This article provides new empirical evidence on the dynamics of price movements and transaction volume in the housing market using data from Finland. While the previous related literature studies the reactions of sales volume and prices to an interest rate shock only, we investigate the responses to income and credit shocks as well. Based on an estimated vector-error correction model, the response of prices to demand shocks is found to be substantially slower than that of sales. The effect of a demand shock on sales peaks within a quarter from the shock. The results show that the differences in the reaction patterns to demand shocks can create the kind of strong positive co-movement between price movements and sales and the kind of negative correlation between price level and sales that has been found in a number of housing markets. It is also found that the direct predictive ability of the fundamentals with respect to housing price growth and sales is overwhelmed by the predictive power of the lagged observations on price changes and sales themselves.