Population projections are performed regularly by national statistics institutes. In France, the most recent projections were produced by Insee in 2016 using a deterministic approach based on 27 different scenarios. In this article, we propose a new approach which combines probabilistic population projections and a greater use of the Bayesian paradigm in order to quantify the uncertainty of future population levels without resorting to scenarios. Using the components method, the mortality rate, fertility rate and net migration are projected independently by sex and age. These three components are modelled, taking account of registry data (number of births and deaths) and net migration data series. The results reveal that the population of metropolitan France will continue to grow, reaching a level of between 66.1 million and 77.2 million inhabitants in 2070, with a probability of 95%.

Citation: Costemalle, V. (2020). Bayesian Probabilistic Population Projections for France. Economie et Statistique / Economics and Statistics, 520‑521, 29–47.
https://doi.org/10.24187/ecostat.2020.520d.2031