Climate change and natural disasters pose global collective action problems: while mitigation requires coordinated effort at a global scale, the impacts of climate-related disasters are unevenly distributed across local communities. This asymmetry raises a fundamental question for collective action under risk: when global cooperation is uncertain, will individuals prefer locally coordinated solutions that offer partial protection at a cost?

We investigate this question using a nested collective-risk dilemma that introduces a local level of cooperation into a high-risk collective action setting. In our experiment, participants face a high probability of collective loss unless a global contribution threshold is reached. Unlike standard collective-risk dilemmas, we allow individuals to contribute either to a global public good or to a local one. Reaching the global threshold fully eliminates the disaster risk for all participants. Reaching a local threshold, instead, provides local protection but does not fully prevent the disaster: if the disaster occurs, locally protected participants incur a cost reflecting the impact of the event.

We implement four conditions. The baseline is a standard collective-risk dilemma without a nested structure. In the remaining treatments, we introduce a local level and vary the cost of disaster impact for locally protected individuals: (i) a low local cost, designed to encourage local cooperation; (ii) a high local cost, expected to shift incentives toward global cooperation; and (iii) an asymmetric condition in which one local group faces double the impact cost of the other. This asymmetric treatment is expected to generate the weakest cooperative outcomes, potentially undermining both global coordination and local cooperation.

Our central hypothesis is that introducing a local level weakens global cooperation by increasing strategic substitution, while simultaneously strengthening local cooperation by reducing social uncertainty and free-riding – except under severe cost asymmetry. In addition, we elicit participants’ beliefs to examine the emergence of social norms at different levels. We ask whether global and local cooperation are governed by the same normative expectations with varying strength, or by distinct norms altogether. By isolating the interaction between risk, social uncertainty, and nested cooperation, our study sheds light on why local solutions often prevail in global climate challenges, even when they are collectively suboptimal. We will collect data in the coming months in order to obtain preliminary pilot results to be presented at the conference.