E-piphany

Below is a summary of my post-CPI tweets. You can (and should!) follow me @inflation_guy, but subscribers to @InflGuyPlus get the tweets in real time and a conference call wrapping it all up by about the time the stock market opens. Subscribe 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. Individual and institutional investors, issuers and risk managers with interests in this area be sure to stop by Enduring Investments! Check out the Inflation Guy podcast!

Note that since the post-8:30am charts were tweeted rapidly and commentary added to it by later re-tweets, the summary below is rearranged to eliminate the redundancy and improve readability.

• Welcome to the #CPI #inflation walkup for April! (March’s CPI figure)
• A reminder to subscribers of the tweet schedule: At 8:30ET, when the data drops, I 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.
• There is a small wrinkle this month: I am going to be a guest on a Twitter space hosted by @Unusual_Whales while I’m busy tweeting. That shouldn’t impact you subscribers. Tune in if you want!
• After the tweeting dies down, 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 .
• I will also record that call for later call-in if you’re not available (and of course later there will be my tweet summary, and my podcast, so you can consume my opinions however suits you).
• Thanks again for subscribing! And now for the walkup.(Some of this I’ve related over the last few days and am summarizing/repeating here.)
• The whole banking-collapse thing seems to have blown over for now, but interest rates are still lower than they were a month ago. And breakevens are higher. This is one reason stocks are doing well – steady infl expectations and lower real yields is a sweet cocktail for equities.

• It’s also likely fleeting, but it helps explain why the market is doing so well for now.
• Today’s CPI print might be very interesting. There are a lot of crosscurrents and everyone seems to be interpreting them differently. The spread isn’t super wide, but the swaps market is almost a full 0.1% below economists’ estimate for headline inflation.

• (The swaps market tends to be more accurate than economists in this regard, but I hope this month they aren’t because I have the over.)
• The drag on inflation is not going to come from food; raw foods are again spiking and there’s still the wage issues for food-away-from-home. I have gasoline adding 3bps, while some others see it flat or subtracting slightly. But the big drag is piped gas.
• As I noted on Monday, piped gas is part of household energy and normally it is too small to matter. But the massive recent decline pulled down February CPI and should pull down March. I have the effect worth 13bps.

• But also, lower utilities implies that primary rents will have a small tailwind UPWARD and most people will ignore that. The reason it happens is the BLS backs out utilities when rents include utilities, so sharply lower utilities implies slightly higher rents.
• Anyway, that’s the big drag. But why does the swap market see it as so much bigger than economists do? That’s odd. Or it could imply the Street sees a real drag on core…but that’s a hard sell right now.
• Last month, Used Cars did not rise along with the private indices, but those indices rose again and so it’s likely we’ve seen the end of the price retracement from Used Cars. Indeed, Core Goods is showing signs that it is not going to gently go to -1%.
• Heck, in my view the economists are too low on core anyway – they’re 0.05% below the traders on Kalshi’s core inflation market, and 0.1% below me. Is it possible we can get 0.4% or lower on core? Sure. But there are a lot of upward pressures.
• This chart shows median wages minus median CPI. For years, it has been stable at about 1%, other than in the aftermath of disaster. Right now it isn’t, b/c Median CPI is still rising while median wages have ebbed although just a little.

• Now, this chart might say something different to you than to me. My interpretation is that employees will fight against further declines in wage growth, until inflation comes down. But you might argue that this gives room for CPI to decelerate.
• Since we are focused on the wage-price feedback loop in core-services-ex-shelter (as I was saying long before the cool kids dubbed it “supercore”), the resolution of this question is very, very important.
• Anyway, I think we will see 0.5% on core inflation. But even if we only see 0.4%, y/y core will rise. Not many will get too exercised about that, though, because the easy comps are coming. By May, we will likely see y/y core start declining again.

• Of course, I’m focused on median CPI, which is still hitting new highs. But it also should start ebbing soon. As always, the question is “how much” and I continue to say “not as much as the market is pricing in.” With breakevens in the low 2s, they’re very cheap in my view.
• We will see what the number brings. But unless it’s even higher than I have it, and with an alarming breadth, I think the Fed is likely done hiking. As I said last month, 25bps doesn’t do anything at this stage anyway.
• But +0.5% on core will be taken very badly by the stock market, I think, and probably pretty bad for bonds as well. Everyone wants fervently to believe with the inflation swaps market that this inflation episode is over.
• Doesn’t look like it to me. Not yet! Good luck today and I’ll be back live at 8:31ET.

