1. Key message  

A review of eight recent modelling reports of the additional investment needed for achieving Sustainable Development Goal 2 (SDG 2) by 2030 reveals a wide range of estimates; from a total of USD 86 billion to over USD 4 trillion. This results in confusion for governments, donors and policymakers. So why are the numbers so different? The answer largely depends on the question the researchers asked.

Three of the eight reports are numbered in the trillions because they go beyond SDG 2 to include a range of other linked SDGs or focus on agri-food systems transformation. The remaining five reports are numbered in the billions because they focus on the SDG 2 sub-targets. However, even within the five reports focusing on SDG 2, the range is USD 86 billion to 760 billion. This is partly because of how many SDG 2 sub-targets are included, and the types of interventions modelled (See Figure 1). But a closer look reveals that the devil can often be in the detail.

The choice of targets and the quantification of those targets, the baselines and the data sources used for baseline projections, and the modelling approaches, lead to different estimates that are confusing policymakers and are sometimes even contradictory. This technical note explores the reasons behind the dramatic variations, examines the factors contributing to the differences, and explains how researchers and modellers could provide more clarity and coherence to policymakers in the future. 

The main reason for the wide range of cost estimates for ending hunger – from USD 86 billion to over USD 4 trillion – is between those that go beyond SDG 2 versus those that focus on SDG 2 sub-targets. 

Figure 1. Comparison of SDG 2 modelling reports, SDG sub-targets, and additional costs 

Source: Author’s own based on Laborde & Torero (2023), High Level Political Forum on Sustainable Development (2023), Food and Land Use Coalition (2019), FAO, IFAD, WFP (2015), IFPRI (2019), ZEF (2024), Ceres2030 (2020), IFPRI & IISD (2016), Global Nutrition Report (2021). 

Note: For all studies, except ZEF (2024), the total cost is calculated by multiplying the additional cost per year by the number of years. The total cost of ZEF (2024) is derived from the report, which applies a 2 percent annual discount to some interventions over five years and to others over six years. 

2. Why do policymakers need accurate cost estimates to achieve SDG 2? 

Knowing the additional investments needed to achieve SDG 2 by 2030 enables donors and governments to allocate resources efficiently and effectively towards their global commitments. This ensures that spending is directed towards the most effective programmes, maximising the impact of every dollar spent. Accurate cost estimates help prevent underfunding, which can lead to incomplete projects and unmet goals, and overfunding, which can result in wasteful spending and missed opportunities to address other critical needs. By optimally allocating resources, donors and governments can achieve maximum impact, making the goal of ending hunger more attainable and sustainable. 

Accurate cost estimates also help policymakers develop effective policies and long-term plans by setting realistic goals, timelines and budgets. These estimates create a baseline for monitoring progress, evaluating the effectiveness of interventions and making data-driven adjustments, ensuring that goals are being met. Consequently, this enhances accountability and transparency, as policymakers can demonstrate to the public and stakeholders that funds are being used efficiently and effectively. 

Accurately estimating the additional investment needed involves complex, multidisciplinary models that can simulate the world we want by 2030 based on the realities of the world today and projections for what the world will look like if we continue with a business-as-usual approach. With advances in modelling frameworks and improved datasets, a proliferation of modelling work is now possible with a higher degree of reliability and robustness than ever before. But, depending on the question asked, the timeframe considered, and the interventions included, the estimates can vary wildly.  

Multiple reports have estimated the additional investments needed to achieve SDG 2. Some focus on ending hunger by focusing on selected SDG 2 sub-targets, including increasing agricultural productivity and the incomes of small-scale food producers, promoting sustainable agriculture, and increasing the consumption of healthy diets. Others aim to end hunger by transforming agri-food systems or eradicating poverty, thus going beyond the scope of SDG 2. This has resulted in varying estimates for the total additional cost needed to achieve SDG 2 ranging from USD 86 billion to over USD 4 trillion. This section provides an overview of eight recent modelling reports, highlighting the different approaches, research questions, and interventions (see Figure 1). 

