On the existence of time lags between changes in real estate prices and explanatory macroeconomic variables. A regression analysis applied to the city of Stockholm

The prices of real estate in a region change over time in response to changes in several economic background variables, such as population, income,unemployment rate, interest rate etc. In this paper a multiple regression analysis between real estate prices and some economic background variables will be described. This regression analysis will be applied to data for residential real estate transactions that took place in the city of Stockholm during the period 1985-1994. From this regression analysis the set of background variables that presents the strongest explanation of the variations will be identified. If a lag dependence between these background variables with the stronger explanatory power can be revealed, the model can be used in an early warning system. Then, it is possible to make predictions for these background variables in order to forecast the future development of the real estate prices. A summary of the main results is presented with a list of the variables with the stronger explanatory power. It was found that the introduction of time lags dramatically increased the explanatory power of the background variables, i. e. the cross correlation factors increased substantially. The values of the time lags were found to be in the range of 3-7 months between the background variables and the prices of the real estates. That means that a change in a background variable explains considerably better a change in the prices of the real estate occurring 3-7 months later than an analysis without a time lag. The validity of these results could be checked by applying the model here developed to other regions and real estate markets