An ultimate mortality table is a life‑table that shows, for each age, the proportion (or number) of a studied population expected to survive to that age. It is built from life‑insurance policyholder data and deliberately excludes early years of policy experience to remove “selection” bias from recently underwritten policies. Ultimate tables are a core actuarial tool used to price life insurance, set reserves, run longevity and mortality stress tests, and help individuals and advisers estimate life expectancy for financial planning.
Key takeaways
– An ultimate mortality table gives survival/death rates by attained age after removing early-policy (select) effects.
– It typically covers ages 0–120 and is based on insured lives rather than the whole population.
– Insurers use ultimate tables to price products, calculate reserves, and manage risk; advisers and individuals use them for retirement planning and longevity estimates.
– Accuracy depends on data breadth, data quality, and how tables are constructed and updated; major producers (e.g., the Society of Actuaries) publish widely used tables.
Sources used: Investopedia (Mira Norian), Society of Actuaries (SOA), David H. LaFever (Encyclopedia of Ecology).
1. Understanding ultimate mortality tables
– Survivorship and death rates: Tables report age‑by‑age metrics such as the number surviving to an age (l_x), probability of dying during an age (q_x), or central death rates (m_x).
– Insured population focus: Data typically come from policyholders—often healthier and subject to underwriting—so results differ from population life tables.
– Exclusion of early experience: The “ultimate” adjective means the table represents mortality after selection effects have disappeared. Newly issued policies (first few policy years) are excluded because applicants tend to be temporarily healthier after medical underwriting.
– Stratification: Modern tables often break out by sex, smoker status, underwriting class, and sometimes by weight, region, or other factors.
2. Short history and context
– Mortality tables have long roots in demography and actuarial science. Raymond Pearl and contemporaries applied life‑table concepts in ecological and biological studies in the early 20th century; actuaries adapted and refined these methods for insurance and public health (see LaFever).
– Professional bodies (e.g., the Society of Actuaries) compile large pooled data and publish standard ultimate tables used across the industry.
3. Difference between select, ultimate, and aggregate tables
– Select tables: include the immediate post‑issue years and capture underwriting selection (lower short‑term mortality).
– Ultimate tables: exclude early years and show “longer‑term” mortality once selection fades.
– Aggregate tables: combine many experience tables into a single table (by pooling or blending) to reflect an entire studied population, sometimes including both select and ultimate experience in blended form.
4. How insurers and financial firms use ultimate mortality tables
– Product pricing: Estimate expected death benefits and set premiums so the present value of premiums covers expected claims, expenses, and profit.
– Reserves and capital: Calculate best‑estimate liabilities and regulatory reserves (including stochastic/longevity risk tests).
– Underwriting policy: Inform underwriting classes and decisions (e.g., smoker vs non‑smoker rate differentials).
– Reinsurance and risk transfer: Determine stop‑loss levels, aggregate exposures, and pricing for mortality reinsurance.
– Investment & retirement planning: Asset managers and advisers may consult tables to estimate client life expectancy and plan withdrawal rates.
5. Practical steps — building or updating an ultimate mortality table (for a carrier or researcher)
1) Collect and prepare data:
• Compile individual policy records: date of birth, sex, underwriting class, smoker status, policy issue date, policy duration, death events, and exposure (time at risk).
• Include a sufficient number of years and broad coverage to reduce sampling variability.
2) Remove selection (create ultimate experience):
• Exclude the first N years of each policy’s experience (N depends on product and observed selection duration—commonly 1–5 years) or use statistical select models to separate select vs ultimate risk.
3) Calculate raw age‑specific rates:
• For each attained age, compute deaths D_x and exposure E_x, then central death rate m_x = D_x / E_x (or q_x = D_x / number exposed at start of year).
4) Smooth and graduate:
• Apply smoothing/graduation methods (moving averages, spline fits, parametric models) to reduce random year‑to‑year noise while preserving trend.
5) Credibility weighting and pooling:
• If data at some ages are sparse, blend with external tables (industry tables such as SOA) using credibility theory.
6) Test and validate:
• Compare to previous tables and industry benchmarks, run back‑testing, test for anomalies (e.g., pandemic years), and check for cohort effects.
7) Include improvement factors:
• Consider mortality improvement scales (future decline in death rates) and scenarios for sensitivity testing.
8) Document assumptions and governance:
• Keep clear documentation: data sources, exclusions, smoothing techniques, credibility weights, and governance approvals.
6. Practical steps — using ultimate mortality tables (for insurers, actuaries, and financial planners)
For life insurers / actuaries:
1) Select appropriate table by risk class (sex, smoker status, underwriting class).
2) Apply appropriate mortality improvement assumptions for pricing and reserving.
3) Use scenario and stress testing (e.g., faster improvement, pandemic shock) to assess capital adequacy.
4) Calibrate to internal experience periodically and adjust pricing or underwriting if credible deviations persist.
For financial advisers and individuals:
1) Choose an appropriate source (SOA tables, national population life tables, or an insurer’s published table) and the correct cohort (sex, smoker status).
2) Adjust for personal health: treat table life expectancy as a baseline and adjust for chronic conditions, family history, and lifestyle.
3) Use life‑expectancy as an input to retirement spending plans (determine safe withdrawal horizon, annuity decisions).
4) Revisit estimates periodically and after major health events.
7. Special considerations and limitations
– Data quality and representativeness: Insurer‑specific tables may not generalize outside that carrier’s book; large pooled tables (e.g., SOA) are more robust.
– Selection and underwriting effects: Differences in underwriting standards across time and companies change the applicability of older tables.
– Temporal change and improvement: Mortality changes with medical advances, public health trends, and events such as pandemics; tables must be updated and incorporate improvement assumptions.
– Sparse data at extreme ages: Death counts at very high ages are small—statistical techniques or external benchmarks are often used for ages >100.
– Socioeconomic and demographic differences: Insured lives are often wealthier or healthier than the general population—use caution when applying insured‑life tables to broader populations.
8. Best practices
– Use pooled, peer‑reviewed tables (e.g., SOA published tables) as benchmarks and blend with internal experience where credible.
– Update tables and improvement assumptions regularly (annual or multi‑year review).
– Document and validate: retain versioning, tests, and governance approvals for any table adopted for pricing or reserving.
– Stress testing: model alternative mortality trajectories (faster improvements, plateauing, shocks) to quantify capital and pricing sensitivity.
– Transparency: for consumer decision‑making, explain assumptions and provide ranges (median and confidence intervals) rather than single deterministic numbers.
9. Practical example (conceptual)
– Suppose a carrier observes 50 deaths and 10,000 policy‑years of exposure at attained age 70 (after removing select years). The central death rate for age 70 is m_70 = 50/10,000 = 0.005. After smoothing and converting to a one‑year death probability, and blending with industry data where necessary, this figure becomes the ultimate mortality rate used in pricing and reserving for age 70.
10. Where to find published ultimate mortality tables and resources
– Society of Actuaries — Mortality and Other Rate Tables: /
– Investopedia explanation (overview):
– LaFever, D.H., Encyclopedia of Ecology, Vol. 3 — on historical context and age‑class models (see bibliographic reference for background).
References
– Investopedia. “What Is an Ultimate Mortality Table?” (Mira Norian).
– Society of Actuaries. Mortality and other rate tables and experience studies. /
– LaFever, David H. Encyclopedia of Ecology, Volume 3: Age‑Class Models. Academic Press, 2019.
Editor’s note: The following topics are reserved for upcoming updates and will be expanded with detailed examples and datasets.