The Zero Hunger Formula
An ecosystem framework for food security resilience and shared prosperity
Updated white paper v0.6 • Research and policy consultation draft • October 22 2026
Sponsor and coordinating organization: World Food Bank Foundation (WFBF). Technical structuring and reporting: Institute for Applied AI & Robotics (IAAIR), where AI means artificial intelligence.
Executive summary
Hunger persists when essential parts of the food system fail to work together. More production cannot resolve missing storage, inaccessible finance, unreliable transport or weak purchasing power on its own. The Zero Hunger Formula (ZHF) provides a common way to diagnose those connected constraints, prioritize coordinated responses and measure progress toward zero hunger.
ZHF combines nine pillars, Regional Food Portfolio Design, an adjoining implementation handbook and a digital accountability system. It supports food security, agricultural development and disaster preparation through one ecosystem perspective. Countries retain their own plans and institutions while using shared definitions, transparent evidence and comparable reporting.
Ecological foundations | Public coordination | Physical systems | Human and financial capacity |
Regenerative agricultureWaterBiodiversity | PolicySurplus redistribution | StorageInfrastructure | EducationCapital access |
Nine pillars → regional food portfolios → coordinated action → measured outcomes
One proposed headline gap score shows proximity to agreed zero-hunger benchmarks. Its outcome, capacity and resilience components remain visible, alongside the annual rate of improvement. Mathematical relationships, weights and benchmarks will be evaluated before any global standard or public ranking is established.
Creative financing is an implementation mechanism. Cross-agency purchasing can support a shared food reserve; commodity-backed payments and separately funded insurance can improve trade and financial protection. Adoption, funding volumes and use of any particular product do not earn score points. Improvements must appear in measured services and outcomes.
The initial research program
WFB and IAAIR propose coordinated research in Zambia, Zimbabwe, Mozambique, Botswana and Namibia, and in Colombia, Costa Rica, Panama, Nicaragua and Honduras. Preparation will take six to eight months, followed by three years of operation and six months of synthesis. The specified staffing and institutional support require $33.40–$34.79 million; additional delivery costs must be budgeted before submission.
This paper consolidates the supplied ZHF materials, the Federal Emergency Management Agency (FEMA) proposal, the American Strategic Food Reserve Initiative (ASFRI) proposal and the FLOR commodity-token proposal, with the sponsor’s subsequent SYLO commodity-token and insurance design clarifications. FLOR and SYLO are product names. Case descriptions are proposed or sponsor-described mechanisms; they are not independent findings of effectiveness. The updated nine-pillar framework supersedes earlier six-component formulations.
1 The nine pillars
The pillars describe services and conditions that a functioning food ecosystem needs. They guide diagnosis; they do not prescribe an identical crop mix, institution or technology for every place. Nutrition, equity and exposure to shocks are measured across the system.
Pillar | Diagnostic question | First pathway and proposed measure |
R Regenerative agriculture | Can production maintain soil function and viable yields? | Assess sourcing land and soil constraints; test suitable practices. Track verified soil-function safeguard coverage. |
W Water | Is safe and sustainable water available when needed? | Map household and production needs, quality and seasonal limits. Track service reliability against each need. |
B Biodiversity | Are biological functions and habitats protected? | Identify ecological dependencies and threats. Track verified biodiversity safeguards in the sourcing footprint. |
P Policy | Do institutions perform essential duties reliably? | Assign mandates, decisions and handoffs. Track verified completion of required public services. |
S Surplus redistribution | Can safe food reach unmet need without encouraging waste? | Verify demand, suitable surplus and recipient access. Track needs met; identify recovered surplus separately. |
D Storage | Can usable food be safely held until required? | Map storage demand, quality control and rotation. Track commodity-specific safe storage service coverage. |
I Infrastructure | Can food and essential inputs reach users reliably? | Map transport, energy, processing and communications constraints. Track complete, on-time food delivery. |
E Education | Can participants apply the skills the system requires? | Assess practical capabilities and provide relevant training. Track demonstrated skills, rather than attendance alone. |
C Capital | Can participants obtain suitable finance when needed? | Map cash cycles, cost, settlement and protection gaps. Track timely, usable working-capital access. |
These are candidate indicators, not established global measures in every case. Their definitions, sampling, denominators, ecological applicability and verification rules require pilot testing. The companion indicator workbook provides collection fields and initial operational pathways.
Keep pillar boundaries explicit: storage measures safe inventory service; infrastructure measures the connections that enable delivery. Education measures applied capability; policy measures execution. Prevent the same underlying observation from being credited repeatedly across pillars.
