Ethanol 2: What Does an Acre Produce?

Farming, ethanol and the economics of living landscapes — essay 2 of 4

Thorsten Arnold is an environmental scientist and farmer in Grey County, Ontario, working with the National Farmers Union on its ethanol campaign.

A farmer grows 180 bushels of corn per acre instead of 150. Michael Grunwald sees less land required for corn [1]. The farmer sees a competitive advantage in serving mortgage payments. The bank sees greater earning capacity. The machinery dealer sees a farm that is susceptible to buying bigger machinery. Of these four readings, only one protects land — and it is the only reading where no one is paid to act on it.

The first essay in this series asked what mechanism turns a 10% yield gain into 10% less land farmed, and answered: none. This essay explains why. Farmers run on three treadmills, each pushing toward more production and more acreage even as yields rise. They lock together into one system vortex. And ethanol turns all three at once.

Treadmill one: technology

Willard Cochrane described the first treadmill in 1958 that was named after him [2]. A new technology — a hybrid, a larger planter, a new herbicide, tile drainage, precision guidance — lowers unit costs. Early adopters profit while prices still reflect the old costs. Then everyone adopts. Total output rises, and because food demand is price-inelastic, commodity prices fall. Any financial gain is competed away to buyers. Non-adopters now earn less than before, and must adopt or exit.

This treadmill manufactures overproduction. Each farmer is a price-taker: cutting their own output does not raise the market price, it only cuts their own income. When margins tighten, the rational response is to produce more, spreading fixed costs over more bushels. What is rational for each individual becomes ruinous for everyone. The farms that exit are absorbed by neighbours who can finance the next round. Grunwald’s own figures describe this without naming the mechanism: corn yields doubled over three decades while corn prices fell 85%, and America lost half its farms while output tripled [1].

This changes the meaning of yield. In commodity agriculture, yield tecnnology is not about land-saving. It is about a competitive edge over the neighbour. The farmer who produces more cheaply survives lower prices, outbids neighbours for rented ground, buys the next farm and spreads costs over more production. Higher yield reduces the hectares mathematically needed per tonne of corn. It does not reduce the hectares the successful farmer wants to farm.

Treadmill two: land and debt

Cochrane saw a second treadmill early on: Expected farm returns — from commodity prices, support programs, tax treatment and expected productivity — capitalize into land values and rents [3]. Early owners capture the windfall; everyone who buys or rents afterward pays for it, and must earn it back. During landprice inflation, owned land leverages new land purchases. The next generation however does not inherit this old economics: it inherits a new mortgage load on land that stiffly and mercilessly drives farmers to squeeze every dime possible out of the land.

This treadmill converts expected asset value into a fixed payment obligations, due every year, in drought and glut alike. The mortgage is denominated in dollars; corn and soybeans appear on the cash-flow statement; groundwater recharge and flood storage and pollinator habitat and carbon storage do not. The hedgerow, ther shelterbelt, the woodlot, the wet corner carry the same land cost per acre as the field around them, and earn nothing on paper. Removing them is the cheapest “new” acre a farmer can acquire: no auction, no commission, no purchase price.

At least, so it looks from the tractor seat. Sub-field analysis in Iowa found that a substantial share of cropped area — edges, wet spots, poor soils — loses money in most years [4]. But the pressure is real. Ecosystem services pay neither the land mortgage nor the tractor loan.

Treadmill three: machinery

The third treadmill is related to Cochrane’s but economically distinct. Modern farm machinery is lumpy capital – you can only buy an entire combine even if can do more area than you own. FCC borrows against farmland value, so landprice inflation gives farmers almost unlimited access to credit for buying machines, and sales people know that. Yet, depreciation, financing and insurance are paid whether the machine runs at capacity or sits in the shed. If the next combine is sized for 3,000 acres and its owner farms 1,500 only, machinery is overcapitalized – the farmer owns more productive capacity than it can utilize.  Once the machine is bought, every additional acre lowers the fixed machinery cost per acre. Empirical farm records confirm the economics: in Illinois, machinery investment runs about US$847 per acre on farms under 500 acres and US$668 on farms over 2,000 acres [10]. In Ontario, a back-of-the-envelope estimate indicates that large tractors alone could likely work its cropland several times over – exact data is not available. And a modern combine needs about 2,800 acres a year to pay for itself on farms that average 249 acres.

The result of overcapitalization is pressure for farmland expansion:

competitive pressure → larger equipment → excess machine capacity → search for more acres → lower fixed cost per acre → expansion → the next round of investment.

Machinery also has a preferred landscape geometry: large, regular fields and long straight passes. Hedgerows, wet spots, isolated trees, irregular corners and small fields add turning, overlap and operating time. An ecological feature therefore carries two apparent costs — the ground it occupies, and the machine inefficiency it creates around itself.

Three treadmills, one system

In one line each:

• Technology says: produce more.
• Land debt says: earn more per acre.
• Machinery says: farm more acres.

