Studies and Reports of the Month: July 2026

This month the evidence points at two absences: what is missing from diets, and what is missing from policy. Underconsumption drives 63% of diet-related life loss in Germany, and youth plant intake has risen since 1990 yet sits at just 2.8 servings a day. On the policy side, 70% of agricultural support funds cheap staples, input suppliers capture 98% of Canadian farm revenues, and WHO prices cancer at US$33.2 trillion without naming a food policy instrument.
Studies
1. “Food Systems Transformation Would Reshape Global Agriculture” Nature — A ten-model ensemble finds that healthier diets, higher productivity and halved food waste would cut the value of global livestock production 60% below business-as-usual by 2050, while the value of vegetable, fruit, nut and legume production rises. 2. “Dietary environmental impacts and quality differ by socio-demographic characteristics” Nature Food — Across 165 countries, diet-related footprints vary up to fivefold within a single country and about twofold at the global median, with plant-rich diets consistently driving the lowest environmental impacts and highest quality. 3. “Global, Regional and National Intake of Plant-Based Foods Among Youth in 185 Countries (1990-2018): Findings from the Global Dietary Database” BMJ Global Health — Youth plant intake rose between 1990 and 2018 but still averages 2.8 servings a day, and high-income countries now record the lowest intake of any world region for ages 2 to 19. 4. “Global Dietary Estimates for Conducting Health, Environmental and Economic Impact Assessments” Nature Food — A new, open-access database reconciles discrepancies across global dietary datasets that swing annual diet-related death models by 1.5 million people. 5. “Impacts of Food Consumption on Life Expectancy and Hidden Health Costs” Cleaner Food Systems — A product-level German framework attributes 11.8 days/year of lost life expectancy (€3,774/capita in hidden costs) to diet, with 63% driven by underconsuming healthy foods vs. 37% from overconsuming harmful ones. 6. “Why Farms Must Expand Production to Survive: Cost-Price Squeeze and the Farm-Level Growth Imperative in Canadian Agriculture” Cleaner Food Systems — A 411% capital spike alongside oligopolistic market capture forces Canadian farms into a growth treadmill, explaining why voluntary sustainability policies fail against structural economic pressure. 7. “Poultry consumption and chronic disease risk: evidence from substitution analyses” Academia Nutrition and Dietetics — A 33-study synthesis frames poultry as a transitional food rather than an optimal endpoint: swapping red meat for chicken lowers cardiometabolic risk, but swapping poultry for legumes/grains lowers it further. 8. “Eco-focused menu labels on full meal orders from fast-food restaurants: a randomized clinical trial” JAMA Health Forum — Traffic-light climate menu labels cut fast-food meal emissions by 15% to 30% at no added consumer cost, but require pairing with structural procurement standards to improve nutritional quality. 9. “Preliminary Trial Findings from Harvest Share: A Cost-Offset Community Supported Agriculture Model to Improve Diet in New York City” International Journal of Behavioral Nutrition and Physical Activity — A $360-$375/season culturally tailored CSA subsidy buffered low-income NYC households against a citywide inflation drop in produce intake, increasing dietary variety rather than total volume.
Reports
Studies
1

“Food Systems Transformation Would Reshape Global Agriculture”

Gibson, M. et al. Nature (2026) · DOI: 10.1038/s41586-026-10775-2
Key Takeaway

Adopting EAT-Lancet 2.0 diets, boosting agricultural productivity, and halving food loss by 2050 would cut global livestock value by 60% ($1.3 trillion) below business-as-usual, shrink ruminant herds by 400 million, and trigger the largest land contraction in over 2,000 years.

About

A 10-model economic ensemble under AgMIP compares a business-as-usual (BAU) trajectory to a transformation scenario (EL2) combining the 2025 EAT-Lancet reference diet, supply-side efficiency gains, and a 50% food waste reduction across 13 regions. The paper reports results for production, land use, prices, emissions, water and fertiliser use, employment, and the value of production, disaggregated across sectors and macro-regions. Historical land use is reconstructed back more than 2,000 years to place the projected changes in long-run context.

