeded the Maximum Federal
Reimbursements, by Participation
Category, Fiscal Years 1996 and 1997
Percent of sponsors that
reported costs exceeding
the maximum reimbursements
--------------------------
Number of Fiscal year Fiscal year
Participation category sponsors 1996 1997
---------------------------- ------------ ------------ ------------
Dropped out between 1996 and 370 76 N/A
1997
Dropped out between 1997 and 305 46\a 76
1998
Participated in 1996, 1997, 3,190 67 73
and 1998
----------------------------------------------------------------------
\a Of the 305 sponsors that dropped out between 1997 and 1998, 3
percent did not participate in the program in 1996.
Table IV.2
Sponsors That Reported Program Costs
Exceeding the Maximum Federal
Reimbursements, by Type of Sponsor,
Fiscal Years 1996 and 1997
Percent of sponsors reporting costs
exceeding the maximum federal
reimbursements
--------------------------------------
Sponsor type Fiscal year 1996 Fiscal year 1997
------------------------------ ------------------ ------------------
School 62 66
Government 58 66
Private nonprofit 69 74
Camp 91 95
National Youth Sports Program 68 75
----------------------------------------------------------------------
MAJOR CONTRIBUTORS TO THIS REPORT
=========================================================== Appendix V
Thomas Slomba, Assistant Director
Leigh McCaskill White, Evaluator-in-Charge
Peter Bramble, Senior Evaluator
Fran Featherston, Senior Social Science Analyst
Alice Feldesman, Supervisory Social Science Analyst
Beverly Peterson, Senior Evaluator
Carol Herrnstadt Shulman, Communications Analyst
John W. Shumann, Evaluator
*** End of document. ***