RAPID: Preferences and Decisions to Evacuate in the Face of Hurricane Harvey

William Marsh Rice University, Houston TX

Investigators

Abstract

This study looks at who decided to leave and who decided to stay in the face of a mandatory evacuation of citizens prior to the landfall of Hurricane Harvey in August 2017. Citizens were given ample warnings that Hurricane Harvey was growing in intensity and was likely to bring record amounts of rainfall. Social scientists often have studied decisions by citizens to evacuate or remain in the face of a natural disaster. We find that when people understand their risk, when they are well informed about the hazards of staying and when they have resources to leave, they will. However, social scientists do not understand how preferences about risk, trust in public officials and cooperation with others inform decisions to leave. Understanding when citizens agree to leave behind their property at the urging of local officials is something that is not well understood. Citizens heeding the recommendations of public officials are a central element of a democratic society. Yet it is well known that a sizeable proportion of the population refuses to go along with mandatory evacuation orders. When the natural disaster is severe, those staying behind place enormous stress on first responders. This research provides insight into what motivates people to leave or stay. This in turn will help public officials in their efforts to convince people to evacuate when necessary. Social scientists have thoroughly explored people's rationalization about whether to evacuate prior to a natural disaster. Much of this research has looked at individuals after an event (a flood, a tornado or a hurricane) and asked them to recall their decisions. These post-event recollections are usually correlated with attitudes that people hold about disasters. Unfortunately, the recollections, the scope of the disaster and post-event attitudes are confounded. The disaster alters attitudes, often in a self-serving manner. People rebuild their own narratives about what drove them to leave or remain. Our research solves this problem by leveraging a sample of individuals living in a hurricane-prone area whose preferences toward risk and loss and attitudes about disasters were measured between 2009 and 2011. Many of these individuals were part of a three-wave panel with repeated measures. We will re-contact these individuals and ask them whether they obeyed a mandatory evacuation prior to the first landfall of Hurricane Harvey. Our measures are not concerned with their reconstructed rationale, but rather what they did. Their behavior will then be linked to the rich set of measures taken well prior to this event. This puts this research in an unusual position to see if fundamental preferences about risk, loss, time discounting, trust and cooperation predict basic evacuation behavior.