E-piphany

Below is a summary of my post-CPI tweets. You can (and should!) follow me @inflation_guy, but to get these tweets in real time on CPI morning you need to subscribe to @InflGuyPlus by going to the shop at https://inflationguy.blog/shop/ , where you can also subscribe to the Enduring Investments Quarterly Inflation Outlook. Sign up for email updates to my occasional articles here. Investors, issuers and risk managers with interests in this area be sure to stop by Enduring Investments! Check out the Inflation Guy podcast!

• It’s #CPI Day – the last one of 2022!
• A reminder to subscribers of the path here: At 8:30ET, when the data drops, I’ll be pulling that in and will post a number of charts and numbers, in fairly rapid-fire succession. Then I will retweet some of those charts with comments attached. Then I’ll run some other charts.
• Afterwards (recently it’s been 9:30ish) I will have a private conference call for subscribers where I’ll quickly summarize the numbers.
• After my comments on the number, I will post a partial summary at https://inflationguy.blog and later will podcast a summary at inflationguy.podbean.com .
• Thanks again for subscribing! And now for the walkup.
• Last month, the CPI was significantly weaker than expected. Against expectations for 0.5% core, we got 0.3%. Apparel and Medical Care (specifically in Health Insurance but there was weakness in other parts of Medical Care) were the main culprits.
• However, Used Cars CPI was also more negative than private surveys had led us to believe. A decline in Airfares rounded out the list of usual and unusual suspects.
• But on the other hand…
• Other than Health Insurance, no services were on the “largest decliners” list. While Used Cars was droopy, New Cars inflation remained solidly positive. Rents were lower than in the prior month, but still increased at annualized monthly rates of 8.7% (Primary) and 7.7% (OER).
• Median inflation was still +0.53%, a 6.4% or so annualized rate of increase. The Enduring Inflation Diffusion Index and other measures showed that inflation pressures remained quite broad.
• This month, economists are calling for a repeat of softer core inflation, although the forecasts have been drifting up slightly as more economists add their estimates. Since economists like to shade vs other economists, this is like sharp money coming in on the “over.”

• …although come to think of it, calling economists “sharp money” is probably wayyyy more generous than they (as a group) deserve.
• Those prints (the economists’ estimates) would take y/y to 6.1% on Core (and 7.3% on headline).
• I think the consensus is giving too much signaling weight to the deceleration in goods. It’s real, it’s important…but it is completely divorced from what is happening in services. There, we have a feedback loop in full swing.
• Inflation leads to higher wage demands and settlements, which leads to higher inflation. Or at least, it slows the deceleration of inflation. Next year, we get an 8.7% increase in Social Security payouts, and wages are rising rapidly.
• Median wage growth is basically steady around 6.5%ish. That’s 0.5% below median CPI, when it’s usually ~1% over. Now, I don’t think Median is about to jump another 1.5%, but another interpretation is that wage settlements suggest workers feel like 5.5% is what they’re seeing.

• That doesn’t seem terribly wrong, and I think Median is in the process of peaking, but the point is that people are getting wage increases that in the Fed’s words are “not compatible with 2% inflation.”
• To reiterate something I’ve been saying recently: I think the peak is in, and will show in Median CPI soon, but the real question is whether core goes back to 2%. This is ASSUMED by many economists these days. Peak=”inflation is done.” I think that’s very unlikely.
• We also have to recognize that rents in the CPI are not going to slow soon, and I think economists are getting ahead of themselves on that one as well.
• Yes private rent indices are declining. So? They were also skyrocketing at +18% when the CPI was not (this chart is sourced from https://en.macromicro.me).

• That’s because only a tiny proportion of rents were turning over at those increases  The CPI was designed to capture the broad trend of expenses to consumers, NOT to mark-to-market the whole rent market. So CPI goes up less, and down less.
• To be sure, rents are higher than my model “expected” them to be, but it’s not really egregious and I don’t expect them to slow markedly and immediately. **I think some economists are mistaking timely data for quality data.**
• Another effect, more minor, I discussed on the private blog a week or so ago: the possibility that Hospital Services has some catch-up this month after not being reported last month. See the tweet at https://twitter.com/InflGuyPlus/status/1600503515121680384 Worth a couple of bps max.
• So, I’m on the ‘over’ for this report, but I can make a case for a higher-than-0.4% core more easily than I can make a case for a lower-than-0.3% number.
• Now since last month’s surprise, breakevens have dropped and so have real yields. It helps that Powell and others have basically committed to decelerating Fed hikes this month, and the market clearly believes (as do I) that they’re nearly done.

