House Price and Household Debt in Norway: An Econometric Analysis

It is commonly acknowledged that historically high levels of house prices and household debt pose risks for macroeconomic and financial stability in Norway. This thesis is aimed at understanding what are the driving forces behind the development in house prices and household debt in Norway since the 1980s. By surveying the empirical literature, the study by Anundsen and Jansen (2013) is found to be of particular interest. Their model of the interaction between house prices and household credit in Norway is studied in detail and re-estimated over an ex- tended sample. Following this, a re-specification of the long-run relationships of their model that seems to better match the extended data sample is introduced. Next, by building on Anundsen and Jansen (2013), a new model for the housing market and household debt is developed. Specifically, the econometric modelling is focused on challenging two assumptions made by Anundsen and Jansen (2013): First, that it is valid to condition on the supply of housing when modelling house prices and household borrowing. Second, that none of the variables in the information set are integrated of an order higher than one. Regarding the first assumption, the results suggest that it is valid to condition on the flow of new dwellings when estimating long-run relationships for house prices and household debt. Regarding the second assumption, two variables of interest in the analysis, real household debt and the housing stock, are found to be I(2) variables. A simple solution is suggested to ensure that all unit roots are accounted for in the econometric analysis. Like Anundsen and Jansen (2013), it is found to be of importance to account for mutual dependence between house prices and household borrowing when modelling the housing mar- ket. Both house prices and household borrowing are found to depend positively on disposable income and negatively on interest rates. Housing supply is found to depend on a proxy for the profitability of supplying new dwellings. Dynamic simulation of the model shows that shocks are amplified by self-reinforcing effects between house prices and credit. Over time, the supply of housing responds to changes in prices and dampen the credit and house-price cycles