We produce enough food to feed every person alive. We waste roughly one third of it — 1.3 billion tonnes every year. Meanwhile 735 million people go hungry. Food waste is the third-largest source of greenhouse gas emissions on earth, behind only China and the United States. This letter is not a climate letter. It is a systems letter: every lever that exists to stop this, from the supply chain to your refrigerator, from policy to technology — and why the two-date labeling system on every package you buy is a quiet catastrophe that a simple fix could largely solve.
Not investment advice. Not a recommendation to buy or sell. Research and long-horizon thinking only. Consult a qualified financial advisor before making any investment decision. Figures cited are sourced from the FAO, UNEP, ReFED, the Harvard Food Law and Policy Clinic, the World Economic Forum, Project Drawdown, and current US and EU legislative databases, current as of June 2026.
Every year, approximately 1.3 billion tonnes of food produced for human consumption is lost or wasted globally. That figure has remained stubbornly consistent across decades of awareness campaigns, voluntary industry pledges, and municipal composting programs. In the United States alone, the USDA estimates that 30-40% of the entire food supply goes to waste — nearly 60 million tonnes annually, more than any other country on earth, at a cost to the American economy of approximately $408 billion per year. At the household level, a family of four wastes an estimated $1,500 worth of food every year. For a single person, the figure runs $300-600. These are not abstract statistics — they are the spinach forgotten in the crisper, the strawberries that went soft before Friday, the leftovers pushed to the back of the shelf.
The climate dimension is even less discussed than the economic one. Food waste generates 8-10% of all global greenhouse gas emissions, according to the IPCC — nearly five times the total emissions from global aviation, a sector that attracts enormous regulatory scrutiny and public attention. Rotting food in landfills contributes up to 14% of global methane emissions, a greenhouse gas more than 80 times more potent than CO₂ over a 20-year window. Project Drawdown, the organisation that ranks climate solutions by impact, places reducing food waste as the single most powerful intervention available to humanity — ahead of electric vehicles, solar power, and plant-based diets combined. 25% of the world's entire fresh water supply is used to grow food that is never eaten. The land that produces wasted food is roughly the size of China.
"Reducing food waste is the number one solution to the climate crisis — above electric cars, solar power and plant-based diets." — Project Drawdown
Walk to any packaged food in your kitchen right now and you will find one or two dates: a manufacturing date, and some version of an expiry — "best before," "use by," "sell by," "enjoy by," "expires on." These phrases are not defined the same way by any two manufacturers, are not regulated consistently in any country except for infant formula, and do not — in the vast majority of cases — indicate the date on which food becomes unsafe. They indicate the date after which the manufacturer no longer guarantees peak quality. A "best before" date on a jar of peanut butter is not telling you the peanut butter will make you sick the next day. It is telling you the manufacturer believes it will taste slightly less good.
The consequence is enormous and entirely preventable. A Harvard Food Law and Policy Clinic and NRDC joint study found that more than 90% of Americans occasionally throw food away based on the "sell by" date out of a mistaken concern for food safety. An estimated 20% of household food waste in the UK — and up to 10% in the EU — is directly caused by misinterpretation of date labels. In Canada, a May 2026 report by food rescue organisation Second Harvest found that 23% of all avoidable food waste — worth over $12 billion annually — is directly linked to date label confusion. "Sell by" dates, it should be noted, are a stock-management tool designed for grocery retailers, not consumers. They communicate when a store should rotate a product off the shelf to ensure residual shelf life after purchase — yet they appear identically to safety-relevant dates on the same packaging, and consumers have no way to tell the difference.
The current two-date system — manufacturing date and a single ambiguous expiry — is structurally insufficient to prevent the waste it creates, because it collapses three distinct and important facts into one or two numbers, forcing consumers, retailers, and food banks to guess which type of information they are looking at. The solution is a mandatory three-date system, where every label carries three unambiguous data points with three different mandatory actions attached to each.
A mandatory framework — not a suggestion, a requirement
AI inventory and demand prediction at retailers and distributors is showing 25-30% reductions in supply chain waste in documented deployments, and up to 50% spoilage reduction in fully optimised cold chains. The bottleneck is adoption, not technology — mid-size retailers and food service operators lag large chains by 5-10 years on this.
Approximately 14% of global food is lost between harvest and retail — almost entirely in countries without adequate refrigerated storage and transport. Cold chain infrastructure investment in Sub-Saharan Africa and South Asia is the single highest-impact supply-side intervention available, reducing post-harvest losses that no amount of consumer education in wealthy countries can offset.
Cosmetic standards at supermarkets cause significant quantities of perfectly safe, nutritious food to be rejected before it reaches shelves purely on appearance. Voluntary "ugly produce" programs exist in most countries and consistently underperform because they require retailer initiative. Making cosmetic rejection of safe food a liability-creating action — rather than standard practice — would shift the incentive structure immediately.
