July CPI provides "more data" that the Fed has asked for to confirm that inflation "is on the way back down to 2%." We're at 2% now, and July's data confirms this yet again. The Fed has already gotten way too cute by not cutting before now, but repeatedly asking for more and more confirmatory data will likely send us into recession. That being said, let's go through today's report.
July's headline and core CPI (the core excludes changes in food and energy prices, as a reminder) increased 0.2% month-over-month, in-line with consensus forecasts. Unrounded, this works out to about +17 bps (or 0.17%). On a year-over-year basis, headline CPI (sometimes referred to as "All Items" CPI) increased 2.9%, while core increased 3.2%. These are the lowest figures since March of 2021 and April 2021, respectively. As we've talked about in recent posts, but will again below, focusing too heavily on the year-over-year figures is a mistake. We're well past being "on the way back down" to 2% inflation. We're there now.
As we normally do, let's look at our three "buckets" within the CPI: goods, services ex. shelter, and shelter. This month, goods contributed -6 bps, services ex. shelter contributed +5 bps, and shelter increased +18 bps, which adds to +17 bps in aggregate. The "core" services, sometimes referred to as "super-core" was definitely the bright spot of the report this month (as it was last month), while shelter (the gnat that we just can't seem to swat away) was the biggest blemish.
Goods was no doubt helped by weakness in used car prices, which contributed -6 bps of the total, so excluding used cars, goods inflation was again 0. As a reminder, this is generally in-line with recent months and with the pre-COVID average of 0 (meaning goods prices tended not to inflate, or go up or down, prior to COVID). This is good (excuse the pun).
Core services ex. shelter, or "super core," as we noted above, was the bright spot of this month's report, increasing +5 bps over the month. While this was higher than the -1 bps contribution we saw in each of the last two months' reports, this still compares favorably to the pre-COVID average of +7 bps (see chart below).
Notably, however, car insurance prices contributed +4 of the +5 total, after contributing +3 bps last month. As we've talked about in prior months, it's very reasonable to exclude this, since car insurance prices (just like home insurance prices) lag changes in new and used car prices, the latter of which have been falling for months now (see below). As an additional reminder, pre-COVID, auto insurance typically contributed nothing (so 0 bps) to monthly inflation reports, largely because car prices didn't inflate much either. To be a bit ridiculous, but to make an important point, the "super super core" services is barely increasing on a monthly basis now. Super core has been the Fed's focus for a long time, predominantly because it was believed that since service businesses tend to have labor as their key cost component, and because services businesses account for ~70% of the economy, higher labor costs due to a strong labor market would keep core services inflation high. Whether because of a coincidence or not, as the labor market has clearly weakened in recent months, core services inflation has notably cooled as well. Regardless of the merits of the connection here, this is welcome news, and we are very much at levels compatible with inflation "on the way back down to 2%." As we've said above, we're more than "on the way back down to 2%." We're there now.
Lastly let's look at shelter, which added +18 bps to the monthly report in July. This was a step backwards from last month's +9 bps, which was the lowest we'd seen since August 2021, but in-line with the four months before that (each of which also contributed +18 bps). There isn't much insight I can provide here as to why this reversion happened, but this continues to be at odds with what's happening in the real economy. Take the following from Apartment List's August 2024 national rent report:
Rent prices ticked up for the sixth straight month, but rent growth over the course of 2024 as a whole remains modest, signaling ongoing sluggishness in the market. And while month-over-month rent growth remains positive, it is decelerating. Prices increased just 0.2% in July and today the nationwide median rent stands at $1,414. It is very possible that rent growth will turn flat or negative in August, and stay there for the remainder of the year.
For another broad market report, see Realtor.com's June report (July's isn't out yet), which stated "June 2024 marks the 11th year-over-year rent decline in a row for 0-2 bedroom properties observed since trend data began in 2020. Asking rents dipped by $7 (-0.4%) year over year." The rent market just isn't hot. It's actually quite soft.
Hopefully this month's shelter component of the CPI is a two- steps forward, one-step back situation, but time will tell. CPI's shelter increases of +18 bps compares to the pre-COVID averages of +11 bps, indicating rent inflation is more than 1.5x normal levels, which just isn't what's happening on the ground.
