In 1911, the German sociologist Robert Michels formulated a conclusion that he called the iron law of oligarchy: any large organization, regardless of how democratic it was originally conceived, over time inevitably comes to be run by a small group of people.
Michels studied social democratic parties—organizations whose explicitly declared value was the equal participation of all members. And he discovered that it was precisely within them that a stable ruling elite formed, gradually detaching itself from ordinary members.
He explained this mechanism not by malicious intent, but by simple organizational necessity. The larger an organization becomes, the harder it is to make decisions collectively: coordination costs grow, specialization is needed, speed is required. The need to delegate arises—and people emerge who handle coordination professionally. These people quickly accumulate what others lack: access to information, control over communication, negotiation experience, and an understanding of internal procedures. After some time, unseating them becomes virtually impossible—not because they arranged it on purpose, but because the entire organization already depends on their accumulated expertise and connections.
For growing companies, this describes a familiar story. A startup of twenty people, where decisions are made in a group chat and everyone can influence the product, grows to two hundred—and suddenly it turns out that approval layers have appeared, an "inner circle" where decisions are actually made has emerged, and regular employees feel that their voice no longer matters. At the same time, the founders genuinely do not understand what happened: nobody deliberately changed the culture, the values were not rewritten, and everyone is still in favor of openness.
The utility of Michels's law lies not in fatalism, but in realistic expectations. A flat culture is not a natural state of an organization that sustains itself if you just don't forget about it. It is a state that requires a constant expenditure of energy to maintain against the natural drift toward the concentration of power. If nothing is done deliberately, the drift will happen.
What helps:
explicitly designing mechanisms for participation rather than relying on culture—regular formats where decisions are discussed broadly, not just in an inner circle;
rotating people in key coordinating roles so that information and connections do not concentrate in the same hands permanently;
transparency in the decision-making process—at minimum, recording who made a decision and why, so that the "inner circle" does not become invisible;
an honest check once a period: how many people actually influence decisions today compared to last year.