---

• Definitely better than expected. Swap market as usual is closer than economists…and core was actually was .053%
• m/m CPI: 0.053% m/m Core CPI: 0.385%
• Kneejerk observations: Used Cars dragged again (?). RENTS WERE SHARPLY LOWER FROM TREND. Medical Care was a drag.
• Last 12 core CPI figures

• Inflation Swap market gets closest-to-the-pin. In fact, Headline rounded UP to 0.1%. Core was actually kinda close to expectations (but lower than I thought!).
• M/M, Y/Y, and prior Y/Y for 8 major subgroups

• The big story here is going to be housing. Housing 0.3% m/m is a big decline. Some of that is piped gas, but…
• Core Goods: 1.53% y/y                       Core Services: 7.13% y/y

• Now, notice that core goods turned up. That’s even though CPI for Used Cars declined. Again, that is unexpected since private surveys have said used car prices are going back up.
• Primary Rents: 8.81% y/y                    OER: 8.04% y/y

• …still not peaked, but peaking? Actually y/y higher this month, so it’s possible there’s some seasonality issue.
• Further: Primary Rents 0.49% M/M, 8.81% Y/Y (8.76% last)         OER 0.48% M/M, 8.04% Y/Y (8.01% last)         Lodging Away From Home 2.7% M/M, 7.3% Y/Y (6.7% last)
• (This really is the big story today. Actually, core being that high despite housing…is surprising.)
• Actually core ex-shelter rose very slightly to 3.81% y/y.
• Here is my early and automated guess at Median CPI for this month: 0.401%

• Some ‘COVID’ Categories: Airfares 3.96% M/M (6.38% Last)          Lodging Away from Home 2.7% M/M (2.26% Last)          Used Cars/Trucks -0.88% M/M (-2.77% Last)          New Cars/Trucks 0.38% M/M (0.18% Last)
• Piece 1: Food & Energy: 2.63% y/y

• A lot of the recent plunge here is piped gas…which is just about done.
• Piece 2: Core Commodities: 1.53% y/y

• Piece 3: Core Services less Rent of Shelter: 5.53% y/y

• Supercore coming down! But just a little. Still not sure this is thrilling enough for the Fed.
• Piece 4: Rent of Shelter: 8.26% y/y

• The distribution here is going to be really important. Unfortunately my data scraper is having a strange issue and that feeds my distribution stuff. Obviously the middle shifted, which is why median CPI decelerated, but I want to see the diffusion stuff. Tech delay for me…
• Piped gas actually fell only -8.0% m/m NSA, versus -9.3% last month. I thought it was going to be greater, so there was a slightly SMALLER drag on headline than I expected there.
• Also encouraging is that Food and Beverages was only 0.02% m/m. I’m a little surprised by that, but it’s good news. Non-core of course.
• I will say the bottom line is that IF the housing data is real, then this is a really happy inflation number. But outside of the housing data…core was still 0.4%! So not GREAT data. The distribution data will be important, which is why it’s even more frustrating atm.
• I can also report that the biggest decliners in core m/m were Car/Truck Rental (-37% annualized monthly change), Energy Services (-24%), Misc Personal Goods (-14%) and Used Cars/Trucks (-10%). Latter I’ve already mentioned is really odd.
• Biggest gainers are Public Transportation (+46%), Lodging Away from Home (+38%), Motor Vehicle Insurance (+16%), Mens/Boys Apparel (+13%), and Personal Care Products (+10%).
• We are obviously not going to have the conference call today…too late to be of any use. But I have some thoughts anyway about the Fed and the positive market reaction.
• Totally understand the positive market reaction. The headline figure ALMOST rounded to unchanged, and core was a little light although not very much. The rally makes sense.
• The dive in longer-term breakevens doesn’t, as much. If you think this big deceleration in shelter is real then it means inflation is probably peaking even in a median sense…but long-term breakevens already impound a 2.2% average inflation rate.
• There is nothing to make me think that rents are going to go flat, with median wages rising at 6% and home prices advancing again. This is not 2009-10 and there is still a big shortage in shelter and plenty of income to support rents. So 2%…is still very unlikely IMO.
• That said, let’s think about the Fed. Start from the premise that their model is assuming high-frequency rent data is predictive, even though it’s been predicting rent deceleration for a long time and this is the first sign of it.
• But if your null is “I’m waiting for rental inflation to turn” and then you see a sign of a turn…well, it’s bad econometrics to “confirm” a hypothesis but that’s how humans work. I think this makes a further hike fairly unlikely unless the Fed wants to make a symbolic gesture.
• With Fed funds at 5% and at least SOME concerns about banking, the juice doesn’t seem to be worth the squeeze to hike again. Which is, of course, why markets are ebullient today.
• I don’t think we’re out of the woods on inflation yet. I should have missed this number by a LOT more than I did given I was 0.25% off on the largest part of core. It means the strength is still broad.
• But the question has never been “WILL inflation go back down someday.” It has been about WHEN. And how far…but not so many people are questioning that when it goes back down, it’ll go to 2%.
• There’s just no natural reason that should happen. It’s a pleasant wish, but there’s no mechanism to cause inflation to go to the Fed’s target naturally. And as I’ve shown recently, there’s actually not much evidence that inflation mean reverts at all…even if the mean IS 2%.
• So…good news today, and the Fed will take it as such. As will markets. But here is the chart of m/m primary rents. This doesn’t seem entirely plausible to me. Give me another month or two and I’ll be a believer.