3. The trillion-dollar reports 

Three of the eight reports are numbered in the trillions (see Figure 2). Laborde & Torero (2023) and Food and Land Use Coalition (2019) offer a comprehensive approach to ending hunger by transforming agri-food systems, extending beyond the traditional focus on SDG 2 components such as food security, dietary diversity and nutrition, agricultural productivity, and sustainable food production systems. This transformation also includes reducing food loss and waste, creating sustainable livelihoods, preserving biodiversity and natural resources, integrating land use planning, and reducing energy consumption and greenhouse gas emissions. By addressing these multiple facets, these reports suggest that transforming agri-food systems can simultaneously tackle various interconnected SDGs, leading to a more comprehensive and lasting solution for ending hunger. Similarly, the FAO, IFAD, WFP (2015) report aims to achieve SDG 2 by eliminating poverty (SDG 1). For this reason, it calls for investment in social protection and pro-poor development, arguing that hunger results from a lack of purchasing power. The total additional investment needed to end hunger while transforming agri-food systems is USD 3.9 to 4 trillion.  

The trillion-dollar reports end hunger by transforming agri-food systems and/or eradicating poverty 

Figure 2. Cumulative additional cost required to end hunger by transforming agri-food systems and/or eradicating poverty, in USD billion 

Source: Author’s own based on FAO, IFAD, WFP (2015), Food and Land Use Coalition (2019), Laborde & Torero (2023), and High-Level Political Forum on Sustainable Development (2023) 

Note: Food and Land Use Coalition (2019) estimate that the additional investment needed to transform land use and food systems is USD 300 to USD 350 billion. The graph shows the average of this range, USD 325 billion annually and USD 3.9 trillion in total. 

Laborde & Torero (2023) aim to enhance the efficiency of agri-food systems by increasing agricultural productivity and reducing food loss and waste. This strategy has the potential to lift 314 million people out of hunger and provide 568 million people with access to healthy diets. The report aims to reduce chronic hunger to a 5% prevalence of undernourishment and enable 3 billion people to afford healthy diets by 2030. To achieve these goals, Laborde & Torero (2023) propose several key interventions, including the implementation of social safety nets, establishment of school feeding programmes, introduction of innovation packages (such as irrigation and livestock training), reduction of food waste and loss, the repurpose of farm subsidies and the reform of consumer incentives. These measures support various SDGs, including reducing extreme poverty (SDG 1), ensuring universal access to nutritious food (SDG 2.1), making healthy diets affordable (SDG 2.2), limiting greenhouse gas (GHG) emissions from agriculture (SDG 2.4), reducing food loss and waste (SDG 12.3), and promoting the sustainable use of biodiversity and ecosystems (SDG 15.1). The total estimated cost for these interventions is over USD 4 trillion, or USD 680 billion per year from 2024 to 2030 (High Level Political Forum on Sustainable Development, 2023). 

Food and Land Use Coalition (2019) aims to transform food and land use systems through ten critical transitions. These include promoting healthy diets, scaling productive and regenerative agriculture, protecting and restoring nature, securing a healthy and productive ocean, investing in diversified sources of protein. Additionally, the report aims to reduce food loss and waste, build local loops and linkages, harness the digital revolution, deliver stronger rural livelihoods and improve gender equality while accelerating the demographic transition. Achieving these goals requires an estimated investment of USD 300 to USD 350 billion annually. These investments are projected to yield total economic gains of USD 5.7 trillion per year by 2030 and USD 10.5 trillion per year by 2050, as well as generate business opportunities estimated to be worth USD 4.5 trillion annually by 2030. 

