Canadian CPI is Good Enough for a Hawkish Cut

• Core inflation gauges remain within BoC’s target range

• Breadth of price pressures has improved

• Why the BoC will cut next week

• Why it shouldn’t, or why it should at least sound hawkish

• Canadian CPI m/m NSA // y/y %, September:
• Actual: +0.1 / 2.4
• Scotia: 0.0 / 2.3
• Consensus: -0.1 / 2.2
• Prior: -0.1 / 1.9
• Trimmed mean CPI (m/m SAAR %): 2.8
• Weighted median CPI (m/m SAAR %): 2.8
• CPI ex-food and energy CPI (m/m SAAR %): 2.3

Canadian core inflation measures remain good enough for the BoC to cut next week when properly evaluated in terms of month-over-month trends and breadth. I’ll explain my views about why it should be a careful, hawkish sounding cut.

Key is that each of the main core measures of inflation were well within the flexible 1–3% headline inflation target range and were likely overstated by mechanistic seasonal adjustments that may not be appropriate.

Chart 1 shows the trimmed mean and weighted median core measures that were both 2.8% m/m at a seasonally adjusted and annualized rate (SAAR). Chart 2 shows the traditional core CPI measure that only excludes food and energy and was 2.3% m/m SAAR.

Chart 3 shows the pattern over 2025 for all three measures. They’ve generally been trending within the target range for a while now.

Core goods inflation was weak as the pressure came from the services side of the picture (charts 4, 5).

The BoC will mostly welcome the breadth readings (chart 6). They show that the share of the CPI basket that is cruising above 3% or 4% m/m SAAR is between roughly one-third and 40%.

Chart 7 demonstrates that traditional core CPI was among the lowest on record for like months of September over time. What propped up the SA reading for traditional core in m/m SA terms was a high seasonal adjustment factor (chart 8). 

Statcan will say that the SA factor is just a mechanistic outcome of applying standard X12A seasonal adjustment methodologies that are commonly used by many data agencies. That’s not the same as saying we should take it at face value.

Chart 9 applies different SA factors drawn from the history of SA factors for months of September to show hoe core CPI would have changed. At most other SA factors in time, traditional core CPI would have been weaker than reported and possibly even negative. 

BoC CALL

There is plenty of precedence to cut after the average of TM and WM cpi landed at about 2¾% m/m SAAR. In fact, they've done so a half dozen or so times in the past (chart 10). I think the BoC will work the flexible inflation target range that Macklem keeps emphasizing and deliver easing next Wednesday but with a hawkish sounding and noncommittal feel.

Why cut?

• They’d need good arguments against a cut that I don’t think they have in relation to disappointing the two-thirds market pricing of a 25bps cut. From a risk-reward standpoint, it may be more difficult to justify holding than to cut and, if they agree, sound like they’re shifting to the sidelines.
• Core inflation measures are in the range on a higher frequency m/m SAAR basis for a while now.
• Their July MPR expected inflation to be around 2% by the end of 2026 in the current tariffs scenario. Developments since then likely have them thinking there is more downside to that forecast than upside, ergo give it a nudge.
• One cut doesn't do it given my longstanding bag of chips metaphor and that you can’t just pull one chip out of the bag.
• Jobs rebounded, but the BoC always fades just one spot number for the volatile household survey and the trend is weak.
• GDP is weakening with basically no growth being tracked in Q3 so slack continues to open up. On the one hand, that's why they were cutting from last July to this March in anticipation of a souring economy. Plus, the economy is broadly tracking their expectations in the July MPR. But key may be Macklem’s ongoing guidance they expect a slight improvement in growth by at rates still below the economy’s potential GDP growth rate which means more slack opening up over time.
• You could argue the Budget's influences both ways. Wait to see if PM Carney is going to prime the pump a lot, versus act now because it might be optically harder to act after the Budget is presented and passed. I think at this juncture, the BoC will say they want to err on the side of combining policy measures.
• Macklem is a dove at heart. I rarely take him seriously when he jawbones inflation risk. He’s a labour market guy at heart as he was when delivering fully inclusive speeches as Senior Deputy Governor under BoC Governor Carney at the time and more recently in the pandemic.
• Trade policy uncertainty will be elevated for a long time yet. Talk of a possible agreement soon sounds very limited, mostly metals, with the negotiators distancing themselves from rumours. Other tariffs have worsened.

Why not cut, or why to sound hawkish?

Then there are plenty of counter-arguments to easing now which at a minimum should have the BoC leaning against extreme dovish views.

• the real policy rate is zero and we're already well within the neutral r* range.
• Financial conditions are buoyant. There is no financial crisis notwithstanding persistent risks.
• Monetary policy entails lagging effects and we’re still inside the full pass through of 250bps of rate cuts that began in June of last year through to March of this year. Give it time for a zero real rate to work through.
• there are more drivers of inflation risk than just output gaps. Costs are under upward pressure through the whole value-added supply chain. Labour settlements are too hot, productivity is not, inventories are high, and supply chains are at a highly nascent stage of being revamped in the US-driven global trade war and revamping them entails higher costs. Someone pays those higher costs. Sensible incidence effects would have one thinking everyone will share in it including higher prices for consumers in a longer wave sense.
• so overall, I'm still of the belief that if I were Macklem, I would not be easing now. I would take a rather BoE-like approach to ‘gradual’. Spread it out, and assess new information along the way.
• and I still think the tariff shock to Canada is often exaggerated. The effective tariff rate has bumped up to 6%. USDCAD over 1.40 is doing what a flexible exchange rate should do to help assuage the shock to the terms of trade. Remove the tariffs on the most affected narrow sectors, and the tariff shock is tiny given the high CUSMA/USMCA compliance rate.
• Canada has the lowest tariff shock against any notable US trade partner by far. Anything can happen to CUSMA/USMCA but you can't conduct monetary policy when the outcome could be bi-modal in nature.

DETAILS

Charts 11–20 give breakdowns of select components of the CPI basket. The recreation category’s 0.6% m/m SA rise was driven by travel tours as the biggest gainer. Shelter cost inflation has subsided but not because of direct primary rent that remains warm. Transportation costs were buoyed by gasoline. 

Charts 20–21 provide a breakdown fo the whole y/y CPI basket in raw terms and in terms of weighted contributions to the overall index’s rise. Charts 22–23 do likewise in m/m terms.