This is the accessible text file for GAO report number GAO-02-328 
entitled 'Foreign Assistance: Global Food for Education Initiative 
Faces Challenges for Successful Implementation' which was released on 
February 28, 2002. 

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

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

United States General Accounting Office: 
GAO: 

Report to Congressional Requesters: 

February 2002: 

Foreign Assistance: 

Global Food for Education Initiative Faces Challenges for Successful
Implementation: 

GAO-02-328: 

Contents: 

Letter: 

Results in Brief: 

Background: 

Lessons Learned from School Feeding Programs Define Conditions for 
Likely Success: 

Pilot Program Did Not Adequately Incorporate Lessons Learned: 

Weaknesses in Structure, Planning, and Management Reduce Chances for 
Pilot Program Success: 

Most Other Donors Currently Uncommitted or Opposed to Major Support of 
GFEI: 

Conclusions: 

Matters for Congressional Consideration: 

Agency Comments and Our Evaluation: 

Appendixes: 

Appendix I: Scope and Methodology: 

Appendix II: Pilot Program Projects' Implementing Partners, Countries, 
Agreement Status, Tonnage, Cost, and Beneficiaries: 

Appendix III: The World Food Program's Role in School Feeding and Food 
for Education: 

Appendix IV: Results from Review of Experts' Findings and Views on 
School Feeding Programs: 
References: 

Appendix V: Costs of School Feeding Programs: 

Appendix VI: Process to Solicit, Evaluate, and Approve Proposals for 
the Pilot Program: 

Appendix VII: Selected Information Contained in Proposals for Approved 
School Feeding Programs: 

Appendix VIII: Donor Views on Uses of Food Aid And How It Is Provided: 

Appendix IX: Sources USDA Uses to Finance Its Implementing Partners' 
GFEI Project Costs: 

Appendix X: Top Food Aid Donating Countries: 

Appendix XI: Key GFEI Events from Announcement of Concept to 
Notification of Project Approvals: 

Appendix XII: Comments from the U.S. Department of Agriculture: 
GAO Comments: 

Appendix XIII: Comments from the U.S. Agency for International 
Development: 

Appendix XIV: Comments from the Office of Management and Budget: 

Appendix XV: GAO Contacts and Staff Acknowledgments: 
GAO Contacts: 
Acknowledgments: 

Tables: 

Table 1: Presence or Absence of Key Factors in USDA's Request for 
Proposals and in Written Criteria for Evaluating Proposals: 

Table 2: Number of Approved PVO and WFP Proposals that Addressed 
Various Key Factors: 

Table 3: Other Donors' Overall Views on the Food for Education
Initiatives: 

Table 4: Results From Selected Studies and Experts on the Impacts
of School Feeding Programs on Enrollment and Attendance: 

Table 5: Results From Selected Studies and Experts on the Impacts
of School Feeding Programs on Learning: 

Table 6: Targeting Factors and School Feeding Program Effectiveness: 

Table 7: Learning Environment Factors and School Feeding Program 
Effectiveness: 

Table 8: Health and Nutrition Factors and School Feeding Program
Effectiveness: 

Table 9: Community and Parental Factors and School Feeding Program 
Effectiveness: 

Table 10: Government Commitment and Sustainability and School Feeding 
Program Effectiveness: 

Table 11: Actual Costs of Various School Feeding Programs (Year 2000 
Dollars): 

Table 12: Estimated Returns on Alternative Nutrition Interventions: 

Table 13: Expert Views on Top 10 Educational Interventions for Latin 
America: 

Table 14: Budgeted Amounts for Nonmeal Program Components Included in 
PVO Proposals Approved by USDA: 

Table 15: WFP Projects' Funding for Nonmeal Program Components: 

Table 16: Number of Approved PVO Proposals With Information on Other 
Donors: 

Table 17: Donor Views on Uses of Food Aid and How It Is Provided: 

Table 18: Top Food Aid Donors Based on Shipments, 1995-1999: 

Table 19: Key GFEI Events from Announcement of Concept to Project 
Approvals: 

Abbreviations: 

CCC: Commodity Credit Corporation: 

CSSD: Consultative Sub-Committee on Surplus Disposal: 

FAO: Food and Agriculture Organization of the United Nations: 

FAPC: Food Assistance Policy Council: 

FAS: Foreign Agricultural Service: 

FNS: Food and Nutrition Service: 

GFEI: Global Food for Education Initiative: 

IFPRI: International Food Policy Research Institute: 

IQ: Intelligence quotient: 

ITSH: internal transportation, storage, and handling: 

LDC: least developed country: 

MCH: Maternal and child health: 

NA: not available: 

NFDM nonfat dry milk: 

PSP: parent school partnership: 

PTA: parent-teacher association: 

PVO: private voluntary organization: 

SFP: school feeding program: 

UMR: Usual Marketing Requirements: 

U.N.: United Nations: 

UNICEF: United Nations Children's Fund: 

UNESCO: United Nations Educational, Scientific, and Cultural 
Organization: 

USAID: U.S. Agency for International Development: 

USDA: U.S. Department of Agriculture: 

VAM: vulnerability analysis and mapping unit: 

WFP: World Food Program: 

WHO: World Health Organization: 

[End of section] 

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

February 28, 2002: 

The Honorable Tom Harkin: 
Chairman: 
The Honorable Richard G. Lugar: 
Ranking Minority Member: 
Committee on Agriculture, Nutrition and Forestry: 
United States Senate: 

The Honorable Richard J. Durbin: 
The Honorable Patrick Leahy: 
United States Senate: 

The Honorable Larry Combest: 
Chairman: 
The Honorable Charles W. Stenholm: 
Ranking Minority Member: 
Committee on Agriculture: 
House of Representatives: 

At the July 2000 Group of Eight industrialized countries' summit in 
Okinawa, Japan, President Clinton proposed a Global Food for Education 
Initiative (GFEI) whereby developed countries would provide school 
breakfasts or lunches to needy children in poor countries. The overall 
goal of the initiative is to contribute to universal education for all 
by using school meals to attract children to school, keep them 
attending once they enroll, and improve learning. An estimated 300 
million children in developing countries are chronically 
undernourished, and many of them are among an estimated 120 million 
who do not now attend school. At the same time, the president 
announced a 1-year, $300-million pilot food for education program to 
be administered by the U.S. Department of Agriculture (USDA)[Footnote 
1] to jump-start the proposed global effort. The pilot's objectives 
are to use school meals to improve student enrollment, attendance, and 
performance. Under the program, the United States is using the 
authority of section 416(b) of the Agricultural Act of 1949 to donate 
surplus agricultural commodities from the Commodity Credit 
Corporation's (CCC)[Footnote 2] inventory for use in existing as well 
as new and expanded school feeding and preschool nutrition projects in 
developing countries. Congress is currently considering whether to 
provide additional funding for the pilot program and/or establish a 
permanent program.[Footnote 3] 

As you requested, we examined (1) lessons that can be drawn from 
existing research and expert views on the effectiveness and cost of 
school feeding programs in promoting increased school enrollment, 
attendance, and performance; (2) the extent to which the U.S. pilot 
program has built upon these lessons to date; (3) whether the U.S. 
pilot program is being operated and managed so as to reasonably ensure 
that the food aid and monetized proceeds are effectively and 
efficiently used; and (4) the views of other major donors regarding 
support for a comprehensive, long-term global food for education 
[Footnote 4] initiative. 

As you also requested, we are providing our analysis in advance of the 
pilot program's completion. The first meals of the pilot were not 
delivered until fall 2001, and USDA expects the program to carry over 
into 2003. As a result, our observations on the pilot program concern 
its design and implementation through December 2001 and do not assess 
the in-country phase of the program. To address the issues outlined 
above we met with and reviewed information from U.S. government 
officials at the Departments of Agriculture and State, as well as the 
U.S. Agency for International Development (USAID), the Office of 
Management and Budget, and the White House. We also gathered 
information from and met with officials of the U.N. World Food Program 
(WFP), some private voluntary organizations, foreign donor 
governments, the European Commission, the World Bank, and private 
research institutions. Appendix I provides detailed information on our 
scope and methodology. 

Results in Brief: 

Research and expert views on the effectiveness of school feeding 
programs indicate that the programs are more likely to have positive 
results when they are carefully targeted and integrated with other 
educational, health, and nutritional interventions. To be effective, 
programs need to be targeted at relatively poor areas where enrollment 
and attendance rates are low and where the value of the food is a 
sufficient incentive to attract children to school. However, actual 
learning requires a facilitative environment that includes enough 
adequately trained teachers, good texts and other learning materials, 
and adequate physical facilities. Other important factors that 
contribute to the effectiveness of school feeding programs include 
interventions that focus on micronutrient[Footnote 5] deficiencies and 
clean water and sanitation facilities. At the same time, school 
feeding programs are costly in terms of both the dollars required to 
fund them and the human resources needed to operate them. As a result, 
school feeding programs may not be cost effective when compared with 
alternative interventions such as providing quality teaching and 
offering nutritional and health packages directed at pregnant women 
and at mothers with their preschool children. 

In designing and setting up the pilot program, USDA did not build on 
some important lessons from previous school feeding programs. For 
example, when USDA solicited proposals for the program, it did not 
require information on several important factors linked to effective 
food for education projects-—although in some cases the sponsors 
supplied some of this information anyway. USDA asked private voluntary 
organizations—but not WFP—-for specific information on whether their 
proposed projects targeted the right communities or populations, but 
only to a limited extent. Information on many other key contributors 
to success, such as whether schools had good learning environments and 
safe and adequate on-site water and sanitation facilities, was not 
required of either the Private Voluntary Organizations (PVO) or WFP, 
in part because of the program's quick start-up and short duration, as 
well as concerns that costly information requirements might discourage 
potential sponsors from applying. Further, the written criteria for 
evaluating proposals did not focus on many of these factors. In 
addition, USDA provided little funding for these and other nonmeal 
components of school feeding programs, which are essential elements in 
effective food for education projects. 

While USDA expects more than 8 million children to benefit from the 
pilot program, we found that the pilot's structure, planning, and 
management thus far do not reasonably ensure that the program's 
objectives of increasing enrollment, attendance, and learning will be 
attained. The administration's decision to use surplus commodities to 
fund the program was an expedient way to get it started quickly but 
may not be sustainable. The selection of USDA to manage the program 
raises concerns because USDA does not have the experience and 
resources for managing food for education development programs. USDA 
officials told us they were under pressure to get the program up and 
running, had little time for planning and consideration of the human 
capital necessary to run the pilot successfully, and had insufficient 
resources to fully address the educational components of school 
feeding. In addition, USDA initiated the pilot without a fully 
developed strategy for monitoring and evaluating performance; and, 
because of the pilot's short duration, USDA says it will not be able 
to monitor and evaluate one of the program's three objectives—
improvements in learning. WFP only recently completed collection of 
relevant baseline data on enrollment and attendance, and USDA is still 
in the process of collecting such information. Other weaknesses in 
project performance data and financial accountability may make it 
difficult to draw clear conclusions about the pilot's effectiveness 
when the program is completed. 

Representatives of most other donor countries that we interviewed 
[Footnote 6] indicated their governments were either noncommittal 
about—-or unwilling to provide—-substantial support for a 
comprehensive, long-term food for education program. This lack of 
support is problematic because the United States envisioned a 
multilateral program with other donors funding about three-quarters of 
the program's total cost. The European Commission and several other 
nations are generally opposed to using food aid for development, 
saying sustainable development assistance requires programs that are 
integrated across a variety of sectors. Several donor country 
representatives said the pilot program seems principally designed to 
dispose of surplus commodities and questioned the sustainability of a 
program that depends on agricultural surpluses. Overall, GFEI seems 
unlikely to attract much support from other donors unless the United 
States adopts a permanent program that is not dependent on surplus 
agricultural commodities and/or unless the pilot program demonstrates 
strong, positive results. 

In this report, we provide matters that the Congress may wish to 
consider as it contemplates legislation on a food for education 
program. 

USDA, in commenting on a draft of this report, said it believes we 
took an overly critical view of how it administered the pilot program 
given time and resource constraints. The Office of Management and 
Budget, the Department of State, and USAID indicated the report's 
findings are essentially accurate. USAID endorsed our matters for 
congressional consideration. We also received technical comments on 
portions of the report from the WFP, six PVOs,[Footnote 7] and the 
World Bank and incorporated changes as appropriate. 

At Jomtien, Thailand, in March 1990, representatives of the global 
education community held the "World Conference on Education for All" 
and adopted a declaration on universal access to education as a 
fundamental right of all people. In April 2000, the "World Education 
Forum"[Footnote 8] met in Dakar, Senegal. Delegates from 181 nations 
adopted a framework for action committing their governments to achieve 
quality basic education for all—-including ensuring that by 2015, all 
children—-especially girls, children in difficult circumstances, and 
those from ethnic minorities-—have access to completely free primary 
education of good quality. 

Also in early 2000, the U.S. ambassador to the U.N. Food Agencies in 
Rome proposed that the United States, within the U.N. framework, take 
the lead in organizing a worldwide school lunch program.[Footnote 9] 
The purpose would be to provide a meal every day for every needy child 
in the world. Doing so, the ambassador said, would attract children to 
school and keep them there under conditions in which they are able to 
learn and grow. The United States would pay 25 percent of the cost, 
and other donor nations would pay the rest. The United States would 
benefit, since Americans produce more food than they can eat or 
profitably sell and since most of the U.S. contribution would be in 
the form of agricultural commodities and thus would strengthen the 
market for cereal grain, dairy products, and livestock. According to 
the ambassador, other farm surplus countries such as France, Canada, 
and Australia would benefit as well. 

In late May 2000, President Clinton met with the ambassador to discuss 
the idea and asked senior advisers to prepare an analysis of how the 
United States might participate. In early July 2000, the advisers 
reported that all relevant agencies recommended that the president 
announce a U.S. pilot program to support the international community's 
goal of achieving universal access to basic education by 2015 and the 
U.N.'s 10-year "Girls' Education Initiative" to help poor countries 
eliminate gender disparities in educational access. The advisers 
recommended spending approximately $300 million in the first year on 
the pilot program, with levels in subsequent years dependent upon 
factors such as the extent of international participation and the 
continued availability of CCC funding. At the Okinawa Summit on July 
23, 2000, the president announced the Global Food for Education 
Initiative and the pilot program. 

According to the White House press release, which was issued the day 
the program was announced, the purpose of the pilot program is to 
improve student enrollment, attendance, and performance in poor 
countries. These objectives were reaffirmed in USDA's September 2000 
request for proposals from cooperating sponsors and, more recently, in 
a December 2001 paper describing the goals, scope, and framework for 
action for monitoring and evaluating the pilot program. 

For the pilot, USDA sought proposals from various potential 
implementing partners, and approved 53 projects[Footnote 10] in 38 
countries covering an estimated 8.3 million children. Partners include 
WFP[Footnote 11] and various cooperating sponsors.[Footnote 12] Among 
the latter are 13 PVOs and 1 foreign government (Dominican Republic). 
As of mid-December 2001, USDA had finalized agreements for 21 of 25 
PVO projects, 26 of 27 WFP projects,[Footnote 13] and 1 project with 
the Dominican Republic. The recent total estimated cost for all of the 
pilot projects was $227.7 million, allocated as follows: WFP projects, 
$92.5 million; PVO projects, $121.1 million; and the government of the 
Dominican Republic, $14.1 million. The total cost is $72.3 million 
less than the originally planned $300 million initiative.[Footnote 14] 
According to USDA officials, the balance will be used in fiscal year 
2002 to expand existing projects that show the most potential, based 
on performance. Appendix II provides more detailed program and cost 
information. 

Lessons Learned from School Feeding Programs Define Conditions for 
Likely Success: 

Research and expert views on school feeding programs indicate that 
these programs are more likely to have positive results when they are 
carefully targeted and integrated with other educational, health, and 
nutritional interventions. There is considerable evidence that school 
feeding programs can increase enrollment and attendance if the 
programs are targeted at the right communities or populations. 
Evidence of the effectiveness of school feeding programs in improving 
learning is somewhat more mixed, possibly because of difficulties 
isolating factors associated with increased learning, the quality of 
studies assessing such relationships, or the quality and settings of 
such programs. Programs are more likely to have a positive result on 
enrollment, attendance, and learning when they are integrated with a 
facilitative learning environment and appropriate health and 
nutritional interventions. Community participation and parental 
involvement also promote these objectives. Taking steps to ensure that 
programs will be sustainable when donor assistance is no longer 
available is important for ensuring long-term effectiveness. At the 
same time, school feeding programs are costly and may not be cost 
effective, relative to other possible interventions. (Appendices IV 
and V provide results from selected studies on these issues.) 

Targeting the Right Population Can Improve Enrollment and Attendance: 

Evidence indicates that school feeding programs can improve school 
enrollment and attendance if they target the right population. In 
general, studies and experts point to the importance of targeting 
programs on low-income communities that lack a secure supply of food 
and have relatively low rates of school enrollment and attendance. 
When school feeding programs do improve enrollment and attendance, 
their contribution is primarily through a transfer of income (the 
food) to families.[Footnote 15] School feeding programs may not have 
much of an impact if children are staying away because the distance to 
school is too far to walk, parents perceive the quality of the 
education to be low, or children are disabled.[Footnote 16] Providing 
national coverage to all children is usually not cost effective. 
Targeting high-risk communities is preferable to targeting individual 
children within schools, which could lead to competition among 
students and parents, dilution of nutritional impact through food 
sharing, and insufficient community support. (See appendix IV for 
results from selected studies on the use of targeting to improve the 
effectiveness of school feeding programs.) 

According to several experts and practitioners, school feeding 
programs can also help reduce the educational gender gap—where the 
proportion of school-age boys attending school significantly exceeds 
that for school-age girls. Many studies have shown that the inability 
of households to cover direct and indirect costs of education results 
in fewer girls attending school. This inequity exists partly because 
parents perceive less value in educating girls, there is greater 
demand for girls' labor at home, and girls are more affected by issues 
of school location and security. Yet some of the highest returns to 
education and other development investments derive from girls' 
education. For example, according to studies cited by WFP: 

* Illiterate girls have an average of six children each while girls 
who go to school average 2.9 children; 

* Infants born to mothers with no formal education are twice as likely 
to die before their first birthday than are babies born to mothers 
with a post-primary school education; 

* Between 1970 and 1995, 44 percent of the decrease in child 
malnutrition was attributable to improvements in female education; and; 

* Educated mothers are more likely to send their own daughters to 
school.[Footnote 17] 

To increase educational opportunities for girls, a "package" of 
strategies is often tailored to meet a country's special needs. These 
packages typically contain some combination of interventions to (1) 
reduce the opportunity costs of sending girls to school; (2) improve 
the quality and relevance of education; (3) increase access to close, 
safe schools equipped with basic infrastructure; (4) educate parents 
and communities about the benefits of girls' education; and (5) 
establish supportive national policies. 

Facilitative Environment Needed for Effective Learning: 

A group of experts and practitioners who convened at USAID 
headquarters in October 2000[Footnote 18] concluded that little 
learning is likely to occur without a facilitative learning 
environment, where teachers engage children in stimulating learning 
tasks, provide frequent feedback and encouragement, and are equipped 
with motivational textbooks and other learning materials.[Footnote 19] 
A facilitative learning environment also requires a suitable physical 
environment and minimal school supplies. Unfortunately, most schooling 
in the developing world is far from this kind of environment.[Footnote 
20] Teaching is frequently of poor quality and is poorly supported; 
and the curriculum often has little relevance to rural life, making 
formal schooling unconnected with the needs of rural communities. 
Thus, most developing countries require investments in teacher 
training; basic supplies (books, blackboards, desks, and chairs); a 
suitable physical environment; and other learning materials. 
Furthermore, many school systems in developing countries are 
dysfunctional, characterized by dispersed or displaced populations (as 
a result of conflict or natural calamities), limited basic 
infrastructure, and endemic child malnutrition.[Footnote 21] Many 
experts and practitioners also conclude that food for education 
programs must take place within the context of broad, national 
education reform programs that focus on essential inputs to education 
and learning, such as teacher development, curriculum reform, and 
student assessment.[Footnote 22] (See appendix IV for results from 
selected studies on the impacts that school feeding programs have on 
learning.) 

Nutritional and Health Measures Are Needed for Effective Programs: 

According to various studies,[Footnote 23] poor nutrition and health 
among schoolchildren contribute to diminished cognitive abilities that 
lead to reduced school performance. According to experts, school 
feeding programs can be effective in reducing short-term hunger 
[Footnote 24]—-which in turn can improve learning capacity-—by 
providing an inexpensive breakfast or small snack, shortly after 
students arrive at school. Meanwhile, using enriched foods or 
complementing commodities in school feeding programs with locally 
available vitamin and mineral-rich foods is an effective route to 
alleviating the complex micronutrient deficiencies that schoolchildren 
in developing countries often suffer. At the same time, school feeding 
programs designed to capture both nutritional and educational gains 
need to invest in adequate water and sanitation at schools, since poor 
water and sanitation give rise to infectious diseases, including 
parasites, which adversely affect school children's enrollment, 
attendance, and learning. These programs also benefit from inclusion 
of deworming treatments and health and nutrition education. (See 
appendix IV for results from selected studies on nutrition and health 
measures that can be used in combination with school feeding programs 
to improve school performance.) 

Community and Parental Involvement Also Can Contribute to Enrollment, 
Attendance, and Learning: 

Community and parental involvement are also important to successful 
school feeding programs. Community involvement in implementing school 
feeding programs[Footnote 25] can increase contact, and hence 
communication, between parents and teachers, officials, and others; 
provide parents an opportunity to become more aware of what goes on at 
schools; help raise the value of education and the school for parents 
and the whole community; and motivate parents to enroll their children 
in school and ensure regular attendance. Parent-teacher associations 
(PTA) or other outreach efforts can be used to educate parents and 
other community groups on issues such as the negative effects of 
temporary hunger on learning or the social and health benefits of 
educating girls. 

Strong Government Commitment Boosts Effectiveness of School Feeding: 

According to WFP, another important ingredient in successful school 
feeding programs is national government commitment to the goal of 
"education for all." This commitment should be put into practice 
through policies, programs, and financial commitments within a 
country's means that support basic education. 

Governments also need to commit to school feeding programs within the 
context of broad, national school reform programs, according to 
practitioners and experts who met at USAID in October 2000. These 
reforms should target essential inputs to education and learning, 
including teacher development, curriculum reform, and student 
assessment. 

Cost of School Feeding Programs Affects Sustainability: 

While the benefits of school feeding programs are recognized, the 
programs are expensive both financially and in terms of the human 
resources required to operate them. In addition to the price of the 
food, costs associated with food logistics, management, and control 
can represent a significant financial burden for recipient country 
governments.[Footnote 26] These costs may be difficult for national 
governments to absorb and thus adversely affect long-term program 
sustainability. 

Estimates of the average cost of school feeding programs vary 
considerably (see appendix V).[Footnote 27] According to WFP, the 
average cost per student of its development school feeding projects in 
2000 was 19 cents per day, or $34 for a 180-day school year (see 
appendix V). Programs costing $34 per pupil per school year are 
substantial when compared with what many developing countries spend on 
education. For example, in 1997 public expenditures of 19 least-
developed countries for both pre-primary and primary education 
averaged only $20 per pupil, according to UNESCO. Average public 
expenditures of five southern Asian countries were reported at $40 per 
pupil. 

According to many experts,[Footnote 28] national ministries of 
education in developing countries should not be encouraged to take on 
school feeding at the expense of other educational inputs. Few 
national governments are committed to school feeding programs over the 
long term, they said.[Footnote 29] In addition, many governments and 
education ministries already are struggling to maintain barely 
functioning education systems; may not be equipped, financially or 
technically, to assume the additional burden of food distribution; and 
do not have the financial resources to sustain feeding programs after 
donor support is withdrawn.[Footnote 30] These experts say that 
getting local communities involved from the beginning and giving them 
ownership of school feeding programs greatly increase the chances for 
long-term program sustainability. According to WFP, its guidelines for 
school feeding programs require both national governments and local 
communities to provide a significant amount of resources and 
infrastructure. 

There are potential detrimental impacts if school feeding programs are 
not effectively implemented. For example, where adequate 
infrastructure is not available, increased attendance may lead to 
overcrowding and actually reduce educational achievement for existing 
students, while providing minimal benefit to new students. In some 
developing country circumstances, the school day is only a few hours. 
In such cases, time taken to prepare a meal may further limit an 
already inadequate period of instruction. In addition, if volunteers 
are not available to provide labor, teachers may be required to 
undertake this task at the expense of instructional time. Since school 
feeding is a highly visible income transfer, it may also be used for 
political purposes by actors in the recipient country. If school 
feeding programs are relatively ineffective, they may result in 
resources being taken away from better performing programs. 

According to several experts, in particular situations, school feeding 
programs may not be as cost effective in promoting learning as other 
possible approaches, such as establishing maternal child health and 
early childhood development programs or providing alternative 
nutritional and educational interventions (see appendix V). 

Pilot Program Did Not Adequately Incorporate Lessons Learned: 

The pilot program has not provided reasonable assurance that lessons 
from previous school feeding and food for education programs have been 
integrated into approved pilot projects. Under pressure to get the 
pilot up and running quickly, USDA gave interested applicants little 
time to prepare proposals, and it did not require them to provide 
basic information on and analysis of various factors important to 
successful food for education programs. Written criteria for 
evaluating proposals similarly did not focus on many of these factors. 
Many of the proposals approved did not address key elements of 
successful school feeding programs. Moreover, USDA provided little 
funding for important nonmeal components of the food for education 
projects, and only a few of the approved PVO proposals indicated they 
had obtained other donors' support for nonmeal components.
	
USDA Did Not Have Sufficient Information for Its Evaluation: 

According to USDA officials with whom we spoke, the agency was under 
pressure to start a new and complex food for education program quickly 
and with far less funds-—$300 million-—than what is needed to fully 
address the educational components of school feeding. As a result, 
USDA did not solicit basic information on various factors linked to 
effective school feeding and food for education programs. Table 1 
lists a set of questions, based on lessons learned, that USDA could 
have used to guide the type of information and analysis requested from 
implementing partners (i.e., cooperating sponsors and WFP) and, 
subsequently, for evaluating proposal quality. As shown in table 1, 
many important factors that experts cited were not addressed 
specifically by USDA in its formal request for proposals, and other 
items were only partly addressed in its request.[Footnote 31] The 
request was made to cooperating sponsors but not to WFP. (Less 
information was sought from WFP because, as a USDA official told us, 
many projects in the WFP proposals had previously been reviewed and 
approved by the U.S. government as part of the process by which the 
WFP Executive Board approves its projects.[Footnote 32]) We derived 
the questions from our review of lessons described in various studies 
and other documents on school feeding and food for education programs 
(see appendix IV, especially tables 4 to 10. Also see appendix VI for 
a more complete discussion of the interagency process used to 
evaluate, and approve proposals.) 

Table 1: Presence or Absence of Key Factors in USDA's Request for 
Proposals and in Written Criteria for Evaluating Proposals: 

Overall area of focus: Targeting; 
	
Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Is program 
targeted on areas/communities with relatively low school enrollment 
and attendance rates? 
Did USDA request that item be addressed in proposal submissions[A]: 
Yes [B,C] 
Was item included in written criteria for evaluating proposals: Partly. 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Is program 
targeted on areas/communities with relatively low rates 
of literacy? 
Did USDA request that item be addressed in proposal submissions[A]: 
Yes{B]; 
Was item included in written criteria for evaluating proposals: Partly. 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Is program 
targeted on low-income areas? 
Did USDA request that item be addressed in proposal submissions[A]: 
Yes[C]; 
Was item included in written criteria for evaluating proposals: No. 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Is program 
targeted on areas where enrollment and attendance are considerably 
lower for girls than boys? 
Did USDA request that item be addressed in proposal submissions[A]: 
Yes{B]; 
Was item included in written criteria for evaluating proposals: Partly. 

Overall area of focus: Learning environment; 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Does proposal 
address whether there are adequate numbers of teachers currently 
available and provisions to increase the number of teachers needed in 
response to expected rise in student enrollment and attendance? 
Did USDA request that item be addressed in proposal submissions[A]: No; 
Was item included in written criteria for evaluating proposals: No[D]. 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Does proposal 
address whether teacher training is adequate and whether actions will 
be taken to provide additional training? 
Did USDA request that item be addressed in proposal submissions[A]: 
No[E]; 
Was item included in written criteria for evaluating proposals: No[F]. 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Does proposal 
address whether there are adequate supplies of good textbooks and 
other learning materials? 
Did USDA request that item be addressed in proposal submissions[A]: No; 
Was item included in written criteria for evaluating proposals: No[F]. 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Does proposal 
address whether classroom space, desks and chairs, lighting, and 
heating/cooling are adequate and, if not, actions that will be taken 
to improve the situation? 
Did USDA request that item be addressed in proposal submissions[A]: No; 
Was item included in written criteria for evaluating proposals: No[F]. 

Overall area of focus: Health and nutrition issues; 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Does proposal 
address whether intestinal parasitic infections are a problem and, if 
so, how to address them? 
Did USDA request that item be addressed in proposal submissions[A]: No; 
Was item included in written criteria for evaluating proposals: No. 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Does proposal 
address whether clean water and adequate sanitation facilities are 
present and, if not, what will be done to address them? 
Did USDA request that item be addressed in proposal submissions[A]: No; 
Was item included in written criteria for evaluating proposals: No. 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Does proposal 
address whether student population has serious micronutrient 
deficiencies and, if so, what will be done to address them? 
Did USDA request that item be addressed in proposal submissions[A]: No; 
Was item included in written criteria for evaluating proposals: No. 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Does proposal 
address need for health and nutrition education and, if appropriate, 
offer to provide it? 
Did USDA request that item be addressed in proposal submissions[A]: No; 
Was item included in written criteria for evaluating proposals: No. 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Does proposal 
specifically discuss the nutrient content of the meal that will be 
provided and identify how it addresses nutritional
needs of the student population? 
Did USDA request that item be addressed in proposal submissions[A]: 
No[C]; 
Was item included in written criteria for evaluating proposals: Yes. 

Overall area of focus: Community and parental involvement; 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Does proposal 
address actions to involve the local community in the program? 
Did USDA request that item be addressed in proposal submissions[A]: 
Yes[E]; 
Was item included in written criteria for evaluating proposals: Partly. 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Does proposal 
address actions to involve parents in the program? 
Did USDA request that item be addressed in proposal submissions[A]: 
Yes[E]; 
Was item included in written criteria for evaluating proposals: Partly. 

Overall area of focus: Government commitment; 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Does proposal 
discuss whether the national government subscribed to "education for 
all" goal? 
Did USDA request that item be addressed in proposal submissions[A]: 
Yes[C]; 
Was item included in written criteria for evaluating proposals: Partly. 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Does proposal 
discuss whether the national government has initiated broad, national 
school reform programs that focus on essential inputs to education and 
learning? 
Did USDA request that item be addressed in proposal submissions[A]: No; 
Partly[F]. 

Overall area of focus: Sustainability; 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Does proposal 
explain what resources, if any, national government is committing? 
Did USDA request that item be addressed in proposal submissions[A]: 
No[E]; 
Was item included in written criteria for evaluating proposals: No. 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Does proposal 
explain what resources, if any, local communities are committing? 
Did USDA request that item be addressed in proposal submissions[A]: 
No[E]; 
Was item included in written criteria for evaluating proposals: No. 

Question related to lessons learned from experience and/or study of 
previous school feeding and food for education programs: Does 
proposing sponsor explain why it believes program can become self-
sustaining for the community and over what period of time? 
Did USDA request that item be addressed in proposal submissions[A]: No; 
Was item included in written criteria for evaluating proposals: Partly. 

Note: Also see discussion in appendix VI on the interagency process 
used to evaluate and approve proposals. 

[A] USDA's requests for specific information were made only to 
cooperating sponsors and not to WFP. 

[B] Sponsors were asked to provide, to the extent possible, 
information on literacy rates for the target population, and 
percentage of school-age children attending school (with special 
emphasis on school-age girls attending school). 

[C] USDA said priority consideration would be given to projects in 
countries that have a commitment to universal free education but need 
assistance in the short run, and where the projects would promote 
significant improvements in nutrition, school enrollment, and 
attendance levels. USDA also said it wanted to target poor countries. 

[D] USAID's written criteria asked for a review of the educational 
components of the proposal and their adequacy. 

[E] Sponsors were asked to provide information on the impact of their 
proposed projects on areas such as teacher training, community 
infrastructure (PTAs and community groups), health and nutrition, and 
other potential donors. 

[F] USAID's written criteria asked for a review of whether proposals 
addressed the host country educational policies and commitment to 
basic education. 

Source: GAO analysis. 

[End of table] 

As table 1 indicates, USDA sought some information on how the projects 
would be targeted. For example, USDA indicated that it wanted to 
target poor countries and that it favored programs that would 
significantly improve enrollment and attendance. However, USDA did not 
require that proposals address how programs would be designed to 
improve educational performance, nor did it seek any information on 
factors that are key to whether learning could occur, such as adequate 
numbers of well-trained teachers and reasonable supplies of good 
learning materials. Similarly, USDA asked requesters how their 
programs would affect health and nutrition but did not specifically 
ask whether the schools had safe water and adequate sanitation 
facilities and whether intestinal parasitic infections in the student 
population were likely to be a problem. A USDA official told us there 
were limits on how much information the agency could require, given 
the short amount of time sponsors had to prepare proposals and the 1-
year duration of the pilot. Further, the agency did not want to make 
the information requirements so costly for sponsors that it would get 
few or no proposals, the official said. 

Regarding the criteria used to evaluate the programs, table 1 also 
shows that U.S. agencies' written criteria did not specifically 
address most of the key factors we derived, based on our review of 
lessons from previous school feeding and food for education programs. 
Of the 20 questions in table 1 on key factors in effective school 
feeding and food for education programs, 1 question was addressed 
specifically in the agencies' written criteria and 8 were partly 
addressed. None of the agencies' criteria specifically addressed the 
four learning environment questions shown in table 1. See appendix VI 
for a discussion of the written criteria used by agencies in 
evaluating the proposals. 

Some PVO and WFP Proposals Included Additional Information on Key 
Factors: 

We also reviewed the approved PVO and WFP proposals and found that 
many included information related to the key factors we identified as 
important to successful food for education programs, although fewer 
than a third of the approved PVO and WFP proposals discussed most of 
the items. In general, the response rate was highest for those factors 
where USDA had solicited information. 