• I don’t think this number will change that trajectory unless it’s, say, 0.7% on core or something like that. Even then, it would be very hard for the Fed to produce 0.75% tomorrow with no time to leak the change…and a quarter point wouldn’t matter much anyway.
• BUT, if we got a crazy number then the market would immediately price a higher peak rate and push the pivot out further in the future. And stocks would get shellacked.
• We’d need a lot of messaging pretty quickly in that case, and liquidity is very thin at this point of the year. Fortunately I don’t think we get anything that outlandish. Knock wood!
• Good luck! Done with the walkup a bit early this month since I started early. Auto charts will follow the print fairly quickly. I still curate the charts rather than totally auto-tweet them; one of these days I’ll trust the Machine but not yet.

---

• Someone is pretty sure they know the number three minutes early! Equity futures just popped 20 points.
• …looks like he did! Weak figure.
• m/m CPI: 0.0963% m/m Core CPI: 0.199%
• Last 12 core CPI figures

• Just to be clear, core at 0.2% almost exactly was the best in years. Doesn’t really feel like that when you are out shopping, IMO.
• M/M, Y/Y, and prior Y/Y for 8 major subgroups

• Apparel back in positive territory, which is slightly surprising. In Medical Care, Medicinal Drugs were +0.08% m/m, and Doctors’ Services +0.04%. Pretty weak, but not negative. The negative is entirely from Health Insurance and I’ve said my piece there.
• Here is my early and automated guess at Median CPI for this month: 0.477%

• Always a caveat here when the median category is a regional housing index. Still, it would be the lowest in more than a year although 5.7% isn’t exactly great.
• Actually, when I calculate this using my spreadsheets I get 0.456% m/m with Recreation the median category. That would put y/y still at 7%, but slightly (very slightly) lower than last month. Fairly easy comp next month, so high might not quite be in, but pretty close.
• Core Goods: 3.68% y/y   Core Services: 6.82% y/y

• story here is that core services reaccelerated a tiny bit. NOT that core goods plummeted. Core goods reverting lower is something we knew already.
• the SIZE of the core goods adjustment is what was surprising. I wonder how much of this involves early Christmas discounting. There was certainly some fear among retailers that they’d over-ordered. I don’t have an easy way to measure that.
• Suffice to say that I’d like this number better, if it was services which had decelerated.
• Primary Rents: 7.91% y/y       OER: 7.13% y/y

• Further:
  • Primary Rents 0.77% M/M, 7.91% Y/Y (7.52% last)        
  • OER 0.68% M/M, 7.13% Y/Y (6.89% last)        
  • Lodging Away From Home -0.7% M/M, 3.2% Y/Y (5.9% last)
• So, rents were HIGHER than last month, 0.77 vs 0.69 on Primary rents and 0.68 vs 0.62 on OER. This is convenient since economists have convinced themselves that they can look past this. Again, the question isn’t whether it decelerates. It’s HOW MUCH, when it does.
• Some ‘COVID’ Categories:
  • Airfares -3.02% M/M (-1.1% Last)
  • Lodging Away from Home -0.71% M/M (4.85% Last)
  • Used Cars/Trucks -2.95% M/M (-2.42% Last)
  • New Cars/Trucks 0.04% M/M (0.37% Last)
• Just want to say that Christmas airfares are way above normal, but nationwide fares are about right for the level of jet fuel prices. Weak Lodging Away from Home too. Note that New Cars is still rising, though weakly this month.
• Piece 1: Food & Energy: 11.5% y/y

• The story here continues to be that it isn’t down more than it is. Food is staying buoyant.
• Piece 2: Core Commodities: 3.68% y/y

• Piece 3: Core Services less Rent of Shelter: 6.33% y/y

• It is funny to me that all of a sudden, this is the category everyone is talking about. And…it’s really not showing anything super positive, especially when you consider that health insurance is a drag. This is actually pretty bad news.
• Piece 4: Rent of Shelter: 7.19% y/y

• OK, so let’s hold the phone here.
• Today’s number is a core goods story. Core goods y/y went to 3.7% from 5.1%. But core services went UP to 6.8% from 6.7%. Used cars large decline (& CPI is now ahead of private surveys a fair amount). And that’s despite health insurance, a large fall in airfares and auto rental.
• Overall Core ex-housing (which includes core goods) is down to 5.2% y/y. That’s the lowest since…well, September 2021. Going the right direction but unless core services start to decelerate, there’s a limit to how good this picture can be.