Key evidence
  • $1.3 trillion livestock value shift. Livestock production value falls 60% below BAU (ruminants drop 81% below). Plant-based output (fruits, vegetables, nuts, legumes) expands from 34% under BAU to 58% of global agricultural value under transformation.
  • 400 million fewer ruminants. Ruminant meat output drops 53% relative to BAU, returning ruminant animal numbers to 1996 levels. Non-ruminant slaughter falls by 11 billion animals. Dairy production declines 27%, non-ruminant meat production 45%.
  • 2,000-year land contraction. Total agricultural land contracts by 274 Mha. Grazing land shrinks at 0.40% per year, 2.5x faster than the 2000-2020 rate. Direct agricultural non-CO2 emissions drop 34% and land-use emissions drop 76% below BAU.
Implications for food systems transformation

For financial analysts the useful output is not the central estimate but the spread: ten independent models put livestock production value between 49% and 83% below business as usual by 2050, which is a wide but bounded transition-risk range for a sector that currently carries none.

As other research strongly supports, current tools (e.g. sugar taxes, labelling and advertising restrictions) are insufficient on their own. Halving the livestock market while doubling plant foods requires dedicated policy to manage rural economic transitions and livestock sector employment declines (-36% vs. BAU).

Restructuring would carry substantial consequences for producers, agricultural employment and rural economies, requiring deliberate policy design to manage the transition. Regional heterogeneity is pronounced: gains in production value (vs. 2020 levels) in Sub-Saharan Africa (+60%) and India (+40%) while China (-42%) and Brazil (-35%) see declines, reflecting their positions as major animal feed producers. Global achievement of healthy diets would prevent ~15 million adult deaths per year.

Gaps and next steps

Assumes a costless consumer shift to EAT-Lancet diets without accounting for cultural preferences, affordability, or accessibility. Excludes aquatic foods and does not model specific policy mechanisms.

← Back to contents
2

“Dietary environmental impacts and quality differ by socio-demographic characteristics”

Taherzadeh, O. et al., Nature Food (2026). · DOI: 10.1038/s43016-026-01389-y
Key Takeaway

Across 165 countries healthier diets are consistently lower-impact: each 5% gain in diet quality is associated with a fall in emissions, land use and water footprint. The highest diet-related emissions belong to young, highly educated urban men, so higher education does not translate into lower-impact diets.

About

A harmonized model linking individual intake data, a synthetic global population and supply chain accounts to estimate diet-related climate, water and land footprints alongside diet quality, scored using the Alternative Healthy Eating Index (AHEI). Covers 72 socio-demographic subgroups defined by sex, age, education and urban or rural residence across 165 countries.

Key evidence
  • Up to a fivefold gap inside the same countries. As a median across all 165 countries, highly educated urban men aged 15-34 record the highest diet emissions at 5.66 kgCO2e/day, vs. 2.85 for low-education rural women aged 35-54. Within individual countries the spread runs wider, exceeding fivefold in Kenya but narrowing to roughly 1.8x in Germany.
  • Co-benefits with climate. Each 5% rise in AHEI score is associated with 7.7% lower greenhouse gas emissions, 7.2% lower land footprint and 4.4% lower water footprint. No trade-off appears between diet quality and impact.
  • Countries cluster, so uniform strategies misfire. Diet quality and environmental impact combine into distinct national patterns, which the authors read as evidence that targeted subgroup strategies will outperform single national messages.
Implications for food systems transformation

Education may not produce lower-impact diets. Targeting the highest-emitting subgroups is sharper than uniform national campaigns, and the health case and the environmental case point the same direction rather than competing.

Cross-references

Springmann (entry 4). Taherzadeh draws intake from the Global Dietary Database, the same survey source Springmann shows over-reports meat and inflates modelled pasture use 22% above FAO land statistics. Energy adjustment addresses part of that bias, so the subgroup rankings here are firmer than the absolute footprint levels.

Gaps and next steps

Intake comes from survey data with documented over-reporting of meat, partly addressed here through energy adjustment. Footprints are modelled rather than measured. Subgroup medians describe populations, not individuals.

← Back to contents
3

“Global, Regional and National Intake of Plant-Based Foods Among Youth in 185 Countries (1990-2018): Findings from the Global Dietary Database”

Yearley, S. et al. BMJ Global Health (2026). · DOI: 10.1136/bmjgh-2025-021543
Key Takeaway

While healthy plant-food consumption among youth rose between 1990 and 2018, they still average just 2.8 servings a day, and legume and fruit intake did not move. High-income countries now record the lowest intake of any world region for ages 2 to 19, at 1.89 servings, after peaking at 3.77 in infancy.

About

The first global analysis of youth intake across five healthful plant-based food groups. Bayesian hierarchical modeling of 1,248 mostly nationally representative surveys across 185 countries (1990-2018) tracking youth intake across 5 healthful plant food groups (fruits, non-starchy vegetables, legumes, nuts/seeds, non-potato starchy vegetables). Estimates are stratified by age, sex, urban/rural residence, and household education, adjusted to age-specific caloric references. Funded by the Gates Foundation and NIHl; the senior author reports extensive commercial affiliations.

