Food miles
A Socratic walk-through of food miles — reasoned out one step at a time, not lectured.
The question we started with
THE QUESTION #Why can food grown far away sometimes cost the planet less than food grown nearby?
The rule feels self-evident: the tomato that travelled two thousand kilometres must have cost more carbon than the one grown down the road. Distance is visible, it is easy to print on a label, and it is the part of the journey we can actually picture. But it is worth asking what the distance is standing in for — and whether the thing it stands in for is the thing that dominates.
Reasoning it through
REASONING #Take the whole life of a food and ask where energy and emissions enter it. There is the farm: fertiliser manufacture, which is energy-intensive; nitrous oxide from fertilised soil; methane from ruminants and from flooded rice; fuel for machinery; heating and lighting if it is grown under glass out of season. There is land: clearing forest or draining peat releases carbon that had been stored for centuries. Then processing, packaging, refrigeration, retail — and, last, transport.
Now ask the obvious follow-up question. Of that whole list, how big is the last item? Across food globally, transport accounts for roughly six per cent of food-related greenhouse emissions — a real share, but a minor one beside production and land use, which together dominate. That single fact does most of the work here. If transport is a small slice, then a difference in transport cannot easily outweigh a difference in how the food was grown.
So consider what a "local" food sometimes has to do to be local. Grow a tomato in a northern winter and you must heat a greenhouse, sometimes for months. Grow the same tomato in southern Europe in the open field and the sun does that work for nothing — after which you truck it north. Which wins? It depends entirely on the heating fuel, the greenhouse's efficiency, the yields, and the transport mode, but the comparison is genuinely close, and several European life-cycle studies have found the heated local crop coming out worse. The point is not that imports always win. It is that the outcome is decided by the growing conditions, not by the kilometres.
Which brings us to the sharper distinction, the one "food miles" conceals. Are all kilometres alike? Not remotely. Moving a tonne of goods one kilometre by container ship emits far less than doing it by road, and one to two orders of magnitude less than doing it by air. A ship is slow, enormous, and moves through water with very little energy per tonne carried; a plane must hold itself up. So a container of bananas crossing an ocean can carry a trivially small transport footprint, while a box of fresh asparagus flown the same distance carries an enormous one. Air freight moves only a tiny fraction of food by weight — overwhelmingly the highly perishable: berries, green beans, some fresh fish and flowers — yet it contributes a disproportionate share of transport emissions.
That reframes the practical advice. "Is it local?" is a weak question. "How was it grown, and did it fly?" is a strong one — and, for most diets, the strongest question of all is not about geography but about what the food is, because the spread between food types is much larger than the spread within one food type by origin.
The analogy
THE ANALOGY #Think of the cost of a wedding. The invitation lists the venue's distance, so guests argue about petrol. But the venue, the catering, the flowers and the photographer swamp the travel for almost everyone — and among those who did travel, the couple who drove and the couple who flew are not in the same conversation at all. Distance was the only number printed on the invitation, so it became the number people discussed.
Wedding costs are all borne in money by identifiable people, whereas emissions from land clearing, methane and fuel are not paid by anyone directly — which is precisely why the visible, countable number came to stand for the whole, and why nobody was forced to notice the error.
Clarifying the model
THE MODEL #Several refinements matter, and one of them cuts against the argument above.
First, "food miles are a small share" is a global average across all food. For a particular product it can be false. A dense, low-emission staple such as dried grain grown with little fertiliser has such a small production footprint that transport becomes a large fraction of a small total; a bulky, watery, low-value food moved a long way by road can also tilt that way. The claim is about where the leverage usually sits, not a law.
Second, the studies that made this argument popular deserve their caveats. The most-cited example — New Zealand lamb shipped to Britain emerging with a lower footprint than British lamb, chiefly because of pasture-based farming and lower on-farm energy use — came from researchers in the exporting country, and its assumptions about farm systems and allocation were contested. Life-cycle assessments are sensitive to boundaries: what is counted, over what time, and how emissions are divided between a carcass and its by-products. Two honest studies of the same food can differ severalfold. Ranges and directions are trustworthy here; precise single numbers usually are not.
Third, carbon is not the only reason to buy locally, and this argument does not touch the others — freshness, varietal diversity, local livelihoods, farmland kept in production near cities. What it does say is that where the reason offered is climate, that reason is usually weaker than it sounds, and the effort would do more good redirected at what is eaten, how much is wasted, and whether it flew.
A picture of it
THE PICTURE #How to readEach slice is a share of the emissions attributed to food as a whole, not of any one product. The thing "food miles" is about does not appear as a slice at all — it sits inside the supply-chain slice alongside processing, packaging, refrigeration and retail, and is roughly six per cent of the whole. Set that against the three larger slices, all of which are decided on the farm and on the land, and the argument becomes visual: changing origin moves a fraction of one slice, while changing what you eat moves between slices.
What became clearer
WHAT CLEARED #Food miles took a genuine cost and mistook it for the main one. Transport is real, and the difference between sea and air freight is dramatic enough to matter to a shopper — but the bulk of a food's climate cost is incurred before it moves at all, in fertiliser, livestock, and cleared land. That is why a heated local greenhouse can lose to an open-field import, and why the useful questions are what the food is and whether it flew, rather than how far it came.
Where to go next
ONWARD #- Why the emissions spread between food types dwarfs the spread within a type, and what that implies for dietary advice.
- How life-cycle assessment boundaries are chosen, and why two honest studies of the same food can disagree severalfold.
Key terms
TERMS #| Term | What it means |
|---|---|
| Food miles | the distance a food travels from producer to consumer, used as a proxy for its environmental cost. |
| Life-cycle assessment (LCA) | an accounting of a product's environmental burdens across its whole chain, from inputs and land through production, transport, use and disposal. |
| Tonne-kilometre | one tonne of goods carried one kilometre; the unit in which the efficiency of transport modes is compared. |
| Land use change | emissions released when land is converted for agriculture, such as clearing forest or draining peat. |
Every term the collection defines is gathered in the glossary.