CEA economics are decided outside the facility itself: where does a vertical farm's profit disappear to?
When we think about the profitability of a vertical farm, we reflexively open the facility’s P&L. There is revenue, there are costs, there is depreciation—and that single sheet is supposed to settle whether you are in the black or the red. Yet the discussions that organize CEA (controlled-environment agriculture) as an economy abroad do not place that sheet at the starting point of profitability. They see the place where revenue is generated as sitting outside the facility instead. Even though they are supposed to be handling the same “vertical farm,” where you measure profitability is misaligned from the very start.
How the farm’s profit flows upstream
The cultivation line is running cleanly. The marketable yield is not bad. And yet when you open the books, you fall just short of the black. If you have worked with PFAL (Plant Factory with Artificial Lighting) farms, you probably recognize this lack of payoff. You cut labor, you revisit the electricity costs, and still that last step won’t close.
This is from a time when we swapped out the equipment. At one lettuce farm, the cultivation line approached the ideal and the marketable yield rose, yet the bottom line barely moved. Look closely, and the profit was not staying with the farm. Usage fees were flowing toward the materials maker that holds the seedlings, the dedicated growing media, and the maintenance contract for environmental control, and the farm was in the position of paying that, month after month. Survey the candidates for investment, and they looked the same way. The farm’s bottom line sits a hair above the red, yet the line items paid upstream pile up neatly—this is a scene I have witnessed many times in the field, in PFAL leafy greens.
That profit tends to stay upstream is not a matter of chance but closer to structure—that is my read. Seedlings, growing media, environmental-control maintenance. These three are line items the farm keeps paying every month for as long as it stays in operation, and they are hard to switch out. Once you build a line, changing to different seedlings or media means rebuilding the cultivation conditions from scratch, and swapping the control system changes even the maintenance contractor wholesale. Seen from the materials maker’s side, it takes on something close to the character of recurring income that is hard to cancel—that is what I infer. I have not directly seen the materials maker’s internal P&L itself, but from the shape of the farm’s payments, you can glimpse a structure in which the take tends to stay upstream.
That the farm’s bottom line sits a hair above the red is not necessarily because the marketable yield is poor. It can also be read as a structure in which what you earn flows upstream every month in the form of usage fees. Lettuce is more often delivered to supermarkets and food service under direct contracts than sold as-is on the market, and the unit price is set by negotiated contract. But it is the sales channel that holds the conditions and the continuity of that contract, and the farm tends to be in the position of following along. If the contract is cut or a price reduction is demanded, the take thins out at once. That the cultivation layer tends to run thin is, more than the price itself, also the flip side of “not holding the sales channel.” The materials side, meanwhile, is on the side where its take grows the more the farm’s utilization rises. The effort spent improving marketable yield was, if anything, neatly piling up the upstream payments—you can even read it that way. If you are going to fix the bottom line, before efficiency within the line, you need to question “which layer you secure your take in.”
To avoid any misunderstanding here, let me lay the numbers side by side as they are. That the cultivation layer tends to run thin-margin is itself shown repeatedly in the research. There is a review noting that vertical farming and CEA hold up sufficiently as technology, yet high upfront investment, operating costs, regulation, and a shortage of specialized expertise are cited as barriers to adoption (see 1). What I want to flag, though, is that there is no paper that has proven the very mechanism described so far—that profit gets siphoned upstream. That is, at the end of the day, a read I built from the shape of payments I have seen in the field, in PFAL leafy greens; what the literature backs up only goes as far as “the cultivation layer tends to be thin-margin and high-cost.” Please receive those two things separately.
On top of that, the point that the farm is a price-taker is also backed up from the research side. Even a single-model estimate that examined the profitability of lettuce factories from the angle of scale can be read as showing that the cultivation layer’s profitability carries a fragility that collapses badly with a slight swing in market price (see 3). It means that as long as you can secure scale, the cultivation layer itself does reach the black (we look at this in detail in a later section), yet that profit is extremely vulnerable to swings in price.
The entry point for winning back your take changes with where you came from
Suppose you have noticed the three line items of seedlings, media, and maintenance. Then, is there really a road other than riding on the upstream? This is the next question that arises. Trying to bring seedlings or media in-house takes both technology and cost. From the farm’s standpoint, where in reality is the entry point for winning back your take?
Holding seedlings yourself is, to be sure, a heavy choice to carry. But if you decide upfront that there is only one entry point, you get stuck right there. There are multiple layers in which you can secure your take, and what’s more, the place that is easy to hold differs by where you came from.