Table 2 shows the number of approved PVO and WFP proposals that 
provided information related to the key factors irrespective of 
whether USDA requested this information. For example, a considerable 
number of the PVO and WFP proposals included information on certain 
health and nutrition issues that were not specifically requested by 
USDA. To a lesser extent, proposals also included information on 
factors associated with the learning environment. Overall, the highest 
response rates were mostly for factors for which USDA had sought 
information (i.e., school enrollment and attendance levels, literacy 
rates, target area based on low economic status, and programs that 
involve the community and parents.) (See appendix VI for additional 
discussion about the information that was included in WFP proposals.) 

Table 2: Number of Approved PVO and WFP Proposals that Addressed 
Various Key Factors: 

Overall area of	focus: Targeting; 

Question related to key factors: Is school feeding program (SFP) need 
in the target area based on low school enrollment or attendance levels? 
Number of PVO proposals addressing question (total approved proposals 
= 25): 19; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 24. 

Question related to key factors: What are the country's literacy rates? 
Number of PVO proposals addressing question (total approved proposals 
= 25): 10; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 9. 

Question related to key factors: What are the target population's 
literacy rates? 
Number of PVO proposals addressing question (total approved proposals 
= 25): 4; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 6. 

Question related to key factors: Is SFP need in the target area based 
on low economic status? 
Number of PVO proposals addressing question (total approved proposals 
= 25): 22; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 25. 

Question related to key factors: Figures on percent of girls attending 
schools in the country or in 
the target population; 
Number of PVO proposals addressing question (total approved proposals 
= 25): 7; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 11[C]. 

Overall area of	focus: Learning environment; 

Question related to key factors: Is current educational experience 
adequate regarding qualified teachers? 
Number of PVO proposals addressing question (total approved proposals 
= 25): 3; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 10. 

Question related to key factors: Is current educational experience 
adequate regarding student-to-teacher ratio? 
Number of PVO proposals addressing question (total approved proposals 
= 25): 4; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 4. 

Question related to key factors: Is current educational experience 
adequate regarding teacher training? 
Number of PVO proposals addressing question (total approved proposals 
= 25): 5; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 8. 

Question related to key factors: Is current educational experience 
adequate regarding textbooks and other learning materials? 
Number of PVO proposals addressing question (total approved proposals 
= 25): 8; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 9. 

Question related to key factors: Are classroom space, desks, and 
chairs adequate?; 
Number of PVO proposals addressing question (total approved proposals 
= 25): 4; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 5. 

Overall area of	focus: Health and nutrition issues; 

Question related to key factors: Do schools have safe water? 
Number of PVO proposals addressing question (total approved proposals 
= 25): 10; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 9. 

Question related to key factors: Do schools have adequate sanitation 
facilities? 
Number of PVO proposals addressing question (total approved proposals 
= 25): 10; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 9. 

Question related to key factors: Is target population likely affected 
by short-term hunger, chronic 
malnutrition, or protein/energy malnutrition? 
Number of PVO proposals addressing question (total approved proposals 
= 25): 14; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 18. 

Question related to key factors: Is target population likely affected 
by micronutrient deficiencies? 
Number of PVO proposals addressing question (total approved proposals 
= 25): 6; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 9. 

Overall area of	focus: Community/parental involvement; 

Question related to key factors: Will program involve the community or 
community groups? 
Number of PVO proposals addressing question (total approved proposals 
= 25): 17; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 16. 

Question related to key factors: Will program involve parents? 
Number of PVO proposals addressing question (total approved proposals 
= 25): 16; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 15. 

Overall area of	focus: Government commitment; 

Question related to key factors: Is national government committed to 
universal free education?[D] 
Number of PVO proposals addressing question (total approved proposals 
= 25): 7; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 6. 

Question related to key factors: Is national government committed to 
educational reform? 
Number of PVO proposals addressing question (total approved proposals 
= 25): 7; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 14. 

Overall area of	focus: Sustainability; 

Question related to key factors: Is national government or local 
community committed to eventually assuming responsibility for 
operating the program? 
Number of PVO proposals addressing question (total approved proposals 
= 25): 2; 
Number of WFP proposals[A] addressing question (total approved 
proposals = 26[B]): 5. 

[A] WFP submitted very brief proposals to USDA that at best addressed 
only a few of the items. However, additional documentation was 
available, to U.S. officials who evaluated the proposals, at WFP's Web 
site for those proposals involving projects that had already been 
approved by WFP's Executive Board. We reviewed that documentation, as 
well as WFP proposals, for the purpose of this table. In its proposal 
submission to USDA, WFP said its selection criteria for the proposals 
included mainly lower-income countries with some economies in 
transition; countries committed to universal free education, with a 
special emphasis on ensuring girls' education; and countries committed 
to sustainable school feeding. In addition, WFP said that projects 
would be implemented in countries where national governments and local 
authorities aim to attract children to schools in areas where 
enrollment rates are lowest and school meals are most likely to make a 
difference. (See appendix VI for further discussion about WFP 
information provided to USDA as part of the proposal process.) 

[B] USDA approved 34 WFP proposals; however, eight were for project 
expansions that had not yet been approved by WFP's Executive Board. 
Documentation on these eight proposals was not available on WFP's Web 
site. According to WFP officials, expansion projects closely parallel 
the original projects. Therefore, our analysis focused on the other 26 
projects. 

[C] WFP proposals for Bolivia, the Dominican Republic, El Salvador, 
Gambia, Kenya, and Peru give the percentage of participants in the 
program who are girls. 

[D] When the pilot program was announced, the director of the U.S. 
National Economic Council said that a prerequisite for countries being 
selected was a commitment to universal, free education. A USDA 
official told us that for all the approved proposals, USDA 
independently verified that the national government was committed to 
universal free education. WFP proposals indicated the following 
countries were committed to universal education but did not specify 
whether that included a free education: Bhutan, Bolivia, Cameroon, 
Cote d'Ivoire, El Salvador, Ethiopia, Gambia, Honduras, Kenya, Nepal, 
and Tanzania. 

Source: GAO analysis of proposal documentation for USDA approved 
proposals. 

[End of table] 

USDA Provided Little Funding for Components Identified as Important to 
Successful Programs: 

USDA provided little funding for nonmeal components—-such as basic 
classroom materials, nutritional education, and treatment of parasitic 
infections-—that are essential elements of an integrated approach to 
food for education programs. Altogether, USDA approved 60 proposals, 
including 34 for WFP,[Footnote 33] 25 for PVOs, and 1 for the 
government of the Dominican Republic. For WFP projects, USDA largely 
funded only school meals and related costs, including storage, 
transportation, and handling of the food.[Footnote 34] For the PVO 
projects, USDA was willing to consider proposals that included nonfood 
components to be funded by monetizing some of the surplus commodities 
or by the PVOs themselves.[Footnote 35] We found that 17 of the 25 
approved PVO proposals included nonmeal components; but of the 17 
proposals, only 10 included in their proposed budget a dollar value 
for resources that would be allocated to some or all of these 
activities.[Footnote 36] (See appendix VII, table 14, for additional 
information on the extent to which PVO proposals included nonmeal 
components and budgeting for these activities.) 

Weaknesses in Structure, Planning, and Management Reduce Chances for	
Pilot Program Success: 

While the U.S. pilot program expects to provide food to more than 8 
million schoolchildren in developing countries, its structure, 
planning, and management to date do not reasonably ensure a program 
that will produce substantial gains in enrollment, attendance, and 
especially learning.[Footnote 37] The administration's decision to 
fund the program through surplus commodities may be appropriate for a 
1-year pilot but is not sustainable for a longer-term program. USDA, 
which was selected to manage the pilot, lacked the expertise and 
resources of USAID--the agency traditionally responsible for foreign 
development aid such as food for education programs. The pressure on 
USDA to get the pilot program up and running quickly did not allow 
time to adequately plan the program and hire additional staff to 
manage it. USDA launched the pilot before fully developing a strategy 
for monitoring and evaluating performance; and, because of the pilot's 
short time frame, USDA officials told us they would not be able to 
evaluate improvements in learning—-one of the program's three 
objectives. This weakness, as well as others related to ensuring 
financial accountability for some parts of the projects, could make 
determining the pilot's effectiveness difficult. 

Surplus Commodities Not Reliable Funding Mechanism: 

The administration's decision to use surplus agricultural commodities 
to fund the pilot was an expedient way to get the program quickly 
under way. However, surplus commodities are not a good vehicle for 
funding a medium- or long-term development program, since surpluses 
cannot be ensured on a regular basis.[Footnote 38] (For example, 
between fiscal years 1996 and 1998, there was no section 416(b) 
program.) Although the pilot was expected to run for just over 1 year, 
the administration contemplated a multiyear food for education 
program, possibly lasting as long as a decade. Under this scenario, 
when surpluses were not available, the administration would have to 
end the program or sustain it through the foreign aid budget, which is 
expected to have many competing priorities in the foreseeable future. 

USDA Lacked Expertise on School Feeding Programs: 

USAID—-traditionally the U.S. agency for providing foreign development 
assistance, including school feeding and food for education programs—-
would normally have been the logical choice to establish and run the 
pilot. However, in light of constraints on foreign aid funding 
generally and other high priority development needs, the 
administration wanted CCC[Footnote 39] to manage the pilot, and to do 
so using available surplus agricultural commodity funding authority 
(i.e., section 416(b) of the Agricultural Act of 1949). 

The administration's decision to assign management responsibility for 
the pilot to USDA rather than USAID follows a recent trend of giving 
USDA a larger role in U.S. food aid programs, primarily because of 
increased section 416(b) program activity. However, USDA lacked 
USAID's resources (such as USAID's overseas development missions) and 
USAID's school feeding/food for education development expertise. 
[Footnote 40] The principal mission of USDA's Foreign Agricultural 
Service (FAS) is to help ensure open markets for U.S. agricultural 
exports; it generally has had little experience in managing school 
feeding development assistance programs.[Footnote 41] USDA has 
previously used section 416(b) authority to provide some commodities 
for international school feeding programs, but we were told the 
amounts were relatively small[Footnote 42] and not for integrated food 
for education programs. In contrast, USAID has been engaged in school 
feeding programs since the 1950s and administers economic and 
humanitarian assistance programs in more than 80 countries.[Footnote 
43] Beginning in the mid-1990s, USAID began reducing its support for 
traditional school feeding programs that provided only meals, citing 
mounting evidence that school feeding, in and of itself, contributed 
little to improving child learning ability or child nutrition on a 
sustainable basis. According to USAID officials, its school feeding 
assistance has evolved into programs designed to improve education 
(i.e., enrollment, attendance, and graduation rates, especially for 
girls) by focusing on national education policy reform, curriculum 
development, and teacher training programs. In 2000, USAID spent $33 
million on PV0-operated food for education programs in eight countries 
that benefited 1.3 million children. 

Pilot Program Was Launched Quickly and Has Been Understaffed: 

President Clinton announced GFEI on July 23, 2000. USDA began to 
implement the pilot program almost immediately, leaving little time 
for planning and relying on existing staff from within the Foreign 
Agricultural Service to work on the assignment. USDA issued its 
request for proposals on September 6, 2000, with a closing date for 
all submissions at the end of September. (See appendix IX for key 
events from the time the concept of an international school lunch 
program was suggested until approval of the GFEI pilot program 
proposals.) 

According to USDA officials, USDA was understaffed when the GFEI pilot 
was launched and a year later still lacked sufficient staff for 
handling food aid matters. For example, in a July 2000 meeting with 
PVOs to discuss the pilot program, the Secretary of Agriculture said 
the lack of staffing in U.S. agencies for running food aid programs 
was acute. At the same time, he said the president wanted to see some 
benefits from the pilot program before leaving office. In November 
2000, a USDA official told us that USDA was generally understaffed for 
monitoring food aid programs. At a July 2001 meeting with PVOs, other 
USDA officials apologized to PVO representatives for having too few 
staff available to negotiate agreements and address other food aid 
program issues in a timely manner.[Footnotes 44, 45] 

According to OMB, in March 2001, the administration authorized USDA to 
use $2.5 million of the $300 million in CCC funds for administrative 
salaries and expenses. According to a USDA official, the funds are 
being used to facilitate monitoring and evaluation of the pilot 
program's impact. As of September 2001, a year after the pilot was 
launched, USDA was still in the planning stage regarding hiring 
regional coordinators and local national staff in PVO recipient 
countries to help monitor pilot program projects.[Footnote 46] 

USDA Policy Change on Funding PVO Projects Delayed Implementation: 

USDA's September 2000 Federal Register notice indicated that CCC funds 
might be available to cover some of the cooperating sponsors' expenses 
related to implementing the school feeding projects. As a result, many 
PVOs submitted proposals based on the assumption that they would 
receive CCC funds to cover part of their expenses. However, in January 
2001 USDA reversed its position, announcing that funding would not be 
available.[Footnote 47] This meant that PVOs' expenses in recipient 
countries would have to be covered by selling (monetizing) commodities 
in the recipient countries and using the resulting local currency 
proceeds to cover in-country costs. The policy change further meant 
that PVO direct administrative headquarters' costs could not be 
covered, since the section 416(b) program does not allow monetization 
of commodities for that purpose. 

USDA's policy shift resulted in several of the proposals having to be 
restructured, causing discontent within the PVO community and leading 
to delays in concluding a number of agreements. In fact, about one-
third of the approved PVO agreements were not signed by the end of 
September 2001. In addition, the change presented problems for some 
PVOs because it required them to monetize increased quantities of 
commodities within recipient countries to recover some of their costs, 
and there were limits on the commodity tonnage that could be monetized 
effectively. Some PVOs were also upset because all of WFP's operating 
expenses, including headquarters' costs, were funded by CCC cash 
payments. Legislative relief in the form of limited CCC funding was 
provided to PVOs in late July 2001; at that time, only 4 PVO 
agreements had been signed.[Footnote 48] (Appendix IX discusses the 
funding sources used for pilot program sponsors in more detail.) 

Weaknesses in Program Management and Short Duration of Pilot Will 
Affect Monitoring and Evaluation: 

To know whether programs are effective, program objectives should 
clearly describe the intended end results and accompanying indicators 
so that changes and progress toward achieving the objectives can be 
tracked over time. However, USDA initiated its requests for proposals 
in September 2000 without having a comprehensive plan for how it would 
monitor and evaluate project performance and has spent much of the 
time since then establishing such a plan. USDA and WFP will collect 
baseline data on school enrollment and attendance for the time before 
the projects began and monitor and assess change in these variables 
over the course of the projects. However, USDA has not set specific 
targets or desired performance levels for enrollment and attendance in 
its agreements with most of its implementing partners. In addition, 
although improved learning is one of the three principal objectives of 
the pilot program, USDA said it will not monitor and evaluate 
performance on this variable, unless improved learning is an element 
within an agreement, because of the program's short duration.[Footnote 
49] 

Officials from USDA's Foreign Agricultural Service told us USDA is 
responsible for evaluating the performance of WFP, PVOs, and the 
Government of the Dominican Republic in implementing GFEI projects. 
According to these officials, FAS' mandate is to monitor and review 
the 25 PVO and 1 country government projects in 20 countries from 
October 2001 through March 2003, and at appropriate intervals report 
to the Congress on the projects' status. They added that FAS 
headquarters staff is also responsible for evaluating WFP's GFEI 
project implementation. They stated that the agency intends to 
complete an interim status report on the pilot for the Congress by 
July 2002 that will address several performance-related issues. 
[Footnote 50] 

In its September 6, 2000, Federal Register notice, USDA said that 
cooperating sponsors would be required to report periodically the 
number of meals served, enrollment levels, and attendance levels, 
including female attendance levels. In addition, USDA said that 
reports should include information on infrastructure relevant to 
sustaining the feeding program, such as establishment of PTAs and 
community groups. However, the notice did not indicate whether 
sponsors would be required to collect baseline data on these 
variables, which would permit comparisons of conditions before a 
project got under way and when it was completed. It did not indicate 
whether or how success would be measured—for example, what percent 
improvement in attendance would represent achievement of the program's 
objectives. In addition, the notice did not discuss whether sponsors 
would be required to report on educational performance, one of the 
program's three principal objectives.[Footnote 51] 

In February 2001, USDA began negotiating final agreements with 
cooperating sponsors and WFP for approved proposals. As of December 
2001, USDA had completed agreements for 21 of 26 approved cooperating 
sponsor project proposals. All 21 proposals contained provisions that 
required reporting on the number of meals served, enrollment and 
attendance levels (including female attendance), and establishment of 
infrastructure relevant to sustaining the feeding program, such as 
PTAs and community groups. However, less than half of these agreements 
indicated a requirement for baseline data; and a majority of the 
agreements did not specify performance targets for enrollment, 
attendance, and female attendance. None of the agreements included 
reporting requirements for educational performance. (According to USDA 
officials, PVOs opposed such reporting, arguing that the pilot was too 
short in duration to permit a meaningful analysis of impacts on 
learning.) By September 2001, 33 of 34 agreements for WFP projects 
were concluded, with 1 deferred until fiscal year 2002. None of these 
agreements specified requirements for measuring project performance; 
in fact, they did not even say that WFP would report the types of data 
USDA had required from cooperating sponsors, such as enrollment and 
attendance data. 

Nonetheless, WFP developed a detailed survey instrument for collecting 
baseline information on its GFEI-funded projects. The survey was pilot-
tested in August 2001, approximately 1 year after USDA received 
proposals from WFP and cooperating sponsors. According to USDA and WFP 
officials, WFP conducted the surveys in a sample of schools[Footnote 
52] for all of its projects before the end of 2001 and before the food 
aid was distributed. In addition to collecting basic information on 
the feeding program,[Footnote 53] the survey sought: 

* detailed baseline and subsequent performance data on school 
enrollment and attendance (broken down by boys and girls and grade 
level); 

* the number of certified and uncertified teachers in the school; 

* the number of classrooms;[Footnote 54] 

* certain baseline information on community and parental involvement 
[Footnote 55] and health and nutrition issues;[Footnote 56] and; 

* whether the school had other ongoing programs related to effective 
school feeding programs and if so, the name of the donor providing the 
program.[Footnote 57] 

The survey also called for the use of focus groups[Footnote 58] to 
collect views on the likely reasons why eligible children did not 
enroll and enrolled boys and girls did not attend school during a year. 

The survey instrument indicates WFP's interest in upgrading monitoring 
and evaluation of its feeding programs, since previous efforts 
revealed some weaknesses.[Footnote 59] However, the survey included 
only two questions focused on the possible impact of the programs on 
improved learning.[Footnote 60] WFP is sharing its survey results with 
USDA. (See appendix III for additional information on WFP activities 
to improve monitoring and evaluation of school feeding programs.) 

During the summer of 2001, USDA was still debating how to monitor and 
evaluate performance for the cooperating sponsors' projects. In August 
2001, it convened a working group of USDA officials and USAID 
consultants with expertise in monitoring and evaluation methodologies 
to discuss the issue. The group recommended use of local school or 
government records for collecting data on enrollment and attendance, 
but it was against collecting quantitative data on indicators for 
measuring educational progress (such as reduced dropout rates, 
retention and/or completion, and promotion to the next grade) and 
level of community participation and infrastructure development. For 
the latter variables, it recommended information be collected through 
a combination of focus groups and structured interviews with school 
staff and parent and community groups. 

In fall 2001, USDA decided to use the WFP survey instrument for the 
cooperating sponsors' projects and, like WFP, apply the survey in a 
sample of the schools in each project. According to USDA officials, 
doing so would allow collection of comparable data, provided USDA's 
sampling strategy was properly designed. USDA also decided to contract 
with about 20 local national monitors (approximately 1 per country) to 
collect the data and 5 regional coordinators to manage the monitors. 
In late December 2001, USDA officials told us they planned to add a 
few more questions to the survey to address concerns about whether 
some of the projects were well targeted.[Footnote 61] They also said 
the surveys would be conducted in early 2002.[Footnote 62] 

USDA officials told us that they ultimately decided not to measure 
change in school learning. They said that from the beginning of the 
pilot, USDA, WFP, and PVOs were concerned about the ability to 
effectively evaluate and judge an increase in student performance 
under a 1-year pilot program. Research that tries to demonstrate 
improvements in academic achievement is lengthy and requires a long-
term approach, they said. USAID officials with whom we spoke were also 
critical of the short time allowed for running the pilot program. They 
said USAID pilot programs usually take 4 to 5 years, with an 
evaluation done in the third year to see if the program is on track, 
and an assessment of the impact conducted in the fourth year. 

Processes to Prevent Disincentive Effects of Food Aid Raise Some 
Concerns: 

An effective global food for education program needs to ensure that 
food aid does not interfere with commercial markets and inhibit food 
production in developing countries.[Footnote 63] USDA uses an 
international consultative process—the Consultative Sub-Committee on 
Surplus Disposal (CSSD)-—to keep the pilot program's food aid from 
interfering with commercial exports. The process involves notification 
of various categories of food aid donations, prior consultation with 
other exporters, and establishment of Usual Marketing Requirements 
(UMR) to ensure that food aid recipients maintain a normal intake of 
commercial imports in addition to the food aid they receive. According 
to the CSSD, in recent years several factors reduced the effectiveness 
of the UMR approach, including (1) lack of uniformity in the 
compliance period (fiscal year, crop year, and calendar year); (2) 
fewer food aid operations covered by the UMR because many transactions 
are exempt; (3) a rise in UMR waivers[Footnote 64] for countries 
facing difficult economic situations; and (4) delays in collecting 
trade data, which make establishment of 5-year average commercial 
imports as a benchmark for current import levels unrealistic. USDA 
officials acknowledged that some countries have expressed concerns 
that GFEI might adversely affect commercial exports but said they have 
not received any specific complaints about the U.S. pilot's food 
exports. 

To address disincentive effects of food aid on local production, the 
United States requires all proposed food aid projects to submit an 
analysis showing the recipient has adequate storage facilities and 
that food aid will not disrupt domestic production and marketing. 
(Technically the analysis is known as a Bellmon determination.) 

We reviewed the analyses by cooperating sponsors whose projects were 
approved for the pilot and found the analyses were not adequate for 
determining disincentives to production of local commodities. All 
cooperating sponsors concluded that the amount of food for their 
projects was so small it was unlikely to significantly affect local 
production. But their analysis of data on local market conditions was 
generally based on production of identical commodities. For example, 
if wheat was not grown in the recipient country, sponsors concluded 
there was no disincentive to importing and monetizing wheat—without 
considering whether the amount of imported wheat would affect price or 
demand for locally produced substitute commodities. Cooperating 
sponsors did not adequately verify that the commodities were in demand 
and would not compete with local markets, other commercial export 
programs, and other donor imports.[Footnote 65] 

USDA officials told us that cooperating sponsors are responsible for 
analyzing the potential disincentive effects of their projects. They 
said USAID no longer has agricultural officers stationed overseas and 
now USDA has to rely on PVOs-—which have on-the-ground, in-country 
staff-—to determine whether the food aid will adversely affect 
recipient country markets. (USAID advised us that while the number of 
agricultural officers overseas has been reduced in recent years, it 
still has such officers in a number of posts.)[Footnote 66] Although 
USDA and/or USAID attaches may review such analyses, USDA does not 
independently verify the results.[Footnote 67] USDA officials also 
noted that the lack of good data could affect sponsors' ability to 
prepare more robust analyses. USDA does not require WFP to conduct or 
submit similar analyses of WFP projects that are partly funded by the 
U.S. pilot program. However, WFP told us a review is required of all 
WFP proposed projects for their potential impact on production and 
markets, and food aid donors (including the United States) 
participate.[Footnote 68] 

Key Weaknesses in Financial Accounting Could Have Negative Impact on 
Pilot Program: 

We identified several weaknesses in how USDA has maintained financial 
accountability over WFP and PVO projects that could adversely affect 
the pilot program. Although USDA advances funds (in the case of WFP) 
or food (in the case of cooperating sponsors) on the basis of their 
estimated needs and requires them to provide regular though different 
forms of financial and project status reporting, WFP in particular has 
not adequately accounted for past Section 416(b) program donations. 
The PVOs provide more detailed financial reporting, in part, because a 
large portion of the commodities they receive are to be monetized in 
country to cover food[Footnote 69] and other expenses. USDA requires 
that PVOs monetize commodities at market prices, but it has not 
systematically tracked whether the PVOs received prices for the 
monetized commodities that were commensurate with their cost or 
whether the funds were spent in accordance with approved program plans. 

WFP Reporting Has Been Inadequate: 

Under a section 416(b) umbrella agreement, WFP is required to account 
for the costs it incurs and charges USDA on food aid donations. WFP is 
supposed to submit annual standardized project reports that provide 
implementation and actual expenditure data for ongoing activities 
similar to what is required of PVOs. We found that WFP had not met its 
obligation to provide USDA with an accounting for past Section 416(b) 
program donations by providing detailed actual cost data. As a result, 
USDA is not in position to know whether its advances to WFP, on the 
basis of initial cost estimates, are consistent with actual project 
costs and to what extent the project objectives are being achieved 
within the approved budget estimates. A similar situation exists with 
USAID-funded donations to WFP. According to a USAID official, WFP has 
not provided actual cost data for direct and indirect project costs at 
the level of project activities and by donors. Such data is needed, 
the official said, to know whether the United States is meeting and 
not exceeding its fair share of a project's total cost, as well as the 
costs of specific project activities. 

In April 2001, U.S. officials reiterated to WFP officials the need for 
disaggregated actual cost data. During the meeting, WFP officials 
noted that WFP was in transition, using a new financial information 
system for new business while still using the earlier system for old 
business. According to a USAID review conducted in June 2001, WFP's 
new system appeared to have the capacity to accurately monitor and 
report on full cost recovery in the aggregate. However, the system was 
not yet fully operational and thus the adequacy of the complete system 
could not yet be determined. In September 2001, WFP told USDA it would 
not be able to provide finalized reports for fiscal year 1999 
obligations that were due by the end of that month. According to 
USAID, pursuant to bilateral consultations between an interagency U.S. 
government delegation and WFP management, the United States agreed to 
a 6-month extension for WFP to report actual cost data for all U.S. 
government contributions to WFP. 

Oversight of PVO Monetized Commodities Is Limited: 

As previously indicated, a substantial portion of the commodities 
provided to PVOs are to be monetized, with the proceeds used to pay 
for other foods and/or other expenses, such as administrative expenses 
and inland transportation, storage, and handling costs. For the first 
17 completed PVO agreements, more than 80 percent of the commodities 
are to be monetized. At issue is whether USDA is sufficiently tracking 
the proceeds that PVOs receive from the commodities they monetize. 
[Footnote 70] Also, if a PVO sells a commodity for less than the 
market value, the commodity could undercut other commercial sales, 
including imports or domestically produced commodities, and fewer 
proceeds would be available for financing the school meals or related 
activities.[Footnote 71] 

USDA regulations require that PVO commodity sales meet local market 
conditions and that PVO and government sponsors provide a report 
showing deposits into and disbursements out of special accounts 
established for commodity sales proceeds. In past Section 416(b) 
programs, USDA did not determine to what extent proceeds compared with 
what sponsors expected to receive as stipulated in the project 
agreements, nor whether the commodities were sold at real market 
prices. However, in September 2001, USDA officials told us they plan 
to conduct such an analysis for the pilot program projects.[Footnote 
72] 

Most Other Donors Currently Uncommitted or Opposed to Major Support of 
GFEI: 

The success of a comprehensive, long-term GFEI strongly depends on 
other donor support, but most other donors are either opposed or not 
committed to supporting GFEI at this time. A few donors have indicated 
support for the food for education initiative but have offered little 
in terms of specific additional contributions. While WFP officials are 
confident of eventual support, most donor countries seem unlikely to 
provide substantial support unless the United States adopts a 
permanent program that is not dependent on surplus commodities and/or 
unless the pilot program demonstrates strong, positive results. Some 
donors are opposed to GFEI on the grounds that developmental food aid 
assistance is ineffective in promoting sustainable development. Others 
are noncommittal for a variety of reasons, including possible adverse 
impacts on commercial agricultural exports to and domestic 
agricultural production in recipient countries. 

Long-Term Program Will Need Substantial Support from Other Donors: 

The U.S.-proposed GFEI challenged other donor countries and 
organizations to join the United States in helping achieve the goal of 
education for all children in developing countries by 2015. Indeed, 
the United States said that its willingness to extend the pilot 
program beyond its first year would depend in part on other donors' 
response.[Footnote 73] Since the initiative was first proposed, U.S. 
officials have indicated they would like to see other donors 
contribute, in aggregate, anywhere from two-thirds to three-quarters 
of the total cost of a global food for education program. 

The Clinton administration estimated that at least 300 million 
children in developing countries need school meals. Assuming an annual 
average cost of $34 per student for a 180-day school year, the annual 
meal cost alone for 300 million children would be approximately $10.2 
billion.[Footnote 74] To put this estimate in perspective, in 1999, 
$10.2 billion represented about 96 percent of the Organization for 
Economic Cooperation/Development Assistance Committee countries' 
official development assistance[Footnote 75] to least developed 
countries, or about 18 percent of development assistance directed to 
all developing countries. In addition, net official development 
assistance has declined during the past decade, from $56.7 billion in 
1991 to $53.7 billion in 2000. 

We estimate the food tonnage required to provide a school meal for 300 
million children (for a 180-day school year) to be in excess of 16 
million metric tons, which would exceed average annual global food aid 
deliveries between 1990 and 2000 by about 40 percent.[Footnote 76] 
(Global food aid deliveries averaged approximately 12 million metric 
tons per year from 1990 through 2000.) Moreover, food aid for 
development programs, only a part of which is for school feeding, 
averaged about 3 million metric tons per year. Thus GFEI would 
represent more than a fivefold increase for these types of programs. 

Donors Have Been Generally Noncommittal to GFEI: 

According to a State Department cable, when the United States proposed 
GFEI at the July 2000 G-8 Summit, the proposal received a cool 
reception. Subsequently, in November 2000, the State Department 
headquarters asked U.S. diplomats in 23 countries to explain the U.S. 
pilot program to foreign governments and encourage their support. In 
addition, the previous U.S. Ambassador to the U.N. Food Agencies in 
Rome sought other countries' support for GFEI through his 
participation in the WFP Executive Board and in official visits to 
food aid donor countries, such as Denmark and Finland. These efforts 
notwithstanding, most donor countries have yet to respond in a 
strongly positive or substantial way. 

Of the top 13 food aid donating countries for the period 1995 through 
1999, the United States supplied more than half of all deliveries, 
with the other donors providing slightly more than 41 percent (see 
appendix X). Table 3 summarizes general views of all but one of these 
other donor countries as well as Finland[Footnote 77] and their plans 
or actions to contribute to GFEI or the WFP's school feeding 
initiative. As table 3 shows, representatives of 4 of the 12 donors 
(Japan, France, Italy, and Finland) indicated general support for the 
food for education initiative. The European Commission, the second 
largest provider of food aid in the world, has said it is against a 
"one-program-fits-all" approach, citing a preference for strategic 
planning that identifies all of a country's development needs and then 
analyzes alternative ways to achieve them.[Footnote 78] According to 
the Commission, education forms an integral part of the European 
Union's development policy, and it is crucial that all shortcomings in 
providing education are tackled at the same time. If analysis 
indicated that a food for education program would have a positive 
impact, the Commission would still want to assess the relative cost 
effectiveness and efficiency of the alternatives. Representatives of 
Germany, the United Kingdom, the Netherlands, and Sweden also 
expressed reservations about GFEI not being an integrated approach to 
development assistance and/or about the ability of recipient countries 
to sustain the programs over the long run. Representatives of 
Australia, Canada, Sweden, and the United Kingdom indicated they would 
like to see whether the U.S. pilot program or WFP program demonstrates 
successful results. Representatives of the European Commission, 
Canada, Germany, the Netherlands, and Sweden expressed concerns about 
or said they thought the U.S. program was being used to dispose of 
surplus commodities.[Footnote 79] 

Table 3: Other Donors' Overall Views on the Food for Education 
Initiatives: 

Donor[A]: European Commission; 
General views on GFEI and/or WFP School Feeding Initiative: It is 
unlikely the Commission will join GFEI; the Commission has no plans to 
do so. Will not encourage or participate in a one-program-fits-all 
approach. The Commission creates regional or country strategic plans 
identifying all development needs and analyzes the best ways to meet 
those needs on a case-by-case basis. U.S. program's use of surplus 
commodities and desire to aid U.S. farm incomes introduce 
inefficiencies and confuse objectives, leading the Commission to 
question the program's legitimacy; 
Plans or actions to contribute to GFEI or WFP School Feeding 
Initiative: If the U.S. program did not use surplus commodities, the 
Commission probably would not oppose GFEI, but it does not foresee 
supporting the program. It will continue to include educational and 
nutritional provisions in its own development programs, and in cases 
where food aid, through school feeding programs is appropriate, it 
would be ready to support SFPs for a limited period of time. 

Donor[A]: Japan; 
General views on GFEI and/or WFP School Feeding Initiative: Basically 
supports GFEI but is not strongly positive. Has no plans to make 
supplemental contributions to the particularly by the use of food aid. 
Plans or actions to contribute to GFEI or WFP School Feeding 
Initiative: There is uncertainty about whether GFEI will continue; 
Japan wants to support development but not initiative. 

Donor[A]: Canada; 
General views on GFEI and/or WFP School Feeding Initiative: Thinks 
there may be better ways to promote food security for development. Has 
some concern about whether GFEI is being used to dispose of surplus 
commodities. Says recent U.S. food aid has not always been used in a 
nontrade distorting way. Believes it would be premature to go forward 
with a global program before WFP and U.S. demonstrate results for 
their SFPs; 	
Plans or actions to contribute to GFEI or WFP School Feeding 
Initiative: Has no planning under way for a direct contribution to 
GFEI. Will await results of WFP and U.S. initiatives.
		
Donor[A]: Australia; 
General views on GFEI and/or WFP School Feeding Initiative: Has made 
no formal public statements supporting	GFEI. Is adopting a wait-and-
see position. The U.S. program has to demonstrate positive results; 
Plans or actions to contribute to GFEI or WFP School Feeding 
Initiative: If the program is sustainable, minimally distorting, and 
helps feed and educate children, Australia would consider it an pilot 
option. 

