Age-specific proportion of shifted deaths and the tempo effect in period mortality

Christian Wegner-Siegmundt, Wittgenstein Centre (IIASA, VID/ÖAW, WU)

Tempo effects remain a controversial but also very interesting topic in mortality research. Although the existence and the origin of tempo effects are theoretically easy to prove, the methods and their need for tempo adjustment are still highly controversial. In this paper I present a method for deriving the age-specific proportions of shifted deaths. These proportions contain the period flow of deaths which were rescued due to the period mortality conditions. According to the logic of Bongaarts and Feeny, these saved deaths cause a tempo distortion in conventional period life expectancy. By using the age-specific proportions of shifted deaths, it is possible to reformulate the different methods for tempo-adjusted life expectancy without the assumption of a proportional change in period mortality. The empirical application presents that the reformulated tempo-adjusted life expectancy reflects more precisely the changes in period mortality conditions than the conventional life expectancy.