In fact, the way we measure the number of housing starts right now is based on the spread footing, or the basic foundation being poured.
Why are we doing this? It's because, once the foundation has been poured, we have clear confirmation that a building will indeed be built there and that it will follow the established plan for the residential units. It's also because, as long as the project isn't at the excavation stage, it could still be revised or the land use could change.
Generally, once the foundation is poured, 99% of the units are built and delivered to the market. So this is a stage where we really have a high degree of certainty that the units promised in the building permit will actually become housing comprising the planned number and size of units.
There are also other ways to measure housing starts. For example, other companies collect data on excavation. With respect to excavation permits, not all municipalities issue them, and those that do don't have a uniform definition for them. When an excavation permit is issued, the number of units associated with the permit isn't specified. So the information available is really quite fragmented. It's also not comparable across municipalities.
What we're going to try to do in the survey I was referring to in response to the previous question is not only determine the number of pre-sales and their rental market equivalent, but also map a building's progression throughout the housing continuum, including the point at which we will define what constitutes an excavation. We will therefore be able to obtain comparable data in all participating urban centres, rather than data that varies widely from one urban centre to another.