FAO, IFAD, WFP (2015), aims to achieve SDG 2 by investing in social protection and pro-poor development, arguing that hunger results from a lack of purchasing power. Therefore, eliminating poverty (SDG 1) is essential for eliminating hunger (SDG 2). The report suggests that ending hunger can be achieved by increasing individual incomes to USD 1.75/day (PPP). To bridge this poverty gap and achieve zero hunger, an additional investment of USD 3.7 trillion is needed, or USD 265 billion per year between 2016 and 2030. This annual spending includes USD 67 billion for social protection to address the immediate income shortfall and USD 198 billion for targeted investments in agriculture and rural development. These investments focus on improving primary agriculture, natural resources, agro-processing, infrastructure, institutional frameworks, and research and development. The increased income from these investments is expected to reduce reliance on social protection. Additionally, the report suggests that increased income levels should support more diversified diets, thereby meeting at least some nutritional needs beyond just ensuring adequate dietary energy. Thus, the report covers all SDG 2 targets as well as SDG 1. The total additional cost needed is USD 3.7 trillion. 

4. The billion-dollar reports 

Five of the eight reports are numbered in the billions and focus on SDG 2 (see Figure 3). Four of them - IFPRI (2019), ZEF (2024), Ceres2030 (2020) and IFPRI & IISD (2016)— aim to end hunger by focusing on direct investments in agricultural productivity, social protection, agricultural research and development (R&D), and infrastructure such as rural roads, storage facilities and irrigation systems. These interventions cover hunger and food security (SDG 2.1), agricultural productivity (SDG 2.3), and sustainable food production (SDG 2.4). However, the four reports do not include interventions specific to sub-target SDG 2.5 (genetic diversity) and, among them, only ZEF (2024) addresses interventions for SDG 2.2 (nutrition). The total additional cost in these reports ranges from USD 165 to 780 billion. Meanwhile, the fifth report, the Global Nutrition Report (2021) covers sub-target SDG 2.2 (nutrition), estimating a total additional investment of USD 86 billion.  

The billion-dollar reports end hunger by exclusively focusing on SDG 2 sub-targets 

Figure 3. Cumulative additional cost required to achieve SDG 2 targets (USD billion)  

Source: Author’s own based on IFPRI & IISD (2016), IFPRI (2019), Ceres2030 (2020), Global Nutrition Report (2021), ZEF (2024) 

IFPRI (2019) aims to reduce hunger in developing countries to 5 percent, and to 10 percent in Eastern and Central Africa, by increasing food availability and income, reducing commodity prices and offsetting the adverse impacts of climate change. Achieving these goals will require additional investments of USD 780 billion, or USD 52 billion per year, between 2015 and 2030 in areas such as agricultural research and development, irrigation, soil and water management, and transportation and energy infrastructure. 

ZEF (2024) aims to lift 700 million people out of hunger and malnutrition by 2030 through targeted investments in the most cost-effective interventions. The report identifies 10 key short-term interventions based on evidence from modeling reports and impact assessments. Using a marginal cost curve (MaCC) analysis, it estimates the additional costs for each intervention required to alleviate hunger and malnutrition. These interventions are ranked from least to most expensive based on their marginal costs, which is the cost per individual lifted out of hunger and malnutrition. Key interventions include investments in social protection, school feeding programmes, female literacy improvement, nutrition enhancement, crop protection and removing trade barriers. Implementing these measures will require an additional USD 512 billion, or USD 93 billion per year, between 2025 and 2030. 

Ceres2030 (2020) aims to end chronic hunger to a 3 percent prevalence of undernourishment, double the incomes of small-scale producers on average, and maintain agricultural GHG emissions to a level aligned with the Paris Agreement by 2030, targeting SDG 2.1, 2.3 and 2.4. 

To achieve these targets, the report proposes increasing public spending in 14 policy interventions. These interventions are categorised into three categories. “On the Farm” interventions aim to directly assist farmers, including provision of farm inputs, research and development, improved livestock feed and irrigation infrastructure. "Food on the Move" interventions target the reduction of post-harvest losses through measures such as storage improvement, enhancing returns from sales and supporting services offered by SMEs. Lastly, "Empower the Excluded" interventions focus on social protection and vocational training programmes.  

Implementing these interventions will require an additional USD 330 billions of public spending by 2030. Using the co-funding rule, the report calculates the difference between the donor and country share. The donor share of this amount is USD 140 billion while the remaining USD 190 billion will be sourced from low- and middle-income countries via increased taxation. This additional public spending is expected to attract USD 520 billion in private investment in primary and processed food sectors. 

