S4.1 General methodology (applied to public and private sources of financing)
Supportive expenditures and weighting
The concept of food security and nutrition is cross-sectoral in nature. As a result, some financing flows were deemed to contribute wholly to food security and nutrition, i.e. to have solely or mainly food security and nutrition outcomes (specific expenditures), while others were deemed to have some food security and nutrition outcomes, but also substantial outcomes related to different policy areas (supportive expenditures). See Supplementary material to Chapter 3 for a detailed list of these expenditures by type.
To reflect as closely as possible the impact of these supportive expenditures on food security and nutrition, a weight was applied to them. The weight, 22 percent, was calculated based on household spending on food as a percentage of GDP in LICs, LMICs and UMICs.g
S4.2 Methodology for the estimation of domestic public spending on food security and nutrition
Among the sample of countries covered in the exercise aimed at developing preliminary estimates of public spending on food security and nutrition, those for which budget data were publicly available for at least the most recent five years were selected. These countries are Benin, Brazil, Georgia, India, Kenya, Mexico, Nigeria, the Philippines, South Africa and Uganda.
Data on public budget were collected from different data sources, in different formats and with varying degrees of detail and were classified according to the food security and nutrition financing definitions, both core and extended, using the keywords list developed and available in Table S3.3 of the Supplementary material to Chapter 3.
The amounts and evolution of public spending on agriculture and on food security and nutrition are presented for some LICs in Section 4.2 of the main report and for middle-income countries (MICs) in this supplementary material.
Data source for public spending on agriculture
In Section 4.2 of the main report, indicators of public spending on food security and nutrition are analysed in comparison with aggregates of public spending specific to agriculture. As an estimate of agricultural spending for Uganda, this report uses a narrow definition of public spending on food and agriculture as defined by FAO’s Monitoring and Analysing Food and Agricultural Policies (MAFAP) programme, which is consistent with the raw data used including both national and subnational (decentralized) spending.h For the other nine countries included in this analysis (Benin, Brazil, Georgia, India, Kenya, Mexico, Nigeria, Philippines and South Africa), budgetary central government data were found to be comparable with: i) the data reported to the FAO Statistics Division on GEA in its annual questionnaire; and ii) the data reported to the IMF using the analytical framework of the IMF’s Government Finance Statistics (GFS) Manual, 2014.39
Estimating public spending on food security and nutrition
For all countries, budget data defined according to the Classification of the Functions of Government (COFOG) serves as the basis for calculating food security- and nutrition-related expenditure. In brief, COFOG comprises a three-level classification with ten divisions at the top level, for example, “Economic affairs”. Each division is broken down into between six and nine groups at the next level of detail, for example, “Agriculture, forestry, fishing and hunting”. In most cases, these groups, in turn, are partly subdivided further into classes, for example – in the case of the group “Fuel and energy” – the class 70435 “Electricity”, covering traditional sources of electricity such as thermal or hydro supplies, and newer sources such as wind or solar heat. Therefore, a review of the divisions, groups and classes show how the broad socioeconomic objectives or functions of general government units are to be achieved through various kinds of expenditure. This facilitates evaluating delivery on the economic policy objectives of the consolidated general government and its subsectors, as relevant.
FAO GEA and IMF GFS both encourage reporting countries to present data by separately identifying expenditure for the consolidated general government along with each of its subsectors (Table S4.1). This focus on institutional coverage allows for determining whether expenditure primarily occurs within the context of the execution of the central government budget or whether other levels of government (e.g. subnational government) lead on food security- and nutrition-related activities. In the future, an effort will be made to expand the institutional coverage of the source data to include subnational government, as relevant. This will better inform the results obtained and discussed in Section 4.2 of the main report on trends and shares of total public spending for food security and nutrition against total budget and against spending on agriculture.
TABLE S4.1Structure of general government and its subsectors under the Classification of the Functions of Government
For Benin and Uganda, data on public spending following FAO’s MAFAP classification approach were already available and therefore used as a starting point for classifying expenditure on food security and nutrition. This was done by matching MAFAP public spending categories against food security and nutrition financing categories, as presented in Table S4.2. The MAFAP methodology accounts for public spending that benefits the food and agriculture sector directly (i.e. agriculture-specific), as well as agriculture-supportive expenditures that support rural development more broadly and may have an indirect effect on agricultural development (e.g. rural roads, health and education services). Agriculture-specific expenditures are divided into: i) payments to agents (e.g. producers, consumers, traders), also defined as expenditure on private goods and consisting mainly of input/output subsidies for producers and cash transfers/food aid targeting consumers; and ii) general support expenditures that benefit the sector through the provision of public goods (i.e. agriculture infrastructure, provision of extension services, research or marketing facilities).40
TABLE S4.2Monitoring and Analysing Food and Agricultural Policies public spending categories and corresponding determinants under the food security and nutrition financing definition
SOURCE: Adapted from FAO. 2015. MAFAP Methodology working paper: Volume II. Analysis of public expenditure on food and agriculture. MAFAP Technical Notes Series. Rome. https://www.fao.org/fileadmin/templates/mafap/documents/Methodological_Guidelines/METHODOLOGY_WORKING_PAPER_Vol2_Public_Expenditures.pdf
For the other eight countries, the process of deriving estimates of public spending on food security and nutrition consisted of combining the data published by national authorities for the period from 2018 to 2022, and classifying those data using an automated Excel-based classification tool. This tool facilitates using the national time series data along with their codes and nomenclature to derive the various food security and nutrition classifications. For example, as illustrated in Table S4.3, in the case of the Philippines, the activities of the Office of the Secretary and the Bureau of Fisheries and Aquatic Resources of the Department of Agriculture are classified under the core definition of “food security and nutrition”, and more specifically under “domestic production” associated with the component “food availability”. Activities of the Agriculture Credit Policy Council and the Center for Post-Harvest Development and Mechanization are instead associated with the extended definition of food security and nutrition.
