Farming Upwards

AFew agricultural proposals of the past two decades have been promoted with as much confidence as the indoor farm, in which crops are raised on stacked trays under coloured diodes and never touched by weather. The case for it is easy to summarise. A hectare of floor space, repeated ten times upward, is said to do the work of many hectares of field, while water that would otherwise drain away is captured and returned to the roots. Pests are excluded, so nothing needs to be sprayed, and because the building can stand inside the city it supplies, the lorries that haul salad across a continent are supposed to become unnecessary. Investors found the argument persuasive: between 2014 and 2021 rather more than three billion euros went into European ventures of this kind. What has proved harder is the arithmetic that ought to sit beneath such claims, since almost every figure quoted depends on a choice of yardstick that is seldom stated aloud.

BThat choice of yardstick is worth pausing over. Yields from indoor farms are conventionally reported per square metre of growing surface, so a building with twelve tiers records twelve times the figure of the field against which it is set. Expressed instead per square metre of ground actually occupied, the advantage remains real but shrinks considerably; expressed per unit of capital invested, it frequently disappears altogether. The saving in water is calculated in the same generous manner, counting the volume delivered to the roots while ignoring what is consumed by the power station that lights the room. Transport deserves similar care, since studies of European supply chains consistently find that moving vegetables from farm to shop accounts for a modest fraction of their total burden, about 6 per cent for leafy crops travelling under 400 kilometres. None of this shows the indoor farm to be a poor idea; it shows that the comparison usually offered is not really a comparison at all.

CThe most careful attempt to redo the arithmetic has come from Marta Quintela, an agricultural economist at the Ravensholt Institute of Food Systems, who maintains that the industry has been defending the wrong claim. Between 2017 and 2023 her group collected metered consumption from 34 commercial facilities across nine countries, rather than the design estimates on which earlier reviews had relied. The median came to 8.4 kilowatt hours of electricity for every kilogram of lettuce leaving the building, against roughly 0.3 for the same crop grown outdoors in season. Water told the opposite story, with indoor crops using about 94 per cent less than field equivalents, a saving Quintela regards as genuine and unusually well documented. Her conclusion is narrower than either camp would like: the indoor farm reduces the burden on land and water while transferring it to the grid, so its climate case holds only where the electricity is already almost free of carbon. Where coal or gas supplies the current, she insists, the salad grown indoors carries a heavier footprint than the salad flown in.

DTeodor Marek, an engineer at the Brenner Centre for Controlled Environment Agriculture, contends that measurements gathered over six years describe a technology which no longer exists. The efficiency of horticultural diodes, expressed as the light usefully delivered for each unit of electricity drawn, improved by roughly four fifths between 2014 and 2022, and the four newest sites he has monitored consume 4.9 kilowatt hours for every kilogram harvested. Quintela's median, on his reading, is an average of obsolete equipment and therefore says nothing useful about what the industry should build next. The objection has force, although his own evidence comes from four sites that all grow basil, all opened after 2021, and all draw current from hydroelectric grids in northern Europe. Marek concedes, too, that the curve he relies on is flattening, since photosynthesis sets a physical ceiling on how much a plant can do with additional light. A further doubling of efficiency, he accepts, is not available.

EElectricity, in any case, is not what has closed most indoor farms. Between 2022 and 2024 at least eleven European operators entered insolvency, and the post-mortems point consistently at the cost of the building rather than the cost of running it. A fitted-out growing hall absorbs many times the capital per square metre that a heated greenhouse requires, and that outlay must be recovered from whatever is grown inside. Only crops that combine a high price per kilogram with a short cycle can carry it, which is why lettuce, herbs, micro-greens and strawberries appear in almost every business plan while wheat and potatoes appear in none. Quintela puts the point sharply: the sector is not competing with agriculture at all, but with the intensive greenhouse, which delivers a good deal of the same control for a fraction of the outlay. Marek does not dispute the comparison, though he observes that greenhouses need land with good light, and such land is scarce beside large cities.

FWhat survives of the original promise is smaller than its advocates hoped and more solid than its critics allow. Both researchers now describe the same narrow band of uses, though they arrive at it from opposite directions. Where sunlight is scarce for months, where imports arrive by air, or where a crop must be certified free of disease, an enclosed room is not an extravagance but the only practical option, and Iceland supplies most of the examples. The most convincing case, however, has nothing to do with feeding cities, since it involves raising seedlings that spend three weeks indoors and are then planted out to finish under the sun. In that arrangement the expensive lighting is applied only during the phase in which a plant is small, uniform and valuable, and the field supplies the rest of the growth free of charge. Neither Quintela nor Marek regards this as the revolution once advertised, and both suspect that it is the version most likely to last.