The Treadmill Nobody Chose

A follow-up to Monocultures Are Dumb by Design

Let me be direct about where I stand.

I work in technology. I have spent over two decades building open-source infrastructure, mesh networks, and digital tools designed to put capability into communities’ hands rather than extract it from them. I believe in technology that serves people. I have seen it do exactly that.

And in the contest between family farmers and the Monsanto playbook, I have no difficulty choosing a side.

Growing food is one of the most skilled, meaningful, and grounded things a human being can do. It requires reading a specific piece of land across seasons and years, building knowledge from observation and experience that no platform subscription can fully encode. This is artisanal work in the most literal sense: the product cannot be separated from the person, the place, and the accumulated practice that produced it.

Technology genuinely helps with this. Acoustic monitoring, satellite sensing, in-situ soil and water sensors: used well, these tools give farmers richer information about their specific place, supporting better decisions rather than bypassing them. I work with some of these tools and believe in what they can do for building more resilient, diverse, and sustainable food systems.

The Monsanto playbook is something else entirely. It is not about giving farmers better information. It is about replacing farmer judgment with patented inputs and proprietary platforms, building dependency rather than capability, and extracting rent at every step. The previous piece laid out where that architecture ends. The question for this one is why the system makes it so hard to choose otherwise.

That is not a future I am interested in celebrating, defending, or being neutral about.

The “farmer’s choice” argument that came up in response to my last post deserves a direct answer, not because the people raising it are wrong to ask, but because the answer exposes exactly how the system is designed to make the alternative feel impossible. The constraint isn’t in the seed bag. It’s in the debt ledger.

“Wheatbelt, Western Australia” (CC BY-SA 2.0) by European Space Agency. Copernicus Sentinel-2 composite: three near-infrared passes mapped to RGB, showing crop growth stages across the WA Wheatbelt — the same agricultural region discussed in this article.

The Dumbing Down Is Already Well Advanced

Before the economics, it is worth understanding what the system has already cost. Because the “farmer’s choice” argument sounds very different once you account for what has already been lost.

This is not a warning about where the system is heading. It is a description of where it has already arrived.

The Food and Agriculture Organization estimates that around 75 percent of plant genetic diversity disappeared from agriculture during the 20th century, as thousands of locally adapted traditional varieties were replaced by a small number of genetically uniform, high-yielding commercial cultivars. Today, just 12 plant species and five animal species provide approximately 75 percent of the world’s food. Three crops, rice, maize, and wheat, account for nearly half of all human calories.

To put that in scale: humans have cultivated between 6,000 and 7,000 plant species for food across our history. Fewer than 200 now make any significant contribution to global food production.

The greatest loss is not species extinction. It is within-species diversity: the thousands of local rice varieties adapted to specific soils, water regimes, and pest pressures, gone, replaced by a handful of commercial cultivars that yield well under ideal conditions and collapse when conditions shift. The same story played out in wheat, maize, potato, and most other staples.

The pattern is documented: single-variety systems fail catastrophically when conditions shift, whether the Irish Lumper in 1845 or the Cavendish banana today. As the previous piece covers in detail, the industry response to each failure is to sell farmers a new resistant variety rather than question the architecture that produced the vulnerability. The disease is the business model.

The foundation for modern diversity collapse was laid by the Green Revolution of the 1960s and 1970s, which replaced thousands of traditional, locally adapted varieties with a small number of high-yielding commercial cultivars across Asia, Latin America, and Africa. The productivity gains were real and significant. So was the diversity cost. What the Monsanto playbook added was the business model: the patent, the lock-in, the subscription, the data rent. It didn’t start the narrowing. It monetised it.

This is not an accident. It is the logical output of a system that rewards uniformity, scale, and control. Every variety that disappears from cultivation is a variety that can no longer compete with the subsidised, patented, agronomist-backed commercial alternative. Every piece of local knowledge lost with it cannot be recovered from a database. It lived in the hands of the people who grew it, in the specific conditions of a specific place, over generations. When those people exit, it goes with them.

The Mechanism Has a Name

Agricultural economist Willard Cochrane described it in 1958. He called it the agricultural treadmill, and sixty-seven years later it remains the most accurate description of how farming economics work. His formulation was blunt: the average farmer, he wrote, was on a treadmill. Running faster did not increase returns. It just made the treadmill turn faster.

Early adopters of new technology, a new seed variety, a new chemical, a new machine, gain a temporary cost advantage. They produce more, or for less, and pocket the margin. As adoption spreads, increased supply pushes commodity prices down to the new cost of production. The early advantage disappears. Laggards who couldn’t afford the technology are now uncompetitive at the new price level. They exit. Survivors consolidate their land. And the cycle begins again with the next technology.