Importantly, however, we do not need to wait for this metric to normalize to realize we're more than "on our way back down" to 2%. And the Fed shouldn't wait either, especially when they readily acknowledge the lag. Here's what Chair Powell said at his June press conference on 6/12/24 about this topic:
You asked about—your second question was really about shelter inflation. So I think if you go back a couple of years—we, we know that there were renters, and then there are, you know, people who own their houses. And we have OER, which is owners’ equivalent rent. And so when market-based rents go up sharply, as they did at the beginning of the—of the—when the economy reopened, they really went up sharply. Those play into rollover rents much more slowly for existing tenants than they do for new tenants. And, and so what we—so we, we’ve found now that there are big lags. So there’s sort of a—there’s a bulge of high past increases in, in market rents. It has to get worked off, and that may take, you know, several years. Nonetheless, as long as market rents remain—are going at—up at relatively low levels—and they still are—you know, this, this is just going to happen. It just is going to happen more slowly than we thought.
You know, we, we also do sort of this thing where we impute a rental value to, to owned homes. And that’s—you know, many countries around the world do that. Some do it differently than us. It’s something that the—you know, that the, the price experts have regularly looked at. It is not something that we’re the only country that does, and it’s not something we’re looking at changing or would look at changing, you know, anytime soon. But it—it’s true. It’s one of the very hardest things in, in inflation and in prices is how to think about, you know, the services that someone is getting by living in a home that they could rent, but they’re actually living in it. And some countries—you should just ignore that. But that’s not our—that’s not how we do inflation here. We, we do it the way we do it, and we’ve been very transparent about it. And, you know, it’s true that we’re—we’re seeing delays in, in realizing what—what’s happening economically. But we understand that—you know, we understand that very well.
Bigger picture, what we're looking for is for our three buckets to add up to roughly +20 bps a month (so 0.2%, exactly what we got today), simply because a 0.2% monthly increase on CPI annualizes to about 2% inflation (particularly after you account for the fact that P-C-E, which is the inflation index the Fed targets, rather than CPI, tends to be 30-50 bps below C-P-I). Goods has been "inflating" at 0 each month for a while now, even excluding cars. "Core" services inflation has now receded to below pre-COVID levels. So let's be conservative and even say we're inflating in core services at pre-COVID levels (so +7 bps). That leaves +13 bps for shelter to inflate each month to get to our +20 bps (0.2%), even though shelter inflation is actually inflating at much lower levels than it was pre-COVID. In reality then, if you take 0 for goods inflation, +7 bps for core services, and let's say +7-8 for shelter inflation, that puts us at +14-15 bps (0.14-0.15%) per month, which annualizes to inflation levels of (roughly) 1.68-1.8%. Adjusting for the PCE to CPI difference of 30-50 bps puts us closer to 1% inflation than 2%.
Before I get off my soap box and stop ranting, let's look at category level inflation, as we do each month and see what that's telling us. As a reminder, what we do here is look at what I call "Category Level 4" and "Category Level 5" goods and services, which are baskets of 55 and 101 goods and services that account for 98% and 77% of core CPI respectively. The thought is, if more categories of goods and services are inflating at, below, or above our 2% inflation target, we can be more confident in the underlying trend in inflation in the economy. In July, the median and average increase for "Category 4" level goods and services was 0.1% and 0%, respectively, while the same figures for "Category 5" level goods and services was 0.1% and -0.04%. All of these are consistent with 2% inflation.
In recent months (and again this month, as we just noted), category level inflation has been meaningfully better than aggregate index level inflation, largely due to rent / shelter and its significant weighting in the index. This month continued to follow that trend. Notice the spreads between the bars and the lines in the charts below (the lines being below the bars is good, as it indicates median and average category inflation is below aggregate, index level inflation).
Now let's look at the 3 month averages for Category 4 and Category 5 level inflation, and annualize them. This chart effectively shows that inflation in the economy has arguably slowed to zero. While this may be overstated a bit, it's not ridiculous to think that this might be true, especially considering that goods is inflating at 0, core services ex. auto insurance is basically zero, and rent increases are modest at best.
In closing, July's CPI report provided further evidence that not only are we "on the way back down" to 2% inflation, but we're there today. The Fed has been more than cute with its decision to continue to wait to cut interest rates, and now as the labor market has weakened in the last 2-3 months, recession risks have meaningfully risen. There is simply no reason to wait any longer, and cuts of more than your run-of-the-mill 25 bps starting in September are probably in order (25 bps should be the minimum). Continuing to argue that we're still well above our inflation target based on the fact that the year-over-year CPI figures are still above 2% is sort of like arguing about how fast we're currently driving our car by comparing it to how fast we were driving 40 miles ago, rather than simply looking at what the speedometer says right now to answer that question. The economy is no longer driving well above the speed limit, and if anything, we're driving below it. Inflation figures, and now labor market figures, indicate as much. Continuing to argue that we're still above inflation targets will very likely put us in recession. The case, and the choice, is pretty clear.
Comments