Reiko Hayashi, National Institute of Population and Social Security Research, Japan
The vital statistics of Japan started in 1899 and continued up to present but the data of 1944 and 1945 lacks consistency. The numbers of total deaths published in the reports are 1,290,513 in 1944 and 2,115,162 in 1945, which are smaller compared to the war victim estimates, and the data is not sex and age disaggregated. There is a need to fill the gap. By using the census of 1940 and 1955 as starting and ending points of population, the number of deaths from 1941 to 1943 and 1946 to 1955 were subtracted and added by sex and birth cohort. The calculation was done for Japanese nationals living within mainland Japan or overseas. The data on late-registration and nationality change was also taken into account. The results show that the total number of deaths was the double of that published in the vital statistics report. The 1944-1945 deaths structure by sex and birth cohort exhibits a high concentration of young (20s) men. The huge death toll towards the end of WWII affected unevenly by age and sex. The estimation could be further refined with additional data and modelling, but the overall structure of death would remain the same.
Keywords: Historical Demography, Mortality and Longevity, Population projections, forecasts, and estimations, Population, Shocks and Pandemics