Are we under- or overreacting to climate change? Adaptation behavior in ambiguous risk with various levels of risk history

Climate change extreme events have characteristics that predispose people to fall prey to biases and hinder optimal adaptation choices. The paper focuses on biases derived from skewed risk perception that can lead to over-and under-updating probabilities. This in turn causes individual adaptation to deviate from what is anticipated by benchmark models. The paper starts by establishing a theoretical framework to substantiate this phenomenon and account for the ambiguous risk entailed by climate change related extremes. It then experimentally explores the effect that different levels of risk history has on individual adaptation choices ex-post.The risk history is produced in a pure risk phase, and the over-and under updating is tested in a subsequent ambiguous risk phase. We conducted a laboratory experiment with 75 subjects and examined their adaptation decisions in a home flood context. We found that a history of higher risk results in a small but significant decrease in the share of subjects who adapt. This trend is observed across three experiment treatments that differ in risk history, i.e., low, medium and high risk levels. Accordingly, we also find that the share of subjects who do not adapt in the ex-post ambiguity phase is higher for subjects who have experienced two floods than for those who have only experienced one. These findings have important implications for the efficient development of policies that target communities with higher frequency of climate change extremes.