IFPRI & IISD (2016) aims to end hunger to a five percent threshold by 2030 by investing in three interventions: (1) social safety nets supporting consumers through cash transfers and food stamps; (2) farm support to increase production and farmers’ incomes; and (3) rural development that reduces inefficiencies along the value chain and enhances rural productivity through investments in infrastructure, education, storage, market access and value chains. 

Implementing these interventions will require an additional USD 165 billion or USD 11 billion annually between 2015 and 2030. Of this annual spending, USD 4 billion is expected to derive from donor contributions, while the remaining USD 7 billion will be sourced from governments of developing countries. This will reduce the number of hungry people to 310 million in 2030. Further, the additional public spending is projected to stimulate an additional USD 5 billion in private investment per year, on average. The report outlines priority levels for donor spending by countries, determined by the severity of hunger. Africa, especially Central Africa and Southeast Africa, requires the most significant support due to the high prevalence of undernourishment in these regions. 

Global Nutrition Report (2021) aims to achieve significant global nutrition targets by 2025. These targets include a 40 percent reduction in child stunting, a 50 percent reduction in anaemia in women, a 50 percent increase in exclusive breastfeeding rates, and reducing child wasting to 5 percent or less. To meet these goals, additional investments are needed in areas such as micronutrient supplementation, the promotion of good infant and young child nutrition practices, and the fortification of staple foods. Implementing these interventions will require an estimated USD 86 billion, or USD 10.8 billion per year, between 2022 and 2030. Yet, the total economic gains to society of investing in nutrition could reach USD 5.7 trillion a year by 2030. 

5. The devil is in the detail  

The big picture helps to understand the broad differences between the different modelling exercises. But a closer look reveals that (1) the choice of targets and the quantification of those targets, (2) the baselines and the data sources used for baseline projections, and (3) the modelling approaches, lead to different estimates that are confusing policymakers and are sometimes even contradictory. 

5.1 The choice of targets and the quantification of those targets  

The reduction of the prevalence of undernourishment (PoU) is a core target for most of the reports. This is understandable since their focus is on achieving zero hunger (sub-target SDG 2.1) and the PoU is the main indicator to measure SDG 2.1. The PoU has a robust and historical dataset and is published annually in the State of Food Security and Nutrition in the World (SOFI). However, a closer look at the different modelling reports shows that the researchers choose reduction targets for the PoU of between 0 percent and 10 percent. This leads to wide differences in cost estimates, since reducing hunger down to zero percent is significantly costlier than leaving 3, 5 or 10 percent of the population in a country in chronic hunger.  

For example, FAO, IFAD, WFP (2015), ZEF (2024) and Food and Land Use Coalition (2019) aim to completely eradicate hunger, achieving zero percent PoU by 2030; Ceres2030 (2020) aims to reduce the PoU to 3 percent in each country; IFPRI & IISD (2016) and Laborde & Torero (2023) reduce the PoU to 5 percent in each country; and, IFPRI (2019) reduces the PoU to 5 percent in developing countries except in Eastern and Central Africa where the PoU is reduced to 10 percent or more. 

There are some important justifications for the different choices in the percentage. However, researchers poorly communicate both the percentage selected and the justification, leaving policymakers with the impression that the additional spending will achieve zero hunger, or confused about the implications of only reducing hunger to 3 or 5 percent.   

One of the reasons for choosing a 3 or 5 percent target is that the PoU does not actually count the number of people that are affected by hunger. Rather, it is a statistical estimate of the percentage of individuals in the population that are in a condition of undernourishment (SOFI, 2022). To compute the PoU in a population, the probability distribution of habitual dietary energy intake levels (expressed in kcal per person per day) for the average individual is modelled as a parametric probability (SOFI, 2022). Given that the PoU is an estimate based on probability of energy intake, the precision of the estimates is generally low, and we do not really know exactly how many people are chronically undernourished in a given country (SOFI, 2022). FAO believes the margins of error are expected to likely exceed 5 percent in most cases (SOFI, 2022). For this reason, FAO does not consider PoU estimates with results that are lower than 2.5 percent as sufficiently reliable to be reported in the SOFI report (SOFI, 2022). Therefore, having a PoU target of between 3 and 5 percent makes sense given the probability of the margin of error in the report. Nevertheless, researchers consistently fail to effectively communicate this reality with policymakers. 

