Why did the Bank of Canada stop cutting interest rates?

Here's why:

As I noted several weeks ago, the low y/y inflation rates we're seeing now are due to a one-off fall in prices that occurred when the CAD hit parity with the USD. Since then, core CPI has been growing faster than the Bank's 2% target. And this trend was clear even before yesterday's May CPI release.

So the surprise isn't that the Bank didn't cut interest rates. The surprise is that the private sector was convinced that it would. From two recent stories in the Globe and Mail:

Why Mark Carney's honeymoon is over: “Our view is that core inflation is a dead issue in Canada and will remain that way for a long time yet,” said Derek Holt, vice-president of economics at Scotia Capital.

Mr Holt was not alone:

How Bay Street got it wrong on inflation: "Twelve out of 12," says Don Drummond, chief economist of the Toronto-Dominion Bank. No, he's not rating the new Sex and the City movie on some bizarre point scale. He's referring to the number of economists at a dozen Bay Street investment dealers who believed the central bank would cut its overnight lending rate again, to below 3 per cent.

I really don't understand how they could have all missed the surge in inflation and/or have convinced themselves that the Bank of Canada wouldn't have noticed it. It's true that no-one at the Bank telegraphed the move. But it's not clear why a central bank with an explicit inflation target would have had to telegraph such a decision; it's something that they should have been able to figure out for themselves.

Comments

[Nick Rowe]

Was it 3 and 4 month core inflation? Or was it rising commodity prices and the flat exchange rate? (Of course, it could be both, and you are right to draw our attention to the inflation data.)

The Bank has made it clear that it will not react to "type 1" changes in the exchange rate (those caused, or at least correlated with [don't shoot!] a change in commodity prices etc.), but will react to "type 2" changes in the exchange rate (those caused by something other than a change in demand for Canadian output). By implication (though, I admit, the inference is not 100% transparent), an increase in commodity prices coupled with no change in the exchange rate should be equivalent to a "type 2" depreciation of the exchange rate, to which the Bank should respond by raising the overnight rate. Which is what the Bank did (or, it raised the overnight rate relative to the cut it had previously planned).

In my opinion, the Bank's policy would be easier to predict if it simply said that it would respond to commodity prices (and other indicators of demand) AND to the exchange rate. If commodity prices rise, raise the overnight rate. If the exchange rate appreciates, lower the overnight rate. If commodity prices rise and the exchange rate appreciates, do nothing, etc. (all understood ceteris paribus, of course).

[J. Douglas]

I realize that this is about a month too late, but... I agree with your point that the shorter-term trends in core CPI have been picking up. However, using non-seasonally adjusted data certainly exaggerates the trends. I usually keep an eye on the 3-month and 6-month annualized trends, and as of May they were both sitting at 2.2% using seasonally-adjusted data.

[Stephen Gordon]

You're right: the core isn't seasonally-adjusted. That's news to me.