ahmadou sekou

PCD Working Paper
Linking school feeding with agriculture development
Technical meeting on school feeding programmes linked to small holder
farmer production
London, 27-29 September 2011.
Meeting report
Acknowledgements
A large number of stakeholders supported the development of this paper. We would like to thank all
the participants of the HGSF technical meeting for their invaluable contributions and enthusiasm
without which this work would not have been possible. We are very grateful for the contributions by
the participants from the Government of Kenya, including Leah Rotich, Nur Guleid and Philomena
Chege, as well as inputs from the Government of Mali, including Mamadou Doumbia and Makiyou
Coulibaly, and Aboubacar Guindo from WFP Mali. We would also like to thank Harold Alderman,
Carmen Burbano, Sally Burrows, Dick Commandeur, Alessia De Caterina, Lawrence Haddad, Emily
Janoch, Andreas Kretschmer, Elizabeth Kristjansson, Afua Kufuor, Edoardo Masset, Leo Nederveen,
Marc Regnault de la Mothe, Shalini Roy, George Scharfeenberger, Emilie Sidaner and Roberta
Sonnino for their presentations. We would also like to thank the PCD team for the logistics,
facilitation, presentations, reporting back, note taking and support before, during and after the
meeting, including Abby Deamer, Amadou Sekou Diallo, Lesley Drake, Iain Gardiner, Alex Hulme,
Angela Kituara, Jane Lillywhite, Nicola Lloyd, Brie McMahon, Wairimu Muita, Daniel Mumuni, Kristie
Neeser. We are also very grateful to Alesha Black, Leah Ashe for assisting with the research agenda
development, and to Jeff Waage for his chairing and contributions to the meeting. We would like to
thank the Bill & Melinda Gates Foundation for supporting part of this work. This report was written
and edited by Aulo Gelli and Brie McMahon.
ii
Contents
Acknowledgements ................................................................................................................................. ii
Contents ..................................................................................................................................................iii
Executive summary ................................................................................................................................. 1
Session 1: Setting the scene................................................................................................................... 3
Introduction and scope of the meeting (Jane Lillywhite, PCD) ........................................................... 3
Exploring different contexts, different programmes (Aulo Gelli, PCD) ................................................ 5
Exploring institutional frameworks for school feeding (Carmen Burbano and Emilie Sidaner, WFP) 6
School feeding: have we been asking the right questions? (Harold Alderman, World Bank) ............ 7
Policy perspective: Government of Kenya .......................................................................................... 8
Policy perspective: Government of Mali .............................................................................................. 9
Session 2: Exploring potential impacts ................................................................................................. 10
Reviewing the effectiveness school meals and cost-outcomes (Elizabeth Kristjansson, U. of
Ottawa) .............................................................................................................................................. 10
School feeding, public procurement and sustainability (Roberta Sonnino, U. of Cardiff) ................. 11
School feeding impact- operational perspectives (Sally Burrows WFP, Office of Evaluation) ......... 12
How “home grown” are Home-Grown School Feeding Programmes? ............................................. 13
Group work: Linking agriculture, health, nutrition and education ...................................................... 14
Session 3: Programme design: Procurement, governance and social accountability .......................... 16
School feeding- procurement and governance (Dick Commandeur, SNV) ...................................... 16
M&E and small-holder farmer income: the P4P experience (Alessia De Caterina, WFP-P4P) ....... 17
Governance & accountability: School feeding example in Mali (Emily Janoch, Harvard University) 18
HGSF impact evaluation in Mali (Edoardo Masset -IDS, Aulo Gelli -PCD) ...................................... 19
Group work: Procurement and social accountability ......................................................................... 20
Session 4: Country context, needs and targeting approaches ............................................................. 22
Analysing the context for school feeding (Marc Regnault de la Mothe WFP, Kristie Neeser PCD) . 22
Reflections on needs and targeting approaches for school feeding (Harold Alderman, World Bank)
.......................................................................................................................................................... 23
Group work: Analysing needs and targeting approaches for school feeding ................................... 24
Session 5: Focus on school feeding programme design, agriculture and nutrition links ...................... 25
Ration design using “home-grown” foods (Leo Nederveen, PCD) ................................................... 25
School feeding and household level nutrition (Shalini Roy, IFPRI) .................................................. 26
Group work 4: Designing school feeding programmes: agriculture and nutrition links ..................... 27
Analysing the HGSF supply chain (Afua Kufuor GIMPA, George Scharffenberger UC Berkeley and
Daniel Mumuni PCD) ........................................................................................................................ 29
Next steps: building the research agenda ............................................................................................. 30
Presentations-References..................................................................................................................... 31
Annex 1: School feeding implementation scenarios ............................................................................. 32
Annex 2: Break-out group session: Research questions on linking agriculture, health, nutrition and
education ............................................................................................................................................... 36
Annex 3: Break-out group session: Research questions on procurement and social accountability ... 37
Annex 4: Break-out group session: Research questions on targeting .................................................. 37
Annex 4: Participants list ....................................................................................................................... 38
Annex 5: PCD HGSF working papers ................................................................................................... 39
iii
Executive summary
Child development requires a life-cycle approach to intervention. School health and nutrition
programmes are a key part of this continuum, providing the foundation for physical, cognitive and
educational development that will allow children to reach their full and equal potential. Today, every
country for which we have information is seeking to provide food, in some way and at some scale, to
its schoolchildren. However, where the need is greatest, in terms of hunger, poverty and poor social
indicators, the programmes tend to be the smallest. Past experience shows that countries do not
seek to exit from providing food to their schoolchildren, but rather to transition from externally
supported projects to nationally owned programmes. Countries that have made a successful
transition have often explored linking school feeding programmes to agriculture development –an
approach also known as “Home Grown School Feeding” (HGSF).
In 2003, African governments included locally-sourced school feeding programs as a key intervention
within the food security pillar of the Comprehensive Africa Agriculture Development Programme
(CAADP). That same year, the New Partnership for Africa’s Development (NEPAD), launched a pilot
HGSF programme. Many countries, including Cote d’Ivoire, Ghana, Kenya, Mali and Nigeria, are
implementing national programmes, and many more are seeking to develop or strengthen scaled
and sustainable HGSF. In response to this demand, the World Bank, WFP, PCD, and other key
stakeholders have been working together since early 2008 to help governments develop and
implement cost effective, sustainable national school feeding programmes.
HGSF provides an integrated framework with multiple impacts across agriculture, health, nutrition
and education, but even with recent efforts, there are several important gaps in the knowledge on
optimal implementation and measures of effectiveness of HGSF. Some key questions include, how
best can the potential of school feeding be maximised to support multi-sectorial integrated
frameworks linking agriculture, health, nutrition and education? How can HGSF be a win-win for
agriculture, education and health?
Meeting objectives and approach
The technical meeting on “Linking School Feeding with Agriculture Development” was part of an ongoing series of meetings framed around the development of a harmonised school feeding research
agenda, with a particular focus on
 Recent research on school feeding and links with small holder farmers, examining linkages
between possible effects on agriculture, nutrition and education
 Forthcoming research, including the design of the school feeding sourcebook, transition
analysis and HGSF impact evaluations
The meeting brought together stakeholders from different sectors and disciplines, bridging policy,
research and practice. The sessions included plenary discussions and group work aimed at
maximising the interaction between participants. The meeting began with an overview of the
different models of school feeding, examining programme design and the institutional framework
perspectives. The content focussed on exploring different programmatic elements of the system
linking small holder farmers to school children, including scoping the potential impact of HGSF across
agriculture, health, nutrition and education. Cross-cutting issues across the implementation models
were examined in detail, including procurement, governance and accountability.
Key messages and next steps
The discussions provided a wealth of content that will be mainstreamed within the on-going
development of the research agenda of the HGSF programme. Key research themes included:
1. HGSF programmes combine programmatic activities with potential education, health and
nutrition, agriculture and social protection objectives. In order to manage the trade-offs explicitly
across these objectives, it is critical that the objectives are clearly defined (including
differentiation between primary and secondary objectives), alongside clear articulation of the
1
programme theory linking the different target groups, the programme activities and the impact
pathways.
2. Clarifying the objectives is also critical in terms of defining HGSF within the context of an impact
evaluation, and in particular, identifying what the control group will be. Are we comparing a local
versus national/central procurement model? Does it involve nutrition, education, and agricultural
comparisons? How would small holder farmers be sampled?
3. Whichever food procurement approach is chosen, its primary objective must be the timely and
stable supply of quality food for school feeding. Support to small holder farmers to respond to
the demand from the school feeding market should be framed under the objective of safe and
stable supply. Research is needed to explore the trade-offs of different pro-small holder
procurement models, analysing the data on the costs and impacts, including issues around
market integration, scale, timeliness, prices, food types (including perishables), and seasonality.
4. Following form the preceding points, as HGSF has potential simultaneous impacts, how do we
aggregate these effects in a single metric? This is particularly important when comparing the
cost-effectiveness of HGSF to other interventions.
5. Issue of decentralisation highlights inherent tension between procurement processes’
prioritization of transparency, accountability and value for money, with the HGSF objective to
prioritize procurement from smallholder farmers. These competing, though not completely
incompatible priorities must be balanced in order to generate stakeholder buy-in and the
intended financial and social returns.
6. More detailed descriptions are needed on the various models of HGSF, including information on
how programmes began and then evolved over time, capturing good practices and lessons
learned. Information on how high and middle income countries are implementing and financing
school feeding programs was also raised as an important area of research.
7. There is an opportunity to document and analyse the process, or transition, from externally
driven programmes to country owned, sustainable programmes, identifying where the transition
works, and where it doesn't. Are there particular characteristics of country governments which
make this transition easier?
8. Clarifying the nature of the programme activities will also enable more accurate estimation of the
full implementation costs of HGSF, including different activities and actors across the supply and
value chains. It is particularly important to capture all the “hidden” costs, including community
level contributions that are often substantial and have been until recently overlooked.
9. Developing a standard costing and impact framework (including standardised indicators) for the
different implementation models will enable improved comparability of the cost and costeffectiveness estimations.
10. A particular challenge in terms of aggregating the impacts of HGSF hinges on the issue that the
value of transfers does not easily aggregate with the other effects. In particular, the aggregation
will require quantifying the weight society puts on consumption of the poor relative to that of the
average citizen, which is not straightforward.
The technical meeting provided a multi-disciplinary platform for the development of a harmonised
research agenda for school feeding, providing key pointers in terms of identifying the gaps in the
evidence for HGSF. PCD and partners now have a solid foundation on which to build on-going and
future research activities, including in particular i) a case study source book; ii) a transition analysis;
iii)a collection of HGSF innovations; iv) impact evaluations; and v) supply chain analyses. These five
products will be developed under the World Bank, WFP, PCD global school feeding partnership.
2
Session 1: Setting the scene
Introduction and scope of the meeting (Jane Lillywhite, PCD)
Child development requires a life-cycle approach to intervention. School health and nutrition
programmes are a key part of this continuum, providing the foundation for physical, cognitive and
educational development that will allow children to reach their full and equal potential1. Today,
every country for which we have information is seeking to provide food, in some way and at some
scale, to its schoolchildren. However, where the need is greatest, in terms of hunger, poverty and
poor social indicators, the programmes tend to be the smallest2. Past experience shows that
countries do not seek to exit from providing food to their schoolchildren, but rather to transition
from externally supported projects to nationally owned programmes. Countries that have made a
successful transition have often explored linking school feeding programmes to agriculture
development –an approach also known as “Home Grown School Feeding” (HGSF)3.
Figure 1: School feeding, global picture (Source: Bundy et al., 2009)
In 2003, African governments included small holder-sourced school feeding programs as a key
intervention within the food security pillar of the Comprehensive Africa Agriculture Development
Programme (CAADP). That same year, the New Partnership for Africa’s Development (NEPAD),
launched a pilot Home-Grown School Feeding and Health Programme. Many countries, including
Cote d’Ivoire, Ghana, Kenya, Mali and Nigeria, are now implementing national programmes, and
many more are currently seeking to develop or strengthen scaled and sustainable HGSF. In response
to this demand, the World Bank, WFP, PCD, and other key stakeholders have been working together
since 2008 to help governments develop and implement sustainable national school feeding
programmes. HGSF provides an integrated framework with multiple impacts across agriculture,
health, nutrition and education4, but even with recent efforts, there are several important gaps in
the knowledge on optimal implementation and measures of effectiveness of HGSF, especially given
1
Bundy DAP (ed). Rethinking School Health: A Key Component of Education for All. World Bank, 2010.
Bundy DAP, Burbano C, Grosh M, Gelli A, Jukes, M and Drake, L. Rethinking School FeedingWorld Bank, 2009.
Sumberg, J. and Sabates-Wheeler, R. (2011) 'Linking Agricultural Development to School Feeding in Sub-Saharan Africa: Theoretical
Perspectives', Food Policy 36.3:341-349.
