Those applicants you declined? Their mortality matters more than you think

The rise of automated underwriting (AUW) presents challenges to carriers that previously relied on medical exams or fluids tests to identify high-risk conditions. Most assume that a small percentage of applicants who would have been declined under previous fully underwritten (FUW) guidelines will now be issued policies.

To price this risk, actuaries must estimate the percentage of declinable risks being issued and the average mortality of those applicants who would have been declined. The industry’s rule of thumb has been 500% of standard, but verifying that number has presented an obvious challenge: The industry has rarely tracked the actual mortality of the “decline” population.

In his white paper, Decline mortality: Shape and severity of mortality for declinable life insurance risks, Partner Re’s Ehren Nagel took a big step towards solving this problem by using a large, proprietary dataset and Irix Risk Score (Rx, Dx) to create a realistic, declinable-risk proxy population and model it for the purposes of ascertaining its mortality.

The result is one of the most creative and interesting white papers we’ve read in a while. Among the observations:

• The conventional-wisdom estimate of 500% of standard or better mortality for declinable risks may be in line with the results for longer-duration products, but mortality varies significantly by product duration, with very steep A/E curves in early durations.
• Aggregate A/E is much higher for females than males.
• In breaking down decline mortality by duration, age, and sex, significant cohorts often showed A/E of over 900%.

Nagel notes that in addition to tobacco use—which was an obvious source of concern as carriers adopted fluidless programs—other factors including build, substance use, diabetes, cholesterol, and high-blood pressure have also driven significant misclassification.

This research supports our position that carriers should use Medical Data in all AUW cases to identify low-frequency/high-severity conditions that may add very high mortality risk to AUW pools.