Key evidence
  • Severe global deficit. Global youth consumed just 1.43 servings/day at \<2 years and 2.80 servings/day at 2-19 years. Intake spans a ~4x range across populous nations, from Spain (1.35) and UK (1.71) to Vietnam (4.28), DRC (4.38) and Mexico (5.18).
  • High-income reversal. In wealthy nations, intake peaks at \<1 year (3.77 servings/day) and collapses by mid-childhood (1.71 at ages 5–9). In these countries, urban and higher-educated households consume fewer plant foods than rural/lower-educated peers.
  • Gains, but in one food group. Total intake rose between 1990 and 2018, globally and in every region except South Asia. Almost all of it came from non-starchy vegetables (+0.44 servings/day) and nuts/seeds (+0.15). Legumes and fruit stayed flat, and starchy vegetables declined.

Implications for food systems transformation.

Nearly three decades of flat legume intake is not what successful information-based intervention looks like. Progress requires changing institutional defaults: mandating plant-rich options in school lunches, public procurement, and early childcare settings. Emphasis needs to shift from information-based interventions toward food environment redesign.

Cross-references

SOFI 2026 (entry 10). Beans and legumes are the one plant food group consumed more in lower-education households globally, which the authors attribute to affordability and nutrient-to-price advantage. SOFI finds perishable nutrient-dense foods drive 70% of healthy diet cost while starchy staples deliver half the basket calories at one-sixth the cost. Read together, the cheapest protective food group is also the one that has not moved in 28 years, which points at availability, not price, as the binding constraint on legumes.

Gaps and next steps

Data ends in 2018 (pre-COVID/price shocks) and does not identify specific causal mechanisms behind high-income demographic reversals.

← Back to contents
4

“Global Dietary Estimates for Conducting Health, Environmental and Economic Impact Assessments”

Springmann, M. (2026). Nature Food · DOI: 10.1038/s43016-026-01388-z
Key Takeaway

The datasets researchers rely on to measure what the world eats disagree enough to swing the modeled global death toll from unhealthy diets by roughly 1.5 million people in a single year. The new Global Dietary Database for Impact Assessments (GDD-IA) reconciles survey, supply, and biophysical data into a reliable baseline for food policy.

About

This study introduces the open-access Global Dietary Database for Impact Assessments (GDD-IA), a new global dataset built specifically for assessing the health, environmental, and economic effects of diets. It combines three imperfect sources: national food availability figures from the UN Food and Agriculture Organization (adjusted for food waste), individual dietary surveys, and estimates of energy requirements based on measured body weight, height, and physical activity. It covers 43 food groups across 171 countries from 1990-2020, broken down by age, sex, and urban or rural residence.

Key evidence
  • 1.5 million death swing. Using unadjusted dietary surveys overreports intake, while raw FAO supply data underestimates it, swinging annual global death estimates by 1.5 million.
  • It changes the environmental answer. Overreported meat consumption in surveys inflated modeled pasture use by 22% above actual FAO land statistics. GDD-IA resolves this error and matches the official figures once food loss and waste are accounted for.
  • High-income diets remain heavily animal-based: In 2020, high-income diets contained 90–130% more meat and dairy and 40% fewer legumes and grains than global averages.
Implications for food systems transformation

Provides policymakers with an accurate, open-access foundation for designing evidence-based dietary guidelines, fiscal policies, and emissions targets.

Gaps and next steps

Relies on older food waste (2010) and survey data (2018). Excludes data-poor nations outside the 171 covered countries.

← Back to contents
5

“Impacts of Food Consumption on Life Expectancy and Hidden Health Costs”

Oebel, B., Stein, L., Kleemann, S. & Gaugler, T. Cleaner Food Systems (2026). · DOI: 10.1016/j.clfs.2026.100053
Key Takeaway

Underconsuming protective, plant-based foods drives 63% of diet-related life expectancy loss in Germany, compared to 37% from overconsuming harmful foods. Current diets cost Germans 11.8 days of life per year and €3,774 per capita in hidden health costs.

About

A new product-level framework linking Global Burden of Disease risk data, national cost-of-illness metrics, and EAT-Lancet benchmarks to convert dietary risks into exact hours of life gained or lost per kilogram of food consumed in Germany. Recommended intakes are drawn primarily from the EAT-Lancet Planetary Health Diet.

