Common Knowledge

In Hans Christian Andersen's "The Emperor's New Clothes," every citizen privately knows the emperor is naked. The swindlers' con has already failed at the individual level — no one actually believes the fabric is invisible to the unworthy. But no one acts on what they know, because each person lacks a critical piece of information: whether everyone else also knows. The child who shouts "He has nothing on!" doesn't add any new information to the system. Every person already possesses the relevant fact. What the child creates is common knowledge — the state where everyone knows, everyone knows that everyone knows, and everyone knows that everyone knows that everyone knows. That recursive certainty is what turns private belief into coordinated action. Before the child speaks, each citizen is trapped in a private assessment. After the child speaks, the crowd laughs in unison. The information didn't change. The knowledge structure did.

Common knowledge is a game theory concept formalised by Robert Aumann in 1976, building on David Lewis's work in Convention (1969). The formal definition is precise: a fact is common knowledge among a group if everyone knows it, everyone knows that everyone knows it, everyone knows that everyone knows that everyone knows it, and so on to infinity. This is not a philosopher's parlour game. It is the difference between a crowd of individuals who each privately disagree with a regime and a crowd that revolts. The difference between a market where every participant privately believes an asset is overvalued and a market that crashes. The information content is identical in both cases. The knowledge structure — who knows what about what others know — determines whether anyone acts.

The application to financial markets is where the concept does its most visible work. A stock can be overvalued, every sophisticated participant can privately believe it is overvalued, and the bubble continues. Not because participants are irrational. Because each participant, making a rational calculation, knows they can profit by staying in the trade as long as others stay — and they lack common knowledge about when others will exit. The bubble persists not despite everyone knowing it is a bubble but precisely because knowing is not enough. You need to know that others know. And you need to know that others know that you know. Without that recursive certainty, the individually rational move is to ride the bubble and try to exit first.

Ben Hunt's "Epsilon Theory" newsletter has spent a decade applying common knowledge theory to financial markets. Hunt's central thesis: the market is not primarily an information-processing mechanism. It is a common knowledge game. What moves markets is not new information about economic fundamentals — it is new common knowledge about what other market participants believe about economic fundamentals. A Fed announcement rarely contains information the market hasn't already priced. What the announcement creates is common knowledge: every participant now knows that every other participant has received the same signal, interpreted through the same models, at the same time. The announcement is a coordination device. The information is incidental.

The mechanism operates identically outside financial markets. A company where every employee privately knows the product is failing but no one raises the issue in meetings is a common knowledge failure — the information exists everywhere, the coordination exists nowhere. A political movement where millions of citizens share a grievance but each believes they are alone is a common knowledge failure — the shared experience has not been made publicly, recursively known. The Arab Spring was triggered not by new grievances but by public acts of protest that made existing grievances common knowledge, enabling the coordinated action that decades of private dissatisfaction could not produce. In every case, the pattern is identical: the information was already there. What was missing was the knowledge structure.

This explains the otherwise baffling phenomenon of markets moving violently on "no new information." When the Fed holds rates steady — exactly as consensus predicted — and the market moves anyway, commentators scramble for explanations in the statement's language, the dot plot, the press conference tone. The common knowledge framework provides a simpler one: the announcement didn't change what anyone knew about rates. It changed the knowledge structure. Before the announcement, each participant held a private estimate of what other participants expected. After the announcement, everyone knows that everyone knows rates are unchanged, which enables coordinated repositioning that private knowledge alone cannot support. The phase transition from "most people expect X" to "everyone knows that everyone expects X" is the transition from a collection of individual beliefs to a coordination mechanism. Markets don't move on information. They move on common knowledge.