In order to identify the variables influencing the Canadian housing market, we estimate a structural model on annual data from 1956 to 2001. This model simultaneously explains the behaviour of the real average housing price and the change in the housing stock as measured by the number of housing units. Four variables are significant in the price equation: real per-capita income, nominal interest rate on residential 5-years mortgage loans, population growth in the 25 to 54 age group, and housing stock. As to the housing stock equation, the only significant variables are real housing price and changes in real per-capita income. The real price dropped during the recessions of 1981-82 and 1990-92 and remained low during the rest of the 90s. The stagnation of the housing market observed during the last decade resulted from a slower population growth and a sluggish economy. The only favourable impact came from the drop in the interest rates. The model’s price forecasts for the next fifteen years do not encompass price decreases because the trend increase in real income will more than offset the impact of the slower population growth.