Donor[A]: Germany; 
General views on GFEI and/or WFP School Feeding Initiative: Prefers a 
multisectoral approach to development that includes health assistance 
and is supported by other donors and international organizations. 
Believes the trigger for the U.S. program was surplus commodities. 
Wonders whether U.S. program will be long lasting; Congress enacting a 
permanent program would address this concern; 
Plans or actions to contribute to GFEI or WFP School Feeding 
Initiative: Does not specifically support WFP's SFPs, but contributes 
to WFP development programs in specific countries and does not exclude 
from consideration WFP programs with an SFP component. Does not plan 
to provide separate support because of Germanys overall budget 
situation. 

Donor[A]: France; 
General views on GFEI and/or WFP School Feeding Initiative: Has 
generally good impressions of the U.S. program, but program could be 
dangerous if it donates only surpluses. GFEI could be important to 
developing countries, but their local agricultural production must be 
respected. Wants donors' actions to be transparent; 
Plans or actions to contribute to GFEI or WFP School Feeding 
Initiative: May undertake initiatives similar to the U.S., working 
through WFP or through French or European NG0s. However, France wants 
to maintain the current level of WFP development activities, not 
expand it. France's contribution could be cash, food, and or technical 
aid. 

Donor[A]: United Kingdom; 
General views on GFEI and/or WFP School Feeding Initiative: Supports 
the "education for all goal but does not think GFEI will help. Sees 
chronic undernutrition as a long-term development issue that must be 
dealt with by alleviating poverty. WFP's SFPs have not worked well due 
to lack of sustainability in the developing countries. Over the long 
run, recipients must be able to pay for teachers, books, etc. 
Plans or actions to contribute to GFEI or WFP School Feeding 
Initiative: Is not going to provide early funding for the initiative. 
Will wait for results of U.S. program. Would only support program if 
it were sustainable and backed by local governments with resources to 
support it. Believes that developing countries are not very 
enthusiastic about these programs because they have not previously 
asked for them. 

Donor[A]: Italy; 
General views on GFEI and/or WFP School Feeding Initiative: Strongly 
supports the WFP school feeding initiative; 
Plans or actions to contribute to GFEI or WFP School Feeding 
Initiative: Contributed approximately $952,000 to the WFP initiative 
in 2001 for three countries in the Horn of Africa region. 

Donor[A]: The Netherlands; 
General views on GFEI and/or WFP School Feeding Initiative: SFPs are 
among the best ways to boost enrollment and attendance but does not 
think the programs are sustainable once donors leave. WFP initiative 
seems driven by the U.S. and is probably a result of surplus U.S. 
commodities. If U.S. did a cash only program, the Netherlands would be 
more receptive; 
Plans or actions to contribute to GFEI or WFP School Feeding 
Initiative: Will not contribute to WFP initiative because of concerns 
about sustainability. In its bilateral programs, which are cash 
funded, the Netherlands would consider a program only if the recipient 
country asked for it and the Netherlands believed the education sector 
would benefit. 

Donor[A]: Denmark; 
General views on GFEI and/or WFP School Feeding Initiative: Has not 
done much about GFEI. Has a sectoral approach to development. Would 
look at SFPs in terms of a wider effort focusing on food security and 
development—including the education sector (e.g., teachers, books) 
because it is not adequate to feed a child if education is not 
improved; 
Plans or actions to contribute to GFEI or WFP School Feeding 
Initiative: Told the U.S. that it would evaluate GFEI for possible 
additional funding. (But U.S. officials in Rome said it would be hard 
to get Denmark to increase its funding, since it already gives 1 
percent of its GNP to foreign aid.) Might reallocate some of its 
existing WFP donation to the SFP initiative. 

Donor[A]: Sweden; 
General views on GFEI and/or WFP School Feeding Initiative: Welcomes 
GFEI but concerned that such a huge program could disturb markets. 
Conducts its development policy through integrated rather than 
separate programs such as school feeding. Sees U.S. initiative as a 
way to get rid of surpluses; 
Plans or actions to contribute to GFEI or WFP School Feeding 
Initiative: Will track GFEI's progress and may contribute to it if the 
program is deemed excellent. But may decide to provide more money to 
education or give additional cash to WFP and let WFP decide how to 
spend it. 

Donor[A]: Finland; 
General views on GFEI and/or WFP School Feeding Initiative: Concept of 
universal school feeding and WFP's program strongly endorsed by 
Finnish President and Minister of Agriculture. Cautioned that food aid 
should not be used to dump excess agricultural product without taking 
into account its effect on other countries; 
Plans or actions to contribute to GFEI or WFP School Feeding 
Initiative: Supports initiative through nondirected contribution that 
WFP can use for whatever development purpose and countries it decides 
upon. But has not indicated it will direct WFP to assign the 
contribution to the school feeding initiative or that it will increase 
its donation. 

Note: Most of the information presented in the table is based on our 
interviews in February and April 2001 with representatives of the 
countries, supplemented in a few cases by information contained in 
U.S. government documents. European Commission views are based 
primarily on a written statement provided to us in July 2001. A few 
donor country representatives said that their governments had not 
formulated an official view on GFEI but agreed to comment on the 
issues. 

[A] Donors are presented in descending order, based on the total 
amount (annual average) of all their global food aid deliveries during 
1995-99 (see appendix X). 

Source: GAO analysis. 

[End of table] 

In addition, some donors indicated they favor using food aid for 
emergency (rather than development) purposes, expressed reservations 
about providing assistance for school feeding programs in the form of 
food or surplus commodities, or indicated they lack convincing 
information on the effectiveness of WFP school feeding activities. 
(See appendix VIII for additional information on donor views on food 
aid.) 

Regarding actual support for GFEI, Italy has contributed nearly $1 
million to the WFP initiative in three African countries. A French 
representative said France might provide some support, either on its 
own or through WFP, but added that France wanted to maintain its 
current level of WFP development activities, which would limit 
France's ability to greatly increase funding for WFP's school feeding 
initiative. Representatives of Japan and Finland, the two other 
supporters, indicated their countries would not increase their current 
level of donations to support the initiatives. Meanwhile, 
representatives of Canada, Australia, the United Kingdom, and Sweden 
all indicated that they would track the progress of the food for 
education initiatives for the results. The German representatives said 
their country's budget situation does not permit providing additional 
support. 

In mid-April 2001, the U.S. Ambassador to the U.N. Food Agencies in 
Rome acknowledged that there had been very little movement by other 
donor countries toward supporting GFEI but said that they were coming 
around to the idea. They want to see an American commitment, which 
will begin with the pilot program's implementation, he said. The 
Ambassador said he thought Denmark, Finland, Norway, and Sweden would 
be on board within the next few months and that France and Germany 
would soon join in. At the same time, WFP officials told us that most 
or all governments, donors and recipients alike, support a global 
school feeding effort and that they were optimistic that additional 
contributions would be forthcoming by the end of 2001. 

At the beginning of August 2001, WFP officials told us the Swiss 
government was contributing 194 metric tons of food, and France 
intended to contribute a total of 5,280 metric tons of rice, beans, 
oil, and corn/soy blend to a Honduran program. In addition, they said, 
Cargill, Inc.,[Footnote 80] had provided a $50,000 donation to assist 
WFP's school feeding operation in Honduras (to be matched by the local 
Cargill affiliate in Honduras). Apart from food donations, the 
Canadian government approved the use of a $250,000 grant facility for 
WFP for a deworming effort in conjunction with WFP school feeding 
efforts in Africa, WFP officials said.[Footnote 81] In addition, an 
international fund offered to consider providing upwards of $300,000 
to fund nonmeal items (such as construction of schools, teacher 
training, training materials, school books, and cooking utensils) in 
least-developed countries. And, the officials said, WFP was 
negotiating new partnerships for school feeding, including the health, 
sanitation, and educational aspects of primary schools, with a variety 
of U.S. government and international agencies.[Footnote 82] 

At the end of December, 2001, the U.S. Mission to the U.N. Food 
Agencies in Rome told us that Italy, France, and Switzerland were 
still the only countries that had agreed to supplement the U.S. 
government contribution to the WFP school feeding program. 

Conclusions: 

In our review of the current GFEI pilot, we found a number of 
weaknesses that make it difficult to evaluate the program's 
effectiveness. For example, our research of past school feeding 
programs indicated that the programs are more likely to improve 
enrollment, attendance, and learning if they are carefully integrated 
with other educational, health, and nutritional interventions—such as 
ensuring adequate numbers of well-trained teachers and providing 
treatments for parasitic infections and micronutrient deficiencies. 
However, USDA began the GFEI pilot quickly and did not require 
potential implementing partners to provide important information on 
the linkages to these other interventions. Since most of the pilot's 
funding is targeted for the school meals, it is unclear whether these 
other important factors that contribute to effective programs are 
adequately addressed. In addition, USDA has not effectively managed 
the pilot in part because of its lack of expertise and resources for 
food for education development programs. It has not set specific 
targets or desired performance levels for enrollment and attendance in 
its agreements with most of its implementing partners. WFP has 
recently collected baseline data on enrollment and attendance, and 
USDA is in the process of doing so. USDA will not try to measure the 
projects' impacts on learning, as it believes the 1-year time frame is 
too short for such an assessment.[Footnote 83] Because of these 
weaknesses, we do not believe the pilot program will yield adequate 
information on whether its projects have succeeded or failed in 
improving enrollment, attendance, and learning—and why. Furthermore, a 
number of other donor countries will not contribute to GFEI until they 
see if the pilot is successful. These are important concerns as the 
Congress considers what actions to take regarding legislation on GFEI. 

Matters for Congressional Consideration: 

As the Congress decides whether to further fund GFEI, it may wish to 
consider: 

* extending the pilot program to permit an assessment of its effects on
learning, as well as a more meaningful review of its impact's on 
enrollment and attendance; 

* deciding whether additional funding for pilot project related 
activities, such as teacher training and textbooks, may be needed for 
effective projects; 

* assuring that the administering agency has sufficient expertise and 
staff resources to effectively manage the program; and; 

* requiring the administering agency to establish measurable 
performance indicators to monitor progress and evaluate project 
results. 

Agency Comments and Our Evaluation: 

We received written comments on a draft of this report from USDA, 
USAID, and the Office of Management and Budget (OMB) that are 
reprinted in appendixes XII, XIII, and XIV. These agencies also 
provided technical comments, which we incorporated in this report as 
appropriate. The Department of State's liaison for GAO told us that 
State believes the report findings are essentially factual and correct 
and opted not to comment further. We also obtained technical comments 
on parts of the report from the World Bank, WFP, and six PVOs and have 
incorporated them as appropriate. 

In its comments, USDA reiterated a number of key points and findings 
that were in the draft report and provided some additional information 
about certain aspects of the pilot program. Beyond that, USDA said it 
believes we have taken an overly critical view of how it has 
administered the pilot program, given time and resource constraints. 
Our draft report cited time and resource limitations as key factors 
affecting the management and possible effectiveness of the program. 
USDA also said it believes the report fails to recognize that the 
president directed a school feeding program, not an entire educational 
program. We disagree with this statement. We clearly said—-as the 
White House did on the day the program was announced and as USDA 
itself did in its comments—-that the pilot is a school feeding program 
with the three purposes of improving student enrollment, attendance, 
and learning. 

USAID said our draft report accurately and fairly depicted the complex 
and formidable challenges confronting the GFEI, fully endorsed our 
matters for congressional consideration, and said the findings and 
matters should be of great use to the Congress as it debates the 
structure of U.S. food assistance. USAID observed that the pilot 
placed priority on getting the program up and running, with program 
designers believing that improvements could then be made that would 
address issues of cost, sustainability, and the need for complementary 
programs. 

OMB commented that the draft report was balanced and generally 
accurate and would serve the Congress and the public in future 
deliberations about school feeding programs. OMB also said that the 
principal criticisms of the pilot program problems may be attributable 
to the urgency with which the program was generated. In addition, OMB 
said, greater emphasis was placed on the nutritional goals of the 
pilot rather than education objectives. One could expect that some of 
these problems could be addressed by a more deliberate approach to 
performance and evaluation, it said. 

We are sending copies of this report to interested congressional 
committees and the secretary of state; secretary of agriculture; and 
the administrator, USAID. Copies will also be made available to others 
upon request. 

If you or your staff have any questions about this report, please 
contact me on (202) 512-4347. Other GAO contacts and staff 
acknowledgments are listed in appendix XII. 

Signed by: 

Loren Yager, Director: 
International Affairs and Trade: 

[End of section] 

Appendix I: Scope and Methodology: 

We obtained information on the Global Food for Education Initiative 
(GFEI) and pilot program from U.S. government officials at the 
Departments of Agriculture (USDA) and State, as well as officials from 
the Agency for International Development (USAID), the Office of 
Management and Budget (OMB), and the White House. We also obtained 
information from officials of the World Food Program (WFP), foreign 
donor governments, and representatives of private voluntary 
organizations. In addition, we met with representatives of the 
European Commission and the World Bank, and experts from private 
research institutions. We conducted our review in Washington, D.C.; 
Rome, Italy; and Brussels, Belgium. 

Our review addressed lessons learned from past international school 
feeding programs, the application of lessons learned to the pilot 
program, an assessment of the design and implementation phase of the 
pilot project, the impact of the GFEI on recipient country 
agricultural markets, and the commitment of other donor countries to 
the initiative. Our review did not address the in-country phase of the 
pilot program because projects were not operational during most of the 
time of our review. Our contact with PVOs was limited because most of 
their agreements were not finalized until we had completed most of our 
field work. 

To examine the lessons learned about the effectiveness and cost of 
school feeding programs in promoting increased school enrollment, 
attendance, and performance, we reviewed studies completed by the U.S. 
government, international organizations, private voluntary 
organizations, and private research institutions. We also met with 
selected experts in international school feeding. We reviewed the 
studies in terms of past programs' impact on enrollment, attendance, 
and learning. In reviewing studies and meeting with experts, we also 
identified key factors common to effective school feeding programs. 
Through our analysis of information from World Bank and WFP, we also 
compared estimated costs of various school feeding programs. 

To examine the extent to which the U.S. pilot program has been built 
upon the lessons learned from previous school feeding programs, we met 
with senior officials of the USDA and State, USAID, the White House, 
and OMB, as well as representatives of private voluntary 
organizations, research institutions, and international organizations. 
We also reviewed program decisionmaking documents. We compared 
information obtained from these sources to key conclusions of past 
international school feeding studies and the views of various experts. 

To determine whether the U.S. pilot program was designed and 
implemented to reasonably ensure that the food aid and monetized 
proceeds were used effectively and efficiently, we gathered 
information and met with officials from the USDA, USAID, the White 
House, and OMB. We also obtained information from private voluntary 
organizations and WFP. We reviewed pilot program guidance, proposals, 
and relevant laws and regulations governing the development and 
administration of the pilot project. We also gathered and analyzed a 
variety of key pilot project information to provide estimates of 
tonnage, project costs, and number of beneficiaries by cooperating 
sponsor. We assessed selected information in proposals for approved 
pilot projects and nonmeal program components of these projects, 
including the amount budgeted and number of project beneficiaries. We 
applied our governmentwide internal control standards in evaluating 
the pilot project's management and financial controls. 

To determine the views of other major food aid donors regarding 
support for a comprehensive, long-term global food for education 
initiative, we gathered information and met with officials from donor 
countries including Australia, Canada, Denmark, Finland, France, 
Germany, Italy, Japan, the Netherlands, Sweden, and the European 
Commission. We developed an analytical framework to summarize their 
individual and collective views on how food aid should be provided in 
terms of emergencies, development, cash, or food-in-kind. 

We conducted our review from November 2000 through December 2001 in 
accordance with generally accepted government auditing standards. 

[End of section] 

Appendix II: Pilot Program Projects' Implementing Partners, Countries, 
Agreement Status, Tonnage, Cost, and Beneficiaries: 
					
Implementing partners: Government of the Dominican Republic; 
Recipient country: Dominican Republic; 
Date of agreement: 6/25/01; 
Estimated tonnage for project[A]: 62,200; 
Estimated project cost[A]: $14,147,783; 
Number of beneficiaries: 1,000,000 

PVO sponsors: 

Implementing partners: AC DINOCA; 
Recipient country: Uganda; 
Date of agreement: Not signed; 
Estimated tonnage for project[A]: 8,710; 
Estimated project cost[A]: $2,776,030; 
Number of beneficiaries: 40,000. 
					
Implementing partners: Adventist Development & Relief Agency (ADRA); 
Recipient country: Bolivia; 
Date of agreement: Not signed; 
Estimated tonnage for project[A]: 6,270; 
Estimated project cost[A]: $2,806,697; 
Number of beneficiaries: 88,000. 

Implementing partners: ADRA; 
Recipient country: Madagascar; 
Date of agreement: 8/16/01; 
Estimated tonnage for project[A]: 4,900; 
Estimated project cost[A]: $1,818,554; 
Number of beneficiaries: 50,000. 

Implementing partners: ADRA; 
Recipient country: Yemen; 
Date of agreement: Not signed; 
Estimated tonnage for project[A]: 5,000; 
Estimated project cost[A]: $2,115,510; 
Number of beneficiaries: 66,000. 

Implementing partners: Cooperative Assistance for Relief Everywhere 
(CARE); 
Recipient country: Albania; 
Date of agreement: 8/13/01; 
Estimated tonnage for project[A]: 18,500; 
Estimated project cost[A]: $4,724,901; 
Number of beneficiaries: 16,000. 
					
Implementing partners: Catholic Relief Services (CRS); 
Recipient country: Albania; 
Date of agreement: 7/27/01; 
Estimated tonnage for project[A]: 740; 
Estimated project cost[A]: $1,279,139; 
Number of beneficiaries: 4,000. 

Implementing partners: CRS; 
Recipient country: Benin; 
Date of agreement: 8/23/01; 
Estimated tonnage for project[A]: 3,350; 
Estimated project cost[A]: $1,788,931; 
Number of beneficiaries: 10,000. 

Implementing partners: CRS; 
Recipient country: Bosnia/Herzegovina; 
Date of agreement: 6/20/01; 
Estimated tonnage for project[A]: 24,630; 
Estimated project cost[A]: $4,804,912; 
Number of beneficiaries: 30,000. 

Implementing partners: CRS; 
Recipient country: Guatemala; 
Date of agreement: 11/5/01; 
Estimated tonnage for project[A]: 27,630; 
Estimated project cost[A]: $5,078,170; 
Number of beneficiaries: 27,000. 

Implementing partners: CRS; 
Recipient country: Honduras; 
Date of agreement: 11/30/01; 
Estimated tonnage for project[A]: 15,100; 
Estimated project cost[A]: $3,957,800; 
Number of beneficiaries: 10,000. 

Implementing partners: Counterpart International (CPI); 
Recipient country: Georgia; 
Date of agreement: Not signed; 
Estimated tonnage for project[A]: 26,600; 
Estimated project cost[A]: $6,138,719; 
Number of beneficiaries: 50,000. 

Implementing partners: CPI; 
Recipient country: Senegal; 
Date of agreement: 8/20/01; 
Estimated tonnage for project[A]: 7,550; 
Estimated project cost[A]: $3,311,178; 
Number of beneficiaries: 54,000. 

Implementing partners: International Partnership for Human Development 
(IPHD); 
Recipient country: Republic of Congo; 
Date of agreement: 7/3/01; 
Estimated tonnage for project[A]: 18,300; 
Estimated project cost[A]: $7,146,715; 
Number of beneficiaries: 100,000. 

Implementing partners: IPHD; 
Recipient country: Moldova; 
Date of agreement: 7/06/01; 
Estimated tonnage for project[A]: 28,400; 
Estimated project cost[A]: $11,796,455; 
Number of beneficiaries: 300,000. 

Implementing partners: International Orthodox Christian Charities 
(IOCC); 
Recipient country: Georgia; 
Date of agreement: 4/20/01; 
Estimated tonnage for project[A]: 10,800; 
Estimated project cost[A]: $2,324,592; 
Number of beneficiaries: 14,000. 

Implementing partners: IOCC; 
Recipient country: Lebanon; 
Date of agreement: 6/26/01; 
Estimated tonnage for project[A]: 27,000; 
Estimated project cost[A]: $4,906,330; 
Number of beneficiaries: 19,400. 

Implementing partners: Land O'Lakes (LOL); 
Recipient country: Bangladesh; 
Date of agreement: 11/15/01; 
Estimated tonnage for project[A]: 34,950; 
Estimated project cost[A]: $11,254,050; 
Number of beneficiaries: 500,000. 

Implementing partners: LOL; 
Recipient country: Vietnam; 
Date of agreement: 8/3/01; 
Estimated tonnage for project[A]: 43,300; 
Estimated project cost[A]: $10,781,997; 
Number of beneficiaries: 400,000. 

Implementing partners: Mercy Corps International (MCI); 
Recipient country: Eritrea; 
Date of agreement: 8/14/01; 
Estimated tonnage for project[A]: 17,430; 
Estimated project cost[A]: $8,841,361; 
Number of beneficiaries: 35,000. 

Implementing partners: MCI; 
Recipient country: Kyrgyzstan; 
Date of agreement: 8/03/01; 
Estimated tonnage for project[A]: 5,440[B]; 
Estimated project cost[A]: $5,169,168; 
Number of beneficiaries: 20,000. 

Implementing partners: Mercy USA (MUSA); 
Recipient country: Albania; 
Date of agreement: 12/10/01; 
Estimated tonnage for project[A]: 10,000; 
Estimated project cost[A]: $4,584,836; 
Number of beneficiaries: 60,000. 

Implementing partners: Project Concern International (PCI); 
Recipient country: Bolivia; 
Date of agreement: 8/23/01; 
Estimated tonnage for project[A]: 8,950; 
Estimated project cost[A]: $7,107,677; 
Number of beneficiaries: 120,000. 

Implementing partners: PCI; 
Recipient country: Nicaragua; 
Date of agreement: 3/16/01; 
Estimated tonnage for project[A]: 3,960; 
Estimated project cost[A]: $2,059,722; 
Number of beneficiaries: 19,200. 

Implementing partners: Save the Children (STC); 
Recipient country: Uganda; 
Date of agreement: 8/24/01; 
Estimated tonnage for project[A]: 640; 
Estimated project cost[A]: $849,135; 
Number of beneficiaries: 5,000. 

Implementing partners: World Share (WS); 
Recipient country: Guatemala; 
Date of agreement: 8/06/01; 
Estimated tonnage for project[A]: 20,980; 
Estimated project cost[A]: $3,661,139; 
Number of beneficiaries: 90,000. 

Total PVOs (25); 
Estimated tonnage for project[A]: 379,130; 
Estimated project cost[A]: $121,083,718; 
Number of beneficiaries: 2,154,000. 

World Food Program: 

Implementing partners: WFP projects approved by WFP's executive 
board[B] 

Recipient country: Bhutan; 
Date of agreement: 4/18/01; 
Estimated tonnage for project[A]: 1,750; 
Estimated project cost[A]: $880,902; 
Number of beneficiaries: 17,500. 

Recipient country: Bolivia; 
Date of agreement: 5/02/01; 
Estimated tonnage for project[A]: 7,880; 
Estimated project cost[A]: $1,917,890; 
Number of beneficiaries: 102,000. 

Recipient country: Cambodia; 
Date of agreement: 3/16/01; 
Estimated tonnage for project[A]: 1,660; 
Estimated project cost[A]: $818,448; 
Number of beneficiaries: 100,000. 

Recipient country: Cameroon; 
Date of agreement: 5/02/01; 
Estimated tonnage for project[A]: 1,058; 
Estimated project cost[A]: $410,588; 
Number of beneficiaries: 49,000. 

Recipient country: Chad; 
Date of agreement: 4/28/01; 
Estimated tonnage for project[A]: 1,170; 
Estimated project cost[A]: $1,342,682; 
Number of beneficiaries: 43,600. 

Recipient country: Colombia; 
Date of agreement: 3/16/01; 
Estimated tonnage for project[A]: 3,655; 
Estimated project cost[A]: $758,298; 
Number of beneficiaries: 30,000. 

Recipient country: Cote d'Ivoire; 
Date of agreement: 3/21/01; 
Estimated tonnage for project[A]: 700; 
Estimated project cost[A]: $313,185; 
Number of beneficiaries: 200,000. 

Recipient country: Dominican Republic; 
Date of agreement: 4/18/01; 
Estimated tonnage for project[A]: 310; 
Estimated project cost[A]: $137,300; 
Number of beneficiaries: 115,000. 

Recipient country: El Salvador; 
Date of agreement: 5/23/01; 
Estimated tonnage for project[A]: 9,040; 
Estimated project cost[A]: $2,707,376; 
Number of beneficiaries: 175,800. 

Recipient country: Ethiopia; 
Date of agreement: 4/26/01; 
Estimated tonnage for project[A]: 3,990; 
Estimated project cost[A]: $2,504,174; 
Number of beneficiaries: 130,000. 

Recipient country: Gambia; 
Date of agreement: 3/27/01; 
Estimated tonnage for project[A]: 900; 
Estimated project cost[A]: $444,557; 
Number of beneficiaries: 37,500. 

Recipient country: Ghana; 
Date of agreement: 3/27/01; 
Estimated tonnage for project[A]: 1,065; 
Estimated project cost[A]: $544,488; 
Number of beneficiaries: 12,900. 

Recipient country: Guinea; 
Date of agreement: 3/23/01; 
Estimated tonnage for project[A]: 150; 
Estimated project cost[A]: $101,200; 
Number of beneficiaries: 82,000. 

Recipient country: Honduras; 
Date of agreement: 5/18/01; 
Estimated tonnage for project[A]: 9,450; 
Estimated project cost[A]: $3,348,209; 
Number of beneficiaries: 164,000. 

Recipient country: Kenya; 
Date of agreement: 4/19/01; 
Estimated tonnage for project[A]: 68,500; 
Estimated project cost[A]: $29,684,445; 
Number of beneficiaries: 1,362,000. 

Recipient country: Kenya; 
Date of agreement: 6/18/01; 
Estimated tonnage for project[A]: 2,400; 
Estimated project cost[A]: $638,544; 
Number of beneficiaries: 47,400. 

Recipient country: Mozambique; 
Date of agreement: 4/12/01; 
Estimated tonnage for project[A]: 2,300; 
Estimated project cost[A]: $1,337,831; 
Number of beneficiaries: 56,800. 

Recipient country: Nepal; 
Date of agreement: 4/26/01; 
Estimated tonnage for project[A]: 200; 
Estimated project cost[A]: $139,495; 
Number of beneficiaries: 250,000. 

Recipient country: Nicaragua; 
Date of agreement: 3/16/01; 
Estimated tonnage for project[A]: 970; 
Estimated project cost[A]: $474,432; 
Number of beneficiaries: 80,000. 

Recipient country: Nicaragua; 
Date of agreement: 5/17/01; 
Estimated tonnage for project[A]: 14,960; 
Estimated project cost[A]: $9,156,468; 
Number of beneficiaries: 351,000. 

Recipient country: Peru; 
Date of agreement: 5/15/01; 
Estimated tonnage for project[A]: 10,000; 
Estimated project cost[A]: $2,589,400; 
Number of beneficiaries: 137,600. 

Recipient country: Tajikistan; 
Date of agreement: 3/23/01; 
Estimated tonnage for project[A]: 380; 
Estimated project cost[A]: $193,467; 
Number of beneficiaries: 16,100. 

Recipient country: Tanzania; 
Date of agreement: 5/23/01; 
Estimated tonnage for project[A]: 2,050; 
Estimated project cost[A]: $956,362; 
Number of beneficiaries: 61,300. 

Recipient country: Uganda; 
Date of agreement: 3/16/01; 
Estimated tonnage for project[A]: 6,060; 
Estimated project cost[A]: $3,621,898; 
Number of beneficiaries: 66,000. 

Total[C] approved (25)[D]; 
Estimated tonnage for project[A]: 151,818; 
Estimated project cost[A]: $65,980,230; 
Number of beneficiaries: 3,638,000. 

Implementing partners: Expansion proposals subject to executive board 
approval[B]: 

Recipient country: Bhutan; 
Date of agreement: 7/03/01; 
Estimated tonnage for project[A]: 1,070; 
Estimated project cost[A]: $583,561; 
Number of beneficiaries: 28,500. 

Recipient country: Chad; 
Date of agreement: 8/03/01; 
Estimated tonnage for project[A]: 3,170; 
Estimated project cost[A]: $2,053,200; 
Number of beneficiaries: 87,400. 

Recipient country: Ethiopia; 
Date of agreement: 7/03/01; 
Estimated tonnage for project[A]: 6,940; 
Estimated project cost[A]: $5,130,872; 
Number of beneficiaries: 233,700. 

Recipient country: Gambia; 
Date of agreement: 7/13/01; 
Estimated tonnage for project[A]: 2,570; 
Estimated project cost[A]: $1,923,249; 
Number of beneficiaries: 112,500. 

Recipient country: Mozambique; 
Date of agreement: 6/28/01; 
Estimated tonnage for project[A]: 6,500; 
Estimated project cost[A]: $3,952,469; 
Number of beneficiaries: 170,300. 

Recipient country: Nepal; 
Date of agreement: 7/03/01; 
Estimated tonnage for project[A]: 5,962; 
Estimated project cost[A]: $4,726,256; 
Number of beneficiaries: 537,000. 

Recipient country: Pakistan; 
Date of agreement: 8/03/01; 
Estimated tonnage for project[A]: 5,860; 
Estimated project cost[A]: $5,785,880; 
Number of beneficiaries: 175,800. 

Recipient country: Tajikistan; 
Date of agreement: 8/09/01; 
Estimated tonnage for project[A]: 4,080; 
Estimated project cost[A]: $2,341,519; 
Number of beneficiaries: 172,700. 

Total expansion (8)[D]; 
Estimated tonnage for project[A]: 36,152; 
Estimated project cost[A]: $26,497,006; 
Number of beneficiaries: 1,517,900
		
Total WFP (33)[D]; 
Estimated tonnage for project[A]: 187,970; 
Estimated project cost[A]: $92,477,236; 
Number of beneficiaries: 5,155,900
		
Grand total: 
Estimated tonnage for project[A]: 629,300; 
Estimated project cost[A]: $227,708,737; 
Number of beneficiaries: 8,309,900. 

[A] USDA estimates as of February 21, 2002, for WFP projects and 
December 11, 2001, for other projects. 

[B] At the time when WFP submitted its proposals to USDA. 

[C] Does not include a late fiscal year 2002 shipment of 2,350 metric 
tons. 

[D] Some projects involve multiple commitments. The United States 
approved 34 WFP proposals covering 27 WFP projects in 23 countries. Of 
the 34 proposals, 8 were for expansions of already existing school 
feeding projects. The United States approved two different projects 
each for Guinea, Kenya, Nicaragua, and Uganda. As of February 21, 
2002, USDA and WFP were still negotiating the terms of the second 
project for Guinea, and no figures for this project are shown in the 
table. 

[End of table] 

[End of section] 

Appendix III: The World Food Program's Role in School Feeding and Food 
for Education: 

The World Food Program (WFP), set up in 1963, is a major U.N. agency 
in the fight against global hunger. In 2000, WFP fed 83 million people 
in 83 countries, including most of the world's refugees and internally 
displaced people. It shipped 3.5 million tons of food; received $1.75 
billion in donations; and had operational expenditures of $1.49 
billion (provisional figures).[Footnote 84] 

WFP provides three basic kinds of food aid: (1) emergency assistance 
to cope with the adverse food effects of natural disasters, civil 
conflict, and war; (2) protracted relief or rehabilitation aid to help 
people rebuild their lives and communities once the causes of 
emergencies recede; and (3) development assistance that aims to make 
communities food secure so they can devote time, attention, and work 
to escaping the poverty trap. When WFP was founded, its food 
assistance primarily focused on development, and for years development 
projects accounted for more than two-thirds of its expenditures. 
However, during the past 15 years, WFP has become increasingly 
involved in responding to humanitarian emergencies. According to WFP 
officials, WFP devoted 28 percent of its resources to development in 
1997, 18 percent in 1999, and only 13 percent in 2000. 

WFP relies entirely on voluntary contributions to finance its 
projects. Governments are the principal source of funding, but 
corporations, foundations, and individuals also contribute. Donations 
are made either as cash, food (such as grains, beans, oil, salt, and 
sugar), or the basic items necessary to grow, cook, and store food—
kitchen utensils, agricultural tools, and warehouses. Since it has no 
independent source of funds, WFP's Executive Board has mandated that 
all food donations, whether in cash or in-kind, must be accompanied by 
the cash needed to move, manage, and monitor the food aid. 

WFP has been running school feeding programs for nearly 40 years. In 
1999, it operated 76 school feeding projects in 48 developing 
countries. These included 33 emergency or protracted relief projects 
that had 5.28 million beneficiaries and 43 development projects that 
had 5.85 million beneficiaries. Thus, total beneficiaries were 11.13 
million. In 2000, WFP operated 68 projects in 54 countries, with a 
total of 12.27 million beneficiaries. According to WFP, the total 
expenditure for its school feeding operations in 2000 was 
approximately $421 million. About $239 million was for development 
projects focused on school feeding, and the remainder was for school 
feeding components of emergency or protracted relief and recovery 
operations. 

WFP welcomed President Clinton's July 23, 2000, announcement of the 
$300 million pilot program to launch a universal school feeding 
program, noted that it had been working closely with the U.S. 
ambassador to the U.N. Food Agencies in Rome to assist in the creation 
of such a program, and expressed the hope that the initiative would 
become a permanent feature of the global community of nations. A few 
days later, WFP's executive director, in testimony before a U.S. 
Senate committee, said a global program needs to be managed by a 
global organization and WFP, as the food aid arm of the U.N., was 
uniquely qualified to manage the initiative.[Footnote 85] 

Regarding its role in implementing a global program, WFP has said that 
much could be done to strengthen the education system in many 
developing countries.[Footnote 86] According to WFP, this a highly 
complex task, one for which food aid is not the most effective 
resource.[Footnote 87] WFP's approach will be to use food aid where 
the food is needed. WFP does not propose to monetize food commodities 
to fund related educational support activities. WFP will monetize only 
to effect an exchange between donated commodities and locally grown 
foods when this is cost effective and does not have an adverse effect 
on local markets. At the same time, WFP recognizes that while school 
feeding can bring children to school and help them learn while they 
are there, school feeding does not ensure qualified teachers, books 
and supplies, or a suitable curriculum. According to WFP, this is the 
role of national governments, often supported by international 
organizations or Private Voluntary Organizations (PVO); and the 
relationship between improvements in an education system and a 
national system of school feeding is one that must be managed by 
governments. 

However, within the broad framework of government cooperation, WFP 
said, it is eager to work with other operational partners and 
experienced in doing so. 