TABLE S4.3Example of classification of food security and nutrition expenditure against the national nomenclature in the Philippines for selected Classification of Functions of Government groups
One major deviation from the food security and nutrition financing classification framework that applies to all countries is the treatment of forestry-related expenditure, which – due to the impossibility of identifying and classifying it under the extended definition and contrary to what was done for ODA data classification – was kept under the core definition, namely under the main determinant of food consumption.
Overall, the sources and methods for each of the ten countries covered in the exercise are summarized in Table S4.4, supplemented with country-specific notes below.
TABLE S4.4Data sources, coverage and classification assumptions for deriving estimates of public spending on food security and nutrition
SOURCE: Authors’ (FAO) own elaboration.
Application of the food security and nutrition financing approach: methodology by country
The following section details the methodology followed for each country to classify public spending on food security and nutrition, as well as data limitations and assumptions.
Benin
Budgetary information used for the Benin case study is sourced from the World Bank BOOST database.41 It includes actual expenditure data at national and subnational levels and covers on-budget expenditures for the period from 2017 to 2021. Expenditure data from the same source is also aggregated for each year and used as total public budget in this report. The database contains a total of 46 583 observations categorized across administrative (vote/section), economic (recurrent/capital), geographical (central level, deconcentrated services) and functional standard classifications.
In the attempt to reduce the number of budget lines to classify, as well as increase the classification consistency – considering that the same project appears in multiple budget lines within a year and across years – unique identifiers (UIDs) are created for each project, whereby each project is associated with one budget line and is thus classified once. The UID for each project is then associated with a combination of budget information at vote/ministry, programme, project and output levels to then create a classification sheet consisting of 5 194 budget lines to classify, from the initial 46 583 lines in the raw budget file. In the classification sheet, an MAFAP classification category and its associated weight are assigned to each budget line that falls within the MAFAP perimeter, and a category “non-MAFAP” is assigned to expenditures that fall outside the MAFAP perimeter. The MAFAP classification categories are then used as starting points for classifying expenditures on food security and nutrition. This is done by matching MAFAP public spending categories to food security and nutrition financing categories and assigning weights accordingly (see Table S4.2). Of the total 5 194 budget lines, 4 516 (or 87 percent) were classified as “non-MAFAP”, while the remaining 678 were classified within one of the MAFAP agriculture-specific or agriculture-supportive categories. Out of the total 5 194 lines, 2 331 (or 45 percent) were classified into food security- and nutrition-related spending.
For expenditures that fall outside the MAFAP perimeter, the lowest identifiable information across the budget classification variables (i.e. Section, Chapitre, Fonction Principale and Fonction Secondaire [Section, Chapter, Primary Function and Secondary Function]) is used to define whether the expenditure is part of the food security and nutrition classification perimeter based on a search of food security and nutrition keywords, and whether it is food security- and nutrition-specific or supportive (therefore to be weighted), as defined and suggested in Chapter 3 of the main report. Nonetheless, the food security and nutrition classification exercise is subject to some assumptions and limitations. The main ones include the following:
- Detailed documentation for projects and programmes under ministries relevant to food security and nutrition (e.g. health) were not consulted due to time constraints. As such, budget lines identified as food security and nutrition specific are relatively limited because some food security- and nutrition-specific words are unlikely to appear in the name of a budget line. Therefore, some food security and nutrition subdeterminants will have zero or smaller figures than they otherwise would.
- Running and administrative costs of ministries beyond the Ministry of Agriculture (i.e. Ministry of Education, Health, Transport, Environment, Trade and so on) are also assumed to broadly contribute to food security and nutrition.
Brazil
To derive public spending on food security and nutrition, spending estimates were referenced from the Projeto de Lei Orçamentária Annual42 (Annual Budget Proposal) presented by the Ministério do Planejamento e Orçamento43 (Ministry of Planning and Budget), whereby each annual budget proposal details actual expenditure from two years prior. In particular, Volume IV (Details of the Executive Branch’s budgetary actions) and Volume VI (Details of the Ministry of Education's budgetary actions) showing expenditure per ministry and containing the programme-level data needed were referenced to derive most of the core and extended food security and nutrition expenditure categories. In total, over 122 budgetary central government food security- and nutrition-related expenditures were identified and classified. A time series covering fiscal years from 2018 to 2022 was used as the basis for developing a preliminary food security and nutrition expenditure series.
