Thank you for that question.
Very much so. Sometimes, as they say, you have to leave the house to understand how good you have it, and also where it is you need to do some renovations, perhaps.
In both those positions it was very clear to me that we had some significant data gaps. It wasn't just that we don't have enough information, but in many cases it's defining what's relevant in today's context. There's the whole clean-tech sector, for example, that has a definitional issue: where does it end and where does it start? What is an innovation that goes toward clean tech? How do we capture that? How do we sustain that over time?
In fact, even when I was at Natural Resources Canada, I started to work with my colleagues at Statistics Canada on how we could improve statistics in the resource sector. It's getting that experience so that when you're moving a policy file or you're dealing with industry and you know there are gaps in what they're telling you, you can actually move things forward. That's not to say these are easy things.
On definitions, we obviously want to try not to be isolated in those definitions, because we live in a global world. Our industries compete globally, so those definitions, if you want to be able to compare with other countries, take time. It's about how we work with our colleagues internationally, how we take leadership roles in those organizations to move forward with things that are in our best interest.
It's the same thing on the health file. In the last couple of years I've been looking at things such as how we reduce sugar or salt and how we market that to kids, which are commitments of the government. Again, there are some real gaps in information about the interrelationships in the factors that contribute to obesity or mental health issues in kids.
There is a real gap. For kids under 12 years old, we really don't have much data, so once again it means working with Statistics Canada to fill those gaps. That takes money, and there are burden issues, so we have to be very cognizant of these things. They're not easy answers. We have to work to look at innovative approaches. In many cases the data are required much more quickly than would be required in a traditional survey, so we have to look at even more innovative ways to do those things, whether that is microsimulation or administrative data. In some ways it's a combination of survey data and administrative data.
Those are all giving me good incentives and imperatives to make some of the changes going forward at Statistics Canada.