The SDG 2.2 target on nutrition is another case in point. Once again, many of the reports include SDG 2.2 as a target. While the Global Nutrition Report (2021), Food and Land Use Coalition (2019), FAO, IFAD, WFP (2015) and Laborde & Torero (2023) all target SDG 2.2, they use diverse sub-targets and interventions in the models. This variation leads to significant differences in the costs associated with each model. For example, the Global Nutrition Report (2021) focuses on achieving the 2012 World Health Assembly (WHA) Resolution nutrition targets by 2025 which are: 40 percent reduction in child stunting; 50 percent reduction in anaemia in women; 50 percent increase in exclusive breastfeeding rates; and reduction in child wasting to 5 percent or less (WHO, 2012). Food and Land Use Coalition (2019), FAO, IFAD, WFP (2015) and Laborde & Torero (2023), on the other hand, use a target of diet quality, but define the target differently and achieve the target through different interventions. Laborde & Torero (2023) use a cost of a healthy diet target of UD 3.75 per person per day and achieve this target mainly through social protection interventions that increase incomes (Herforth et al, 2020). FAO, IFAD, WFP (2015) use a prevalence of extreme poverty target of USD 1.75 per person per day and achieve this target mainly through social protection interventions that increase incomes. Meanwhile, Food and Land Use Coalition (2019) uses a target of a "human and planetary health diet" definition; a primarily plant-based diet, aligning with the EAT Lancet Commission’s definition of a planetary health diet (EAT, 2019). 

5.2 The baselines and the projections used for the business-as-usual scenario  

Another reason for the differences in the cost estimates in the reports is the different baselines and projections used for the business-as-usual scenario. This is a bigger problem. For instance, each report uses a different baseline for the projected number of undernourished people in 2030 under a business-as-usual (BaU) scenario. Unfortunately, the source for the baseline projection is not always clearly detailed in the methodology.  

IFPRI (2019) and IFPRI & IISD (2016) use a baseline projection of around 600 million people undernourished in the business-as-usual scenario in 2030. Ceres2030 (2020) and FAO, IFAD, WFP (2015) have a baseline projection of around 660 million people. Meanwhile, ZEF (2024) estimates a baseline of 700 million people will be undernourished by 2030. The Food and Land Use Coalition (2019) projects a baseline of 475 million people. The differences in the extent to which hunger is eradicated and the baseline number of people projected to be undernourished in 2030 under a business-as-usual scenario lead to very different estimates of the number of people who will be lifted out of hunger due to additional investments (see Figure 4). 

The estimated number of people undernourished in 2030 in the reports varies from 475 million to 700 million while the number of people to be lifted out of hunger differs by a factor of nearly 4. 

Figure 4. Projected number of undernourished people by 2030 under a business-as-usual scenario and the number of people lifted out of hunger due to additional investment (in millions)  

Source:  Laborde & Torero (2023), Food and Land Use Coalition (2019), FAO, IFAD, WFP (2015), IFPRI (2019), ZEF (2024), Ceres2030 (2020), IFPRI & IISD (2016)  

Note: The Global Nutrition Report (2021) does not provide a projected number of undernourished people 

In some cases, these differences make sense. The number of hungry people changes every year, according to the annual SOFI report, which publishes the prevalence of undernourishment. Therefore, a report published in 2019 will have a different baseline compared with one published in 2024. But for reports published in the same year, such as Food and Land Use Coalition (2019) and IFPRI (2019), such significant differences highlight a problem. The problem is further compounded by the fact that not all reports use SOFI projections and, in some cases, do not provide the source for their baseline projections. 