Each ministry has a Programa de Gestão e Manutenção do Poder Executivo (Programme for Management and Maintenance of the Executive Branch) as well as programmes with titles such as Operações Especiais: Outros Encargos Especiais (Special Operations: Other Special Charges) and Operações Especiais: Gestão da Participação em Organismos e Entidades Nacionais e Internacionais (Special Operations: Management of Participation in National and International Organizations and Entities). These management programmes and special charges have been applied to administrative food security and nutrition categories. Brazil spends roughly 4 percent of its total annual budget on social protection, seen in both the annual budget proposals and the COFOG series submitted by authorities to the IMF. For this exercise, the relevant programmes identified under social protection (extended definition) were Social Protection under the Unified Social Assistance System (SUAS), Consolidation of the SUAS and Productive Inclusion of People in Situations of Social Vulnerability, which in various years were found in either the Ministry of Development and Social Assistance, Family and Combating Hunger or the Citizenship Ministry. For the purposes of calculating core versus extended food security and nutrition expenditure, it would have been useful to have a breakdown of programme data to allow for the extraction of food-specific social protection spending.
A programme entitled Food and Nutritional Security appears in several ministries, and it was allocated to food affordability in the core definition, under the assumption that spending targets the subsidization of food and food price support. However, the authorities would be able to clarify whether some of these expenditures go towards domestic production or food-specific social protection. The subprogramme was seen under the Presidency of the Republic, Ministry of Agriculture, Ministry of Health, Ministry of the Environment, Ministry of Regional Development, Citizenship Ministry and Ministry of Development and Social Assistance, Family and Combating Hunger.
Compared to other countries in the exercise presented here, Brazil at the central government level executes relatively low expenditure on categories like transport and water and sanitation, but this appears to be explained by the large portion of spending on social protection. Under water and sanitation in the core definition, the Basic Sanitation programme in the Ministry of Environment and the Ministry of Regional Development was identified as relevant, as was the Environmental Quality programme in the Ministry of Health since this is described online as involving the monitoring of water quality and ensuring compliance with water standards, among other non-food security and nutrition functions. However, it is unclear how much of the health subprogramme is relevant to water and sanitation.
Georgia
For deriving preliminary estimates of public spending on food security and nutrition, the extensive set of fiscal data disseminated by the Department of Budget pertaining to budgetary central government expenditures was consulted.44 In particular, Chapter VI: State Budget Payments for 12 Months of 2022 According to the Program Classification was referenced as the basis for developing a time series that encompasses each of the fiscal years from 2019 to 2022. Government expenditure on agriculture comprised less than 1 percent of total government expenditure in 2022. However, if seen from the broader perspective of the consolidated general government, GEA increases from 1.8 percent to 3.3 percent of total outlays.
Overall, the process of deriving the preliminary food security and nutrition estimates comprised the classification of more than 150 separate types of budgetary central government outlays. Issues that could be usefully clarified with the authorities are: i) the nature of the various expenditures under budget code 24-08 “Development of innovations and technologies in Georgia”. The majority of the components pertain to energy which is “supportive” but some outlays, such as budget code 24-08-02 “Supporting the development of innovation and technology in Georgia” could possibly be food security- and nutrition-specific in nature, and thereby warrant inclusion; ii) under budget code 31-00 “Ministry of Environment Protection and Agriculture of Georgia”, certain outlays, such as budget code 31-06 “Modernization of melioration systems” require further information to ensure proper treatment; iii) the nature of the World Bank-financed budget code 32-08 “Innovation, Inclusivity and Quality Project – Georgia I2Q” in relation to food security and nutrition; and iv) partitioning budget code 56-13-09 “Adjara Villages Water Supply and Sewerage Program”, funded by the European Union, among the relevant food security and nutrition core and extended components would be analytically useful.
India
Among the countries considered, India is particular. While the national presentation of expenditure is provided in the annual finance accounts, Union Government45 allocates revenue and capital expenditure into three broad types (general, economic and social services), and Schedules 9 and 10 detailing expenditure by minor heads provide sufficient detail to derive many of the core and extended food security and nutrition components. As such, a time series covering the fiscal years from 2018 to 2022 was used as the basis for developing a preliminary food security and nutrition expenditure series.
