Why do people underestimate how much food they throw away?

Ask someone to guess how much food their household throws away in a month, and then have them actually keep a running note of everything that goes in the bin — the guess is almost always well below the real number. This is not a special failing of careless people. Households that think of themselves as careful come up short too. The reason has very little to do with honesty and a lot to do with the fact that food waste almost never arrives as a single event large enough to notice, remember, and add to a mental total.

Waste rarely happens as one countable moment

Picture the version of food waste that would actually be easy to estimate: a whole chicken, untouched, thrown out on one identifiable day. That version is rare and, because it is rare, it is exactly the kind of loss that does get remembered — which quietly convinces people that memorable losses are the main kind there are. The far more common version looks nothing like it: a spoonful of rice scraped off a plate, the last third of a bag of spinach gone slimy, the heel of a loaf nobody finished, a sauce jar with two tablespoons dried onto the bottom that got rinsed out instead of used. None of these register, on their own, as "wasting food." Each one is small enough to feel like tidying up rather than losing anything.

Disposal has no weigh-in

A grocery bill gives you a number without asking — it is printed on a receipt, whether you want to look or not. Waste has no equivalent. It goes into a bin that also holds packaging, peelings, and everything else from a normal kitchen, and it leaves the house a day or two later without ever being separated out, counted, or looked at as a category of its own. Compare that to the one thing that reliably does get an accurate estimate: what a household actually eats, because eating is a series of countable events you were present for. Throwing food out is comparatively passive and quick, often done while thinking about something else entirely, and it leaves behind no receipt, no photograph, nothing that would let you reconstruct the tally later even if you wanted to.

Some of it never gets classified as "food" at all, even in your own head

A good portion of what actually gets thrown out sits in a mental category that is not "wasted food" but "kitchen mess" — trimmings, an odd end of a vegetable, a sauce that separated, a container nobody could identify by smell alone. Calling that waste requires a small act of reclassification most people never perform, because it does not look like the thing the word "waste" brings to mind, which tends to be a whole, recognisable, still-good-looking item. The half-used jar and the shrivelled pepper both count. Only one of them looks like it counts.

Memory keeps score of the wrong sample

The instances that do get remembered are the dramatic ones — the takeout container found weeks later, the whole loaf gone furry, the moment that made someone say "we really need to stop doing this." Those moments are genuinely rare, which is exactly why they stand out enough to be recalled. But rare and memorable is not the same as representative. When someone estimates their household's waste from what they can actually recall, they are unknowingly sampling from the small pile of dramatic incidents and ignoring the much larger pile of small, forgettable ones that never got flagged as noteworthy in the first place. A biased sample produces a biased guess, every time, no matter how honestly the person is trying to answer.

This is a visibility problem, not a character one

None of this means people do not care, and it is not a reason to feel worse about the number once it is finally counted — feeling bad about a total you could not see coming does not do anything useful with it. What actually closes the gap between the guess and the reality is not more self-scrutiny; it is something that happens to be counted automatically, at the moment it happens, rather than something a person has to notice, categorise, and remember to tally later. That is a structural fix, not a willpower one, because the entire reason the guess was wrong in the first place is that the underlying events were never designed to be countable.

Munch does not put a number on how much a household wastes, and it is not built to. What it does track is the plainer thing underneath the estimation problem: what is actually in the fridge, freezer and pantry right now, with its own date, updated the moment something is opened or used up rather than reconstructed later from memory. A lot of the small, forgettable losses above happen precisely because nobody had an accurate, current picture of what was already there to work from. Seeing that picture clearly, before the fact, is a different thing from tallying the losses after it — but it removes a fair amount of the reason those small losses happen in the first place.