The consistent winner in this dynamic is not the farmer. It is the technology vendor.

This is not a conspiracy. It is a structural feature of commodity markets where the key technologies are developed, owned, and priced by someone else. Every wave of “productivity-enhancing” AgTech sold to farmers on the promise of higher margins reliably delivers the opposite: temporarily higher margins for early adopters, then lower commodity prices for everyone, then consolidation, then debt for those who couldn’t keep up.

The Monsanto playbook didn’t break this cycle. It accelerated it, while extracting rent at every step.

Squeezed from Both Ends

The farmer operating inside the industrial input system faces two markets in which they have zero pricing power.

On the input side: four companies, Bayer (which absorbed Monsanto), Corteva, Syngenta (now ChemChina), and BASF, collectively control more than half of global commercial seed sales and a similar share of agrochemicals (ETC Group, 2019; ETC Group, 2025). This is not a competitive market. It is an oligopoly that sets prices.

On the output side: in Australia, Coles and Woolworths alone account for around 67 percent of grocery retail volume, with Woolworths holding 38 percent and Coles 29 percent (ACCC Supermarkets Inquiry, 2024-25). Add Aldi and Metcash and four buyers account for nearly all mass retail food. The farmer, again, is a price taker.

The farmer is not negotiating with the market. They are operating between two sets of price makers extracting value in opposite directions. The input oligopoly sets a floor on costs. The retail oligopoly sets a ceiling on prices. The margin in between is what the farmer lives on, and it has been shrinking for decades while farm debt has grown. Australia’s total farm debt reached $120.5 billion in 2022-23, up from $109.9 billion the previous year, and industry data suggests it has continued climbing since (ABARES, farm debt tracking). That is not a sign of investment confidence. It is a sign of structural squeeze.

This is not a market arrangement in any meaningful sense. It is a risk transfer. The input oligopoly profits whether or not the harvest succeeds. The retail chains negotiate price regardless of what the season cost. The banks collect regardless of what commodity prices did. When the system fails, as it periodically does through drought, disease, or price collapse, the concentrated downside lands on the one party with no pricing power on either side of the ledger. The farmer is the shock absorber for an extractive architecture they did not design, did not ask for, and cannot exit without losing everything they have already invested. In any other financial context, we would call this arrangement by its proper name.

Debt Is Not a Choice

This is where the “farmer’s choice” argument collapses.

When a farmer takes on debt to finance a planting season, they are not in a position to experiment with systems the bank doesn’t recognise. The loan structure assumes a conventional crop package: certified seed, registered chemical programme, proven yield history. Deviation from that package puts the loan at risk.

The bank talks to the agronomist. In many cases that agronomist is aligned with the input industry: employed by it, or commission-paid by it. The input supplier sets the package. The farmer signs.

Nobody is forcing anyone to use the products. That is true in the same way nobody forced anyone to sign a subprime mortgage in 2006. The contract was signed freely. The system that made the alternative impossible was built deliberately.

Farm financial stress in Australia is structural, not exceptional. A farmer inside a debt spiral is not choosing inputs freely. They are choosing within a system designed to make leaving cost more than staying.

The Hotel California logic applies exactly here: you can check out any time you like. Just refinance your debt, survive a three-to-seven year transition period on reduced yields while your soil biology recovers, renegotiate your supermarket contracts for varieties cold-chain retail doesn’t want, and retrain yourself after decades of relying on company agronomists.

Technically free. In reality a trap.

What Genuine Choice Actually Requires

Real market choice requires three conditions: accurate information, viable alternatives, and reversible decisions.

On information: agricultural extension services in Australia have been progressively defunded or co-sponsored by the input industry. The institutions whose job it is to give independent advice are funded by the companies selling the inputs. This is not a neutral information environment.

On alternatives: they exist and they work. Ian and Dianne Haggerty’s 26,000-hectare regenerative operation in Western Australia. Dr Terry McCosker’s 8,000 trained farmers across four decades. The Open Food Network connecting producers directly to communities. But these receive a fraction of the institutional support, research funding, and bank finance that conventional high-input systems access. The playing field is not level. It is structurally tilted toward the model that generates the most input sales.

On reversibility: degraded soil biology takes years to rebuild. Knowledge outsourced to corporate agronomists takes years to reconstruct. Markets for diverse, direct-sale produce take years to develop. These are decade-long commitments made under conditions of radical uncertainty. Framing this as a simple consumer preference, like switching phone plans, doesn’t survive contact with a farm balance sheet.