By analysing individual budget heads and the components to which the related expenditures pertain, it was possible to classify food security- and nutrition-specific items. Overall, the preliminary food security and nutrition estimates are the result of classifying about 110 separate types of budgetary central government outlays. For example, under the broad category Social Services (a) Education, Sports, Art and Culture, outlays could be separately identified on the National Programme of Mid-Day Meals in Schools under Budget Head 2202 General Education as being relevant for inclusion under Food Access. Similarly, under Social Services (b) Health and Family Welfare, Budget Head 2210 Medical and Public Health includes a variety of outlays, one of which pertains to Prevention of Food Adulteration that relates to Food Utilization in the core food security and nutrition classification structure. Although the proper allocation across food security and nutrition categories related to Social Services was generally clear, there were a few items warranting further consideration. Among these, Social Services (e) Budget Head 2225 Welfare of Scheduled Castes, Scheduled Tribes, Other Backward classes and Minorities is currently being classified under Pathway 6 as it seems to relate more to Equitable Access to Resources, Assets and Essential Services than it does to the core food security and nutrition category Food Access that includes poverty (e.g. food-specific social protection). This could usefully be clarified with the authorities and, if appropriate, the outlays partitioned. Similarly, Social Services (g), Social Welfare and Nutrition comprises Budget Heads 2236 Nutrition and 2245 Relief on account of Natural Calamities, with the latter further disaggregated by outlays pertaining to Relief on account of Natural Calamities – Drought (2245-01) and Relief on account of Natural Calamities – Floods, Cyclones, and so on (2245-02), with outlays on the associated “of which” items being food security- and nutrition-cross-cutting in nature: Gratuitous Relief, Repairs and Restoration of Damaged Roads and Bridges, Repairs and Restoration of Damaged Water Supply, Drainage and Sewerage (systems); Ex-Gratia Payments to Bereaved Families; and Evacuation of Population. The proper food security and nutrition treatment could usefully be clarified with the authorities and, if appropriate, the outlays reclassified.
Regarding the Economic Services presented in the Finance Accounts, the outlays presented under section (a) Agriculture and Allied Activities could be allocated across various core and extended food security and nutrition categories. This was also the case for section (b) Rural Development, (c) Special Area Programmes, and (d) Irrigation and Flood Control, while other sections (e–j) allowed the identification of some specific expenditures ranging from New and Renewable Energy (including Renewable Energy for Rural Applications) to Fertilizer, Chemical and Pesticides industries, to outlays on the Government of India’s Climate Change Action Programme.
As a next step, and with the aim of ensuring the broadest institutional coverage possible for the food security and nutrition estimates, the accounts published for the state governments (30) of India as presented in the State Accounts Report of the Comptroller and Auditor General of India shall be examined.46
Kenya
For Kenya, the National Treasury’s sector budget proposal reports47 pertaining to executed budgetary central government expenditures were consulted. The national budget is classified into sectors that are largely aligned with COFOG categories: Agriculture, Rural and Urban Development, Education, Energy, Infrastructure and ICT, Environment Protection Water and Natural Resources, General Economic and Commercial Affairs, Governance Justice Law and Order, Health and Public Administration, and International Relations. Each annual sector report shows actual expenditure for the previous three years by ministry, programme and subprogramme. This was used to develop a fiscal time series for the period from 2017 to 2022. According to the subsector reports and the COFOG series submitted by Kenyan authorities to the IMF, Kenya spent roughly 2 percent of its total budget on GEA during the referenced period.
Overall, the process of deriving the preliminary food security and nutrition estimates comprised the classification of more than 263 separate types of budgetary central government outlays. The subprogramme level data allowed for a generally clear allocation of expenditures to food security and nutrition categories, but issues that could be usefully clarified with the authorities are: i) the nature of the expenditures under subprogramme Transboundary Waters in the programme Water Resources Management in the Water and Sanitation Sub-sector (the spending under this subprogramme is relatively small but was ultimately allocated to water access in the core definition); and ii) whether the expenditures under subprogramme Water Harvesting are potentially spread across domestic water use, agricultural use and/or environmental conservation (this subprogramme falls under the Water Storage and Flood Control in the Water and Sanitation subsector; currently, the entire subprogramme is allocated to water and sanitation in the core definition).
Another possible area of clarification is the subprogramme Child Community Support Services in the programme Social Development and Children Services, which falls under the Ministry of Social Security and Protection. It was unclear whether these expenditures involve community services or social worker support for children, but online sources seem to support that they are development related in the form of social protection for children, therefore falling under the poverty category in the core definition.
Mexico
To derive estimates of public spending on food security and nutrition, a review of the various sources (Informacion Presupuestaria [Budget Information]) presented by the Secretaría de Hacienda y Crédito Público (Secretariat of the Treasury and Public Credit) relating to expenditure of the Federal Government of Mexico were referenced.48 In particular, the Estado Analítico del Ejercicio del Presupuesto de Egresos Clasificación Funcional/Programática (Analytical Status of the Expenditure Budget Exercise Functional/Programmatic Classification) for each year between 2018 and 2022 was found to contain the level of detail needed to derive most of the core and extended food security and nutrition expenditure categories. In total, over 85 food security- and nutrition-related expenditures were identified and classified.
