Global professional services company Towers Watson announced the release of The Older Age Mortality Study 3 (TOAMS 3), the company’s third study of U.S. life insurance industry mortality. TOAMS is the most comprehensive, longest-running study of older age mortality experience in the market. The original TOAMS, released in 2005, covered mortality experience for insureds attained ages of 50 and higher during 2000, 2001 and 2002. The subsequent TOAMS 2 report, issued in 2008, covered older age mortality during 2003, 2004 and 2005.TOAMS 3 spans the five-year experience period from 2006 to 2010 for insureds attained ages of 50 and above. The 21 participating companies provided $8.2 trillion of face amount for 48 million policy years. Over 700,000 deaths in the study represent $40 billion of death claims. The study was performed on an issue age and duration basis, using a 25-year select period, consistent with the Society of Actuaries 2001 and 2008 Valuation Basic Table (VBT).
“This study is very timely for life insurers,” said Elinor Friedman, director, Life practice, Towers Watson. “It’s no secret that the low interest rate environment has placed downward pressure on life insurers’ profitability. Insurers have had to reprice many products with profit streams vulnerable to low interest rates, heightening the need for greater pricing accuracy. Our study gives insurers the tools to address trends in life expectancy and fine-tune their pricing assumptions.”
One new feature and important enhancement to TOAMS 3 is the inclusion of predictive modeling analysis of the participant-submitted data using Towers Watson’s proprietary generalized linear modeling (GLM) tool, Emblem™. Emblem uses advanced modeling procedures to determine what function set of independent risk factors best predicts a chosen target variable — in this case, the expected mortality rate. TOAMS 3 is the first industry-wide mortality study to combine traditional experience analysis with GLM.
“TOAMS 3 was ideal for predictive modeling, since many correlated risk factors influence mortality rate, and multivariate analysis determines the underlying effect of these correlated risk factors. The potential impact of predictive modeling for participants in the industry is very significant, as it helps guide a more precise view of how relationships among underlying factors can affect mortality results,” said Karen King, senior consultant, Life practice, Towers Watson.
Key findings of the study include:
- Overall, mortality is 62% of the 2001 VBT by face amount and 75% of the 2001 VBT by policy count. Experience as a percentage of the 2008 VBT showed higher percentages (83% by face amount, 90% by policy count).
- Male nonsmokers have considerably more exposure and a higher average face amount than the other gender/smoking status combinations.
- 76% of the exposure by face amount was in the two highest bands ($250,000 and higher), while 81% of the exposure by policy count was in the four lowest bands (up to $250,000).
- The predictive model analysis shows that expected mortality levels by plan are not as large as the traditional analysis would suggest, indicating that other correlated variables explain the difference.
About the Study
TOAMS 3 provides detailed results focusing on gender, smoking status, level of underwriting, risk class, type of reinsurance and cause of death as well as distribution channel, issue state and refined plan types. All mortality results are expressed as actual-to-expected ratios, with expected equal to the appropriate version of 2001 VBT or 2008 VBT.
TOAMS 3 is now available for purchase by nonparticipants. The TOAMS 3 package includes the detailed report that shows results from both the traditional experience analysis and the predictive modeling analysis, and also includes the TOAMS 3 Query Tool. With the tool, you can use specific criteria to create customized reports from the TOAMS 3 database.
To purchase the complete TOAMS 3 report with predictive modeling analysis, e-mail mortality.study@towerswatson.com.