Regional differences in housing price dynamics: panel data evidence

The aim of this study is to examine empirically the regional differences in housing price dynamics by estimating panel data models that allow for regional variation in the dynamics. In particular, we investigate whether there are large regional differences in the housing price ímomentumîù (íbubble builderî) and in the speed of adjustment of housing prices towards their long-run fundamental level (íbubble bursterî). Using annual data for 14 Finnish cities for the period 1988-2009, we estimate fixed-effects models that include both main effects and interaction effects to allow for regional differences in the dynamics. While, the results indicate that the differences across regions are generally quite small in Finland, the analysis shows that the adjustment towards the long-run sustainable price level is significantly slower in the Helsinki Metropolitan area (HMA), the largest urban area in the country, than in the other cities. The results also provide evidence for cointegration between housing prices and income.