Key evidence
  • 11.8 days lost annually. Average German dietary patterns cause a loss of 11.8 days of life expectancy per person per year (~2.6 years over a lifetime), led by cardiovascular disease (8 days per year), followed by neoplasms (2.1 days) and type 2 diabetes (1.6 days).
  • Product-level impact (per kg): Red meat: -1.9 hours. Processed meat: -1.7 hours. Legumes: \+0.7 hours. Whole grains: \+0.5 hours.
  • Deficit beats excess. Low whole grain intake (1.8 days lost) harms population health more than high processed meat intake (1.6 days lost). In aggregate, underconsumption of protective components accounts for about 63% of the per capita loss and overconsumption of harmful components for about 37%.
  • €3,774 hidden cost: National diet-attributable costs reach €314 billion annually (per capita €567 direct healthcare/productivity; €3,207 monetized mortality).
Implications for food systems transformation

Public health strategies should pivot from restriction-focused messaging (“eat less meat”) toward active substitution (“eat more protective plant foods”) and leverage health insurance incentives.

Gaps and next steps

Point estimates omit confidence intervals. Analysis excludes non-communicable diseases like chronic kidney disease and relies on a high valuation per DALY averted (€99,000). Findings reflect average population-level dietary patterns and do not capture heterogeneity by age, sex, socioeconomic position, or baseline health. The 63/37 decomposition is specific to the current German misconsumption pattern relative to Planetary Health Diet benchmarks. The framework is transferable but the numeric findings are not.

← Back to contents
6

“Why farms must expand production to survive: Cost-price squeeze and the farm-level growth imperative in Canadian agriculture”

Chevalier, H. Cleaner Food Systems (2026). · DOI: 10.1016/j.clfs.2026.100040
Key Takeaway

Concentrated input suppliers captured 98% of Canadian farm market revenues from 1986-2007, trapping producers in an imperative to continuously expand capital and output simply to stay solvent. Voluntary corporate sustainability programs consistently fail against this structural constraint.

About

Integrates political economy and social metabolism with 1976–2021 Canadian financial data and a 1996–2024 longitudinal study of Quebec cranberry farming to explain agricultural growth dynamics.

Key evidence
  • 411% capital escalation. Average Canadian farm capital rose 411% (to $3.08M) between 1976 and 2021. Farmers generated $820B in gross output from 1986 to 2007, but input suppliers captured 98% of market revenues.
  • Oligopoly structures squeeze farm margins on both ends. Canada’s five largest grocery chains capture close to 80% of food retail sales, two firms virtually monopolize beef slaughter, and the top four firms control ~70% of hogs. Upstream, the top four firms hold 50-60% of the global commercial seed market, ~70% of the pesticide market, and 45% of farm equipment sales, while the top five control 70-90% of global grain trade.
  • Subsidy inflation. Each $1 of per-acre government support inflates farmland values by $8–$72, worsening mortgage debt and forcing output expansion.
  • Quebec cranberry case. Production expanded 23-fold (destroying 24% of local peatlands), demonstrating how market dynamics coerce environmental destruction.
Implications for food systems transformation

Voluntary eco-certifications cannot break the treadmill. Policy must tackle input/retail monopolies, redirect subsidies away from output expansion, and protect producers from price squeezes.

Gaps and next steps

Single-author synthesis without formal systematic review methods. Uses aggregate national financial averages that mask within-province variation by farm size, commodity, and tenure. Does not model policy counterfactuals or quantify what alternative regulatory, fiscal, or land-tenure arrangements would deliver.

A key underdeveloped question is whether the growth imperative applies with equal force to feed-crop-dependent livestock production and to direct-human-food crop production, and whether reallocating cropland from feed to food would ease or intensify farm-level constraints.

Cross-references

Kortleve et al. (2026) on stranded EU agricultural assets (January issue). This paper supplies the farm-level mechanism, a coerced growth imperative, that helps explain why voluntary transition is structurally difficult even where it would be economically rational at the sector level.

Foodrise CAP subsidies analysis (February issue). Extended here by Canadian empirical data showing that each additional dollar of per-acre subsidy support raises farmland value by $8 to $72, demonstrating how public funding inflates land values and incentivizes intensification vs. sustainability.

Yang et al. (2026, Science) on midstream actor power (April issue). Connected here through the cost-price squeeze mechanism, which translates midstream oligopoly power directly into on-farm expansion pressure.

