Agent-based models have facilitated greater understanding of flood insurance futures, and will continue to advance this field as modeling technology develops further. As the pressures of climate-change increase and global populations grow, the insurance industry will be required to adapt to a less predictable operating environment. Complicating the future of flood insurance is the role flood insurance plays within a state, as well as how insurers impact the interests of other stakeholders, such as mortgage providers, property developers, and householders. As such, flood insurance is inextricably linked with the politics, economy, and social welfare of a state, and can be considered as part of a complex system of changing environments and diverse stakeholders. Agent-based models are capable of modeling complex systems, and, as such, have utility for flood insurance systems. These models can be considered as a platform in which the actions of autonomous agents, both individuals and collectives, are simulated. Cellular automata are the lowest level of an agent-based model and are discrete and abstract computational systems. These automata, which operate within a local and/or universal environment, can be programmed with characteristics of stakeholders and can act independently or interact collectively. Due to this, agent-based models can capture the complexities of a multi-stakeholder environment displaying diversity of behavior and, concurrently, can cater for the changing flood environment. Agent-based models of flood insurance futures have primarily been developed for predictive purposes, such as understanding the impact of introductions of policy instruments. However, the ways in which these situations have been approached by researchers have varied; some have focused on recreating consumer behavior and psychology, while others have sought to recreate agent interactions within a flood environment. The opportunities for agent-based models are likely to become more pronounced as online data becomes more readily available and artificial intelligence technology supports model development.
The rapid increase in losses from flooding underlines the importance of risk reduction efforts to prevent or at least mitigate the damaging impacts that floods can bring to communities, businesses, and countries. This article provides an overview of how the science of disaster risk management has improved understanding of pre-event risk reduction [or disaster risk reduction (DRR)]. Implementation, however, is still lagging, particularly when compared to expenditure for recovery and repair after a flood event. In response, flood insurance is increasingly being suggested as a potential lever for risk reduction, despite concerns about moral hazard. The article considers the literature that has emerged on this topic and discusses if the conceptual efforts of linking flood insurance and risk reduction have led to practical action. Overall, there is limited evidence of flood insurance effectively promoting risk reduction. To the extent there is, it suggests that more complex behavioral aspects are also at play. Further evidence is required to support this potential role, particularly around data and risk assessment, and the viability of different risk reduction measures.