AI Can’t Replace Free Markets
Algorithms process data from the past while economic decisions are dynamic and forward-looking.
By Marian L. Tupy and Peter Boettke
Illustration: David GothardImagine artificial intelligence controlling the economy.
That’s the future envisioned in three recent manifestos.
Law professor Ted Parson introduces “Max,” an AI that overlays markets with Pigouvian price tweaks—taxes here, subsidies there—until every externality is neutralized.
Computer scientist Spyridon Samothrakis proposes a mesh of data hubs and reinforcement-learning schedulers to guide economic coordination, resource allocation, and production.
And economist Leo Schlichter argues that an AI system could reduce output, respect ecological limits, and meet human needs through participatory dashboards and feedback loops.
Their pitch is straightforward: AI can help to augment or even replace the function of prices and the free market that generates them.
Not so fast.
Economic coordination isn’t a problem to be solved by computing an optimal answer.
It emerges from the decentralized decisions and adjustments made by billions of economic actors—each with their own plans, preferences, and knowledge—in an ongoing, evolutionary process.
Certain rules and institutions are essential for transforming decentralized decision-making into orderly and socially beneficial outcomes.
The three Ps—property rights, prices, and profit and loss—provide the three Is—information, incentives and innovation.
Prices enable people to engage in economic calculation, which forms the basis for the rational allocation of scarce resources among alternative ends.
Prices also function as decentralized feedback loops.
“A price is a signal wrapped up in an incentive,” note Tyler Cowen and Alex Tabarrok.
This dual signal communicates information about relative scarcities and simultaneously encourages economic actors to adjust their plans accordingly.
When lithium prices rise, producers and consumers conserve, recycle, innovate, and explore alternatives.
The belief that AI can achieve comparable results to free markets, let alone surpass them, reflects a misplaced confidence in computation and a misunderstanding of the price system.
The problem for the would-be AI planners is that prices don’t exist like facts about the physical world for a computer to collect and process.
They arise from competitive bidding over scarce resources and are inseparable from real market exchanges.
Moreover, prices aren’t fixed inputs to be assumed in advance.
They are continually being discovered and formed by entrepreneurs testing ideas about future consumer wants and resource constraints.
Economic models that treat prices as given overlook the entrepreneurial actions that create them in the first place.
Ludwig von Mises made this point in 1920: Without real market exchange, central planners lack meaningful prices for capital goods.
Consequently, they can’t calculate whether directing steel to railways rather than hospitals adds or destroys value.
AI can process vast amounts of data—but always from the past.
Economic action, by contrast, is forward-looking.
An algorithm may extrapolate trends, but it can’t anticipate innovation and changing tastes.
It can’t discover what hasn’t been imagined.
Free markets, by contrast, continuously produce real and reliable price information.
That happens through the interplay of the three Ps.
These institutions force participants to put skin in the game—bearing real costs for mistakes and earning profits for insight.
Simulated markets can’t replicate this feedback.
Without consequences, algorithmic outputs fail to elicit true valuations or meaningful behavioral adjustments.
AI’s economic champions confuse data processing with discovery and overlook how incentives shape the data AI receives.
If political actors influence prices, then the input into algorithms is already distorted.
“Garbage in, garbage out” still applies—only now the garbage is processed faster and packaged in technical jargon.
AI may appear precise, but it has the same blind spots that doomed prior central planning efforts.
Centralizing decisions also distorts behavior.
Entrepreneurs anticipating expropriation or opaque regulations may withdraw, reduce investment, or exit entirely.
Consumers may hoard or barter.
The very data planners rely on become unreliable as people adapt their behavior to avoid being captured by the system.
Our research on post-socialist transitions shows that meaningful price signals only re-emerged after private exchange and budget discipline were restored.
Computational power didn’t restore order—institutional reform did.
Crucially, markets coordinate existing knowledge and generate new information.
The price system reveals hidden scarcities and helps discover untapped opportunities.
That discovery process is the engine of growth.
Central planning by bureaucrats or algorithms can’t substitute for it.
As Friedrich Hayek observed, “the value of freedom rests on the opportunities it provides for unforeseen and unpredictable actions.”
Economics and engineering don’t substitute.
If allocation becomes a technical problem and AI the solution, society may shift talent from exploration to optimization.
But prosperity depends on experimentation, not blueprint execution.
Economists should embrace what Hayek called catallaxy—order born from exchange among strangers, each pursuing new ends with evolving means.
Centralized intelligence freezes that process, replacing dynamic evolution with rigidity.
AI is a powerful tool for recognizing patterns and improving processes.
But it can’t replace free markets.
It can’t generate genuine prices, account for opportunity costs, or bear entrepreneurial risk.
Economic vitality still depends on free exchange, not on optimization routines run in sterile data centers.
Rather than resurrect central planning with AI, policymakers should focus on strengthening the institutional foundations that make real market coordination possible.
Mr. Tupy is a senior fellow at the Cato Institute. Mr. Boettke is a professor of economics at George Mason University. Kyle O’Donnell contributed to this article.
0 comments:
Publicar un comentario