September's CPI report this morning again came in hotter than expected, with the Core CPI increasing 0.6% month-over-month vs. expectations of 0.4%. A deeper look provides some grounds for optimism, however.
First, as we've discussed before, I focus on the seasonally adjusted monthly price changes in the core index because they provide a more "updated" view of inflationary momentum in the economy. Looking at the year-over-year numbers in this kind of environment is not as helpful, simply because the economy is in a completely different place today than it was even 6-12 months ago. Looking at 1 month and trailing 3 month core figures therefore provides a better reality check on inflation than the more commonly cited year-over-year figures.
With respect to the actual data this month, there's really two points to make. First, while the headline number came in hotter than expected, if you look at the median and average increases across a broader array of categories, those continue to come in well below the aggregate index increases. This suggests that certain categories (like rent) with larger weightings are driving disproportionate increases in the aggregate index. Notice the spreads between the lines (which are average and median increases) versus the bars (which represents the aggregate core CPI index change) in the charts below. Note that "SI" = Sub-Indent 4, or Category Level 4 in Table 2 of the CPI releases, and encompasses 55 goods and services baskets that comprise about 98% of the core CPI index. Sub-Indent 5 in the chart below that represents a deeper layer of 101 baskets of goods and services but only comprises about 77% of the core CPI. Both are useful but 4 is probably a safer metric.
Here is the monthly chart. Look at the gaps between the lines and the bars the last 2-4 months. This tells you inflation is probably below the levels implied from the index (so maybe ~4% annualized rather than the ~7% implied by the 0.6% aggregate index change).
Now here's the 3 month average to smooth things out, except this time, I've included Sub-Indent 5 that we discussed above, and annualized it as well to make things easier. Again, notice the gaps between the lines and the bars. This indicates inflation is actually getting better faster than the core index indicates. To be clear, inflation is still clearly too hot, but these broader metrics indicate we are probably running closer to 3-4% inflation right now rather than the ~6-7% that's indicated by the aggregate core index changes.
The second key take from the data today is around the goods versus services dynamic. As a quick reminder, when the lockdowns happened, people shifted their spending from services to goods, and then as the economy has normalized, they did the reverse of that. This has increasingly shown up in the supply chain for goods first (which has pressured trucking rates, ocean freight rates, etc.), but may be increasingly starting to show up in actual consumer prices as well.
This month, goods inflation continued to slow and actually returned to its pre-COVID contribution level, which was essentially zero. Prior to COVID, goods essentially contributed zero to inflation in the US economy, and we appear to be returning towards that trajectory. Notice in the charts below the steady decline down to zero in the current month of September. This was achieved in spite of the fact that Household Furnishings and New Vehicle prices both continued to increase at significantly higher than normal levels (for example, at 0.04% per month, New Vehicles are still contributing over 0.4% to aggregate core inflation on an annual basis). Home Furnishings and Supplies are still contributing about 0.4% on an annualized basis as well. Prior to COVID, both of these categories essentially contributed zero to inflation. With interest rates where they are, it seems likely that both of these categories will be under pressure in coming months, as both housing market and auto market softness continues. This should provide a tailwind of potentially 0.8% (80 bps) to aggregate core inflation in coming months, especially if we have a soft holiday season, which feels more likely than not. Used car prices coming down would be another source of tailwind.
That being said, while goods inflation is moving in the right direction, services remain an issue. If goods inflation was zero this month, that means services contributed the entirety of the 0.6% monthly increase. This is problematic on the surface because services prices tend to be stickier than goods prices, and thus harder to change. What's driving the services inflation? 3 things: rent, medical services, and transportation services. For reference, Medical Services consists of things like doctors visits, dental services, hospital and related services, health insurance, etc. Transportation services includes things like leased cars and trucks, car and truck rental, auto maintenance services, parking and tolls, bus fares, and airline fares, among other things.
These three baskets are all inflating at levels well above where they were pre-COVID, and while some are likely to remain inflationary (auto body care, for example, feels sticky), as we've highlighted in other posts (see particularly here: https://www.citizenanalyst.com/post/inflation-macro-datapoints-and-company-commentary-for-the-week-ending-10-2), rent inflation is almost certainly coming down, and in many cases rents actually appear to be decreasing. Home prices are also now starting to decrease. Given the Shelter component of CPI is contributing 15 bps extra per month to core inflation than it was prior to COVID, this amounts to an excess of 1.8% of inflation in aggregate each year. Given this should reverse in coming months as the lag works its way through the CPI price collection process, that's a meaningful tailwind in the pipeline.
The same can be said for Medical Care and Transportation services with respect to their excess contributions, but in these categories, there's less visibility there into a correction. Because of the way the CPI is constructed with respect to health insurance, however, Omair Sharif makes the case here (https://www.bloomberg.com/news/articles/2022-10-03/odd-lots-podcast-omair-sharif-explains-rent-healthcare-and-cpi-data?srnd=oddlots-podcast) that a large tailwind may be coming our way in Medical Services too. It's complicated, but if you're interested, its a fascinating listen. Labor inflation in the medical care space has likely been a key source of price inflation as well, but this seems to have improved in recent months, which may bode well for slower increases there as well. Time will tell there.
Transportation Services remains well above pre-COVID levels, but it's hard to know whether or when this might correct.
All other services outside of our three problematic baskets seem to be generally inflating at pre-COVID levels at this point, which is good.
Bottom line, while getting into the category-by-category business can be tricky, there's grounds to be hopeful now that don't rely on just one category improving. If the median and average category increases produces inflation closer to 4% right now instead of 7%, and rent is potentially contributing an excess of 1.8% a month as it stands today, the path back down to 2% may be right in front of us. If we get more help from car prices (new or used), and / or Home Furnishings prices, or if medical care or transportation services starts to revert as the labor headwinds in those sectors fade, it feels like there's a decent case to be made that we could see "run-rate" inflation levels of below 3% within the next 6 months. Time will tell if that plays out, and as we've learned, things could change quickly, but it does seem like we're moving in the right direction here. Today's headline therefore may not be as rotten-to-the-core as it might seem.
Comentarios