After a real crisis or quality failure, a leader pulls the decisions at fault back to the center in the name of stability, and the authority tends to stay there once the crisis passes.
Borderline: whether this move is justified depends on the situation, and only the person in it can judge.
An example
Two listeria recalls in one year cost a frozen foods maker its largest retail account for a quarter, and both traced back to plants that had set their own environmental testing schedules. Within three weeks the new chief quality officer proposed moving release decisions for every finished lot from the six plant managers to her central team. She brought the recall timelines to the executive committee, showed that the two plants with the lightest testing had produced both positive lots, and asked for authority over testing frequency and lot release only, with a review in eighteen months against a published defect target. The plant managers objected that central release would slow shipments, and she agreed to a four-hour release standard. The committee approved the change in one meeting.
Why it works
A crisis gives centralization the vocabulary of quality and stability, which makes it look better and protects the person proposing it from the charge of grabbing power. The autonomous arrangement looks worse, because its strengths, such as local speed and ownership, now look like the cause of the failure. Defending autonomy becomes costly, because any later incident at an autonomous unit will be blamed on whoever argued for it. Real turmoil is required, along with few strong defenders of the decentralized model, pressure to decide fast, and a proposer with standing to speak for quality.
The thinking behind it
Daniel Kahneman Daniel Kahneman's Judgment under Uncertainty: Heuristics and Biases (1974), written with Amos Tversky, describes availability: people judge how likely something is by how easily examples come to mind. Two recent recalls make another failure vivid, so a proposal to centralize the decisions behind them looks prudent. It explains why the move works while the incident is fresh, and why it should be confined to what the incident implicated.
William Samuelson William Samuelson and Richard Zeckhauser's Status Quo Bias in Decision Making (1988) found that people choose an option disproportionately often when it is framed as the existing state; a later replication supported three of the four scenarios it retested. Once central release is in place it becomes the status quo, and returning authority to the plants must overcome that bias. A review date and a published target written in from the start offset it.
Barry M. Staw Barry Staw's Knee-Deep in the Big Muddy: A Study of Escalating Commitment to a Chosen Course of Action (1976) found that people personally responsible for a losing course committed the most additional resources to it. A quality officer who centralized release decisions is badly placed to judge, at the review, whether the arrangement should continue. That is why the review should rest on a published target and include people who did not make the original call.
When it fits
After a real incident that traces to decentralized decisions, when the business cannot afford a repeat and you have the standing to speak for quality. It fits best when you can name the specific decisions at fault. It fails if the crisis looks manufactured or overstated, or if strong defenders of local autonomy can show the incident had another cause.
How to use it well
Tie the case to evidence. Show how the incident traces to the decisions you want to move, and leave out the ones that had nothing to do with it.
- Ask for the narrowest authority that fixes the problem.
- Put a review date and a published target in the proposal, so the units know what would return authority to them.
- Meet the units' strongest objection with a concrete service commitment, such as a turnaround standard.
- Move while the incident is fresh, since crisis decisions are made quickly and the room for deliberation narrows.
Honor the review date when it comes. Your credibility in the next crisis depends on whether you kept the terms of this one.
What it costs and where it backfires
Centralized authority is hard to give back. New policies, headcount, and reporting lines settle in, the burden of proof flips, and units keep their titles while losing their decisions. You inherit every failure in the area you took over. Staying clean means a real crisis and a request no larger than it requires. Inflating or staging a crisis, blaming autonomous teams for failures they did not cause, or using the moment to collect unrelated decisions turns a quality case into a power grab, and the pattern shows when every crisis ends with more authority at the center.
When it is used on you
Ask for a sunset date and the criteria for returning authority at the moment decision rights are pulled from your unit, usually within weeks of an incident. Check whether the authority taken matches what the incident implicated. Track how often turmoil has produced permanent shifts of authority, and name the pattern in public if it keeps repeating.