How long will men in France who are now 75 live? Or women? Or men or women who are now 60 or 40? Or men and women in Denmark, Germany, Italy, Japan or the United States? What will the lifespans be of newborns in these and other countries with long life expectancies?
The answers to such questions require forecasts for cohorts, i.e., groups of people the same age. Demographers more often make forecasts for periods - e.g., what will life expectancy be 50 years from now? Period calculations pertain to synthetic cohorts - imaginary generations - that live their entire lives at the death rates that prevail in a particular year. Both cohort and period forecasts of longevity are of interest to insurance - and reinsurance - companies, to governments and businesses that have pension obligations, to investors worrying about the solvency of such governments and businesses, and to individuals pondering how much to consume, how much to save and how to plan the time of their lives. For many purposes, cohort forecasts are more useful - but usually period forecasts are easier to make.
- James W. Vaupel
SCOR Paper #20
This document is available in format Acrobat. Learn more :