Underfunding of Projects: 

WFP told us that many of its school feeding projects have shortfalls. 
[Footnote 88] Funding for all components of approved projects, 
including current school feeding programs, depends on the level of 
contributions received. When and where possible, WFP will allocate 
unearmarked donations to underfunded projects, taking into 
consideration the urgency of the need and a need to comply with the 
executive board's approved allocation formula. 

According to WFP, it usually is not technically feasible to identify 
how many children were not fed due to under-resourcing. An unstable 
resourcing situation often compels project managers to temporarily 
adjust the on-site ration size or the number of food distribution 
days, rather than reducing the number of beneficiaries, it said. When 
under-resourcing is of a more permanent nature, the project plan is 
revised and a formal change in the beneficiaries occurs. 

WFP'S Approach to Certain Key Factors Associated with Effective School 
Feeding Programs: 

WFP has developed several documents that describe its policies for 
establishing school feeding programs and which guide the project 
development and approval process for all WFP school feeding 
activities.[Footnote 89] The following is a brief summary of some of 
the points presented in these documents, or provided directly to us by 
WFP in response to questions that we provided to the agency, regarding 
certain key factors associated with their school feeding programs. 

* Targeting: The focus of WFP's world school feeding initiative is on 
feeding preschool and primary school age children. On an exceptional 
basis, food aid activities designed to encourage girls to continue 
their education beyond primary school will be considered. Some 
fundamental issues to be examined in determining the problems to be 
addressed are (1) enrollment and dropout rates in primary education 
broken down by gender, region and sociocultural groups, to the extent 
possible, and factors explaining these rates; (2) extent of, and 
factors contributing to, short-term hunger; (3) average distances 
walked by the students, who will be covered in the school feeding 
activity, between their homes and their school; and (4) cultural 
practices affecting enrollment/attendance, especially of girls. 

As a general rule, targeting within school feeding projects will be 
conducted at the level of geographic areas, with no selection of 
individual pupils within schools. The only exception for this may be 
when the effectiveness of an incentive for a particular category 
(e.g., girls) can be demonstrated. According to WFP, it requires at 
least 50 percent of its resources in education to be targeted for 
girls, and WFP has been very successful in achieving this requirement. 

WFP has a vulnerability analysis and mapping unit (VAM) to identify 
people most vulnerable to hunger and to target their needs. According 
to WFP, VAM uses state of the art satellite imagery of rainfall and 
crop conditions, as well as monitoring of food prices in local 
markets. WFP has VAM sub-units in more than 50 developing countries. 
According to WFP, this system is also used in targeting its school 
feeding programs. 

* Facilitative learning environment: WFP told us that it does not 
require a facilitative learning environment to be in place or provided 
as part of its programs, but such an environment is highly desired and 
encouraged. According to WFP, the presence of school feeding in 
schools helps bring attention to other school conditions (e.g., 
classrooms, materials, sanitary facilities, teachers, curricula, and 
health conditions) and, in turn, helps WFP and its partners to bring 
attention to problems and attract other needed resources. 

* Safe water and sanitation: WFP guidelines say basic water supply and 
sanitation standards must be met if food is to be safely stored and 
prepared for school feeding, and safe water supply should be available 
on the school premises at all times. WFP provides detailed information 
on optimal and minimal standards for a safe water supply and 
sanitation at schools. However, WFP told us it does not require safe 
water and sanitation facilities to be in place in order to implement 
school feeding in a given school and, as a rule, does not provide 
water and sanitation facilities. However, WFP said, it does work with 
the national and local governments and with other U.N. agencies, 
donors, and nongovernmental organizations who have the appropriate 
skills and resources to "trigger" action where the lack of such 
facilities is a problem. 

* Deworming treatments: According to WFP guidelines, WFP will 
generally support deworming in a school feeding program when more than 
50 percent of the children have intestinal parasites. Treatment is 
with a single dose of the proper medicine, up to three times a year, 
and should be combined with improved sanitation and safe water supply, 
as well as health education on prevention. In April 2001, WFP told us 
that it did not yet have complete information regarding which of its 
school feeding programs had already initiated deworming activities 
(due to decentralized decision-making and no prior requirements for 
reporting such information). However, WFP said it did know that most 
or all of its school feeding operations in Latin America and the 
Caribbean and two or more in Asia had at least implemented limited 
deworming activities. WFP estimated that by the end of 2001, it would 
have initiated deworming in its school feeding programs in 15 or more 
countries, in partnership with WHO and the World Bank, and assisted, 
in part, by a Canadian grant. WFP said that it hopes to achieve 
deworming activities in most or all GFEI, as well other WFP school 
feeding operations. WFP also noted that national, regional, or local 
governments may require deworming to be in place. 

* Micronutrient supplementation: WFP guidelines note that school 
feeding can be a vehicle for micronutrients in countries where school 
children are affected by and/or at high risk of developing 
micronutrient deficiencies. WFP provides information on micronutrient 
deficiencies that have been shown to affect school attendance and 
performance, recommended levels of intake of these micronutrients for 
3- to 12-year old children, and guidance on how to use them in school 
meals. WFP told us that micronutrient supplementation is most often 
handled as an additive to the commodities that are distributed. In 
cases where the commodities that arrive are not fortified, WFP most 
often works locally to fortify the food or seeks other remedies. WFP 
collaborates with groups that have expertise and resources to bring to 
bear, especially UNICEF, WHO, a Canadian micronutrient initiative, and 
certain NGOs. WFP noted that national, regional, or local governments 
may require micronutrient supplementation to be in place. 

* Health and nutrition education: WFP told us that this is not 
strictly required in all WFP school feeding operations. However, such 
activities are highly encouraged, are frequently planned and 
implemented, and will be further strengthened through collaboration 
with appropriate partners and coworkers on the ground. WFP noted that 
national, regional, or local governments may require health and 
nutrition education to be in place. 

* Community and parental participation: WFP told us that community and 
parental participation are not strictly required in all WFP school 
feeding operations. However, WFP said, such activities are highly 
encouraged,[Footnote 90] are frequently planned and implemented, and 
are and will be further strengthened through collaboration with 
appropriate partners and coworkers on the ground. WFP noted that its 
data indicates that as girls' enrollment and attendance increases, so 
does parental participation. WFP also noted that national, regional, 
or local governments may require parental involvement to be in place. 

* Education for All: WFP expects recipient governments to have 
demonstrated a commitment to Education for All. 

* Sustainability: WFP requires that plans be in place for eventual 
takeover of a feeding program by recipient countries. WFP generally 
insists that programs be supported by national governments and local 
communities and that resources and infrastructure be provided as 
counterpart contributions. However, WFP will consider providing school 
feeding activities in some emergency and protracted relief situations 
where full government support is not possible. In addition, for low 
income countries, it is probably necessary to provide most or all of 
the food commodities, technical assistance, and equipment.
According to a WFP official, sustainability depends on the economic 
status of the recipient country. There are countries where the 
national government has been able to take over a program. However, in 
the poorest, least developed countries, he said, sustainability is 
only possible where there is substantial community involvement. In 
many least developed countries, government expenditure on the 
education sector often represents up to 30 percent of the national 
budget; it is difficult enough for such countries to maintain the 
physical infrastructure and teachers. For least developed countries, 
sustainability is a long-term process. A realistic estimate is 10 to 
15 years, he said. 

Monitoring and Evaluation: 

WFP officials told us that there had been some problems in the past, 
but WFP is working hard to overcome them for both the U.S. pilot 
program and its other school feeding activities. As an example of 
problems, collection of baseline date had varied, depending on the 
country, the specific goals of the school feeding activity, and the 
resources available. Principal performance indicators that WFP tended 
to use were increased enrollment and attendance, reduced dropout 
rates, and improved performance (such as number of students who had 
completed primary school the previous year and gone on to higher 
education). WFP had looked at these indicators, especially as they 
relate to girls' education, and had been able to report some notable 
successes. However, WFP had only done that in isolated 
cases/countries. Therefore, WFP intends under GFEI to standardize the 
indicators and upgrade its monitoring and evaluation systems so as to 
be able to regularly collect and report comparable and up-to-date data 
for its school feeding operations. WFP also said that data collection 
and analysis in developing countries is challenging and requires 
additional resources and capacity building of national counterpart 
staff. 

WFP's guidelines for its new World School Feeding Initiative require a 
baseline monitoring study to establish the situation prior to the 
onset of the initiative, followed by periodic updates as a program is 
implemented.[Footnote 91] To this end, WFP developed a detailed survey 
instrument for collecting baseline information on its GFEI-funded 
projects. The survey was pilot-tested in August 2001, and WFP 
conducted the surveys in a sample of schools[Footnote 92] for all of 
the U.S. pilot program projects before the end of 2001 (details of the 
survey instrument are discussed in the letter). 

In addition, according to WFP, during 2001, it developed and 
successfully pilot-tested a new system of collecting key monitoring 
data on a timely basis directly from the schools involved in its 
feeding programs. The system involves school staff entering key data 
directly into devices, installed at the schools, that transmit the 
data via satellite to a data collection center in France, using the 
ARGOS satellite system (that is jointly managed by the governments of 
France and the United States). Country data is then reported from the 
data collection center to the country's relevant ministry of education 
and to WFP. WFP is seeking donors to fund implementation of the system. 

WFP also conducted a major, global survey of national school feeding 
programs (not specific projects) between May and December 2001. The 
survey collected information on countries' school feeding programs and 
related information on their demography; education system; 
nongovernmental program assistance; health-related education services 
at school; and evaluations, studies, and surveys about school feeding 
and related topics. According to WFP, the survey provides a focal 
point for school feeding information, which WFP will use to promote 
dialogue with governments and nongovernmental organizations concerning 
the use of food aid for education and related issues. WFP will also 
use the data to produce special reports and identify country specific 
needs and coordinate partnerships between countries with experience in 
school feeding and those in need. WFP is posting country-specific 
results on its Web site. WFP is seeking donors to fund installation of 
the system in its schools. 

Regarding evaluations, WFP's central evaluation office generally does 
not conduct separate evaluations of the school feeding projects that 
WFP assists. (Occasionally separate evaluations of school feeding 
projects are undertaken if specifically requested by the executive 
board.) WFP mandates that evaluations of its country programs[Footnote 
93] be conducted about every 4 years, on average. The evaluations are 
submitted to WFP's Executive Board for review. If a country has a 
school feeding project, the project's role, relevance, and performance 
as an activity is to be included in the review. 

[End of section] 

Appendix IV: Results from Review of Experts' Findings and Views on 
School Feeding Programs: 

This appendix provides additional information on our review of 
experts' findings and views concerning (1) the effect of school 
feeding programs on enrollment and attendance, (2) the effect of 
school feeding programs on educational performance or learning, and 
(3) key factors contributing to effective school feeding programs (see 
tables 4 and 5). It also provides further information on key factors 
associated with effective school feeding programs (see tables 6 
through 10). (See also appendix V, which discusses the costs and cost 
effectiveness of school feeding programs.) 

Our review relied considerably on the views of two experts who have 
reviewed the results of many school feeding program studies;[Footnote 
94] WFP, which has conducted school feeding programs for 4 decades and 
also reviewed the results of other studies;[Footnote 95] and the 
summary views of a meeting of experts and practitioners held at USAID 
in October 2000.[Footnote 96] We also conducted literature searches, 
reviewed the results of individual studies on school feeding programs, 
and spoke with experts and practitioners. 

Table 4 summarizes the results of studies and expert views on the 
relationship between school feeding and school enrollment and 
attendance. 

Table 4: Results From Selected Studies and Experts on the Impacts of 
School Feeding Programs on Enrollment and Attendance: 

Country: Bangladesh; 
Finding: Student enrollment in schools with a food for education 
program increased by 35 percent immediately after the program was 
introduced, while enrollment in other schools increased by about 7 
percent. The overall rate of school attendance was 71 percent in 
schools with the program and 58 percent in other schools; 
Source[A]: IFPRI, 2001. 

Country: Developing countries (generally); 
Finding: Properly designed and effectively implemented school feeding 
programs (SFP) can motivate parents to enroll their children in school 
and have them attend regularly; 
Source[A]: Del Rosso, 1999. 

Country: Developing countries (generally); 
Finding: Review of literature prior to 1986--SFPs seemed to make a 
difference in enrollment and attendance when there was a good fit 
between the design of the program and the environment in which it 
operates; 
Source[A]: Levinger, 1986. 

Country: Developing countries (generally); 
Finding: Cites several studies of programs that have increased 
enrollment or lowered absenteeism and dropout rates; 
Source[A]: Whitman et al, 2000. 

Country: Developing countries (generally); 
Finding: SFPs have the advantage of bringing children into school in a 
way that other interventions (e.g., safe water and sanitation and 
health packages) do not. Under the right circumstances, favors feeding 
to get and keep children enrolled in school; 
Source[A]: Levinger, 2001. 

Country: Developing countries (generally); 
Finding: The evidence strongly suggests that SFPs can increase 
attendance rates, especially for girls, and that school feeding or 
take-home rations serve as incentives for enrolling children in school 
and encouraging daily attendance; 
Source[A]: Stakeholders, 2000. 

Country: Developing countries (generally); 
Finding: SFPs can be effective in increasing attendance and 
enrollment. However, they may not overcome other factors that lead 
parents to keep their children at home, such as poverty, a lack of 
roads to the school, or perceptions of low school quality; 
Source[A]: World Bank, 2001. 

Country: Developing countries (generally); 
Finding: Most research, despite limitations in design and validity of 
findings, supports a positive effect of school feeding programs on 
school attendance and enrollment; 
Source[A]: World Food Program, 1999. 

Country: Dominican Republic; 
Finding: Up to 25 percent of children-—especially children from rural 
areas and girls—-dropped out of school during a period without a 
school feeding program; 
Source[A]: King, 1990. 

Country: Jamaica; 
Finding: SFP had no discernible effect on school attendance; 
Source[A]: Chambers, 1991. 

Country: Kenya; 
Finding: The hypothesis that children in a school with a lunch program 
would have better nutritional status and school attendance compared 
with a school without a lunch program was not confirmed; 
Source[A]: Meme et al, 1998. 

Country: Malawi; 
Finding: Small pilot school feeding program over a 3-month period led 
to a 5 percent increase in enrollment and up to 35 percent improvement 
in attendance; 
Source[A]: WFP, 1996. 

Country: Pakistan; 
Finding: Providing one or two tins of oil to families whose girls 
attended school for 20 days per month improved enrollment by 76 
percent in participating schools, compared with 14 percent in the 
province overall. Attendance increased from 73 percent to 95 percent 
among participants; 
Source[A]: WFP, 1995 (b). 

Country: Cape Verde, Gambia, Mauritania, Niger; 
Finding: Impact of school canteens on enrollment was difficult if not 
impossible to determine. However, impact on attendance was easily and 
statistically ascertainable; 
Source[A]: WFP, 1995 (c). 

Country: India; 
Finding: Results suggest that the program did not positively affect 
aggregate enrollment but had a positive impact on attendance and drop-
out rates; 
Source[A]: Rajan & Jayakumar, 1992. 

[A] See references at the end of this appendix for complete citations 
for the tables in this appendix. Source: GAO review of the literature. 

[End of table] 

Table 5 summarizes the results of several studies and expert views on 
the relationship between school feeding and school performance. 

Table 5: Results From Selected Studies and Experts on the Impacts of 
School Feeding Programs on Learning: 

Country: Benin; 
Finding: Children in schools with food services scored significantly 
higher on second-grade tests than did children in schools without food 
services; 
Source[A]: Jarousse & Mingat, 1991. 

Country: Burkina Faso; 
Finding: In 22 of 30 provinces, the success rate on a national exam 
for sixth grade pupils was higher for schools that had school feeding. 
The eight exceptions were for schools in moderate or better-off 
provinces; 
Source[A]: Moore, 1994. 

Country: Chile; 
Finding: Studied 279 children, from low socioeconomic background—-and 
categorized as normal, wasted, or stunted-—on the effects of breakfast 
omission on cognitive performance. No consistent association was found 
between study condition and performance in short-term visual memory, 
problem solving, and attention tasks in any of the three nutritional 
groups. Results suggested that given a motivating short-term task and 
maintaining routine conditions, missing breakfast does not affect the 
cognitive performance of children; 
Source[A]: Lopez et al, 1993. 

Country: Chile, Jamaica, Great Britain, Peru, United States; 
Finding: Selective review of the literature on the effects of 
breakfast on cognition and school performance after 1978 and before 
1995. In at-risk subjects (defined by clinical history and 
anthropometry), a morning and overnight fast had adverse effects on 
cognition. Contradictions in the data from different studies prevent 
definitive conclusions on whether well-nourished children experience 
functional deficits. Well-conducted evaluations suggest the 
availability of feeding programs in public schools throughout the 
academic year increases the probability that children will eat 
breakfast and improve their educational status; 
Source[A]: Pollitt, 1995. 

Country: Developing countries (generally); 
Finding: The impact of SFPs on meeting educational objectives is 
uncertain, since little work evaluating them has been done. However, 
experience shows that properly designed and effectively implemented 
SFPs can alleviate short-term hunger in malnourished or otherwise well-
nourished schoolchildren. This effect helps to increase students' 
attention and concentration, producing gains in cognitive functioning 
and learning; 
Source[A]: Del Rosso, 1999. 

Country: Developing countries (generally); 
Finding: The level of a student's cognitive performance is, in part, a 
function of the adequacy of his diet. However, meaningful cognitive 
development will occur only when a facilitative learning environment 
is present to complement the food a child receives; 
Source[A]: Levinger, 2000. 
	
Country: Developing countries (generally); 
Finding: Evidence for the positive effect of preschool nutrition 
programs on educational performance, particularly when the programs 
reach very young children, is quite strong and there is strong 
evidence for an impact on school performance of nutrition 
interventions targeting short term hunger, especially breakfast or mid-
morning snack programs. At the same time, schools need adequately 
trained teachers, motivational textbooks, and other learning materials 
for adequate learning to take place; 
Source[A]: World Bank, 2001. 

Country: Developing countries (generally); 
Finding: The existing literature on the effects of school feeding on 
education is not fully conclusive. Although studies based on an 
appropriate experimental design usually succeed in capturing the 
positive effects of school feeding, most of the ordinary field 
evaluations of SFPs seem to be too crude to yield significant results. 
Sufficient evidence does exist to suggest that school feeding can 
enhance children's cognitive function by offsetting the effects of 
short-term hunger, especially among already undernourished children; 
Source[A]: World Food Program, 1999. 

Country: Developing countries (generally); 
Finding: Only when hunger is addressed and the child is in school can 
other factors--such as the quality of the teaching--become relevant. 
For a child who attends school but is hungry, it does not matter 
whether the schools are stimulating settings that encourage 
development and learning; 
Source[A]: World Food Program, 2000. 

Country: India; 
Finding: Problems of malnutrition and health could not be overcome by 
a school meal program, which provided less than 15 percent of the 
recommended daily allowance for calories. However, the program did 
improve school attendance and academic performance, as well as reduce 
the dropout rate; 
Source[A]: Agarwal et al, 1987. 

Country: Jamaica; 
Finding: Study examined the effects of omitting breakfast on the 
cognitive functions of three groups of children: stunted, nonstunted 
controls, and previously severely malnourished. Results indicated that 
cognitive functions are more vulnerable to missing breakfast in poorly 
nourished children; 
Source[A]: Simeon & Grantham-McGregor, 1989. 

Country: United States and other countries; 
Finding: Though not definitive, existing research suggests that 
omitting breakfast affects performance of specific cognitive tasks, 
particularly those involving memory. Effects appear more pronounced 
after a period of fasting and in more vulnerable subgroups of 
children, such as those nutritionally at risk. Long-term assessments 
of breakfast omission and cognitive function have not been conducted. 
Studies on the U.S. school breakfast program could not definitively 
conclude that participation in this program caused improvements in 
either long- or short-term cognition and school performance. The 
inconclusive findings reflected limitations in the studies themselves; 
Source[A]: Briefel et al, 1999; Ponza et al, 1999. 

[A] See references at the end of this appendix for complete citations 
for the tables in this appendix. 

Source: GAO review of the literature. 

[End of table] 

Table 6 provides results and views on how targeting factors can affect 
school feeding program effectiveness. Ways to target programs include 
focusing on areas/communities that are (1) low-income and food 
insecure, (2) have relatively low levels of school enrollment and 
attendance, and (3) where girls' enrollment and attendance are 
considerably lower than boys'. 

Table 6: Targeting Factors and School Feeding Program Effectiveness: 

Targeting factors: Target programs on areas/communities with 
relatively low school enrollment and attendance rates: 

* A food ration provided to only poor households who sent their 
children to food-for-education primary schools, in Bangladesh. 
Enrollment rates increased by 20 percent; attendance rates increased 
from 63 to 78 percent; dropout rates fell from 19 percent to 11 
percent. Results were statistically significant when compared to non-
food-for-education schools; 
Source[A]: Ahmed & Billah, 1994. 

* Best practices for SFPs include targeting countries or regions with 
lowest enrollment/attendance statistics; 
Source[A]: Nazaire, 2000. 

* The target for SFPs is where the proportion of children enrolled is 
low or the percent that leave school early is high; 
Source[A]: Levinger, 2001. 

* Targeting the most underserved, food insecure areas, with relatively 
low rates of school attendance and where reasons for lack of 
attendance relate to lack of income and not lack of a facility, seems to
make the most sense; 
Source[A]: Stakeholders, 2000. 

* Experience indicates SFPs have generally been successful in 
providing an incentive for families to send children to school, but it 
is unclear whether SFPs alone have helped to increase enrollment
overall. Studies show that unless SFPs are targeted properly, 
enrollment will increase at schools that provide SFPs, yet decrease in 
surrounding, non-SFP schools; 
Source[A]: Janke, 1996. 

Targeting factors: Target programs on low-income areas: 
* Targeting SFPs to poorest and most insecure families has proved 
problematic. Past research has indicated that children attending 
primary school are more likely to come from less vulnerable
backgrounds, suggesting that SFPs may even discriminate against the 
neediest. To be successful, targeting systems must use truly needs-
based criteria; 
Source[A]: Pillai, 2000. 

* Best practices for SFPs include targeting low-income, food-deficit 
countries and regions where the principal reason children do not 
enroll in or attend school regularly is economic; or where primary
school children arrive at school hungry; 
Source[A]: Nazaire, 2000. 

* Targeting is essential if SFPs are to reach families and communities 
that lack the resources to adequately provide for their school-age 
children or that need to be motivated to enroll their children in
school and have them attend more regularly. Pressure to maintain 
almost universal coverage in Gambia has resulted in a less effective 
WFP SFP; 
Source[A]: Del Rosso, 1999. 

* It may be possible to alleviate hunger in schoolchildren without an 
SFP. Encouraging and educating parents to feed their children before 
sending them to school or to provide a bag lunch or money for them to 
purchase food at school may be an appropriate objective; 
Source[A]: Del Rosso, 1999. 

Targeting factors: Target programs on areas where girls' enrollment 
and attendance are considerably lower than boys': 

* Best practices for SFPs include giving priority to countries/regions 
where girls and minority groups	have traditionally been marginalized 
from access to primary education; 
Source[A]: Nazaire, 2000. 

* School-based food distribution has been used successfully to improve 
enrollment and attendance among school-aged children, particularly 
girls; 
Source[A]: Del Rosso, 1999. 

* In India, an SFP attracted more girls to school and improved the 
attendance of those already in school; 
Source[A]: Devadas, 1983. 

* A study of four long-term WFP projects in West Africa found that 
school feeding alone could not be viewed as a motive for sending girls 
to school if, because of sociological prejudices, their parents were 
not convinced of the usefulness of giving them an education. 
Sensitization programs on the importance of education for girls were 
recommended to be used in association with the programs; 
Source[A]: WFP, 1995 (c). 

[A] See references at the end of this appendix for complete citations 
for the tables in this appendix. 

Source: GAO review of literature. 

[End of table] 

Table 7 provides results and views on how learning environment factors 
can affect school feeding program effectiveness, including ensuring 
adequate numbers of teachers, teacher training, supplies of textbooks 
and other learning materials, and school infrastructure. 

Table 7: Learning Environment Factors and School Feeding Program 
Effectiveness: 

Learning environment factors: Ensure adequate numbers of teachers, 
including for responding to expected increase in student enrollment 
and attendance; 

* Best practices for SFPs include targeting primary schools where 
quality teaching is taking place and the learning environment is 
positive or interventions are being implemented to ensure this becomes 
the case; 
Source[A]: Nazaire, 2000. 

* In the face of deteriorating education infrastructures and the 
shortage of qualified teachers and materials, food aid agencies are 
increasingly compelled to examine, not only their effectiveness at 
getting children to school, but also their effectiveness in helping 
children maximize their time there; 
Source[A]: Janke, 1996. 

* Evidence strongly suggests that SFPs can increase attendance rates, 
especially for girls, but this is likely a short-term solution; if 
there is no change in the quality of schooling, attendance will likely 
drop once the food incentive is removed; 
Source[A]: Stakeholders, 2000. 

* In Bangladesh, the quality of education was lower in schools with a 
feeding program than in	nonfeeding schools largely because enrollment 
was greater in the former. Student achievement test scores were 
slightly lower in schools that received the food aid; 
Source[A]: IFPRI, 2001. 

Learning environment factors: Provide adequate teacher training; 

* Poorly trained teachers provide a strong disincentive to students 
and their families; 
Source[A]: Janke, 1996. 

* Improvements in child learning only come when a facilitative 
learning environment is present to complement the food a child 
receives. A facilitative learning environment is one in which teachers 
are trained to engage children as active learning partners in 
stimulating learning tasks; 
Source[A]: Levinger, 2000. 

* The drive for Education for All highlights the need for quality, in 
terms of relevance and academic performance, and for providing a 
school environment which encourages children to learn how to improve 
their lives; 
Source[A]: UNESCO, 2001. 

Learning environment factors: Ensure adequate supplies of textbooks 
and other learning materials; 

* Inappropriate curricular materials are a strong disincentive to 
students and their families; 
Source[A]: Janke, 1996. 

* A facilitative learning environment requires minimal supplies, 
including blackboards, desks, and chairs. Without these components, 
programs may increase enrollment in what in reality will be poorly 
administered day care centers; 
Source[A]: Levinger, 2000. 

* Educational achievement is believed to be strongly determined by 
factors such as the availability and quality of textbooks and other 
learning materials; 
Source[A]: UNESCO, 2001. 

* Access to books and other learning materials is the most cost-
effective means of raising the level of educational achievement. 
Textbooks are a rare commodity in most developing countries; 
Source[A]: UNESCO, 2001. 

Learning environment factors: Ensure adequate classroom space, desks 
and chairs, lighting, and heating/cooling; 

* Dilapidated school buildings provide a strong disincentive to 
students and their families. Among best practices is school 
infrastructure improvement initiatives; 
Source[A]: Janke, 1996. 

* A facilitative learning environment requires a suitable physical 
environment, including blackboards, desks, and chairs. Many schools 
have no blackboards, chairs, or desks and frequently no classrooms; 
Source[A]: Levinger, 2000; Stakeholders, 2000. 

* In Yemen, especially in areas where girls' enrollment is low, 
communities are constructing new classrooms, contributing financially, 
and providing labor for the building; 
Source[A]: UNICEF, 2001. 

[A] See references at the end of this appendix for complete citations 
for the tables in this appendix. 

Source: GAO review of literature. 

[End of table] 

Table 8 provides results and views on how health and nutrition factors 
can affect school feeding program effectiveness, including through 
treating intestinal parasitic infections, ensuring clean water and 
adequate sanitation facilities, addressing micronutrient deficiencies, 
and ensuring health and nutrition education. 

Table 8: Health and Nutrition Factors and School Feeding Program 
Effectiveness: 

Health and nutrition factors: Treat intestinal parasitic infections; 

* In 1989, an examination of the global distribution of parasitic worm 
infections revealed that large parasitic burdens were associated with 
impaired cognitive function as well as absenteeism, underenrollment, 
and attrition; 
Source[A]: Bundy & Guyatt, 1989. 

* In the West Indies, a single chemotherapy treatment for whipworm 
infection, given to children at school without nutritional supplements 
or improvements in education, improved the children's learning capacity
to the point that their test scores matched those of children who were 
uninfected; 
Source[A]: Bundy & Guyatt, 1989. 

* In Jamaica, a double-blind placebo trial was conducted to determine 
the effect of moderate to high loads of whipworm infection on the 
cognitive functions of 159 school children. Results suggest that
whipworm infection has an adverse effect, which is reversible by 
therapy; 
Source[A]: Nokes et al, 1994. 

* Mass treatment of parasitic infections given to children in their 
schools is considered a powerful tool for improving health; 
Source[A]: UNDP, 1992. 

* To maximize benefits, SFPs should be integrated, when relevant, with 
intestinal worms control programs. Conclusive evidence exists on the 
nutritional and educational benefits of relatively inexpensive 
deworming interventions; 
Source[A]: WFP, 1995 (a). 

Health and nutrition factors: Ensure clean water and adequate 
sanitation facilities are present; 

* Education about water/sanitation/hygiene in schools can encourage 
the construction of facilities and their subsequent use in school and 
in the community; 
Source[A]: Hubley, 1998. 

* Without clean water and adequate sanitation facilities, schools may 
be a major disease vector, and hygiene education is meaningless. By 
providing clean water and sanitation, schools can act as an example to 
both students and the wider community; 
Source[A]: Levinger, 2000. 

* Inadequate sanitation and water in schools jeopardize not only 
students' health but also their	attendance. Girls in particular are 
likely to be kept out of school if there are no sanitation facilities; 
Source[A]: Khan, 1997. 

* Safe water supply should be available on the school premises at all 
times; 
Source[A]: WFP, 1999. 

Health and nutrition factors: Address micronutrient deficiencies; 

* In Ghana, iron supplements for 6 weeks led to a significant 
improvement in school performance, compared with a placebo group; 
Source[A]: Berg, 1999. 

* In China, iodine supplementation brought the average hearing 
capacity of iodine-deficient schoolchildren close to that of non-
iodine-deficient children; 
Source[A]: Berkley & Jamison, 1991. 

* SFP integrated a 6-month dose of an anti-parasite, vitamin A, and 
daily iron supplements. Results included substantial reduction in 
parasitic infection, decline in vitamin A deficiency, and improvement in
children's growth; 
Source[A]: Gopaldas & Gujaral, 1996. 

* Nutritional interventions such as micronutrient supplementation and 
treatment of intestinal worms helped increase students attention, 
cognitive problem solving, and test scores; 
Source[A]: Whitman et al, 2000. 

* School aptitudes among three- to six-year-old children are affected 
by iron-deficiency anemia. These effects continue into the school 
period if the nutritional deficit is not corrected. There are no data 
to support the contention that cognitive deficits observed among pre-
schoolers will persist after appropriate treatment. Chronic iron-
deficiency anemia during the pre-school period will have cumulative 
adverse effects on learning variables that interfere with school 
performance; 
Source[A]: Pollitt, 1990. 

* In India, study of the impacts of iron supplementation on 163 anemic 
girls; significant improvements in cognitive function scores after 8 
months; 
Source[A]: Seshadri & Gopaldas, 1989. 

* In Malawi, when the diets of primary school children were 
supplemented with iron as well as iodine, the gain in IQ scores was 
greater than with iodine supplementation alone; 
Source[A]: Shrestha, 1994. 

Health and nutrition factors: Ensure adequate health and nutrition 
education; 

* Clinical trials show a critical link between learning and school 
children's health and nutrition. Education that addresses specific 
nutrition and health practices is a critical element of SFPs and helps 
to complement and sustain the benefits of deworming and micronutrient 
supplementation, which will in turn increase the benefits of SFPs; 
Source[A]: Del Rosso, 1999. 

* Schools can, with community participation, provide the necessary 
learning experiences to encourage children to practice good hygiene in 
school, in their community, and later in life; 
Source[A]: Hubley, 1998. 

* Experience in a number of countries has shown that unless 
collaborating education institutions include nutrition and hygiene 
information in the curriculum and provide teachers with adequate 
training in these areas, the additional hygiene and nutrition 
education focus of SFPs has little impact; 
Source[A]: Janke, 1996. 

* Nutritional deficiencies (e.g., vitamin A and iodine deficiency) and 
health problems, such as parasitic infections and malaria, affect 
school participation and learning. Most of these issues can be addressed
effectively through health, hygiene, and nutrition policies and 
programs for students and staff; 
Source[A]: Whitman et al, 2000. 

* To enhance the impact of SFPs on children's learning, the programs 
should be part of a larger school health and nutrition intervention. 
Possibilities include, among others, offering health/nutrition 
education; 
Source[A]: WFP, 1995 (a). 

Health and nutrition factors: Ensure nutrient content of meals 
addresses nutritional needs of the student population; 

* The nutritional quality and quantity of a ration should always be 
assessed as well as the effects of the timing of the delivery. Other 
factors such as local food habits, logistical considerations, food 
availability, and food cost will also affect the ration selection; 
Source[A]: Del Rosso, 1999. 

* Best practices include meals of a sufficient size and composition to 
override potential losses from the meal substitution effect; and 
identifying the particular nutritional needs of the targeted 
population and providing meals that directly correspond to local need; 
Source[A]: Janke, 1996. 

[A] See references at the end of this appendix for complete citations 
for the tables in this appendix. 

Source: GAO review of literature. 

[End of table] 

Table 9 provides results and views on how community and parental 
involvement can impact the effectiveness of school feeding programs. 

Table 9: Community and Parental Factors and School Feeding Program 
Effectiveness: 

Community and parental factors: Involve the local community; 

* SFPs can improve educational quality and efficiency by increasing 
community involvement in schools. Schools with community support are 
more effective; 
Source[A]: Del Rosso, 1999. 

* Best practices include community participation in education through 
parent teacher associations (PTA), school infrastructure projects, and 
integrated income generation projects; 
Source[A]: Janke, 1996. 

* Getting the community involved from the beginning and giving it 
ownership of SFPs greatly increase the chances for program success and 
sustainability. Parents see the need for feeding their children and 
want to help. Communities can assist in planning the program as well 
as preparing and distributing meals; 
Source[A]: Stakeholders, 2000. 

* Promoting a positive interaction between the school and community is 
fundamental to the success and sustainability of any school 
improvement process. Community partnerships engender a sense of
collaboration, commitment, and communal ownership. Such partnerships 
also build public awareness and demand; 
Source[A]: UNICEF, 2000. 

* International and national education initiatives have focused on 
integrated school health and education interventions, including 
mobilization of parents and communities; 
Source[A]: WFP, 1995 (a). 