← Back to contents
7

“Poultry consumption and chronic disease risk: evidence from substitution analyses”

Goldman, D. & Nagra, M. Academia Nutrition and Dietetics (2026). · DOI: 10.20935/AcadNutr8436
Key Takeaway

Replacing red or processed meat with chicken lowers cardiometabolic risk, but replacing chicken with nuts or whole grains lowers it further, and legumes lower diabetes risk specifically. Poultry is a transitional step on the protein gradient, not the endpoint.

About

A review of 33 prospective cohort studies, meta-analyses and one randomized trial, identified through PubMed to March 2026, using poultry as the reference comparator across cardiovascular disease, coronary heart disease, type 2 diabetes and all-cause mortality. Substitution models estimate the effect of exchanging one food for an equal amount of another.

Key evidence
  • Chicken beats red meat. Replacing 100 g/day of processed meat with poultry was associated with 29% lower CVD risk and 16% lower all-cause mortality. Substituting for total red meat: 17% lower CHD, 24% lower stroke.
  • Plants beat chicken. Replacing a daily serving of poultry with nuts was associated with 9% lower CVD and 14% lower CVD disease risk. For type 2 diabetes: whole grains 13% lower, nuts 10%, legumes 5%.
  • The one trial found no chicken advantage. In a controlled-feeding RCT (n=113), lean poultry and red meat raised LDL and apoB equally. Both were worse than legumes, nuts, grains and soy, with diets matched for saturated fat and fiber.
  • The benefits don’t travel. Cohorts in other countries showed null or reversed CVD effects when poultry replaced red meat, which the authors attribute to baseline poultry intake, high-heat preparation methods and/or local meal accompaniments (e.g., refined starches vs. vegetables).
Implications for food systems transformation

Dietary guidelines treat poultry as a healthy protein endpoint. The evidence places it mid-gradient. Procurement standards and menu defaults that stop at chicken capture only part of the available health gain.

Gaps and next steps

This is a narrative review. Observational evidence rated low to very low certainty; the single feeding trial ran four weeks. Poultry is treated as one undifferentiated exposure regardless of cut, processing or cooking method. Cohorts are almost entirely US and European.

← Back to contents
8

“Eco-focused menu labels on full meal orders from fast-food restaurants: a randomized clinical trial”

Wolfson J.A., et al. JAMA Health Forum (2026). · DOI: 10.1001/jamahealthforum.2026.2108
Key Takeaway

Traffic-light climate labels on fast-food menus cut meal-level greenhouse gas emissions by 15-30% at no extra cost to consumers, but had virtually no effect on nutritional quality.

About

A randomized controlled trial of 6,210 US adults ordering from hypothetical fast-food menus across 5 labeling conditions (control, low-impact, high-impact, traffic-light, grade A–F).

Key evidence
  • 15-30% emissions drop. Traffic-light labels cut meal emissions by 15.2% at burger restaurants and 29.8% at sandwich shops vs. control.
  • High noticeability. Traffic-light labels outperformed all other designs: 78% of participants noticed the traffic-light label (vs. 43% for the control group) and 89% correctly understood their meaning.
  • Zero cost / zero health gain. Customer meal spend did not increase. Nutritional healthfulness scores improved by \<1 point on a 100-point scale.
Implications for food systems transformation

Traffic-light climate labels are an effective, cost-neutral demand-side tool for cutting greenhouse gases. However, improving diet health requires direct menu reformulation and procurement rules.

Gaps and next steps

Tested in an online hypothetical ordering environment without real money. Does not track choice persistence over repeat exposures.

← Back to contents
9

“Preliminary Trial Findings from Harvest Share: A Cost-Offset Community Supported Agriculture Model to Improve Diet in New York City”

LeCroy, M. et al. International Journal of Behavioral Nutrition and Physical Activity (2026). · DOI: 10.1186/s12966-026-01906-z
Key Takeaway

A $360–$375 seasonal subsidy for culturally tailored produce boxes protected low-income NYC residents from a citywide drop in fruit and vegetable intake during 2023 price hikes, increasing dietary variety rather than total volume.

About

An evaluation of the Harvest Share cost-offset CSA in Sunset Park, Brooklyn, comparing 153 low-income Chinese and Mexican American participants ($7–$11/week share) against 389 control residents using skin carotenoid biomarkers across the 2023 inflation spike.

