The Determinants of Life Expectancy: An Analysis of the OECD Health Data

This study considers an aggregate life expectancy production function for a sample of developed countries. We find that pharmaceutical consumption has a positive effect on life expectancy at middle and advanced ages but is sensitive to the age distribution of a given country. Thus, ignoring age distribution in a regression of life expectancy on pharmaceutical consumption creates an omitted-variable bias in the pharmaceutical coefficient. We find that doubling annual pharmaceutical expenditures adds about one year of life expectancy for males at age 40 and slightly less than a year of life expectancy for females at age 65. We also present results for lifestyle inputs into the production of life expectancy. For example, decreasing tobacco consumption by about two cigarettes per day or increasing fruit and vegetable consumption by 30% (one-third pound per day) increases life expectancy approximately one year for 40-year-old females.