Consider a company that entered from manufacturing. There are examples of holding the environmental-control devices and sensors in-house and leaning toward the equipment and operational-data layers. A move where you put a vertical farm’s operating know-how out into the world as data and turn into the side that supplies other farms with the devices and everything around them is also conceivable. For a company that entered from food service or food products, the picture changes. Even if it outsources the cultivation itself, it holds its own stores and processing lines—a sales channel—from the start. Being able to take in leafy greens like fixed-spec lettuce in-house, it is likely to be in a position that can shrug off market-price swings to some degree.
Bringing seedlings and media in-house is no more than one of many entry points. Materials, environmental-control data and operating know-how, the sales channel, equipment. It is more realistic to work backward from which of these layers you are strong in to begin with. Not everyone can hold every layer; the layer contiguous with where you came from is the entry point easiest to win back.
When you compare layers, it helps your judgment to lay out a few yardsticks—that is one way to organize it. How much capital and technology it takes to enter that layer. How much the take you secured is shaken by the economy and by prices. And how contiguous it is with where your company came from. For instance, the materials and equipment layers take heavy investment and technology to enter but tend to become recurring income; the sales-channel layer is hard to enter without the asset of an existing channel but is easy to shrug price swings off in—you can lay them out and size them up that way. Compare them by the same yardstick, and where it is realistic for your company to win back from begins to come into view.
A case study that lined up the business models of domestic vertical farms also reports a tendency for the effective direction of strategy to change with where the company came from (see 4, 5). Companies that entered from manufacturing bring in their environmental-control technology and go after building a new high-value-added market; companies that entered from food service or food services go after locking in stable supply to their own existing channel. Even within the same undertaking of “doing a vertical farm,” there is a diversity that cannot be drawn on the single axis of cost reduction alone. That said, these are tendencies that emerged from a limited set of domestic cases, and one cannot go so far as to say they hold the same way everywhere.
And one more thing. It is not that holding an upstream layer always pays off either. There is also a survey noting that for models that sell devices and data externally as a service, it remains unclear whether the operators are turning a stable profit (see 2). As with the standalone-farm success stories touched on later: the story of a layer that worked out surfaces easily, while the side that did not is hard to see. Whichever layer you choose, whether the take really stays in that layer has to be confirmed separately.
CEA points not to a vertical farm but to an economy
So far I have talked about layers as the farm’s practical work, but where this view really starts to bite is when you read overseas discussions. So let me touch once on the matter of words. Read overseas materials and the term “CEA” shows up frequently. It is an abbreviation for controlled-environment agriculture, often rendered in Japanese as “vertical farm,” but the range CEA actually points to is far wider than that rendering suggests. In this article, we grasp CEA not as a single farm building but as one economy that includes materials, data, and the sales channel. When overseas IR and investment reports speak about the CEA economy, too, they can be read as arguing not about a single farm building’s bottom line but about which layer of that economy you secure your take in.
When you transpose overseas material into your own discussion, mapping a few words across keeps you from getting lost. When the other side says “CEA,” it points not to a single “vertical farm” building on your side but to the whole economy that includes materials, data, and the sales channel. The cultivation layer itself that appears in their discussion corresponds to what you call the “cultivation layer (the farm layer)” on your side, with the materials layer, the data-and-control layer, the sales-channel layer, and the equipment layer strung before and after it. The internal rate of return and net present value they use as investment yardsticks are, in your words, indicators that measure the efficiency and size of profitability: at what annual rate the funds you put in grow (internal rate of return), and how much profit it comes to when you convert the future recovery back into today’s value (net present value). Keep just this mapping in hand, and you can read which layer the overseas IR and investment reports are talking about while translating it into which layer of your own you are talking about.
Open overseas material thinking CEA is “the English for vertical farm,” and you simply cannot decode it. That is because it is used as a word pointing to one economy that bundles together everything from seedlings and media to environmental-control devices and data, equipment, and the sales channel. The side securing its take with devices and data, the side securing it with the sales channel, the side securing it with materials can be read as each setting separate strategies of “where do I earn.” Tie that together with an equals sign as “CEA = vertical farm,” and the discussion of what should have been multiple layers is crushed into a single sheet, and everything looks like “a story about the vertical farm’s bottom line.” Read it looking only at the farm’s bottom line, and the crucial part the writer of the overseas material is really arguing—“which layer to go after”—vanishes wholesale from view.
The view of seeing “which layer to secure your take in” as a structural problem also meshes with research on optimal scale in economics. The economics of a vertical farm is decided by the balance between economies of scale in construction cost and operational transaction costs such as customer acquisition and labor management; it is not a matter of just making it bigger. An optimal scale exists, and going large is not always advantageous (see 3, 6). So profitability surfaces as a problem of placement—“in which layer and in what structure you place your take”—rather than “how much you cut.”
How to read the numbers in standalone-farm profitability and success stories
When you look at the cultivation itself—that is, the farm layer—on its own, how far can profitability be expected? Even if the idea of securing your take by layer has settled in, the question remains of how much room there is for the farm layer itself to go into the black. In overseas investment reports you do see success stories of a high internal rate of return or a short payback, but how should those numbers be received?
