Consumption patterns create hidden costs: i) the health pathway, where unhealthy dietary patterns contribute to undernutrition and NCDs, leading to productive and healthy years of life lost; and ii) the social pathway, where distributional failures in food supply and insufficient revenues for agrifood workers lead to undernourishment, as discussed in Chapter 1. The resulting hidden costs permeate all impact domains – environmental, socioeconomic and health – creating an interconnected web of effects. In addition, each of these impact domains is also affected by other non-diet-related pathways, such as the inappropriate use of pesticides in primary production, leading to biodiversity loss, occupational hazards and poor health outcomes. This chapter, however, focuses on how shifts in consumption patterns can drive agrifood systems transformation by exploring the links between diets and these interconnected impacts.
From an environmental perspective, researchers agree that achieving sustainability in agrifood systems requires more than just transforming production methods.20 A report by the EAT-Lancet Commission highlights diets as pivotal in transforming agrifood systems beyond their health impacts,21 and the conclusions linking diets and the environment are shared by other research.22–25 While the study acknowledges other agricultural measures to lessen adverse effects, it argues that sustainable agrifood systems cannot be achieved solely through improvements on the production side. Research shows that dietary shifts, such as reducing animal product consumption in countries where it is excessively high, can significantly lower GHG emissions and mitigate other environmental harms, such as biodiversity loss, land-use change and nutrient runoff.25–33
Such discussions tend to be mired in controversy, as they are based on historical consumption patterns in industrialized or transition countries that have led to significant environmental damage globally due to the interconnected nature of agrifood systems. This raises questions about fairness with regard to distributional issues among cost producers and cost bearers. Consequently, it is important to recognize the heterogeneity of dietary quality around the world; moreover, in some places, the consumption of animal products needs to increase to achieve a nutrient-adequate diet and the burden of countering the current environmental damage cannot be equally distributed.
In Bangladesh, a country with traditional agrifood systems, a study on the potential transition to healthier diets reveals trade-offs between environmental, socioeconomic and health indicators.34 The study compared the transition from current diets (high in animal products and sugars and low in vegetables, fruits, legumes and nuts) to the EAT-Lancet diet or a diet based on the food-based dietary guidelines between 2022 and 2050. A diet with more plant-based protein and fewer staples was found to have positive health effects, as well as positive impacts on most of the environmental footprint indicators. However, there are environmental trade-offs between the EAT-Lancet and FBDG diets, with the former leading to higher land and phosphorus use and the latter to a greater rise in GHG emissions. In socioeconomic terms, the FBDG diet scenario scored best on national self-sufficiency objectives, cereal affordability and low-skilled wages, underlining the importance of customizing global guidelines to local needs and national priorities.
The various socioeconomic and health impacts associated with consumption speak to the multifaceted nature of malnutrition, which encompasses both insufficient and excessive intake of nutrients, a lack of balance in essential nutrient levels, and hindrances to nutrient utilization due to repeated instances of disease.4, 35 Malnutrition manifests itself as undernutrition – being underweight for one’s age, too short for one’s age (stunted), dangerously thin for one’s height (wasted), or deficient in vitamins and minerals (suffering from micronutrient deficiency) – as well as overweight and obesity.4 Many countries are facing a double burden of malnutrition, where undernutrition coexists with overweight, obesity or diet-related NCDs.36, 37 While the prevalence of the double burden of malnutrition decreases across agrifood systems types (from 70 percent in countries with protracted crisis and traditional agrifood systems to 27 percent for expanding agrifood systems, and zero for formalizing and industrial agrifood systems), that of adult obesity and overweight increases (from 30 percent to around 60 percent).37
Unhealthy diets are ubiquitous across all weight categories. Individuals with a healthy weight may consume diets that are low in healthy foods and high in unhealthy foods or nutrients (for example, high in sodium). Meanwhile, individuals with overweight and obesity may consume a healthy diet. As a result, their weight may be more responsive to other factors (such as changes in lifestyle habits). Using 24-hour dietary recall data from Ethiopia, Mexico and the Philippines – countries with different types of agrifood systems – Box 23 presents a case study on how the dietary quality of these populations is associated with NCD risk and weight.
