I study the general equilibrium of the housing market in an economy populated by overlapping generations of households. A contribution of the present paper is to solve for the housing market equilibrium in the presence of aggregate (interest rate) uncertainty with a realistic mortgage contract. In addition, households also face idiosyncratic uncertainty resulting from stochastic changes over the lifecycle in tastes (or needs) for housing. In this environment, profit-maximizing banks offer fixed-rate mortgage (FRM) contracts to homebuyers. As seems plausible, each housing market transaction is subject to a fixed cost, which gives rise to S-s policy rules for housing transactions: existing homeowners change the size of their houses only if there is a sufficiently large change in the state of the economy (i.e., in interest rates, in their taste for housing, etc.). A plausibly calibrated version of the model is consistent with three empirically documented features of the housing market: (i) highly volatile housing prices and transaction volume, (ii) a strong positive correlation between transaction volume and housing prices, and (iii) a significant negative relationship between interest rates and housing prices, which can rationalize a large part of the recent boom in housing prices in the US and around the world.