Public institutions depend on specialised knowledge held by individuals. This is neither unusual nor inherently problematic: expertise, professional judgement and differentiated responsibilities are essential to complex institutional work. A continuity risk emerges, however, when a consequential institutional function depends so heavily on knowledge held by one person, or by a very small number of people, that the institution cannot reproduce the required capability once those individuals become unavailable. Retirement, resignation, reassignment, rotation, restructuring or unexpected absence can then produce what appears to be sudden knowledge loss even though the underlying vulnerability may have existed for years.
Personnel transition often exposes institutional dependency rather than creating it. While an experienced employee remains available, the organisation can compensate for weaknesses in its own knowledge architecture simply by consulting that person. Formal procedures may appear sufficient because the employee supplies missing context; difficult cases remain manageable because accumulated judgement is still accessible; and relationships continue to function because the individual through whom they were built remains present. The institution may therefore appear capable while part of that capability remains dependent on continued access to particular people.
The evidence reviewed for this Policy Paper shows that turnover-related knowledge loss is a genuine but heterogeneous organisational phenomenon. Not every departure produces meaningful loss, and not every form of lost knowledge has significant institutional consequences. Whether a transition weakens capability depends on the importance of the affected function, the concentration and substitutability of relevant knowledge, the availability of alternative pathways, and the difficulty of reproducing or reacquiring what the institution still requires (Galan, 2023a). The appropriate unit of analysis is therefore not the departing employee alone, but the relationship between the person, the knowledge and the institutional function that depends upon it.
Different dependencies also create different reproduction burdens. Some knowledge can be represented effectively through records and documentation, while other forms depend more heavily on context, experience, practice or relationships. Continuity may therefore require representation, structured social transfer, practice-based reproduction, distributed capability or, where proportionate, a credible route to reacquisition. The objective is not universal duplication of expertise, but sufficient institutional reproducibility.
PP-001 consequently recommends that public institutions govern material person-dependent knowledge as a continuity risk. This principle translates into four connected areas of action. Institutions should see the risk before transition by identifying consequential functions that depend excessively on particular knowledge holders and prioritising those dependencies according to criticality and vulnerability. They should design the right continuity response by matching mechanisms to the reproduction burden and creating sufficient alternative pathways. They should make continuity actionable by acting early enough for the selected pathway to work and assigning clear institutional ownership. Finally, they should test whether capability survived rather than treating completion of transfer activity as evidence of success.
Implementation can follow a common logic while remaining proportionate to context: own the risk, identify the dependency, decide whether it matters enough to act, determine what must survive, build the lightest credible alternative pathway, activate it while the necessary conditions still exist, and test the result after transition. Some dependencies will justify little more than accessible representation and confirmation that another route exists. Others may require structured transfer, practice or additional holders. A smaller number of high-consequence dependencies may justify multiple pathways or deliberate redundancy. Sudden departures require a corresponding shift from preventive reproduction toward recovery through records, colleagues, external expertise, temporary workarounds or redesign.
Completion of transfer activity is not evidence that continuity has been achieved. The relevant test is whether an alternative institutional pathway actually functions after transition and whether the resulting continuity remains proportionate to its organisational cost and effects. Capability gain should therefore be distinguished from implementation activity and from the mere existence of a planned mechanism.
This proportionality is essential because the objective of knowledge continuity is not institutional stasis. Mobility can introduce new expertise, inherited practices may become obsolete, and some knowledge can be reconstructed more efficiently than it can be preserved. Public institutions therefore need minimum sufficient continuity: preserving or reproducing enough capability-relevant knowledge to keep consequential functions viable without attempting to retain everything the institution currently knows.
The central test is straightforward. When a consequential knowledge holder becomes unavailable, can the institution still do what it materially needs to do with sufficient understanding, reliability and independence? If it can, and the resources required to create that continuity were proportionate to the institutional risk, capability has improved. If transfer activities have been completed but the function remains dependent on privileged access to the former holder, the underlying vulnerability remains.
