What Do Hierarchies Actually Select For?

Jordan Peterson's claim, made often and confidently, is that hierarchies are hierarchies of competence. People rise because they are good at the thing, and the resulting order roughly tracks ability.

This is, at minimum, a question rather than an answer — and it is a question with an empirical answer that nobody seems interested in establishing.

The Obvious Objection

In some organisations hierarchies do track competence. In many others, the dominant factor is skill at organisational politics: managing upward, claiming credit, avoiding blame, cultivating the right sponsors, and being visible at the right moments.

These are genuine skills. They are simply not the skills the organisation exists to exercise. A hospital that promotes on political skill still promotes able people — able at politics.

I want to be careful not to overstate this, because the overstatement is common and it is lazy. I am not claiming hierarchies never promote competence. They plainly sometimes do, and an organisation that never did would collapse.

The claim is that they promote competence in some cases, political skill in others, a blend in others still, and possibly other things nobody has identified. Which is a different claim from Peterson's, and it is the one the evidence actually supports.

Why the Answer Probably Varies

Here is what I suspect is true, and it would be the interesting result if someone established it.

Different hierarchies select for different things, and the variation is predictable from structure.

Where performance is measurable, attributable, and quickly visible, competence should dominate — because a person who cannot do the thing is exposed rapidly and cheaply. Surgery, competitive sport, sales on commission.

Where performance is diffuse, collective, and slow to reveal itself, other factors should dominate — because there is no signal to promote on, and something must fill the gap. Senior management, policy, most of the civil service, and anything where the results arrive after the promotion decision.

That is a testable prediction rather than a slogan, and it would explain why people's intuitions about this differ so violently: they are generalising from the sector they happen to know.

Why This Matters

The Peterson claim is not doing idle work. It is load-bearing in an argument about the legitimacy of existing distributions.

If hierarchies track competence, then the people at the top are largely there because they should be, and inequality of position is substantially deserved. If they track something else, that inference collapses — and with it a common justification for leaving arrangements as they are.

So the question is worth settling rather than asserting. This should be researched. There is a literature on promotion and performance, but I have not found anyone who has attacked the general question with the structural hypothesis above.

I would want to know: across sectors, what predicts advancement? How does that correlate with measurable output? And does the correlation vary with feedback speed the way I have guessed?

Until someone does that work, "hierarchies are hierarchies of competence" is a hypothesis being deployed as a conclusion — and it is being deployed by people who have done well out of hierarchies, which is worth noticing without being decisive.