In this paper we re-examine the structure of internal migration flows in New Zealand. We use data on gross migration flows between the 39 main and secondary urban areas for the four intercensal periods between 1986 and 2006. We confirm that the gravity model – in which migration is inversely related to distance – fits the data well, and that the way in which distance is measured (travel time, road distance or straight-line distance) matters to some extent for the goodness of fit of the model. We also show that some forms of socioeconomic differences influence migration in the same way as distance. Perhaps surprisingly, we find that the deterrence effect of distance on migration was increasing, at least until 2001. Improvements in connectivity through reduced travel time have not increased migration flows.
Road distance and travel time for an improved house price Kriging predictor
