E-piphany

Below is a summary of my post-CPI tweets. You can (and should!) follow me @inflation_guy. Or, 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!

• Once again, it is #CPI Day! As the summer shutdown continues to recede (even as a winter shutdown potentially looms), the data is starting to clear up a little.
• There are still huge dislocations, though, in the global supply chain and we’re seeing anecdotal evidence of that all over the place. Check out the price of polymer grade propylene if you don’t believe me. PGP1 Comdty on Bloomberg.
• Container shortages are also causing price and quantity ripples in global trade. And of course, huge amounts of money chasing these fewer goods. y/y M2 is +25% in the US. 10% in Europe. 9% in Japan.
• Often, I go into the CPI report with just a vague sense of looking at the whole number and then breaking it down. This month, I have a little different plan of attack.
• The word of the day today is “compartmentalize.” There is housing, and then there is ex-housing. Housing is one thing. There seems to be near-term pressure on rents even outside of the big cities, as delinquencies are at last rising (as many have long predicted).
• Measured rental inflation is lower (even if quoted rents aren’t) to the extent that landlords don’t expect to collect the full rent, so that’s a downward effect.
• But longer-term, housing is doing just fine – home price changes, in fact, are accelerating – so I am not concerned that rental deflation will stay around very long.
• For a discussion of housing, see my post from october 23 here: https://inflationguy.blog/2020/10/23/the-outlook-for-housing-inflation-from-here-oct-2020/
• So shelter will probably be soft though we’re due for a bounce in Lodging Away from Home. But outside of shelter, the “non-sticky” is what I really want to see. As I mentioned, there are supply chain problems out there and it’s affecting prices because it affects supply.
• Used Cars might have some more upside, though the early-summer surge looked to be past when that subcomponent declined slightly last month. But the surveys of used car prices are headed back up.

• Last month, overall core CPI was weak largely because Medical Care tanked. That was the real outlier. I’m keenly interested to see if it rebounds. It’s implausible to think that medical care prices aren’t inflating much.
• Broadly, consensus this month is for 0.1% on core and just a smidge over 1.5% y/y. No real opinion there – as I said, I’m compartmentalizing. I want to see ex-shelter.
• Do remember, as I constantly remind: the #Fed doesn’t care one bit about inflation. But if YOU do, and have interests in how to hedge/invest in the inflationary period approaching, visit https://enduringinvestments.com
• There’s a growing tide of stories about inflation, both anecdotal ones and stories of the “smart people worried” variety. And I’m seeing lots of interest in new product development in inflation space, and working on several with customers. 2021 will be busy.
• OK, let’s see what the BLS has in store for us this month. Good luck. And in case I forget to say it later, Happy Holidays. Now let’s light this candle!!
• Core CPI +0.22%, higher than expected. y/y at 1.647%…rounded to 1.6%, but almost ticked up to 1.7%. But remember: let’s compartmentalize. See the breakdown.

• Just a quick glance tells me this will be interesting. Used Cars, a usual suspect when we’re above consensus, wasn’t it. -1.28% m/m, which means next month probably it adds. And rents were soft.
• Owners’ Equivalent Rent was +0.03% m/m, dropping the y/y to 2.28% from 2.50%! That’s a huge drop. Primary Rents +0.04%, y/y foes to 2.45% from 2.67%. Also huge. All of that temporary though.
• Lodging Away from Home did rebound as expected: +3.93% on the month. So total housing was 0.27% m/m, even though the big pieces belly-flopped.
• And Medical Care was down again, -0.13% m/m. So what was UP??
• Well, airfares were +3.5% m/m. Motor Vehicle Insurance rebounded 1.2% m/m after falling last month.
• Core Goods inflation rose to 1.4% y/y. Core services was flat at 1.7%. You’re looking at supply chain and classic too-much-money-chasing-too-few-goods.

• Core ex-shelter rose to 1.47%, reversing the deceleration from last month. Still not terribly high, but heading the wrong way.
• Shelter, and core ex-shelter. Outside of shelter, there’s no sign that covid is causing anything that looks like deflation or even disinflation.

• Apparel was +0.92% on the month, which is a whopper and a part of the ‘how is this happening’ story, even though it’s only 2.8% of the CPI.
• Biggest declines on the month were jewelry and watches (-14.5% annualized) and used cars and trucks (-14.4% annualized). But there’s a long list of categories annualizing over 10%:
• Motor Vehicle Insurance (13.8%), Footwear (17.6%), Men’s/Boys Apparel (28.5%), Misc Personal Goods (+31.9%), Public Transportation (+34.4%), Infants’/Toddlers’ Apparel (+51.3%), Lodging AFH (+58.9%), Car/Truck Rental (+66.8%).
• So here’s the thing. Because there were small categories with BIG rises and big categories with small declines, the “average” CPI is exaggerated. Median CPI in fact should be very soft, maybe even +0.04% this month. And Median y/y should fall, pretty sharply even.
• To be sure, if the skewness goes from being on the downside where it’s been for years, to being on the upside – that’s partly what inflation looks like. More on this later when I show distribution stuff.
• Appliances inflation, part of that core goods component, continues to accelerate. CPI for Major Appliances is now +17.2% y/y, which you probably know if you’ve been remodeling.

