Soft Canadian CPI is Stale on Arrival

• CPI inflation was weaker than expected

• Core measures conflicted again as BoC’s review reassesses which ones are best

• Breadth of inflation pressures eased off somewhat

• Next month’s headline CPI will surge and do so again in April

• Key to the BoC is the permanence of the oil shock…

• ...and how it transmits through core…

• ...as empirical questions that will take time and data to answer while on hold

• Canadian CPI m/m % / y/y %, NSA, February:
• Actual: 0.5 / 1.8
• Scotia: 0.9 / 2.1
• Consensus: 0.7 / 1.9
• Prior: 0.0 / 2.3
• Trimmed mean CPI m/m % SAAR: 0.6 (prior unrevised from 1.1%)
• Weighted median CPI m/m % SAAR: 1.6 (prior unrevised from 1.6%)
• Traditional core ex-food & energy % SAAR: 2.3 (prior 0.8 revised from 1.5%)

The rearview mirror showed inflation undershooting expectations in February, ‘February’ being the operative word here. The readings are as stale as week old bread sitting on the discount shelf and headed for the pigeons. Core measures conflicted once again, while the breadth of price pressures eased off somewhat. The rally in bond yields should be more focused upon future risks to the inflation outlook.

FIRST, THE DETAILS

Headline CPI was up by just 0.5% m/m seasonally unadjusted, or 0.1% m/m SA. The year-over-year rate ebbed to 1.8% from 2.3% previously and with the expiration of the temporary GST cut after mid-February 2025 contributing to this morning’s weaker year-over-year figure. The BoC will look through that effect.

The two keys were further evidence of soft core inflation (charts 1–3) and weaker breadth (chart 4) but with a big caveat on the former point. 

That caveat involves guesswork around what measure of core inflation is best. Our house research argues traditional core is best. The BoC has in the past published research saying the central tendency measures of core (trimmed mean and weighted median) are best, while press conferences emphasize a flexible approach that looks at everything. BoC guidance points to core measures being reevaluated in this year’s 5-year review and with possibly more added.

This matters because traditional core CPI (ex-food and energy) was up by 2.3% m/m SAAR which isn’t light, but follows a downwardly revised reading of 0.8% m/m SAAR in January from 1.5% previously. The three-month moving average of traditional core CPI is 2.1% m/m SAAR.

And yet trimmed mean CPI was just 0.6% m/m SAAR for a three-month moving average of 0.7%. Weighted median CPI was just 1.6% m/m SAAR for a three-month MA of 1.3%.

So what’s going on with core inflation? It depends, it seems, on which measure the BoC prefers.

Chart 5 shows that services inflation accelerated and chart 6 shows that core goods inflation decelerated, both behaving as directionally expected but on net by less than expected.

WHAT DOES THE FUTURE HOLD?

Regardless, the BoC will be more intensely focused upon forward-looking inflation risks as follows. Assessing them will take time for a central bank that is likely to be in data watch mode for an extended period while doing nothing at least for several meetings.

1. How much upside to headline inflation might arrive and for how long? The magnitude of the oil price increase and its effects on energy components within CPI (gasoline, home heating) could give rise to a 0.75–1% lift to inflation if moderately persistent over time based on the rule of thumb that each $10 lift gives a few tenths of upside. In the nearer term, I figure that next month's (March) inflation reading will get about a 0.5 ppt lift to the y/y rate from gasoline prices based on tracking so far to mid-month (chart 7). All else equal, that would drive the y/y headline CPI rate up toward 2¼%. It will go even higher in April when last April's elimination of the consumer portion of the carbon tax shakes out and that base effect will persist for many months.

2. Will the energy price spike stop there or get passed into core measures and when? A little of it may get passed through in first-round effects but the second-round effects could be more complicated and pointed toward the downside if real wages take a sustained hit. The ultimate outcome will be an empirical question settled by data and that will take time.

3. Canada—being a net energy exporter—gets a positive terms of trade shock from higher energy prices that usually causes hawkish concerns at the BoC. It's a positive imported income shock to broad national income that trickles down into higher incomes in the household, business (retain earnings) and government (improved balances) sectors. That will translate into higher incomes and usually translates into higher spending on net. It would be super unusual if a sustained positive energy shock didn't give rise to higher activity in Canada so be careful with arguments that lean to the contrary to typical experiences for commodity producers. If that happens, then the income effect could lift core inflation pressures independent of the effects of any pass through of higher energy prices into core measures. Again, an empirical question that will take time to assess.

MORE DETAILS!

Charts 8–25 break down parts of the CPI basket. Charts 26–27 do likewise in terms of m/m changes and weighted contributions. Charts 28–29 do the same thing with y/y rates. Also see the table showing more details and micro charts.