Since 2022, SOFI has been publishing the business-as-usual projections to 2030 using the MIRAGRODEP CGE model. For future modelling exercises, researchers should agree to use the SOFI baseline and the 2030 business-as-usual projections for future modelling work on hunger and the SDGs.  

Another example is the use of different climate change scenarios. Representative Concentration Pathways (RCPs) are climate change scenarios to project future greenhouse gas concentrations, based on the Intergovernmental Panel on Climate Change (IPCC) Fifth Assessment Report (IPCC, 2014). These climate change scenarios were updated in the IPCC Sixth Assessment Report. The resulting Shared Socioeconomic Pathways (SSP), build on the RCPs and are climate change scenarios of projected global socioeconomic changes up to 2100 (IPCC, 2021). There are five pathways:  

• SSP1: A shift towards sustainable development and reducing inequality. 

• SSP2: A "middle of the road" scenario where trends generally follow historical trajectories. 

• SSP3: A fragmented world characterized by resurgent nationalism. 

• SSP4: A scenario of increasing inequality. 

• SSP5: A world experiencing rapid, unrestricted growth in economic output and energy consumption. 

Each of these pathways has different assumptions about future global development, which in turn impacts the estimated costs of the models. For example, ZEF (2024) uses SSP3 which projects a pessimistic future with high population growth, economic stagnation and significant income inequality. This scenario might result in higher costs due to the need to address more severe social and economic challenges to end hunger. In contrast, Food and Land Use Coalition (2019) and IFPRI (2019) use SSP2. This scenario assumes that social, economic and technological trends will continue to follow historical patterns, leading to more moderate costs as development progresses unevenly but steadily across regions. Some reports project severe climate impact scenarios which necessitate high interventions costs.  

IFPRI (2019) estimates the cost of the interventions needed to achieve SDG 2 targets based on the most pessimistic climate change scenario (RCP 8.5), where the planet's temperature could rise by up to 4.9°C above preindustrial levels. This scenario implies the need for higher expenditures on interventions to mitigate the adverse effects of climate change on agriculture and food security. In contrast, ZEF (2024) incorporates climate change impacts at a level consistent with RCP 6.0, where temperatures are projected to increase by 3-4°C above preindustrial levels by 2100. This scenario suggests significant intervention costs, though slightly lower than those associated with IFPRI (2019).  

There are multiple strong arguments for using different climate change scenarios in hunger-related modelling exercise, but researchers are confusing policy makers by not being upfront or clear about the assumptions and scenarios selected. Researchers need to be much clearer about the scenarios selected and the rationale to help policy makers make sense of the results. 

5.3 The different modelling approaches  

The reports use three different modelling approaches to estimate the costs: the computable general equilibrium model (CGE), the partial equilibrium model, and the marginal abatement cost curve (MACC). In some cases, there are a mix of model approaches. Using different model approaches and asking different modellers to answer similar questions is a good thing. All models are a simulation of a potential future world, none are perfect, each has their limitation, but they are all useful to support more evidence-based decision making.  

Using multiple different model approaches enables policy makers to have more confidence in the results, especially when they are coherent and well-aligned, or to provide a more complete answer to the question, since some models are better at simulating different things. For example, CGE models are typically better at simulating macro-economic scenarios, such as trade policy shocks, while partial equilibrium models are much better at simulating productivity and yield scenarios for crops or forests at the sub-national level.  

Often, the question being asked by policy makers determines the best model approach. If, for example, policy makers want to know what will happen to maize yields in Malawi as a result of climate change, it would be much better to use a partial equilibrium model, whereas if the policy maker is interested to know what will happen to trade in Africa as a result of the African Continental Free Trade Area (AFCFTA), then it would be preferable to use a CGE model. Policy makers often do not know which model approach is best to answer a question and modellers should educate policy makers on the usefulness of the different modelling approaches rather than trying to promote their own model approach as superior to others and capable of answering all questions.  

