The Production and Use of Population Projections: Assumptions – Uncertainties – Communication – Policy Making

Population projections are among the most controversial population data. The assumptions on fertility, mortality and migration required for population projections and the ambiguity about their future development are a major source of uncertainty. Even relatively small deviations in the assumptions from actual developments can lead to considerable uncertainty margins over medium to long projection periods. Nevertheless, population projections are often much more accurate than other forecasts in the economic and social sciences because part of the future population development is already inherent in the age structure of a population (demographic momentum). Thus, population projections are often interpreted as unavoidable predictions and used as forecasts they are often the basis for far-reaching planning and policy decisions. By making assumptions, they are prone to the risk of generating results that can negatively impact planning or policy decisions. In order to avoid such misuse, the uncertainty in the selection of assumptions and the subsequent uncertainty ranges of the results should be communicated transparently and the limits of forecasts should be made clear (they are ‘if then statements’). Scenarios with different assumptions that show possible influences and limits of planning or policy decisions on population development are suitable for this purpose; alternatively, more recently probabilistic methods are more widely used for this purpose. The presentation addresses these challenges using UN population projections for the world population (WPP), BiB/BBSR projection scenarios for regional population development in Germany and small-scale projections for Dutch municipalities from PBL/CBS.