Food Waste in Institutional Dining: The Numbers Behind the Problem

Roughly a third of the food available at the retail and consumer level went uneaten in 2010, the latest USDA estimate we found. USDA’s Economic Research Service estimated that in 2010, 31 percent of the available US food supply at the retail and consumer levels, 133 billion pounds, went uneaten (USDA ERS). That figure covers grocery stores, restaurants, institutions and households together, so it is context, not a benchmark for a dining program.

Most of that waste ends up in landfills. The EPA estimates that in 2019 about 66 million tons of wasted food were generated in the food retail, food service and residential sectors, and that about 60% of it was sent to landfills (EPA). EPA’s food service sector explicitly includes hospitals, nursing homes, colleges and universities, schools (K-12) and office buildings alongside restaurants. In 2018, food made up 24 percent of the municipal solid waste landfilled in the US, more than any other single material in everyday trash (EPA).

US foodservice generates 12.5 million tons of surplus food a year — 78% of the resulting waste goes straight to landfill, worth $122 billion. (ReFED, 2024)

Those figures come from ReFED’s foodservice fact sheet. Two cautions when you quote them. First, ReFED’s “surplus food” is broader than waste: by its own definition it includes food that is donated, fed to animals or repurposed. Second, ReFED reports foodservice as a single sector and does not break out schools, hospitals or campuses, so do not present these as institutional-dining numbers.

In the most detailed campus study we found, plate waste was about twice the size of kitchen waste. Researchers at the University of Missouri collected pre- and post-consumer food waste from four all-you-care-to-eat dining facilities over three months in 2014. Overall, 5.6% of the food reaching those facilities was lost at the pre-consumer (kitchen) stage and 10.7% at the post-consumer (plate) stage (Costello, Birisci and McGarvey). Put another way, roughly two-thirds of the dining-facility food waste in that study was plate waste.

Fruits and vegetables were the largest source of waste by weight, followed by grains. Meat and protein ranked fourth by weight but carried the largest share of the embodied greenhouse gas emissions. This is one university’s all-you-care-to-eat halls, not a national rate. Retail, trayless or made-to-order service will look different, which is one reason to measure your own operation. Campus teams can find more on our higher education page.

Patient plate waste can be substantial, but the multi-hospital data we found is not from the US. A 2025 study in Frontiers in Sustainable Food Systems measured plate waste for 155 inpatients across 16 hospitals in Lebanon between April and September 2023, and found that plate waste represented 31.4% of the amount served (Abiad et al.). Vegetables were the most wasted food group, at 55.5%.

Read it as evidence that patient tray waste is worth measuring, not as a US benchmark. The authors used a convenience sample, measured some plates by weighing and others by visual estimate, and note that some hospitals lacked calibrated scales. Patient menus, diet orders and tray-delivery models differ by country and by hospital. More on hospital food service is on our healthcare page.

USDA’s School Nutrition and Meal Cost Study found that about 21 percent of the calories in school lunches went uneaten in school year 2014–15, with vegetables and milk the most-wasted components (USDA). For the full set of school figures, including audit-program numbers, sample sizes and their limits, see School Food Waste Statistics: What the Data Shows.

In three places: the kitchen, the serving line and the plate. Pre-consumer waste is trim, spoilage and overproduction. Unserved surplus is food that was prepared but never plated. Plate waste is food diners take but do not eat.

Plate waste is where the cost concentrates, because procurement, preparation and service have all been paid for by the time the food reaches the dish return. Unserved surplus sits in between: the food and cooking labor are spent, but not the service.

The drivers differ by sector. Schools plan menus around USDA meal pattern requirements for reimbursable meals, and students do not always eat every component they take. Self-serve university dining lets students choose their own portions. Hospital kitchens plan around individually prescribed patient diets and a census that shifts day to day. Corporate cafeterias often sell meals that diners choose and pay for themselves. How those drivers net out varies from one operation to the next, which is why your own plate-waste data matters more than any sector average.

More than the food. Every wasted plate carries the cost of purchasing, storing, preparing and serving that food, plus hauling it away. The ingredient cost is only the most visible part.

The environmental cost follows the same chain. Food that decomposes in a landfill produces methane, which the EPA describes as at least 28 times more effective than carbon dioxide at trapping heat over a 100-year period. Municipal solid waste landfills were the third-largest source of human-related methane emissions in the US, approximately 14.4 percent of those emissions in 2022 (EPA). Composting keeps food out of the landfill, but only preventing the waste avoids the water, energy and farm inputs that went into producing the food.

The numbers above make the case that the problem is real. They cannot tell you which items your diners leave behind or whether a menu change worked; that takes measurement in your own operation. We compare the options in food waste tracking methods compared.