The relatively hawkish language in the Bank of Canada's interest rate decision -- most notably the removal of the word 'eventually' from the sentence describing the conditions in which interest rates will increase -- took financial markets by surprise.
Central banks try to avoid surprises when they can, but in this case the Bank has the best of excuses: the facts changed.
The Bank of Canada updates its estimates for the output gap -- the difference between actual and potential GDP -- each time it releases its Monetary Policy Report. These estimates play a key role in the Bank's decisions, but they are also based on data that are subject to revision -- and revisions to output gap estimates can be very large. What may seem like a sensible policy stance given the data that are currently available can turn out to be a mistake when those numbers are revised.
Statistics Canada's release of its national estimates for the first quarter of the year are accompanied by a significant revision of the previous year's numbers, and these changes induce a revision of the Bank's estimates for the output gap. It turns out that the new data suggest that the output gap is and has been much smaller than what the Bank had previously thought; see the accompanying graph.
These new numbers may well be revised away in the coming months, but policy makers have to work with the data they have before them. If you take an output gap that is shrinking much faster than you thought and add it to a core inflation rate that is drifting towards and perhaps past the Bank's 2 per cent target, you will find yourself in a position where you have to start preparing to increase interest rates earlier than you had planned.
Follow Economy Lab on twitter