• So here’s the distribution story. Here is the overall distribution. You can’t tell much from this unless you have the prior chart handy. But there was a shift in the middle.

• In red is the weight of components above 6% y/y growth. In blue, the weight of components above 5% y/y growth. This doesn’t tell you much about the monthly figure exactly but it tells you the middle of the distribution is shifting left. Still pretty high though!

• Let’s see. Biggest monthly decliners in core were Used Cars and Trucks (-30% annualized monthly ROC), Car/Truck Rental (-26%), and Public Transport (-22%). Nothing else in the Median set declined faster than 10% at an annualized rate (Health Insurance is one level lower).
• There were actually a lot of big gainers: Misc Personal Goods (+27%), Infants/Toddlers Apparel (+21%), Personal Care Services (+18%), Vehicle Maintenance/Repair (+17%), Communication (+13%), Jewelry/Watches (+11%), Vehicle Insurance (+11%), and the South Regional OER (+11%).
• Lots of decliners in Recreation/Goods: TVs (-3.8% m/m), Other Video Equipment (-4.1%), Audio Equipment (-1%), Sports Equipment (-0.9%), Photographic Equipment/supplies (-1.6%), Toys (-1.8%)…see any common theme there? That looks like XMas.
• Now, those are NSA, so some of that is the natural seasonal discounting of Christmas. But that is usually bigger in December.
• First real pullback in the Enduring Investments Inflation Diffusion Index. So that’s also supportive of the notion that the peak is in.

• Let me sum up. This supports the idea of a Fed taper, but I didn’t think there was much chance of derailing that unless we got a BIG number. But it’s not all it’s cracked up to be. I suspect early seasonal discounting had a lot to do with this.
• Core services ex-rents is the fly in the ointment and will continue to be so until wages start to decelerate. No sign of that yet. I think next month we are unlikely to see another 0.2% on core.
• But that’s not the market story. The market is celebrating because the Fed is nearly done.  Now, they are not going to start easing unless there’s a market crack-up and there’s no sign of that happening while people are happy about rates peaking.
• The story is intact, despite the fact I was surprised by the overall figure: inflation is peaking, the Fed is nearly done…but inflation isn’t going back to 2% any time soon. *Nothing in this number suggests it is.* The sticky stuff is all still ugly.
• To me…that’s a story of a steepening curve next year. Short rates aren’t going to go up when the Fed is sidelined but long rates will eventually have to adjust to a higher-inflation reality (and increasing deficits along with a balance sheet taper).
• I’m going to give this summary verbally if anyone wants to listen! Call the conference number at <<REDACTED>>  Access Code <<REDACTED>>. We will start at 9:40ET (9 minutes).

This CPI print was definitely a surprise, but let’s just tap the breaks a touch. It was a one-tenth surprise on core CPI – certainly welcome, but it hardly changes the overall narrative. Let’s review the points of the overall narrative:

1. Inflation is in the process of peaking, or has already peaked.
2. Goods price inflation is decelerating markedly, for both demand- and supply-side reasons.
3. Rents will eventually decelerate, of course, but private surveys seriously overestimate the degree of the deceleration and the timing.
4. Core services ex-rents, where wage inflation lives, is going to prove sticky.
5. All of this means that after the peak, median and core inflation will drop…but not to 2%. More like 4%-5%, where they will be disagreeably stubborn about declining further.

In today’s number, nothing in that list really changed. The deceleration in goods price inflation was sharper than I expected, but a lot of that was used cars and a lot of it were in categories that smell a lot like early Christmas discounting. Notably, rents reaccelerated from last month and core services ex-rents showed no signs of weakness.

What does this mean for the Fed? 50bps tomorrow, probably 25bps at the next meeting and possibly one more 25bps hike after that. And then the Committee stays on hold for most of the rest of 2023, unless something breaks. The bond market is pricing the former, but not the latter. The Fed is very unlikely to overreact to an 0.1% miss in core CPI, especially when their expectation is that inflation is decelerating.

So nothing really changes about the story on the basis of today’s number. I will add a few final thoughts, though. (a) part of the miss today came from Used Cars being down more than it “should” have given private surveys. That’s likely going to be a give-back in the future. (b) if part of the miss was due to early Christmas discounting, then that will come back in December or January. (c) someone really needs to look into the huge trades right before the number was released. This wasn’t an accidental post on the website. And you don’t put that much money into an illiquid market on a guess. Someone knew something. Do I expect anyone to investigate? Not really.