Canada's September 2026 Inflation: CPI Rises to 3.1% While CPI-trim and CPI-median Fall

Headline inflation in Canada edged up again in September 2026, to 3.1% year over year. Yet all three core inflation measures the Bank of Canada tracks moved lower. Our article on the August 2026 CPI, with shelter at 1.5% and headline at 3% covered the previous month. This one explains the gap that opened up in September: what each core measure actually tracks, why it diverges from headline CPI, and why it is the number the central bank watches most closely.

September 2026: headline up, core down

The figures come from two primary sources. Statistics Canada released "Consumer Price Index, September 2026" in The Daily on October 19, 2026. The Bank of Canada publishes, on its core inflation measures page, the three indicators it calculates from the same consumer price data.

The two sets of numbers are heading in opposite directions. Headline CPI gained a tenth of a point, while CPI-trim and CPI-median each lost one and CPI-common lost two. The gap between headline and core inflation is widening, not closing.

What drove headline CPI: gasoline and energy

September's increase was highly concentrated. According to The Daily, gasoline was up 27.8% year over year, after 22.8% in August, and energy as a whole was up 19.0%, after 15.4%. Increases that large are enough to move the headline index on their own, even though they touch only part of the basket.

The proof shows up in the indexes that leave them out. CPI excluding gasoline slowed to 2.2%, from 2.4% in August. CPI excluding energy came in at 2.1%. In other words, strip out energy and price growth is close to 2%, and slowing. Another sign pointing the same way: groceries rose 1.7%, down from 2.8% in August, their smallest increase since May 2024.

Month over month, CPI rose 0.1% on a raw basis and 0.4% seasonally adjusted. The second series removes seasonal patterns, which accounts for the difference between the two.

What CPI-trim, CPI-median and CPI-common actually measure

All three core measures start from the same prices as headline CPI, but they process them to bring out the underlying trend rather than one-off shocks.

• CPI-trim: the average of price changes after excluding the most extreme moves, up and down. A spike in gasoline like September's is trimmed out.
• CPI-median: the median price change, the one sitting in the middle of the distribution. A handful of categories surging does not move it.
• CPI-common: the common component, meaning the movement shared across price categories as opposed to what is specific to each one.

None of the three is better than the others in absolute terms. They complement each other: when all three ease together, as they did in September, the signal that the underlying trend is cooling is stronger than if only one of them moved.

That also tells you what to check next month. When the October figures come out, the useful question is not just where headline CPI lands, but whether CPI-trim, CPI-median and CPI-common keep moving in the same direction. Three measures easing together a second month would reinforce the September signal; a split among them would weaken it.

Why the Bank of Canada watches core inflation

A central bank acts on inflation with a lag: a rate decision plays out over several quarters. So it needs to know whether the price increases it sees are broad and persistent, or concentrated in a few items whose prices can reverse as fast as they rose.

Gasoline is the textbook case of the second kind. Its price is set in global markets that a policy rate cannot reach directly. Reacting to every swing at the pump would mean tightening or easing credit for a shock monetary policy cannot fix. Core measures let the Bank look through that noise.

What September 2026 puts in front of the Bank is therefore a split picture: headline inflation above 3%, and underlying inflation around 2% and cooling.

Shelter in the September CPI

Shelter rose 1.8% year over year, up from 1.5% in August. The breakdown shows very different moves line by line:

• rent: +2.8%;
• homeowners' home and mortgage insurance: +4.9%;
• mortgage interest cost: −0.2%;
• homeowners' replacement cost: −1.9%.

Two of these lines are rising and two are falling. The Daily does not weight them against each other, so the size of each line's contribution to the 1.8% total cannot be read from these figures alone.

Mortgage interest cost, which reflects what households pay in interest on their loans, is now slightly lower than a year ago. It is rent and insurance doing the lifting. These are national figures: The Daily does not support a Quebec-specific reading, and this article does not offer one.

What it means, and doesn't mean, for borrowers

For a buyer or a homeowner coming up for renewal, the practical question is simple: does this report make a rate hike more likely? The honest answer is that it gives ammunition to both sides. Headline CPI above 3% argues for caution; core measures easing toward 2% argue against tightening on the back of energy alone.

This report is only one of the inputs the Bank of Canada will weigh at its October 28, 2026 decision, whose outcome is not known at the time of publication. The next marker comes on November 16, 2026, when Statistics Canada releases the October CPI.

In the meantime, the best protection is a budget that works across several rate scenarios rather than a bet on one. Your lender or mortgage advisor can stress-test your own situation; a broker can help you target a purchase price that leaves that margin.