Community and parental factors: Involve parents; 

* Africa has a wealth of community associations, including PTAs, that 
can be a channel for community participation and responsibility. Early 
involvement of such organizations in program development maximizes the 
community's commitment and sustainability; 
Source[A]: Del Rosso & Marek, 1996. 

* Parental support and cooperation allow education about health to be 
shared and reinforced at home; 
Source[A]: UNICEF, 2000. 

* SFPs probably do make a difference in enrollment and attendance if 
their design takes into account	the environment in which they operate, 
including the importance of parent education and involvement; 
Source[A]: Levinger, 1986. 

[A] See references at the end of this appendix for complete citations 
for the tables in this appendix. 

Source: GAO review of literature. 

[End of table] 

Table 10 provides results and views on the effect of government 
commitment and sustainability on the effectiveness of school feeding 
programs. Among the factors addressed are national government 
commitment to broad, national school reform programs, resource 
commitments by national governments and local communities, and plans 
for program sustainability. 

Table 10: Government Commitment and Sustainability and School Feeding 
Program Effectiveness: 

Government commitment and sustainability factors: National government 
commitment to broad, national school reform programs; 

* School feeding must take place within the context of broad, national 
school reform programs. These reforms should focus on other essential 
inputs to education and learning, such as teacher development, 
curriculum reform, and student assessment; 
Source[A]: Stakeholders, 2000. 

* For WFP assistance, governments must demonstrate—through 
promulgation of policies, programs, and financial commitments within 
their means—that high priority is accorded to human resource 
development through basic education, as reflected in the World 
Declaration on Education for All; 
Source[A]: WFP, 1999. 

Government commitment and sustainability factors: Resources committed 
by national government and local communities; 

* Aside from the costs of the food, SFPs have high costs associated 
with transportation, warehouses	and distribution, and storage 
facilities. These costs are often borne by recipient governments. Human
resources may be as much of a constraint as cash and physical 
facilities. Parent-teacher or other community associations can play a 
significant role and ultimately assume some of the costs; 
Source[A]: Del Rosso, 1999. 

* For WFP assistance, programs must be supported by national 
governments and local communities, with a significant amount of 
resources and infrastructure provided as counterpart contributions; 
Source[A]: WFP, 2000. 

* SFPs are expensive. On-site feeding is costly because it requires 
daily preparation and delivery of food; but it is also a model that 
can invite or require community participation; 
Source[A]: Del Rosso, 1999. 

* Most parents, even in the poorest communities, are willing to 
provide whatever resources they can spare to support programs for 
their children, especially when those programs meet a need they
recognize and value; 
Source[A]: Young, 1995. 

Government commitment and sustainability factors: Plan for achieving a 
self-sustaining program; 

* It is unlikely that host governments would continue funding for most 
programs at the same level, if at all, were aid to be withdrawn.; 
Source[A]: Pillai, 2000. 

* The cost of school feeding is a major issue for both governments and 
donors. Feeding programs of any kind are expensive. Financing may 
include international assistance; but in all cases, available public 
resources, or the potential to draw on them, are required; 
Source[A]: Del Rosso, 1999. 

* Even if project objectives are successfully achieved, their long-
term sustainability will still be in doubt because of the high 
proportion of recurrent costs; 
Source[A]: Pillai, 2000. 

* Little evidence exists to support the notion that participation of 
the community, beneficiaries, and government in programs will 
successfully transfer the responsibilities for funding and operating 
SFPs; 
Source[A]: Select Committee on Hunger, 1987. 

* The choice of commodities for SFPs should be determined primarily by 
(1) the acceptability of the food to beneficiaries and (2) their cost, 
with a view toward ensuring takeover by governments and/or communities 
after the phasing out of assistance; 
Source[A]: WFP, 1999. 

* From the beginning, host governments will be expected to contribute, 
within their means, and to put forward a realistic plan to gradually 
increase their contribution and eventually assume full responsibility 
(an "exit strategy"); 
Source[A]: WFP, 2000. 

* Maintenance and continuity of program activities, after outside 
assistance is phased out, are more likely when the participants have 
played an active role in designing and implementing the program. If 
they feel that the program is important and they share responsibility 
for making it work, they will ensure that the activities are continued; 
Young, 1995. 

[A] See references at the end of this appendix for complete citations 
for the tables in this appendix. 

Source: GAO review of literature. 

[End of table] 

References: 

Agarwal, D.K.; Upadhyay, S.K.; Tripathi, A.M.; and Agarwal, K.N. 
Nutritional Status, Physical Work Capacity and Mental Function in 
School Children. Nutrition Foundation of India, Scientific Report 6 
(1987). As cited in Del Rosso, 1999. 

Ahmed, A.U. and Billah, K. Food for Education Program in Bangladesh: 
An Early Assessment. International Food Policy Research Institute, 
Bangladesh Food Policy Project. Dhaka, Pakistan: 1994. 

Berg A. "School Daze," New & Noteworthy in Nutrition 34 (1999). 

Berkley, S. & Jamison D. A Conference on the Health of School Age 
Children. Sponsored by the United Nations Development Programme and 
the Rockefeller Foundation, held in Bellagio, Italy, August 12-16, 
1991. As cited in Whitman et al, 2000. 

Briefel, R; Murphy, J.; Kung, S.; & Devaney, B. Universal-Free School 
Breakfast Program Evaluation Design Project: Review of Literature on 
Breakfast and Reporting. Mathematica Policy Research, Inc. Princeton, 
New Jersey (December 22, 1999). 

Bundy, D.A.P., & Guyatt, H.L. Global Distribution of Parasitic Worm 
Infections. Paris: UNESCO (1989). As cited in Whitman et al, 2000. 

Chambers, C.M. "An Evaluation of the World Food Program (WFP)/Jamaica 
2727 School Feeding Program." Cajunas 24(2)(1991) pp. 91-102. As cited 
in Del Rosso, 1999. 

Del Rosso, J.M. & Marek, T. Class Action: Improving School Performance 
in the Developing World through Better Health and Nutrition. 
Washington, D.C.: The World Bank (1996). 

Del Rosso, J.M. School Feeding Programs: Improving Effectiveness and 
Increasing the Benefit to Education: A Guide for Program Managers. The 
World Bank (August 1999). 

Devadas, R.P. The Honorable Chief Minister's Nutritious Meal Program 
for Children of Tamil Nadu. Ciombatore, India: 1983. As cited in Del 
Rosso, 1996. 

Gopaldas, T., Gujral, S. "The Pre-Post Impact Evaluation of the 
Improved Mid-Day Meal Program, Gujarat (1994-continuing)." Tara 
Consultancy Services, Baroda, India (1996). As cited in Del Rosso, 
1999. 

Hubley, J. "School Health Promotion in Developing Countries: A 
literature review." Leeds, U.K.: Self-published (1998). As cited in 
Whitman et al, 2000. 

IFPRI. Feeding Minds While Fighting Poverty. Washington, D.C.: IFPRI 
(2001). 

Janke, C. "SFPs and Education: Establishing the Context." Catholic 
Relief Service (CRS) School Feeding/Education Companion Guidebook. 
1996. 

Jarousse, J.P., & Mingat, A. "Assistance a la formulation d'une 
politique nationale et d'un programme d'investiseement dans le secteur 
de l'education au Benin," Project UNESCO/PNUD Benin/89/001. Paris: 
UNESCO (1991). As cited in Whitman et al, 2000. 

Khan, A. "The sanitation gap: Development's deadly menace," The 
Progress of Nations 1997. New York: UNICEF (1997). 

King, J. Evaluation of School Feeding in the Dominican Republic. Santo 
Domingo, Dominican Republic: CARE (1990). As cited in Whitman et al, 
2000. 

Levinger, B. School Feeding Programs in Developing Countries: An 
Analysis of Actual and Potential Impact. MD Evaluation Special Study 
No. 30. USAID (January 1986). 

Levinger, B. Statement of Beryl Levinger before the Committee on 
Agriculture, Nutrition, and Forestry. U.S. Senate, July 27, 2000. 

Levinger, B. GAO interview with Beryl Levinger, March 9, 2001. 

Lopez I.; de Andraca, I.; Perales, C.G.; Heresi, M.; Castillo, M.; and 
Colombo, M. "Breakfast Omission and Cognitive Performance of Normal, 
Wasted and Stunted Schoolchildren." European Journal of Clinical 
Nutrition 47 (1993). 

Meme, M.M.; Kogi-Makau, W.; Muroki, N.M.; and Mwadime, R.K. "Energy 
and Protein Intake and Nutritional Status of Primary School Children 5 
to 10 Years of Age in Schools with and without Feeding Programs in 
Nyambene District, Kenya, "Food & Nutrition Bulletin Vol. 19, Number 
4, 1998. 

Moore, E. & Kunze, L. Evaluation of Burkina Faso School Feeding 
Program. Catholic Relief Services, consultant report (February 1994). 

Nazaire, J. CRS Select Targeting and Design Guidelines for School 
Feeding and Other Food Assisted Education Programs. Catholic Relief 
Services (2000). 

Nokes, C.; Grantham-McGregor, S.M.; Sawyer, A.W.; Cooper, E.S.; 
Robinson, B.A.; & Bundy D.A. "Moderate to High Infections of Trichuris 
Trichura and Cognitive Function in Jamaican School Children" 
Parasitology Vol. 104, June 1992. 

Pillai, N. "Food Aid for Development? A Review of the Evidence." In 
Food Aid and Human Security, Clay, E., Stokke, 0., eds. London, 
England: Frank Cass Publishers (2000). 

Pollitt E. "Does Breakfast Make a Difference in School?" Journal of 
the American Dietetic Association, Vol. 95, October 1995. 

Pollitt, E. "Malnutrition and Infection in the Classroom: Summary and 
Conclusions," Food and Nutrition Bulletin Vol. 12, No. 3, 1990. 

Ponza, M.; Briefel, R; Corson, W.; Devaney, B.; Glazerman, S.; 
Gleason, P.; Heaviside, S.; Kung, S.; Meckstroth, A.; Murphy, J.; & 
Ohls, J. Universal-Free School Breakfast Program Evaluation Design 
Project: Final Evaluation Design. Mathematica Policy Research, Inc. 
Princeton, New Jersey (December 20, 1999). 

Rajan, S.I, Jayakumar, A. "Impact of Noon Meal Program on Primary 
Education: An Exploratory Study in Tamil Nadu." Economic and Political 
Weekly (1992). As cited in Del Rosso, 1999. 

Select Committee on Hunger, United States House of Representatives, 
Alleviating World Hunger: Literacy and School Feeding Programs. U.S. 
Government Printing Office (1987). As cited in Del Rosso, 1999. 

Seshandri, S. & Gopaldas, T. "Impact of Iron Supplementation on 
Cognitive Functions in Pre-School and School-aged Children: The Indian 
Experience." The American Journal of Clinical Nutrition, Vol. 50 
(1989). 

Shresta, R.M. "Effects of Iodine and Iron Supplementation on Physical, 
Psychomotor, and Mental Development in Primary School Children in 
Malawi." Ph.D. thesis, University of Malawi, Wappeningen (1994). As 
cited in Whitman et al, 2000. 

Simeon, D.T., & Grantham-McGregor, S. "Effects of Missing Breakfast on 
the Cognitive Functions of School Children of Differing Nutritional 
Status." American Journal of Clinical Nutrition 49. (1989). 

Stakeholders. "School Feeding/Food for Education Stakeholders' 
Meeting." Summary proceedings of a meeting at USAID of 50 
practitioners and experts from USAID, USDA, the World Bank, UNICEF, 
the World Food Program, and other organizations that either administer 
or implement school feeding programs. October 3, 2000 (unpublished). 

UNDP. Partnership for Child Development: An International Program to 
Improve the Health and Education of Children through School-Based 
Services. Project document, interregional project. New York (1992). As 
cited in Whitman et al, 2000. 

UNESCO. Basic Learning Materials Initiative. www.unesco.org 
(downloaded Nov. 2001). 

UNICEF. Focusing Resources on Effective School Health: A FRESH Start 
to Enhancing the Quality and Equity of Education (2000). 

UNICEF. "Basic Education Fund Raising Kit." www.unicetorg (downloaded 
March 12, 2001). 

Whitman, C.V., Aldinger, C., Levinger, B., Birdthistle, I. Thematic 
Study on School Health and Nutrition. Education Development Center, 
Inc. (March 6, 2000). 

World Bank. GAO interviews with World Bank officials, May 15 and 
August 9, 2001. 

World Food Program (a). Implementation of Operational Guidelines for
WFP Assistance to Education (1995). 

World Food Program (b). "Project Pakistan 4185: Promotion of Primary 
Education for Girls in Baluchistan and NWFP," (1995). As cited in Del 
Rosso, 1999. 

World Food Program (c). Thematic Evaluation of Long-Term School 
Canteen Projects in West Africa. WFP Office of Evaluation, (1995). 

World Food Program. "Report on Pilot School Feeding Programme," 
Evaluation Report, WFP/MALAWI, (1996) (unpublished). As cited in Del 
Rosso, 1999. 

World Food Program, UNESCO, and World Health Organization. School 
Feeding Handbook. Rome, Italy (1999). 

World Food Program. "School Feeding/Food for Education." World Food 
Program comments in Response to Oct. 3, 2000, Stakeholders' Meeting" 
(2000) (unpublished). 

Young, M.E. "Integrated Early Child Development: Challenges and 
Opportunities." World Bank, 1995. 

[End of table] 

Appendix V: Costs of School Feeding Programs: 

This appendix discusses actual costs of school feeding programs as 
determined by two World Bank studies, as well as World Food Program 
(WFP) cost estimates of its programs and our own estimates of school 
feeding programs based on WFP guidelines and cost factors and other 
data. It also provides information on situations where school feeding 
programs may not be as cost-effective in promoting learning as certain 
other approaches. 

Table 11 provides figures on the actual costs of more than 30 school 
feeding programs in 21 countries that were reported in two World Bank 
studies. Table 11 shows the annual cost of providing 1,000 calories 
per student on a daily basis for a 180-day school year; dollar values 
have been expressed in 2000 dollars. As the table shows, costs vary 
significantly, ranging from a low of $4.29 for one program to a high 
of $180.31 for another. All but four of the programs cost more than 
$23 per pupil, and the average cost for all programs was $58.66 per 
student. Cost differences can be due to a variety of factors, such as 
differing program objectives, type of food served, and costs in 
transporting the food to the country and, once there, to its final 
destination. 

Table 11: Actual Costs of Various School Feeding Programs (Year 2000 
Dollars): 

Country: Bolivia (food only); 
Per-pupil cost for a 180-day program, 1,000 calories per day: 
$47.22	 

Country: Bolivia-WFP; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $74.51 

Country: Bolivia-four programs; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $31.48 

Country: Chile; 
Per-pupil cost for a 180-day program, 1,000 calories per day: 
$126.65	 

Country: Colombia; 
Per-pupil cost for a 180-day program, 1,000 calories per day: 
$51.52	 

Country: Costa Rica; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $75.13 

Country: Dominican Republic; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $51.52 

Country: Ecuador-government; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $45.08 

Country: Ecuador-WFP; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $27.91 

Country: Ecuador-collection; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $35.93 

Country: El Salvador; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $77.28 

Country: Gambia; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $47.91 

Country: Guatemala; 	
Per-pupil cost for a 180-day program, 1,000 calories per day: $25.76 

Country: Guatemala; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $11.32 

Country: Honduras; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $55.81 

Country: Honduras-two programs; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $14.34 

Country: Jamaica; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $169.58 

Country: Madras-mid-day meal; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $77.91 

Country: Morocco-WFP; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $56.99 

Country: Nepal (MCH and SFP); 
Per-pupil cost for a 180-day program, 1,000 calories per day: $33.23 

Country: Panama; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $4.29 

Country: Paraguay; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $180.31 

Country: Paraguay; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $122.50 

Country: Peru-WFP; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $6.44 

Country: Tamil Nadu-mid-day meal; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $39.41 

Country: Tunisia-WFP; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $61.57 

Country: Uruguay; 
Per-pupil cost for a 180-day program, 1,000 calories per day: $88.01 

Country: Venezuela-1(lunch); 
Per-pupil cost for a 180-day program, 1,000 calories per day: $25.76 

Country: Venezuela-2 (snack); 
Per-pupil cost for a 180-day program, 1,000 calories per day: $23.61 

Country: Venezuela-3 (milk); 
Per-pupil cost for a 180-day program, 1,000 calories per day: $70.84 

Average cost of all programs: 
Per-pupil cost for a 180-day program, 1,000 calories per day: $58.66 

Source: World Bank. "Feeding Latin America's Children: An Analytical 
Survey of Food Programs," Report No. 9526—LAC. Human Resources 
Division, 1991; and Horton, S. (1992) "Unit Costs, Cost-Effectiveness, 
and Financing of Nutrition Interventions," PHN Working Paper 952. 
World Bank, Human Development Department, Washington, D.C. 

[End of table] 

In April 2001, WFP officials told us they estimated the current 
average cost of WFP school feeding programs ranged between about $22 
to $27 per student, for a 180-day school year. They said WFP did not 
have precise figures available on the average costs of its school 
feeding programs because it has not required data to be reported in 
the specific category of school feeding. Many large projects have a 
school feeding component, they noted, but are not entirely devoted to 
school feeding. Subsequently, in July 2001, WFP issued a paper that 
reported the average cost of its school feeding development projects 
in 2000 at 19 cents a day (or $34.20 for a 180 day program).[Footnote 
97] 

We prepared a separate estimate of the cost of school feeding programs 
using some WFP guidelines and cost factors and other data. According 
to WFP, the recommended daily school feeding ration for full-time 
primary school students can range between 600 to 2,000 calories, 
depending on whether schools are half day, full day, or boarding. For 
day school, the recommended acceptable range is between 1,200 to 1,500 
calories (i.e., 60 to 75 percent of the daily energy requirements of 
school-age children). The guidelines also indicate that a minimum of 
10 percent of calories should be obtained from consumption of edible 
fats. In addition, the guidelines for day schools recommend that 
school feeding programs provide 28 to 36 grams of protein; 13 to 17 
grams of fat; and no more than 300 grams of cereals, 30 grams of 
pulses, and 15 grams of vegetable oil. 

We analyzed the nutritional value of typical food aid commodities and 
determined that the least costly mix of commodities—consisting of corn 
and vegetable oil--that met the above requirements for primary day 
schools would cost 3.72 cents per child per day (based on USDA 
valuations of the commodities for 2001).[Footnote 98] If this diet 
were supplied for 180 days, the food alone would cost approximately 
$6.69 per child. On the basis of overall WFP costs for its various 
food aid programs in 1998 to 1999, we estimated that administrative, 
storage, and transportation costs would result in an additional cost 
per child (for a 180-day school meal program) of $7.70. The total 
average cost of this diet would be $14.39 per student. When factoring 
in the nutritional requirements of school-age children to include 
other essential elements, such as vitamins, micronutrients, and 
minerals, we found the lowest-cost, most nutritionally-complete recipe 
would cost $29.67 per child ($13.80 for the food and $15.87 for 
administrative and transportation costs.)[Footnote 99] 

Situations Where Other Approaches May Be More Cost Effective: 

According to a number of experts, school feeding programs may be less 
cost effective than other possible approaches, such as establishing 
maternal child health and early childhood development programs and 
providing alternative nutritional or educational 
interventions.[Footnote 100] 

According to USAID and World Bank officials, maternal and early child 
feeding programs cost about the same as school feeding programs but 
have far greater impacts on both child and life-long learning 
capabilities than school feeding programs.[Footnote 101, 102] Health- 
and nutrition-related programs directed at maternal care and early 
child development are positively associated with physical growth, 
basic cognitive abilities, school readiness, and positive classroom 
behavior. Such programs can help prevent malnutrition before it occurs 
or address it in its early stages[Footnote 103] and thus increase the 
likelihood that children will be healthy when they reach school age. 
[Footnote 104] 

Table 12 provides an estimate of the cost effectiveness of nutrition-
related interventions for a typical developing country, in terms of 
the return on each program dollar spent, as reported by the World 
Bank. (Impact is estimated in terms of wages rather than learning per 
se.) As shown in table 12, school feeding has one of the lowest return 
($2.80) of the 11 interventions. Interventions with the highest 
returns on each program dollar spent are iron fortification of flour 
($84.10), vitamin A supplementation for all children under age 5 
($50), nutrition education ($32.30), and iodized salt ($28). 

Table 12: Estimated Returns on Alternative Nutrition Interventions: 

Nutrition intervention: Iron fortification of flour; 
Return to program dollar (In wages, discounted to the present)[A]: 
$84.10. 

Nutrition intervention: Vitamin A supplementation for all children 
under age 5; 
Return to program dollar (In wages, discounted to the present)[A]: 
$50.00. 

Nutrition intervention: Nutrition education; 
Return to program dollar (In wages, discounted to the present)[A]: 
$32.30. 

Nutrition intervention: Iodized salt; 
Return to program dollar (In wages, discounted to the present)[A]: 
$28.00. 

Nutrition intervention: Supplementation of pregnant women with iron 
pills; 
Return to program dollar (In wages, discounted to the present)[A]: 
$24.70. 

Nutrition intervention: Vitamin A fortification of sugar; 
Return to program dollar (In wages, discounted to the present)[A]: 
$16.00. 

Nutrition intervention: Iodine supplementation for women of 
reproductive age; 
Return to program dollar (In wages, discounted to the present)[A]: 
$13.80. 

Nutrition intervention: School feeding; 
Return to program dollar (In wages, discounted to the present)[A]: 
$2.80. 

Nutrition intervention: Nutrition as part of primary health care; 
Return to program dollar (In wages, discounted to the present)[A]: 
$2.60. 

Nutrition intervention: Food supplements; 
Return to program dollar (In wages, discounted to the present)[A]: 
$1.40. 

Nutrition intervention: Food subsidies; 
Return to program dollar (In wages, discounted to the present)[A]: 
$0.90. 

Note: The methodology and assumptions used in making the estimates 
were not described in the article. 

[A] The discounted present value of wages represents the current value 
of future wages. 

Source: Judith S. McGuire, "The Payoff from Improving Nutrition" 
(updated January 1996), as reported in The World Bank Group, 
"Nutrition as a Sound Investment," To Nourish a Nation (The World Bank 
Group Web site, March 30, 2001). 

[End of table] 

In a study of the cost effectiveness of 40 educational interventions 
in Latin America, the authors surveyed a panel of 10 world experts on 
educational research and practical attempts at educational reform in 
the region, as well as 30 Latin American planner/practitioners working 
primarily in education ministries. Of the 40 interventions, 4 were 
variations on school feeding programs.[Footnote 105] None of the 
school feeding options were identified as being among the top 10 
interventions for increasing learning, taking account of the estimated 
likelihood of adequate implementation (see table 13). The school 
feeding options were ranked between 23 and 34 in terms of increasing 
learning and between 34 and 40 when cost effectiveness was also 
considered.[Footnote 106] 

Table 13: Expert Views on Top 10 Educational Interventions for Latin 
America: 

Top interventions for increasing achievement, based on the estimated 
likelihood of adequate implementation: 

1. Provide standard textbooks and train teachers in usage. 

2. Pay teachers in rural schools salary increment of 50 percent. 

3. Provide multiple interventions: learning packages, school-based 
management, training, testing. 

4. Provide learning materials for individualized instruction. 

5. Assign best teachers to first grade. 

6. Extend daily schedule by 1 hour. 

7. Decentralize schools with supervision. 

8. Provide developmentally oriented preschooling (100 percent unit 
cost of primary school). 

9. Provide classrooms with standard textbooks. 

10. Raise teachers' salaries by 20 percent. 

Source: Ernesto Schiefelbein, Laurence Wolff, and Paula Schiefelbein, 
Cost Effectiveness of Education Policies in Latin America (Washington, 
D.C.: Inter-American Development Bank, Dec. 1998). 

[End of table] 

According to Beryl Levinger, an expert on school feeding and food for 
education programs, there are children in developing countries that 
can effectively and efficiently benefit from school feeding programs. 
Short-term hunger is a genuine problem, and school feeding is one way 
to get and keep children enrolled in school, she said. At the same 
time, success in improving school enrollment, attendance, and learning 
is context driven, and many external factors can affect and interfere 
with these outcomes, she said. Therefore, according to Levinger, one 
needs to assess the total picture and identify the most important 
needs and best solutions for addressing them. For example, if the 
quality of education in a particular community is low and resources 
are limited, it is possible that resources could be better spent on 
improving education than addressing short-term hunger. As learning 
tasks become more interesting, she noted, learning goes up. Levinger 
estimated that providing motivational textbooks and other learning 
materials and training teachers in active learning methods would cost 
roughly about $5 per pupil per year. For an additional $2, she said, 
one could also provide some micronutrient supplementation and 
deworming treatments. 

Multiple studies of treatments for intestinal parasite infections, 
through iron supplementation and regular deworming, have shown 
benefits of lower absenteeism and higher scores on tests of cognition 
or school achievement at a cost of about $1 per child per year. 
[Footnote 107] This is considerably less costly than school feeding 
programs that average $34 per child per year. However, we are not 
aware of any studies that assess and compare the relative impacts of 
programs that only treat for parasite infections to programs that 
provide a school meal. 

In April 2000, the World Health Organization, the U.N. Educational, 
Scientific, and Cultural Organization, the U.N. Children's Fund, and 
the World Bank proposed a strategy for Focusing Resources on Effective 
School Health (FRESH) to give a fresh start to improving the quality 
and equity of education and promoting the Education for All goal. They 
noted that poor health and malnutrition are important underlying 
factors for low school enrollment, absenteeism, poor classroom 
performance, and early school dropout. The agencies identified a core 
group of activities that they said captured the best practices from 
their programming experiences, were highly cost-effective, and a 
starting point to which other interventions might be added as 
appropriate. The agencies recommended that the following basic 
components of a school health program be made available together, in 
all schools: (1) health related school policies;[Footnote 108] 
provision of safe water and sanitation;[Footnote 109] (3) skills based 
health, hygiene, and nutrition education;[Footnote 110] and (4) school 
based health and nutrition services. 

Regarding the latter component, the agencies said schools can 
effectively deliver some health and nutritional services provided that 
the services are simple, safe, and familiar and address problems that 
are prevalent and recognized as important within the community. For 
example, they said, micronutrient deficiencies and worm infections may 
be effectively dealt with by infrequent (6-monthly or annual) oral 
treatment. As another example, they said changing the timing of meals, 
or providing a snack to address short-term hunger during school-—an 
important constraint on learning—-can contribute to school 
performance.[Footnote 112] 

In commenting on a draft of portions of this report, WFP officials 
said there has been no more cost-effective approach identified than 
school feeding for the combined objectives of increasing enrollment, 
attendance, and performance in developing countries--especially in 
areas of food insecurity. Further, when the key resource available is 
food, the case for school feeding to accomplish these objectives is 
indisputable, they said. 

[End of section] 

Appendix VI: Process to Solicit, Evaluate, and Approve Proposals for 
the Pilot Program: 

USDA used a considerably different process to solicit, evaluate, and 
approve program proposals from interested cooperating sponsors and 
WFP. Cooperating sponsors, including Private Voluntary Organizations 
(PVO) and the government of the Dominican Republic, underwent an 
expedited two-stage qualification and proposal review process that 
either did not apply to or generally was different from that applied 
to WFP. Proposal formats and criteria applied to them by reviewers 
varied considerably. An interagency Food Assistance Policy Council 
(FAPC) made the final selection of project awards. 

Proposal Process and Information Required: 

On September 6, 2000, USDA published a notice in the Federal Register 
requesting proposals from interested cooperating sponsors to carry out 
activities under GFEI. (See appendix XI for key events under GFEI.) 
USDA said it would use section 416(b) of the Agricultural Act of 1949 
to provide surplus agricultural commodities in support of an 
international school feeding program to improve student enrollment, 
attendance, and performance in poor countries. Proposals would be 
reviewed on an expedited basis. Given time constraints and the 
considerable effort and time involved in preparing and evaluating 
proposals, USDA invited interested sponsors to present an initial 
submission that contained only information intended to demonstrate, 
based on experience, the organizations' administrative capabilities 
for implementing and managing school feeding or monetization of 
commodities for school feeding. USDA identified nine types of 
information that should or could be provided. The deadline for initial 
submissions was September 15, 2000. 

USDA said that sponsors found to be most capable of successfully 
implementing school feeding activities under step one would then be 
invited to provide a supplemental submission addressing their specific 
proposed activities. The deadline for the step-two submission was 
September 29, 2000. USDA said the submissions should provide 
information that supported the goal of establishing a preschool or 
school feeding program to draw children into the school environment 
and improve access to basic education, especially for females. 
Priority consideration would be given to: 

* countries that had a commitment to universal free education but 
needed assistance in the short run; 

* places where preschool or school feeding programs would promote 
significant improvements in nutrition, school enrollment, and 
attendance levels; 

* projects involving existing food for education programs; and; 

* projects where the likelihood of support from other donors was high. 

USDA requested that sponsors provide, to the extent possible, 
information on (1) literacy rates for the target population; (2) 
percentage of children attending schools, with special emphasis on 
school-age girls; (3) public expenditure on primary education; (4) 
whether the country currently operated a school feeding initiative 
(either through USAID, with assistance from the World Bank, or through 
internal resources); (5) program impact on areas such as teacher 
training, community infrastructure (e.g., PTAs and community groups), 
health, and nutrition; and (6) other potential donors. USDA also 
referred interested parties to the Code of Federal Regulations, which 
describes the requirements for the standard 416(b) program. These 
regulations provide additional guidance on factors to address in 
preparing a proposal. 

Twenty-nine PVOs submitted part one of the proposal application within 
the required time frame. On September 22, 2000, USDA announced that 20 
PVOs had qualified for further consideration and invited them to 
submit the second part of the application on the specific projects 
they were proposing. In addition, USDA announced that the government 
of the Dominican Republic had submitted an application, which had been 
approved for further consideration, and that WFP was eligible to 
participate in the pilot program. 

The September 6, 2000 Federal Register notice stated that the pilot 
program was also open to WFP. USDA did not require WFP to provide 
either the initial or supplemental submission. WFP had already 
submitted a set of proposals to USDA in August 2000, following 
consultations with USDA officials.[Footnote 112] These proposals (1) 
were abbreviated; (2) concerned already existing or approved WFP 
school feeding projects that had not been fully funded, as well as 
planned expansions of these or other projects;[Footnote 113] (3) and, 
in general, did not address many points that USDA had asked 
cooperating sponsors to address in the second-stage submission. The 
proposals typically contained a brief half-page description of the 
project, accompanied by a summary budget for the commodities 
requested. Some, but not all, U.S. agency officials charged with 
reviewing the proposals were told they could obtain additional 
information describing the projects on WFP's Web site. However, some 
projects had been approved by WFP's Executive Board in prior years. 
Information posted on the Web site was sometimes incomplete and/or out 
of date.[Footnote 114] 

USDA officials noted that the United States is a member of the WFP 
Executive Board and as such has a vote on which WFP proposed projects 
should be approved. They also noted that a vote by a donor country to 
approve a project does not mean that the country intends to donate to 
that project. In addition, they noted that approved WFP projects 
submitted to the pilot program in August 2000 would have been approved 
by the executive board prior to the U.S. announcement of the pilot 
program and GFEI. 

According to WFP officials, WFP is strongly committed to addressing 
the key factors associated with effective food for education programs 
discussed in this report. The U.S. government is well aware of this 
commitment, and as a result WFP did not deem it necessary to make 
repeated reference to this commitment in the country-specific 
information included in its proposals. WFP officials noted that 
proposals submitted to USDA for projects that had already been 
approved by WFP's Executive Board had gone through a long vetting 
process, adding that approval of a WFP project requires unanimous 
consensus from all executive board members, including the United 
States. The officials also noted that written documentation on its 
projects had been provided to U.S. government representatives during 
previous WFP Executive Board sessions when the projects had been 
reviewed and approved, as well as in sessions to review projects that 
had been operational. As a result, WFP officials said, the U.S. 
government had plenty of documentation for evaluating WFP proposed 
projects apart from documentation available at WFP's Web site. 

However, USAID told us that when the United States concurs in an 
executive board decision to approve a project, the United States 
frequently states its concerns or reservations about the feasibility 
or sustainability of program activities and has done so in the case of 
school feeding programs. Therefore, the fact that a particular project 
had been approved by WFP's Executive Board did not necessarily mean 
the project was a good candidate for the U.S. food for education pilot 
program. In addition, according to a USAID official, though in 
principle U.S. government personnel responsible for evaluating WFP 
proposals could have gone to WFP's Web site to look up additional 
documentation, there was little time to do this because of the push to 
get the pilot program up and running so quickly. He added that he knew 
of no one who used the Web for this purpose. He also said the 
evaluation task force members did not receive hard copies of 
documentation beyond the abbreviated set of proposals provided by WFP 
to USDA. 

Proposal Evaluation Process And Agencies’ Criteria: 

USDA/Foreign Agricultural Service (FAS) staff evaluated the initial 
PVO submissions on the basis of criteria in USDA's September 6, 2000, 
Federal Register notice. USDA/FAS assigned different weights to the 
criteria. PVOs that scored above a certain level were invited to 
submit the second part of the requested proposals. Of 20 PVOs invited 
to make a second submission, 19 responded and 1 declined, citing a 
lack of adequate time to prepare the type of careful proposal the 
organization wanted to submit. 

The 19 PVOs submitted a total of 62 project proposals. The government 
of the Dominican Republic also responded with a proposal.
For the second part of the proposal process, which covered the actual 
programs sponsors proposed to implement in various developing 
countries, USDA/FAS employed a more elaborate review procedure. The 
Food Assistance Policy Council (FAPC)[Footnote 115] was designated to 
make the final project selections. An FAPC working group was 
established to evaluate the PVO, government of the Dominican Republic, 
and WFP proposals and make recommendations on which ones to approve. 
The working group consisted of staff from FAS and its Food and 
Nutrition Service (FNS), the Department of State, USAID, OMB, and the 
White House. 

USDA/FAS provided the other members of the working group with copies 
of all of the second-stage as well as WFP set of proposals. USDA/FNS 
assigned a nutritionist to review all of the proposals from a 
nutrition perspective. The Department of State assigned two staff to 
review the proposals. Four offices within USAID were involved in 
evaluating the proposals: a country backstop officer, the appropriate 
regional bureau, a nutritionist analyst from the Bureau of 
Humanitarian Response, and an education specialist from USAID's Global 
Bureau, Field Support and Research. USAID's Food for Peace Office 
within the Bureau of Humanitarian Response coordinated the process 
within USAID. The Food for Peace Office is responsible for USAID's 
food aid programs, including any programs that have funded school 
feeding or food for education programs. Each member of the working 
group conducted an evaluation of the proposals separately during 
October 2000 and met in early November to discuss their results and 
reach consensus on which proposals to submit to the FAPC for final 
approval. 