Although the structure of the functional/programmatic classification facilitates identifying most key outlays, to allow for an even more robust set of food security and nutrition estimates, further clarity on some specific programmes could prove useful. For example, in 2022 there were significant outlays related to the Censo Agropecuario (Agricultural Census) being carried out under the programme Información Nacional Estadística y Geográfica (National Statistical and Geographic Information) that could be fully allocated to Food Availability (Domestic Production) rather than treated as Supportive, particularly if the agricultural census was completed in the fiscal year. Similarly, the large expenditure on the Censos Económicos (Economic Censuses) in 2019 could warrant investigation to ascertain the extent of agrifood and food chain impacts. All expenditure related to the Instituto Nacional para la Evaluación de la Educación (National Institute for Educational Evaluation) was also found to have ended in 2019. However, as these outlays would generally be considered food security- and nutrition-supportive, understanding whether the activity or programme was terminated or whether the source data are incomplete would be useful.
An area that could be usefully clarified as to the exact nature of the related expenditure relates to programmes under Gobernación (Governorate). Two programmes were identified: Atención a refugiados en el país (Support to refugees in the country [E006]) and Política y servicios migratorios (Immigration policy and services [E008]) that appear to pertain to Pathway 1. However, other programmes such as Determinación, ejecución y seguimiento a las acciones de búsqueda de personas desaparecidas y no localizadas (Planning and monitoring of search for missing persons [P026]) and Subsidios para las acciones de búsqueda de personas desaparecidas y no localizadas (Funding for search for missing persons [U008]) could be classified as Food Access should they largely comprise anti-poverty food security-related outlays.
Another area potentially impacting Pathway 6 relates to the various programmes under Hacienda y Crédito Público (Treasury and Public Credit). While the subprogramme related to Agropecuaria, Silvicultura, Pesca y Caza (Agriculture, Forestry, Fishing and Hunting) falling under Desarrollo Económico/Otros Asuntos Sociales (Economic Development/Other Social Issues) clearly relates to providing equitable access to resources, assets and essential services, the subprogramme Programa de Inclusión Financiera (Financial Inclusion Programme [F035]), while deemed relevant to food security and nutrition could be investigated further, given the major outlays included therein.
Nigeria
Estimates for Nigeria were referenced from the quarterly budget performance reports of the federal government available on the Open Treasury49 fiscal data portal. Budget execution data are available by administrative, economic or functional classification for the years from 2018 to 2021. Unfortunately, no data have been published on the portal since 2021. The functional classification data only contain aggregates for the functions, which is not enough detail to produce food security and nutrition estimates according to the exercise presented here. Therefore, administrative ministries, departments and agencies (MDAs) data are used to derive the preliminary food security and nutrition estimates through the classification of about 311 separate types of federal government outlays.
For the Federal Ministry of Agriculture and Rural Development Headquarters, expenditure is presented as a ministry total without further detail on programmes and subprogrammes. This expenditure was allocated to domestic production in the core definition as it seemed to mostly apply there, but it would be useful to see a breakdown of the activities of this MDA. There are several farmer producer organizations (FPOs) around the country that appear as MDAs in the budget. According to online sources, the activities of FPOs include the cultivation from inputs, technical services, processing and marketing. These MDAs are allocated to food access, under the assumption that most of the spending is for food marketing and improving food distribution networks. A breakdown of the FPOs’ spending would allow for a clearer split into more than one food security and nutrition category.
There are several technology business incubator centres listed as MDAs in cities across the country. These expenditures have not been allocated to any food security and nutrition categories as it was unclear whether their function relates in part to any technology or infrastructure for food and nutrition. Overall, Nigeria provided an extensive database of MDA budget performance, but a further breakdown into programmes and subprogrammes would allow for a more robust analysis.
Philippines
In the case of the Philippines, the extensive set of fiscal data disseminated by the Department of Budget and Management pertaining to budget of expenditures and sources of financing was consulted.50 In particular, Table B.5.c Details of the Classification of the Functions of Government was referenced for each of the fiscal years from 2019 to 2022. Overall, this approach facilitated deriving the preliminary food security and nutrition estimates through the classification of about 175 separate types of budgetary central government outlays.
The authorities have included a variety of programmes under General Services that could be placed under Economic Affairs as well. For example, activities identified as specific to the Department of Agriculture, such as the Agriculture Credit Policy Council, the Bureau of Fisheries and Aquatic Resources, the Fertilizer and Pesticide Authority, the National Meat Inspection Service, the Philippine Center for Post Harvest Development and Mechanization, the Philippine Council for Agriculture and Fisheries and the National Fisheries Research and Development Institute could all be considered related to Domestic Production under the Food Availability dimension. Similarly, general services subprogrammes of the Department of Environment and Natural Resources, such as the Environmental Management Bureau, the National Mapping and Resource Information Authority and the National Waste Water Resources Board, appear to relate to the COFOG division Environmental Protection. Understanding why the national authorities have treated these outlays as general services would ensure that they are properly classified for deriving food security and nutrition estimates.
