Rare risks: what experience alone cannot reveal
A team may have worked without incidents for a long time and still be poorly informed about a rare risk. The management question is precise: how much weight should that experience carry when deciding whether to maintain, strengthen or relax a safeguard?
What the literature establishes
Ralph Hertwig, Greg Barron, Elke U. Weber and Ido Erev (2004, Psychological Science) compared risky choices based on described probabilities with choices based on personal sampling of outcomes. In their experiments, rare events generally carried less weight in decisions based on experience than in those based on a description. Observing outcomes first-hand therefore does not guarantee that infrequent events will receive greater consideration.
What lived experience leaves out
Ralph Hertwig and Ido Erev (2009, Trends in Cognitive Sciences) examine the mechanisms behind this description–experience gap, including limited sampling and the weight of recent observations. A rare event may never appear in the sample consulted; it may also carry little weight when recent outcomes dominate judgement. It is therefore important to distinguish what a team has not encountered from what it has encountered but given little weight.
The hasty conclusion
One might conclude that a written probability should always take precedence over operational experience. These studies do not justify that hierarchy: a description may be incomplete, while experience may reveal local conditions missing from the model. The useful conclusion is narrower: the absence of an event within limited experience is not enough to establish that its probability is negligible. (our executive and employee training programmes)
The limits of the evidence and incident records
The experiments reviewed mainly concern simplified risky choices, rather than regulatory, public health or industrial decisions made in Berne. They provide neither a universal threshold for safeguards nor a direct measure of the effectiveness of an incident register. In practice, an incident-free register remains difficult to interpret unless exposure duration, opportunities for failure and reporting rules are documented.
A practical check in Berne
Within SHR's decision-making programme, an exercise applicable in Berne is to review, over one month, every decision to maintain, strengthen or relax a safeguard: continuity of a federal administration service, a public health alert system, telecom redundancy or a precision-industry inspection. For each decision, record the rare risk concerned, the exposure observed without incident, the source of the risk estimate and the remaining uncertainty. Then measure the proportion of these decisions whose records explicitly address these elements, stating “unknown” where necessary. This indicator checks the traceability of the reasoning, not the accuracy of the probabilities or a demonstrated reduction in incidents. To go further: explore the Decision-making training in Bern, or browse our executive and employee training programmes in Switzerland.
In pictures: Decision-making in Bern
- decision-making
- Bern
- research
- training bern