4
New Partnership for Africa’s Development. 2003. “The NEPAD Home-Grown School Feeding Programme: A Concept Note.”
2
3
3
the complexity of the system. Some key questions include, how best can the potential of school
feeding be maximised to support multi-sectorial integrated frameworks linking agriculture, health,
nutrition and education? How can HGSF be a win-win for agriculture, education and health? There is
a need to answer these research questions operationally, building the evidence base to help policy
makers manage the trade-offs across the multiple school feeding objectives5.
The technical meeting on “Linking School Feeding with Agriculture Development” is part of an ongoing series of meetings framed around the development of a harmonised school feeding research
agenda, with a particular focus on
 Recent research on school feeding and links with small holder farmers
 Understand linkages between possible effects on agriculture, nutrition and education
 Explore possible impacts on community development, governance and accountability
Meeting overview
The meeting brings together stakeholders from different sectors and disciplines, bridging policy,
research and practice and included plenary discussions and group work aimed at maximising the
interaction between participants. The meeting structure was designed to first explore the potential
impact pathways for HGSF quite broadly and from a systems perspective and then progressively drill
down on key programme design components including; procurement and governance of the supply
chain, understanding the context for HGSF and targeting the intervention; ration design and
household level nutrition. In particular:
 Day 1 begins with an overview of the different school feeding models with links to
agriculture, looking at both programme design and the institutional framework perspectives.
Panel presentations will explore the potential impact of HGSF across agriculture, health,
nutrition and education, and provided the backdrop for group exercises aimed at
consolidating the impact theory of HGSF.
 Day 2 continues the analysis of the design and implementation of different models,
focussing on systems strengthening to improve procurement, governance and
accountability. These cross-cutting issues are examined in detail within the context of the
design of an impact evaluation of HGSF in Mali. This is followed by an overview of country
level benchmarking exercises, and break-out groups examining needs and targeting
approaches across different target groups including school children and small-holder
farmers.
 Day 3 explores the evidence and gaps in terms of school and household level nutrition and
then focusses on supply chain analyses for school feeding programmes linked to smallholder farmers. This is followed by a breakout session aimed at consolidating the school
feeding research agenda, with a particular focus on key products that will be delivered
during the course of the next 12-24 months.
Outputs
 Meeting report summarising presentations and discussions
 Validated design of on-going research, including school feeding source book, transition
analysis and HGSF impact evaluations
 Working paper developing the school feeding research agenda
This report is the first output of the technical meeting, summarising the key findings from the
meeting, with the content of the report following the overall structure meeting.
5
Gelli A, Neeser K and Drake L. Home Grown School Feeding: linking small holder agriculture to school food provision. London: Partnership for
Child Development, 2010.
4
Exploring different contexts, different programmes (Aulo Gelli, PCD)
School feeding can be defined as the provision of food to children through schools. In general, school
feeding programmes come in one of two basic modalities, on-site meals or snacks, or take home
rations. In some contexts school feeding programmes combine on-site meals/snack programmes
with an extra incentive from take-home rations targeting a specific group of vulnerable children
identified in the problem analysis (e.g. orphans, or older girls).
School feeding and HGSF programmes exhibit different, context-specific models or configurations.
Different approaches can even co-exist within the same country, where, for instance, programme
implementation is owned by decentralised institutions (e.g. individual states in Brazil or India), or
where agencies like WFP are complementing the national programmes (e.g. Ghana and Kenya). PCD
and partners have been undertaking a scoping analysis to understand HGSF system using a
standardised approach. Figure 2 shows stylised supply chain linking food production to food
distribution in schools shown alongside examples of different implementation models in three
countries.
Organization
of farmers
Production
of food
Agriculture sector and food
production
Fully decentralised
school-based model
(e.g. Kenya HGSM)
Partially decentralised
model (e.g. Mali)
Integrated
farm to school model
(e.g. Cote d’Ivoire)
Wholesale,
Trading
Food procurement
Transportation
& Storage
Processing &
distribution
to schools
Logistics and processing
Food
Preparation
Distribution
To Children
Food preparation and
feeding
Schools responsible for procurement and preparation using funds
from the central level
Traders
Schools
Women’s groups receive supply side package and provide
schools with increasing supply of food.
Figure 2: Stylised HGSF supply chains and example country models
The emerging policy consensus amongst the stakeholders involved in the scoping analysis suggested
that HGSF in sub-Saharan Africa is a key tool in the transition towards “nationally owned” school
feeding programmes. Three distinct target groups were identified for HGSF, including not only school
children, but also small holder farmers and community based groups delivering support services to
school feeding. At impact level, HGSF had the potential to improve food security for small holders
and other community groups, however in order for this to happen an explicit programme
component, other than food procurement, was required to support agriculture and community
development. This component at a minimum included providing sensitisation campaigns around
improved production practices, income generation activities in support to school feeding and on
improved nutrition practices. This perspective confirms the key role of Ministries of Agriculture, the
relevance of HGSF as a key intervention within Pillar 3 of the CAADP framework, and the importance
of mainstreaming HGSF within country level CAADP compacts. The evidence base of the impact of
HGSF on food security, alongside the associated incremental costs, however, is still missing and will
need to be built to inform policy and programme design.
5
Exploring institutional frameworks for school feeding (Carmen Burbano and Emilie Sidaner, WFP)
In response to the increased country demand for school feeding, international partners have joined
forces to help governments improve the efficiency and sustainability of on-going programs and to
scale them up as needed. In 2008, the World Bank Group launched a Global Food Crisis Response
Facility that mobilized US$ 1.2 billion to help countries respond to shocks, including scaling up school
feeding programs. Education funding, including Fast Track Initiative funding, has been allocated to
national school feeding programmes as well. In 2009, WFP – one of the largest implementers of
school feeding programs in the world- reset its corporate policy on school feeding emphasizing
sustainability, government ownership and move from food aid to food assistance. The Bill and
Melinda Gates Foundation has provided resources to WFP to strengthen local procurement through
a program called Purchase for Progress (P4P), and to the Partnership for Child Development to
support school feeding programs that help small farmers to connect with school feeding markets. At
the regional level, NEPAD through the Comprehensive Africa Agriculture Development Programme
and the Alliance for a Green Revolution in Africa, have been actively supporting the connection
between school feeding and domestic agricultural production. Other important partners are Brazil,
Chile and Russia, which provide South-to-South technical support to low-income countries.
Situation analyses have been conducted in a number of countries in order to understand where
countries are in the transition towards national ownership. For illustrative purposes, focus on
findings of work in Haiti, Kenya, and Mali, three countries with important differences in institutional
frameworks for their school feeding programmes.
Figure 3: Stylised school feeding institutional frameworks in Mali and Kenya.
This analysis highlighted a number of cross cutting issues focussing on the important trade-offs
associated with alternative models, including management and administration, institutional home of
the programme, financing, accountability, quality assurance and M&E, amongst others. Although still
at the initial stages of collaboration, and with several important research pieces in progress or yet to
come including an in-depth analysis of the transition to national ownership, this analysis highlights
that there is a gap in knowledge about the institutional set up and governance for school feeding
within national public administration systems. The transition to government ownership of school
feeding programs is an iterative learning process both for national institutions and for development
partners, which depends on a myriad of endogenous and exogenous factors. There is a need to study
how countries are transitioning to national ownership not with the intention of defining a roadmap
or a fixed set of milestones, but with a focus on gathering information about how to create an
enabling environment in which these learning processes can take place in a given context. In
addition, the transition process has often occurred in the context of decentralisation, which in turn
has highlighted constraints in terms of management and implementation capacity, as well as in
opportunities to structure the market for small-holder farmers.
6
School feeding: have we been asking the right questions? (Harold Alderman, World Bank)
“Rethinking School Feeding” analysis included a comparison of school feeding and education
expenditures. Is this a fair comparison? School feeding is not just an education intervention, should
we view school feeding as a cost to education or as a cost to some larger development goal? To
answer this question we conducted a review of the recent evidence on school feeding.
Do the results reviewed imply that school feeding is among the best investments in nutrition?
Despite new evidence indicating favourable externalities to siblings of students, and the clear
benefit in addressing hunger in schoolchildren, the fair answer to this question is no. While school
feeding can provide iron and other key micronutrients, these programs are not designed to address
the most critical nutritional constraints in low income settings, simply because they are not targeted
at the most vulnerable period in child development, which is between conception and two years of
age.
Do the results imply that school feeding is the best way to use funds for education? Again, the quick
answer is likely no. However, in this case, the answer is more nuanced. School feeding is not a
substitute for a well-organized education system and teacher performance. However, there is
extensive evidence that school feeding can complement a good education program. So although
school feeding may not be the best education response it may be an important element in achieving
an effective education system.
“Declining ratio of school feeding
to education expenditures: Food
budgets increase somewhat over
GDP range but other schooling
expenditures
increase
more
rapidly. The ratio is surprisingly
constant at 10 to 20 percent for
middle-income and rich countries,
but for a few countries, school
feeding cost per beneficiary is as
much as is spent on the average
student in basic education or
nearly so.”
Figure 4: Ratio of per child cost of school feeding versus per child cost of basic education, plotted against GDP per capita
Do the results imply that school feeding is a plausible candidate for a social protection investment on
a par with conditional cash transfers (CCTs)? Here the fair answer appears to be: quite likely. School
feeding can increase human capital investments while also providing support to poor households.
Thus they serve as a support to current poverty reduction while making the need for future transfers
and assistance less likely. The dual objectives of raising current consumption while promoting
investments, however, make it difficult to compare outcomes of either CCTs or school feeding with
direct investments. The value of transfers does not easily aggregate with outputs in a benefit cost
assessment. For one thing such a summation requires a quantification of the weight society puts on
consumption of the poor relative to that of the average citizen. Absent this calculation, a direct
comparison of demand-side interventions for education or direct investments in health with a school
feeding transfer does not put both categories of expenditures on the same metric. There are also
issues around the reliability of the program. Beryl Levinger in the 1980’s reviewed school feeding
programs and none of the programs were delivering food for 180 days per year. The evidence she
assembled could support increased efforts to achieve the 180 day target or it could support the view
that school feeding is inherently ungovernable- the interpretation of her studies depends on how
one views the implementation constraints more than the efficacy of school meals themselves.
7
Policy perspective: Government of Kenya
Kenya is a country in Eastern Africa classified by FAO as Low-Income Food Deficit country (LIFDC),
with a population of approximately 39 million people, over 40% of whom are under 15 years of age.
Kenya ranks 147th on the United Nations Development Programme (UNDP) Human Development
Index table, with an average life expectancy at birth of 54 years, an adult literacy rate of 74%, and a
gross domestic product (GDP) purchasing power parity (PPP) per capita of US$1,542 (UNDP, 2009).
Access to primary education has dramatically increased since the introduction of free primary
schooling in 2003. According to the latest Population and Housing Census, in 2009 primary school
net enrolment ratio was estimated at 77%, with considerable gender disparities varying across the
country. Kenya has a rich history of school feeding programmes and there are currently 3 main
models of school feeding implementation:
 Home Grown School Meals Programme (HGSM): Implemented by the Ministry of Education
reaching about 600,000 children; Undertaken at school level and co-ordinated by members
of the School Meals Committee; Cash is transferred directly to schools in arid and semi-arid
areas on a biannual basis; financed through the state budget. Expanded this programme as
part of emergency response to food security crisis by reaching an additional 900,000
children.
 WFP SF programme: Initiated in 1980, the WFP school feeding program reached 1.3 million
children at its peak but has since scaled back to 678,000 in 2010. Food is sourced from
outside the country and the program is financed through donor support. WFP is phasing
down its programme, progressively transferring 50,000 children to the HGSM programme on
a yearly basis.
 Njaa Marufuku Kenya Home Grown School Meals (NMK): A collaborative initiative by
Agriculture sector Ministries together with Ministry of Education and Ministry of Public
Health. A cash grant is provided directly to the school, which then purchase local foods.
Funding is scaled out after a period of 3 years and the community then has complete
responsibility for the program. Significant focus is given to capacity building for the
community, volunteers and the local farmers to ensure programme sustainability.
Institutional capacity and coordination: the lead Ministry for the HGSM programme is the Ministry of
Education and the lead Ministry for the NMK programme is the Ministry of Agriculture. Both
Ministries have similar management structures. The HGSM programme is currently managed by a
technical unit, within the Directorate of Basic Education. The NMK programme is managed by a
technical unit within the Ministry of Agriculture.
Financing: The HGSM programme has received KSh 400 million per year in the last two financial
years (2009/10, 2010/11) from the education sector budget. THE NMK programme has received KSh
116.9 million for the first 5 years of implementation from the coordinating partnership (MoA, MoE,
MoPH and FAO).
The main successes and challenges for the national programme are summarised in Table 1.
Table 1: Kenya national school feeding programme: Successes and challenges.