Key evidence
  • Defensive protection. Biomarker-measured intake dropped significantly in comparison neighborhoods but remained stable among CSA participants (-27.3 vs. -7.4 skin carotenoid score points).
  • Variety and cooking skills moved, volume did not. CSA participants added 3.74 new produce items to monthly diets and improved cooking confidence, though total cups consumed did not increase.
  • High retention. 84% retained enrollment through 20 weeks, with 88% average weekly pickup and 30% paying via SNAP.
Implications for food systems transformation

Produce subsidies are usually framed for raising intake, but they should be framed and funded as insurance buffers against food inflation rather than magic levers for dramatically increasing intake volume.

Gaps and next steps

Non-randomized design; control group was lower-income and hit harder by inflation. Variety gains were self-reported. Data from all four planned growing seasons will provide the stronger test.

← Back to contents
Reports
10

“The State of Food Security and Nutrition in the World (SOFI) 2026: Understanding and Addressing the High Cost of a Healthy Diet”

FAO, IFAD, UNICEF, WFP and WHO (2026). Rome: FAO. · DOI: 10.4060/cd8306en
Key Takeaway

What a healthy diet costs and what governments fund are misaligned. Most of the world’s $540 billion in annual farm support goes to grains, sugar, dairy, beef and feed crops. Redistributing it, without spending more, could raise fruit and vegetable intake 10% in OECD countries.

About

The UN’s flagship food security report updates the Cost and Affordability of a Healthy Diet indicator for 166 countries, drawing on International Comparison Program price data, FAOSTAT, and demographic and health survey data.

Key evidence
  • Global unaffordability: 2.69 billion people (32.7% of the world) could not afford a healthy diet in 2025, down from a peak of 3.13 billion in 2020 and from 2.97 billion (37.4%) in 2021. The improvement reflects incomes rising faster than food prices, not falling costs: the cost of a healthy diet rose in every region and in every year since 2017. Prevalence fell in Asia (28.9%), Latin America and the Caribbean (25.7%) and Northern America and Europe (6.8%), and rose in Africa, the only major region moving the wrong way, to 66.6%.
  • Cost structure: The global average cost of a healthy diet reached $4.28 PPP per person per day in 2025 (up from $2.94 in 2017).
  • Subsidy misalignment: Global support to agricultural producers runs to about $540 billion a year, roughly 70% of it tied to specific commodities, predominantly grains, sugar, dairy, beef and oil crops used in livestock feed. Fruit and vegetable producers receive very little support, and support for legumes has historically reinforced cereal pathways at the expense of legume diversification.
  • Midstream value-chain upgrades yield highest cost reductions: A 10% efficiency gain in fruit and vegetable value chains lowers healthy diet costs by 7.5% globally. Transport, processing, and retail account for 70–75% of consumer food costs.
Implications for food systems transformation

Governments must repurpose agricultural support away from staple monocultures and toward high-cost food groups (fruits, vegetables, legumes) and post-farmgate logistics. The report’s central policy signal is that broad-spectrum fiscal support to low-cost staples is not the same instrument as targeted support for affordable healthy diets, and the two can move in opposite directions.

Regional priorities differ: Africa on animal source foods (69% of countries in the highest global cost tercile for this group), Asia on fruits, Latin America and the Caribbean on vegetables (66% of countries in the highest tercile), and Northern America, Europe and Oceania on labour costs in horticulture.

Gaps and next steps

Uses a rigid normative diet basket rather than modeling price-saving substitutions (e.g., swapping expensive animal proteins for lower-cost legumes). Excludes true-cost environmental externalities. The report identifies animal source foods as the highest-cost group in Africa and prescribes structural transformation of the animal source food sector as the response, but does not model whether within-basket compositional flexibility (for example, shifting a share of animal source food calories toward the legumes, nuts and seeds allocation, where per-kilocalorie prices are lower in most regions) would deliver additional affordability gains alongside supply-side interventions. Demand-side and food environment interventions are explicitly deferred to a forthcoming FAO report on healthy diets.

← Back to contents
11

“Global status report on cancer 2026: the future we choose together”

World Health Organization and International Agency for Research on Cancer. Geneva: WHO (2026) who.int/publications/i/item/9789240123977
Key Takeaway

Cancer cases are projected to rise 67% by 2050, to nearly 35 million a year, at a cumulative cost of US$ 33.2 trillion between 2020 and 2050, with WHO putting the return on prevention at US$ 9.50 for every dollar invested. The report’s own obesity package lists public food procurement, marketing regulation and labelling as proven interventions. None reaches its seven formal recommendations, where the words “diet”, “food” and “nutrition” do not appear once.

