The behavioral economics of extreme event attribution

Can Attribution Science, a method for quantifying - ex post - humanity's contribution to adverse climatic events, induce pro-environmental behavioral change? We conduct a conceptual test of this question by studying, in an online experiment with 3,031 participants, whether backwards-looking attribution affects future decisions, even when seemingly uninformative to a consequentialist decision-maker. By design, adverse events can arise as a result of participants' pursuit of higher payoffs (anthropogenic cause) or as a result of chance (natural cause). Treatments vary whether adverse events are causally attributable and whether attribution can be acquired at cost. We find that ex-post attributability is behaviorally relevant: Attribution to an anthropogenic cause reduces future anthropogenic stress and leads to fewer adverse events compared to no attributability and compared to attribution to a natural cause. Average willingness-to-pay for ex-post attribution is positive. The conjecture that Attribution Science can be behaviorally impactful and socially valuable has empirical merit.