2 Regional Food Portfolio Design
Regional Food Portfolio Design connects the nine-pillar diagnosis to practical choices about what to produce. Begin with what a region does well, identify how its strengths can address nutritional needs, and select a complementary mix that improves nutrition and durable income through sustainable or regenerative practices. Revisit these choices together rather than allocating all resources to one priority before considering the others.
The portfolio includes fruits, vegetables, grains, pulses, roots, herbs, edible leaves, livestock, aquaculture and other appropriate foods. Evaluate varieties, rotations, intercropping and integrated production systems as well as individual commodities. A profitable crop can support household purchasing power; a locally consumed food can address a seasonal dietary gap; a complementary enterprise can use suitable by-products or diversify income.
Design priority | Question | Decision rule |
Regional strengths | What can this area produce reliably and efficiently? | Assess climate, soils, water, skills, realistic yields, delivered costs and ecological limits. High yield alone is insufficient. |
Nutritional contribution | Which foods and nutrients are unavailable or unaffordable, and when? | Use population needs, cultural preferences, seasonal supply and household access to identify gaps. |
Income and resilience | Which combinations improve livelihoods while sustaining food access? | Compare net income, cash timing, demand, risk, diversification and environmental performance. |
Agro-ecological zoning by the Food and Agriculture Organization of the United Nations (FAO) and the International Institute for Applied Systems Analysis (IIASA) provides a starting point for crop suitability and production potential. Validate regional maps with field observations, trials and producer knowledge; supplement them with livestock, aquaculture and integrated-system assessments. Nutrition-sensitive value-chain methods connect production to consumer access. [S23; S24]
Select a portfolio within real limits
Compare nutrition-priority, income-priority and resilience-priority portfolios. Each must respect land, sustainable water availability, labor, capital, rotations, feed requirements, processing, storage, buyer demand and environmental safeguards. Agree minimum nutrition-access commitments and ecological limits before comparing financial returns. Show trade-offs instead of concealing them in a single unexplained ranking.
Decision variables describe production areas or units, seasonal timing, processing, local supply, purchases and trade. Estimate edible output and nutrient retention after losses; count feed and by-products once. Model households’ ability to obtain the food, not only nutrients produced. Net income includes full production and delivery costs, with payment timing and working-capital needs tracked separately.
A region need not produce every food it needs. Coordinate complementary production and dependable trade across localities and countries, retaining diversity and reserves against shared shocks. A city can shape procurement, urban production and distribution while partnering with its wider food-sourcing region.
3 A practical process for local adoption
Step | Local team action | Working output |
1 Define | Map communities, production zones, markets and trading partners; distinguish ecological zones from administrative boundaries. | Food-system map |
2 Assess strengths | Combine soil, water, climate, existing performance, producer knowledge and ecological constraints. | Suitable production shortlist |
3 Identify needs | Assess seasonal availability, affordability, cultural preferences and nutritional gaps by population group. | Nutrition-needs calendar |
4 Profile options | Document yield ranges, edible products, nutrients, inputs, costs, labor, timing, losses and buyers. | Commodity and system profiles |
5 Compare portfolios | Combine suitable products, rotations, livestock or aquaculture and trade into two or three feasible alternatives. | Portfolio comparison |
6 Coordinate dependencies | Identify the skills, finance, water, storage, processing and delivery each alternative requires. | Prerequisite and investment schedule |
7 Pilot and select | Test with willing producers and buyers; review actual costs, margins, access and environmental effects. | Locally approved operating plan |
8 Review and adapt | Update after production cycles and material changes in markets, resources or household needs. | Revised portfolio and evidence |
Four simple working documents
Use a regional strengths map; a seasonal nutrition-needs calendar; standardized commodity and production-system profiles; and a portfolio with its dependency schedule. The approved plan specifies what is produced, where, by whom, in what quantities and seasons, for which consumers or buyers, with which prerequisites and budget.
Each commodity profile records the production zone and practice, yield range and evidence quality, edible portion, nutrient contribution, seasonal water and labor needs, soil and biodiversity effects, time to harvest or sale, storage life, demand limits, expected prices, full costs, cash cycle and major risks. Record local knowledge with consent and test uncertain assumptions through pilots.
Connect choices to the nine pillars and results
Ecology defines feasible production; education and capital enable it; storage and infrastructure connect supply to users; policy and redistribution support coordinated access. Portfolio analysis identifies the constraints to address first. Selecting or publishing a portfolio does not improve the ZHF score; measured services, food-security outcomes and resilience must change.