Successes
Efficient and timely disbursement of funds
Establishment of school based committees to
coordinate procurement of food from local market
Economic benefit: emerging evidence that local
farmers have access to market for their produce
Multi-sectoral approach: coordinating linkages
between Government Ministries and partners.
Sustainability: gradual handing over from donor to
government
Challenges
Strengthening
capacity
for
governance
and
accountability at the local level
Expanding linkages with local farmers
Expanding HGSM to reach more beneficiaries
High poverty rates, harsh climate (droughts)
Sustainability of programmes
Hard evidence of impact: plan for collaborative allinclusive impact evaluation
8
Policy perspective: Government of Mali
Mali is a land-locked country in Western Africa, classified by FAO as LIFDC, with a population of 14
million people, over half of whom are under 15 years of age. According to UNDP, Mali is ranked
178th in the Human Development Index table, with an average life expectancy at birth of 48 years,
adult literacy rate of 26 percent and a GDP per capita (PPP) of $1083 USD. At country level, Mali has
seen remarkable progress in terms of access to school (net enrolment ratios increased from 20
percent in 1990 to 66 percent in 2007), though the levels of enrolment are still well below the
average for Sub-Saharan Africa and completion rates are very poor. A large proportion of children,
girls in particular, are excluded from the schooling system. In Mali, the majority of farmers are
involved in the production of subsistence crops, mainly millet and sorghum. Production is carried out
with poor technology and minimal inputs, and access to credit is a major constraint. Production
levels are not only poor but also highly variable. Output is largely dependent on rainfall and
fluctuates considerably from year to year and season to season. The Government of Mali is looking
to respond to some of these educational and agricultural challenges and has highlighted its
commitment in the school feeding policy approved by Cabinet Ministers in November 2009.
Different school feeding programmes are implemented in Mali, including:



National school feeding programme: Implemented at commune level; financed through the
state budget; procurement at commune level through national and local level suppliers
WFP SF programme: Implemented through school management committees (CGS) and
commune level education centres (CAP); financed by WFP through donor support;
procurement mainly at central level through national level suppliers, and through small
holder organisations linked to P4P programme
CRS food for education programme: Implemented by CRS through CGSs and CAPs; financed
by donors including USDA mainly through food aid donations, piloting school level
procurement
Institutional capacity and coordination: Until the recent creation of the National School Feeding
Centre and the appointment of a national director, the school feeding programme was the
responsibility of the Ministry of Education, with one national school feeding coordinator and focal
points at decentralised levels. National coordination framework including key stakeholders
(Ministries, DP, NGO’s, private sector) meeting on monthly basis.
Financing: The Government of Mali has demonstrated clear ownership by also allocating nearly $5
million USD (2.6 billion CFA) in 2010 and $5.8 million USD in 2011 (3.1 billion CFA) for programme
implementation.
The main challenges and opportunities for the national programme are summarised in Table 2.
Table 2: Mali national school feeding programme: Challenges and opportunities.
Challenges
In general
Roll-out of new national school feeding centre
Involvement of other key Ministries, including
Agriculture and Health as outlined in the National
Policy
HGSF specific
Weak organization of small holder farmers;
Securing regular supply ;
Resources available;
Gaps in coordination among different sectors and
local authorities at local level ;
Small holders have insufficient storage capacity;
Access to inputs, to credit
Opportunities
In general
Information, reporting and accountability systems need
to be developed and strengthened;
MoE capacity is limited (financial and in HR);
Monitoring and evaluation
HGSF specific
HGSF can offer a stable, structured and relatively
predictable demand to small holders;
Commitment from government;
Supportive institutional framework;
Several experiences of local procurement from which
to learn;
Markets and local production available
9
Session 2: Exploring potential impacts
Reviewing the effectiveness school meals and cost-outcomes (Elizabeth Kristjansson, U. of Ottawa)
Undertook a Cochrane review of school feeding programmes in disadvantaged children included
trials from five continents and spanned eight decades. Although we found that the programmes
have significant positive effects on growth and cognitive performance, the trials had many different
designs and were implemented in varying social contexts and educational systems; by staff with
different backgrounds, skills, and cultural beliefs; and with huge variation in the prevailing social,
economic, and political context. Simply knowing that feeding programmes work is not enough for
policymakers to decide on the type of intervention that should be implemented. We therefore
looked at the trials more closely to determine the aspects that determine success and failure in
various situations. Also looked at analysing the standardised costs of school meal programmes and
combined these estimates with results of the systematic review to calculate cost per outcome.
Table 3: School feeding: cost and outcomes (Source: Galloway et al., 2009).
School meal programs should be well-designed, and provide adequate nutrient content, including
micronutrients. The amount and - and maybe amount and type of fat and cholesterol should also be
taken into account given their role in structure and function of parts of the brain which continue to
develop into adolescent and influence cognitive outcome. Food should be appealing and
demonstrate healthy and affordable dietary choices for later in life. As shown in this review, it is
very important to encourage full participation, and to measure participation and consumption
carefully. We might tentatively conclude for future trials that measuring non-consumption of food is
best done at the level of the individual child and include displacement. Palatability and special needs
of the target population are also extremely important. Food should be appealing, acceptable, and
locally available. We noted that most of the researchers in included studies went to a great deal of
trouble to ensure that foods were culturally acceptable and tempting to the palate. School meal
programs take place in a context which may significantly impact on their effectiveness. Thus, in
addition to school meals, schools should have the basic amenities necessary to good education;
adequate materials, good teaching, and an emphasis on development of the whole child. We
conclude that, as other authors have suggested, school meals should be one of multiple
interventions designed to improve the health and development of poor and marginalized children.
10
School feeding, public procurement and sustainability (Roberta Sonnino, U. of Cardiff)
“The banking crisis of 2008 led the world to the brink of financial disaster and shook the dominant economic
model to its foundations. […] The model was always unstable ecologically. It has now proven itself unstable
economically. […] An appropriate response is to question the underlying vision of a prosperity built on continual
growth. And to search for alternative visions.” Tim Jackson, UK SDC, 2009.
The challenge of sustainable development lies at the heart of HGSF- or school feeding programmes
linked to small holder agriculture and community development. Seen from the wider context of
global food systems, where the bulk of agriculture subsidies are generally geared to a small number
of basic commodities, there has been an emergence of a “new food equation” since the food price
crisis of 2008.
East Ayrshire, a deprived rural county in Scotland has been the focus of a recent reform of a school
food programmes that can provide useful insights in terms of innovative, multi-sectoral planning
aimed at environmental stewardship, connecting children with food and with where the food comes
from. In 2005, school meals in 12 of the County’s 45 primary schools were reformed, including the
following measures:
 Pasta, rice, potatoes, fruit and vegetables were increased to promote health benefits
 Fat, sugar, and salt were reduced
 Added colourings, artificial flavourings and GM-foods were banned
 Fresh and unprocessed ingredients were prioritized
 Local suppliers actively involved to build relationships between producers and the Council
and to provide guidance on tendering.
Strict “straightness” guidelines for class 1 vegetables made more flexible. The bidding contract was
divided into 9 lots with innovative award criteria, including the ability to supply to deadlines, quality
and range of foods offered, food handling arrangements and facilities, and proposed use of
resources. Eventually, 45 primary and 2 secondary schools involved in the school food reform where:
 50% of the ingredients utilized are organic, 70% are locally sourced and 90% of the food
served is unprocessed
 In addition, an “Ethical incentive scheme” was developed to engage children with the world
food system.
The benefits of local procurement included:
 Multiplier effect of £ 160,000/12 schools
 Prevalence of obese and overweight children dropped by 30% and 22% respectively
between 2008 and 2010
 Food miles curbed by 70%, as well as less packaging waste, leading to projected annual
savings of 38 tonnes of CO2 emissions/school
 Social return on investment Index of 6.19
“I like the school [market], because […] if you educate them to good eating, then it starts to affect the whole
structure of the economy. Later on, when they grow up and they have children, it gets passed on.” School food
provider, East Ayrshire.
“The business we do with East Ayrshire is very important to us in terms of dealing with a local customer, we
don’t want to lose that customer […]. We need to look over each other. […] Children are the future.” School
food provider, East Ayrshire.
Some of the key lessons from the East Ayrshire experience highlight the role of Home-Grown School
Feeding as a dynamic process of change that:
 Integrates production, consumption, health and education
 Involves all actors in the food chain
 Is based on a long-term vision that builds our collective commitment to sustainable
development
11
School feeding impact- operational perspectives (Sally Burrows WFP, Office of Evaluation)
New WFP school feeding policy framework identifies 5 possible objectives for school feeding:
Education, nutrition, value transfer, alongside capacity development and wider economic
development (including agriculture). WFP Office of Evaluation identified 7 common challenges from
mixed-method impact evaluations of school feeding conducted in 5 countries, namely Bangladesh,
Cambodia, Cote d’Ivoire, The Gambia and Kenya.
1. Different modalities of school feeding contribute to the various objectives in different ways:
School feeding can contribute to three main objectives but no single modality is best for all
objectives. Programme should be designed according to priority objectives: Is school feeding
the best social protection intervention for the particular case?
2. School environment matters: Quality of education and the school environment are major
contributing factors in effectiveness concerning education goals and household decision
making. School feeding cannot compensate where these are poor but effectiveness of
school feeding is enhanced where they are good. This is particularly relevant as WFP/UNICEF
‘Essential Package’ is not implemented consistently.
3. Understand the real value transfer- factoring in “hidden” costs: Real cost of the meal should
include contributions made by the community (cash or in-kind labour, food …etc…). Factor in
costs of schooling, both financial costs (e.g. books, uniforms) and opportunity costs (e.g.
income foregone from child labour). What influences the decision to send children to
school? Balance willingness to pay, costs of education and value transfer. Value transfer at
household level is particularly difficult to quantify, developed different methodologies
accounting for net transfer, taking costs and willingness to pay into account. Other
operational issues can affect value transfer: pipeline breaks, non-attendance, children’s cash
contributions, target children not eating, non-target beneficiaries eating (e.g. cooks).
Positive effects on value transfer can arise from household food savings, more food available
for other household members and freed-up time for income generating activities.
4. Targeting is crucial: Some improvements have been made but more can be done. For safety
net effect, targeting must be based on household level analysis (see point 3).
5. Seasons matter: For safety nets, cover the lean season and/or when in school holiday,
and/or labour seasons (e.g. harvest).
6. Community engagement can help: Where blockages are cultural or religious and for
reducing risk of food diversion/dissipation. Action: encourage use of school for adult
learning and community asset/resource.
7. Elaborate capacity development: Whose capacity? Who provides the technical assistance?
What kind of assistance (‘doing it for’ does not build capacity)? What funding?
It is particularly important to analyse the trade-offs between the different objectives. For example,
social protection equates to needs based approach (including education incentives, nutrition, and
value transfer). However, agricultural development may focus on agriculture potential based, which
might not be compatible with other objectives. Another example of possible tension comes from the
question of where will school feeding have the greatest impact? Is it enabling girls to transition to
secondary school? Or getting the last ‘per cent’ of primary school age children into school?
12
How “home grown” are Home-Grown School Feeding Programmes?
Home-Grown School Feeding (HGSF) tries to set up school procurement of food for school meals in
such a way that it stimulates the local food economy while improving school enrolment, attendance
and achievement as well as food and nutrition security. Achieving just one of these outcomes
(educational, nutritional, local economy) is a challenge. Nevertheless we do know that there are
certain conditions under which all three objectives can be realised (see Adelman et. al. 2006 for a
nice review).
The HGSF programme is working with governments and other national partners to build on regular
school feeding programmes (where food is sourced nationally but not locally) to implement and test
various HGSF approaches. Here are a few reflections from the three preceding presentations:
 HGSF is such a neat idea, there is a danger that we can get a bit evangelical about it—we
need to give the evidence every chance to speak
 Assessing impact is difficult, because there is no real way of aggregating across multiple
outcomes–the PCD team is working on this
 The systematic reviews of school feeding programmes (not the programmes that rely on
locally sourced food) show how sensitive impact is to key design features such as calling the
food a snack versus a meal (this has a big impact on food substitution away from the child in
the home—families being more likely to treat something framed as a snack in an additive
way). This means that there is a need to build in these variants into impact evaluations,
which is expensive if done using surveys (as it increases sample sizes dramatically)
 This design sensitivity has implications for capacity development and for sustainability.
Innovation and adaptation require capacities at the institutional, organisational and
individual levels, so capacity development efforts are intrinsic to the scaling of HGSF
 The design sensitivity also implies that unless the HGSF programmes are “doubly” home
grown (i.e. developed locally as well as using local food) they may well fall apart when the
outsiders leave
 Ultimately this potential for scaling and sustainability via established institutions (i.e.
schools) might be the trump cards that HGSF has when we benchmark its impacts against
interventions like public works and cash transfers.