To say the conclusion first, it is not that the farm layer cannot reach the black. Rather, there is an estimate that can be read as showing that as long as you can secure scale, the cultivation layer itself goes into the black. In the earlier single-model estimate, for leafy greens like lettuce that turn over fast and fetch a price by spec, it has entered a stage where it holds up commercially, and beyond an average scale (3,000 m2 in that model) an annual profit of 28-37% can be expected on the premise of laying neither subsidies nor tax (see 3). The problem is that this profit is extremely vulnerable to swings in price. In the same estimate, when lettuce’s market price falls by twenty percent, the minimum floor area needed to break into the black leaps from 38 m2 to 1,700 m2. Even though it is certain that profitability holds up in itself, it carries a fragility in which that profitability line shifts by tens of times with a small move in price. So it is not that “leafy-greens farms don’t make money”; reading it as “secure scale and it reaches the black, but profitability is extremely fragile against price swings” is the reception faithful to the source.
Here I need to draw the line of type. The profitability story I just gave is about the closed system (PFAL), grown indoors with applied electricity. What runs commercially in a closed system is mostly leafy greens like lettuce and herbs. Fruit vegetables like tomatoes, on the other hand, already hold up as a core commercial crop in sun-using greenhouse systems. An overseas review, too, reports that greenhouse-grown tomatoes can produce a yield exceeding 500 tonnes per hectare (see 7), and this is a domain where profitability holds up on a separate track from closed-system leafy greens. Lump it together as “the only thing that goes into the black in CEA is leafy greens” and you skip over greenhouse fruit vegetables entirely. Limit the story to the closed system and the accurate reception is, with the type attached, “what runs commercially is almost entirely leafy greens.”
When it comes to staple grains, the story is different again. Grow grains like rice or wheat indoors with applied electricity and the production cost mounts up; on the premise of current market prices, it doesn’t pay off at all. There is an estimate that growing wheat in a vertical farm, its production cost could come to roughly 50 times that of outdoor cultivation (see 8). This is a figure spoken with a “might be” hedge and is not a fixed value, but as a sense of the order of magnitude it shows that, under current prices and current technology, staple grains are structurally hard to make hold up in a closed system (see 7). Incidentally, this non-viability of grains and the argument that “it doesn’t hold up without subsidies” come from research targeting grains in general, including outdoor cultivation. Laying it directly over leafy-greens PFAL farms and reading it as “leafy greens are subsidy-dependent too” goes too far—as we saw earlier, leafy greens (lettuce) in fact achieve the black on the premise of laying neither subsidies nor tax. The grain story and the leafy-greens story need to be kept with their attribution separated.
The numbers you come across overseas of a high internal rate of return or a short payback are not to be denied out of hand. But it is wiser not to take them at face value. Those are often cases that happened to mesh under a specific farm, specific region, or contract conditions, and what’s more, farms that did not work out do not come out into the open. So success cases alone end up being what catches your eye. The more spirited the number, the more it tends to be a case study targeting just a single farm, or carried in a venue whose peer-review reliability cannot be called high, and it is not the kind of thing you can rely on as-is. Such case analyses do not even bring farms that ran into the red onto the table for analysis in the first place. Because of that, you may learn that “it held up under specific conditions,” but how high the probability of hitting is across the whole population stays, in truth, unknown. This is why treating an individual success number directly as your company’s premise misses the mark.
That said, this is not a story of “so give up on the farm.” The leafy-greens farm reaches the black if you secure scale, but is vulnerable to price swings. Accept that fragility as a given, and rebuild it into a design that combines other layers to spread out the take. That is where the starting point lies.
Separating the carving-up of layers from the scrutiny of individual deals
Finally, let me draw one boundary line. The way of grasping things I have described so far—“seeing the take by layer”—is useful as a starting point for re-organizing your company’s position. But this only shows the starting point; it does not hold the answer beyond it, whether an individual deal really pays off. Whether to actually put money into this farm, this equipment, is the domain of scrutiny that works through the location, the power contract, and how firmly the sales channel can be locked in, one case at a time—and that is now the work of specialized due diligence.
Deciding which layer you fight in, and working through whether an individual deal in that layer pays off. These two are better kept as separate work. Skip the former and do only the latter finely, and it is wasted effort if you have mistaken the very layer your company should fight in; conversely, decide only the layer and skip the scrutiny, and it ends up as pie in the sky.
What I have been able to hand you in this article is the carving-up of layers that sits before that. Start by writing out which layer of the CEA economy you place your own company in—farm, materials, data, sales, or equipment—working backward from where you came from. That first question of where the profit was disappearing to becomes, only then, something you can solve with your own hands.