Age preferences for life-saving programs: Using choice modeling to measure the relative values of statistical life

Document Type


Publication Date



This study employed the survey-based choice modeling (CM) approach to determine people's age preferences for life-saving programs in Metro Manila, Philippines. Two social welfare models — social welfare as a function of the number of lives saved in each of the age groups (Model 1) and social welfare as a function of the number of lives saved irrespective of age and total life-years saved (Model 2) — were specified and ran using binary probit regression. Our results from both models indicate a general preference for saving younger lives. Based on Model 1 estimates, saving a child aged below one year was considered to be equivalent to saving four retirees (60 years old and above), and saving a life in the 1–19 years age group was equivalent to saving five retirees. The working age group's (20–59 years old) life was judged equivalent to three lives in the retired age group. Regression results for Model 2 indicated that both number of lives and total number of life-years saved (and hence age group of lives saved) significantly affected the choices of the respondents.