Reflections on Home-Grown School Feeding by Lawrence Haddad, Director of the Institute of Development
Studies and member of the HGSF steering committee.
http://www.developmenthorizons.com/
13
Group work: Linking agriculture, health, nutrition and education
Three implementation scenarios were used to explore some of the trade-offs associated with the
different school feeding models throughout the meeting (see annex 1 for details). The scenarios
referenced the stylised supply chain shown in Figure 1, and included:
1. Fully decentralised, school based model (e.g. Kenya MoE)
2. Partially decentralised model (e.g. Mali)
3. Integrated farm to school (e.g. Cote d’Ivoire)
In this session the three break-out groups explored some of the programme theory on the links
between agriculture, nutrition and education in each of the three different scenarios, with a
particular focus on identifying the potential short, medium and long term effects. The purpose of
this work was not to determine which model is 'best', but rather what efficiencies or innovations
could be shared across the different scenarios. Questions that guided this scoping session included:




What do you believe the long term goals and objectives of the school feeding programme
should be?
What do you believe the programme activities to achieve the goals and objectives? Are
these activities appropriate for the goals and objectives that have been identified?
What would be good indicators for measuring programme results?
What are the key gaps in the evidence base?
Key messages
The group work highlighted that HGSF, in principle, provides an integrated framework with multiple
impacts across agriculture, health, nutrition and education, including potential direct benefits, spillovers, and trade-offs. However, the nature of these multiple effects and interactions is not yet very
well understood, particularly when we focus on small-holder agriculture, highlighting the need for
further research in this area.
There was a general consensus amongst the three break-out groups on framing the education
objectives around improved school participation (including enrolment, attendance and drop-out),
and cognition and learning. This finding reflects the current state of the evidence base of school
feeding programme impact.
From the nutrition and health perspectives, the break-out groups highlighted that by combining food
provision with behaviour change messaging and sensitisation school feeding programmes had the
potential to improve nutrition and health both in school and at household level. In school, the
programme activities would focus on delivering nutritionally balanced menus throughout the school
year, with a particular focus on adjusting the service delivery to account for seasonality, both in
terms of food production and diet diversification, and in terms of seasonal hunger. The groups also
highlighted the opportunity to focus on specific age groups with the biggest potential to benefit
from the intervention, for example pre-schoolers or adolescent girls. The school could also provide
an entry point to reach households in the wider community through campaigns aimed at improving
health and nutrition practices at household level, including, for example, diet diversification, water
and sanitation and healthy eating practices.
Discussions on the agricultural side were broader and are summarised in Table 4. The 3 groups had
increasing small-holder farmer food production as a common objective. Similarly, the groups all
highlighted building small holder farmer production capacity mainly through extension services.
14
Table 4: Break-out groups- HGSF: agriculture impacts and activities
Fully decentralised
Goals and
Programme features
objectives
and activities
Farmers
are Sensitization of farmers
able to access
Procurement standards
school feeding
allow for access by local
markets
farmers
Increased
Purchases from farmers
production
who have surplus to
Sustained
avoid significant
quality of food
impacts on the local
market
Strengthened
farmer
Working with Partners
associations,
to build the capacity of
groups and
the farmers who do not
networks
have surplus.
Infrastructure provided
to farmers to assist
them in moving food
(transport, storage,
etc.)
Partial decentralisation
Goals and
Programme features
objectives
and activities
To enhance
Each mayor forms a
the country’s
committee that would
capacity to
include officials from 3
produce food. sectors so that better
links.
To enable
small farmers
Improve information
to become
systems on the
more
demand and supply.
competitive
To increase
To deliver
productivity of local
more
farmers and to better
perishable
link them to traders to
(fresh) food to be able to comply with
target
purchasing conditions.
schools.
To improve links
between MoA and
other stakeholders.
Integrated farm to school
Goals and
Programme features
and activities
objectives
Increasing
Guaranteed
food
market
production
Extension
through the
services
capacity of
provided (inputs,
the women’s
seed, support for
group.
legal creation)
To have a higher
percentage of food
bought directly from
local farmers.
Local farmers are
connected with the
national cereal board
To break down
barriers in
procurement
conditions, making it
more flexible so that
local farmers can
supply.
Training on both sides.
Cross cutting research themes
 What is the “control group”, what do we compare HGSF to? Is it traditional school feeding
programs, or the absence of a programme? This in turn is linked to what the objectives of
the intervention are.
 How can we clearly articulate objectives and benefits across the sectors? This will likely
mean exploring from multi-sectorial programs that have been running for some time.
 These are complex programs/systems with the opportunity to reach multiple goals, but
there are trade-offs which clearly emerge- how do you manage the tradeoffs between
different sectors?
 How do we account for benefits across these sectors? There is a need for indicator
development that can account for multi-sectorial objectives. Which are valid and feasible in
terms of money and timing?
 Translating multi-sectorial outcomes into economic benefits isn’t straightforward.
Particularly when looking at transfers – how do you value the consumption of a poor child
compared to an average person?
 There is a need for longitudinal studies and data sets as time frame for potential benefits is
beyond the 1-2 year period covered in recent evaluations.
 Lack of data on complementary activities (e.g. deworming, supplementing, fortification) and
opportunities for synergies across the sectors- need for mapping of activities and potential
partners in the different intervention areas.
Some of the more specific research issues from the three break-out groups are captured in Annex 2.
15
Session 3: Programme design: Procurement, governance and social accountability
School feeding- procurement and governance (Dick Commandeur, SNV)
SNV in partnership with Crown Agents have recently launched a new initiative focussing on
strengthening the procurement governance of HGSF programmes in 50 districts in Ghana, Kenya and
Mali. The project also has strong focus on learning and dissemination on issues that relate to smallholder farmer participation in the HGSF process.
What is ‘procurement governance’? Concept
brings together public procurement issues
(including: state buying that ensures an open
and transparent process with value for money;
Competing needs for public investment), with
governance dimension (including issues related
to state and civil society; social accountability of
public resources, efficiency, development
dimension, prevention of corruption; multistakeholders processes with multi-interests;
Involvement and joining forces.
There is a tension between procurement
processes’ prioritization of transparency,
accountability and value for money, with the
Figure 5: Procurement governance: Inherent tension?
HGSF objective to prioritize procurement with targeted supplier groups in this case, smallholder
farmers. These competing—though not completely incompatible—priorities must be balanced in
order to generate stakeholder buy-in and the intended financial and social returns.
Why don’t school feeding programmes procure foodstuff of small holder origin? A broad number of
constraints, including, for example, lack of consistency in production (quantity, quality),
procurement requirements (see Table 5).
Table 5: Challenges and opportunities for small holder farmers accessing HGSF market
CHALLENGE FOR SMALLHOLDER
FARMER ACCESS
No information on procurement
notifications
Informal status and consequently
non-eligibility
Lack of experience with bidding
proposals and competitive
processes
Insufficient quantity, quality,
continuity and processing capacity
to meet the requirements
Gender and cultural gap with public
procurement officers
Higher transactions costs
Lack of liquidity to pre-finance
delivery
Lack of bank guarantees
POSSIBLE MEANS TO
ADDRESS CHALLENGES FARMER SIDE
Access to ICT/mobile devices,
focus on procurement
opportunities
Formalization according to
requirements public tenders
Partnership with private
company, invest in good
proposal writing, link with
technical assistance
Collect products of organized
smallholders, connect with
other smallholder
organizations, invest in quality
and processing, partnerships
with private companies
Self-esteem, gaining
experience, involve
professionals, including
women
Efficient management for
activities at organization level,
scaling up of collection
Link with credit institutes, sales
contract as collateral,
arrangements with members
for delayed payments
Link with financial institutes,
formalize assets as collateral
16
POSSIBLE MEANTS TO ADDRESS
CHALLENGES - SCHOOL FEEDING
PROGRAM SIDE
Information campaigns (radio, specific
targeting)
Simplify registration requirements and
process
Simplify proposal requirements, training,
easy templates, limited waivers for
smallholders and alliances between
smallholders and private sector
Reduce lot size, adapt to period of
availability, menu based on local
products, adjust standards for packaging,
use of cereal banks, promote
partnerships and association
Change of attitude, creativity, leadership
for linking with smallholders
Incentive mechanisms, preferences,
facilitate transport and storage
Timely and frequent payments to the
suppliers, flexibility for including credit
institutes in contract arrangements
Reduce requirement for financial
guarantees (e.g. along with reduced lot
sizes), guarantee from local government
M&E and small-holder farmer income: the P4P experience (Alessia De Caterina, WFP-P4P)
Purchase for Progress (P4P) is a 5 year pilot programme aimed at using WFP’s purchasing power to offer
smallholder farmers opportunities to access agricultural markets, to become competitive players in those
markets and thus to improve their lives. Currently implemented in 21 countries and targeting 500,000
smallholder farmers. Food procurement is coupled with over US$ 140million for technical assistance. One of
the 4 P4P objectives targets small-holders income explicitly. The fundamental components of P4P include:
1. Innovative procurement modalities to ensure farmers find a fair market for their produce.
2. Strong supply-side partnerships to strengthen productivity farmer participation in markets. Working
with supply-side partners such as FAO, NGOs and national governments who specialize in enhancing
agricultural production.
3. P4P is a learning programme and has a rigorous monitoring and evaluation system designed to
effectively assess the impact of P4P on livelihoods and markets.
P4P procurement modalities include:
1. Adjusted competitive tendering practices for farmers’ organisations (FOs) or small traders (e.g.
reducing tender sizes, waiving bag markings or performance bonds, purchasing from warehouses).
2. Direct purchasing (from FOs) WFP can engage in negotiation and contracting directly with FOs by
waiving competition (less than three suppliers are invited to make offers).
3. Forward contracting to reduce risk and provide farmers with greater market certainties and
developing partnerships with micro credit and insurance schemes.
4. Processing options – WFP is working with the private sector and other stakeholders to encourage the
establishment of processing units that ensure added value.
5. Engaging with nascent marketing platforms like commodity exchanges and warehouse receipt
systems and linking these to small-holder production.
A number of assumptions in the theory
Profitable access to markets
Sustainable and
Smallholder
increased
profitable
Productivity
of change for P4P highlight complexity
f(organisation, markets,
engagement in
Increased
enabling environment)
markets
of the issues. For example, the impact
theory
assumes
that
increased
Increased Income = f{increased productivity; enhanced group marketing capacity;
productivity is necessary for farmers to
market development; enabling environment}
engage in markets in a sustainable and
Figure 6: P4P impact theory
profitable way. Pure subsistence
farmers are not really targets, P4P targets instead those who can develop a surplus. Farmers will fare better
when marketing as a group; WFP demand needs to be predictable and stable; the market for quality will exist
beyond the WFP market; FOs are plentiful in areas where WFP needs to deliver food assistance (see slides for
full list of assumptions).
The P4P monitoring and evaluation strategy includes a broad set of analysis and learning components,
including impact assessments comparing the P4P farmers groups (treatment vs. control) with in-depth surveys
in a subset (3-7) of countries, with sample size to be about 400 households; Monitoring surveys in years 3 and
5 in all countries; developing farmer organisation records, complemented with other quantitative and
qualitative studies. Measuring impact on household income though has been challenging for a number of
reasons, including the difficulty in establishing the counterfactual and maintaining the integrity of
counterfactual over time, as well as difficulties in sourcing data on market prices. Other challenges involve
difficulties with recall e.g. farmers generally don’t keep records of their production costs. As a result use a
number of proxies e.g. changes in consumption, production, marketing capacity, amongst others. M&E
activities have been very resource intensive for WFP country offices, as a result, a data analysis hub has been
developed to provide quality assurance in sampling, data analysis and consolidation, training and supervision.
Some lessons to draw on M&E front include:
 Use mixed methods
 Understanding the context within which public procurement operates as well as how traders
operate/where they purchase/from whom/why, including them from the beginning
 Importance of having reliable market price information in areas of intervention
 Don’t assume only members of targeted FO will contribute to contracts
 Importance of keeping records – invest in records keeping to complement surveys
 Be aware of high risk of control group erosion
 Take into account real capacity of those who will need to implement the M&E system
17
Governance & accountability: School feeding example in Mali (Emily Janoch, Harvard University)
Community engagement is important for a number of reasons, especially because it can lead to
better local ownership, uptake, and improved programme outcomes. Additionally, any organization
operating with a rights-based framework acknowledges the community’s right to have some over
the kinds of programming that another actor implements and the way in which an implementer
raises money on its behalf. Community engagement (or downward accountability) is critical to
achieving appropriate, effective, and sustainable programme results. This case study addresses the
question of how implementers of HGSF systems can create and operationalise feedback systems
between communities, governments, and external partners that ensure the HGSF programmes are
meeting communities’ needs.
Mali’s school feeding programme has showed some success in fostering community engagement.
Three national-level partners—WFP, CRS, and the Government of Mali—work with communities to
implement school feeding programmes in over 1500 schools. Through the School Management
Committees (CGS), community participation is fairly robust in day-to-day programme management.
