Curbing household food waste and associated climate change impacts in an ageing society

The study situates household food waste (FW) within the broader context of global environmental sustainability, noting that food systems account for over one-third of global GHG emissions and exert significant pressures on water, land use, and biodiversity. Approximately one-third of food produced globally is lost or wasted (about 1.3 Gt), contributing roughly half of food-system GHG emissions. In developed nations, households are a key locus of food waste, and policies and consumer behaviors (e.g., meal preparation, portioning, label interpretation) are central to mitigation. Japan, a highly developed nation reliant on food imports, induces a substantial share of food-related GHG emissions overseas and reports that nearly half of domestic supply-chain food waste originates from households. The Government of Japan targets a 50% reduction in household food waste (relative to 2000) by FY2030. However, detailed attribution of how much, which foods, and by whom household FW and associated life cycle GHG (FWGHG) are generated has remained unclear. This study aims to quantify the structure of household FW and FWGHG in Japan by age group of the household head, clarifying links between demographic aging, dietary preferences, and FW/FWGHG to inform decarbonization strategies in an aging society.

Prior work using life cycle assessment has estimated that the global food chain contributes over one-third of GHG emissions (about 17 Gt CO2eq), with increases since 1990 when including land-use change. Global analyses highlight substantial shares of food loss and waste (around one-third of production; 17% of total production wasted), with associated GHG emissions near half of food-system totals (~9.3 Gt CO2eq) and significant environmental and social costs. Studies emphasize consumer roles in developed countries and the need for behavioral and policy interventions (e.g., labelling clarity, food banks). For Japan, 34% of food-consumption life cycle emissions occur overseas due to import dependence; 47% of domestic food loss/waste is from households, and 22% of total losses are still edible. Existing Japanese studies estimated overall food waste structures and emissions but did not disaggregate by household age or detail what foods and which households contribute most. Comparative national estimates of per-capita household FW vary widely (e.g., Germany ~60–100 kg/cap-yr, US 124, Finland 232, China 39), providing context for the Japanese estimates by age developed here. Demographic shifts toward aging populations have been linked to changing dietary patterns and carbon footprints, underscoring the importance of age-sensitive analyses.

The study integrates multiple Japanese socioeconomic and environmental datasets to quantify household food waste (FW) and associated life cycle GHG emissions (FWGHG) by age group for 2015, and projects demographic impacts to 2040. - Data sources: Food Loss Statistics Survey (FLSS) for household FW ratios by waste type; Standard Tables of Food Composition in Japan 2015 (STFC) for inedible ratios; Family Income and Expenditure Survey (FIES) for expenditures on ~197 food items; National Survey of Family Income and Expenditure (NSFIE) for age-bracketed expenditures; Retail Price Survey (RPS) for unit prices and conversion of expenditure to physical amounts; Institute of Population and Social Security Research (IPSS) for household counts and projections; Japanese LCI database IDEA v3.1.0 for life cycle GHG intensities; 3EID for wholesale-to-retail adjustments. - FW intensity calculation: Monetary expenditures from FIES/NSFIE were converted to physical quantities using RPS unit prices and representative item weights. The survival ratio (1 − inedible fraction) from STFC and FW ratios by waste type from FLSS were combined to compute average FW per expenditure for each food item and waste type. Concordance tables mapped items across datasets; for items lacking direct RPS data, representative products and official prices were used. - Normalization and totals: Item-level FW estimates were normalized to match the government-published total household FW (W_gov) to reconcile methodological differences between FLSS-derived scaling and official totals. Per-capita FW by age bracket was computed using NSFIE expenditures and adjusted by household counts and optimized family sizes with OECD square-root equivalization. - FWGHG quantification: Life cycle GHG intensities (IPCC 2013 GWP100) from IDEA were applied to FW amounts to estimate FWGHG, with system boundaries from raw materials to retail (excluding cooking). For agricultural, meat (non-chicken), and fishery items where IDEA covers up to wholesale, additional emissions from wholesale-to-retail (including processing) were added using 3EID (factor β_i). For meats (beef, pork, other raw meats), intensities were adjusted using reciprocal yield ratios to allocate to edible parts. For ready meals and other monetized intensities, monetary-to-physical conversions were applied. - Demographic projections: Future FW and FWGHG for 2016–2040 were projected assuming fixed base-year consumption patterns and technologies, using age-specific per-household FW (2015), projected numbers of households by age (IPSS), and optimized family sizes p^b(t), capturing effects of changing household sizes. Family sizes were optimized to ensure consistency with national population totals and constraints (e.g., non-increasing over time), via quadratic minimization subject to equality/inequality constraints. - Scope: Household FW includes edible parts discarded directly by households; excludes eating out, takeout, and school lunches. FWGHG covers cradle-to-retail; cooking-related emissions are excluded. Results are aggregated to 11 categories and 28 food groups for reporting.

