Why Vertical Farming Keeps Losing Money

In a six-month window between April and September 2023, four of the world's most prominent vertical farming companies filed for bankruptcy. AeroFarms (Newark, NJ). Kalera (Orlando, FL). AppHarvest (Morehead, KY). Infarm (Berlin, Germany). Combined, these four had raised over $1.7 billion in venture capital and public market funding. Combined, they grew lettuce.

They were not alone. 5th Season, a Pittsburgh robotics-focused vertical farm backed by $35 million in funding, shut down abruptly in January 2023. Several smaller operators in Europe and Southeast Asia quietly wound down operations through the same period. The wreckage was not random. It followed a pattern visible in the unit economics from the start.

The vertical farming pitch was seductive: grow food in stacked indoor layers, using 95% less water than field agriculture, achieving 390x yield per acre-equivalent, with no pesticides, no weather risk, and no seasonal constraints. Every claim was technically true. The Association for Vertical Farming and Grand View Research confirmed the numbers. But technical truth and economic viability are different things, and the gap between them turned out to be measured in electricity bills.

This is not a story about technology that does not work. It is a story about technology that works but cannot yet pay for itself growing the wrong crops. The distinction matters for the green transition, because understanding why green projects fail is as important as celebrating the ones that succeed.

Venture capital poured into vertical farming during the same 2020-2022 window that inflated alternative protein, SPAC-mania, and the broader climate tech bubble. The six most prominent companies raised over $2.6 billion. Here is what happened to that capital.

AppHarvest is the most instructive case. The company went public via SPAC merger in February 2021 at a $2.5 billion implied valuation. By July 2023, it was in Chapter 11. The stock went from over $40 per share to effectively zero. AppHarvest did not operate pure vertical farms (it used large controlled-environment greenhouses), but it shared the same fundamental problem: the operating cost of replacing sunlight with artificial systems exceeded the revenue from commodity produce.

Infarm's trajectory was different in structure but identical in outcome. The Berlin-based company raised $604 million to deploy modular vertical farming units inside supermarkets across Europe. The modular approach was supposed to eliminate distribution costs by growing lettuce at the point of sale. It did not eliminate electricity costs. Infarm cut 500 employees (half its workforce) in late 2022 and filed for insolvency in 2023.

Every vertical farming pitch deck contains the same set of impressive metrics. All of them are real. And none of them address the core economic problem.

The disconnect is not subtle. A field in California's Salinas Valley produces lettuce at roughly $0.80-1.50 per pound, using free sunlight, natural air circulation, and rain supplemented by irrigation. A vertical farm in Newark produces the same lettuce at $2.50-5.00 per pound, paying for every photon of light, every degree of temperature control, and every cubic foot of air movement.

The 390x yield-per-acre figure is real but misleading. It counts vertical stacking (8-16 growing layers) and year-round production (12 harvests per year vs. 2-3 outdoors). The productivity per unit of growing area is genuinely extraordinary. But productivity per dollar of input cost is the metric that determines whether the business survives, and on that metric, field agriculture still wins for commodity produce.

The sun delivers approximately 1,000 watts per square meter of light energy to the Earth's surface at peak. It does this for free. A vertical farm must replicate this light artificially, paying for every watt-hour. Even with modern LED efficiency gains (up from 1.7 to 3.0+ micromoles per joule since 2015), electricity remains the dominant cost driver.

Here is what the cost stack looks like inside a typical vertical farm operation:

The electricity problem has a cruel feedback loop built in. LED grow lights convert electricity into light, but they also convert a large fraction into heat. That heat must be removed by HVAC systems, which consume additional electricity. In hot climates, the cooling load alone can equal the lighting load. A vertical farm in Phoenix or Dubai faces higher cooling costs than one in Scandinavia, but the Scandinavian facility faces higher heating costs in winter to maintain growing temperatures.

The good news: LED efficiency is improving steadily. Efficacy has roughly doubled since 2015, from 1.7 to over 3.0 micromoles per joule. Simultaneously, the cost of renewable electricity continues to fall, with solar LCOE now below $0.03/kWh in optimal locations. The combination of cheaper light and cheaper electricity is narrowing the gap. But narrowing is not closing. At current trajectories, vertical farm lettuce reaches cost parity with field lettuce only in regions with extremely cheap renewable power (under $0.04/kWh) and extremely expensive field-grown produce (northern Europe, Gulf states, island nations).