It was also observed that the activities of the Food and Nutrition Research Institute, operating under the Department of Science and Technology, appear under various COFOG divisions. It would be good to clarify why this is the case in order to ensure proper treatment in deriving food security and nutrition estimates. It would also be useful to better understand why the activities of the agricultural colleges and universities appear under some of the different COFOG divisions that comprise the division Education. For example, it was observed that, among others, the Southern Philippines Agri-Business and Marine and Aquatic School of Technology appears under Tertiary education, Education not definable by level, Subsidiary services to education, R&D Education, and Education n.e.c. Understanding why the national authorities have treated the outlays of these educational institutions in this way would ensure that they are properly classified for deriving food security and nutrition estimates.
South Africa
To derive the estimates of public spending on food security and nutrition, actual budget execution was referenced from the estimates of national expenditure51 (ENE) presented by the National Treasury as part of their annual budget.52 Each ENE details the main budget expenditure execution for the previous three years. In particular, the section entitled “Expenditure trends and estimates” in each department’s chapter, where spending is available per programme and subprogramme as well as per economic classification, was referenced. This approach facilitated the derivation of preliminary food security and nutrition estimates through the classification of about 220 separate types of budget outlays for the fiscal years 2017–2021.
According to the ENE data, South Africa spent approximately 1 percent of its total budget on GEA during the referenced period, and this outcome is also seen in the COFOG data submitted by South African authorities to the IMF. Although the structure of the data by subprogramme in the ENE facilitates the identification of most key outlays, to allow for an even more robust set of food security and nutrition estimates, further clarity on some specific programmes could prove useful. It was unclear whether the subprogramme Ingonyama Trust Board in the Food Security, Land Reform and Restitution programme is in part related to food security or production, or whether it is purely for restitution purposes. Online sources indicate that this subprogramme in the Department of Agriculture, Land Reform and Rural Development administers land to members of traditional communities, but the purpose of the land is not clear.
Another potential area of improvement is the National Disaster Management Center in the Department of Cooperative Governance and Traditional Affairs where funds are allocated for disaster relief and recovery. Given the reforms that have been implemented in South Africa’s disaster funding, it is realistic to believe that at least some of these funds go towards food provision as well as temporary shelter in the case of immediate disaster relief funding. It may be difficult for authorities to extract, but it would have benefited the exercise presented here to see the portion spent on food for displaced people following natural disasters.
Uganda
Budgetary information used for the Uganda case study is sourced from the World Bank BOOST database.53 It includes actual expenditure data at national and subnational levels and covers on-budget expenditures for 2018–2022. Expenditure data from the same source are also aggregated for each year and used as total public budget in this report. In the raw budget file, expenditure data are structured by standard budget classifiers, namely: administrative (i.e. vote/ministry), economic (recurrent/capital), functional (i.e. education, agriculture, health and so on) and geographical (i.e. national, district/county and so on), and account for 133 392 budget lines.
As done for Benin (see above), UIDs are created for each project, which is associated with a combination of budget information at vote/ministry, programme, project and output levels to then create a classification sheet consisting of 43 216 budget lines that are classified from the initial 133 392 lines in the raw budget file. The MAFAP classification is then used as a starting point following the matching categories in Table S4.4, while the category “Non-MAFAP” is assigned to expenditures that fall outside the MAFAP perimeter. For these, the lowest identifiable information across the budget classification variables (i.e. vote/ministry, programme, project and output levels) is used to define whether the expenditure is part of the food security and nutrition classification perimeter based on a search of food security and nutrition keywords, and whether it is food security- and nutrition-specific or supportive (therefore to be weighted), as defined and suggested in Chapter 3 of the main report. The same limitations that apply for Benin are also valid for Uganda and include the following:
- Detailed documentation for projects and programmes under ministries relevant to food security and nutrition (e.g. health) were not consulted due to time constraints. As such, budget lines identified as specific to food security and nutrition are relatively limited because some food security- and nutrition-specific words are unlikely to appear in the name of a budget line. Therefore, some food security and nutrition subdeterminants will have zero or smaller figures than they otherwise would.
- Running and administrative costs of ministries beyond the Ministry of Agriculture (i.e. Ministries of Education, Health, Transport, Environment, Trade and so on) are also assumed to broadly contribute to food security and nutrition.
Data on public spending on food security and nutrition
The public spending on food security and nutrition for selected countries is presented in Figures S4.1 to S4.8.