Automated markdown systems that discount products approaching their quality date — already deployed by UK chains including Marks & Spencer and Waitrose — demonstrably reduce waste at the retail stage. Apps like Too Good To Go (operating in 17 countries, 100M+ registered users) extend this model to restaurants and bakeries. The technology exists; adoption is the bottleneck.
The US Bill Emerson Good Samaritan Food Donation Act of 1996 gives federal liability protection to food donors. Thirty years later, most food businesses still do not donate because they lack dedicated logistics, tax incentives are complex, and local liability law varies. Streamlined donation tax credits and simplified liability frameworks would unlock tens of billions in currently-wasted near-expiry food that sits in warehouses because donation is administratively harder than disposal.
South Korea's volume-based food waste fee system — households pay per kilogram of food waste — achieved a 95% food recycling rate and a 30% reduction in household food waste after introduction. France mandated food waste reporting and redistribution for large retailers in 2016 and has seen consistent reductions since. Japan achieved a 31% per-capita reduction using a combination of reporting mandates and public campaigns. The evidence base for mandatory policy intervention is substantially stronger than for voluntary programs.
Mimica Lab and similar companies produce packaging that changes colour in real time based on actual food condition — responding to temperature, time, and bacterial activity rather than a date printed at manufacture. This is a direct replacement for the "best before" date with a physically accurate indicator — and is already commercially available in limited deployment in UK supermarkets. Scaling requires either consumer demand or regulatory mandate.
AI crop monitoring, yield prediction, and precision harvest timing tools are reducing field-level food loss — food that never reaches the supply chain — particularly for fruit and vegetable crops where harvest timing critically affects shelf life downstream. Crops harvested at optimal ripeness last significantly longer post-harvest, reducing waste at every subsequent stage of the chain.
Households generate 60% of all food waste by volume globally. The highest-impact household interventions are practical: buying less more frequently rather than weekly bulk shopping, first-in-first-out pantry rotation, and using senses (look, smell, taste) rather than dates to assess food safety. A family of four following these practices can reduce food waste by 40-50%, recovering $600-750 per year. This requires no technology and no policy — only habit change.
Composting is frequently presented as the solution to food waste. It is the last resort after prevention, redistribution, and repurposing have been exhausted. Composting food waste is significantly better than sending it to landfill — it avoids methane generation — but it recovers none of the water, land, labour, or energy that went into producing the food. The hierarchy matters: prevent first, donate second, repurpose third, compost last.
A growing category of food businesses converts retail and manufacturing surplus into new products — brewery spent grain into flour, surplus bread into beer, cosmetically-rejected fruit into juice and jam. Toast Ale (UK), Rubies in the Rubble, and similar companies demonstrate commercial viability. At industrial scale, this represents a genuine value-recovery pathway rather than a write-off — and reduces the raw material cost of a new product category.
Consumer behaviour at the household level — responsible for the majority of waste by volume — is formed by understanding of where food comes from, how it is produced, and what waste costs. Countries with mandatory food literacy in school curricula consistently show lower per-capita household waste in adult populations. Finland, Japan, and South Korea are the benchmark cases. Food systems education is the longest-cycle but most durable lever on this list.
The reason 1.3 billion tonnes of food has been wasted every year for decades despite universal awareness of the problem is not ignorance — it is that the current system is structured so that the cost of waste falls on those who did not make the decision to waste. Manufacturers set conservative dates that protect their brand from quality complaints; retailers benefit from the turnover that ambiguous labels produce; households pay in wasted money; food banks go short; the climate absorbs the emissions. The incentive structure points entirely away from solutions, and voluntary programs consistently fail to overcome it at scale. Mandatory intervention with real enforcement is the mechanism that works — the South Korea, France, and Japan evidence is unambiguous — and voluntary programs are often the mechanism that politics chooses instead, specifically because it does not disturb the commercial interests that created the problem.
The three-date mandatory system proposed in this letter is not a radical idea — it is a logical extension of what California and the US Food Date Labeling Act are already building toward, with one critical addition. The existing legislation focuses on clarifying what dates mean. The Donate-By date closes a different gap: it creates a legal obligation to act between the quality-date and the safety-date, converting a window in which food currently goes to landfill into a window in which it must go to people who need it. The legislative mechanism exists. The policy will exists in some jurisdictions. What is missing is the mandatory bite that makes it an obligation rather than an option.
We produce enough food today to feed every person alive on earth. The 735 million people who go to bed hungry tonight do not do so because agriculture failed to generate sufficient calories. They do so because those calories were lost in supply chains without cold storage, discarded by retailers confused about the difference between quality dates and safety dates, thrown away by households who trusted a number printed months ago over a smell they could have checked in seconds, and sent to landfill by systems whose incentive structure makes waste cheaper than prevention. Every one of those causes has a known fix. The obstacle is not knowledge. The obstacle is the decision to act.
Long-horizon thinking on capital, technology, and the forces shaping the next decade of wealth creation. Written from first principles. Not consensus. Not noise.