There’s a story technologists like to tell about AI agents: software will do most cognitive work humans currently do. Transaction costs, the friction of coordinating economic activity, collapse toward zero. Markets become faster, smarter, more efficient. Everyone benefits.
It’s tidy but conveniently missing about half the economics.
The moment you treat agents as an economic object, the familiar frameworks stop fitting. Agents aren’t workers with lower wages. They aren’t software with better features. The market they’re creating doesn’t behave like any market we’ve built institutions around.
In an endpoint when transaction costs drop so far that the logic of firms and employment starts to dissolve, the “Coasean singularity” named by the MIT and Harvard researcher, the supply and demand in Economics 101 and what people take for granted in labor economics become… strange.
Demand has two regions
Demand for AI agents splits into two structurally different regions:
Substitution demand: agents doing things humans currently do: drafting, summarizing, screening, supporting. This has a natural price ceiling: no buyer pays more for an agent than the human alternative costs. The whole substitution, or a more fashionable word today, displacement market trends toward commodity, and it is deflationary. As agents displace workers, human wages fall, and the ceiling they’re priced against falls with them.
Frontier demand: agents doing things humans couldn’t do at scale regardless of cost. Monitoring every transaction across a financial system in real time. Personalizing every customer interaction for millions of people simultaneously. Operating in hazardous mining environments. There was no human version to substitute. These get priced against the value of the outcome, not the cost of an alternative, hence no ceiling, no floor.
This echoes MIT’s research that identifies the two scenarios for agent deployment: making decisions of similar quality at dramatically lower cost, or making higher-quality decisions than humans. The first maps to Substitution demand. The second is Frontier.
The boundary between the two regions is not fixed. As agents improve, tasks that once required human judgment slide into the substitution category. The frontier keeps moving. The people on the wrong side of that line don’t get much warning.
boundary
*assumes agents keep improving
replaceability →
Supply has a gap where the middle used to be
In a normal labour market, supply is continuous. Buyers choose along the spectrum based on need and budget.
AI agent supply doesn’t work this way.
At one end: the commodity tier. Once an agent is trained, copies deploy at near-zero marginal cost. Supply, at least in terms of quantity, is effectively unlimited.
At the other end: the differentiated tier. Agents with genuine domain depth, trained on proprietary data, fine-tuned on years of professional feedback. These are scarce, not because copying is expensive, but because the inputs that make them valuable are slow to accumulate. You can’t rush the training data and pile garbage data that makes a legal agent reliable in contested litigation.
Between these two tiers, there is a chasm. AI does not optimize for competent generalist. An agent that is almost as good as the best one is nearly worthless if buyers can identify the best one, exactly the same as why Google dominates the search engine market and confirmed by this NBER paper.
middle tier
Zero friction is real.
Agents are, for most practical purposes, location-less. An agent trained in Tashkent deploys in Copenhagen at no incremental cost. For cognitive tasks, there is no Geneva wage premium anymore. There is just a global market.
Labour economics has always depended on friction. Workers can’t instantly move between markets. That friction sustains wage differentials and creates floors. Agents collapse the friction on the cognitive side entirely. Wage arbitrage that took decades in manufacturing, a.k.a. move production to cheaper labour, takes months or even shorter for knowledge work.
One writer put it bluntly: agents create “a new hierarchy enforced by API keys instead of labour contracts”. When the marginal cost of running an agent approaches zero, the scarce resource shifts to where the agent can run and what it can access. The firm that owns the data or controls the model becomes the landlord of the agent economy.
For regions where human labour is cheap, this creates a more complicated picture than a simple “AI won’t displace workers here”. It provides a temporary buffer. It also means those regions accumulate less deployment experience, less training data, less institutional knowledge of how to build valuable agents. According to the latest Anthropic Economic Index research on learning curves, this could mean lagging behind in outcomes from AI use: lower success rate in conversations and more concentrated in simple task types. The global agent market will run at two speeds, and the second speed is building a structural disadvantage.
What an agent market actually looks like.
Three features together: segmented supply with no middle, demand with a moving frontier, and zero friction making every capability immediately global.
In a normal labour market, supply and demand move toward equilibrium: wages rise when demand exceeds supply, workers retrain, the market adjusts. Imperfectly, slowly, but directionally.
Agent markets don’t have these stabilizers. Quality supply can’t expand quickly in response to price signals, as quality supply is constrained by data, ideally, the right kind of data, not willingness to produce. Zero friction means geographic arbitrage doesn’t create equilibrating pressure. Congestion erodes efficiency gains faster than markets can respond.
The standard policy tools designed for labour market correction, including but not limited to reskilling programmes, minimum wages, labour unions and the intentional inflexibility to fire, were built for a different kind of market. Applying labour market frameworks to this reality is approximately as useful as applying 19th-century railroad regulation to a platform economy.
One scope note: this piece starts with knowledge workers because that’s where the argument is most visible. But the majority of the global employment is not knowledge work. The economics described here will arrive in different industries at different times. The scope is a starting point, but will travel through sectors and geographies.
It will be naïve to still question whether the agent market develops. It already is. The question is whether anyone setting the rules understands they’re working with a different animal entirely.