The Question Worth Asking

The question the “farmer’s choice” framing refuses to ask is: who designed this architecture, who benefits from it, and who is funding the next generation of tools to deepen it?

The Tenacious Ventures analysis that prompted my original post is explicit: the Monsanto playbook is a model worth replicating. More control points. More lock-in. More dependency. Sold as innovation.

The next generation adds a third squeeze. The precision ag platform aggregates the farmer’s own field data, builds proprietary models from it, and retains ownership of the intelligence the farmer generated. Exit the platform and you leave without your own data history. This is not a side effect of the architecture. It is the point of it.

The alternative already exists, and it works. In rural northern Italy, the wine in a trattoria came from a farm you can see from the window. The cheese came from a specific valley. The bread came from someone’s cousin’s mill. The prosciutto has a protected designation that encodes the exact geography, breed, and curing method that makes it what it is. This is not nostalgia. It is a functioning food economy, where the connection between land and table has market value and cultural meaning, and where the farmer’s knowledge about a specific place has not been subscribed away.

Compare that to some Australian regional towns surrounded by grape-growing land where the bottle shop stocks imported European wines and nothing from the surrounding region. Or to Woolworths supermarkets in fruit-growing districts selling produce shipped through a distribution warehouse in Geelong, or sourced interstate entirely, rather than from the farms that literally surround the building. This is not a market failure. It is the logical outcome of a system that replaced local food economies with uniform supply chains, stripped place-based value out of pricing, and handed the judgment of the farmer who knows their land to a logistics algorithm that doesn’t care where it came from.

My father read the land. He paid attention to a specific place over decades, and that attention became knowledge that no input schedule could replicate. The system being celebrated in AgTech investment circles is designed to make exactly that kind of knowledge unnecessary: to replace it with a subscription, to turn the farmer into an operator and the farm into a supply node.

That is not progress. That is a trap.

Farmers on the treadmill don’t need more of the same architecture with a software interface bolted on. They need tools that build capability rather than rent it back to them. Data infrastructure they own rather than subscribe to. Investment in transition pathways rather than faster treadmills.

Sources

Food diversity loss

• FAO (2010). The Second Report on the State of the World’s Plant Genetic Resources for Food and Agriculture. Food and Agriculture Organization of the United Nations, Rome. Source for the 75% plant genetic diversity loss figure and the 12 plant / 5 animal species statistic.
• FAOSTAT, Food Balances: underlying data confirming rice, maize, and wheat as sources of approximately half of global caloric supply.

The agricultural treadmill

• Cochrane, W.W. (1958). Farm Prices: Myth and Reality. University of Minnesota Press. The foundational text on the treadmill mechanism: early adopters gain a temporary advantage that diffuses into lower commodity prices as adoption spreads, benefiting only the technology vendor. Cochrane described the average farmer as being on a treadmill where running faster did not increase returns but simply made the treadmill turn faster.

Input and retail market concentration

• ETC Group (2019). Blocking the Chain: Industrial Food Chain Concentration, Big Data Platforms and Food Sovereignty: post-merger consolidation data for global seed and agrochemical markets.
• ETC Group (2025). Top 10 Agribusiness Giants: Corporate Concentration in Food and Farming: updated market share figures showing Bayer, Corteva, Syngenta, BASF controlling more than half of global commercial seed sales.
• ACCC (2024-25). Supermarkets Inquiry: Woolworths 38%, Coles 29%; retail market concentration and pricing power in Australian grocery.

Australian farm debt and financial stress

• ABARES. Farm sector debt reaches $109.9 billion: ABARES tracking of Australian farm sector debt; $109.9 billion at 30 June 2022, rising to $120.5 billion in 2022-23.
• National Centre for Farmer Health, farmerhealth.org.au: Australian data on rural mental health, debt stress, and suicide rates in farming communities.

Regenerative alternatives

• Ian and Dianne Haggerty, Prospect Pastoral Company: 26,000 ha of zero-input commercial production in the Eastern Wheatbelt, WA.
• Resource Consulting Services: Dr Terry McCosker OAM’s GrazingforProfit™ school; 8,000+ graduates across nearly four decades.
• Open Food Network Australia: open-source platform connecting farmers directly with communities.

The Monsanto playbook

• Tenacious Ventures / Shane Thomas, Business Model Breakdowns: The Monsanto Playbook: the analysis that prompted the original post and this follow-up.

• Regenerative Agriculture
• Agtech
• Monocultures
• Food Sovereignty
• Corporate Power
• Rural Australia
• Food Systems
• Farm Debt
• Commodity Farming