- Total household food waste (2015): 2.89 Mt/yr. - Category contributions to FW: Vegetables largest at 43% (1.23 Mt/yr); fruits second. Key items: cabbage 0.12 Mt/yr; other leafy greens 0.11; onion 0.10; tomatoes 0.093; banana 0.097; apple 0.083; tangerine 0.060. Rice 0.070; other cooked meal 0.058; milk 0.086; lunchbox 0.046; egg 0.042. Vegetables + fruits comprise 57% of FW. - Per-capita FW by age (2015): increases with age—20s and younger 16.6 kg/cap-yr; 60s 44.4; 70s and older 46.0 kg/cap-yr (highest). Older households purchase more fresh vegetables, fruits, and seafood. Waste-type structure: Excessive preparation is the main driver for most age groups (rising from 34% to 50% with age). 20s and younger predominantly waste via leftovers (43%). Direct disposal is ~20% across ages. - Total FWGHG (2015): 6.06 Mt-CO2eq/yr. Leading categories: vegetables 1.28 Mt-CO2eq/yr; ready meals 1.13; fishery and seafood 0.70; meats 0.67; fruits ~0.45; grains ~0.45. Top item contributors: other cooked meal 0.38; beef 0.22; other bread 0.20; pork 0.17; other mushrooms 0.16; milk 0.15; tofu 0.10; strawberry 0.10; cucumber 0.078; banana 0.076. - Per-capita FWGHG by age (2015): lowest for 20s and younger 39.2 kg-CO2eq/cap-yr; highest around 60s and 70s (~90 kg-CO2eq/cap-yr), with 60s slightly higher than 70s. Relative to FW, leftovers and disposal contribute a larger share to FWGHG than excessive preparation. - Meat vs vegetables in FW vs FWGHG: Meats account for only ~2.8% of FW (<3%; ~one-fifteenth of vegetable FW) but nearly 11% of FWGHG. Cutting FW of beef, chicken, and pork alone could reduce >7% of total FWGHG. - Demographic projections (2015–2040): Total FW increases slightly to 2020 (+0.7% vs 2015), FWGHG rises to 2019; number of households peaks ~2025 then declines. By 2040, national population −13% vs 2015; total FW −5.3% and FWGHG −6.2% vs 2015. Shares from 60s and 70s+ households grow; FW and FWGHG by 70s+ increase by 18% (2040 vs 2015), while those by 40s decrease by 29%. - Item-level projection changes (2015–2040): Fresh fruits show the smallest reduction in FW and FWGHG (−1.2%), followed by other processed vegetables and seaweed (−2.5%) and salted/dried seafood (−2.5%). Largest decreases: other soft drinks (−10.4%) and instant staple foods (e.g., frozen pasta) (−10.2%). - Policy gap to target: To meet the government’s 2030 household FW target (2.16 Mt), a reduction of 0.73 Mt from the 2015 base is needed. Demographic impacts alone would yield only a −0.8% (~−0.02 Mt) reduction by 2030, indicating a need for stronger measures.

Findings demonstrate that aging drives higher per-capita household FW due to increased purchases of fresh vegetables and fruits, which are prone to waste via excessive preparation and spoilage. Although per-capita FWGHG is highest among older households, the overall FWGHG constitutes about 20% of household food-related life cycle emissions and roughly 1% of total per-capita household GHGs across all commodities, suggesting limited absolute climate mitigation via FW alone—yet the associated emissions are still unnecessary and avoidable. Effective mitigation requires targeted, age-tailored interventions: for older households, recurrent education on FW and environmental impacts, and producer-side offerings such as age-appropriate, easily prepared, and freezable meal kits to reduce excessive preparation and spoilage. For younger households, strategies should address higher consumption of confectionary and ready meals and dining out habits. Prioritizing items with high FWGHG intensity (e.g., meats, milk, mushrooms, tomatoes, tofu) can yield disproportionate benefits; reducing FW of beef, pork, and chicken could cut FWGHG by over 7%. The study also highlights a potential trade-off: health-oriented dietary shifts toward higher vegetable and fruit intake can increase household FW absent complementary education and storage/cooking guidance. Demographic projections indicate that, despite a shrinking population, aging will prevent FW from declining in proportion to population and households; thus, policy instruments must exceed demographic effects to meet national targets.

This study integrates household economic statistics with LCA to quantify, by age group, the structure of household food waste and associated life cycle GHG emissions in Japan, and projects demographic impacts to 2040. It reveals that per-capita FW increases with age due to dietary preferences for fresh produce, and that FWGHG hotspots do not perfectly align with FW volumes (e.g., meats contribute relatively low FW but high FWGHG). Demographic aging alone will not deliver the reductions needed to meet Japan’s 2030 household FW target. The study recommends age-tailored policies—education for seniors, optimized meal kits and portioning, and priority actions on high-impact items (especially meats)—as well as guidance to mitigate FW risks associated with healthier diets richer in vegetables and fruits. Future research should incorporate econometric analyses to capture behavioral and policy effects, and further integrate health outcomes with environmental impacts to design synergistic interventions.

Key limitations include: (1) data constraints and inconsistencies across sources necessitating normalization to government totals and the use of concordance mappings between monetary and physical units; (2) scope limited to edible parts of food wasted at home, excluding eating out, takeout, and school lunches; (3) FWGHG system boundary from cradle to retail, excluding cooking emissions; (4) for some items (e.g., ready meals), life cycle intensities available only in monetary terms required conversion to physical intensities; (5) LCI coverage for agricultural, meat (non-chicken), and fishery products up to wholesale required adjustments to include wholesale-to-retail stages and edible-yield corrections for meats; (6) demographic projections assume fixed base-year consumption patterns and technologies and no major policy or lifestyle shocks; and (7) partial truncation of detailed limitations in the provided text suggests additional nuances (e.g., discrepancies between FLSS and government accounting) that warrant careful interpretation.