Contents

Markets: Optimization Against Optimizers

A shop lowers a price and attracts customers. Its competitor notices, matches the price, and changes its product bundle. The first shop responds with faster delivery. A supplier raises a fee to capture part of the new margin. Nothing in this sequence stays a private improvement for long. Each move changes the environment in which everyone else’s next move will be judged.

Markets are full of adaptive people and firms optimizing against other optimizers. A better product, forecast, advertisement, or trading strategy doesn’t merely improve one participant’s position. It changes the signals and opportunities facing the rest.

Rows of changing stock-market figures on a financial screen

A field of moving signals. Market feedback reflects the actions of other people who are also learning from it. Photo by Daniel Brzdęk on Unsplash

The target moves

In a solitary engineering problem, a stronger method may keep working. In a market, success attracts imitation and response. A profitable trading pattern can disappear as others discover it. A novel product feature becomes a category expectation. An advertising message loses force when every competitor adopts it. Improvement is strategic because the object being optimized includes other people’s likely reactions.

Friedrich Hayek’s essay on the use of knowledge in society described prices as a way of coordinating dispersed knowledge that no central planner possesses in full. Prices convey compressed information about scarcity and demand. They also provoke adaptation: consumers substitute, producers invest, and entrepreneurs search for alternatives. The signal changes behavior, and changed behavior produces a new signal.

That loop can improve coordination without any market participant understanding the whole. It can also produce an arms race. Faster traders invest in speed because rivals are fast. Platforms subsidize one side of a market to gain scale, prompting competitors to do the same. Sellers learn the ranking system of a marketplace; the marketplace changes the ranking system in response; sellers learn the revision.

In a market, becoming better often means becoming better at anticipating how others will become better.

Discovery and distortion

Competition can reward genuine improvement. A firm that makes batteries cheaper or delivery more reliable creates value that rivals must answer. Failed experiments lose money, while useful discoveries can spread through imitation. The market supplies feedback with unusual force.

But profit is a signal, not a complete moral judgment. A company may externalize costs onto neighbors, exploit an information gap, or make a product more compulsive without making its users better off. The market can reveal what people will pay under existing rules and circumstances; it cannot decide by itself which rules are just or which desires deserve cultivation.

No fixed finish line

Competitive advantage is relational. A ten-percent improvement can be decisive when rivals stand still and irrelevant when they improve by twenty percent.

Recursive capability can deepen both the discovery and the distortion. Firms improve experimentation, personalization, logistics, and pricing. Then they improve how those systems learn from customers. The resulting service may become wonderfully responsive. It may also become very good at finding the point where convenience turns into dependence, or where tailored pricing becomes extraction.

The strongest market institutions don’t eliminate adaptation; they shape its boundaries. Property rights, contracts, liability, competition law, professional norms, and ordinary moral restraint affect which strategies pay. Those rules also provoke adaptation and must be revised with care, because participants will optimize around the revision too.

Markets show recursive improvement without a single improving self. The intelligence is distributed among participants, and the feedback emerges from their exchange. That makes markets powerful discovery systems and unreliable moral authorities. They can tell us that an optimizer has found a better move. They cannot, on their own, tell us whether the game is making better people or a better world.