The effect of credit conditions on the Dutch housing market

It is widely perceived that the supply of mortgages, especially since the extensive liberalization of the mortgage market of the 1980s, has had implications for the housing market in the Netherlands. In this paper we introduce a new method to estimate a credit condition index (CCI). The CCI represents changes in the supply of credit over time, apart from changes in interest rates and income. It has been estimated by an unobserved component in an error-correction model in which the average amount of new provided mortgages per period is explained by the borrowing capacity and additional control variables. The model has been estimated on data representing first time buyers (FTB). For FTB we can assume that the housing and non-housing wealth is essentially zero. The CCI has subsequently been used as an explanatory variable in an error-correction model for house prices representing not only FTB, but all households. The models have been estimated on quarterly data from 1995 to 2012. The estimated CCI has a high correlation with the Bank Lending Survey, a quarterly survey in which banks are asked whether there is a tightening or relaxation of (mortgage) lending standards compared to the preceding period. The CCI has explanatory power in the error-correction model for house prices. In real terms house prices declined about 25% from 2009 to 2012. The estimation results show that nearly half of this decline can be attributed to a decline in the CCI.