Average impacted heavily by outliers during the pandemic. Median close to zero suggests that removing those outlier observations would suggest that there is very little bias in the flash estimates.

What’s the median one-month-later revision?

I didn’t have a strong opinion ex-ante, but I remember thinking easier monetary policy and a weaker currency should have helped the economy boost non-residential investment elsewhere and export growth. Instead, Canada seemed to absorb the slack via more housing activity.

The biggest problem is that nothing has filled the gap of the oil sector in ten years. I think even pessimists would have assumed the Canadian economy was dynamic enough to recover some of the lost ground.