Communities demonstrate serious commitment to the canteen by contributing of vegetables, food,
water, fuel, and labour. Communities even pitch in to cover the gap when national level partners are
late delivering food, or deliver too little, an occurrence that happens a few times a year in most
places. Strong ties between local and district levels make communities feel comfortable going to
their school director or district canteen technician with any problems.
Focus areas
Procurement
Communications
and
Monitoring and Evaluation
Policy frameworks
Challenges
Centralized
Fixed enrolment projections
Bottleneck at district
Recommendations
Locally-controlled procurement
Flexible projections
Creative feedback mechanisms (radio, cell phones)
One-way
3 separate systems
Two-way
Harmonize selection and implementation systems
No Ministry of Agriculture
Establish formal links to agriculture
Despite its successes, efforts to integrate community participation into the programming have had
mixed results, depending on the program phase. Defining community needs and designing
interventions happen largely at the national level, which makes it challenging for communities to
participate at these stages. Each of the three main partners acts largely independently from the
others, with its own needs definition and programme design process, and few of the communities
are involved in this process in a meaningful way. Communication is another challenge. If
communities raise a complaint, they rarely hear back from anyone higher than the district level.
There is little programme flexibility, and implementers may address community concerns by saying
that planning is done on a three- to five-year basis, and that any other needs cannot be met until the
new planning cycle. The monitoring and evaluation phase shows similar top-down decision making.
Communities are active in the evaluation stage, but have no mechanism for receiving information or
feedback from the national partners.
18
HGSF impact evaluation in Mali (Edoardo Masset -IDS, Aulo Gelli -PCD)
Providing food through schools has well documented effects in terms of the education, health and
nutrition of school children. However, there is limited evidence in terms of the benefits of providing
a reliable market for small-holder farmers through “home-grown” school feeding approaches. This
study aims to evaluate the impact of school feeding programmes sourced from small-holder farmers
on small-holder food security, as well as on school children’s education, health and nutrition in Mali.
In addition, this study will examine the links between social accountability and programme
performance.
A randomised field experiment around the scale-up of the national school feeding programme,
involving 100 primary schools in 60 communities in food insecure areas of Mali. The randomly
assigned interventions are 1) school feeding programme group, including schools and villages where
the standard Government programme is implemented; 2) “home-grown” school feeding and social
accountability group, including schools and villages where the programme is implemented in
addition to training of community based organisations and local government; and 3) control group,
including schools and household from villages were the intervention will be delayed by at least two
years, preferably without informing schools and households. Primary outcomes include small-holder
farmer income, school participation and learning, and community involvement in the programme.
Other outcomes include nutritional status and diet-diversity. The evaluation will follow a mixed
method approach, including household, school and village level surveys as well as focus group
discussions with small-holder farmers, traders, school children, parents and community members.
Ministry of
Finance
Ministry of
Finance
Enrolment/food requirements
Enrolment/food requirements
Funds
Mayors
Credit advance
Enrolment/food requirements
Food requirements
CAP
Mayors
Enrolment/food requirements
Enrolment
Monitoring
Training
Training
Funds
Enrolment
Credit advance
Monitoring
Feedback
Traders
Food
Market
CGS (schools)
Food
Traders
Food
Food
Small holder
farmers
Market
CAP
CGS
Food
Training
Food
Small holder
farmers
Training
Figure 7: Stylised food procurement process in the national and HGSF+ programme
The impact evaluation will be incorporated within the national M&E system strengthening activities
that are currently underway in Mali. Baselines surveys are planned for November/December 2011.
Monthly process monitoring visits, spot checks and quarterly reporting will be undertaken as part of
the regular programme monitoring activities. Evaluation surveys are planned for 2013.
19
Group work: Procurement and social accountability
In this session, the three groups focussed on some of the potential effects and trade-offs associated
with different school feeding models in terms of procurement and social accountability.
Questions that guided this scoping session included:
 What do you believe the objectives of the procurement system should be?
 At what scale should the procurement occur? Where? And by whom?
 Based on these objectives, what frequency of tendering and/or purchasing would be
optimal?
 Based on these objectives, what would the optimal size of the lots purchased be?
 What kind of competition, tendering and pricing mechanisms might be encouraged?
 What restrictions on potential suppliers might be introduced?
 What food management and quality control issues need to be considered?
 What governance, oversight & financial control systems might be introduced?
Key messages
The group discussions identified a number of issues cross-cutting the three scenarios. The main
objective of the food procurement activities hinges on the timely and stable supply of quality food
for the school feeding programme. The specific objective of supporting small holder farmers to
respond to the demand from the school feeding market is framed under the high level objective of
safe and stable supply. The groups identified the opportunity to explore the trade-offs of different
pro-small holder procurement models as much of the data on the costs and benefits, including
issues around food types (both staples and perishables), prices, timeliness, seasonality and scale, still
need to be collected and analysed. A number of capacity strengthening activities would be provided
to increase the feasibility of pro-small holder procurement, including supply side interventions for
farmer organisations and quality assurance, aimed at improving the cost-efficiency of the
procurement process. Providing mechanisms for improved information flow and coordination across
the sectors, including mapping of supply and demand actors, was also identified as a key programme
activity.
The discussions in the three break-out groups on social accountability converged to a consensus
around three interlinked objectives, also supporting the main procurement objective of safe and
uninterrupted food service delivery. These objectives included increasing:
1. Community ownership and participation in the management of the programme
2. Transparency of financial flows at all levels
3. Sustainability of the programme
Timely, uninterrupted supply of quality food with links to small holder farmers
Steady supply:
Develop resilient supply
chain system
-Procurement system
allows explicit management
of seasonality, scale and
geographical context
e.g. Guidelines for different
purchasing models and
different commodities
(including perishables)
Quality assurance:
Process standards applied
across supply chains
-Quality standards are
developed
e.g. Nutrition and food safety
standards developed
-Quality standards are
monitored
e.g. testing for aflatoxin,
menu composition is
monitored
Small-holder access:
Small holder farmers
participate in supply chain
-Improved production capacity
e.g. Farmer field schools,
inputs…
-Improved awareness of school
feeding market
e.g. Sensitisation campaigns
Community ownership
Transparency of financial flows
Sustainability
Figure 8: Example of stylised HGSF procurement objectives
20
Table 6: Break-out groups- HGSF procurement: impacts and activities
Fully decentralised
Goals and
objectives
Partial decentralisation
Programme features
and activities
Goals and
objectives
Ensure food is
nutritious and
of high quality
There is an assurance of
the frequency of
procurement
Small holder
farmers are
able to supply
the demand
from school
feeding
program
Frequency can be
flexible
Timely delivery,
safe/high
quality foods,
cost effective.
Procurement
model allows
for stability and
a frequency of
procurement
Procurement
system allows
for direct
purchase from
the farmer in a
sustainable
matter giving
best value to
the children
and the farmer.
Building capacity of the
small holder farmers to
comply with the
procurement system
Support farmers in
forming farmer
associations
Link with partners to
support farmers build
their capacity
Information to farmers
are widely
communicated – farmer
sensitization occurs
Sensitization for
community members
Procurement officers
are sensitized to
become more pro-small
holder actors –
including school
committees
Programme features
and activities
Increase
feasibility of
procurement
from
smallholder
farmers by
mayors (directly
or indirectly)
Strengthen
capacity of
Farmer
Organizations
through
information,
storage
capacity, credit
etc
Increase
procurement of
perishables in
support of
smallholder
farmers and
children’s
nutrition
Target a higher
proportion of
procurement from
FOs (from traders or
directly)
Increase information
for stakeholders in
the supply chain
Increase funding to
schools for
procurement of
perishables
Map supply side
interventions in the
area and take steps to
link to the
programme
Review procurement
procedures and make
more appropriate to
smallholder farmers
and FOs e.g.
securities,
documents,
registrations etc
Review timing of
procurement in light
of storage capacity,
seasonality etc
Integrated farm to school
Programme
Goals and
features and
objectives
activities
Women’s
Sell the produce
groups
for a fixed price.
supplying
Anything in
food to
excess women
schools
are free to sell at
the market.
Quantities
supplied to the
school by
women is
flexible
Capacity building
– community
and women
empowerment.
Agriculture
support.
Transaction
costs are
minimal as most
comes from the
farm gate to the
school
Food is not
stored for
prolonged
periods of time.
Cross cutting research themes
The group discussions identified a large number of research issues on procurement that will be the
focus of a follow-up working paper. However, the discussion focussed on a number of key research
issues, including:
 Acknowledging that there are multiple scenarios for the different procurement models, with
very context specific nuances that are not very well documented or understood. Begin by
mapping different procurement processes in detail.
 Important to separate staples and perishables conceptually, as there are clear
differentiations in the supply chains and associated procurement processes.
 Who are the smallholders, and how do you measure whether small holder farmers are
smallholders? Quality, land size, knowledge, identification of key barriers and key areas of
improvement. Opportunity to build scenarios for improved guidance for policy and
programme development.
 How does the inclusion of the smallholder farmers generate new businesses along the
supply chain?
 Tendering process and what we can learn from P4P and what is unique within government
systems.
 Fortification makes a big difference, very cost-efficient but not universally implemented.
Why is this not occurring?
21
Session 4: Country context, needs and targeting approaches
Analysing the context for school feeding (Marc Regnault de la Mothe WFP, Kristie Neeser PCD)
The System Assessment and Benchmarking for Education Results (SABER) program is a flagship
initiative of the World Bank’s new Education Strategy 2020. SABER enables countries to
systematically examine and better understand their education system’s policies, which in turn helps
them to identify where investment is needed. By leveraging global knowledge, SABER fills a gap in
the availability of policy data and evidence on what policies matter most to improve the quality of
education and achievement of better results.
SABER provides a system to examine education policies, of which school feeding is now included. As
such, a great deal of research is being done to explore how school feeding will be examined within
the SABER program. Based on the findings reported in Rethinking School Feeding, it was identified
that those countries who have successfully transitioned from externally owned to government
owned school feeding programs had key standards:
 Policy
 Financial capacity
 Institutional capacity
 Program design and implementation
 Community participation
These key standards have become the basis of analysis for school feeding within the SABER program.
Each standard is currently being analyzed further to identify clear indicators and develop
benchmarks. This information is gathered through the following organizing principles:
 Reviewing the evidence – what makes a good school feeding program and what metrics can
be used to identify countries current positions.
 Drawing on the theory
 Examining top performers
 Identifying the indicators
 Collecting policy data
Following a peer review by key stakeholders, the school health and school feeding questionnaires
were beta-tested with government and development partners in Kenya in late July. The participants
were asked to: (i) review the questionnaire tool, and (ii) provide overall guidance on how to improve
the instrument. Overall feedback was that such a tool would be very useful in assessing country
status in the areas of school health and school feeding and in identifying areas of focus. Further
pilot exercises are being undertaken with CARICOM and East & Southern African countries later this
year.
22
Reflections on needs and targeting approaches for school feeding (Harold Alderman, World Bank)
Upon initial reflection it may seem that targeting is a non-issue with regards to HGSF, after all, the
goal is to feed school children, is this not targeted enough? However, upon deeper analysis, there
are a few themes which arise with regards to targeting, both among beneficiaries and also among
suppliers. Targeting among both groups can and should be considered, particularly with regards to
geographic and within school targeting. In all cases, it is key that the objective of the program is
thoroughly understood. From a food security and social protection point of view, targeting will likely
be critical – however from an education point of view, this may be less of a priority. Therefore fully
understanding the objective, as well as the local context, is needed prior to any targeting exercise.
Geographic targeting
The question which should be asked here is how do you reach the food deficit and poorest areas? In
reality, if you are targeting the poorest areas, a lack of infrastructure may result in an unsuccessful
program. Therefore more investment may be needed in these areas and in this case, not every
region should be getting the same opportunity. In fact, the easiest areas to target will be those that
have grain surplus, however these may not be the areas which need the support most. Therefore
when targeting schools based on geographic location, multiple factors must be considered including
the communities ability to implement the program, the need of the community, available surplus
and the overarching question – will you prioritise some areas over others? It must also be
considered that geographic targeting has the risk that it can be interpreted as being based on
ethnicity. Geographic targeting is also seen with regards to the suppliers, where farmers in surplus
areas are targeted to supply deficit areas. There is an opportunity for researching this issue in these
novel programmes as they are being scaled-up.
Targeting within schools
There are a few examples of how school feeding programs are targeting children within schools
around the world. In many cases, the most in need children have their costs subsidized. To ensure
anonymity, coupons are used by all children, reducing the visibility of children who are receiving
support. However, this approach is not always feasible in low income countries. If targeting within
schools is a necessity, then perhaps the best way to do this is by grade. Age targeting can prioritize
either young children who are particularly sensitive to hunger in the classroom and also attempt to
support adolescent growth spurts and the increased requirements for iron for young women. This
approach, however, is not without flaws as the age of the children in the school grades often varies
substantially.