IFPRI & IISD (2016), Ceres2030 (2020), and Laborde & Torero (2023) use the MIRAGRODEP model, a dynamic multi-country, multi-sector Computable General Equilibrium (CGE) model, which incorporates extensive data on income levels, production opportunities, and consumption patterns at macroeconomic, regional, sectoral, and household levels. This comprehensive data integration enables a thorough analysis of household heterogeneity and socio-economic factors.  

By contrast, FAO, IFAD, WFP (2015) and IFPRI (2019) use partial equilibrium models. FAO, IFAD, WFP (2015) use a global partial equilibrium model approach focusing on food supply and demand projections. IFPRI (2019) uses the IMPACT model. However, this model does not account for the broader macro-economic implications. To overcome this limitation, IFPRI (2019) integrates IMPACT with the GLOBE-Energy model, a dynamic computable general equilibrium model. GLOBE assesses the macroeconomic effects of various agricultural productivity scenarios and feeds the resulting simulated income paths back into IMPACT. This integration enables a more comprehensive analysis of the economic spillovers associated with agricultural investments, leading to a detailed cost estimate.  

Food and Land Use Coalition (2019) and ZEF (2024) estimate the additional cost needed to end hunger based on mixed evidence-based interventions from various literature sources and expert assessments, resulting in variable estimates based on the scope of the interventions and assumptions. Food and Land Use Coalition (2019) uses data from GLOBIOM, a partial equilibrium model, to refine the cost estimates to its Better Future scenario while ZEF (2024) uses a marginal abatement cost curve (MACC) model to assess the cost effectiveness of various hunger reduction interventions. 

The Global Nutrition Report 2021 quantifies the cost of achieving the World Health Assembly targets based on unit costs data from actual programme financing needs, incremental program coverage, and target populations in each country. The unit costs data is supplemented by information from peer-reviewed publications, grey literature, national nutrition plans, and primary data collected by the World Bank. The report employs cost-benefit analyses to assess the monetary value of benefits relative to the costs of nutrition-sensitive interventions on stunting, wasting, anaemia, and breastfeeding targets (Shekar et al, 2017). 

6. Conclusion 

Knowing the additional investment needed to achieve SDG 2 by 2030 enables donors and governments to allocate resources efficiently and effectively towards their global commitments. This ensures that spending is directed towards the most effective programmes, maximising the impact of every dollar spent. Accurate cost estimates help prevent underfunding, which can lead to incomplete projects and unmet goals, and overfunding, which can result in wasteful spending and missed opportunities to address other critical needs.  

A review of eight recent modelling reports of the additional investment needed for achieving SDG 2 by 2030 reveals a wide range of estimates; from a total of USD 86 billion to over USD 4 trillion. Five of the eight reports are numbered in the billions because they focus on the SDG 2 sub-targets, while three reports are numbered in the trillions because they go beyond SDG 2 to include a range of other linked SDGs or include agri-food systems transformation. However, even within the five reports focused on SDG 2, the range is USD 86 billion to 760 billion. This is partly because of how many SDG 2 sub-targets are included and the types of interventions modelled.  

However, a closer look reveals that the devil can often be in the detail. The choice of targets and the quantification of those targets, the baselines and the data sources used for baseline projections, and the modelling approaches, lead to different estimates that are confusing policymakers and are sometimes even contradictory. The biggest problem is the confusion in the baselines and the data sources used for baseline projections. The eight modelling reports reveal significantly different numbers for the PoU and the projections of the PoU by 2030, even between reports published in the same year. This is a problem.  

For future modelling exercises on hunger and the SDGs, researchers should agree to use the SOFI baseline and the 2030 business-as-usual projections from the year that their report is published. Researchers need to be much clearer about the scenarios selected, particularly climate change, demographic and economic scenarios and their rationale to help policymakers make sense of the results. Using different model approaches and asking different modellers to answer similar questions is a good thing. But some models are better at answering some types of questions than others. Policymakers often do not know which model approach is best to answer a question and modelers could play a much better role in educating policymakers on the usefulness of the different modelling approaches rather than trying to promote their own model approach as superior to others and capable of answering all questions. Researchers need to agree on the data sources, baselines, and projections they use in their modelling exercises, and be much clearer about the reason for different choices of targets, interventions, and assumptions. Communicating more effectively and coherently with policymakers will facilitate a more optimal allocation of resources, making the goal of ending hunger more actionable and sustainable. 

