This paper introduces a method for the evaluation of the occurrence of tipping points based on the combination of probability intervals from experts judgments elicited in face-to-face interviews. The computation of such conditional probabilities is based on the aggregation of imprecise probability judgments through the Steiner point. The probability of a tipping point can be updated by the standard Bayes rule to generate tipping point scenarios. Our results suggest that tipping events may happen with relatively large probabilities, in contrast with the view that tipping points are low-probability-high-impact events. This suggests that mitigation and containment policies cannot be further postponed.
JEL Classification: Q54; D81; C10