USDA/FAS did not score but recommended approval of WFP proposals for 
all 27 countries in which WFP had established, but unmet, food aid 
requirements. However, USDA scored and divided the non-WFP proposals 
into three distinct categories (i.e., strongly recommended, recommend 
approval, or not recommended). In conducting its second-stage 
evaluation of the non-WFP proposals, USDA/FAS employed a considerable 
number of written criteria, nearly all of which were taken from its 
standard approach to evaluating 416(b) programs. The standard criteria 
do not focus on school feeding or food for education programs. Apart 
from the standard criteria, USDA's evaluation included some criteria 
that related to school feeding/food for education. (All of USDA's 
second-stage criteria were weighted.) USDA considered whether: 

* Objectives supporting the goal of establishing preschool or school 
feeding programs to draw children into the school environment and 
improve basic education for females were clearly stated. 

* The proposal targeted a country with existing food for education 
programs in the host country's development plan. 

* The method for choosing beneficiaries (whether for preschool or 
school feeding) activities was clear and justifiable; emphasis on 
females. 

* The cooperating sponsor provided indicators to measure program 
impact, including baselines and expected outcomes. Potential 
indicators might include literacy rates for target populations, 
percentage of school-age children attending school (emphasis on 
females), and public expenditure on primary education. 

* The cooperating sponsor included specific performance targets as 
part of its proposal, such as magnitude of change in number of meals 
served; enrollment levels, specifically female enrollment; attendance 
levels; capacity building in areas necessary to sustain the feeding 
program, such as development of PTAs and other community groups; or 
infrastructure development for delivery of service. 

Agriculture officials told us they did not have time and adequate 
staff to study lessons learned from past school feeding/food for 
education programs given the short lead time they had to get the 
program up and running. Instead, they said, USDA relied considerably 
upon USAID for this aspect of the evaluation, since USAID had 
extensive experience with school feeding programs. 

Most of USAID's written criteria did not focus specifically on food 
for education. Evaluators in the Regional Bureaus were asked to review 
how the proposals fit with the bureau priorities for the country and 
how a proposed project might affect (positively and/or negatively) 
USAID programs in the country. The bureaus were also responsible for 
providing each country proposal to the respective cognizant field 
mission and for incorporating mission responses and concerns into 
their review. Field missions were also responsible for providing input 
regarding the Bellmon analysis.[Footnote 116] Country backstop 
officers were asked to review each country proposal regarding 
commodities, monetization, and logistics and how these issues might 
affect (positively and/or negatively) USAID's Title II food aid 
programs in country. The USAID nutritionist was asked to review the 
nutritional components of the proposal and their adequacy. USAID's 
Global Bureau was asked to review the educational components of the 
proposals and their adequacy, as well as host country policies and 
commitment to basic education. All of the USAID evaluators were 
instructed to indicate briefly whether they approved or disapproved of 
a proposal and, if they approved, to indicate the priority they 
thought the proposed program should have (low, medium, high, very 
high). 

In USAID's weighting scheme, the Global Bureau's assessment of the 
educational component could have accounted for about 25 percent of a 
proposal's total score. However, for several reasons, its analysis did 
not contribute to USAID's evaluation of which proposals were the best. 
The USAID staff person assigned to rate this dimension of the 
proposals told us that although he had expertise in the education 
area, he was not an expert on school feeding programs. In addition, he 
said that nearly all of the proposals did not provide adequate 
information to judge the quality of the educational component. He told 
us it might have been possible to obtain this information if 
discussions could have been held with the sponsors. However, the 
evaluation process did not provide for such interaction. As a result, 
he assigned the same score to all but one of the proposals. Since 
virtually all proposals were scored exactly the same, education was 
not a discriminating factor in the Global Bureau's overall ranking of 
the proposals. 

No formal record was kept of the interagency working group's 
deliberations, but a summary of its consensus recommendations was 
forwarded to the FAPC for action. This summary contained a brief 
description of the proposed food aid to be delivered to each country, 
its cost and rationale, economic assessments, and prior aid. In the 
end, the FAPC approved 34 WFP proposals covering 23 countries. Of the 
34, 26 were for approved WFP projects with unmet food aid needs and 8 
were for expansion projects. FAPC approved 25 PVO projects and the 
only proposal submitted by a government entity (the Dominican 
Republic). FAPC allocated almost equal program value to WFP (about 
$138 million) and the other sponsors (about $150 million), with 
instructions that the remainder be first offered in support of 
additional WFP proposals. However, cost estimates that FAPC used in 
its award determinations were too high and have since been reduced by 
USDA in implementing agreements. The total cost of WFP agreements was 
recently estimated by USDA at about $92.5 million; cooperating 
sponsors' agreements were estimated at about $135 million. 

[End of section] 

Appendix VII: Selected Information Contained in Proposals for Approved 
School Feeding Programs: 

This appendix discusses selected information in school feeding program 
proposals approved by USDA, including proposed nonmeal components of 
the program, proposed funding of nonmeal components, and comments on 
other donor assistance. 

In its request for proposals, USDA indicated that PVOs could monetize 
some of the food to cover certain other elements important to food for 
education programs. Table 14 provides information on the PVOs that 
proposed funding for nonmeal components, including the specific 
components and the overall proposed funding amount for these 
components. As the table shows, for 17 of the 25 approved proposals, 
PVOs proposed to include a variety of nonmeal components. Examples 
include repairs to school buildings, investments in teacher training 
and school supplies, treatments for parasite infections, and health 
and nutrition education. Ten of the 17 proposals included a budget 
amount for some or all of these components. 

Table 14: Budgeted Amounts for Nonmeal Program Components Included in 
PVO Proposals Approved by USDA: 

PVO and	country: Counterpart International (CPI) Senegal; 
Proposed nonmeal components of the project: Small grants program, 
including a scholarship fund for girls of low-income families who 
would otherwise not attend school, improvement of school and school 
feeding infrastructure, teacher enrichment training, and provision of 
agricultural inputs for other income-generating activities; 
Proposed budget amount: $270,000. 

PVO and	country: Counterpart International (CPI) Senegal; 
Proposed nonmeal components of the project: Nutritional and 
sustainable agricultural technical assistance and community training. 
Teach improved sustainable dry land agricultural techniques in rural 
areas. Provide deworming medicine to children; 
Proposed budget amount: $143,500. 

PVO and	country: Counterpart International (CPI) Senegal; 
Proposed nonmeal components of the project: Provide information 
campaign on the importance of education and nutrition; 
Proposed budget amount: $85,000. 

PVO and	country: Save the Children (STC) Uganda; 
Proposed nonmeal components of the project: None; 
Proposed budget amount: N/A. 

PVO and	country: Mercy Corps International (MCI) Eritrea; 
Proposed nonmeal components of the project: Parent school committee 
capacity building. Provide training to at least 100 parent 
school	committees, including training on how to write proposals and 
budgets for rehabilitation of their schools; 
Proposed budget amount: $180,000 

PVO and	country: Mercy Corps International (MCI) Eritrea; 
Proposed nonmeal components of the project: Facility repair. Provide 
grants to 90 parent school committees that will be used to repair 
primary	schools and distribution warehouse facilities; 
Proposed budget amount: $270,000. 

PVO and	country: Land O'Lakes (LOL) Bangladesh; 
Proposed nonmeal components of the project: Education program. Deliver 
nutritional education curriculum (approved by the government) to 
participating schools to be integrated into classroom discussions; 
Proposed budget amount: Not specified. 

PVO and	country: LOL Vietnam; 
Proposed nonmeal components of the project: Education program. A 
nutritional education curriculum (approved by the government) to 
participating schools to be integrated into classroom discussions; 
Proposed budget amount: Not specified. 

PVO and	country: International Partnership for Human Development 
(IPHD) Moldova; 
Proposed nonmeal components of the project: None; 
Proposed budget amount: N/A. 

PVO and	country: ACDI/VOCA Uganda; 
Proposed nonmeal components of the project: Mobilize and provide 
organizational and leadership training to PTAs. Assist in developing 
sound organizational structures, school and community engagement 
strategies, and action plans; 
Proposed budget amount: $17,500. 

PVO and	country: ACDI/VOCA Uganda; 
Proposed nonmeal components of the project: Provide seeds, tools, and 
technical assistance to establish and maintain school gardens. 
Potential uses include on-site breakfasts for students, sale of the 
produce to purchase school supplies and materials, or supplemental 
take-home commodities for most needy families in the community; 
Proposed budget amount: $64,000. 

PVO and	country: ACDI/VOCA Uganda; 
Proposed nonmeal components of the project: Feeder road linkage to 
schools. Link current PL480-supported community feeder road 
rehabilitation program in Gulu with the program implemented by CRS. 
Feeding effort is intended to increase attendance, and the feeder road 
activity will complement this by increasing access to the selected 
schools; 
Proposed budget amount: Not specified. 

PVO and	country: ACDI/VOCA Uganda; 
Proposed nonmeal components of the project: Organize and deliver 
teacher/administrator training. Develop training modules for teachers 
and administrators, together with national and district education 
officials and other relevant organizations. Pedagogy experts from the 
Ministry of Education and Sports will conduct training modules through 
this program; 
Proposed budget amount: $136,200. 

PVO and	country: IPHD Republic of Congo; 
Proposed nonmeal components of the project: None; 
Proposed budget amount: N/A. 

PVO and	country: International Orthodox Christian Charities (IOCC) 
Georgia; 
Proposed nonmeal components of the project: None; 
Proposed budget amount: N/A. 

PVO and	country: IOCC Lebanon; 
Proposed nonmeal components of the project: None; 
Proposed budget amount: N/A. 

PVO and	country: Catholic Relief Services (CRS) Albania; 
Proposed nonmeal components of the project: Tie school feeding program 
to a longer-term CRS initiative, a Parent School Partnership (PSP) 
program, which aims to increase community involvement in the education 
process. The program includes training for parents and teachers on 
writing project proposals, fund raising techniques, and long-term 
strategic planning; grants given to Parent Councils (maximum of $1,500 
per grant) for small projects in their kindergartens. Mini-projects 
might include new school materials, seminars and publications on 
nutrition for young children, new playground equipment, of child-sized 
furniture in classrooms; 
Proposed budget amount: Not specified. 

PVO and	country: CRS Bosnia/Herzegovina; 
Proposed nonmeal components of the project: Within the framework of 
PSP program, promote community involvement, diversity, aid to the 
poor, sustainable structures, and quality education, by working to 
empower local citizens, partnering with local communities to 
reconstruct and repair school buildings damaged by the war, working 
with mixed communities where stakeholders are actively engaged with 
each other to focus on the education of their children, providing 
education in conflict transformation and peace-building strategies, 
and promoting gender inclusiveness; 
Proposed budget amount: $256,000. 
	
PVO and	country: CRS Benin; 
Proposed nonmeal components of the project: Within PSP focus on 
training and strengthening PTAs and teachers to improve the 
educational system of the targeted area. The program will include the 
completion of up to 30 small grant projects. The types of needs 
targeted are basic school infrastructure, and school materials such as 
textbooks and chalk boards; 
Proposed budget amount: $433,307. 

PVO and	country: CRS Guatemala; 
Proposed nonmeal components of the project: Infrastructure 
improvements. Examine, and when feasible, make repairs to basic 
school	infrastructure. Community education committees will provide 
labor for these repairs and GFFEI Guatemala will provide basic 
materials. Seek to provide a girl-friendly environment in schools by 
considering factors such as latrine privacy and security; 
Proposed budget amount: $40,000. 

PVO and	country: CRS Guatemala; 
Proposed nonmeal components of the project: Community education 
committee. Form a committee that gives parents and community leaders a 
role in program activities such as infrastructure improvements to the 
schools and food distribution; 
Proposed budget amount: Not specified. 

PVO and	country: CRS Guatemala; 
Proposed nonmeal components of the project: Teacher training. Provide 
quarterly training sessions to enhance teachers' awareness of 
different learning styles and pedagogical options, creating an 
environment that will respond to the needs of the children; 
Proposed budget amount: $22,368. 

PVO and	country: CRS Guatemala; 
Proposed nonmeal components of the project: Educational materials. 
Reference materials, such as dictionaries and encyclopedia sets, will 
be provided for schools participating in the GFFEI program; 
Proposed budget amount: $170,000. 

PVO and	country: CRS Guatemala; 
Proposed nonmeal components of the project: Health and nutrition in 
the classroom. Train teachers in health and nutrition. Integrate 
improved health and nutrition practices into daily classroom 
activities; 
Proposed budget amount: $7,000. 

PVO and	country: CRS Honduras; 
Proposed nonmeal components of the project: Deworming/vitamin A. 
Provide deworming treatment and vitamin A supplementation to all 
children; 
Proposed budget amount: Not specified. 

PVO and	country: CRS Honduras; 
Proposed nonmeal components of the project: Promotion of girls' 
education. Develop tools for understanding the economic and cultural 
factors that hinder girls' access to education in the region. On the 
basis of findings, introduce consciousness-raising about the cultural 
and attitudinal factors regarding girls' participation in formal 
education into teacher training; 
Proposed budget amount: Not specified. 

PVO and	country: CRS Honduras; 
Proposed nonmeal components of the project: Establish an improved 
educational services committee in each school (consisting of school, 
personnel, parents, and municipal authorities to enhance a health, 
hygiene, and nutrition program, organize broad-based support for 
school improvement activities, and develop strategies for sustaining 
the benefits of the project). In addition, promote establishment of 
community libraries and identification of teacher training 
opportunities; 
Proposed budget amount: Not specified. 

PVO and	country: CRS Honduras; 
Proposed nonmeal components of the project: Enhancement of learning 
environment. Ensure that participating schools have access to clean 
water and appropriate sanitation facilities. Program resources will be 
made available for the construction of latrines for students, with 
separate facilities available for boys and girls. Particular emphasis 
will be placed on creating an environment attentive to the sanitary 
needs of girls. The project will also facilitate repair of school 
infrastructure. In addition, it will provide a computer and basic 
complementary hardware/software for each school during its first year 
in the program as well as school supplies and key educational 
materials; 
Proposed budget amount: Not specified. 

PVO and	country: CRS Honduras; 
Proposed nonmeal components of the project: Training of community and 
school personnel. Provide health, hygiene, and nutrition training to 
committees enabling them to carry out research and supervision tasks. 
Instruct classroom teachers on information delivery for children; 
Proposed budget amount: Not specified. 

PVO and	country: Project Concern International (PCI) Bolivia; 
Proposed nonmeal components of the project: Establishment of 
greenhouses, school gardens, etc. Provide greenhouses, gardens, and 
demonstration plots, with the aim of ensuring program sustainability. 
PCI has experience in these activities with the USAID Title II 
program. Due to limited financing, only a small number of these 
projects will be implemented. For example, in the area of Uncia, 
Potosi, in which 359 schools will participate in the program, 25 
greenhouses, 80 school gardens, and 25 demonstration plots, will be 
built in coordination with municipalities and school districts. Only 
schools that show interest and have the capacity to manage this 
infrastructure will be selected; 
Proposed budget amount: Not given. 

PVO and	country: Project Concern International (PCI) Bolivia; 
Proposed nonmeal components of the project: Training of teachers. 
Carry out periodic workshops for municipal officials and Ministry of 
Education specialists of each district. Participants will be trained 
in the goals of the program. Additional training will cover improved 
educational methodologies, improved educational environments for 
girls, program supervision. Immediately after these workshops, PCI and 
the Ministry of Education specialists will train teachers in each 
participating school about educational reform, curriculum development, 
educational methodologies, and classroom supervision; 
Proposed budget amount: $100,000. 

PVO and	country: Project Concern International (PCI) Bolivia; 
Proposed nonmeal components of the project: Specific support for 
girls' attendance. Special recognition will be given to schools that 
achieve the goals of increased inscription and attendance of girls in 
a public ceremony given by the Ministry of Education and 
municipalities in coordination with Project Concern International; 
Proposed budget amount: Not specified. 

PVO and	country: Project Concern International (PCI) Bolivia; 
Proposed nonmeal components of the project: Complementary health 
activities. Nurses from the nearest health post will make at least two 
visits per year to participating schools to perform vision and hearing 
exams, provide deparasitization medication, and provide health 
education. Constant monitoring of health and nutrition activities will 
be made by permanent PCI staff as well as education and training in 
nutrition and proper food preparation; 
Proposed budget amount: Not specified. 

PVO and	country: PCI Nicaragua; 
Proposed nonmeal components of the project: Training component. 
Purchase educational materials to improve the learning environment in 
participating primary schools. PCI has 9 years of experience in 
Nicaragua with participative methodologies that are useful and 
attractive to teachers and their students. Training in health, 
environment, and food security will also be given to the 
beneficiaries. During the baseline study, teachers will identify 
special needs related to methodologies and food security. PCI staff 
will then design workshops that will be carried out periodically (five 
during the school year, including a final workshop to evaluate the 
program) with the authorization of the Ministry of Education. Teachers 
will receive training in the organization and operation of the school 
feeding program, and the importance of encouraging enrollment of new 
students and attendance of current students. An average of two 
teachers from each school will be trained. Schoolchildren will be 
trained in hygiene, nutrition, food security and school gardens; 
Proposed budget amount: Not specified. 

PVO and	country: Cooperative Assistance for Relief Everywhere (CARE) 
Albania; 
Proposed nonmeal components of the project: Improved access, quality 
and relevance of basic education. Focus on providing educational 
opportunities for girls and participation of girls. Support improved 
access, relevance and quality of basic education for all children 
through the restoration of basic school infrastructure facilities; 
development of curriculum content that includes life skills 
development (such as preventive health, nutrition, sanitation, 
personal and family care and development) and addresses student well—
being, including psycho-social trauma healing, conflict resolution, 
and gender equity; child-centered learning methods, child-to-child and 
child-to-family learning approaches; and a learning environment with 
high emphasis on gender awareness and equity; 
Proposed budget amount: $340,000. 

PVO and	country: Cooperative Assistance for Relief Everywhere (CARE) 
Albania; 
Proposed nonmeal components of the project: Support formation of PTAs 
in communities and establish community-based accountability for basic 
education through improved communication between community members 
(parents, in particular mothers of school-aged children) and teachers, 
as well as other school and government authorities; 
Proposed budget amount: $45,000. 

PVO and	country: Cooperative Assistance for Relief Everywhere (CARE) 
Albania; 		
Proposed nonmeal components of the project: Provide teacher training 
to 300 primary school teachers to strengthen the quality of child-
centered learning approaches in basic education; 
Proposed budget amount: $67,000. 

PVO and	country: CPI Georgia; 
Proposed nonmeal components of the project: Public information 
campaign. Organize a mass media campaign (television and radio) on 
improving nutrition, cooking methods, and sanitation. Program expected 
to reach 400,000 viewers or approximately one-third of the adult 
female Georgian population (who prepare the majority of meals). 
Training films for local farmers will be produced and shown locally in 
targeted villages and/or on television. Films will cover nutrition, 
health, and improved agricultural techniques, such as increasing plot 
yields and creating school gardens. Local print media campaigns, 
including newspaper articles, advertisements, and brochures, also will 
be developed to reach population in mass media-inaccessible regions; 
Proposed budget amount: $129,992. 

PVO and	country: CPI Georgia; 
Proposed nonmeal components of the project: Income generation. Assist 
development of small-scale local industries (production and sale of 
clothing, shoes, and food) by purchasing these local commodities for 
distribution to families with school-age children; 
Proposed budget amount: $30,000. 

PVO and	country: CPI Georgia; 		
Proposed nonmeal components of the project: Improved school 
infrastructure and supplies: Oversee basic school infrastructure 
repair and maintenance activities including fixing broken windows, 
doors, flooring, roofs, and minor wall damage. Provide necessary 
sources for heat, as well as manage improvements to the school 
cafeteria/kitchen and basic sanitation infrastructure. Provide 
assistance to low-income families in the form of locally manufactured 
items such as clothing and uniforms, shoes, books, a school supplies; 
Proposed budget amount: $1,930,000. 

PVO and	country: Adventist Development & Relief Agency (ADRA) Bolivia; 
Proposed nonmeal components of the project: Training. ADRA will 
provide support and training for schools and school directors on the 
organization of school boards, program management and administration, 
and food preparation. Training will be provided to volunteer community 
education promoters in promoting the enrollment and regular attendance 
of students with priority given to females. Parents will be trained in 
themes that support and encourage the education of their children. 
Parents, municipalities, and education authorities will receive 
training on environmental consciousness, especially focusing on 
mitigation of the widespread use of firewood in the preparation of 
food; 
Proposed budget amount: Not specified. 

PVO and	country: ADRA Madagascar; 
Proposed nonmeal components of the project: Promote increased 
commercial poultry production. Also promote, through the monetization 
of corn, increased commercial production of chickens and eggs in the 
region, and stimulate market demand by introducing these high protein 
foods into the school feeding program on a periodic basis. Project 
will reduce land pressure, improve economic conditions of producers, 
increase the dietary protein intake, and promote program 
sustainability; 
Proposed budget amount: Not specified. 

PVO and	country: ADRA Yemen; 
Proposed nonmeal components of the project: None; 
Proposed budget amount: N/A. 

PVO and	country: ADRA Yemen; 
Proposed nonmeal components of the project: Food for work. Encourage 
active participation of local communities and parent groups in the 
upgrading of school infrastructure and facilities for their children. 
Food for work labor will be used to complete the school repair, 
sanitation and water, and equipment projects for the schools; 
Proposed budget amount: Not specified. 

PVO and	country: ADRA Yemen; 
Proposed nonmeal components of the project: Food security grants. Work 
closely with kindergartens and boarding schools to implement a 
sustainable, self-help program designed to ensure institutional food 
security through income generation and food production activities. 
Award small grants to schools demonstrating their commitment and 
capacity to develop and maintain a self-sustaining school feeding 
program; 
Proposed budget amount: $50,000. 

PVO and	country: ADRA Yemen; 
Proposed nonmeal components of the project: Infrastructure repair. 
Assist schools that are in danger of closure due to poor or 
nonfunctioning heating systems with infrastructure repairs; 
Proposed budget amount: $250,000. 

PVO and	country: ADRA Yemen; 
Proposed nonmeal components of the project: Sanitation/water. Improve 
overall sanitation at schools by installing safe, functioning toilet 
and water facilities; 
Proposed budget amount: $35,000. 

PVO and	country: ADRA Yemen; 
Proposed nonmeal components of the project: School equipment and 
supplies. Replace missing, broken, and/or obsolete classroom equipment 
and supplies in schools suffering from insufficient funding for these 
materials; 
Proposed budget amount: $150,000. 

PVO and	country: World Share (WS) Guatemala; 
Proposed nonmeal components of the project: Continued education 
activities. Engage adolescents (aged 12-18) who are no longer enrolled 
in school in weekly informal educational programs (cultural, sports, 
sex education, and skills training). Develop a curriculum for these 
interventions, and train and compensate participating teachers; Offer 
literacy programs to adolescents and young adults; 
Proposed budget amount: Not specified. 

PVO and	country: World Share (WS) Guatemala; 
Proposed nonmeal components of the project: Food scholarships. Provide 
a "food" scholarship to families and secondary schools to enable 
increased enrollment. Families of students in the target areas who 
demonstrate 75 percent attendance in secondary school will receive a 
monthly food ration. Food will also be contributed to help fund the 
costs of secondary boarding school. 1,000 partial scholarships will be 
available for fiscal year 2002 and fiscal year 2003; 
Proposed budget amount: Not specified. 

PVO and	country: World Share (WS) Guatemala; 
Proposed nonmeal components of the project: Food for work. Provide 
food for work rations for each day worked on construction of 35 school 
classrooms. Provide construction materials (just under $2,000 worth of 
materials in addition to skilled labor) that are not locally 
available. Communities will provide locally available materials and 
unskilled labor; 
Proposed budget amount: Not specified. 

PVO and	country: World Share (WS) Guatemala; 
Proposed nonmeal components of the project: School supplies. Provide 
preschool and primary school students with basic school supplies. 
Preschool students will receive educational games and other materials 
that can be shared; Materials will be targeted toward the most needy 
communities; 
Proposed budget amount: Not specified. 

PVO and	country: Mercy USA (MUSA) Albania; 
Proposed nonmeal components of the project: None; 
Proposed budget amount: N/A. 

Note 1: N/A represents not applicable. 

Note 2: The table provides information based on information included 
in sponsors' proposals to USDA. It does not include information based 
on final agreements negotiated between sponsors and USDA. 

Source: GAO analysis of approved PVO proposals. 

[End of table] 

According to information from USDA, it provided little funding for 
nonmeal components of WFP projects. WFP requested funding for the 
underfunded school meals of already existing projects or for meals for 
expansion of existing projects or start-up of new projects. These 
requests included funding for the commodities and related costs, 
including ocean freight and overland transportation costs to the 
recipient countries; internal transportation, storage and handling 
costs for the commodities within the recipient countries; direct 
support costs; and administrative costs. 

According to WFP, its projects often include funding for nonmeal 
components, which can be obtained through donor countries, partnership 
arrangements with other international donors, or by recipient country 
governments. WFP officials told us they are working to develop more 
partnerships with other donor agencies to address nonmeal aspects of 
their food for education projects. Table 15 provides information on 
planned funding of nonmeal components for the pilot program approved 
WFP projects, based on WFP documentation that was available at WFP's 
Web site. Nonfood components typically involve training, construction 
or rehabilitation of school facilities, or health related activities 
(such as deworming). 

Table 15: WFP Projects' Funding for Nonmeal Program Components: 

Country: Bhutan; 
Proposed nonmeal components of the project: Boarding facilities. The 
project also provides support for the maintenance and construction of 
boarding facilities, which are crucial to bringing education within 
reach of children from remote areas; 
Amount budgeted: $1,100,000 (Gov't of Bhutan). 

Country: Bhutan; 
Proposed nonmeal components of the project: Headmaster/teacher 
training. The project will provide gender training for primary school 
personnel, including headmasters of assisted primary schools and 
primary teachers who are teaching in the nonformal education program; 
Amount budgeted: Not specified. 

Country: Bhutan; 
Proposed nonmeal components of the project: Increasing the 
capacity/standards of hostels. Basic equipment for girls' hostels at 
junior high and high schools will be supplied. Girls often cannot 
enroll, simply because there are not enough beds or because existing 
equipment is inadequate for the needs of girls; 
Amount budgeted: Not specified. 

Country: Bolivia; 
Proposed nonmeal components of the project: Training. Project 
activities will be implemented in centers for up to 30 children with 2 
educators and 1 cook. Food aid will be given to the educators, cooks, 
and children. A new phase of the project envisages more appropriate 
training and materials for the rural facilitators/educators, including 
functional literacy and food preparation and storage. NGOs will be the 
major implementing partners, and several U.N. agencies will provide 
material support and participate
with training activities; 
Amount budgeted: Not specified. 

Country: Cambodia[A]; 
Proposed nonmeal components of the project: None; 
Amount budgeted: N/A. 

Country: Cameroon; 
Proposed nonmeal components of the project: Community grain storage. 
In collaboration with CARE, WFP will provide assistance of an initial 
stock of 10 tons of cereals to grain storehouses and train 1,600 
persons to manage both storehouses and village grain stores. The 
objective is to reduce food insecurity in the north and far northern 
provinces, reduce grain losses, and stabilize prices; 
Amount budgeted: Not specified. 

Country: Chad; 
Proposed nonmeal components of the project: None; 
Amount budgeted: N/A. 

Country: Colombia; 
Proposed nonmeal components of the project: None; 
Amount budgeted: N/A. 

Country: Cote d'Ivoire; 
Proposed nonmeal components of the project: None; 
Amount budgeted: N/A. 

Country: Dominican Republic; 
Proposed nonmeal components of the project: Government programs for 
deworming, safe water and sanitary facilities, and health/nutrition 
education will be integrated in the school feeding project; 
Amount budgeted: Not specified. 

Country: Dominican Republic; 
Proposed nonmeal components of the project: Provision of gas stoves to 
schools initiated within the current project has resulted in an 
estimated 80 percent reduction of community wood consumption with 
benefits for the environment and natural resources. Many households 
have now acquired gas stoves for their
own use. To the extent possible, the project will continue to provide 
gas stoves to schools to reduce the use of charcoal and firewood; 
Amount budgeted: Not specified. 

Country: El Salvador; 
Proposed nonmeal components of the project: None; 
Amount budgeted: N/A. 

Country: Ethiopia; 
Proposed nonmeal components of the project: None; 
Amount budgeted: N/A. 

Country: Gambia; 
Proposed nonmeal components of the project: In coordination with the 
government, PTAs, the World Bank, UNICEF, and NGOs, school 
infrastructure will be rehabilitated and/or expanded; the school 
environment for girls will be improved; more female teachers will be 
trained; the curriculum will be revised; and community-
based organizations will be supported; 
Amount budgeted: Not specified. 

Country: Gambia; 
Proposed nonmeal components of the project: PTA training. Several NGOs 
will be tasked with campaigns and training sessions for PTAs to ensure 
that the community, and particularly women, are actively involved in 
the management of school canteens and are generally aware of 
educational problems and their solutions; 
Amount budgeted: Not specified. 

Country: Ghana; 
Proposed nonmeal components of the project: None; 
Amount budgeted: N/A. 

Country: Guinea; 
Proposed nonmeal components of the project: None; 
Amount budgeted: N/A. 

Country: Honduras; 
Proposed nonmeal components of the project: Through the school 
feeding, WFP is also supporting the national deworming program; 
Amount budgeted: Not specified. 

Country: Kenya; 
Proposed nonmeal components of the project: None[B]; 
Amount budgeted: N/A. 

Country: Mozambique; 
Proposed nonmeal components of the project: Training. WFP will finance 
provincial training seminars for appropriate school personnel, staff 
members, and coordinators (30 people per session of 1 week with 3 
trainers) on topics such as school management, food purchase, food 
storage and handling, nutrition and monitoring, access and 
infrastructure, gender, etc. Local capacity to prepare and manage 
schools budgets will be strengthened through appropriate training; 
Amount budgeted: $160,000[C]. 

Country: Nepal; 
Proposed nonmeal components of the project: Treatment for parasitic 
infections. This deworming component will be carried out with the 
World Health Organization, which will also be responsible for training 
health staff. This component will cost an estimated $200,000 and be 
funded by the Canadian International Development Agency; 
Amount budgeted: Not specified. 

Country: Nepal; 
Proposed nonmeal components of the project: Basic Primary Education 
Project. Objectives include school construction and repair, curriculum 
reform, teacher training, provision of textbooks, strengthening of 
girl's education, etc. Project will be implemented in close 
coordination with the Basic Primary Education Project (BPEP), which is 
supported by a consortium of donors including the World Bank, the 
Asian Development Bank, the Danish International Development Agency, 
UNICEF, the European Union, and others; 
Amount budgeted: Not specified. 

Country: Nicaragua[A]; 
Proposed nonmeal components of the project: None; 
Amount budgeted: N/A. 

Country: Peru; 
Proposed nonmeal components of the project: Student instruction. This 
project will also include nutrition, health, and hygiene instruction 
with manuals and demonstrations at schools; 
Amount budgeted: Not specified. 

Country: Tajikistan; 
Proposed nonmeal components of the project: None; 
Amount budgeted: N/A. 

Country: Tanzania; 
Proposed nonmeal components of the project: Training courses for 
school committees will be organized and cover the importance of 
women's participation; their role in promoting education in the 
community, particularly education of girls; the role of the school 
committee in project implementation; and general aspects of project 
management at school level (food handling, hygiene, recordkeeping, 
etc.); 
Amount budgeted: Not specified. 

Country: Tanzania; 
Proposed nonmeal components of the project: WFP will attempt to supply 
non-food items such as cooking pots, buckets, water storage tanks, 
drinking cups, and cutlery for schools to complement the few items 
provided by local project authorities and communities; 
Amount budgeted: $41,600. 

Country: Tanzania; 
Proposed nonmeal components of the project: The school feeding project 
will have direct links with WFP-assisted food for work project. It 
will support food security and rural infrastructure in drought-prone 
areas through self-help schemes, which in turn will support the 
construction and rehabilitation of school infrastructure such as 
classrooms, hostels (particularly for girls), pit latrines, and 
storage rooms; 
Amount budgeted: Not specified. 

Country: Uganda; 
Proposed nonmeal components of the project: Informal education for 
pastoral families. WFP will collaborate with a voluntary organization 
to support a nonformal program designed to impart and encourage the 
families to acquire functional and basic life skills that are relevant 
to their lifestyles, such as animal husbandry,
water and rangeland management, literacy, and numeric know-how; 
Amount budgeted: Not specified. 

Country: Uganda; 
Proposed nonmeal components of the project: School gardens/farms. WFP 
will support 10 schools annually to nurture school gardens/farms and 
establish a tree nursery. Each benefiting pupil and adult will be 
responsible for planting and maintaining one tree per year. WFP will 
provide funds for seeds, watering cans, water hoses, hoes, shovels, 
hand spades, plastic tubing for seedlings, wire and material to 
protect saplings/seedlings, etc.; 
Amount budgeted: Not specified. 

Country: Uganda; 
Proposed nonmeal components of the project: WFP will provide fuel-
saving cooking pans and stoves to schools with more than 400 students. 
One cup and plate per pupil and cooking pans will be provided for 
those schools that do not have adequate cooking facilities; 
Amount budgeted: Not specified. 

Note 1: Projects include only those approved for USDA's pilot program. 
Nearly all of the funding information in the table was not included in 
the proposals provided to USDA. However, some members of the 
interagency task force that evaluated the pilot program proposals were 
told that additional information on the WFP proposals could be found 
at WFP's Web site. (See appendix VI for additional information on this 
issue.) We obtained the information for this table from WFP's Web 
site. Information for some of the projects may not be complete because 
of outdated and/or insufficient documentation at the site. 

Note 2: N/A represents not applicable. 

[A] Some of the available documentation was for larger assistance 
efforts providing food aid that may or may not include school feeding 
programs. 

[B] No nonmeal component could be determined for Kenya from the 
available paperwork. 

[C] Amount budgeted is for 2 or more years. 

Source: GAO analysis of WFP information. 

