Good evening. I have had a busy week and a busy day today, and so today’s post will be short. We will take a brief look at vertical farming, which is the practice of stacking crops on top of each other. Because of this, the primary source of light for the crops must be artificial. This is the main distinction between vertical farming and greenhouses; both are forms of indoor farming, but the latter relies mostly on sunlight, while the latter relies mostly on artificial light.
For this reason, the energy bill from lighting for vertical farming is high—prohibitively high in most cases. Although the problem of vertical farming’s high energy needs is not a new revelation, several recent articles have highlighted this issue.
Vertical farming is meant to address the land, water, and pesticide requirements of conventional agriculture, and it does far better on each of the three metrics. It comes at the cost of much greater energy consumption and also of labor intensity. An analysis by Faraz Moghimi finds the following.
It is not clear to me what a “unit” is—costs are merely given per-kilogram—but most vertical farming ventures grow leafy green vegetables and herbs. Staple crops such as wheat, rice, barley, and soy can theoretically be grown in a vertical farm as well, but the cost is such that these crops are out of the question for now. It also seems that this cost comparison is incomplete, as capital costs are not shown.
It can nevertheless be seen what vertical farming eliminates most water costs. Worldwide, about 70% of freshwater withdrawals go to agriculture; the rest goes to industry, particularly cooling for thermal power plants, and municipal uses such as watering lawns and drinking water. Reducing the water needs of agriculture would seem to be the most effective solution to water challenges, and vertical farming is clearly one way to do that, even if it doesn’t look like a cost-effective way.
On the water front, consider that lettuce requires an estimated 225-275 liters of water for each kilogram of lettuce produced. Desalinated water typically costs around $0.50 to $1.50 per cubic meter. A cubic meter is 1000 liters. If we take the high estimates of the water needs of lettuce and the cost of desalinated water, then for an otherwise conventional farm to supply all its water from desalination would cost around $0.41/kg, which is about half of the gap between the cost of conventional and vertically farmed crop. It thus appears that desalination is a more cost effective solution to water scarcity than vertical farming, even for leafy greens. This is true insofar as water scarcity is a major problem; this is much more true in water-parched areas such as California, South Africa, and the Middle East than in many other parts of the world.
What about land? At first glance, land saving would look like a compelling argument for vertical farming. According to Our World in Data, over 90% of human land use goes to agriculture (OWID fails to take into account land for mining, but I think this statistic would hold even if mining was accounted for). According to the IUCN (International Union for the Conservation of Nature) Red List, agriculture and aquaculture are by far the most significant factors behind extinctions and extinction risks.
But the same OWID data indicates that 77% of agricultural land use goes to livestock, including animal feed, and of the 23% for plant based crops, about 12% is for fruit and vegetables, so that would be less than 3% of all agricultural land use goes to crops that would be plausible candidates for vertical farms, and to describe all fruits and vegetables as plausible candidates is very generous.
I have mused on what effect Pigouvian pricing of land use might have on basic land questions, such as whether it would make the economics of vertical farming significantly more favorable. Last year, I wrote briefly on practical problems with ecosystem service pricing, which is the idea of adding a price on land use that captures the monetary value of the ecosystem services that such land use would compromise. I still contend that ecosystem service valuation will never be practical, but we can at least attempt to ballpark a monetary value of the land would that vertical farming would save.
With extreme imprecision, we could say, based on the estimate of Costanza et al., that the land to be converted to cropland has an ecosystem service valuation of around $5000 per hectare. There is an estimate that a 400 acre lettuce farm yields 4000 tons of lettuce per year. If we put these numbers together and do unit conversions, we get a value of around $0.20 per kilogram of lettuce grown. That’s significantly less than the cost differential between vertical farming and conventional agriculture, but at least it is plausible that with significant cost reductions in labor, which could happen with advances in robotics that are on the horizon, and cost reductions in energy, which might happen e.g. with nuclear power, ecosystem valuation would be a decisive factor. However, for reasons beyond today’s scope, I suspect that the Costanza et al. numbers are much too large.
I like the idea of vertical farming. Dickson Despommier’s farm skyscrapers look cool, and there are a whole bunch of less sexy but more practical ideas such as using otherwise empty buildings. But vertical farming seems like a solution in search of a problem. The case for vertical farming as a way of saving water and land isn’t really there, and there are more effective ways to accomplish these goals, such as greenhouses.
Quick Hits
There are a lot today.
In response to last week’s post, a reader pointed me to Kevin Kuruc’s research, which focuses on the economic effects of population size, as well as topics including economics of animal welfare and many others. This recent paper, for instance, demonstrates that the impact of birth rate decline on greenhouse gas emissions will be of insignificant magnitude for the foreseeable future.
The same reader asked about the “urban doom loop”. The reference I included on the topic was a bit of a throw-away reference. It does not appear to me that it is a very rigorous concept—even the name “urban doom loop” doesn’t sound rigorous. This is something that I would want to explore better if I make another attempt at the topic.
After decades of the “rise of China”, China’s share of world GDP share is declining. Earlier I noted with some surprise that China had posted a 5% GDP gain in the last quarterly statistics, but the article asserts that the 5% number is too high due to some improper inflation adjustment. I don’t understand all the details. It wouldn’t surprise me, though; many official statistics from China are not reliable. The article speculates that China’s more conciliatory attitude, on display at the recent summit in San Francisco, may reflect a realization of the limits of their power.
Adam Stein and Ted Nordhaus at The Breakthrough Institute discuss the problems that advanced nuclear energy, including NuScale’s small modular reactors, is running into. The money quote comes at the end.
For these reasons [interest rates, commodity prices, supply chain issues, improper regulation, etc.], the cancellation of the NuScale project with UAMPS should be a wake-up call not only for nuclear advocates, the nuclear industry, and nuclear allies in both parties, but also for anyone who cares about mitigating climate change, increasing energy security, and improving public health.
Jason Crawford at Roots of Progress considers how the effective altruism movement got off track on risks from artificial intelligence, and how a movement called “effective accelerationism” (e/acc), which has dismissive of the threats, has great shortcomings of its own. I first heard of e/acc after reading Marc Andreessen’s thoroughly unimpressive manifesto.
A few months ago, Justin Trudeau of Canada went on a limb by accusing the Indian government of assassinating a Canadian national and Khalistan separatist Hardeep Singh Nijjar. Now the United States has accused India of attempting to assassinate another separatist in the U.S. The Biden administration has made good relations with India a high priority, as this is seen as a crucial element of their containment policy against China. If Chinese power wanes and U.S.-China relations improve, for reasons (very speculatively) noted above, then so too will wane the rationale for tolerating bad Indian behavior.
There have been several significant deaths recently, including that of Henry Kissinger. News reports on Kissinger’s death have highlighted his brilliant mind and amoral realpolitik approach to foreign policy. Foreign Affairs has a good profile.