No decision is made without institutional memory
Decision quality depends less on the individual deciding than on the memory of the institution deciding. Institutions that lose their memory repeat old mistakes with new justifications.
Political debate gives far more room to individuals than to institutions. Yet what usually determines whether a decision was sound is not the personal quality of whoever made it, but the memory it was made with.
Memory is not the archive
Institutional memory gets confused with the archive. An archive is where past documents are stored. Memory is the accumulated people and procedure that know why a given document matters.
Three signs tell you memory is alive in an institution:
- When a new proposal arrives, someone asks why the last version of it failed.
- The reasoning behind a decision remains accessible after the decider has moved on.
- The dissenting unit's opinion is recorded — it leaves a trace even when the outcome does not change.
Without those three, the institution starts from zero every time.
The invisible cost of rotation
Staff turnover is natural and necessary. But when its pace outruns the learning curve, an institution stays permanently new to its own work. Then:
- Short-term, visible tasks are preferred; long-horizon preparation is deferred.
- The easily defensible decision displaces the risky but correct one.
- External consultancy substitutes for the institution's own knowledge.
Eventually the institution buys back knowledge of its own past from outside.
An institution that loses its memory does not stop making mistakes. It makes the same mistake each time with a fresh justification.
Merit is not a question of morality
Debates about merit are usually squeezed into a moral frame. The issue is first of all technical: where merit works, knowledge accumulates; where it does not, accumulation disperses. The moral dimension matters — but the first casualty of its absence is not a sense of justice, it is foresight.
Procedure that leaves a trace
A functioning decision process contains this minimum:
| Stage | Question | Record |
|---|---|---|
| Preparation | What was tried before? | Prior assessment notes |
| Decision | Which options were dropped, and why? | Reasoned decision text |
| Review | Did the expected outcome occur? | Implementation report |
The table looks unremarkable. Yet behind many contested decisions lies exactly one of those three rows, left blank.
Conclusion
Institutions do not produce good decisions because good people work there. They produce them to the extent that they can accumulate and transmit what those people knew. Institutional memory is not a management luxury; it is the precondition of decision quality.
Related Articles
- Economy
The real problem in fighting inflation is credibility
The fate of a disinflation programme is decided less by the level of interest rates than by how predictable the decisions behind them are. Credibility is monetary policy's cheapest instrument — and, once lost, its most expensive one.
- International Affairs
The regional balance is being rebuilt: three axes, one uncertainty
Regional order is not remade by a single rupture but by the slow displacement of three overlapping axes. That shift will set the foreign policy agenda for the coming decade.
- Analysis
The cost of policy noise
Every unexplained reversal has a price. It is not paid at the moment of the announcement but months later, in the risk premium — and it is paid by everyone.