[End of table] 

Although USDA said that priority would be given to proposals where the 
likelihood of other donor support was high, neither USDA nor USAID 
included this factor in written criteria for evaluating the proposals. 
We reviewed the PVO proposals to assess whether sponsors in fact 
provided such information in their proposals. As table 16 shows, only 
five of the approved proposals indicated that other donors might 
support the project. Of the five, two proposals said other donors 
would support the project and identified the expected amount of 
support. 

Table 16: Number of Approved PVO Proposals With Information on Other 
Donors: 

Item: Proposal indicated whether other donors may support the program; 
PV0s (total approved proposals = 25): 5. 

Item: Proposal stated that the likelihood of other donors supporting 
the program is high; 
PV0s (total approved proposals = 25): 1. 

Item: Proposal stated that other donors will support the program; 
PV0s (total approved proposals = 25): 2. 

Item: Proposal identified the dollar amount of support expected from 
other donors; 
PV0s (total approved proposals = 25): 2. 

Source: GAO analysis of approved PVO proposals. 

[End of table] 

[End of section] 

Appendix VIII: Donor Views on Uses of Food Aid And How It Is Provided: 

This appendix discusses the views of food aid donating countries other 
than the United States regarding the use of food aid and how it is 
provided. 

Table 17 lists donor countries' views on whether food aid should be 
used for emergencies, development, or both and whether food aid should 
be provided as cash or food-in-kind. 

Table 17: Donor Views on Uses of Food Aid and How It Is Provided: 

Donors[A]: European Commission; 
Views on whether food aid should be used for emergencies or 
development: Prefers to limit food aid to emergencies and 
rehabilitation situations where there is a nutritional crisis so as to 
get a country back on the development track. Prefers not to provide 
food aid for development per se; 
Views on whether food aid should be provided as cash or food-in-kind: 
Does not exclude the use of food-in-kind from donor countries in 
appropriate circumstances, where this is the best source. However, 
food aid should not be used as a systematic mechanism for surplus 
disposal.[B] 

Donors[A]: Japan; 
Views on whether food aid should be used for emergencies or 
development: Believes food aid for development is effective but 
prefers to give aid for emergencies; 
Views on whether food aid should be provided as cash or food-in-kind: 
Says food-in-kind raises questions about possible adverse effects on 
recipients' markets. Supports use of cash to purchase food locally and 
regionally. 

Donors[A]: Canada; 
Views on whether food aid should be used for emergencies or 
development: Makes core contribution to WFP for development projects 
and supports WFP's enabling development policy, but needs to see 
program results. Food aid may not be the most effective way to reach 
the poorest and most vulnerable; 
Views on whether food aid should be provided as cash or food-in-kind: 
Supports food-in-kind contributions when they are delivered 
effectively and in a nontrade-distorting way. Has some concern that 
U.S. use of surplus commodities may not always meet these criteria.
		
Donors[A]: Australia; 
Views on whether food aid should be used for emergencies or 
development: Strongly supports WFP development programs; 
Views on whether food aid should be provided as cash or food-in-kind: 
Has reservations about food-in-kind's capacity to adversely affect 
commercial markets. Is concerned about donor sustainability if program 
depends on surplus disposal of commodities. 

Donors[A]: Germany; 
Views on whether food aid should be used for emergencies or 
development: Believes WFP should primarily provide food for 
emergencies and then contribute to food security for vulnerable groups 
like children, lactating mothers, and the elderly. However, Germany 
does provide a regular donation to WFP development programs and is one 
of the biggest contributors to such programs; 
Views on whether food aid should be provided as cash or food-in-kind: 
Prefers cash contributions so as to encourage local or regional 
purchases of the food aid. Provides only cash, which is not tied to 
the purchase of German commodities. Food-in-kind donations sometimes 
change local taste preferences and attitudes toward foods, which can 
deprive local farmers of their traditional markets. 

Donors[A]: France; 
Views on whether food aid should be used for emergencies or 
development: Believes it is useful for WFP to provide both emergency 
and development assistance. Development is not the first task of WFP but
should be part of its activity; 
Views on whether food aid should be provided as cash or food-in-kind: 
Provides both food-in-kind and cash. 

Donors[A]: United Kingdom; 
Views on whether food aid should be used for emergencies or 
development: Is very skeptical of food aid for development and 
generally does not believe in it; 
Views on whether food aid should be provided as cash or food-in-kind: 
Generally does not approve of providing food commodities for 
assistance. Commodities are cumbersome, inefficient, and not a viable 
development tool. Cash is flexible and efficient and does not have a 
negative impact on the local community. Food-in-kind lacks positives, 
other than helping the farmers in the donor country. 

Donors[A]: The Netherlands; 
Views on whether food aid should be used for emergencies or 
development: Does not see food aid as an important tool for 
development. Does not provide any contribution for WFP development 
projects; 
Views on whether food aid should be provided as cash or food-in-kind: 
Feels that food-in-kind has a negative impact on local agricultural 
production in the recipient countries but has not studied the matter 
sufficiently to prove it. 

Donors[A]: Denmark; 
Views on whether food aid should be used for emergencies or 
development: Feels that food aid can be used for either emergency aid 
or development assistance;	
Views on whether food aid should be provided as cash or food-in-kind: 
Provides both cash and food to WFP. However, if Denmark sponsored an 
SFP, it would probably buy the food locally or regionally since it 
would be cheaper, contribute to local agricultural production (if 
available) and in turn sustainability, and would address cultural food 
preferences. 

Donors[A]: Sweden; 
Views on whether food aid should be used for emergencies or 
development: Believes that all of WFP's assistance should go for 
emergencies but is not against other countries contributing to 
development projects. Sees WFP as a U.N. organization that should 
concentrate on emergencies; 
Views on whether food aid should be provided as cash or food-in-kind: 
Gives only cash to WFP. 

Donors[A]: Finland; 
Views on whether food aid should be used for emergencies or 
development: Provides assistance to WFP for both emergencies and 
development projects; 
Views on whether food aid should be provided as cash or food-in-kind: 
Provides both food and cash to WFP. Has cautioned that food aid should 
not be used to dump excess agricultural product without taking into 
account its effect on recipient countries. 

Note: Most of the information presented in the table is based on our 
interviews in February and April 2001 with representatives of the 
countries. European Commission views are based on a written statement 
provided to us in July 2001. We did not obtain Italian government 
views on the issues in this table. 

[A] Donors are presented in descending order based on the total amount 
of their global food aid deliveries during 1995-99 (see appendix X). 

[B] In a May 4, 2001, demarche to the Department of State, the 
Commission said it had been concerned for several years by the U.S. 
tendency to use food aid to dispose of domestic surplus agricultural 
production to countries not experiencing any food emergency. Among 
other things, it said, such donations destabilize local markets and 
undermine agricultural production in some extremely fragile regions of 
the world where farming is of far greater importance to economic and 
social stability than in the United States. 

Source: GAO. 

[End of table] 

[End of section] 

Appendix IX: Sources USDA Uses to Finance Its Implementing Partners' 
GFEI Project Costs: 

USDA uses three funding sources to pay for implementing partners' 
(PVO/government cooperating sponsors and WFP) operating costs under 
the GFEI pilot program. These costs cover the distribution of surplus 
commodities acquired under Commodity Credit Corporation Charter Act 
(CCC) authority and donated under Section 416(b) authority to friendly 
and developing countries. The funding sources are (1) local currency 
proceeds derived from monetization (sale) of the commodities, (2) 
direct cash payments made by CCC under commodity surplus removal (CCC 
Charter Act 5(d)) authority, and (3) direct cash payments made by CCC 
pursuant to specific limited appropriations authority granted to 
sponsors in July 2001. 

Section 416(b) of the Agricultural Act of 1949, as amended, is the 
authority that CCC uses to pay for most of the cost of removing and 
disposing of donated surplus commodities in connection with the GFEI 
pilot program. This authority allows CCC to directly pay freight 
forwarders selected by implementing partners for the cost of ocean 
transportation and reasonably related expenses of moving the 
commodities to a designated discharge port or point within the 
country's border where the food aid is to be distributed. This cost is 
the largest except for the commodities themselves and is estimated to 
be roughly one-third of the overall pilot program. In the case of 
urgent and extraordinary relief requirements, CCC may also pay the 
partners for internal transportation, storage, and handling (ITSH) 
expenses but not for nonemergency development assistance, which is the 
principal type of aid provided by the pilot. In addition, under 
section 416(b) authority, CCC funds cannot be used to pay partners' 
direct administrative headquarters costs of running the program. 

In lieu of getting CCC funding to recover their ITSH expenses for 
nonemergency programs and administrative costs, partners are permitted 
to monetize (i.e., sell) all or a portion of the commodities in the 
country or region. Local currency proceeds generated from the sale of 
section 416(b) commodities can be used to finance most of the 
sponsors' operating costs—as long as they are specifically approved by 
USDA in program agreements. Monetization is generally how the PVOs and 
government sponsors recover their operating costs. Furthermore, these 
sponsors' budgets and provisions for financial statement and 
monetization reporting as well as limitations on budget adjustments 
without prior USDA approval are incorporated into the program 
agreements. 

USDA's treatment of WFP on these matters differs from that of PVOs and 
a government sponsor. USDA pays cash to WFP for all of these costs, 
including headquarters' administrative expenses. In doing so, it 
relies on section 5(d) of the CCC Act. This section provides authority 
for CCC to expend funds in connection with disposal of surplus 
commodities if such expenditure is required to aid in removing the 
surplus. WFP's general policy, as approved by its executive board, is 
not to monetize commodities. Thus WFP requires cash to cover its 
expenses. In addition, WFP operates under a "full cost recovery" 
policy, which requires that the country making a donation cover its 
full cost. According to USDA's Office of General Counsel, if USDA 
wants to dispose of surplus commodities through WFP, it may pay 
associated costs using section 5(d) authority. 

Specifically, USDA costs incurred in connection with providing 
commodities to WFP under the GFEI program are governed by an agreement 
between CCC and WFP that covers matters related to donation of 
commodities furnished under section 416(b) during calendar years 2001 
and 2002. Under this agreement, CCC agreed to pay WFP not only ocean 
transportation but other authorized expenses incurred by WFP in 
connection with distribution of commodities donated to it. 
Collectively, these other authorized expenses include internal 
transportation, storage and handling,[Footnote 117] direct support 
costs, other direct operational costs, and indirect support costs, up 
to the maximum amount approved by CCC.[Footnote 118] For the GFEI 
program, these costs amounted to about $35 million. 

When USDA requested sponsor proposals for the GFEI pilot program in 
September 2000, it said CCC cash funds might also be available to 
cover expenses related to implementing activities supported with 
commodities acquired under section 5(d) of the CCC Charter Act. USDA 
delivered the same message in a meeting with PVOs to discuss the 
planned pilot program. As a result, most PVOs submitted proposals that 
were based on receiving cash to cover some of their expenses. However, 
in January 2001, USDA informed PVOs with approved proposals that cash 
would not be available to them.[Footnote 119] Although USDA said it 
was prepared to adjust approved sponsors' proposals to permit greater 
monetization of commodities to cover costs, the USDA reversal posed a 
few problems. First, monetized commodities cannot be used to cover the 
sponsors' direct U.S. headquarters' administrative expenses.[Footnote 
120] Second, depending on the situation in a recipient country, 
additional monetization of commodities might risk disrupting 
commercial sales. Representatives of one PVO told us the organization 
had submitted proposals for two countries where it was not possible to 
monetize commodities; therefore, without cash to cover its expenses, 
the PVO could not go forward. Several PVOs were also upset because 
they felt that USDA was providing preferential treatment to WFP. 

USDA noted that its long-standing policy for section 416(b) projects 
was not to provide cash to PVOs unless the country is deemed urgent 
and extraordinary. It further said that PVOs and WFP were treated 
differently because they were fundamentally different in nature and in 
how they acquired their funding. USDA said that whereas PVOs are 
operated privately and have access to other funding sources, WFP is 
governed and funded only by its donor nations and thus not subject to 
or constrained by the limitations of the section 416(b) regulations. 
These reasons notwithstanding, USDA did not explain why it had earlier 
indicated an intention to provide cash to the sponsors. 

USDA's policy reversal led to delays in USDA's negotiating agreements 
for implementing approved proposals for a number of PVO projects. Some 
PVOs were not satisfied with the policy change and made their views 
known to members of Congress. Subsequently, in July 2001, the Congress 
approved legislation (P. L. 107-20) that included a provision 
authorizing USDA to approve use of CCC funds up to about $22.9 million 
for financial assistance to sponsors participating in the pilot 
program. Funds could be used for internal transportation, storage, and 
handling of commodities, as well administrative expenses deemed 
appropriate by the secretary of agriculture. As a result of the 
congressional action, USDA agreed to consider renegotiating agreements 
that it had already concluded with some of the PVOs if they so desired. 

[End of section] 

Appendix X: Top Food Aid Donating Countries: 

This appendix provides details on the top food aid donating countries 
in recent years. Table 18 lists the top 20 food aid donors based on 
shipments for the period 1995 through 1999. Apart from the United 
States, which supplied more than half of all deliveries, the other 19 
donors provided about 43 percent of the food assistance during this 
period. 

Table 18: Top Food Aid Donors Based on Shipments, 1995-1999: 
			
Donor County/Organization: United States; 
Average Annual Metric Tons (1,000): 4,936; 
Percent: 53.9; 
Cumulative percent: 53.9. 

Donor County/Organization: European Union; 
Average Annual Metric Tons (1,000): 1,483; 
Percent: 16.2; 
Cumulative percent: 70.1. 

Donor County/Organization: Japan; 
Average Annual Metric Tons (1,000): 553; 
Percent: 6.0; 
Cumulative percent: 76.2. 

Donor County/Organization: Canada; 
Average Annual Metric Tons (1,000): 416; 
Percent: 4.5; 
Cumulative percent: 80.7. 

Donor County/Organization: Australia; 
Average Annual Metric Tons (1,000): 239; 
Percent: 2.6; 
Cumulative percent: 83.3. 

Donor County/Organization: Germany; 
Average Annual Metric Tons (1,000): 237; 
Percent: 2.6; 
Cumulative percent: 85.9. 

Donor County/Organization: France; 
Average Annual Metric Tons (1,000): 175; 
Percent: 1.9; 
Cumulative percent: 87.8. 

Donor County/Organization: United Kingdom; 
Average Annual Metric Tons (1,000): 129; 
Percent: 1.4; 
Cumulative percent: 89.2. 

Donor County/Organization: China; 
Average Annual Metric Tons (1,000): 128; 
Percent: 1.4; 
Cumulative percent: 90.6. 

Donor County/Organization: Italy; 
Average Annual Metric Tons (1,000): 119; 
Percent: 1.3; 
Cumulative percent: 91.9. 

Donor County/Organization: Netherlands; 
Average Annual Metric Tons (1,000): 113; 
Percent: 1.2; 
Cumulative percent: 93.2. 

Donor County/Organization: Denmark; 
Average Annual Metric Tons (1,000): 91; 
Percent: 1.0; 
Cumulative percent: 94.1. 

Donor County/Organization: Sweden; 
Average Annual Metric Tons (1,000): 82; 
Percent: 0.9; 
Cumulative percent: 95.0. 

Donor County/Organization: Norway; 
Average Annual Metric Tons (1,000): 52; 
Percent: 0.6; 
Cumulative percent: 95.6. 

Donor County/Organization: Switzerland; 
Average Annual Metric Tons (1,000): 48; 
Percent: 0.5; 
Cumulative percent: 96.1. 

Donor County/Organization: Belgium; 
Average Annual Metric Tons (1,000): 41; 
Percent: 0.4; 
Cumulative percent: 96.6. 

Donor County/Organization: Spain; 
Average Annual Metric Tons (1,000): 18; 
Percent: 0.2; 
Cumulative percent: 96.8. 

Donor County/Organization: Greece; 
Average Annual Metric Tons (1,000): 15; 
Percent: 0.2; 
Cumulative percent: 96.9. 

Donor County/Organization: WFP[A]; 
Average Annual Metric Tons (1,000): 14; 
Percent: 0.2; 
Cumulative percent: 97.1. 

Donor County/Organization: Finland; 
Average Annual Metric Tons (1,000): 12; 
Percent: 0.1; 
Cumulative percent: 97.2. 

Donor County/Organization: All other; 
Average Annual Metric Tons (1,000): 254; 
Percent: 2.8; 
Cumulative percent: 100[B]. 

Donor County/Organization: Total; 
Average Annual Metric Tons (1,000): 9,155; 
Percent: 100[B]. 	 

[A] Includes only WFP quantities purchased from its own funds. 

[B] Numbers may not add to 100 due to rounding. 

Source: GAO analysis of the U.N. Food and Agriculture Organization 
data on food aid shipments. 

[End of table] 

[End of section] 

Appendix XI: Key GFEI Events from Announcement of Concept to 
Notification of Project Approvals: 

This appendix outlines key events related to the GFEI pilot from the 
time the program was announced until early January 2001, when USDA 
notified proposal winners. 

As table 19 shows, USDA's expedited schedule allowed interested 
cooperating sponsors at most 8 business days to prepare and submit the 
first part of the proposal. Sponsors who began preparing for the 
second part of the proposal at the earliest possible time (i.e., 
without waiting to learn whether they qualified to do so), had a 
maximum of 18 business days to complete and submit it to USDA. 

Table 19: Key GFEI Events from Announcement of Concept to Project 
Approvals: 

Date: Feb. 27, 2000; 
Event: Ambassador George McGovern proposes that the United States take 
the lead in organizing a worldwide school lunch program and a 
supplementary feeding program for pregnant and nursing women and their 
children under the age of 5 (article in the Washington Post). 

Date: May 26, 2000; 
Event: President Clinton decides to go forward in principle with the 
proposal for a universal school lunch program. 

Date: July 11, 2000; 
Event: Decision memorandum prepared for the president on international 
school feeding proposal.	
			
Date: July 23, 2000; 
Event: President Clinton announces, at G-8 Summit, USDA GFEI pilot 
program to improve student enrollment, attendance, and performance in 
developing countries; challenges other donor countries to support a 
global program. 

Date: August 17, 2000; 
Event: USDA met with interested PV0s and provided an advance draft 
notice of the Federal Register notice. 

Date: Sept. 6, 2000; 
Event: USDA notice in the Federal Register invites proposals from 
cooperating sponsors to carry out GFEI activities. 

Date: Sept.15, 2000; 
Event: Deadline for cooperating sponsors to present an initial 
submission that contained only information intended to demonstrate the 
organizations' administrative capabilities. 

Date: Sept. 22, 2000; 
Event: USDA announces the cooperating sponsors selected to present 
detailed proposals for specific GFEI projects. 

Date: Sept. 29, 2000; 
Event: Deadline for invited sponsors to present second submission to 
USDA. 

Date: Oct. 12, 2000; 
Event: Deadline for USAID internal review and scoring of GFEI second-
stage proposals. 

Date: Oct. 27, 2000; 
Event: Initial Foreign Assistance Policy Committee (FAPC) Working 
Group meeting discusses which proposals to recommend for approval by 
the committee. 

Date: Nov. 2-3, 2000; 
Event: FAPC Working Group meeting reaches consensus on which proposals 
to recommend for approval. 

Date: Nov. 9, 2000; 
Event: FAPC decides which proposals will be approved. 

Date: Dec. 28, 2000; 
Event: USDA announces approved PVO proposals and number of approved 
WFP proposals. 

Date: Jan. 5, 2001; 
Event: USDA sends letters to individual PV0s advising them of results 
of the evaluation process. 

Source: GAO analysis. 

[End of table] 

[End of section] 

Appendix XII: Comments from the U.S. Department of Agriculture: 

Note: GAO comments supplementing those in the report text appear at 
the end of this appendix. 

USDA: 
United States Department of Agriculture: 
Farm and Foreign Agricultural Services: 
Foreign Agricultural Service: 
1400 Independence Ave., SW: 
Stop 10xx: 
Washington, DC 20250-1034: 

February 8, 2002: 

Mr. Loren Yager	
Director, International Affairs and Trade: 
U.S. General Accounting Office: 
441 G Street, NW, Room 4932: 
Washington, D.C. 20548: 

Dear Mr. Yager: 

Thank you for the opportunity to respond to the General Accounting 
Office's (GAO) draft report to Congress on the Global Food for 
Education Initiative (GFEI). The U.S. Department of Agriculture (USDA) 
welcomes the opportunity to review the lessons learned from the pilot 
phase of the GFEI program, However, we believe that the GAO has taken 
an overly critical view of how USDA administered the GFEI given the 
time and resource constraints. Also we believe the report fails to 
recognize that the President directed a school feeding program, not an 
entire educational program. Our comments on a number of issues raised 
in the report are outlined below. We are also developing a list of 
technical corrections, which will be forwarded to you separately. 

Short Time Frame for Program Implementation: 

USDA's Foreign Agricultural Service (FAS) assumed responsibility for 
administering a pilot Global Food for Education Initiative in response 
to the Presidential Directive announced on July 23, 2000. At that 
time, FAS was also administering an exceptionally high level of food 
assistance under the Section 416(b) program. Nevertheless, FAS moved 
quickly to implement the GFEI without requesting additional staff 
resources. Within three months of the President's announcement, FAS 
developed program criteria and issued a Federal Register notice 
detailing procedures for selecting proposals under the GFEI. By 
December 2001, FAS had evaluated the institutional capacity of 33 
private voluntary organizations (PVOs) and subsequently, with an inter-
agency working group, reviewed and ranked about 60 proposals. Of 
those, 49 programs in 38 countries were selected and commodity 
shipments began in the summer of 2001. 

Food for Education Versus School Feeding: 

The GFEI has three purposes — to improve student enrollment, 
attendance, and performance. Given the time and funding limits, USDA 
believed that the only feasible goal for the one-year pilot was a 
school feeding program to increase school attendance and enhance 
students' ability to learn. The report, however, emphasized the 
educational component of the program and has evaluated FAS against an 
unrealistic standard regarding that performance element rather than 
the other two objectives. A much longer time frame (at least three to 
five years) would be required to address all of the educational 
factors mentioned in the report including teacher training, 
infrastructure, learning materials, health and nutrition programs, and 
community involvement. [See comment 1] 

Measurements to Evaluate the Pilot: 

Within the context of a one-year program, USDA included realistic 
measurements to evaluate proposals and program impact. Measurement 
criteria as outlined in Table 1 of the report were not required due to 
the time limitation and the uncertainty of the future of a school 
feeding program. Upon request, USDA received feedback from the PVOs 
concerning the inability to adequately measure learning progress 
within a one-year pilot effort. USDA implemented a school feeding 
program to increase the attendance and enrollment of school children 
in the targeted communities. Being a pilot program, participating 
organizations were given the flexibility to address specific needs and
constraints within their countries of expertise, Each agreement will 
be evaluated independently once agreement objectives have been met. 
These program evaluations will then be incorporated into the broader 
GFEI program review. 

Partnerships and Staffing: 

USDA views private voluntary and international organizations as 
valuable partners in implementing USDA food aid assistance throughout 
the world. USDA is only able to implement large programs by relying on 
the resources of their partner organizations and their in-country 
experience. Program controls are maintained by completing enforceable 
agreements and monitoring and evaluating programs. This private-public 
partnership was essential in initiating the $300 million pilot GFEI 
initiative within a year's time GAO notes that FAS experienced 
staffing issues and received no additional staffing resources during 
the implementation. The staff shortages and the need to maintain 
quality slowed the negotiation of agreements. 

Donor Support: 

While additional donor support is important for a large, global 
effort, local government support is critical for program 
implementation and sustainability. The GFEI program for Honduras 
provides an example of this local support. The Honduran Minister of 
Education will be asking USDA for technical assistance and support in 
the development of a permanent countrywide school feeding policy to be 
fully implemented in four years. The country's request is based on the 
positive progress of the Honduras GFEI program's operation and the 
regional coordinator's facilitating role. USDA hopes to work with the
government of Honduras to support their school feeding goals. As our 
regional coordinators continue to visit with the governments involved 
in the GFEI program, USDA hopes to see more donor support and local 
support including school feeding policies. 

FAS Management Capabilities of Development & Fond Assistance Programs:: 

The GAO report incorrectly states that USDA lacked expertise in 
managing development and humanitarian assistance such as food aid. In 
fact, USDA has substantial expertise and long experience in managing 
international development and assistance programs including
Public Law 480 (since 1954), Food for Progress and Section 416(b) 
(since the mid1980s). The Foreign Agricultural Service's International 
Cooperation and Development area also has considerable expertise in 
international development programs including education, scientific 
research and technical exchange. The Food and Nutrition Service of 
USDA administers the nation's successful school lunch program. (See 
Attachment 2 for further description of expertise provided by USDA 
agencies.) FAS has also reached out to agencies outside USDA for 
assistance in the educational component of implementing the GFEI. 
These include the Agency for International Development and the 
Department of Education. Additionally, FAS drew upon the technical 
expertise of organizations such as the World Food Program and 
international PVOs with a history of operating overseas school feeding 
and educational programs. [See comment 2] 

All of the agencies noted above provided input on developing realistic 
and measurable criteria that could be addressed within the limited 
time frame. Additionally, these agencies, as well as our embassy teams 
overseas, took part in the proposal review process and had the 
opportunity to provide input on the educational, nutritional, and 
developmental factors involved. 

Proposal Phase Versus Agreement Phase: 

The GAO report focused on the inadequacies of the proposals submitted 
to USDA, It is important to note that the proposal is the beginning 
text of a negotiated contractual process. Once this negotiation has 
been completed, the signed agreement fully explains the activities to 
be implemented and meets all regulatory requirements. [See comment 3] 

The Federal Register notice provided guidance on the minimal content 
of proposals. Very little time was provided to the organizations that 
responded to this solicitation. Yet, many PVO proposals that were 
approved exceeded the minimal information required. In fact, the 
information outlined in Table 1 of the report captures the key 
elements that GAO would have recommended as proposal review criteria. 
While FAS did not require this information, Table 2 of the report 
shows that many of the proposals did in fact contain information not 
required by FAS but suggested by GAO. 

The PVOs and USDA added other requirements and ensured that 
performance targets were included in the final agreements. USDA 
believes that GAO's comparisons between proposals and the recommended 
program elements understate the quality of the GFEI programs. 
Comparisons between the final agreements and recommended program 
elements are more accurate. 

Differences between Private Voluntary Organizations and the World Food 
Program: 

There is an inherent difference in operating with PVOs and a United 
Nations organization such as the World Food Program (WFP). The United 
States sits on the WFP Executive Board, which approves all projects, 
including budgetary requirements, but the U.S. is not part of any PVO 
boards, Additionally, it is through this WFP Executive Board that 
operating procedures and practices governing the WFP are established. 
Monitoring and evaluation procedures are integral to this process. 
Based on GAO's draft report, USDA is confident that the information 
submitted by WFP contains the required information listed
in the Federal Register notice or the regulations governing USDA food 
assistance programs. [See comment 4] 

Processes to Prevent Disincentive Effects of Food Aid: 

The GAO report provided suggestions for improving the analysis of 
disincentive effects of food aid such as reviewing the impact of 
donations on alternative food sources. While this can certainly be a 
goal for future projects. FAS stands by the analysis that significant 
market disruptions do not result from the GFEI pilot. On the contrary, 
in places such as Georgia, the pilot program has expanded flour 
consumption by local bakeries and also increased employment in these 
bakeries. [See comment 5] 

Also, the statement that PVOs would be biased in their analysis 
because they would like to provide the food aid is correct, but in the 
opposite direction suggested by GAO. The PVOs have to implement their 
program within a community. Program success depends upon community 
support which is not going to occur if markets are disrupted. PVOs are 
therefore going to more rigorously analyze the food needs of an area 
and the potential market impact of a program to assure the 
acceptability of the commodity and the project's success. [See comment 
6] 

We appreciate this chance to review and evaluate the critical 
components of an effective school feeding program and of a food for 
education initiative. A better understanding of these components 
should be useful to members of Congress as they consider the future of 
this pilot program. 

Sincerely, 

Signed by: 

Mary T. Chambliss: 
Acting Administrator: 

Attachment: 

The following are GAO's comments on the letter from the Department of 
Agriculture dated February 8, 2002: 

1. USDA noted that GFEI has three purposes - to improve student 
enrollment, attendance, and performance, but indicated it is not 
possible to improve learning in a 1-year pilot program. According to 
USDA, GAO evaluated USDA against an unrealistic standard performance —
rather than the objectives of enrollment and attendance. In addition, 
USDA said, a much longer time frame would be required to address all 
of the factors mentioned in the report (examples cited include teacher 
training, infrastructure, learning materials, health and nutrition 
programs, and community involvement). We disagree with USDA's 
statements for two reasons. First, our conclusion is that school 
feeding programs are more likely to improve enrollment and attendance, 
as well as learning, if they are carefully integrated with other key 
factors and interventions. Second, we conclude that the pilot program 
could have been improved by determining in advance which proposals 
were for communities where key factors were already in place or would 
be addressed during the projects themselves. 

2. USDA disagreed with our statement that USDA lacked expertise in 
managing development and humanitarian assistance such as food aid. We 
have revised that statement to specify expertise in food for education 
development programs. At the same time we note that a recent USDA 
study of its food aid monetization programs cited difficulty 
evaluating the programs' impacts because of limited personnel 
resources, high staff turnover, and increasing demands to implement 
large food aid programs. In addition, the limited presence of overseas 
agricultural attaches has adversely affected USDA's ability to oversee 
some of its sponsors' monetization projects, the study said. USDA's 
Inspector General has also expressed concern about this matter. 

3. USDA said it believes that GAO's comparisons between the proposals 
and the recommended program elements understate the quality of the 
GFEI programs, since the proposal is only the beginning text of a 
negotiated contractual process. We focused on the proposal process to 
determine to what extent USDA secured information for judging and 
selecting proposals that offered greater promise of improving school 
enrollment, attendance, and learning. 

4. Regarding differences in the treatment of PVOs and WFP, USDA 
reiterated (as discussed in our draft report) that the United States 
sits on the WFP Executive Board, which approves all projects. However, 
executive board approval does not mean that the United States may not 
have concerns about a particular project. As USAID advised, even when 
the United States concurs with an executive board decision to approve 
a project, the United States frequently states its concerns or 
reservations about the feasibility or sustainability of program 
activities and, according to USAID, has done so in the case of school 
feeding projects. USDA also said it is confident that the information 
submitted by WFP contains the required information listed in the 
Federal Register notice or the regulations governing USDA food 
assistance programs. However, WFP did not have to address requirements 
of the Federal Register notice; the notice did not require as much 
information as we believe would have been useful for evaluating 
proposals; and USDA's 416(b) regulations did not include specific 
information requirements for assessing food for education programs. 

5. USDA indicated agreement with our finding that analysis of the 
disincentive effects of food aid projects should include the impact of 
commodity donations on alternative food commodities. USDA said doing 
so could improve analyses and be a goal for future projects. At the 
same time, USDA said it stands by the pilot project assessments that 
significant market disruptions will not occur—even though such 
analysis was not conducted. 

Our report notes that cooperating sponsors are responsible for 
analyzing the potential disincentive effects of their projects and 
that USDA does not independently verify the results of such analyses. 
In addition, we noted that USDA officials acknowledged that because 
PVOs want to provide the food aid, these organizations may not be 
completely unbiased in preparing analyses of disincentive effects. In 
its letter, USDA said the latter statement is correct but in the 
opposite direction suggested by GAO. According to USDA, PVOs are going 
to more rigorously analyze the food needs of an area, because program 
success depends upon community support, which is not going to occur if 
markets are disrupted. We agree that the latter is one possible 
interpretation of the statement and therefore removed the statement 
from the letter. 

[End of section] 

Appendix XIII: Comments from the U.S. Agency for International 
Development: 

USAID: 
U.S. Agency For International Development: 
1300 Pennsylvania Avenue, NW: 
Washington, DC 20533: 

February 11, 2002: 

Mr. Loren Yager: 
Director: 
International Affairs and Trade: 
U.S. General Accounting Office: 
Washington, DC 20548: 

Dear Mr. Yager: 

I am pleased to provide the U.S. Agency for International 
Development's (USAID's) formal response on the draft GAO report 
entitled "Foreign Assistance: Global Food for Education Initiative 
Faces Challenges for Successful Implementation" (January 2002). 

The report accurately and fairly ,depicts the complex and formidable 
challenges confronting the Global Food for Education Initiative 
(GFEI). USAID fully endorses GAO's recommendation matters for 
congressional consideration (p.44). The findings and recommendations 
of this report should be of great use to the Congress as it debates 
the structure of U.S. food assistance. 

As the report acknowledges, USAID has extensive experience in school 
feeding and food for education activities. USAID shared its 
experience, including results on current USAID food for education 
programs, with USDA as the pilot was designed. USAID's comments, much 
of which dealt with the cost and the need for complementary programs, 
were well received. Nevertheless, priority was given to getting a 
program up and running. The designers believed improvements could then 
be made that would deal with the issues of cost, sustainability and 
the need for complementary programs. 

While endorsing the basic analyses and findings of the report, we have 
some points of clarification that we are enclosing in this letter. 

Thank you for the opportunity to respond to this GAO draft report and 
for the courtesies extended by your staff in the conduct of this 
review. 

Sincerely, 

Signed by: 

John Marshall: 
Assistant Administrator Bureau for Management: 

Enclosure: a/s: 

[End of section] 

Appendix XIV: Comments from the Office of Management and Budget: 

Executive Office Of The President: 
Office Of Management And Budget: 
Washington, D.C. 20503: 

February 8, 2002: 

Mr. Loren Yager: 
Director: 
International Affairs and Trade: 
United States General Accounting Office: 
Washington, D.C. 20548: 

Dear Mr. Yager: 

Thank you for providing the Office of Management and Budget the 
opportunity to comment on the GAO's draft report, Global Food for 
Education Initiative Faces Challenges for Successful Implementation. 

The report is balanced and generally accurate, and will serve the 
Congress and the public in future deliberations about school feeding 
programs. The report criticizes the GFEI pilot program undertaken by 
the Department of Agriculture (USDA) for 1) failing to adequately 
incorporate lessons learned from previous school feeding programs, 2) 
exhibiting weaknesses in structure, planning and management; and 3) 
failing to generate support from other nations. While the report 
presents evidence for these conclusions, it is worth emphasizing that 
these problems may be attributable to the urgency with which the pilot 
program was generated. It is also worth noting that greater emphasis 
was placed on the nutrition goals of the pilot—as distinct from 
education objectives. One could expect that some of these problems 
could be addressed a more deliberate approach to performance and 
evaluation. 

