Assessing the Accuracy of National Population Projections

A Case Study of Norway

Authors

• Michael Thomas Statistics Norway, Research Department
• Astri Syse Norwegian Institute for Public Health, formerly Statistics Norway, Research Department
• Adrian Rogne University of Oslo, Department of Sociology and Social Geography
• Rebecca Gleditsch Norwegian Institute of Public Health, formerly Statistics Norway, Research Department

DOI: https://doi.org/10.23979/fypr.109057

Keywords:

Accuracy, Errors, Methods, Fertility, Mortality, Migration, Population projections

Abstract

Few producers of official population projections provide regular evaluations of past projection inaccuracies. This paper assesses deviations between the projected and registered total population for Norway (1996–2018), as well as deviations in the age structure, total fertility rate and number of births, period life expectancy at birth and number of deaths, and net international migration. Projected life expectancy was consistently lower than the real development. Few systematic deviations were observed for fertility up to 2009, but thereafter fertility has been consistently overprojected. However, the deviations between projected and realised trends in births and deaths have been relatively small as compared to those for net international migration. The projections produced between 1996–2005 underestimated long-term population growth due primarily to the unforeseen increase in immigration following EU expansion in 2004. More recent projections contain no consistent under- or overprojection of net migration and the deviations for the total population have been moderate.