Disaster Risk Insurance: Insights and Recommendations

Project Overview

Using the Kaggle/Our World in Data natural-disaster dataset, I quantified the global protection gap (economic losses − insured losses) and compared payout ratios across regions and peril types. Asia shows frequent, high-loss events (especially earthquakes and floods) but persistently low insurance coverage; North America/Oceania are more mature, yet uninsured losses are still rising with event frequency. Recommendation: prioritize innovative/parametric products and public-private partnerships in Asia and other vulnerable regions to narrow the protection gap and strengthen resilience. This direction aligns with the World Bank Climate Finance Mobilization goals.

What I Did

• Built an end-to-end R/tidyverse pipeline: cleaning, encoding payout ratios and protection-gap indicators; publication-ready charts.
• Constructed regional time series of uninsured losses; contrasted payout ratios with disaster frequency and peril mix.
• Peril-focused slices (earthquake/flood) to map loss drivers to product design opportunities (incl. parametrics/PPPs).
• Statistical inference: Welch one-sided t-test (Asia vs North America coverage), p = 0.04638 < 0.05 → Asia significantly lower.
• Synthesized external evidence (Swiss Re sigma) to validate trends and support an action-oriented recommendation.
• Packaged findings for a policy audience (World Bank): where to intervene, which peril to target, and why parametrics help.

Reflection

Working with noisy, partly missing self-reported/loss data forced disciplined claims: I documented assumptions, kept estimates conservative, and triangulated with Swiss Re sigma. The biggest learning was turning analytics into a concrete product roadmap: for Asian markets, simulate parametric triggers (e.g., quake intensity, flood gauge thresholds) and test affordability/payout adequacy under PPP structures. If iterating, I would integrate exposure layers and higher-resolution peril data, then run sensitivity analyses on trigger design to balance basis risk, speed of payout, and fiscal impact. Keeping the message policy-relevant—where to act, what to build, and how to measure impact—was key, and it’s also why this project earned recognition.