WFB country leaders coordinate local participation; agriculture advisors assess production and practices; economics advisors assess demand, margins and cash flow. IAAIR supports data standards, scenario comparisons and evaluation. Producers and authorized institutions choose the plan. Analytical tools may compare alternatives; the dashboard monitors evidence and alerts without authoring plans or allocating resources.
4 A simple headline with transparent mathematics
The proposed Zero Hunger Gap Index G ranges from 0 to 100; lower is better. It combines outcome gap O, capacity gap K and resilience gap H. Equal weights are a transparent initial hypothesis, not an established statement of their relative importance. [S01]
G = (O + K + H) / 3
View | Proposed construction | Interpretation |
Outcome gap O | Average of food insecurity, healthy-diet unaffordability and nutrition gaps; the nutrition gap initially averages child stunting and wasting gaps. | Use separately normalized indicators. Different population denominators remain visible; O is not a population prevalence. |
Capacity gap K | Derived from the nine pillars, grouped below. | Shows ecosystem capability against specified service benchmarks. |
Resilience gap H | Average unmet nutritional-service share under registered stress scenarios. | Shows modeled continuity of food service; validate against observed disruptions. |
Proposed relationships among pillars
Normalize each pillar’s achievement to the interval 0–1, using lower-case r, w, b, p, s, d, i, e and c. The four group scores are:
g1 = (r × w × b)^(1/3) g2 = (p × s)^(1/2)g3 = (d × i)^(1/2) g4 = (e × c)^(1/2)
Q = (3g1 + 2g2 + 2g3 + 2g4) / 9 K = 100 × (1 − Q)
Multiplication represents complementary conditions within a group. Roots make groups with two and three pillars comparable; without roots, more factors mechanically depress a product of values below one. Addition permits partial trade-offs across groups. The 3:2:2:2 weights give equal local pillar influence when all achievements are equal, not equal causal importance.
Division forms averages and coverage ratios. Climate, conflict and other risks enter stated scenarios or documented transformations; they are not placed in an undefined denominator. These are modeling choices to test against additive, geometric and less compensatory alternatives.
Distance and speed
For an adverse measure x, target T and worse benchmark U: gap = 100 × clip((x − T)/(U − T), 0, 1). For a beneficial measure, achievement = clip((x − L)/(T − L), 0, 1), where L is the worse benchmark. Targets and anchors must be agreed and fixed for comparable releases.
Annual progress = (previous G − current G) / elapsed years. Positive values indicate improvement. Publish raw measures, component changes and uncertainty beside the headline. Capacity improvement alone must not be described as demonstrated hunger reduction. Zero means the specified benchmarks have been met; it is not proof that every form of hunger has disappeared.
5 Global comparability and local action
One measurement standard at every scale
A city, district, province, state, country and region use the same indicator definitions, units, versioned methods and quality rules. Local teams customize interventions and operating assumptions. Locally preferred weights may inform decisions but must be labeled separately from the common headline score.
Geographic boundaries and food-system boundaries differ. An importing city depends on farms and water outside its borders. Record its sourcing footprint and supply shares rather than treating the absence of city farmland as failure. Unknown origins remain explicit.
Aggregate underlying quantities with the right denominators, then recompute the parent score. People, children, hectares and service units cannot be pooled indiscriminately. Do not average ranks or double-count overlapping areas. Regional resilience must test shared networks and simultaneous demands; the same reserve cannot be promised in full to several countries.
Publish comparisons only when they are credible
Require complete, sufficiently current core data, comparable methods and demonstrated contextual equivalence before publishing a rank. Report missing data as unknown, never as zero or silently redistributed weight. Show uncertainty bands; withhold rankings when differences or data quality do not support an ordering.
The adjoining handbook
Step | Human decision | Required output |
Assess | Where are needs, constraints and dependencies? | Baseline, pillar profile, affected groups, service map and data gaps. |
Prioritize | Which urgent needs and catalytic bundles merit action? | Options compared on expected outcomes, equity, timing, evidence, costs and adverse effects. |
Coordinate | What must happen first, and what resources are available? | Dependency schedule, seasonal windows, named owners and a funded operating plan. |
Deliver and adapt | What has changed, and what needs correction? | Verified milestones, outcome updates, cause reviews and documented decisions. |
Protect urgent nutritional needs first. For other choices, compare bundles and their separate components against a common alternative. A highly connected project is a candidate for catalytic benefit, not proof of it. Record maintenance, replacement and recurring finance alongside construction or startup costs.