Seasonal targeting and community contributions
HGSF programs run differently in times of surplus than deficit, therefore the entire scale of a
program may need to be adjusted based on the food available. This may influence the targeting
process, forcing stricter targeting to occur during low yield periods throughout the year, or from one
year to the next.
Those who pay for the school feeding program are also targeted. This type of targeting can be seen
as targeted tax as particular groups within a community are responsible to contribute for a local
service. For example, where women’s groups pay for the school feeding program, they are in
essence being taxed above the rest of the community.
Take home rations
Take home rations have been used in school feeding programs as a method of targeting girls
primarily, but could also be used to target those in poverty as has been done in Bangladesh. Take
home rations often include a condition on the number of days attended but in practice they are
targeted more on enrolment than attendance.
23
Group work: Analysing needs and targeting approaches for school feeding
In this session the break-out groups explored some of the challenges and opportunities for national
governments developing their targeting approaches, including:
 What trade-offs between universal, geographic and individual targeting should be
considered?
 What other factors should be examined in terms of making links between school feeding and
agriculture (e.g. traditionally the smallholder farmer end has not been included in the
targeting equation at all)?
 If for example we target schools geographically based on food insecurity data, are we closing
the door in terms of linking school feeding with community level food production?
 If individual targeting is appropriate in middle (and high) income countries, is this entirely
unfeasible in low income settings?
 How will the targeting affect the programme design?
Key messages
The group work recognised that key to managing the possible tension across agriculture, health and
nutrition, and education involved being explicit about the different target groups and targeting
criteria. On the agriculture side, there was a consensus in terms of focussing on areas with good
agricultural potential and on working with farmers to increase their marketable surplus. Not
surprisingly, there was also a consensus in terms of targeting areas with the highest potential impact
on education, meaning areas with lowest school participation. Therefore mapping overlaps between
areas of good agricultural potential and low school participation would provide a first pass analysis
in terms of potential target areas for HGSF. A number of other challenges were also raised in terms
of reaching the different target groups, including managing the possible migration between schools
with and without school feeding, as well as managing the seasonality of production when assessing
the ability of the different communities to contribute to the programme.
Table 7: Break-out groups: Targeting approaches for agriculture
Targeting criteria
Based on production potential:
Smallholder farmers who have
the capacity to produce surplus
To give most reliable feeding
source, target food surplus
areas (or at least potential to
produce surplus), not fooddeficit areas; areas with least
potential for pipeline breaks.
There is some land available to
be transferred
Political will
Require a canteen and water
source
Based on vulnerability:
Women farmers/existing
women’s groups in the area
Poor communities
Challenges
Is the program reaching the intended
targeted group?
Ownership challenges for
women, socio-cultural issues
How do you know you are
targeting women if the men
own the land?
Is the program targeting the correct
target group?
Most in need cannot produce
surplus and therefore cannot
access market
As communities must commit
before any resources come to
them – is the program
targeting the poorest
communities?
Measuring the surplus?
Storage? Whose responsibility is this?
Security will be a major challenge
Quality of the land may be poor
Limited resources and extension
services
Opportunities
Farmer Groups
Using farmer networks
as an entry point to
empower farmers
(other partners can
support)
Allows for capacity
building of smallholder
groups and women in
particular
Farmer Associations
Farmers with opportunity to
produce surplus due to
underutilized land
For entrepreneurship
development along the supply
chain.
The main research themes on targeting across the three scenarios are captured in Annex 4.
24
Session 5: Focus on school feeding programme design, agriculture and nutrition links
Ration design using “home-grown” foods (Leo Nederveen, PCD)
Based on UNICEF’s conceptual framework, there are 3 levels of causes of malnutrition immediate
(inadequate dietary intake or disease), underlying causes influencing households and communities
(food insecurity, inadequate care, lack of health services, income poverty) and the basic causes
around the structure and processes of society (lack of capital, social, economic and political context).
All of this must be considered when planning a school feeding program to ensure the nutritional
needs of the children, and the broader community, are met.
Phase I: Develop criteria, propose approach, and assess dietary intakes of schoolchildren
Consider public
comment
Phase II: Use criteria to define nutrient
targets and ration requirements
Select age-grade groups
Define nutrient targets
Consider costs
Identify crops
Consider student
acceptance
Recommend ration
options and
requirements
Consider practicality
(transport, storage)
Consider seasonality,
availability
Consider food
preparation
limitations
(facilities, fuel,
water)
Adapted from School Meals: Building Blocks for Healthy Children, NAS, 2010
Figure 9: Proposed pathway for planning school meals (Source: Galloway, 2010).
However, as part of the entire planning process, it is critical that the goal of the school meal ration is
determined. This will include educational and agricultural goals. In addition, the nutritional goals are
important and should consider:
 Provide nutritious food for growth, physical activity and to maintain adequate health
 Reduce short term hunger
 Reduce malnutrition
 Reduce micronutrient deficiencies (for example anaemia)
 Contribute to household food security through value transfer
 Improve food habits of children and households
Furthermore, the nutrient targets required to reach these goals, as well as the nutritional content of
“home-grown” crops, need to be identified. A valuable, easy to use prototype ration design tool is
available on www.hgsf-global.com which can assist in identifying the how a particular type of food
can assist in attaining the desired nutrient target. It is important to remember that providing
nutritional food to children will not on its own reach the desired goals. School feeding programs
need to be linked with school health and nutrition programs more generally, including hand washing
and deworming. The programs need to run in parallel with sensitization training for the community
and school to ensure that parents understand the food provided is not a substitute for what the
children receive at home. Children should also receive nutrition focused education so that they can
bring this information back to the household. Finally, in many cases, fortification may be needed in
the form of iodized salt or micronutrient powder. All of these factors need to be considered prior to
implementation for a school feeding program to adequately meet the nutritional needs of school
children.
25
School feeding and household level nutrition (Shalini Roy, IFPRI)
One of the challenges with HGSF is linking the various goals of the programme. This includes
improving educational outcomes, bolstering local smallholder farmers and promoting health and
nutrition. The challenge is that the links between these goals are not automatic and the target
groups are not the same. In particular, there is widespread concern about HGSF as a nutritional
intervention. School feeding programs intervene after the critical period (prenatal and ages 0-2)
when crucial mental and physical growth occurs. Malnutrition and micronutrient deficiency at this
point of life can cause permanent damage. Moreover the goal of improving children should not be
limited to school age children but to the entire household, including the younger and more
vulnerable infants and children who soon will be school aged.
There are, however theoretical potential channels for nutritional impacts of school feeding on school
age children. There is the potential that school feeding programs could meet the high energy
requirements during ‘growth spurts’, such as during adolescents when children can gain 15% of
adult height and 50% of adult weight. This is particularly true for adolescent girls who have high iron
needs during puberty to prevent anaemia. Additionally, there is the potential for school feeding to
have an impact on primary school age children. Reduction of short-term hunger and micronutrient
deficiencies may improve cognitive function and resistance to intestinal and respiratory infections.
However for school age children, existing evidence shows very small nutritional impacts, with
negligible impacts on height-for-age, small impacts on cognitive function (requires micronutrients),
small impacts on weight-for-height, and small impacts on anaemia status of adolescent girls
(requires micronutrients).
There is the potential for an impact on the household level through spill-overs from school feeding
to younger siblings of school children. This could be achieved through intra-household reallocation
of more food or more nutritious food to pre-school age children. However this is not the best way to
reach the nutritional needs of the household as it is very indirect, there are limited gains and it is not
self-sustaining.
A more direct approach is needed to target the critical groups of prenatal to age 2 for nutrition
interventions through complementary activities. This could include additional behaviour change and
communication activities targeting the mothers of school children (who may be pregnant or have
small children); coordination with traditional school feeing facilitates access to this group; more
direct targeting of this group where there is the potential for the largest impact.
There is also the possibility of targeting early childhood development centres for pre-school age
children. This approach has the potential to have larger nutritional gains and would increase
enrolment, which is highly promising for educational gains.
Whether feeding school children only or also feeding pregnant mothers, children ages 0-2 and/or
children ages 3-5 in early child development centres, there are some constants which should be
considered:



Food basket should be fortified if not naturally micronutrient-rich
Use of sprinkles even if not possible to locally fortify
Iron is particularly important (crucial for cognitive function)
26
Group work 4: Designing school feeding programmes: agriculture and nutrition links
In this session, the three groups explored the agriculture and nutrition linkages related to the design
of different types of HGSF programmes. Unlike more “traditional” models of school feeding in lowincome settings that focus on benefits to school children, the beneficiaries of HGSF can also include
smallholder farmers, food processors and cooks, for example.
By incorporating community level nutrition education in addition to food procurement, HGSF
interventions can potentially deliver both immediate benefits (in terms of increasing the household
income of smallholders through increased demand for their food production), and long-term, lasting,
intergenerational, benefits (by supporting household level nutrition).
Additionally, in designing the food ration, planning with farmers is needed to assess what is currently
being grown and the types of foods that farmers will need to produce in order to meet the future
educational and nutritional goals of the HGSF programmes.
Designing the food ration- challenges and opportunities in the supply chain
Some of the questions considered in this session included:
 How might the type of HGSF programme that your group is looking at provide opportunities
in designing the food ration to achieve nutrition and education outcomes for schoolchildren?
 How might the type of HGSF programme that your group is looking at constrain the design of
the food ration in achieving nutrition and education outcomes for schoolchildren?
 What more do we need to know to answer the above? How could we go about answering
those outstanding research questions?
Key messages
The break-out group discussions identified a number of cross-cutting issues including, for example,
the opportunity to develop an improved understanding of using traditional foods as part of the HGSF
programme. This is particularly relevant for fruits and vegetables that could provide micronutrients
in forms that are more readily acceptable to the target communities. Identifying opportunities for
processing at the village level, including drying, milling (and fortifying) could also provide
opportunities for business development. However, using perishables and village level processing is
challenging in terms of quality assurance, including storage (e.g. food safety) and ensuring
nutritionally balanced meals.
Table 8: Break-out groups: challenges and opportunities for HGSF ration design.
Fully decentralised
Opportunities
Constraints
Local research
Fortification
organisations have may not be
information on the possible given
nutritional value
cultural cooking
of local foods
practices
Fresh vegetables
and fruit
Outsourcing the
services of cooking
Teachers are able
to ensure
fortification occurs
Teachers are
taken away
from the class
for food
procurement,
therefore the
education could
suffer as a
result
Purchasers are
not professional
nutritionists.
Partial decentralisation
Opportunities
Constraints
Use existing and
Perishables
traditional foods
have different
in region
characteristics
and need
Include
different
perishables in the
supply chains
food basket and
(e.g. more
have separate
frequent
budget for
procurement
perishables for
or donations)
school feeding
committee /
Issues with
parents
seasonality and
scale may be
Drying of fresh
too low for
products may be
some solutions
possible
Use local
structures and
enterprises
27
Integrated farm to school
Opportunities Constraints
Locals are
There is no
aware of what check to
is available
ensure food
and what the
has been
preferences
appropriately
are
stored
Food is fresh
as it is grown
in the
community
and not
transported
It is difficult to
keep the
nutritional
level constant
with perishable
foods.
Implementers
are able to
control the
value of the
seed
There is no
standardization
of menu across
times of the
year.
Community level nutrition education
Some of the questions considered in this session included:
 How might the type of HGSF programme that your group is looking at provide opportunities
for incorporating community level nutrition education?
 How might the type of HGSF programme that your group is looking at constrain
opportunities for incorporating community level nutrition education?
 What more do we need to know to answer the above? How could we go about answering
those outstanding research questions?
The group discussions converged on three main opportunities:
1. Schools can provide an entry point for the education and sensitization for children, parents
and greater community on health, nutrition and agriculture
2. School gardens can be used to promote improved agriculture and nutrition practices at
school and with the community
3. School feeding can also provide the entry point for other school based health and nutrition
interventions like deworming
A number of concerns were raised in the group discussions with regards to the capacity of the
extension staff that would be undertaking the training and sensitisation, highlighting that this is a
great deal is expected from single individuals. The capacity constraints also involved the training
resources, in terms of what materials are available, the quality of these materials and whether they
would require extensive efforts to make them locally appropriate.
Cross cutting research themes
A number of research issues that were raised in the discussions, including:
 Do we dully understand the nutritional value or indigenous foods as well as the traditional
customs around these foods?
 Are we expecting too much from the teachers? How is their role in the school feeding
program affecting the educational activities?
 How sustainable is this approach and will we be seeing the same impacts years from now?
 Is the education having an impact on behaviour and what education is being provided in the
school?
 What are the successful models for nutrition education used elsewhere?