Box 23Assessing diet quality through 24-hour recalls and associations with overweight and obesity and diet-related non-communicable disease risk factors
While the Global Burden of Disease data provide an appropriate approach for estimating global trends based on disability-adjusted life year (DALY) estimates, there are some limitations to the included dietary data, which draw on several national and household sources as noted in Box 4. Where data exist, hidden costs can be estimated based on individual dietary risk factors using robust data of dietary intake. The Global Diet Quality Score (GDQS) is a comprehensive measure of diet quality, validated against nutrient inadequacy and selected diet-related non-communicable disease (NCD) risks using nationally representative surveys.38 Such national surveys are particularly appropriate for the use of true cost accounting to inform policy options in specific targeted contexts. The case study presented here provides a detailed analysis of the associations between diets, overweight and obesity and diet-related NCD risk factors* by constructing the GDQS using individual 24-hour dietary intake data for adults over the age of 20 from nationally representative nutrition surveys for Ethiopia, Mexico and the Philippines.39–41
The results indicate that most of the adult population in all three surveys have medium and high NCD risks, largely due to very limited consumption of healthy food groups protective against NCD risks (Figure A). Ethiopia, in the protracted crisis agrifood systems category, has a smaller share in the high-risk category than the other two countries; however, forthcoming results from a 2021/22 survey indicate that there has been a significant increase since the 2011 survey. As shown in Figure A, Ethiopia has overall the highest score on the GDQS, largely due to very low consumption of unhealthy food groups (GDQS−), despite a very limited variety of healthy food groups consumed (GDQS+). As agrifood systems develop, the variety and quantity of healthy food groups consumed may increase, but this is often offset by greater increases in variety and quantity of unhealthy food groups. This is consistent with low rates of overweight and obesity in Ethiopia (7.2 percent in 2011), and higher rates in Mexico (71 percent in 2012) and the Philippines (31.1 percent in 2013) – the latter having agrifood systems categorized as expanding and the former as diversifying.
FIGURE A The Global Diet Quality Score and its submetrics by country
SOURCE: Authors’ own elaboration.
The analysis shows some limitations of using overweight and obesity and diet-related NCD risk factors as proxies when calculating health hidden costs. First, within countries, groups of individuals who are overweight or obese do not have lower quality diets than those with healthy weights (as demonstrated by patterns of diet-related NCD risk factors in Figure B). Second, using national survey data from the Philippines, regressions of fasting blood glucose and of blood pressure as dependent variables on diet quality (controlling for body mass index and sociodemographic variables) indicated that the effects of GDQS were generally small and not significant, potentially due to the use of cross-sectional data (that is, reverse causality) and single-day dietary intake assessment (which may limit the potential to understand the direction of causal relationships).
FIGURE B Nutrient inadequacy and diet-related non-communicable disease risk using Global Diet Quality Score by body mass index category, for Ethiopia, Mexico and the Philippines
Estimating the cost implications of unhealthy diets highlights the urgency of addressing diet quality. Already in Ethiopia, overweight or obesity in adults aged 20–49 years rose from 7.2 percent in 2011 to 12.0 percent in 2023, and according to the results of the 2021/22 survey, diet quality has worsened (as measured by overall GDQS).
To guide agrifood systems policies that enable access to and consumption of healthy diets, nationally representative surveys provide critical insights into the causal relationships between dietary patterns and health outcomes and their related hidden costs. To improve these estimates and their potential to inform policy options, better evidence is needed on the impacts of measures to increase the production of and access to healthy diets, as well as those measures aimed at regulating food intake so as to moderate or avoid foods high in sugars, salt and fats and foods high in energy but low in nutrients such as fibres and micronutrients.
The social hidden costs of undernourishment – the condition in which an individual’s habitual food consumption is insufficient to provide the amount of dietary energy required to maintain a normal, active, healthy life42 – are significant. The 2024 edition of The State of Food Security and Nutrition in the World estimates that between 713 million and 757 million people in the world may have faced hunger in 2023.4 While the measurement of undernourishment pertains to the total population, special consideration needs to be given to the nutritional status of children. Children that suffer from undernutrition, particularly before the age of five, face profound and lasting impacts on their physical and cognitive development.43, 44 Worldwide, in 2022, an estimated 148.1 million children under five years of age (22.3 percent) were stunted, 45 million (6.8 percent) were wasted and 37 million (5.6 percent) were overweight.o, 4
A methodology applied by the World Food Programme, known as “Cost of Hunger”, estimates the social and economic impacts of child undernutrition, focusing on the health, education and labour sectors.45, 46 While the approach includes a wider range of costs than those hidden from market transactions, as outlined in The State of Food and Agriculture 2023, the results highlight the cross-sectoral need for early childhood nutrition interventions. Box 24 summarizes results from several African and Latin American countries, underlining how they complement the estimates on the hidden costs of undernutrition in this report.