A proposed operating rhythm is monthly delivery review, quarterly portfolio review and outcome review when valid surveys become available. The dashboard displays evidence, dependencies and alerts. It does not write the operating plan, choose projects or allocate funds.
6 Evidence data and digital accountability
Objective reporting requires traceable observations and reproducible calculations. Causal evaluation requires a defensible estimate of what would have happened without the intervention. A global data standard strengthens objectivity but cannot make correlation equivalent to causation. [S08]
A disciplined evaluation approach
Register questions, outcomes and methods before results are known. Use randomized or phased rollout where feasible and ethical; otherwise justify the comparison design and its assumptions. Account for market spillovers, weather, concurrent programs, selection and differing implementation quality. Report negative and null findings alongside positive results.
For a bundle A+B, catalytic interaction = benefit(A+B) − benefit(A) − benefit(B), measured against the same baseline alternative, outcome and horizon. Count the bundle’s total benefit once. Test transfer to new settings rather than assuming local success applies everywhere.
The minimum data backbone
Link geography, indicators, observations, programs, dependencies, milestones, budgets and transactions, physical flows, evidence, alerts, stress scenarios and score releases. Each observation carries its source, date, unit, denominator, uncertainty, geography, method and verification status. Existing activity-reporting and geographic-monitoring structures provide useful starting points. [S10; S11]
Every important handoff records who owes what, to whom, where, when and to which service standard. An alert identifies a suspected breakdown. A review distinguishes delivery failure, an incorrect causal assumption, an external shock and measurement error, then records corrective action and appeal. Digital accountability should establish responsibility through evidence.
Local knowledge and shared artificial intelligence
A permissioned evidence repository can support specialized small-language-model agents for agronomy, economics, nutrition, logistics and evaluation. Local tacit knowledge enters as attributed, consented observations or hypotheses. Each suggestion links to its evidence, context, prerequisites, expected delay, potential harms and uncertainty.
Agents may retrieve, compare and propose; they must not invent causal coefficients, authorize expenditure or change published scoring rules. Use deterministic, versioned calculations for scores. More adoption does not automatically strengthen an innovation’s evidence: independent replication, study quality and context determine weight. Protect community authority over reuse and training; apply FAIR metadata principles (Findable, Accessible, Interoperable and Reusable) and CARE governance principles (Collective Benefit, Authority to Control, Responsibility and Ethics). [S17; S18]
A compact reporting set
Publish G and its annual change; O, K and H; the nine-pillar profile; raw food-security and nutrition measures; service-failure and crisis alerts; data quality; and reviewed implementation progress. Maintain established crisis classifications separately. Retain prior releases and explain corrections so others can reproduce the result.
7 Integrated financing across agencies
Financing is a means of improving the ecosystem. ZHF should make the structure of capital visible: who supplies it, when it arrives, what service it purchases, who bears loss and how resources circulate. It does not add financing volume or institutional adoption to the formula.
Instrument | Function | What must be demonstrated |
Coordinated public purchasing | Several agencies buy distinct services from shared infrastructure. | Authorized commitments, allocated costs, clear service obligations and no duplicate billing. |
Contracted offtake and working capital | Expected sales support procurement and operating finance. | Credible demand, payment timing, delivery conditions and affordable debt service. |
Revolving inventory finance | Sales replenish capital for the next procurement cycle. | Stock ownership, quality, losses, cash reconciliation and replenishment rules. |
Guarantees and insurance | Defined risks are allocated to funded counterparties. | Coverage terms, funding, exclusions, claims and stress-tested obligations. |
Digital receipts and settlement | Verified assets support claims and lower-friction trade. | Custody, liabilities, redemption, liquidity and actual user costs. |
Each structure needs a Financing Design Record: ecosystem constraint; beneficiaries and payers; funding and repayment flows; legal and institutional dependencies; risk allocation; recurring costs; stress cases; and evidence of additional benefit. Countries can adapt structures while maintaining common outcome measurement.
The FEMA and ASFRI example
The proposed chain purchases suitable farm surplus, dries and processes it into regionally appropriate meals, stores verified lots, and supplies both disaster response and routine feeding. Agriculture agencies can support eligible farm and processing functions; disaster agencies can purchase readiness and activation; feeding programs can purchase eligible meals. Shared agreements coordinate distinct commitments and accounts. [U03; U06; S19]
Protect a minimum emergency reserve while rotating designated inventory through routine demand. Opening usable stock + released production − routine issues − emergency issues − losses = closing usable stock. Track lot quality, nutritional suitability, preparation water, expiry, delivery and replenishment. A long shelf life does not remove the costs of storage, inspection, insurance or readiness.