FIGURE S4.1Public spending on agriculture and on food security and nutrition in Brazil
SOURCE: Authors’ (FAO) own elaboration based on Government of Brazil. 2024. Orçamentos Anuais PLDO I LDO I PLOA I LOA - Atos Normativos. In: gov.br – Ministério do Planejamento e Orçamento. [Cited 30 April 2024]. https://www.gov.br/planejamento/pt-br/assuntos/orcamento/orcamento/orcamentos-anuais
FIGURE S4.2Public spending on agriculture and on food security and nutrition in Georgia
SOURCE: Authors’ (FAO) own elaboration based on Ministry of Finance of Georgia. 2024. Ministry of Finance of Georgia. [Cited 30 April 2024]. https://www.mof.ge/en/
FIGURE S4.3Public spending on agriculture and on food security and nutrition in India
SOURCE: Authors’ (FAO) own elaboration based on Ministry of Finance, Government of India. 2024. Accounting information. In: Controller General of Accounts, Department of Expenditure. [Cited 30 April 2024]. https://cga.nic.in/index.aspx#account-section
FIGURE S4.4Public spending on agriculture and on food security and nutrition in Kenya
SOURCE: Authors’ (FAO) own elaboration based on The National Treasury & Economic Planning, Republic of Kenya. 2021. Sector budget proposal reports. [Cited 30 April 2024]. https://www.treasury.go.ke/sector-budget-proposal-reports
FIGURE S4.5Public spending on agriculture and on food security and nutrition in Mexico
SOURCE: Authors’ (FAO) own elaboration based on Gobierno de México. 2024. Investor Relations Office of the Ministry of Finance and Public Credit. In: Gobierno de México. [Cited 9 May 2024]. https://www.finanzaspublicas.hacienda.gob.mx/es/Finanzas_Publicas/Ingles
FIGURE S4.6Public spending on agriculture and on food security and nutrition in Nigeria
SOURCE: Authors’ (FAO) own elaboration based on Federal Government of Nigeria. 2024. Federal Government of Nigeria. [Cited 30 April 2024]. https://opentreasury.gov.ng
FIGURE S4.7Public spending on agriculture and on food security and nutrition in the Philippines
SOURCE: Authors’ (FAO) own elaboration based on Republic of the Philippines, Department of Budget and Management. 2022. Budget of expenditures and sources of financing FY 2023. Manila. https://www.dbm.gov.ph/index.php/2023/budget-of-expenditures-and-sources-of-financing-fy-2023
FIGURE S4.8Public spending on agriculture and on food security and nutrition in South Africa
SOURCE: Authors’ (FAO) own elaboration based on Republic of South Africa, National Treasury Department. 2024. National budget. In: National Treasury. [Cited 30 April 2024]. https://www.treasury.gov.za/documents/national%20budget/default.aspx
S4.3 Methodology for estimating international development finance flows to food security and nutrition
International development financing comprises two main types of flows: official development assistance (ODA), which is humanitarian and development aid from governments or multilateral organizations that meets a minimum grant element requirement;54 and other official flows (OOF), which are official sector transactions that do not meet ODA criteria (excluding official export credits).55
The following main databases were used for ODA and OOF: the OECD’s Creditor Reporting System (CRS) database,56 which tracks humanitarian and development ODA and OOF from Development Assistance Committee (DAC)57 countries and multilateral organizations (as well as from several other countries, such as Türkiye and the Russian Federation, that do not belong to DAC or indeed to OECD but nonetheless report into this database);58 and, for China, the AidData database,59 developed by an international network of researchers based on publicly available information. This tool covers only the period from 2000 to 2017 (for project approval, with implementation until 2021). Other official flows from this source were not used for this analysis, due to the difficulty of determining the extent of their development assistance purpose.
Limited numbers were also obtained from OECD’s Total Official Support for Sustainable Development (TOSSD) database60 (where data are available only from 2019) and, for Brazil’s international development assistance, from the Brazilian Institute of Applied Economic Research.i
Few significant donors are excluded from these databases. For the largest of them, India, OECD estimates that bilateral development cooperation outflows in 2021 amounted to USD 1 billion. The estimate for South Africa is USD 34 million.63
There are limitations to the several ways that the CRS database56 offers to identify existing food security and nutrition financing, as detailed in Chapter 3 of the main report. In addition, TOSSD uses the CRS purpose codes, but AidData59 only uses the CRS’s first level of codification, i.e. sector codes. It was therefore decided to use the following methodological steps to identify the interventions in these databases that match this report’s definitions of financing for food security and nutrition.
First, for the CRS and TOSSD databases, a list of purpose codes was selected (see Supplementary material to Chapter 3). Second, an artificial intelligence (AI) tool was trained on a sample of database entries, based on keyword searches (see Supplementary material to Chapter 3), to recognize entries relevant to food security and nutrition thanks to text patterns it learned to identify in long and short project descriptions. This tool was then used for several purposes: to identify flows relevant to food security and nutrition in non-selected purpose codes in the CRS and TOSSD databases; to identify flows relevant to food security and nutrition in AidData and in Brazil data; and to classify all the flows identified in all data sources, through both purpose codes and the AI tool, between the core and extended definitions. Third, keyword searches were performed on the results of this screening, on a new sample to measure accuracy and on the full results to sort them between sublevels of the food security and nutrition definition framework as well as between specific and supportive expenditures. When several keywords assigned to different subdeterminants or drivers were found in the same database entry, a semantic similarity analysis was performed to classify the entry in the subdeterminant or driver that contained the most semantically similar entries.
Any double counting between the earmarked contributions that international organizations receive from donor countries and those they pay to recipient countries was eliminated.
S4.4 Private sector financing
Definitions and references for philanthropic flows and for remittances
To estimate philanthropic flows, the same methodology was applied to the CRS database56 as to ODA/OOF flows (see Section S4.3).