Vertical farming's viability depends on three crop characteristics: retail price per kilogram (must be high enough to absorb the electricity premium), growth cycle (shorter is better for facility throughput), and plant architecture (must stack efficiently in trays). Only a narrow band of crops passes all three filters.

The grid reveals the fundamental constraint. Only four cells are green. Vertical farming's addressable market is not "all of agriculture" as pitch decks implied. It is premium leafy greens and herbs in urban markets where consumers will pay a significant premium for local, pesticide-free produce. That is a real market, but it is a fraction of the total food system.

The "impossible" category is not a matter of waiting for technology to improve. Growing wheat indoors would require roughly 4,000 kWh of electricity per tonne of grain. At $0.10/kWh, that is $400 in electricity alone for a product that sells for $300/tonne on commodity markets. The math does not work at any plausible electricity price. The same logic applies to rice, corn, potatoes, and every other staple crop that feeds the world.

This is where the comparison to other food system innovations becomes instructive. Precision fermentation faced a similar scaling wall: the technology works in the lab, but unit economics at industrial scale proved far harder than projected. Black soldier fly farming, by contrast, converts waste into protein using organisms that do not require artificial lighting, giving it a fundamentally different cost structure.

Not every vertical farming company went bankrupt. The survivors share three characteristics that the failures lacked.

Crop discipline. Bowery Farming (New York) focused exclusively on leafy greens and herbs from the start. It never tried to grow tomatoes or strawberries. Plenty (Compton, CA), backed by $941 million from SoftBank, similarly restricted its crop range to baby greens and leafy varieties after early experiments with broader produce. The survivors accepted the narrow viable-crop band rather than trying to force economics that physics would not allow.

Energy strategy. Companies that survived either co-located with cheap renewable power or negotiated long-term power purchase agreements. Some Nordic operators benefit from electricity prices below $0.04/kWh via hydroelectric and wind power. In the Middle East, sovereign-backed vertical farms use subsidized solar power in regions where field agriculture is impossible due to water scarcity. The survivors treated electricity as a strategic input requiring dedicated procurement, not a line item to be absorbed.

Capital discipline. The bankruptcies followed a pattern: raise large rounds, build facilities ahead of demand, expand geographically before proving unit economics in a single location. Infarm's strategy of deploying modular units in supermarkets across Europe meant it was scaling distribution costs before proving that any single unit was profitable. The survivors built one facility, proved the unit economics, and only then considered expansion.

The Asian market tells a different story. Japan, South Korea, and Singapore have vertical farming industries that predate the Western VC boom. Japan alone has over 200 plant factories. These operations tend to be smaller, focused on extremely high-value crops (wasabi, shiso, premium herbs), and integrated into existing food distribution systems rather than positioned as disruptive alternatives. Their survival owes more to cultural premium pricing and land scarcity than to solved unit economics.

Vertical farming is not a failed technology. It is a technology that was funded as if it could replace field agriculture, when it can only supplement it for a narrow range of high-value crops in specific geographic contexts.

The 2023 bankruptcy wave was not a verdict on indoor growing. It was a verdict on venture capital's tendency to fund TAM fantasies over unit economics. The companies that died were the ones that tried to grow commodity produce indoors, expanded before proving profitability, or assumed electricity costs would fall faster than they did.

Two structural tailwinds are slowly improving the economics. First, LED efficacy continues to improve at roughly 10% per year, directly reducing the largest cost input. Second, renewable electricity prices continue to fall globally, with solar LCOE now below $0.03/kWh in optimal locations. The crossover point, where vertical farm lettuce reaches cost parity with field lettuce in major markets, is approaching but has not arrived for most geographies.

For the green transition, the lesson is precise: not every technology that reduces environmental impact is economically viable at every scale for every application. Vertical farming uses 95% less water and zero pesticides. Those are real environmental benefits. But regenerative agriculture improves soil health, sequesters carbon, and reduces input costs across all crop types, not just premium leafy greens. The two approaches are complementary, not competing. Vertical farms for urban herb and leafy green production. Regenerative methods for the other 99% of the food system.

The $1.7 billion in bankruptcies is the cost of learning where the technology's economic boundaries actually are. That knowledge is now priced into the market. The next generation of vertical farming companies will build inside those boundaries. The ones that respect the physics will survive. The ones that pitch TAM slides showing "total global agriculture" will not.