Test farmer receipts, full delivered meal cost, waste, reserve availability, response time and recipient outcomes. The total ASFRI budget cannot also be treated wholly as farmgate purchasing power. Calculate procurement savings against relevant food and logistics spending. Large response, health and cost claims in source proposals require independent validation before inclusion as expected benefits.
8 SYLO as an optional supporting mechanism
SYLO is a proposed commodity-backed trade and payment system with a separately held insurance reserve. It aims to let users transact without token-transfer fees, hold value linked to a diversified pool of commodities and cash, and receive defined partial protection against eligible losses. Conversion to other currencies carries disclosed charges. It is an optional implementation example, not a condition of ZHF participation.
The sponsor’s latest SYLO instructions govern this description. The attached FLOR draft is background; earlier governance-token, yield and product claims are not adopted by implication. No country approval, deployed performance or universal product uniqueness is asserted here.'
Three distinct financial functions
Function | Assets and responsibility | Boundary |
Token backing pool | Verified commodities, available cash and appropriately valued net positions; managed by the operator and professional risk managers. | Determines commodity-linked net asset value. Includes all related obligations; excludes the separate insurance reserve. |
Insurance reserve | Separately held protection assets and defined claims obligations under the approved insurance structure. | Supports capped eligible claims. Cannot also be counted as ordinary token backing or operator working capital. |
Operator resources | Agreed revenue share and operating reserves. | Funds transactions, technology and administration; sustains zero token-transfer fees during low-income periods. |
Indicative token net asset value (NAV) = (verified commodity value + backing-pool cash + net derivative value − other backing-pool liabilities) / outstanding tokens. Count derivative obligations once. Commodity-linked NAV varies; a fixed currency redemption value is not promised by commodity backing alone.
Trading and replenishment policy
The proposed standard strategy writes out-of-the-money calls against eligible exposure. Managers may close and roll positions when the combined economics meet policy, or accept assignment and reinvest proceeds in the backing pool. Gains retained as cash and the remaining commodities together determine NAV. Physical inventory floors and processor commitments constrain how much stock can be exposed or released.
A roll closes an existing obligation and opens a new one. Evaluate cumulative premiums, repurchase costs, asset value, strike, storage, financing and settlement costs. A higher strike or additional premium can improve terms but does not guarantee subsequent portfolio gains. Assignment profit must be tested against the relevant cost basis and all costs. [S20]
Exchange-traded options on futures may create futures positions on assignment rather than directly transfer local grain. Experienced managers must reconcile exchange exposure with physical sales, quality, location, basis risk, margin liquidity and settlement. Receipt and inventory audits remain necessary alongside digital records. [S21]
9 SYLO protection and financial sustainability
The sponsor proposes allocating 75% of qualifying net trading proceeds to the separate risk reserve, with 20% paid to the operator. The remaining 5%, treatment once the reserve target is reached, and whether the same allocation applies to every futures and options strategy require explicit rules. The 20% operator share is distinct from a maximum payout equal to 20% of covered crop value.
Define distributable proceeds after losses, closing costs and provisions, with rules for prior losses and open positions. New premiums are not automatically profit; futures opening transactions do not generate option premiums.
Reserve size and claims are different quantities
Let V be the defined reference value of the covered crop, L the eligible loss, and c the issued coverage rate, initially low and potentially rising to 20%. With no deductible, policy entitlement = min(L, c × V), with c ≤ 20%. This reimburses eligible loss up to a share of crop value, not a share of the loss. Set the valuation date and method before coverage begins. Issued obligations require adequate claims-paying resources and an approved capital framework.
Illustrative example on $100 exposure | Amount |
Separate reserve before new earnings | $5.00 |
Qualifying net option earnings after worthless expiry | $4.00 |
Reserve contribution at 75% | $3.00 |
Reserve before claims | $8.00 |
Eligible commodity-value loss at 15% | $15.00 |
Illustrative issued coverage at 8% of crop value | $8.00 |
Payment / residual loss / reserve remaining | $8.00 / $7.00 / $0.00 |
At issued 20% coverage, the same $100 crop has a $20 cap: a $15 eligible loss pays $15; a $30 loss pays $20. The $8 reserve alone cannot fund either obligation. This early-stage example assumes 8% issued coverage, one exposure, no other claims or costs, and is not a solvency recommendation. Reserve growth does not automatically change existing policy terms.
