The Federation of Canadian Municipalities' submissions—I'm sure you'll be hearing from them—are much more detailed in terms of the pieces of legislation that require accurate data.
The city, in particular, is governed by the planning act, and it requires that we establish a housing strategy, housing targets, and that we have a transportation plan, all of which rely on the data that we will no longer be able to rely on.
The Health Protection and Promotion Act also requires that we provide certain information, that we undertake surveillance of our populations around specific health issues. In particular, under section 7 of the Health Protection and Promotion Act, we have to be able to provide this kind of data. I know that Ontario Public Health Standards and Protocols require us to ensure that we are meeting the health standards. In order to do that, we must be able to assess the populations that we're reaching.
We must complete the public health assessment and surveillance protocol. That requires us to actually gather data on education, employment income, housing, and so on, and what I listed earlier. We simply will not be able to provide that information.
For our H1N1 pandemic planning, it's critical that we understand the populations and the characteristics of the populations by neighbourhood. The small geographic data will be far less reliable, and even Statistics Canada has said so. I'm not sure if you've heard what Statistics Canada itself has said, but even though they anticipate undertaking some different kind of sampling to try to mitigate against the non-response rate, the national household survey is anticipated to achieve a response rate of 50%, and there is a substantial risk of non-response bias. They talk about how they might mitigate; however, it is certain that there will be residual significant bias that will be impossible to measure and correct.
Even Statistics Canada itself recognizes that the bias in the national household survey will be difficult, if not impossible, to overcome through different sub-sampling.