Perhaps I could add to that.
From an analytical point of view in trying to figure out what's happening, this is not from what should be our official poverty line because we're not going to choose that.
As Sylvie has said, the market basket measure, because it's more of a cost-of-living-based measure, is definitely more useful when comparing cities or provinces in Canada, for instance, than the LICO, because the LICO doesn't really account for cost of living differences, and that's important. The MBM is good for that.
The LICO has the time series and you can see a long history. If you want to see what's happening now as compared to 20 years ago, you're pretty well stuck with the LICO.
There are two other measures that I would definitely pay attention to. One is the persistent low-income measure; that is, using longitudinal data to ask who is in chronic low income. A lot of people who show up in the LICO, for instance, may be in low income only for a few months or a year. Who are the people who are in low income for four or five years? I would definitely want to focus on chronic low-income measures, and that's exactly what HRSD has done in the identification of those five groups.
There is one other measure we haven't mentioned here at all today. We've been talking in terms of the measurement issue about the low-income cut-off, or what the line is below which you are said to be in stricken economic circumstances. We haven't talked about the depth of low income.
If the low-income cut-off, for the sake of argument, is $20,000, but most of the poor people are at $18,000, that's one situation. You could still have a 10% low-income rate. If the low-income cut-off is at $20,000 and you still have 10% of the population below that but they're at $10,000 average income and not at $18,000, that's a very different situation. Those people are much poorer. Nowhere, if you're only talking about low-income cut-off, do you actually see that, because it's going to look the same; those two scenarios will look the same at a 10% low-income cut-off. So the other issue is the gap or the depth of low income.
Another reason that's important is that a lot of government programs, such as social assistance, for instance, will often increase their income. They stay below this cut-off sometimes, but their income rises. If you're only looking at the low-income rate, you won't see that improvement. You have to look at the depth of low income or how far below the cut-off people tend to be.
That's another measure I would pay attention to if I were trying to figure out what's happening.