Build coverage through research and experience
Start with limited coverage and increase future coverage as revenue, reserves, claims experience and capital adequacy support it. Do not retrospectively reduce issued benefits. Test the suggested historical-low-price benchmark, defined for a weighted basket and covered quantities, against current exposure, tail losses, correlated claims and liquidity needs. No final reserve target or premium yield is established.
A reserve target based on historical low-price values is distinct from the policy payout cap of up to 20% of the defined covered crop value. Crop-production loss is also different from a commodity-price decline and requires its own insured event. Define valuation dates, beneficiaries, claims evidence, exclusions and payment deadlines. Country-specific insurance authorization and professional actuarial review are part of implementation. [S22]
10 Hamara and coordinated regional ecosystems
The sponsor-described Hamara model in Zambia and Zimbabwe links grain production, feed milling, livestock and aquaculture, household income, cooperative reinvestment and grocery or wholesale markets. Farmers sell maize, soy and millet to the mill; purchase suitable feed for chickens, cattle, goats or fish; and use retained income to support additional production.
A practical sequence
Sequence | Decision and evidence |
1 Verify supply and demand | Confirm grain quality, seasonality, feed demand, species-specific formulations, fair purchase terms and household food needs. |
2 Establish farm economics | Measure feed conversion, animal health, water, labor, survival, prices, margins and cash timing by production system. |
3 Secure market connections | Confirm buyers, grades, transport, processing capacity and payment reliability for livestock, fish and crops. |
4 Reinvest through cooperatives | Agree member distributions, governance, planting windows, working capital and buyer commitments. |
5 Evaluate the full chain | Test income, food access, nutrition, resilience and environmental outcomes; monitor feed–food competition and market concentration. |
Producer cash available = grain receipts + animal or fish receipts + crop receipts − feed − other inputs − labor and transport − debt service − household withdrawals. Inventory and receivables timing must be modeled separately. Transfers between ecosystem participants are not additional system-wide income when accounts are consolidated.
Financing may include buyer-backed working capital, warehouse receipts, cooperative guarantees or SYLO where suitable and authorized. Select mechanisms by demonstrated cost, usability, risk and outcomes. No token or financing product is mandatory. Feed safety and species-specific nutritional requirements remain prerequisites. [S15]
Regional coordination without identical national plans
The Africa research group will include Zambia, Zimbabwe, Mozambique, Botswana and Namibia. The Latin America group will include Colombia, Costa Rica, Panama, Nicaragua and Honduras. These are proposed research jurisdictions, not confirmed government commitments. Latin America includes Colombia alongside four Central American countries.
Map trade corridors, shared processors, water dependencies, storage, border procedures and joint shock exposure. Countries can coordinate supply and evidence while retaining separate budgets, authorities and operational responsibilities. Test where a shared reserve or service improves outcomes and where national redundancy is necessary.
Hamara, ASFRI and SYLO are learning cases with different functions. Their claims enter the evidence repository with provenance and verification status. Their usefulness rests on the problems they solve and the results they demonstrate, rather than preferential treatment in scoring or procurement.
11 Governance and the research program
WFB sponsors and coordinates the initiative, country relationships, operating teams and dissemination. IAAIR supports technical structuring, research methods, model testing, data architecture and reporting. A proposed joint steering committee will include participating-country and independent methodological representation. Country teams retain planning responsibility.
Separate commercial interests from evidence review. WFB or related providers may supply candidate mechanisms, but procurement decisions, claims validation and evaluations require disclosed conflicts and independent review. Neither WFB nor IAAIR should unilaterally determine that its own intervention has succeeded.
Phases and decisions
Phase | Duration | Outputs and decision gate |
Preparation | 6–8 months | Country and data agreements; staffing; common protocol; baseline and comparison design; strengths maps, nutrition calendars and commodity profiles; complete delivery budget. |
Full operation | 36 months | Ten country diagnostic programs and locally approved food portfolios; two regional network assessments; pilot evaluations; dashboard releases; annual independent reviews. |
Synthesis | 6 months | Validated datasets; comparative findings; revised scoring and handbook; policy presentations; replication and continuity plans. |
The midpoint schedule is 49 months: seven months preparation, 36 months operation and six months synthesis. Foundation or endowment funding is sought for research and coordination. Investment in mills, warehouses, commodity pools or insurance capital is separately specified and financed.