But the treadmills do not run side by side. They drive each other:

1. the technology treadmill forces adoption of larger, costlier equipment — which starts the machinery treadmill;
2. machinery needs acres to spread its cost, and those acres must be bought or rented at prices the land treadmill has already inflated;
3. every gain in yield or efficiency raises the expected return per acre, which the land treadmill capitalizes into land values and rents;
4. higher land costs make maximum yield per acre compulsory — which returns the farmer to the technology treadmill.

Each treadmill’s escape route is another’s entrance. And every loop runs through the same point: the non-crop acre. With each turn, the private opportunity cost of a wetland, hedgerow or woodlot rises, while its carbon opportunity cost — the one Grunwald counts — appears on no farm ledger. That is the two-ledger gap this series keeps returning to.

The fuel tank

The result is exactly what land-use data show. Across countries, rising yields were generally not accompanied by declining cropland [5]; in integrated global markets, productivity gains can even increase cropland use and emissions [6]. Grunwald concedes the point through Searchinger’s “wicked problem”: farmers who grow more per acre “can usually make more money by clearing more acres” [1]. He treats it as the exception. Agricultural economists have long identified this dynamic as a system rule. The same technology package can raise yield per acre and raise the acreage an individual farm wants to operate.

Now add ethanol.

Ethanol turns all three treadmills at once

Ethanol feeds every threadmill loop from above. Each loop, turning on its own logic, ends in the same place: the non-crop acre becomes more expensive to keep.

Ethanol appears to solve a genuine problem. Technology keeps raising productive capacity; commodity markets cannot absorb the output at remunerative prices; land and machinery leave farmers with fixed costs; and governments want to support farm income without telling anyone how much to produce. Putting corn into cars looks elegant. Fuel markets can soak up volumes that food markets never could.

But ethanol does not stop the treadmills. It hypercharges them:

more ethanol demand → tighter commodity balances → higher crop prices → more output → higher expected returns to cropland → higher land values and rents → larger mortgages and machinery investment → more pressure for cash production → more acreage.

The land effect is measurable. The U.S. Renewable Fuel Standard raised corn prices by 30% and expanded U.S. cropland by 2.1 million hectares [7]; grassland and wetland margins across the Western Corn Belt went under the plough [8]. Ontario’s land-cover records show the same direction at smaller scale, as the next essay documents. The first causal estimates of the land-value effect put the post-2005 increase at about US$1,147 per acre in counties best suited to corn, and up to 44% in ethanol-producing states [9].

Then the new production erodes the very scarcity that propped up prices, and the political pressure to manufacture demand returns — higher blends, low-carbon fuel credits, aviation fuel, new biomass markets.

Ethanol appears to solve Cochrane’s problem by absorbing surplus. But it converts much of that relief into higher asset values and another round of investment. It does not get farmers off the treadmill. It finances another turn.

Why “higher yield = spared land” misses the economic reality

Grunwald’s arithmetic holds: more food per hectare means fewer hectares required. The treadmills explain why that is not the same as fewer hectares actually farmed. Technology rewards scale, land debt makes every acre earn, machinery rewards more acres and simpler fields — and ethanol turns all three at once. Sparing land requires institutions and markets that actually retire land when supply saturates markets. The market system, left to itself, paradoxically incentivizes exactly the opposite. This is the context for ethanol policies.

That is also why the NFU’s moratorium on further ethanol expansion in Ontario matters beyond its climate accounting. The moratorium stops a policy that deliberately accelerates all three treadmills. Yet the moratorium is only a first step. The market system needs to change such that individual and societal rationality are aligned, not pitted against each other.

Next in the series: “The Acre That Produces No Corn Is Not Empty” — why stopping expansion is necessary, and not sufficient.

References

1. Grunwald, M. (2025). We Are Eating the Earth: The Race to Fix Our Food System and Save Our Climate. Simon & Schuster.
2. Cochrane, W. W. (1958). Farm Prices: Myth and Reality. University of Minnesota Press.
3. Levins, R. A. & Cochrane, W. W. (1996). The Treadmill Revisited. Land Economics 72(4): 550–553.
4. Brandes, E. et al. (2016). Subfield profitability analysis reveals an economic case for cropland diversification. Environmental Research Letters 11: 014009.
5. Rudel, T. K. et al. (2009). Agricultural intensification and changes in cultivated areas, 1970–2005. PNAS 106(49): 20675–20680.
6. Hertel, T. W., Ramankutty, N. & Baldos, U. L. C. (2014). Global market integration increases likelihood that a future African Green Revolution could increase crop land use and CO2 emissions. PNAS 111(38): 13799–13804.
7. Lark, T. J. et al. (2022). Environmental outcomes of the US Renewable Fuel Standard. PNAS 119(9): e2101084119.
8. Wright, C. K. & Wimberly, M. C. (2013). Recent land use change in the Western Corn Belt threatens grasslands and wetlands. PNAS 110(10): 4134–4139.
9. Le, H. & Galvez-Soriano, O. (2026). First causal estimates of the ethanol boom on farmland values. Applied Economic Perspectives and Policy.
10. Langemeier, M. (2025). Crop Machinery Investment. farmdoc daily, University of Illinois.