The report's description of international reluctance to support the 
pilot program is troubling. it appears that the short-term funding 
mechanism, using USDA authority, may limit the program's effectiveness 
in several ways. This includes discouraging foreign donations that are 
vital to a broad-based, sustainable program. In addition, there appear 
to be problems with USDA's legal authority to make certain payments to 
the World Food Program (WFP). These problems are mentioned in a USDA 
letter to the WFP on January 16, 2001, and the subject of US 
negotiations with the WFP. 

The Administration cares about feeding and educating vulnerable 
children, and therefore, we intend to undertake a careful evaluation 
of the pilot. It is important to learn from this and other school 
feeding efforts. The GAO report is a valuable step toward making sure 
we get the results we intend from this program and making humanitarian 
aid more effective for those in need. 

Detailed technical comments are being provided separately. 

Sincerely, 

Signed by: 

Marcus Peacock, P.E. 
Associate Director: 
Natural Resource Programs: 

[End of section] 

Appendix XV: GAO Contacts and Staff Acknowledgments: 

GAO Contacts: 

Phillip Thomas (202) 512-9892: 
Wayne Ferris (202) 512-5169: 

Acknowledgments: 

In addition to those named above, Gezahegne Bekele, Janey Cohen, Stacy 
Edwards, Mary Moutsos, and Rolf Nilsson made key contributions to this 
report. 

[End of section] 

Footnotes: 

[1] USDA's Foreign Agricultural Service has overall responsibility for 
managing the pilot program. 

[2] CCC, a funding mechanism for U.S. farm income support and disaster 
assistance programs, has no staff. Its activities, including 
acquisition, storage, and disposition of surplus commodities, are 
carried out primarily by personnel of USDA's Farm Service Agency. 

[3] H.R. 1700, McGovern-Dole International Food for Education and 
Child Nutrition Act of 2001, was introduced in the House of 
Representatives on May 3, 2001. S. 1036, which has the same title, was 
introduced in the U.S. Senate on June 13, 2001. The House and the 
Senate are currently considering whether to incorporate versions of 
the proposed legislation in the new multi-year farm bill. 

[4] The term "food for education" refers to a school feeding program 
that includes the specific objectives of improving school enrollment, 
attendance, and/or performance or learning. The term "school feeding" 
may refer to a program in which the only objective is to provide food 
to children or to a program that also seeks improvements in school 
enrollment, attendance, and/or performance. 

[5] A micronutrient is an organic compound (such as a vitamin or 
mineral) essential in minute amounts to human growth and welfare. 

[6] We interviewed representatives of the European Union, Australia, 
Canada, Denmark, Finland, France, Germany, Japan, the Netherlands, 
Sweden, and the United Kingdom. 

[7] All 13 PVOs participating in the program were invited to comment. 
Comments were received from ACDI/VOCA, Catholic Relief Services (CRS), 
International Partnership for Human Development (IPHD), Land O'Lakes 
(LOL), Mercy Corps International (MCI), and Mery USA (MUSA). Project 
Concern advised us that it had no comments. 

[8] The forum is an interagency body established in 1990 by the U.N. 
Development Program, the U.N. Educational, Scientific, and Cultural 
Organization, the U.N. Population Fund, the U.N. Children's Fund, and 
the World Bank. 

[9] George McGovern, "Too Many Children Are Hungry. Time for Lunch," 
the Washington Post, Feb. 27, 2000. 

[10] An interagency committee, chaired by USDA, evaluated the 
proposals and selected the winners. 

[11] WFP has been doing school feeding projects for nearly 40 years. 
Since 1995, WFP has tried to focus the objectives of its school 
feeding projects on attaining educational objectives, including 
increased enrollment and attendance (especially for girls) and 
improved cognitive functions. (See appendix III for additional 
background information on WFP.) 

[12] Under USDA's 416(b) program, a cooperating sponsor may be either 
(1) a foreign government; (2) an entity registered with USAID in 
accordance with its regulations; or (3) an entity that demonstrates to 
CCC's satisfaction that it has the organizational experience and 
resources to implement and manage the type of program proposed, has 
experience working in the targeted country, and has experienced and 
knowledgeable personnel who will be responsible for implementing and 
managing the program. WFP is not a cooperating sponsor. CCC has a 
separate umbrella agreement with WFP that governs U.S. Section 416(b) 
donations to WFP. 

[13] Some of the WFP projects involve multiple agreements. 

[14] In early January 2001 initial cost estimates for the projects 
totaled about $289 million. According to USDA officials, implementing 
partners tended to overestimate project costs. According to WFP (1) 
commodity prices fell between the time proposals were submitted and 
when the agreements were negotiated and signed; (2) not all requested 
commodities were available and substitutions were made at lower cost; 
and (3) because of long delays between when proposals were submitted 
and projects approved and commodities shipped, many projects had to 
alter and/or reduce the originally requested tonnage. 

[15] The value of the food needs to offset the cost to the family of 
sending their children to school. Where poverty and the need for child 
labor are particularly high, the financial value of the school meals 
needs to be very significant to offset the opportunity costs of 
schooling. See World Food Program, Implementation of Operational 
Guidelines for WFP Assistance to Education (Rome, Italy: 1995). 

[16] According to WFP, the decision to enroll a child at school and, 
thereafter, for the child to attend regularly is influenced by many 
factors, such as the direct and indirect costs of schooling, education 
level of parents, perceived value of education, availability of 
employment opportunities, and availability and quality of school 
facilities. 

[17] WFP, School Feeding Works for Girls' Education (Rome, Italy, 
2001). 

[18] School Feeding/Food for Education Stakeholders' Meeting, Oct. 3, 
2000. 

[19] Several reviewers have noted uncertainties about the impact of 
school feeding on learning; yet they have concluded there is evidence 
that school feeding can positively affect learning by offsetting the 
effects of short-term hunger. For example, WFP, the United Nations 
Educational, Scientific, and Cultural Organization (UNESCO), and the 
World Health Organization (WHO) find that the existing literature on 
school feeding programs' effects on learning is not fully conclusive, 
but they attribute this to weaknesses in the design of studies, 
difficulty in obtaining certain types of information, and other 
factors. See WFP, UNESCO, and WHO, School Feeding Handbook, (Rome, 
Italy: 1999). 

[20] Statement of Beryl Levinger before the Committee on Agriculture, 
Nutrition, and Forestry, U.S. Senate, July 27, 2000. 

[21] One consequence of these factors is that children often attend 
school irregularly because the demand for labor from school-age 
children, their poor health, the difficulties associated with getting 
to school, and the limited benefits accrued from being at school 
conspire to reduce demand for schooling. See World Bank, Effective 
Schooling in Rural Africa, Project Reports (Basic Education Cluster, 
World Bank, 2000). 

[22] School Feeding/Food for Education Stakeholders' Meeting, Oct. 3, 
2000. 

[23] See Joy Miller Del Rosso, Partnership for Child Development, 
School Feeding Programs: Improving Effectiveness and Increasing the 
Benefit to Education: A Guide for Program Managers (Human Development 
Network and the World Bank: August 1999). 

[24] There are several reasons why school feeding alone is unlikely to 
overcome chronic undernourishment or protein-energy malnutrition: (1) 
parents may provide less food at home, with the school meal simply 
replacing a home meal; (2) the school meal may not address the complex 
nutritional deficiencies in the children's diets; and (3) meals may be 
too irregular (at best, the programs are in place only during days 
within the school year). 

[25] Community and parental involvement can include planning school 
feeding programs and/or preparation and distribution of meals. 

[26] The value of the food donated for consumption in a school feeding 
program may be less than half of the cost of the program. For example, 
in 1998-99, commodity donations to WFP and WFP commodity purchases 
accounted for 47 percent of total WFP budgetary expenditures. Nonfood 
costs—such as ocean freight; inland transportation, storage, and 
handling; direct support costs; and indirect support costs—accounted 
for the rest. 

[27] See appendix V. A variety of factors affect costs, including 
differing program objectives, type of food served, and transportation 
costs. 

[28] School Feeding/Food for Education Stakeholders' Meeting, Oct. 3, 
2000. 

[29] The group also noted that once a school feeding program begins, 
it may be politically difficult to discontinue it. A World Bank 
official said she has seen situations where countries have taken over 
school feeding programs and, because of the cost, ignored more 
important priorities. 

[30] Available data indicate it is not easy to achieve long-run, 
sustainable school feeding programs. For example, WFP, which has been 
funding school feeding programs for several decades, estimated that it 
has phased or closed out of programs in only 10 to 15 countries. (In 
1999, WFP had school feeding projects in 55 countries.) Moreover, most 
of these were not clear-cut cases where a country was ready and able 
to take full responsibility for its program. In some cases, officials 
said, WFP has closed down a program only to reopen or reconsider it, 
based on changing circumstances. 

[31] The request was made in a Federal Register notice on September 6, 
2000. 

[32] See appendix VI for additional discussion of this matter. 

[33] The 34 proposals covered 27 projects in 23 countries (see 
appendix II). 

[34] According to WFP officials, WFP projects often include funding 
for nonmeal components that is obtained through donor countries, 
partnership arrangements with other international donors, or by 
recipient country governments. Table 15 in appendix VII provides 
information on such planned funding for the pilot program approved WFP 
projects that was available at WFP's Web site. 

[35] According to USDA, under the section 416(b) program, WFP may also 
sell some of the commodities within the recipient country to provide 
local currency resources for in-country expenses, including, but not 
limited to, administrative, storage, transportation, and handling 
expenses, as well as direct costs of their humanitarian and 
developmental projects. However, WFP's general policy is not to 
monetize food commodities and WFP has said it does not propose to 
monetize food commodities to fund related educational support 
activities. 

[36] According to ACDI/VOCA, USDA was not responsive to funding 
complementary activities necessary for the overall improvement of 
education. Such complementary activities should be funded by GFEI in 
the future, it said. 

[37] Although this section focuses considerably on USDA as manager of 
the pilot, the overall design and structure of the program was the 
result of an interagency process that involved staff from the Office 
of the White House, the National Security Council, the National 
Economic Council, the Office of Management and Budget, USDA, USAID, 
and the Department of State. 

[38] USDA's September 6, 2000 Federal Register notice requesting 
proposals said CCC would consider multiyear proposals subject to an 
annual review of commodity availability and program performance. 
Several PVOs complained about confusion about the time frame of the 
pilot (1 year versus 3 years), based on statements made by USDA 
officials, and said the ambiguity adversely affected their ability to 
develop proposals and negotiate project agreements. 

[39] CCC is a wholly owned government corporation, which has the legal 
authority to borrow up to $30 billion at any one time from the U.S. 
Treasury. CCC must eventually repay the funds it borrows from the 
Treasury. But because CCC spends more than it earns, its losses must 
be replenished periodically through a congressional appropriation so 
that its borrowing authority is not depleted. The Congress generally 
provides this infusion through the regular annual USDA appropriation 
law. However, in recent years CCC has received a "current indefinite 
appropriation," which in effect allows CCC to receive such sums as are 
necessary during the fiscal year for previous years' losses and 
current year's losses. See Congressional Research Service, RL31001: 
Appropriations for FY2002: U.S. Department of Agriculture and Related 
Agencies (Washington, D.C.: August 3, 2001). 

[40] Although CCC is part of USDA, the administration's decision to 
use the Commodity Credit Corporation's surplus disposal program to 
fund the pilot program did not necessitate that USDA manage the food 
for education projects that were funded by the program. For example, 
under an interagency agreement, USDA used USAID to administer its 
overseas section 416(b) activities until 1992. USAID noted that, like 
USDA, it did not have the financial and human resources to undertake 
the pilot program. Thus, if a new interagency agreement had been 
signed, USAID would have required additional resources to design and 
implement a successful pilot. 

[41] A comparison of USAID and USDA programming requirements shows 
that USAID ties program planning to agency strategic objectives in a 
much more integrated way than does USDA. For example, USAID requires 
sponsors to submit more detailed project proposals and reporting on 
project results, conducts more open and interactive project planning 
with sponsors, and has a broader and more formal decisionmaking and 
review process. USAID project proposal requirements that are not 
required by USDA include discussions of key assumptions and risks, 
relationship to existing programs, performance indicators, 
sustainability, and lessons learned. In addition, USAID requires that 
annual results be linked to the agency's strategic objectives. 

[42] USDA officials told us they have not kept track of how much money 
the department has spent on previous school feeding programs. 

[43] Two PVOs said they preferred USDA's management of food aid. IPHD 
said USDA was less bureaucratic, more responsive to field needs, and 
allowed more innovation by PVOs. Land 0' Lakes said GFEI programs can 
involve significant private sector involvement, and USDA's substantial 
authority for procurement and shipping of U.S. food aid provides for 
greater continuity. 

[44] USDA's Foreign Agricultural Service has managed the pilot with 
existing Program Development Division staff resources, which were 
already stretched thin because of a recent section 416(b) program 
expansion, personnel turnover, and slow hiring of replacements. During 
our review, a significant portion (ranging from between 25 percent to 
33 percent) of the division's permanent staff positions were vacant. 

[45] WFP and IPHD noted that many of the recipient countries were well 
into their academic year before USDA commodities were procured, 
shipped, and available for distribution. 

[46] To provide a perspective on the pilot's staffing and resources, 
before the pilot program was announced, USAID estimated its Food for 
Peace Office would require a minimum of 6 months and more likely a 
year to establish a food for education pilot program that would 
operate in only 8 to10 countries. USAID projected a need for 16 
additional headquarters staff, 4 additional regional field officers, 
and 16 to 20 Foreign Service nationals to support the regional 
officers. Additional staff and related costs were estimated at $5.5 
million. 

[47] According to an executive branch official, the actual decision 
was made by November 2000. 

[48] According to Mercy Corps International, many developing countries 
have strict laws about the conversion of local currency; as a result, 
it said, many of the approved proposals could not be implemented until 
after limited CCC funding was permitted. The alternative, according to 
CRS, was to decrease the scope of their programs. 

[49] According to a U.S. government official, as the pilot program 
evolved, the original emphasis on education appeared to shift toward 
feeding in a school setting, as evidenced by views expressed by USDA 
staff. According to CRS, the primary and secondary objectives of GFEI 
have never been clarified. (CRS said this was similar to other school 
feeding and food for education programs supported by other donors.) 

[50] These include attendance and enrollment changes, gender equity, 
best practices, and other donors and their activities. The report will 
cull information from WFP, PVOs, the government of the Dominican 
Republic, monitors, regional coordinators, and FAS staff. 

[51] This objective was cited on the day the president announced the 
program at the G-8 summit; in USDAs Federal Register notice inviting 
project proposals; and in a fact sheet accompanying the announcement 
of approved proposals. 

[52] According to USDA, WFP planned on sampling a total of 3,700 
schools in 23 countries, or roughly 161 sample schools per country. 
Actual country sample sizes were to range from 60 to 388 schools. 

[53] Basic information included (among other items) whether the 
program is in a preschool, primary, or boarding school; whether it 
includes a breakfast, snack, dinner, or take-home ration; number of 
feeding days during the year; and whether the school has a kitchen. 

[54] For enrollment, attendance, certified teachers, and classroom 
numbers, data were to be collected for the current year and the 3 
preceding years if appropriate (most of the GFEI-funded WFP projects 
existed before the pilot, some for many years). 

[55] Whether the school has a PTA; the number of men and women on a 
PTA executive; how many men and women (further differentiated by 
whether school teachers) are involved in managing the feeding program 
and distributing the ration; and whether parents contribute to the 
school financially or in kind. 

[56] Type of water source for the school and used by children; 
sanitation and type of toilet facilities used by the children; and 
whether health and nutrition are part of the school program. 

[57] Teacher training; books; classrooms; curriculum development; 
nutrition; worm eradication; water supply; sanitation; HIV/AIDS 
education; and reproductive health issues. 

[58] Of school teachers and, for the latter indicator, a separate 
focus group of pupils as well. 

[59] Standard and up-to-date data had not been collected across its 
many projects. Our review of several WFP evaluations of country 
programs that included school feeding projects that were completed in 
2000 indicated problems in monitoring of project activities and 
objectives. For example, an evaluation of a project in Peru concluded 
the project did not have a good monitoring and evaluation system, 
lacked baseline surveys, and lacked impact indicators to measure 
improvements. An evaluation of a school feeding project in Yemen found 
that monitoring by the Ministry of Education and the WFP country 
office "is weak, irregular, and altogether severely deficient." WFP 
officials said actions are being taken to correct these weaknesses and 
noted that transparent and self-critical evaluations are an integral 
part of their management system. 

[60] One question seeks data on the number of boys and girls who had 
completed primary school the previous year and had gone on to higher 
education. The second question, to be addressed by separate focus 
groups (teachers and pupils), seeks views on the significance of the 
school feeding in relieving short-term hunger and helping maintain 
attention in school. 

[61] For example, one national monitor questioned a project to provide 
only cow's milk to 96 schools in Vietnam. According to the monitor, 
most of the children had never tasted milk, and a number of health 
professionals were concerned that there would be an initial problem 
with tolerance. In addition, school officials told him that 97 percent 
of the school-age children in the areas were enrolled and attendance 
was not a problem. In another case, involving Honduras, USAID said 
that teachers reported to work during less than half of the school 
year, raising concern that attendance might be adversely affected. 

[62] According to USDA officials, a number of management issues 
delayed finalizing a monitoring and evaluation method, such as the 
time needed to conclude agreements with implementing partners, assess 
options for how USDA could monitor and evaluate the projects 
(including developing baseline data), establish external monitors and 
regional coordinators, and secure financing for the additional 
personnel. 

[63] The sale of U.S. commodities in a recipient country can have food 
security benefits for the recipient country, depending on how the sale 
is handled, due to increased food availability in local markets. 

[64] The UMR can be waived or reduced in unusual situations, such as 
severe drought, floods, balance of payment difficulties, or the 
absence of reliable import data. 

[65] According to CRS, the extremely short time frame from the 
announcement of the GFEI proposal due date might have been a factor in 
the depth of the cooperating sponsors' Bellmon analyses. In addition, 
it said, such analyses can be quite costly if done right, and sponsors 
have to take into consideration the return on investment. It is more 
likely that sponsors would undertake detailed analyses after projects 
are approved or contingently approved, when they could be assured of 
project support. Land 0' Lakes, Inc., said Bellmon analyses would be 
more useful if they were made after or at the time agreements were 
awarded rather than early in the proposal process. 

[66] According to ACDI/VOCA, its Bellmon analysis for Uganda was 
prepared by the USAID Mission in Kampala, and addresses the levels and 
types of commodities that could be used for direct distribution 
without creating a disincentive. 

[67] According to USAID, it requires the USAID Mission Director or, in 
the absence of a USAID mission, the principal officer at post certify 
that there will be no disincentive to local agricultural production or 
marketing prior to a cooperating sponsor in that country receiving 
P.L. 480 Title II food aid commodities. USAID/Washington has provided 
temporary duty assistance to posts to assist them in performing the 
necessary analyses. 

[68] According to USAID, a recently issued report, by the Royal Danish 
Ministry of Foreign Affairs, looked at WFP programs in Bangladesh, 
Bolivia, Nicaragua, Vietnam, and Zambia, and found no examples of 
market disruption associated with the programs. 

[69] If a donated surplus commodity does not match cultural food 
preferences or needs of the recipient community, the commodity may be 
sold in the recipient country or neighboring countries and the 
proceeds used to buy local foods. 

[70] A recent USDA study of its food aid monetization programs cited 
difficulty evaluating the programs' impacts because of limited 
personnel resources, high staff turnover, and increasing demands to 
implement large food aid programs. In addition, the limited presence 
of overseas agricultural attaches has adversely affected USDA's 
ability to oversee some of its sponsors' monetization projects, the 
study said. USDA's Inspector General has also expressed concern about 
this matter. 

[71] All other things being equal, monetization of commodities to pay 
for other program costs is less efficient than direct funding because 
of the added costs involved in transporting, storing, and handling the 
commodities, as well as additional costs to monetize them. However, 
the purchase of additional commodities to pay for other program costs 
has some benefits for U.S. farmers, if the purchase does not undercut 
farmers' commercial sales in the food aid recipient countries, and for 
U.S. commodity processors, domestic transporters, and U.S. shippers. 
According to a U.S. government official, any benefits are likely to be 
negligible given that the volume of commodities involved is small 
relative to U.S. and world agricultural output. According to Land 0' 
Lakes, Inc., in the case of nonfat dry milk (NFDM) donations, any 
inefficiencies would be offset, since the U.S. government purchases 
NFDM though domestic milk support programs and has to pay storage and 
carrying costs for the commodity. In addition, Land 0' Lakes said, 
monetization has supported the introduction and expansion of the use 
of American commodities in international markets. 

[72] USAID's approach for its sponsors who monetize food aid requires 
that PVOs sell the commodity at a value equal to at least 80 percent 
of the commodity, insurance, and freight cost or at 100 percent of the 
free alongside ship value of the commodity. In addition, USAID 
reevaluates market conditions after the food aid has been monetized to 
assess whether this condition has been met. USAID's policy is said to 
concern some sponsors and commodity groups since it may limit the 
countries and type of commodities available for monetization. Because 
it is difficult for a PVO to know the local market price for a 
commodity in a country where the activity is insufficient to 
constitute a market, commodities may be sold at prices that are below 
the world market price. 

[73] Under the pilot, which had been scheduled to start in late 2000, 
the initiative was to be assessed in 2001 and a determination made on 
whether to proceed with a truly global program. 

[74] As discussed earlier, costs can vary significantly, depending on 
the type of food used and other factors. In July 2001, WFP reported 
the average cost of its school feeding development projects in 2000 as 
19 cents per day ($34 for 180 days). (See appendix V.) The $10.2 
billion estimate is higher than that cited by President Clinton in 
December 2000, who said it would cost about $6 billion to $7 billion 
annually to provide a meal to 300 million children every single day 
for a year. 

[75] Official development assistance includes grants or loans to 
developing countries at concessional financial terms (if a loan, 
having a grant element of at least 25 percent) to promote economic 
development and welfare. Technical cooperation is included in aid. 
Grants, loans, and credits for military purposes are excluded. 

[76] According to WFP, the recommended daily school feeding ration for 
full-time primary school students can range between 600 to 2,000 
calories, depending on whether schools are half day, full day, or 
boarding. For day schools, the recommended acceptable range is between 
1,200 to 1,500 calories. Our food tonnage estimate was based on meals 
comprised of corn and vegetable oil and providing 1,200 calories. The 
average weight of a single meal was estimated at 300 grams of corn and 
12 grams of vegetable oil. Thus, for 300 million children in school 
180 days a year, the combined tonnage is 16.8 million metric tons. 

[77] Table 3 does not include China, which ranked ninth in the world, 
based largely on its large cereals deliveries to North Korea between 
1996 and 1999. We included Finland, which ranked twentieth, because it 
was cited by U.S. officials as a country strongly supportive of the 
pilot program. 

[78] Individual member states of the European Union can take a 
different position on GFEI from that of the Commission. 

[79] At a January 29, 2002 symposium, Dan Glickman, former U.S. 
Secretary of Agriculture when the pilot program was initiated, said 
while 70 percent of food aid worldwide had been supplied by the United 
States in recent years, it was the result of domestic political 
pressures to raise commodity prices for the benefit of American 
farmers, not out of any long-term plan to improve diets in poor 
countries. He also said that efforts to make development policy a 
higher priority during the Clinton administration were eclipsed by 
budget constraints imposed by Congress. At the same meeting, John 
Podesta, President Clinton's former chief of staff, said establishing 
a coherent and sustainable development policy was a daunting task 
partly because it would be hard to sustain current levels of public 
interest, also because of interagency turf wars and lack of overall 
leadership and accountability. See: "Making Development Policy in the 
New Era: Priorities, Politics, and Structures of U.S. Policymaking on 
Global Poverty and Hunger," Resources for the Future News Release 
(Washington, D.C.: January 29, 2002). 

[80] Cargill is an international marketer, processor, and distributor 
of agricultural, food, financial, and industrial products and services 
with 90,000 employees in 57 countries. 

[81] WFP is using the grant in part to work with WHO and the World 
Bank to quickly and comprehensively expand deworming activities to a 
much larger number of countries and students. In early August, WFP 
told us the grant was used to fund workshops involving seven 
Anglophone African countries, WFP, WHO, and the World Bank. The 
Canadian grant was used to fund the workshop costs for up to $50,000 
per country for deworming treatments in WFP-assisted schools. In 
addition, WFP said, World Bank education loans are being used to fund 
deworming programs in the countries. WFP also said that World Bank 
education loans with a school health/FRESH component would be the 
source of the bulk of funds necessary to expand the program, and the 
Canadian government also may provide additional funding. (FRESH is a 
partnership developed by the World Bank, WHO, UNICEF, and UNESCO, to 
improve the health and nutritional status of school-age children.) 

[82] The July 2001 G-8 Summit established an Education Task Force to 
provide advice to G-8 leaders in cooperation with developing 
countries, relevant international organizations, and other 
stakeholders on how the G-8 can best support the achievement of the 
education for all goals. WFP was invited to and agreed to participate 
in the task force's February 18, 2002, meeting. 

[83] Unless improved learning is an element within an agreement. 

[84] Donation and expenditure data as reported in WFP Annual Report of 
the Executive Director: 2000, Apr. 18, 2001. 

[85] Statement of Catherine Bertini before the Committee on 
Agriculture, Nutrition and Forestry, U.S. Senate, July 27, 2000. WFP 
has characterized its involvement in GFEI as the "World School Feeding 
Initiative." 

[86] WFP, "The World School Feeding Initiative: WFP's Framework for 
Action," August 2000. 

[87] According to WFP, monetization is an inefficient use of food 
resources. Until school children's hunger is alleviated, monetization 
of foodstuffs is not the answer. In addition, WFP told us that WFP's 
general policy not to monetize commodities is out of concern about 
disrupting commercial trade either internationally or in developing 
countries with relatively fragile agricultural sectors. 

[88] Although WFPs' development and protracted relief and recovery 
projects are reviewed and approved by its executive board, the 
approval for a project does not mean that all requested resources will 
be contributed by donors. According to a WFP official, on average, WFP 
programs are funded at about 70 percent of their requirements. 

[89] Two key examples are the following. World Food Program, UNESCO, & 
World Health Organization, School Feeding Handbook (Rome, Italy: 
1999). WFP, "The World School Feeding Initiative: WFP's Framework for 
Action." 

[90] According to WFP, as a general rule, the essential services 
required for operating a school feeding program—cooks, kitchen 
helpers, guards—should be covered by the community, either by 
providing such services itself or by contributing cash to compensate 
those engaged to perform the services. Beyond that, broader community 
participation will be built into projects wherever feasible. 

[91] In addition, WFP's guidelines require in-depth reviews of 
activities in a sampling of countries and on specifically identified 
issues of interest to WFP, the beneficiary countries, and/or donors. 

[92] According to USDA, WFP planned on sampling a total of 3,700 
schools in 23 countries, or roughly 161 sample schools per country. 
Actual country sample sizes were to range from 60 to 388 schools. 

[93] A country program includes the different projects that WFP 
sponsors within a country For example, a country program might include 
a school feeding program, a maternal/child health and nutrition 
program, and a food-for-work program. 

[94] See, for example, statement of Beryl Levinger before the 
Committee on Agriculture, Nutrition and Forestry, U.S. Senate, July 
27, 2000. Joy Miller Del Rosso, School Feeding Programs: Improving 
Effectiveness and Increasing the Benefit to Education: A Guide for 
Program Managers, (World Bank: August 1999). 

[95] World Food Program, Operational Guidelines for WFP Assistance to 
Education. 

[96] School Feeding/Food for Education Stakeholders' Meeting, Oct. 3, 
2000. 

[97] According to WFP, these costs are all inclusive, ranging from the 
values of all food aid commodities purchased by WFP, costs of 
transportation and monitoring, to internationally and locally 
recruited personnel. In February 2002, WFP officials told us that 
figures on its costs are still estimates, as WFP record keeping has 
not differentiated school feeding within larger categories of 
development, emergency, and protracted relief and recovery operations. 

[98] This diet would provide 1,200 calories and meet or exceed the 
required daily amounts of proteins, fat, magnesium, selenium, niacin, 
and vitamin B6. The diet also would provide iron (60 percent), 
phosphorus (53 percent), zinc (43 percent), thiamin (98 percent), 
riboflavin (43 percent), folate (38 percent), vitamin A (71 percent), 
and vitamin E (51 percent). The diet would provide less than 2 percent 
of the required daily allowance of calcium, vitamin C, vitamin B12, 
vitamin D, and iodine. 

[99] This diet-—consisting of corn-soy blend, corn, wheat, vegetable 
oil, and beans—-would provide 1,200 calories of energy and meet or 
exceed the required daily amounts of protein, lipids, calcium, iron, 
magnesium, selenium, vitamin C, thiamin, niacin, vitamin B6, folate, 
vitamin A, vitamin E, vitamin D, and iodine. The diet also would 
supply significant portions of phosphorus (69 percent), zinc (97 
percent), riboflavin (80), vitamin B-12 (92 percent), and iodine (69 
percent). 

[100] According to Joy Miller Del Rosso, the cost of school feeding 
programs is a major issue for both governments and donors, since 
feeding programs of any kind are expensive. Cost alone can indicate 
little about the value of a school feeding program, she said; but 
unfortunately cost-effectiveness analyses that assess costs of school 
feeding programs relative to their impact on nutrition and education 
outcomes are for the most part unavailable. See School Feeding 
Programs: Improving Effectiveness and Increasing the Benefit to 
Education: A Guide for Program Managers, (Washington, D.C.: World 
Bank, August 1999). 

[101] According to a USAID official, if nutrition is the problem, 
maternal child health and preschool feeding programs are more cost 
effective than school feeding programs. If education is a major 
weakness, investments in educational reform, teacher training, and 
learning facilities are more cost effective. 

[102] In 2001, a USAID contracted evaluation of its school feeding 
program in Haiti, covering the period 1996 to 2000, was completed. 
(The program was primarily a school feeding only operation; however, 
some resources were devoted to food for education activities.) The 
report concluded there is no causal connection between school feeding 
and improved educational performance. Other factors such as school 
quality and parental variables, have a more direct influence on 
educational outcomes, it said. The report found the food for education 
approach to be very promising, provided that food is used as leverage 
to improve school quality. The report recommended USAID consider 
devoting all of the school feeding resources to food for education 
activities. However, USAID decided to phase out school feeding 
activities over a 3-year period. According to a USAID official, Haiti 
was loosing too many kids before they ever got to school. As a result, 
USAID concluded it would be more cost effective to employ the 
resources in a maternal and child health program. 

[103] Early malnutrition coupled with inadequate intellectual 
stimulation and care are likely to result in severe and possibly 
irreversible damage to physical and emotional capacities fundamental 
to further learning. Consequently, programs in elementary schools and 
even kindergarten may be too late to develop these capacities in 
children. 

[104] In commenting on a draft of this report, WFP indicated it agreed 
with the views expressed in this paragraph. WFP noted that it has 
implemented a large number of maternal and child health programs. 

[105] The options were 50 percent of students receive a free snack; 
100 percent of students receive a free snack; 50 percent of students 
receive a free lunch; and 100 percent of students receive a free lunch. 

[106] The 30 planner/practitioners' rankings closely paralleled those 
of the experts. 

[107] World Bank, "School Health" (Washington, D.C., March 2001). 

[108] Some examples that have been cited include school policies to 
increase the number of schools with adequate water and sanitation 
facilities; increase family life education and access to family 
planning services; reduce school dropouts because of pregnancy; and 
reduce discrimination against people with HIV/AIDS and their families. 

[109] Described as the essential first steps toward a healthy 
physical, learning environment. 

[110] In 2001, WFP was included in the FRESH initiative. 

[111] In 2001, WFP was included in the FRESH initiative. 

[112] WFP has an umbrella agreement with USDAs Commodity Credit 
Corporation under which the corporation can donate 416(b) agricultural 
commodities to WFP to carry out activities in specific countries in 
accord with projects approved by the U.S. government through the WFP 
Executive Board approval process and for which the U.S. government has 
agreed to provide commodities. This agreement provided a basis by 
which USDA could make pilot program commodities available to WFP, 
since WFP only requested funding for projects that had already been 
approved by the executive board or, in the case of proposed expansion 
projects, agreements that would require the board's approval before 
they could become operational. 

[113] Altogether WFP submitted 80 proposals covering 48 countries. Of 
the 80 proposals, 30 were for WFP approved projects that had unmet 
food aid needs and 50 were for expansion or new projects that still 
required approval by WFP's Executive Board. 

[114] According to WFP, any budget information that was not available 
on WFP's Web site was made available to USDA upon request. 

[115] The FAPC membership included USDA, USAID, the Department of 
State, the National Security Council, the Office of the President, the 
Council of Economic Advisors, the Office of Management and Budget, and 
the Office of the White House. 

[116] See letter for a discussion of the Bellmon analysis. 

[117] According to OMB, USDA uses CCC Charter Act funds to pay WFP for 
internal transportation, storage, and handling costs. 

[118] According to WFP, this agreement reflects a commitment made by 
the U.S. government in WFP's Executive Board to fully cover the costs 
of delivering its food aid donations. 

[119] According to an executive branch official, the decision had been 
made by November 2000. 

[120] According to USDA, direct administrative headquarters' costs 
could not be covered. USDA allows for indirect headquarters costs to 
be covered through provision of Indirect Cost Recovery principles. 

[End of section] 

GAO’s Mission: 

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

Obtaining Copies of GAO Reports and Testimony: 

The fastest and easiest way to obtain copies of GAO documents at no 
cost is through the Internet. GAO’s Web site [hyperlink, 
http://www.gao.gov] contains abstracts and fulltext files of current 
reports and testimony and an expanding archive of older products. The 
Web site features a search engine to help you locate documents using 
key words and phrases. You can print these documents in their entirety, 
including charts and other graphics. 

Each day, GAO issues a list of newly released reports, testimony, and 
correspondence. GAO posts this list, known as “Today’s Reports,” on its 
Web site daily. The list contains links to the full-text document 
files. To have GAO e-mail this list to you every afternoon, go to 
[hyperlink, http://www.gao.gov] and select “Subscribe to daily E-mail 
alert for newly released products” under the GAO Reports heading. 

Order by Mail or Phone: 

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

U.S. General Accounting Office: 
441 G Street NW, Room LM: 
Washington, D.C. 20548: 

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

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

Public Affairs: 

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