28
Analysing the HGSF supply chain
(Afua Kufuor GIMPA, George Scharffenberger UC Berkeley and Daniel Mumuni PCD)
Setting the objectives of a HGSF programme is a critical step in the design process. The effectiveness
of the programme, however, will hinge on the efficiency of the supply chain linking small holder
farmers to school children. An innovative program has been launched in partnership with GIMPA, UC
Berkley and PCD to engage business schools to analyse the supply chain for HGSF, with a particular
focus on the Ghana School Feeding Program (GSFP). A joint GIMPA/Berkeley research team
examined the GSFP operations in 8 out of 10 regions (and 30 out of 170 districts) during the course
of 2010. The scope of the work included exploring how caterers currently interact with farmers, and
identify ways in which the supply chain could be made more efficient and small-holder inclusive. The
team also developed a costing tool to guide GSFP operations. The project was also aimed at
developing a sustainable model linking school feeding with small-holder agricultural production in
Ghana. The research team developed a detailed understanding of the different value chains, also
highlighting a great deal of hidden costs in the program which are not normally identified by
operations and are often bourn by the communities. The financial flows were also examined as well
as opportunities to improve the overall accountability of the programme. The research was
embedded in the PCD supported technical assistance planning process, ensuring strong buy-in from
Government and in-country partners.
Programmatically, the research highlighted that there is a missing link between demand and supply.
Caterers are contracted at the district level to supply several schools and they are responsible for
doing the procurement. The farmers interviewed in this work were in general not aware of the GSFP.
Some of the key challenges in terms of small-holder farmer engagement in the GSFP included:
 Lack of credit and cash flows constraints
 Lack of trust between farmers and caterers
 High transaction costs: No structure to facilitate caterers and farmers negotiations making it
difficult for caterers to link with farmers
The research team also found that there is very little understanding of the financial flows. To
examine cash flow issues from the planner and caterer perspective, the research team has also
developed a business model and associated tool costing. It allows users to size the demand and to
examine the cost structures of the different menus available for programme delivery. This tool also
enables the Government to monitor the cost-efficiency of the programme, and the trade-offs
associated with locally sourced foods.
The research also identified a clear need for a more open forum where caterers and farmers can
come together and discuss procurement openly. Building on the findings of this work, the GSFP,
through ECOSARD and SNV have launched a pilot project that hinges on developing a formalized
agreement between the caterers and the farmers, formalising the links between supply and
demand. There are, however still challenges as the caterers and the farmers are not properly
organised and caterers are generally not literate and therefore struggle with signing contracts. There
is also the issue of payment. The farmer wants immediate cash, but the caterers are not able to
provide timely adequate payment. The pilot is trying to bring in local banks within the community to
see how they can support the caterers with low interest loans. There is progress as in one district
where a contractual agreement has been developed. A contract is written which means that the
caterer is bound to purchase from the farmer and the farmer is set to sell the food to the caterer at a
specific cost. This means that they trust each other. This process is currently monitored by SNV,
however it is intended that GSFP will take this over. It is being piloted in 3 districts.
Looking at school feeding from the perspective of a business school is not common; however it
provides an opportunity to examine the programme from various angles and identify viable
solutions. Business schools also benefit from involving a new generation of students in dealing with
complex real life programmes, combining “doing good and doing well”.
29
Next steps: building the research agenda
The technical meeting provided a multi-disciplinary platform for the development of a coherent and
harmonised research agenda for school feeding, providing key pointers in terms of identifying the
gaps in the evidence for HGSF. The discussions provided an enormous wealth of content that will be
mainstreamed within the on-going development of the research agenda of the HGSF programme.
Some of the key themes highlighting the complex nature of HGSF that were raised in the discussions
included:
1. HGSF programmes combine programmatic activities with potential education, health and
nutrition, agriculture and social protection objectives. In order to manage the trade-offs
explicitly across these objectives, it is critical that the objectives are clearly defined
(including differentiation between primary and secondary objectives), alongside clear
articulation of the programme theory linking the different target groups, the programme
activities and the impact pathways.
2. Clarifying the objectives is also critical in terms of defining HGSF within the context of an
impact evaluation, and in particular, identifying what the control group will be. Are we
comparing a local versus national/central procurement model? Does it involve nutrition,
education, and agricultural comparisons? How would small holder farmers be sampled?
3. Whichever food procurement approach is chosen, its primary objective must be the timely
and stable supply of quality food for school feeding. Support to small holder farmers to
respond to the demand from the school feeding market should be framed under the
objective of safe and stable supply. Research is needed to explore the trade-offs of different
pro-small holder procurement models, analysing the data on the costs and impacts,
including issues around market integration, scale, timeliness, prices, food types (including
perishables), and seasonality.
4. Following form the preceding points, as HGSF has potential simultaneous impacts, how do
we aggregate these effects in a single metric? This is particularly important when comparing
the cost-effectiveness of HGSF to other interventions.
5. Issue of decentralisation highlights inherent tension between procurement processes’
prioritization of transparency, accountability and value for money, with the HGSF objective
to prioritize procurement from smallholder farmers. These competing, though not
completely incompatible priorities must be balanced in order to generate stakeholder buy-in
and the intended financial and social returns.
6. More detailed descriptions are needed on the various models of HGSF, including information
on how programmes began and then evolved over time, capturing good practices and
lessons learned. Information on how high and middle income countries are implementing
and financing school feeding programs was also raised as an important area of research.
7. There is an opportunity to document and analyse the process, or transition, from externally
driven programmes to country owned, sustainable programmes, identifying where the
transition works, and where it doesn't. Are there particular characteristics of country
governments which make this transition easier?
8. Clarifying the nature of the programme activities will also enable more accurate estimation
of the full implementation costs of HGSF, including different activities and actors across the
supply and value chains. It is particularly important to capture all the “hidden” costs,
including community level contributions that are often substantial and have been until
recently overlooked.
9. Developing a standard costing and impact framework (including standardised indicators) for
the different implementation models will enable improved comparability of the cost and
cost-effectiveness estimations.
10. A particular challenge in terms of aggregating the impacts of HGSF hinges on the issue that
the value of transfers does not easily aggregate with the other effects. In particular, the
30
aggregation will require quantifying the weight society puts on consumption of the poor
relative to that of the average citizen, which is not straightforward.
The technical meeting provided a multi-disciplinary platform for the development of a harmonised
research agenda for school feeding, providing key pointers in terms of identifying the gaps in the
evidence for HGSF. PCD and partners now have a solid foundation on which to build on-going and
future research activities, including in particular i) a case study source book; ii) a transition analysis;
iii)a collection of HGSF innovations; iv) impact evaluations; and v) supply chain analyses. These five
products will be developed under the World Bank, WFP, PCD global school feeding partnership.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
Presentations-References
Lillywhite, J. Introduction to the technical meeting.
Gelli, A. Exploring different implementation models for school feeding.
Burbano, C. and Sidaner, E. Towards national school feeding programs: insights from the field.
Alderman, H. and Bundy D. School feeding programmes and development: Are we framing the
question correctly? World Bank, 2011.
Chege, P., Guleid, N. and Rotich, L. Kenya school meals programme.
Kristjansson, B. Report to the technical meeting. Results of the systematic review of
effectiveness of school meals and cost-outcomes of school meals.
Sonnino, R. School feeding, public procurement and sustainability.
Burrows, S. Operational Perspectives: Evaluations of School Feeding 2009-2011.
Haddad, L. How “home grown” are Home Grown School Feeding Programmes?
Commadeur, D. Procurement Governance in Homegrown School Feeding
De Caterina, A. Purchase for Progress (P4P) uses WFP's purchase power to connect farmers with
markets
Johnson, C and Janoch, E. Evaluating Social Accountability in School Feeding Programmes
Masset E., and Gelli, A. Community participation and the links between agriculture, nutrition
and education: Design of a randomised field experiment of “home-grown” school feeding in
Mali
Neeser, N. and Regnault de la Mothe, M. SABER.
Nederveen, L. HGSF ration design tool.
Roy, S. Home-Grown School Feeding: Potential Impacts on Household-Level Nutrition
GIMPA/Haas. Ghana School Feeding Programme.
31
Annex 1: School feeding implementation scenarios
Scenario 1: Fully decentralised model (see Figure 9). In this scenario, the food procurement process is
undertaken at school level and co-ordinated by members of a School Meals Committee (SMC), which includes
teachers and community members. The food procurement process replicates the process used in the
procurement of school instructional materials in the country. For food purchases, cash is transferred directly
from the Ministry of Education to school accounts using the existing school-based management systems on a
bi-annual basis. Schools are informed of how much money has been credited into their accounts. Procurement
is co-ordinated by the SMPCs through a competitive process with registered/licensed local farmers or
suppliers. Preference is given to suppliers from “vulnerable groups” including; widows, HIV groups and
women-led groups though it is not clear how this preference is enforced. Procurement is guided by circulars
from the Ministry of Finance. The frequency of procurement by schools is influenced by their storage facilities
and availability of food. The construction of storage facilities is considered the responsibility of the parents.
Organization
of farmers
Production
of food
Wholesale,
Trading
Transportation
& Storage
Processing &
distribution
to schools
Food
Preparation
Distribution
To Children
Schools responsible for procurement and preparation using funds
from the central level
•Funds for food
procurement
•Training
•M&E
•Advocacy
•Fundraising
•M&E
•Training
•M&E
•Food procurement,
preparation and
distribution
SMC
District Coordination
teams
MoE
(including SF technical unit)
Supporting ministries
(MoH, MoA)
Technical SHN&M committee
MoE, MoA, Water and Irrigation, Public Health and
other stakeholder
Figure 10: Stylised fully decentralised scenario
Programme activities along the supply chain
Organisation of farmers
 No focus is given to how farmers will be organised.
Production Capacity
 No focus is given to how farmers’ production capacity will be increased.
Wholesale and trading
 Commodity prices used for budgeting purposes are set by an average over selected regions and
integrated with figures provided by the National Cereals and Produce Board (NCBP). Average per child
budget set by the MoE is US$0.09 per day.
 Cash released from the MoE to schools biannually: procurement is expected to be undertaken when
prices are at their lowest.
 Suppliers must have: trading license, bank account, registration certificate, 3yrs of trading experience
and permanent premises.
 Procurement follows simple tendering system: publishing the commodity description and constraints
of supply, submission of tenders, subsequent purchase and payment via cheque.
 Procurement is mainly maize, common beans, pigeon peas, green grams, soya beans and rice.
 Purchase list is formed by the SMC, whose members visit the local market to collect samples.
 SMC activities are guided by the of Ministry of Education manuals.
 Procurement process replicates the school instructional material procurement process.
 Dedicated school feeding programme account requires 3 signatures for fund management
 Procurement is from the market and occurs immediately after school receives funds.
 Procurement is undertaken at school level by the SMC.
32
 Determination of purchase prices by the SMC is via government guideline, surveys of trader prices,
gross margins for produce as set by Agro experts.
Transport and Storage
 Successful suppliers are responsible for delivery of commodities to the school, no additional costs are
charged for transport and delivery of commodities.
 Storage facilities in schools assumed to be of adequate quality to store a three month supply of food.
Processing and distribution to schools
 Commodities will be received at the school, where inspection of quality is undertaken in the presence
of the supplier. Relevant receipts are also exchanged at this point.
Food Preparation
 Many of the schools were previously supported by internationally administered school feeding
programmes and in theory have storage facilities and kitchens that meet definite standards.
 Cooks are hired or provided on a voluntary basis by the community.
Distribution to children
 Children will receive either mid-day meals or a mid-morning snack of porridge.
Scenario 2: Partial decentralisation (see Figure 10). In this scenario, food provision consists of a school lunch
served at noon. The food consists mainly of staples enriched with vegetables and fruits. Enrolment figures are
collected by the district education officers through the SMCs and passed to mayors with estimates of food
requirements. The SMCs operate at village level and have the role of day to day management of the
programme. They are subdivided into subcommittees and include a stock manager, a representative of the
cooks and a treasurer. Mayors, who receive a budgetary allowance from the Ministry of Finance on the basis of
the food requirements, then issue tenders, on the basis of a credit advance, to certified service providers
(traders) to procure the staples (cereals and legumes). The service providers (traders) then purchase the food
from the market or from smallholders, and deliver it to the relevant schools. Perishables are purchased by the
SMC at village level on a weekly basis or provided as an in-kind donation by the community.
Organization
of farmers
Production
of food
Wholesale,
Trading
Transportation
& Storage
Processing &
distribution
to schools
Food
Preparation
Traders
Distribution
To Children
Schools
•Funds for food
procurement
•M&E
•Training
•M&E
•Perishables
•Food preparation
and distribution
SMC
District Coordination
teams
Mayors
Ministry of
Finance
MoE
(including SF technical unit)
Supporting ministries
(MoH, MoA)
Technical committee
MoE, MoA, Water and Irrigation, Public Health and
other stakeholder
Figure 11: Stylised partially decentralised scenario
Programme activities along the supply chain
Organisation of farmers
 No focus is given to how farmers will be organised.
Production Capacity
 No focus is given to how farmers’ production capacity will be increased.
Wholesale and trading
 Commodity prices used for budgeting purposes are set by an average over selected regions and
integrated with figures provided by the National Cereals and Produce Board (NCBP). Average per child
budget set by the MoE is US$0.09 per day.
