Deciding under ambiguity: what quantified risk does not capture
Choosing a technology partner or a new market often means comparing options supported by unequal amounts of information. The managerial question is specific: are we rejecting an option because of its possible consequences, or because we know less about their probabilities?
What the literature establishes
Daniel Ellsberg (1961, The Quarterly Journal of Economics) distinguishes situations where probabilities are known from those where they are ambiguous. Using choice problems involving urns, he shows how a preference for known probabilities can conflict with the axioms of subjective expected utility. His analysis does not imply that every poorly documented option is advantageous: it identifies a difficulty in how uncertainty is represented.
The role of comparison
Craig R. Fox and Amos Tversky (1995, The Quarterly Journal of Economics) demonstrate experimentally that ambiguity aversion depends on the comparison context. It is particularly pronounced when people compare an ambiguous situation with a better-known one, or their own knowledge with that of better-informed people. Reluctance therefore reflects more than the option’s characteristics: it may also express a relative sense of incompetence.
The premature conclusion
One might conclude that managers should overcome this reluctance and choose the unfamiliar more often. The cited research does not justify that recommendation: insufficient information can legitimately influence a choice, particularly when safety or regulatory obligations are involved. The useful distinction is between identified adverse consequences and uncertainty about their probability, without turning that uncertainty into an artificially precise estimate. (our executive and employee training programmes)
What the evidence does not allow us to transfer
Choices between gambles reproduce neither the responsibilities of a medtech company nor the constraints of an international sports organisation. These articles provide no universal threshold for acceptable ambiguity and do not demonstrate that lower aversion improves organisational performance. A risk matrix therefore becomes misleading if it gives a documented probability the same status as a mere judgement.
A practical check in Lausanne
In Lausanne, SHR’s “Decision-making” programme can draw on an exercise applicable to medtech companies, higher education institutions, international sport and scale-ups across the Lake Geneva region: for four weeks, document the probabilities invoked in partnership decisions. For each probability, record its source or explicitly mark it as unknown, then note the reason for selecting the chosen option. Next, measure the proportion of decisions in which every probability invoked has either an accessible source or an explicit “unknown” label, using all decisions reviewed as the denominator. This check verifies the traceability of the reasoning, not the superiority of the choices made. To go further: explore the Decision-making training in Lausanne, or browse our executive and employee training programmes in Switzerland.
In pictures: Decision-making in Lausanne
- decision-making
- Lausanne
- research
- training lausanne