Could I make two quick points in response to Professor Sarlo?
First of all, on the measurement issue, I think he's wrong. I think there are good data out there. I've used them myself. You can go and read my report, published by the C.D. Howe Institute, on anti-poverty policies. I can give you the precise reference.
There are different kinds of data. He referred to one data set that was based on surveys. That's how data used to be collected, because that was the only way we could do it. There are now other ways to collect data, in particular, and it's done very carefully, very confidentially. All individual identifiers are eliminated. But referring to tax data, where we can get it, many of the issues he's talking about.... Most of the income data now coming out of Statistics Canada use those data because they are available, because they're able to use them, and because they provide better measures. So many of the problems that Professor Sarlo was talking about are history. It's not like that anymore. There is a variety of data sets. That's basically what Statistics Canada is generally doing.
Secondly, should there be an official line? First of all, I don't speak for Statistics Canada; I'm a visiting fellow there, as an academic, but I'm not an employee. The issue there is that they don't think it's the job of Statistics Canada to define the measure of poverty, precisely because it is subjective. It's not a statistical exercise, it's not a scientific exercise; it's a subjective exercise. So it is the job of other organizations, and individuals such as Professor Sarlo himself, to come up with a measure, which is a good one, among many. I believe that is why Statistics Canada has done that over the years.
If the government wished to define poverty itself, that would be the job of the government, I think, not the job of Statistics Canada. In fact, I think one of the advantages of studying poverty in Canada is that we now have a suite of measures. None is right; they're all reasonably good and have their advantages and their disadvantages. So we can use Professor Sarlo's measure; we can use the LICO; we can use the LIM; we can use the market basket income.
In the end, though, we're still going to have Renaissance Montréal, which sees that there are things you can do to help people who are poor get back into the mainstream. So we can still do it; we can still adopt measures.