About

WHO’s flagship global cancer assessment detailing burden projections, modifiable risk factors, and national policy recommendations across 280 pages. It succeeds WHO’s first cancer report in 2020 and is the reference document that will inform national cancer control plan revisions and the indicators of WHO’s forthcoming Global Cancer Monitoring Framework.

Key evidence
  • Incidence surge. Global cancer cases are projected to rise from 20.6 million (2024) to 35 million by 2050, an increase the IARC put at ~67%, and surging 133% in low-income nations. The global cost of cancer is projected to exceed US$ 33.2 trillion between 2020 and 2050.
  • Diet is named, but not quantified, as a risk factor. The report attributes 38% of cases to modifiable risks (tobacco 15%, alcohol 3%, high BMI 2%), but assigns 0% directly to diet and downplays processed and red meat risk factors.
  • Processed meat is used to illustrate low risk: A box on page 82 distinguishing hazard from risk selects processed meat as its contrast against tobacco (both group 1 carcinogens), noting tobacco’s much higher absolute risk than processed meat, but says nothing about processed meat exposure prevalence being far greater than smoking prevalence in most populations.
  • Complete action omission: From our read, across 15 pages of formal recommendations, the words diet, food, and nutrition do not appear once. The only fiscal instrument named is an earmarked tax on tobacco, alcohol and sugar-sweetened beverages, and it appears as a way to finance cancer services rather than as a prevention measure.
Implications for food systems transformation

The gap is translation, not evidence. WHO supplies the premise, that more than a third of new cancers are preventable and prevention is the cheapest intervention available, then names no food policy instrument in its recommendations even though its own obesity package already lists procurement, marketing regulation and labelling as proven interventions. The stated reason is insufficient data. For the indicator the WHO has chosen, that is fair. As a general claim it is not. Zhang et al. (2019) estimated more than 80,000 new US cancer cases in 2015 associated with suboptimal diet, 4.4% from direct dietary associations plus 0.82% obesity-mediated, and Islami et al. (2024) put the diet-attributable share at 3.4% in women and 4.9% in men. Either figure is comparable to or larger than the 3% the report assigns to alcohol and does quantify. Attributable fractions drive attention and budgets: a risk factor left unquantified and absent from the recommendations will not be resourced, and national plans drafted against this report will treat food as a communications problem rather than a policy one.

Cross-references

Kassam et al. on low awareness among oncologists and food (January). Of 150 oncology professionals surveyed, ~85% had received no formal nutrition training and 17.3% were unaware that any cancer-specific dietary guidelines exist, yet 77.4% considered diet important for cancer risk reduction.

Aune et al. on vegetarian diets and cancer incidence (March). The WCRF grading for vegetarian diets and total, colorectal, colon and breast cancer was upgraded from inconclusive to probable causal.

Son, Deziel and Bell on animal feeding operations and cancer incidence (March). There is a production-side cancer risk from food, too: counties with the highest density of animal feeding operations showed 4-8% higher all-cancer incidence across three US states.

US withdrawal from IARC (March). The U.S. moved to cut ties with IARC, the WHO body that classifies red and processed meat as carcinogenic and co-produced this report. Capacity to quantify diet-related cancer risk is contracting (while the WHO report cites insufficient data for not quantifying it).

Gaps and next steps

Relies on older 2018 WCRF evidence reviews. Fails to incorporate updated comparative risk assessment methodologies that quantify diet-attributable cancer burdens nationally (3-5%).

← Back to contents
12

“The Carbon Footprint of Leather: A Comprehensive Reassessment Using Global Livestock Data and Meta-Analysis”

Roncevich, M.K., Hayek, M.N. & Hinestroza, J.P. (2026). ACS Sustainable Chemistry & Engineering. · DOI: 10.1021/acssuschemeng.5c12631
Key Takeaway

Bovine leather carries a carbon footprint ~12x higher than vegan alternatives. Correcting standard industry rules that wrongfully assign hide emissions to dairy cattle doubles leather’s calculated farm footprint.

About

Pairs a systematic LCA meta-analysis of bovine vs. synthetic/plant leathers with a top-down re-estimation of farming emissions using FAO global livestock data and regional economic allocations.

