What would it actually take to scale up urban farming?

3 minute read

Published:

A recurring tension in debates about urban agriculture is the question of scale. Advocates argue that urban farming could play a meaningful role in feeding cities. Skeptics counter that cities simply don’t have the space. Both positions have historically been more comfortable with assertion than evidence. A new paper I published with Dimitrios Gounaridis, Joshua Newell, Benjamin Goldstein, Silvio Caputo, and a large international team in Landscape and Urban Planning tries to bring some empirical clarity.

We focused on five cities in the Global North — London, New York City, Paris, Dortmund, and Gorzów Wielkopolski — cities that I’ve been working on since the early days of the FEW-meter project. The question was simple: if we expanded urban agriculture to all of the space that’s physically suitable, how much food could these cities produce, and what would it actually require?

The first finding that surprises most people is how much suitable land there is. We found that between 12 and 24% of city area is physically suitable for urban agriculture. That’s not a rounding error — it’s a substantial share of the urban landscape. Individual and home gardens make up the dominant portion of that expandable space in every city we looked at, which says something important about where the realistic opportunity lies: not primarily in commercial urban farms or greenhouses, but in the millions of backyards, balconies, and small plots that already exist but aren’t currently growing food.

The food production potential is large under optimistic scenarios. If expanded UA achieved high productivity levels, it could supply between 16 and 95% of current vegetable demand in these five cities. There’s a huge range there, and it reflects real variation in city density, climate, available land, and existing agricultural practice. The point isn’t to take any single number as gospel, but to establish that the physical potential is not negligible — and in some contexts, is actually quite striking.

But here’s the most important finding, and the one I think has the clearest implications for policy: the binding constraint is not land. It’s people. Our analysis consistently found that labor availability — the willingness and capacity of residents to actually grow food — is the primary factor limiting how much of that theoretical potential is realizable. Even if you mapped out every suitable square meter and secured its tenure, the question of whether enough people would take it up and grow productively determines the actual ceiling.

This is a significant reframe for urban agriculture policy. If labor is the binding constraint, then the interventions that matter most are not about finding more land (though land security matters). They’re about supporting growers — technical assistance, training, extension support, and the cultural and institutional environments that make urban farming accessible and appealing. On the resource side, expanded UA would demand modest electricity but potentially significant potable water in some cities; pairing expansion with composting and water-harvesting infrastructure would help manage both.

The scaling potential is real. Getting there is primarily a question of will — and of supporting the people who do the growing.


Hawes, J. K., Gounaridis, D., Newell, J. P., Goldstein, B. P., Caputo, S., Cohen, N., Fargue-Lelièvre, A., et al. (2026). Assessing the potential to scale-up urban agriculture in the Global North. Landscape and Urban Planning, 272, 105657. https://doi.org/10.1016/j.landurbplan.2026.105657