33
 Cash release from the MoF to Mayors biannually: procurement is expected to be undertaken when
prices are at their lowest.
 By law suppliers (traders) must be certified service providers.
 Procurement follows a simple tendering system: publishing the commodity description and
constraints of supply, submission of tenders, subsequent purchase and payment via credit advance.
 Procurement is mainly maize, millet and beans.
 Food quantities for tenders are established by Mayors and district education offices based on school
enrolment figures on an annual basis.
 Procurement is from the market.
 Determination of purchase prices is via government guidelines.
Transport and Storage
 Successful suppliers are responsible for delivery of commodities to the school, no additional costs are
charged for transport and delivery of commodities.
 Storage facilities at the school are assumed to be of adequate quality and size to store a three month
supply of food.
Processing and distribution to schools
 Commodities will be received at the school, where inspection of quality is undertaken in the presents
of the supplier.
Food Preparation
 Many of the schools were previously supported by internationally administered school feeding
programmes and in theory have storage facilities and kitchens that meet definite standards.
 Cooks are hired or provided on a voluntary basis by the community.
Distribution to children
 Children receive a mid-day meal.
Scenario 3: Integrated farm to school model (see Figure 11). In this scenario, women’s groups are mobilised
to support the supply of the school feeding programme. The groups are supported by the government
(through national directorate of school feeding programme) over a 5 year cycle with capacity building activities
aimed at increasing agricultural production. In turn the women’s groups progressively supply increasing
quantities of food to school feeding in the village, beginning with 25% of cereal requirements in the first year
and increasing to 100% in year 4. Additional production is sold on the market or to the suppliers of the
national school feeding programme at a fixed price established by a central government board.
Organization
of farmers
Production
of food
Wholesale,
Trading
Transportation
& Storage
Processing &
distribution
to schools
Food
Preparation
Schools
Women’s groups
•Training
•M&E
•Advocacy
•Fundraising
•Funds for food
procurement
Ministry of Finance
•M&E
•Training
Coordination teams at communes
level
Suppliers
Distribution
To Children
MoE
(inc. national directorate
for SF)
Supporting ministries
(MoA, MoH)
Technical working group
MoE, MoA, WFP, UNDP
Figure 12: Stylised integrated farm to school scenario
34
•M&E
•Perishables
•Food preparation
and distribution
SMC/PTA
Programme activities along the supply chain
Organisation of farmers
 A special focus is given to how smallholder farmers will be organised around a school. The national
directorate for school feeding (NDSF) directly works with the women’s groups from sensitization
through the actual legal creation of the group.
Production Capacity
 A special focus is given to how farmers’ production capacity will be increased. The national
directorate works in close collaboration with the National Agency for Rural Development to ensure
the women’s groups’ production capacity is enhanced. In these efforts, the women’s groups benefit
from technical agricultural support and training with the objective of meeting an increasing portion of
the schools food requirements: 25% in year 1, 50% in year 2, 75% in year 3 and 100% in year 4 and
beyond. The women groups will trade their remaining production in open markets or to the NDSF
suppliers. The NDSF has made a commitment to purchase the remaining produce from these groups
at a fixed price so that price fluctuations on the open market do not affect these groups.
Wholesale and trading
 Food quantities are established on an annual basis by the SMC and commune level education offices
based on school enrolment figures. Food commodities are provided directly by the women groups to
the schools, in the communes where these groups are fully functional. The remaining commodity
needs are met by the NDSF.
 Commodity prices are set by the central market board, a national level market regulation entity
following government guidelines.
 By law suppliers (traders) must be certified service providers.
 The NDSF works directly with traders on the procurement exercise, and subsequent purchase and
direct payment is conducted through the MoF.
Transport and Storage
 The women’s groups are responsible for delivery of commodities to the school.
 For the NDSF purchase from suppliers, transport service providers are contracted at the national level
and are responsible for delivery to the schools
 Storage facilities at the school are generally of adequate quality and size to store a three month
supply of food. Storage facilities are also arranged by the women’s groups in their vicinities
surrounding the school (to store their excess production).
Processing and distribution to schools
 Food commodities will be received at the school, where inspection of quality is undertaken.
Food Preparation
 Many of the schools in theory have storage and kitchens that meet definite standards.
 The nutrition unit under the NDSF is responsible for regulating the nutritional value of the meals
provided in schools. The NDSF provides the SMC’s with nutritional guidelines and tools which
emphasize the importance of good nutritional intake, promote the consumption of local foods which
are rich in micronutrients and take into account the cultural dietary practices.
 Cooks are provided on a voluntary and rotational basis by the community.
Distribution to children
 Children receive a mid-day meal.
35
Annex 2: Break-out group session: Research questions on linking agriculture, health, nutrition and education
Fully decentralised
How do you have a
contingency supply which can
be
managed
by
the
community?
How to mitigate risk for small –
borderline farmers in case of
decrease in production?
Practical research to see what
works in a specific context.
Organic HGSF programs?
Partial decentralisation
Information on costs, qualitative work on what
leads traders to purchase from small –holders.
Integrated farm to school
Does the anaemia rate of school
children change?
Mapping of what procurement officials actually
do in terms of buying from local and in-country
farmers. Where is food coming from right now?
What happens to the nutrition
status of children 0-5?
What is productivity right now? What affects it?
What barriers exist to prevent farmers from
working with local traders?
Study on the effects on prices and other poor
households
Is the procurement that is being done making a
difference at all to local economy?
What interventions are needed on the supply
side?
What are complementary activities ongoing right
now?
We need to look at long-term effects and fully
understand the context.
How best to quantify physical activity? Does
school feeding improve physical activity?
What are the nutrient rich local foods?
What are the barriers to using fortified foods?
What are the barriers to hygeine?
Which types of education work best in which
context?
What happens at household level as a result of
food that is eaten at school?
How to change the culture around food in
various communities?
What are the nutritional impacts at various ages
(e.g. is there an increased impact at
adolescence)?
36
Is the labour of women (working the
farms) effect the health of the
younger children?
Does HGSF have any additional
benefits for children compared to
regular school feeding?
Does this type of community based
school feeding program do anything
different to regular school feeding
programs?
Annex 3: Break-out group session: Research questions on procurement and social accountability
Fully decentralised
How do you calibrate frequency in terms
of the tendering and delivery to schools?
Can you work along different timelines
through different crops? How do you
balance what you’re paying for the
food?
What is the adequate scale which can
support farmer engagement? Within this
is the issue of frequency?
At what level is most appropriate to
adequately meet the scale required?
Is it cheaper to secure frequently with
small storage facilities or is it better to
secure less frequently with larger
storage facilities?
How can we effectively integrate
smallholder farmers into the
procurement process?
What are the trade-offs of different
procurement systems within the same
scenario?
Partial decentralisation
Map supply side interventions in the
area
Study optimum timing of procurement
considering
storage,
seasonality,
quality issues
Cost-benefit analysis of provision of
perishables
through
various
procurement options or community
contributions – in terms of cost,
nutrition, farmer income etc
Comparison of procurement through
traders or Farmers
Organizations in terms of impact on
farmer income
Identify optimal way of organising
social accountability system
Should social accountability be specific
to school feeding or wider? Use of links
to existing mechanisms
Questions around frequency, ration
design, fresh produce?
Integrated farm to school
What are the kinds of checks and
balances regarding selling food at a
fixed price, supplying schools and
selling to market/ what are the
regulations?
Does the system support the groups
in terms of enabling them to
interact better with the market?
(capacity building)
Assessment of the level of efficiency
compared to the centralised system
and other semi-centralised systems
Sustainability – do the women
choose to continue, or do they drop
out?
Opportunity cost – what are the
opportunity costs of the women
who engage with the system
compared to other monetary
activities.
How sustainable are these groups
sustainability and the school feeding
program?
Do you alter your procurement model to
meet the needs of the farmers, or do
you build the capacity of the farmers to
meet the needs of your procurement
model?
What is the role of the larger
community in supporting this
venture?
How are the groups progressing to
the 100% target
Are we clear that this is something that
the small holder farmers want?
Does smallholder involvement effect
positively or negatively the impact on
the children (quality received by
children)?
How do we prevent farmers from
becoming dependent on one market?
Annex 4: Break-out group session: Research questions on targeting
Fully decentralised
Where are the surplus and potential
surplus areas? Where can this be linked to
school feeding?
How can farmer networks support
targeting activities of school feeding
programs?
Partial decentralisation
Why is fortification not being
done? Is this due to costeffectiveness of implementing at
a small scale? Is it unclear who
will do it locally, etc.?
What can we learn from the private sector
networks already actively procuring within
the country?
Integrated farm to school
Who is truly bearing the cost and who is
benefiting? (Cost-benefit analysis)
Is it cheaper to do this model given the
cost of input?
Is the gain you are hoping for actually
occurring amongst the poor?
Is there a cost in terms of women’s time
and labour which is observable in the
under 2 children?
How does the inclusion of smallholder
farmers generating new business locally
along the value chain?
37
Annex 4: Participants list
Contact
Joy Kiiru
Alesha Black
Philomena Chege
Leah Rotich
Nur Guleid
Afua Kuffour
George T. Scharffenberger
Emily Janoch
Edoardo Masset
Lawrence Haddad
Rachel Sabates-Wheeler
Richard Longhurst
Shalini Roy
Liz Allen
Andrew Westby
Dick Commandeur
Simbo Keita
Fati Seidu
Eliana Vera
Elizabeth Kristjansson
Beth Thompson
Sally Burrows
Carmen Burbano
Marc Regnault de la Mothe
Emilie Sidaner
Alessia de Caterina
Harold Alderman
Roberta Sonnino
Leah Ashe
Hugh Waddington
Jeff Waage
Birte Snilstveit
Andreas Kretschmer
Aulo Gelli
Kristie Neeser
Iain Gardiner
Daniel Mumuni
Leo Nederveen
Alex Hulme
Nicola Lloyd
Brie McMahon
Abigail Deamer
Organisation
AERC
BMGF
G. of Kenya
G. of Kenya
G. of Kenya
GIMPA
HAAS
Harvard
IDS
IDS
IDS
IDS
IFPRI
LSHTM
NRI
SNV
SNV (Mali)
SNV (Ghana)
SNV
U. of Ottawa
Wellcome Trust
WFP (Evaluation)
WFP (Rome)
WFP (Rome)
WFP (Rome)
WFP P4P
World Bank
U. of Cardiff
U. of Cardiff
3ie
LIDC
3ie
WHU
PCD
PCD
PCD
PCD
PCD
PCD
PCD
PCD
PCD
38
Email
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
[email protected]
Annex 5: PCD HGSF working papers
1. Gelli A, Neeser K and Drake L. Home Grown School Feeding: linking small holder agriculture
to school food provision. London: Partnership for Child Development, 2010.
2. Sumberg, J., and R. Sabates-Wheeler. “Linking Agricultural Development to School Feeding.”
London: Partnership for Child Development, 2010.
3. Galloway, R.. Developing the rations for Home Grown School Feeding. Washington, DC:
Programme for Applied Technology, 2010.
4. Devereux, S., R. Sabates-Wheeler, and A. Pascual Martínez. 2010. Home Grown School
Feeding and Social Protection. London: Partnership for Child Development, 2010.
5. Gelli A. 2010. Food provision in schools in low and middle income countries: developing an
evidence based programme framework. London: Partnership for Child Development.
6. Shaad B, Jaisinghani N and Gelli A. Osun State Home Grown School Feeding and Health
Programme Case study. London: Partnership for Child Development, 2010.
7. DNC/WFP/PCD. A case study of the HGSF programme in Cote d’Ivoire. London: Partnership
for Child Development, 2010.
8. Espinoza M. A. and Palma J.I. Case Study on Local Development Initiatives of the School
Catering Program in Chile. London: Partnership for Child Development, 2011.
9. Johnson, C. and Janoch, E. Engaging communities: Evaluating Social Accountability in School
Feeding Programmes. London: Partnership for Child Development, 2011.
10. Ghana Insitute of Management and Public Administration and UC Berkeley Haas School of
Buiness. Ghana School Feeding Program: Re-Tooling for a Sustainable Future. London:
Partnership for Child Development, 2011.
11. Masset, E. and Gelli, A. Community participation and the links between agriculture, nutrition
and education: design of a randomised field experiment of “home-grown” school feeding in
Mali. London: Partnership for Child Development, 2011.
12. MS Swaminathan Research Foundation. Case study of the school feeding programme in
India. London: Partnership for Child Development, 2011.
13. Botswana Institute for Development Policy Analysis. Home Grown School Feeding in
Botswana: A Country Profile Case Study. London: Partnership for Child Development,
forthcoming.
14. Kenya Institute for Public Policy Research and Analysis. A case study of the Njaa Marufuku
Kenya Home Grown School Feeding Programme. London: Partnership for Child
Development, forthcoming.
39