Through you, Chair, an affordability metric like 30% is a relative metric. It depends on people's income, so it varies across income bands.
I think moving to a more explicit affordability criterion, as the U.S. has used for many years.... They defined affordability relative to a percentage of median income. A very low income is up to 30% of the median, which is a very low level. There is some analysis from colleagues here in Canada. The HART initiative is actually developing that same kind of metric in Canada.
I think we need to think more carefully about how we define affordability levels when we are asking both non-profits and developers to include affordable units in their projects. Currently, in the national housing strategy, it gets a little bit confusing because they have a number of different streams of funding that all use different definitions. For example, the RCFI definition, which uses 30% of household income for all households, is twice as high as the MLI Select criterion that Dan Dixon spoke about, which uses 30% of renter income, because renter incomes are half that of owners.
If we are concerned about renter affordability, we do need to examine the metric and come up with more a precise and also an absolute number. Rents below $1,000, rents below $800—we need some number as opposed to percentages of income because they mean all things to all people.