• Anyway, thanks for tuning in, and bearing with me despite the tech issues. I will update the diffusion index when I get the problem fixed.

Today’s inflation data was clearly positive, but how positive it is depends on whether rents are suddenly decelerating in the way the data says they did in March. That seems implausible to me, but it’s possible. As I said above, the question was never whether inflation would stop going up, but when, and how far it falls back. We thought median inflation had peaked in September, and then it went higher. It now looks like it has peaked again – and this is likely the case. But we’ve been fooled before.

Here’s a crucial point to keep in mind, though, when we are predicting Fed action. What’s their null? If my null hypothesis is that inflation is unlikely to slow below 4%, say, then I need a lot more evidence before I stop hiking rates. I know that many of you reading this fall into that camp. But does that mindset characterize the central bank’s thinking? What I think we know about the Fed right now is that they are moderately (but only moderately) concerned about the banking system; they are concerned about core services ex-shelter because of the wage-price feedback loop I’ve been highlighting since long before they did; and they believe that higher-frequency data on rents suggests that rent inflation should be ebbing ‘soon.’ Chairman Powell has said all of these things.

So if that’s the case, how does it frame today’s data?

There’s nothing new in this about banking. But there does seem to be information which would confirm what I am assuming to be the Fed’s ‘priors’ about rents. To me, that one month doesn’t mean a lot, but to someone who has been expecting a deceleration, this probably looks like one. There’s also nothing here about wages per se, although “supercore” is decelerating some. However, I think the Fed already believes wages are declining, because they tend to focus more on “Average Hourly Earnings” from the Employment report. That’s a terrible measure, but it’s widely used. (In fact, for most economic data you want to ignore “average” measures if the composition can change a lot from report to report, like the employment report can). Here’s a chart of AHE, against my preferred measure of median wages of continuously-employed persons, from the Atlanta Fed (in blue).

If I’m right and the Fed is focusing on the black line rather than the blue line, and I’m right about how they are thinking about rents, then I think if you took a poll of Fed thinkers you’d find that most of them think they’ve broken the back of inflation and the only question is how quickly it gets back to 2%. I suspect most of them would prefer to keep rates where they are, and not lower them quickly, because you want to keep the pressure on…but I believe the argument for pushing rates a lot higher is substantially weakened by recent data – that is, if you share those priors.

My view is unchanged, although I will keep an eye on rents. My model has them coming down to 4% or so, but then my model never had them getting much higher than 5%. Some of that is an overshoot thanks to the correction after the eviction moratorium was lifted, but a lot of that in my opinion is supported by the big shortage of shelter and by strong wage growth. I’m not sure why we’d expect rents to fall drastically, especially if a landlord’s cost of financing and of maintenance are still rising. Overall, I think inflation is in retreat thanks to a contracting money supply although that is offset by the rebound in money velocity. But I don’t expect inflation to get to 2% any time this year or in 2024. More likely, we will settle in around 4%-5% later this year. That’s my null hypothesis!