Key evidence
  • Roughly twelve times the footprint of vegan leather. Bovine leather averages 187 kg CO2e per kg of finished leather vs. 14.9 for vegan options. The highest-emitting vegan leather still emits less than the animal tanning stage alone.
  • Accounting flaw more than doubles farm-stage emissions. Existing Product Environmental Footprint rules assume hides come from dairy cows and assign them \<0.5% of emissions. Reallocating emissions to reflect that most leather comes from beef cattle raises farm emissions to 139 kg CO2e per kg of finished leather.
  • Tanning emissions vary sharply by region. Tanning-stage emissions depend heavily on the local energy mix, ranging from ~19 kg CO2e per kg of finished leather in Europe to ~140 in Turkey and 165 in India, due to energy grid coal intensity.
Implications for food systems transformation

Debunks the fashion industry claim that leather is a low-impact byproduct. Essential evidence for updating the EU Green Claims Directive and environmental footprint rules.

Gaps and next steps

LCA meta-analysis relies on a small sample size (4 studies per category). Excludes land-use change, chemical tanning pollution, and end-of-life disposal.

← Back to contents
News
Business & Markets
  • Super El Niño threatens a multi-year food price shock. Economists warn that a developing El Niño could disrupt yields severely enough to push food inflation into 2028.
  • Extreme weather reshapes global crop output. Carbon Brief’s 15 July round-up covers drought in Uganda with at least 16 deaths from hunger, India’s driest June in 12 years, a double-digit French production decline, UK growers cutting inputs to retain soil water, and a $200m FAO and WFP appeal across 22 African countries.
  • JBS abandons its climate and deforestation targets. The world’s largest meat producer formally dropped its net-zero and zero-deforestation commitments.
  • A handful of firms control Europe’s meat supply. A cross-border investigation finds eight companies hold roughly 70% of UK meat processing turnover, seven control half of Spain’s pork slaughtering, and five control half of Italy’s poultry, pig and beef chain.
  • General Mills, Walmart and ADM commit to regenerative Midwest wheat. A joint pledge to expand regenerative practices across Midwest wheat acres.
  • Australia’s largest insect farm enters liquidation. Goterra was wound up after a creditors’ vote, another marker of commercialisation headwinds in insect protein.
  • Inflation control and cash transfers cut South American hunger. Targeted safety nets and price stabilisation drove measurable falls in regional food insecurity.
Policy and Governance
  • New York City bans processed meats across city agencies. Effective July 1, updated NYC Food Standards remove processed meat and raise the requirement for whole or minimally processed plant proteins across more than 219 million meals a year in 11 agencies, including schools, hospitals and shelters.
  • Copa-Cogeca lobbying kills the EU pesticide law. Europe’s most powerful farming lobby derailed the proposed pesticide reduction regulation, a significant setback for Farm to Fork.
  • USDA moves to repeal chicken farmer protections. The department has begun rolling back fair-competition rules that shield contract growers from corporate integrators.
  • USDA accelerates closure of its flagship science campus. The agency is speeding up the shutdown of its primary agricultural research facility, narrowing public research capacity at a moment of rising climate and market volatility.
  • California budget invests in healthy school meals. The updated state budget funds expansion of healthy, high-quality and locally sourced school meals statewide.
  • EU publishes its livestock strategy and protein plan. The Commission set out its long-term vision for the livestock sector alongside a protein self-sufficiency roadmap.
Technology & Innovation
  • Cultivated duck reaches multi-tonne scale. PARIMA reported a multi-tonne duck run on Vow’s 22,000-litre Sydney line at 99% lower cost than earlier runs, figures that are company-reported.
  • Generative AI maps the taste, health and climate trade-off in burgers. A Stanford diffusion model trained on 2,216 recipes produced top-rated designs that stayed meat-based, a mushroom burger with an order-of-magnitude lower impact but lower sensory scores, and a bean burger with nearly double the Healthy Eating Index and the largest sensory penalty (Tac, Gardner and Kuhl, npj Science of Food).
  • Welfare-driven egg policy carries an environmental cost, at constant output. A UK life cycle assessment finds that moving production from cages to barn or free-range systems raises national emissions by 5.8% to 13.6%, driven mainly by feed. Every scenario holds egg output fixed, and the authors note that lower egg intake would improve environment and welfare together but do not model it (Martinic et al., Royal Society Open Science).
  • Lebanon’s agriculture ministry launches an anti-plant-based AI campaign. The ministry released an AI-generated video discouraging plant-based diets and promoting animal protein.
  • Dietitians register in proportion in local food networks. A social network analysis of a peri-urban Australian food system mapped 242 actors and found one of 11 key actors was a nutritionist or dietitian – close to the sector’s 9.5% network share (Pickles et al., Journal of Human Nutrition and Dietetics).