The key metric that is often looked at is the business expenditure on research and development, the so-called BERD. You are right, that has been in decline, and most recently StatsCan has released a report with a further small decline. It is important, though, to nuance the interpretation of those results for a number of reasons.
First of all, innovation is not necessarily just investments in R and D. It is about a broader concept, a number of which Gerard has highlighted that relate to innovation on process and management capacity that wouldn't be captured in those metrics. Certainly the most recent figures for 2014 are only an indication of intent. They are revised frequently within the course of a year, and we look at those numbers regularly because one of the challenges in the innovation field is that there is an important lag effect between when an initiative is undertaken, or a bundle of initiatives, such as the work that was undertaken in response to the Jenkins panel, and seeing an effect in the economy.
Typically an expected lag can be on the order of about five years, two or three years for something to take effect, and then another two years to measure it. So the lag effect is also an important component, and what we hope to see is the effect of some of the more recently taken initiatives coming online, so to speak, and being visibly seen in the statistics around 2017-18. There's a lag, unfortunately.