Created a new approach and methodology to explain adverse mortality experience.
Case Studies / Insurance / Understand and Manage Mortality Risk
A global life insurance carrier was looking to better understand adverse mortality experience that could not be explained by traditional Actuals to Expected (A/E) studies.
Used Data Science techniques to identify pockets of consistent deviation from expected outcomes. We considered multiple variables, such as underwriting time, the agent involved with the policy, etc.
We were able to invalidate existing hypothesis about the effect of environmental factors like weather, economy, and health facilities in explaining mortality experience. Instead we included additional external and internal data to expand list of factors under consideration and derive consistent, learnable patterns of deviations that did a much better job of explaining adverse mortality experience.
With the information we provided, our client was able to broadly redesign several components of their process. We created a closer alignment between expectation setting and pricing, and improved reserving and distribution. We created a repeatable process to identify deviations between actual and expected mortality.