Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Insurance contracts may fail to perform, leading to a default on valid claims. We relax the standard assumption of known probabilities for such defaults by allowing for uncertainty. Within a large behavioral experiment, we show that introducing risk and uncertainty each leads to significant reductions in insurance demand and that the effects are comparable in magnitude (17.1 and 14.5 percentage points). Furthermore, risk- and ambiguity-averse participants are affected most. These findings are in line with models incorporating ambiguity attitudes or, alternatively, pessimistic beliefs. An analysis of the belief and decision dynamics suggests persistent pessimistic priors and disregard of peer experiences, leading to a stable uncertainty effect.