Box 24The Cost of Hunger methodology for Africa and Latin America
A number of studies reveal the extensive economic toll of child malnutrition in Africa and Latin America. Spanning 21 African nations from 2013 to 2018, the Cost of Hunger in Africa research by the African Union and the World Food Programme (WFP) delves into the profound health consequences of stunting and underweight in pre-school-age children, shedding light on the cascading losses in terms of education, health care and workforce productivity.45 The same methodology, but considering low birth weight and underweight, was also applied in some Latin American countries in a parallel study by the Economic Commission for Latin America and the Caribbean and WFP in 2009.46 It is crucial to recognize that the methodology diverges significantly from the methodology used in this and the 2023 edition of The State of Food and Agriculture, so findings should not be compared with the hidden costs of undernourishment, but serve as complementary insights.
The first major difference is the type of undernutrition and the populations considered. Whereas the hidden costs of undernourishment in both the 2023 and the 2024 edition of this report refer to the total population experiencing insufficient food intake, the Cost of Hunger methodology examines the incidence of underweight and stunting before the age of five. The Cost of Hunger includes an “incidental retrospective dimension” for assessing current-year economic burdens of undernutrition for people who were underweight before the age of five.
The Cost of Hunger approach results in cost estimates that are significantly higher than those quantified in this report (on average, around ten times greater), mainly because its analysis of the hidden costs of undernutrition consider the additional negative effects of undernutrition. These negative effects include increased risk of pathology (for example, respiratory disease and malaria), impact of reduced education attainment on productivity, and lower productivity in manual labour. The figure provides an overview of these results across the studies in Africa and Latin America. It also includes health care costs – unlike The State of Food and Agriculture, which focuses on hidden costs. The monetization of the hidden costs is another difference: while The State of Food and Agriculture monetizes lost productivity using average gross domestic product per worker, the Cost of Hunger research tends to use average wages or the minimum wage.
The Cost of Hunger studies highlight that, while the most direct consequence of undernutrition is the morbidity and mortality caused, accounting for the indirect effects on the health, education and productivity of workers can yield much higher estimates of the economic costs of undernutrition. This is a relevant insight for future researchers aiming to motivate policy action.
FIGURE Proportion of cost category by country
While unhealthy diets are a common cause of all forms of malnutrition,47–49 many other direct (health and care) and indirect (for example, poverty, and health and education services) causes exist simultaneously. Actions to address all forms of malnutrition must address these in parallel. Box 25 draws on evidence from Ethiopia and the Philippines on how public nutrition and health interventions play a vital role, complementary to interventions to enable healthy diets, while Box 26 explores the role of agrifood systems in creating an enabling environment to support breastfeeding for better infant and young child outcomes. Governments have a role to play in inspiring and empowering consumers by creating an enabling environment for change and raising awareness on the role of healthy diets in driving broader societal goals. These complement governments’ role in shaping food environments by setting incentive structures in food supply chains within a systems approach, as discussed in Chapter 3.
Box 25Healthy diets are essential, but not sufficient to eliminate stunting
A study using the Lives Saved Tool assessed the potential impact of dietary, public nutrition and health interventions on significant rates of child stunting in Ethiopia and the Philippines from 2024 to 2030.50
The findings reveal that while a healthy diet could avert an estimated 14 percent of child stunting in Ethiopia and 9 percent in the Philippines, it is not enough in and of itself (see the figure). When combined with full population coverage of essential public nutrition interventions, these percentages increase to 24 and 17 percent, respectively. A further reduction in stunting can be achieved by scaling up public health interventions in areas such as water, sanitation and hygiene practices, antenatal care and immunization. Notably, the impact varies due to differences in coverage of existing public health interventions, which are much higher in the Philippines.
FIGURE Potential for averting stunting through full population coverage by intervention type in Ethiopia and the Philippines
SOURCES: Black, R.E., Victora, C.G., Walker, S.P., Bhutta, Z.A., Christian, P., de Onis, M., Ezzati, M. et al. 2013. Maternal and child undernutrition and overweight in low-income and middle-income countries. The Lancet, 382(9890): 427–451. https://doi.org/10.1016/S0140-6736(13)60937-X; Johns Hopkins & Bill & Melinda Gates Foundation. 2024. The Lives Saved Tool (LiST). In: The Lives Saved Tool. [Cited 21 March 2024]. https://www.livessavedtool.org
The simulations underscore the critical importance of the first 1 000 days from conception to age two years in preventing stunting, emphasizing the necessity of early food systems to ensure access to a healthy diet. However, the study highlights that a healthy diet alone cannot fully address child stunting; essential public nutrition and health interventions are essential complements.
