Forecasting Expected Annual Loss (EAL) for Citizens Insurance in Florida
Haoxiang Liu
Co-Presenters: Individual Presentation
College: College of Business and Public Management
Major: Finance
Faculty Research Mentor: Huaibing Yu
Abstract:
In 2017, Hurricane Irma wrought unprecedented damage across Florida, resulting in over $50 billion in insured losses and underscoring the critical need for robust risk quantification frameworks in the insurance industry. Grounded in theoretical principles of catastrophe modeling and risk management, this study is motivated by the imperative to enhance underwriting strategies through precise loss estimation. We develop a comprehensive model to analyze the Expected Annual Loss (EAL) by integrating key parameters—annual policy count, average coverage per policy, hurricane frequency, and the historical loss ratio (HLR). Using data primarily sourced from Citizens Property Insurance Corporation, our model serves as a validation case to assess its predictive capability.The approach comprises three components: (1) Historical Trend Analysis, which reveals a U-shaped trajectory in policy counts and Total Insured Value (TIV) from 2011 to 2024, marked by a downturn until 2017 followed by a substantial recovery; (2) Scenario-Based Forecasting, projecting EAL for 2025–2034 under Low (conservative HLR), Normal (historical mean HLR), and High (extreme HLR) scenarios; and (3) Risk Sensitivity Assessment, which examines the impact of variations in key risk parameters. Under the Normal scenario, annual EAL is forecasted to rise from approximately $528.57 billion in 2025 to $667.42 billion in 2034, yielding a cumulative loss of about $5.25 trillion over the decade. The Low scenario projects a cumulative loss of around $3.35 trillion, while the High scenario suggests an extreme cumulative loss nearing $24.29 trillion.These findings highlight the model's sensitivity to fluctuations in loss ratios and hurricane frequency, reinforcing the importance of integrating theoretical insights with empirical data to inform proactive risk management and underwriting practices tailored to the unique risk profile of insurers like Citizens.