The following sources of information and assumptions were used on cross-border remittances:
- for remittances to low- and middle-income countries – World Bank and KNOMAD brief Leveraging diaspora finances for private capital mobilization;64
- for the share of remittances going to rural areas (40 percent, based on unpublished data) and for the share of remittances used to finance income-generating activities (15 percent) – The International Fund for Agricultural Development (IFAD) report Sending Money Home: contributing to the SDGs, one family at a time (2017);65
- for the share of food systems in rural areas employment in LICs/LMICs/UMICs, in the absence of comprehensive research in the area, this was assumed to be 50 percent;
- for the share of agriculture in the GDP of LMICs – World Bank indicator66 (8.9 percent, rounded up to 10 percent); and
- for food consumption in percentage of household consumption – FAOSTAT’s26 sample of 37 LICs/LMICs/UMICs (excluding China and India); the indicator for medium-tercile households was used, assuming that bottom-tercile households have less access to migration and that fund transfers from migrants from top-tercile households are more likely to represent philanthropy or business investments than remittances.
Definitions and references for foreign direct investments
Section 4.1 of the main report analyses foreign direct investment for announced greenfield projects. Each of these terms is explained here.
United Nations Trade and Development (UNCTAD) defines foreign direct investments (FDIs) as investments made by a private entity resident in one economy in an enterprise resident in another. The investments must involve a long-term relationship and reflect a lasting interest and control, and be made directly rather than through capital markets.67 UNCTAD also defines them as equity investments (or intercompany loans), yet includes in them international project finance, despite the debt component.
UNCTAD classifies FDIs in four purpose categories, following the OECD framework:68 cross-border mergers and acquisitions (M&A), greenfield projects, extension of capital and financial restructuring. UNCTAD provides data for the first two, and the first one is excluded for the purposes of this report, for the reason mentioned above. The OECD framework notes that a sharp distinction is often drawn between greenfield investment, providing fresh capital and additional jobs, and M&A that are perceived to include only a change of ownership in an existing corporate entity. This theoretical distinction between the types of FDIs, however, may differ in practice, and in a number of instances the acquisition of existing enterprises can provide important additional economic benefits.68 Yet, any redirection into food security and nutrition of the proceeds from the sales of a company by its former shareholders is hypothetical and, if confirmed, would be observed at a different step and through different indicators.
UNCTAD encourages a focus on announced rather than closed projects. On project finance, it notes:
It is clear that using announcement data will tend to overestimate the numbers and values of projects, because some projects never reach the stage of construction or completion. In contrast, using financial close data will lead to underestimation, because many projects have open-ended financing arrangements or financial close data are not reported […]. Looking at all project finance (including domestic deals) shows that the degree of underestimation from the use of financial close data is actually larger than the degree of overestimation from the use of announcement data. This is not the case for international projects, where relatively more projects reach financial close, but the degree of underestimation remains substantial.69
Definitions and references for blended finance
Blended finance is categorized in Section 4.1 of the main report under private financing, to reflect its purpose of attracting private funds to investments in sustainable development, despite its limited success so far in doing so (see Chapter 5 of the main report).
Convergence tracks blended finance deals based on three criteria:
- The transaction attracts financial participation from one or more private sector investor(s).
- The transaction uses catalytic funds in one or more of the following ways:
- –public or philanthropic investors provide concessional capital, bearing risk at below market returns, to mobilize private investment, or provide guarantees or other risk mitigation instruments;
- –transaction design or preparation is grant funded;
- –transaction is associated with a technical assistance facility (e.g. for pre- or post-investment capacity building).
- The transaction aims to create development impact related to the Sustainable Development Goals (SDGs) in developing countries.70
Definitions and references for indirect private finance mobilization
Chapter 4 of the main report cites numbers for “private indirect mobilization” in LICs and MICs provided in the joint report Mobilization of private finance 2020+2021,71 by the multilateral development bank (MDB) Task Force on Mobilization of Private Finance, a group of multilateral development banks and development finance institutions. This report, published in June 2023, breaks down total private mobilization (which it also calls private co-financing) between private direct mobilization and private indirect mobilization, which it defines as follows:
Private direct mobilization:
Financing from a private entity on commercial terms due to the active and direct involvement of an MDB leading to commitment. Evidence of active and direct involvement includes mandate letters, fees linked to financial commitment, or other validated or auditable evidence of an MDB’s active and direct role leading to commitment of other private financiers.71
Private indirect mobilization:
Financing from private entities provided in connection with a specific activity for which an MDB is providing financing, where no MDB is playing an active or direct role that leads to the commitment of the private entity’s finance.71
To avoid any double counting, private direct mobilization is here assumed to be identical to blended finance.
The MDB Task Force also differentiates mobilization from catalysation, which it defines as:
Private sector financing that results from (as opposed to happening in connection with) an activity or multiple activities of an MDB. It includes investments made because of an operation up to three years after completion.71
The numbers provided are not broken down by sector or by SDG but only between investments in infrastructures (which probably contain some supportive investments) and other investments (which may contain some specific investments in agriculture or in food security and nutrition).
