Advances in the sharing economy – such as transportation network companies (e.g., Lyft, Uber) and home sharing (e.g., Airbnb) – have coincided with the increasing need for evacuation resources. While peer-to-peer sharing under normal circumstances often suffers from trust barriers, disaster literature indicates that trust and compassion often increase following disasters, improving recovery efforts. We hypothesize that trust and compassion could trigger willingness to share transportation and sheltering resources during an evacuation.

To test this hypothesis, we distributed a survey to individuals impacted by the 2017 Southern California Wildfires (n=226) and the 2018 Carr Wildfire (n=284). We estimate binary logit choice models, finding that high trust in neighbors and strangers and high compassion levels significantly increase willingness to share across four sharing scenarios. Assuming a high trust/compassion population versus a low trust/compassion population results in a change of likelihood to share between 30% to 55%, depending on scenario. Variables related to departure timing and routing – which capture evacuation urgency – increase transportation sharing willingness. Volunteers in past disasters and members of community organizations are usually more likely to share, while families and previous evacuees are typically less likely. Significance of other demographic variables is highly dependent on the scenario. Spare seatbelts and bed capacity, while increasing willingness, were largely insignificant. These results suggest that future sharing economy strategies should cultivate trust and compassion before disasters via preparedness within neighborhoods, community-based organizations, and volunteer networks, during disasters through communication from officials, and after disasters using resilience-oriented and community-building information campaigns.

共享经济的进步（例如交通网络公司（例如Lyft，Uber）和家庭共享（例如Airbnb））与对疏散资源的需求不断增长相吻合。尽管在正常情况下对等共享通常会遇到信任障碍，但灾难文献表明，灾难后信任和同情心通常会增加，从而改善了恢复工作。我们假设信任和同情心可能会导致疏散期间共享交通和庇护资源的意愿。

为了检验该假设，我们对受2017年南加州野火（n = 226）和2018年卡尔·野火（n = 284）影响的个人进行了调查。我们估计了二进制logit选择模型，发现对邻居和陌生人的高度信任以及同情心的程度极大地提高了在四个共享场景中共享的意愿。假设高信任度/同情心人口与低信任度/同情心人口导致共享可能性的变化在30％到55％之间，具体取决于场景。与发车时间和路线相关的变量（反映了撤离的紧迫性）提高了运输共享的意愿。过去灾难中的志愿者和社区组织的成员通常更有可能分享，而家庭和以前的被疏散者则很少。其他人口统计学变量的重要性高度依赖于该场景。备用的安全带和床的容量虽然增加了意愿，但在很大程度上没有意义。这些结果表明，未来的共享经济战略应在灾前通过邻里，社区组织和志愿者网络内部的准备来培养信任和同情心，在灾害期间通过官员沟通与交流，以及在灾后使用面向抵御能力和社区建设的信息运动。