Workstreams
The program will develop regional food portfolios through suitability assessment, seasonal nutrition analysis and pilot production plans; test indicator definitions and score behavior; document country and regional dependencies; evaluate coordinated intervention bundles; build the digital accountability backbone; and investigate integrated financing and risk protection. Financing studies will compare existing payment and procurement arrangements with candidate structures, including full costs and distribution of benefits.
The first-stage research should confirm sampling and statistical power before promising detectable effects. Establish appropriate comparison areas or rollout sequences, protect essential services, and allow longer observation for soil, nutrition and institutional change. Country participation does not by itself mean every intervention can be causally evaluated.
Durability and adoption
Commercial functions should demonstrate recurring margins, affordable debt service and replacement capacity. Public food services should demonstrate dependable recurring funding and accountable delivery. Sustainability does not require every public service to charge beneficiaries. A standards steward, independent review process and country participation agreement are prerequisites to presenting ZHF as an adopted global standard.
12 Core resource requirements
The supplied compensation assumptions fund five full-time WFB personnel in each country: one country leader at $120,000, one economics advisor at $120,000, two agriculture advisors at $120,000 each, and one local data entry officer at $40,000. Each country costs $520,000 annually; ten countries cost $5.20 million.
WFB home-office roles comprise Africa and Latin America Operations Directors at $175,000 each, full time, and a WFB Director at $220,000 for half-time service. IAAIR comprises five part-time independent researchers at $50,000 each, a Head of Research at $180,000 assumed full time, and a Program Director at $150,000 for half-time service.
Annual core requirement | WFB | IAAIR | Total |
Personnel and research contractors | $5,770,000 | $580,000 | $6,350,000 |
Additional institutional support | $1,000,000 | $1,000,000 | $2,000,000 |
Annual total | $6,770,000 | $1,580,000 | $8,350,000 |
49-month total | $27,644,167 | $6,451,667 | $34,095,833 |
All figures are in United States dollars (USD). The half-time figures are treated as actual annual pay for the stated service, not full-time-equivalent rates. If they are intended as full-time-equivalent rates, the combined annual amount falls by $185,000 and the 49-month amount by $755,417. Rounded components may differ from totals by $1.
The model assumes all positions and institutional allocations continue throughout preparation, operation and synthesis, with no escalation. A phased hiring and closeout schedule may change the final request. The 48-month total is $33.40 million; the 50-month total is $34.79 million.
Institutional support and full delivery costs
The additional $1 million per organization per year is explicit requested grant support for organizational capacity and continuity, additional to listed personnel. Eligible uses, milestones and reporting must be defined with the sponsor. It is not assumed earned revenue from SYLO or another commercial mechanism, and the same expense must not also be billed elsewhere.
Employer costs and benefits, travel, offices, equipment, surveys, translation, software engineering and hosting, independent evaluation, audit, specialist risk and actuarial expertise, inflation, currency allowances and contingency remain to be costed. Identify which costs institutional support will cover. The current figure is a core management and research envelope, not a fully costed implementation grant.
Immediate research decisions
Agree country partners and the legal grant recipient; appoint an independent standards and review function; fix baseline and comparison designs; test score weights and targets; specify risk scenarios; and complete country operating budgets. For SYLO, finalize coverage terms, reserve governance, earnings allocation, trading controls, redemption and liquidity rules before setting pilot commitments.
The purpose of the initial grant is to establish a credible, adaptable framework and test the mechanisms that support it. Progress toward zero must remain visible in food-security outcomes, supported by stronger capacity, resilience and accountable institutions.
13 Technical specification and validation agenda
Outcome and resilience inputs
Use established food-insecurity, affordability and child-nutrition measurement systems where suitable. Preserve their original definitions and populations. Stunting and wasting also reflect health and care conditions, so changes cannot automatically be attributed to food interventions. The demonstration outcome construction is a candidate for validation. [S02–S04]
Initial resilience scenarios may test a 20% reduction in food inflows for 30 days and loss of the largest distribution node for seven days within a 30-day horizon. These are proposed research cases, not approved global thresholds. Define nutritional service units, shared-resource constraints, substitution, duration and affected populations; verify scenario results against real disruptions. [S06; S07]
Dynamic models support explanation
A candidate specification is x(t+1) = x(t) + B u(t−lag) + Γ interaction(t) − D shock(t) + error. Here x describes conditions, u describes interventions, and coefficients represent estimated direct and interaction effects. Specify units, boundaries and delays. Unestimated links remain hypotheses; a language model must not supply invented coefficients.
