From opinion to action

Abstract

Despite efforts to mitigate climate change, adaptation becomes critical. Among climate-induced hazards, flooding is the most costly and widespread, calling for adaptation across scales: from government-led to household-led adaptation. Private adaptation measures, if taken, reduce damages and speed up recovery. Empirical evidence suggests that among socio-behavioral factors, social norms and peer influence are crucial for households’ decisions to adapt. Yet, the role social networks play in household-level adaptation has not been studied systematically, even less so the interplay between private household adaptation and public information policies in the presence of social influences, the exchange of opinions and information within a social network that impacts individuals’ decisions. To improve the understanding of the impact of social influence on private household adaptation uptake, we build an empirically informed agent-based model of household behavior. We leverage household survey data collected in Harris County (Texas, USA) together with information on flood hazard scenarios to study the impact of private adaptation diffusion on flood damages under different social influence scenarios. Furthermore, we use the model to test how different information policy design choices—such as targeting specific households or all, and communicating different aspects of flood adaptation (e.g., flood damage, costs, or effectiveness of measures) --influence private adaptation diffusion and impact regional residual flood damage. We find that regardless of the structure of the social network, social influence triggers higher adaptation uptake by households, resulting in a 5–10% extra reduction in regional residual flood damage. Notably, the effect of social norms depends on the type of information exchanged within networks, where the opinion exchange on the effectiveness of measures and potential damages results in more private adaptation compared to the discussion on perceived costs of adaptation measures or the worry about flooding. Moreover, while information campaigns influence individual perceptions that facilitate household-led adaptation, the provision of information solely on expected damages is not as effective in steering public opinion and, thereby household adaptation, as its combination with information on coping measures that enable action. Our results demonstrate that social influence and information policy can shape the success of private action and highlight the importance of understanding the interaction among scales in climate change adaptation.