• The BoC’s preferred core gauges fell back to one-handled m/m annualized gains
  • Breadth of price pressures remains significant
  • The assumed bounce back in recreation prices didn’t happen…
  • ...as steep travel discounting weighed on the category
  • Rail strikes and methodological changes to cellular services lie ahead
  • Markets barely reacted to a largely priced outcome
 
  • Canadian CPI m/m // y/y %, NSA, July:
  • Actual: 0.4 / 2.5
  • Scotia: 0.5 / 2.6
  • Consensus: 0.4 / 2.5
  • Prior: -0.1 / 2.7
  • Trimmed mean CPI: 2.7% y/y (2.8% prior revised from 2.9%); m/m SAAR 1.7% (2.9% prior)
  • Weighted median CPI: 2.4% y/y (2.6% prior); 1.1% m/m SAAR (3.5% prior revised from 2.9%)

Canadian core inflation ebbed in July using the proper measure calculated as the month-over-month percentage changes in seasonally adjusted trimmed mean and weighted median CPI at an annualized rate (chart 1). They landed at 1.1% m/m SAAR for weighted median and 1.7% for trimmed mean. Traditional core CPI ex-food and energy was also soft but a touch firmer at 2.4% m/m SAAR (chart 2).

Chart 1: BoC's Preferred Core Measures; Chart 2: Canadian Core Inflation

The figures are volatile. Both of the BoC’s preferred core readings were soft from January through to April before they accelerated to revised averages of 3.5% in May and 3.2% in June. Falling back to an average of 1.4% in July is welcome relief but continues to merit caution along a volatile path.

There was no market effect on September rate cut expectations or the market’s view that 75bps of cuts will be delivered over the three meetings in September, October and December. Our forecast is for two more cuts this year including September, ending the year at 4%. We tentatively assume a skip in December partly based on uncertainty in the aftermath of the US election, domestic data uncertainty, and the potential contents of a Fall fiscal update into an election year.

Details

Headline CPI m/m SA was 0.3%, meaning that most of the 0.4% m/m seasonally unadjusted rise in prices was seasonally unusual.

Breadth is shown in chart 3. While off from the peaks, a still considerable portion of the basket is showing price gains over 3% m/m SAAR and over 4% m/m SAAR.

Chart 3: CA Inflation Breadth

Transportation prices including vehicles and gas were up 0.5% m/m SA. Household ops/furnishings were up 0.5%. Food was up 0.3%. Clothing increased by 0.3%.

There was a mild pick-up in goods inflation but it was small effect (chart 4), and this was offset by less services inflation (chart 5).

Chart 4: Canadian Goods Inflation; Chart 5: Canadian Service Inflation

On the services side, there were softer pressures in core services (chart 6) and also somewhat less pressure on shelter costs including rent (charts 7, 8).

Chart 6: Canadian Services ex Shelter Inflation; Chart 7: Shelter Cost in Canada; Chart 8: Rents in Canada

A small part of the reason for the weak m/m weighted median was due to an upward revision to June's m/m SAAR WM rate to 3.5% from 2.9% previously.

Apparently we're not having much fun this summer. Or at least not in terms of travel related activities. We got 0.4 m/m headline CPI instead of my 0.5 estimate in part because I had factored in an assumed rebound in the recreation/education/reading category that carries about a 10% weight in CPI. That didn't happen. We're getting seasonally unusual weakness in that leisure category that fell -0.7% m/m SA in June and then another 0.2% in July. Falling prices for travel tours are a key part of it in the breakdown (chart 9).

Chart 9: Breakdown of Monthly Changes within Recreation Education and Reading CPI Category

It's easy to imagine a bunch of potential factors behind this category's softness but I'm still a bit surprised. CAD makes travel abroad more expensive but should incite more foreign travel to Canada especially from the US given Canada is a very price sensitive travel destination. Pressures on the finances of some households are not helping, but other households like the two-thirds of households without a mortgage are in better shape. Airfare is expensive in Canada as a perennial factor. Maybe weather also contributed to price discounting as it has been a stinkin' hot summer (with quite a lot of rain) in big parts of the country.

COMING RISKS

Figures for August and September may be interesting on three counts. One is regular data uncertainty in the context of ongoing upside risks to inflation (see morning note), but the other two are unusual.

Near-term price pressures partly depend on how long the likely rail strike lasts and how pervasive its effects may be. What has me worried about the duration is that both sides are saying they remain far apart and that the labour regs have changed given the left wing Federal government’s pro-union policies that are against intervention. This raises the risk of supply shocks at a key time for retailers that are ordering their goods for the holiday season. A short strike is no big deal. A prolonged one could be quite disruptive to stocking and prices. Ditto for some key commodities like potash, wheat, etc. Oil by rail is less susceptible with those flows roughly tied with the lows of the past decade in part due to TransMountain.

Of much lesser concern is that Statcan has stated there will be methodological changes to how it captures cellular services prices in next month's CPI for August (here). Charts 10–11 when combined with a 1.5% weight on cellular services in CPI show why it may matter somewhat. For example, when the category was down by about -25% y/y earlier this year it was contributing -0.3% to -0.4% to the deceleration in y/y inflation. My eyes started to glaze over while reading their methodological update but I'm unsure what effect the changes may have on this category and its weight in CPI.

Chart 10: Canada CPI: Cellular Services; Chart 11: Canada CPI: Communications

The cellular category is a great example of the difficulties associated with measuring inflation. Recall it's not just about measuring prices, but also quality changes that get factored into CPI. What you've observed on your bills is only part of the issue. More data, broader networks, faster speeds, etc also factor into inflation stats through hedonic adjustments.

Please also see the collection of charts on the following pages and the detailed table at the back of this publication.

Chart 12: Canada CPI: Gasoline; Chart 13: Canada CPI: Food Prices; Chart 14: Canada CPI: Transportation; Chart 15: Canada CPI: Private Purchase of Passenger Vehicles
Chart 16: Canada CPI: Air Transportation; Chart 17: Canada CPI: Homeowners' Replacement Cost; Chart 18: Canada CPI: Clothing & Footwear; Chart 19: Canada CPI: Alcoholic Beverages & Tobacco Products
Chart 20: July Detailed Category 12-Month Change in Canadian CPI; Chart 21: July Detailed Category Contributions to 12-Month Change in Canadian CPI
Chart 22: July Detailed Category Monthly Change in Canadian CPI; Chart 23: July Detailed Category Contributions to Monthly Change in Canadian CPI
Chart 24: July Single-Month Components Included and Excluded from Bank of Canada Trim Core CPI Measure

Markets Barely Flinched

There were relatively modest market moves in the aftermath of the release. USDCAD moved up (ie: CAD depreciated) by about a third of a cent. The 2-year GoC moved down by 4–5bps and is matching the performance of US 2s this morning. The September BoC meeting was priced for -25bps and remains as such. Only 2–3bps was added onto market pricing for -75bps more to year-end.

Now onto things that matter more, like Jackson Hole...

Table: Canadian Inflation Component Breakdown
Table: Canadian Inflation Component Breakdown