Observed score change is not a causal coefficient. A difference-in-differences estimate compares treated change with comparison-group change, subject to parallel-trend and other assumptions. Other designs may be preferable. For compatible independent studies, evidence synthesis can use precision and heterogeneity, but bias, duplicated samples and transportability require separate judgment. [S08]
Core tests before public ranking
Test indicator validity and redundancy; sampling and missingness; normalization and benchmark sensitivity; alternative weights and aggregation; uncertainty in scores and changes; ecological and administrative comparability; and out-of-sample performance. Do not validate a composite solely by its correlation with outcomes already contained in that composite. [S01]
Financing and SYLO stress tests
Compare fee savings with all conversion and operating costs. Test option premiums by actual strike and tenor; costly rolls; assignment and replenishment; basis and currency changes; storage losses; rapid redemptions; margin calls; simultaneous claims; long periods of low revenue; and reserve rebuilding after payouts. Evaluate the combined balance sheets without double counting reserve assets or liabilities.
For variable insurance, distinguish the reserve target, coverage rate applied to crop value, resulting monetary payout cap and capital requirement. Future coverage can increase as evidence and resources support it. Issued obligations, independent reserve custody, eligible investments and claims priority must remain clear. Reserve investment returns and reserve losses enter the reserve accounts rather than being assumed away. [S22]
Publication and learning
Publish methods, verified findings and limitations in accessible form. Share replicable solutions with their conditions for success and failure. Preserve community control over sensitive knowledge. Update the handbook through documented review, and version
changes so countries can reproduce earlier results and compare progress honestly.
Sources and provenance
Sources support measurement and operating principles. The combined ZHF method and the SYLO specification are proposals; citing an institution does not imply its endorsement. Online source review: October 2026.
S01 Organisation for Economic Co-operation and Development (OECD) and European Commission Joint Research Centre (JRC). Handbook on Constructing Composite Indicators. 2008
S02 FAO and United Nations (UN) Statistics Division. Sustainable Development Goal (SDG) 2.1.2 metadata for the Food Insecurity Experience Scale (FIES). 2024 metadata
S03 World Bank. Food Prices for Nutrition DataBank. Accessed October 2026
S04 United Nations Children’s Fund (UNICEF), World Health Organization (WHO) and World Bank. Joint Child Malnutrition Estimates. 2025 edition
S06 FAO. Resilience Index Measurement and Analysis. Accessed October 2026
S07 United Nations Office for Disaster Risk Reduction (UNDRR). Monitoring the Sendai Framework. Accessed October 2026
S08 World Bank and Inter-American Development Bank (IDB). Impact Evaluation in Practice second edition. 2016
S10 World Bank. Geo Enabling initiative for Monitoring and Supervision. 2021
S11 International Aid Transparency Initiative (IATI). Activity Standard version 2.03. Accessed October 2026
S15 FAO WHO Codex Alimentarius. Animal Feed thematic guidance. Accessed October 2026
Sources and provenance continued
S17 Wilkinson and colleagues. FAIR Guiding Principles. 2016
S18 Global Indigenous Data Alliance. CARE Principles for Indigenous Data Governance. Accessed October 2026
S19 United States (US) Treasury Bureau of the Fiscal Service. Intra-governmental Services.
S20 Fidelity. Rolling Covered Calls.
S21 CME Group (the derivatives exchange operator; CME originated as Chicago Mercantile Exchange). Options on Futures Exercise and Assignment.
S22 International Association of Insurance Supervisors. Insurance Core Principles and ComFrame. December 2024
S23 FAO and IIASA. Agro Ecological Zoning and Land Evaluation.
S24 FAO. Nutrition Sensitive Value Chains.
Supplied materials
U01 ZHF Executive Summary dated July 18 2026 and ZHF Technical Reference Companion. U02 ZHF v3 white paper. U03 Supplied preparedness proposal for FEMA, 2020. U04 Regenified 6 3 4 Verification Standard dated August 2025. U05 Sponsor descriptions of Hamara and ZHF implementation. U06 World Food Bank Group (WFBG). ASFRI Proposal 2026 Final. U07 FLOR by World Food Bank with Notes. U08 Sponsor clarifications of SYLO reserve segregation, a payout cap based on crop value, trading policy and gradual coverage growth, October 2026.
Companions: ZHF Framework and Handbook v0.2; ZHF Indicators and Model v0.2; ZHF Grant Proposal Summary. This white paper updates their narrative context. Existing calculators remain illustrative and do not implement the newly clarified SYLO insurance terms.
Comments