Internal migration in the United Kingdom: an analysis of scale and zonation effects

John Stillwell, University of Leeds
Konstantinos Daras, University of Leeds
Nik Lomax, University of Leeds
Martin Bell, University of Queensland

Population redistribution through internal migration is an important and ubiquitous global phenomenon. The magnitude, intensity and spatial pattern of internal migration in any country depends on the size and shape of the areas for which data are collected. In this paper, we will use software developed as part of the IMAGE (Internal Migration Around the GlobE) project to examine what effect changes in the spatial scale (the number of areas) and spatial pattern (the configuration of areas) can have on different migration indicators in the UK. The IMAGE studio allows the computation of a suite of local and global indicators, including the mean distance migrated and the distance decay parameter calibrated using a doubly constrained spatial interaction model which provides a measure of the frictional effect of distance on migration. The aim of this paper is to compare different streams of migration using the same set of 406 areas, called Basic Spatial Units (BSUs), which are used for local government administration and central government resource allocation across the UK. These migration streams will include both aggregate flows from the 2001 Census and flow estimates for 12 month periods over the inter-censal decade commencing in 2001, as well as flows disaggregated by selected demographic and socio-economic variables. Initial analysis of the aggregate data suggests that there is a significant scale effect evident in the mean distance of migration which shows an exponential increase as the number of regions declines, but the zonation effect is minimal. On the other hand, the scale effect of the friction of distance on migration is relatively small when the spatial system contains over 40 regions but varies more with lower numbers of regions. Similarly, the aggregation effect is also more apparent when the spatial system contains relatively low numbers of regions.