Understanding Food Loss Patterns Across Developed and Developing Countries Using a GDP, Growth Rate, and Health Expenditure-Based Typology

Abstract

Food loss and waste (FLW) threaten progress toward Sustainable Development Goals (SDG) 12.3, yet their distribution by development stage remains under-quantified. We created a time-weighted K-means typology for 105 countries (2000-2022) using Gross Domestic Product (GDP) per capita, GDP growth, and per-capita health expenditure-indicators chosen to capture economic capacity, growth momentum, and institutional investment. The scheme classified nations as developed (n = 13), developing (n = 92), or hybrid, with > 98% membership stability across weighting parameters. Linking this typology with FAO's FLW data, we modelled food loss percentages (FLP) across ten commodity groups and eight supply-chain stages using multilevel mixed-effects regression. Developed countries lost the most food at consumption (22.5%), dwarfing developing (6.8%) and hybrid cases (9.0-14.2%), whereas developing nations suffered greater upstream losses at harvest/on-farm (3.7%). FLP in developing economies was significantly lower for grains (beta = - 8.02, p = 0.007), oilseeds (beta = - 19.29, p = 0.016) and pulses (beta = - 5.43, p = 0.021). From 2000 to 2022, oilseed and sugar losses rose (beta = 0.26, p < 0.001), while roots/tubers and dairy/eggs declined (beta = - 0.31, - 0.89; p < 0.01). Stage analyses revealed pronounced development gaps at consumption (beta = - 16.06, p < 0.001) and processing (beta = - 5.58, p = 0.014), alongside a rising trend in marketing/retail losses (beta = 0.25, p = 0.005). Country-level random effects explained up to 90% of variance, underscoring the dominance of local conditions. The evidence supports consumer-behaviour interventions in high-income settings, upstream infrastructure investment in developing regions, and dual-track strategies in hybrids. Our typology provides a scalable, policy-ready lens for designing targeted FLW actions aligned with SDG 12.3.