References 

Daniel Mason-D'Croz, Timothy B. Sulser, Keith Wiebe, Mark W. Rosegrant, Sarah K. Lowder, Alejandro Nin-Pratt, Dirk Willenbockel, Sherman Robinson, Tingju Zhu, Nicola Cenacchi, Shahnila Dunston, Richard D. Robertson. (2019) Agricultural investments and hunger in Africa modeling potential contributions to SDG2 – Zero Hunger. World Development, Volume 116, Pages 38-53. [IFPRI 2019] 

EAT. (2019). Healthy Diets From Sustainable Food Systems Food Planet Health. Adapted summary of the Commission Food in The Anthropocene: the EAT-Lancet Commission on Healthy Diets From Sustainable Food Systems. The entire Commission can be found online at: thelancet.com/commissions/EAT. 

FAO, IFAD and WFP. (2015). Achieving Zero Hunger: the critical role of investments in social protection and agriculture. Rome, FAO. [FAO, IFAD, WFP 2015] 

Food and Land Use Coalition. (2019a). Growing better: Ten critical transitions to transform food and land use. The Global Consultation Report of the Food and Land Use Coalition September 2019. [Food and Land Use Coalition 2019] 

Food and Land Use Coalition. (2019b). Growing Better: Ten Critical Transitions to Transform Food and Land Use. Annex B: Technical Annex. The Global Consultation Report of the Food and Land Use Coalition September 2019. [Food and Land Use Coalition 2019] 

Global Nutrition Report. (2021). The state of global nutrition. Bristol, UK: Development Initiatives. [Global Nutrition Report 2021]  

Herforth, A., Bai, Y., Venkat, A., Mahrt, K., Ebel, A. & Masters, W.A. (2020). Cost and affordability of healthy diets across and within countries. Background paper for The State of Food Security and Nutrition in the World 2020. FAO Agricultural Development Economics Technical Study No. 9. Rome, FAO.  

High Level Political Forum on Sustainable Development. (2023). Food Systems Transformation: Transforming food systems for a sustainable world without hunger.  

IPCC. (2023). Climate Change 2023: Synthesis Report. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, H. Lee and J. Romero (eds.)]. IPCC, Geneva, Switzerland, 184 pp.  

IPCC. (2013). Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 1535 pp. 

Laborde, D., Bizikova, L., Lallemant, T. and Smaller, C. (2016). Ending Hunger: What would it cost? International Institute for Sustainable Development (IISD) and International Food Policy Research Institute (IFPRI). [IFPRI & IISD 2016] 

Laborde, D., Parent, M., & Smaller, C. (2020). Ending Hunger, Increasing Incomes, and Protecting the Climate: What would it cost donors? Ceres2030. International Institute for Sustainable Development (IISD) and International Food Policy Research Institute (IFPRI). [Ceres2030 2020]  

Laborde, D., Torero, M. (2023). Modeling Actions for Transforming Agrifood Systems. In: von Braun, J., Afsana, K., Fresco, L.O., Hassan, M.H.A. (eds) Science and Innovations for Food Systems Transformation. Springer, Cham. [Laborde & Torero 2023] 

Walters, Dylan & Shekar, Meera & Kakietek, Jakub & Eberwein, Julia. (2016). An Investment Framework for Nutrition: Reaching the Global Targets for Stunting, Anemia, Breastfeeding, and Wasting. World Bank Group. 

World Health Organization. (2012). Sixty-fifth World Health Assembly: Resolutions and decisions, annexes. Geneva, Switzerland.  

von Braun, J., Chichaibelu B.B., Laborde, D. and Torero Cullen, M. Cost of Ending Hunger – Consequences of Complacency and Financial Needs for SDG2 Achievement. ZEF – Discussion Papers on Development Policy No. 347, Center for Development Research, Bonn, May 2024, pp. 103. [ZEF 2024]