Dependency Isn’t a Delegation Failure
Dependency Isn’t a Delegation Failure
Why your team won’t take ownership, and why teaching you to let go won’t fix it
There’s a complaint you hear in almost every executive conversation, usually with some exasperation attached. They won’t take ownership. Everything escalates. Decisions the team understands far better than the executive does arrive on the executive’s desk anyway, wrapped in a request for a steer. The calendar fills with judgements that shouldn’t be theirs to make, and the more of them they make, the more arrive.
The explanation offered is nearly always the same, and it’s aimed at the executive. You haven’t delegated properly. You’ve made yourself indispensable. You’re the bottleneck. You need to let go, develop your people, stop being the answer to every question. There is a large industry built on this diagnosis, and it is well-meaning, and it fails with impressive consistency.
It fails because dependency is not a delegation failure. It’s a rational response to a structure, and it will survive every attempt to coach it away.
What the team is actually reading
Start with the people who won’t take ownership, and assume – as you should always assume – that they are behaving sensibly.
Taking ownership of a decision means carrying its consequence. In a structure where decision rights are thin but accountability is generous, that is an unattractive proposition: you own the outcome, but you don’t hold the authority that would let you produce a good one. You can be overruled at any point, your judgement can be reversed by someone with more standing, and the resources you’d need may or may not be released. You can rely on the fact that if it goes wrong, it will be traceable to you.
Escalating, in that structure, is not laziness. It is the correct move. It converts a decision you’d own into a decision someone else owns, at the cost of some delay and a little visible passivity, both of which are far cheaper than exposure. The team isn’t failing to step up. They’ve read what stepping up costs and priced it accurately.
Which is why the standard remedies bounce off. Delegate more, and you’ve handed people the consequence without the authority, which is the very thing they were avoiding. Tell them you trust them, and they’ll weigh your reassurance against the structure and believe the structure. Send them on a course about ownership and accountability, and they’ll return to a system where ownership is still expensive, and behave exactly as before within a fortnight.
The Ostrich Farm
There’s a version of this from the other side, and it’s the more revealing one.
A manager described her team as an ostrich farm. Everyone looked to her to absorb the uncertainty surrounding the business: the shocks from outside, the turbulence from within. So she did. She worked harder and harder, shielding them from it all. And the more she absorbed, the more they expected to be shielded, and the less ownership they took.
The obvious reading is that she created their passivity. She over-protected; they became dependent; the fix is for her to stop.
That reading is wrong, and the diagnosis matters, because the remedy follows from it. Her exhaustion and their passivity are not cause and effect. They are two outputs of the same structural fault: an architecture that made her answerable for outcomes without distributing the authority to act on them. She wasn’t generating dependency by shielding too much. She was absorbing pressure the structure had routed to her; she stood at the point where the load concentrated, and there was nowhere for it to go but through her. Meanwhile, her team was declining exposure to the same structure, which made it irrational to accept. One fault, two symptoms, appearing at different ends of the same reporting line.
The test that settles it
There is a way to check which reading is right, and that’s why this distinction isn’t academic.
If dependency is a delegation failure – if her shielding causes their passivity – then removing the shielding should produce ownership. Stop absorbing. Let the shocks reach them. Watch people rise to it.
That is not what happens. What happens is paralysis, or blame, or both. The team is now exposed to shocks they still have no authority to act on, which is worse than before, rather than better, and the organisation concludes that they were indeed not ready for ownership, which confirms the original misdiagnosis and produces another round of development programs.
The shielding, it turns out, was load-bearing. She was compensating for a structural gap, and removing the compensation reveals the gap rather than closing it. This is the signature of a symptom being mistaken for a cause: take away the symptom and the underlying fault simply presents somewhere else, usually worse.
People take ownership when they hold enough authority to make ownership survivable. Not when protection is withdrawn.
Diffusion and concentration are the same fault
There’s a second version of this, apparently the opposite, and seeing that they’re the same thing makes the whole pattern legible.
The CEO of Deutsche Bahn recently named her organisation’s central problem as organisierte Verantwortungslosigkeit – organised irresponsibility. Responsibility had been pushed back and forth between the centre and the operating level for years, until nobody in particular had to answer for anything. Where no one must stand behind a result, the persuasive explanation becomes the actual achievement, and the result itself becomes optional.
Set that beside the ostrich farm, and they look like opposites. In one, responsibility is diffused until nobody carries it. In the other, it’s concentrated until only one person does. But they’re the same fault seen from either end: authority and consequence have come uncoupled, and an organisation resolves that in one of two ways. Either it spreads responsibility so thin that nobody can be pinned to it, or it piles it onto whoever is least able to refuse.
Both produce the same executive complaint – nobody takes ownership – and both are immune to the same remedy. You cannot coach an organisation out of diffusion, and you cannot delegate your way out of concentration, because in both cases, what’s missing is the coupling, not the willingness.
Why the misdiagnosis persists
It’s worth asking why the delegation explanation is so durable, given how reliably it fails.
Partly because it’s visible. You can see an executive making too many decisions; you cannot see the decision rights that were never distributed. The behaviour is on the surface, and the architecture isn’t, so the explanation attaches to what’s observable.
But mostly because of who it implicates, and how gently. “You need to delegate better” is a personal development point, addressed to one individual, solvable with a coach and some deliberate practice. “Your organisation has never coupled authority to accountability” implicates the design, which means the executive team that sets the decision rights, the governance that reviews them, and the incentives that punish being wrong. One of those is a conversation you can have on a Tuesday afternoon. The other requires someone senior to concede that the structure they built is producing the behaviour they’re complaining about.
The comfortable explanation wins, as it usually does, and the dependency continues.
What actually moves it
If the fault is uncoupled authority and consequence, then the remedy is to re-couple them. Which is a design task, not a leadership-style adjustment.
It means asking, for each decision currently arriving on the wrong desk: who has the information to make this well, and do they hold the authority to make it stick? If the answer to the second half is no, that’s your escalation explained, not a motivational problem but an authority gap, and one that no amount of encouragement will close.
It means making sure that when authority is handed down, it isn’t quietly retrieved. Delegated authority that gets reversed the moment someone senior disagrees isn’t authority; it’s a test people learn not to take. The team is watching what happened to the last person who decided something without checking, and reading it correctly.
And it means examining what it actually costs to be wrong. Ownership is only rational where the downside of a reasonable decision that doesn’t work out is survivable. Where a single bad call is career-defining, escalation isn’t timidity; it’s the only sensible strategy available, and the people using it are the ones paying closest attention.
Change those, and something happens that no development program has ever reliably produced: the escalations thin out on their own, not because people were told to take more ownership, but because ownership stopped being the expensive option. Same people, different architecture.
The complaint, restated
So when the complaint arrives – they won’t take ownership – it’s worth hearing it as a piece of information about the structure rather than about the people.
Nobody is failing to step up. They have read what stepping up costs here, and they’ve priced it correctly. The exhausted executive at the other end of the same fault has read their position correctly too: the load has been routed to them, and there is nowhere for it to go.
The question isn’t how to make people take more responsibility. It’s what’s in the architecture that makes responsibility so expensive to hold.