Canada’s Inflation Progress Continues…Maybe

Canada’s year-on-year inflation rates broke lower in April with a strong decline in Canada’s CPIx that excludes eight volatile components from the CPI. The CPIx has lower volatility than the CPI headline (1.7%) but its volatility is higher than the core (1.4% vs. 1.1%).

Economists -including some central banker economists among them- have expended some effort in looking at inflation measures that exclude certain things beyond the traditional core omissions (food, energy…and, maybe, tobacco). The Dallas Fed in the U.S. has offered a trimmed mean; the Cleveland Fed has a damped measure as well. There are other efforts in the U.S. to exclude certain components such as housing costs linked to high mortgage rates (one of the items jettisoned by Canada’s CPIx as well); during COVID, there was the removal of surging truck prices in the U.S. gauge.

My view of these efforts is that I have a lot more respect for Canada putting out a price index that eliminates volatile items at all times, offering a structural parallax view of inflation, compared to the U.S. approach of eliminating items that are pushing up the rate of inflation the most. The traditional approach is to look at core inflation instead of headline inflation from time-to-time when commodity prices flare (food or energy particularly). This has been done to take the focus away from the part of the inflation process that is most volatile. However, when inflation is really rising, these components likely will be the most volatile part of the index. By definition, the most volatile components will surge the most when inflation rises. So, if inflation - when it gets going - always is stimulated through the same components, downplaying the structurally volatile items will lead to a misappraisal of inflation and of how it may be gathering momentum.

Canada’s indexes show that CPIx inflation is the lowest now (year-over-year), but it was traditional inflation (CPI) that rose the most. The question is whether CPIx will continue to lead the way lower.

Despite the year-on-year picture, the table shows that over the last two months the Canadian CPIx inflation rates, smoothed to three-month averages of monthly changes, have picked up and exceeded both headline and core inflation on the same basis. In the U.S., for a while, it was pointed-out that the Dallas Trimmed Mean (DTM) inflation rate showed lower inflation than even core inflation rates in the U.S., but the DTM eventually caught up. I remain unconvinced that these ‘trimmed’ rates are all that useful. To the extent that the eliminated items are important, I think use of a trimmed mean or CPIx type measure may even provide a too-rosy picture of the economy, its problems, and its tradeoffs. For example, energy is notably hard to substitute away from in the short run. Housing costs are what they are and if higher interest rates (especially such as a ‘higher for longer’ policy in the U.S.) are exacerbating housing costs, isn’t that part of an enduring U.S. reality? The same is true, of course, for Canada and its housing costs – but Cananda does not labor under a higher-for-longer policy.

In short, Canadian inflation, like inflation in the U.S. and in Europe, is coming down. However, as elsewhere globally the near-term trend is still not well-determined and remains highly speculative. The CPIx in Canada with a lower gain over 6 months and 12 months but with a 3-month pace surpassing the headline and core rates may be a red-flag. It is not clear that the CPIx still points the way lower based on its trend that applies to generally less volatile CPI elements. And since this measure -less the most volatile items- is now rising fastest- does this indicate a growing inflation problem? However, as I note above when these ‘individual’ volatile items are removed from the package called ‘the CPIx’ it did not become automatically less volatile than the headline. In terms of volatility, the CPIx occupies a middle ground between the headline and the core. That must make the entire process of relying on the CPIx somewhat suspicious, less reliable, and less useful. The Bank of Canada looks at a variety of indicators to judge inflation. While that may give us less clarity on precisely how it sees inflation, it certainly informs the bank better and leaves it less in the grip of the foibles of any one measure. The bank of Canada formally targets the CPI in a 1% to 3% range.