• The overall Housing subcomponent inflation rose to 2.00% y/y from 1.95%. But that’s despite big drops in primary and owners’ equivalent rent. BECAUSE of things like appliances, tools, housekeeping supplies, furniture…all up.
• As noted, a big part of the “small categories, big changes” came from apparel subcategories. But here’s the overall apparel category. Not exactly terrifying (this is price level, not inflation). The price of clothing is back to roughly what it was in the late 1980s.

• Totally have skipped over medical care so far. At least we saw increases in some categories although it remains…improbably soft. Medicinal Drugs -0.19% m/m. Doctors’ Services +0.13% m/m. Hospital Services +0.33% m/m. I still don’t buy it.
• Skipping back to rents – the decline is getting to the implausible level. Again, this isn’t really quoted rents – this is being caused by rent delinquencies.

• It’s really, really important to remember that home prices don’t follow rents, rents follow home prices (usually). With home prices shooting higher, rents will not keep decelerating. These are substitutes. Over time they move together.
• So, this distribution is starting to look different. Tails are starting to extend to the high side while the big middle is temporarily moving left b/c of rents.

• But having the long tail on the upside, which will cause median CPI to eventually be BELOW core instead of above it as it has been for years, is part of what inflation looks like. Kernels popping.
• I’m going to do the four-pieces charts, then the perceived inflation index, then wrap up.
• I see we have a number of new joiners this month so to set up the “four pieces” charts: this is just one way of slicing the data into reasonable categories so that each piece is around one quarter of CPI. Food & Energy, Core Goods, Core Services ex-Shelter, and Shelter.
• Piece 1: Food & Energy. About 20% of CPI.

• Piece 2, core goods, also about 20%. Talked about this earlier. This is supply chain disruption, and money chasing goods. Appliances. Furniture. But not medicines, weirdly.

• Core services less Rent of Shelter – weak, as Medical Care Services is weirdly soft. I really don’t understand this. Some of it is auto insurance, which is soft because people aren’t driving as much, but this just seems odd.

• Rent of Shelter, 1/3 of CPI and a plurality of core CPI. This is where you’ll see really strident arguments on both sides over the next month. But remember, it’s largely about delinquencies. If there’s another stimulus and folks get current on rent, this will reverse.

• Perceived inflation still running about 1.1% above core inflation.

• So to wrap up: Core surprised to the upside, because of large changes in small categories like apparel, furniture, and appliances. This means median CPI, which we pay more attention to, will be softer this month.
• Compartmentalizing: rents continue to be soft, but I don’t think they’ll stay that way. Core goods are clearly pushing higher in a way they haven’t for years, and everyone sees this especially during the shopping season. People don’t shop for services for Christmas.
• Core services, even ex-shelter, remain curiously weak. Softness in medical care remains a conundrum to me.
• Bottom line is that this upside surprise isn’t as alarming as it could be, in the same way last month’s downside surprise wasn’t really helpful to deflationistas. But the kernels popping to the high side is an interesting phenomenon that is becoming more common.
• That might be related to the COVID economy, or it might be a shift to an inflationary price dynamic from the disinflationary price dynamic we’ve seen for decades, where the middle is steady but we get occasional downside tails from price-cutting.
• Time will tell. However, there is nothing here to make Fed governors wake up early to catch the next CPI. They don’t care about inflation, and it will be a while before it gets on their radar screens. By the end of 2021, maybe.
• Thanks for tuning in. I will post a summary of this string of tweets at http://mikeashton.wordpress.com in a half hour or so. Please stop by and peruse the blog, or better yet come by https://enduringinvestments.com and drop me a note so we can talk about how to work together. Happy holidays!

This was a really different CPI in a number of ways. For one thing, Apparel actually contributed to the upside. The number of core goods categories – small ones – showing large price increases is really unusual, and fascinating. Some of this is clearly temporary: the global supply of shipping containers is all in the wrong places, and there aren’t enough of them anyway, and this is causing sharp increases in shipping costs, delays in shipping, and therefore shortage in end product markets. This would ordinarily be merely inconvenient, but pressing against that shortage of many sorts of consumer goods is a large increase in the amount of money. When helicopters drop a few thousand dollars in everyone’s pocket and many of them run out to spend it, but the shelves are sparse – you get price increases. I am not sure this is as temporary as we want to think. In the initial helicopter drop, a lot of that money was saved at least temporarily, and money velocity fell. But that was partly because we were all shut-ins; we’re also finding out it was partly because there weren’t enough products to buy. I think we’re going to continue to see that money come gradually out of savings and into spending, and I am not sure the global supply chain can keep up yet.

We aren’t yet seeing the broad inflationary pressures. Obviously, rents are soggy but core services in general are weak. I don’t think that will persist; I wonder how much is due to the hangover from the lockdowns. But the lengthening of the upside tails is one characteristic of an inflationary process. I wrote about this recently in “Are the Inflation Kernels Starting to Pop?” and that’s worth a quick read. It has been a long time since we have seen a true inflationary process even when we’ve seen occasional accelerations in inflation itself, so we tend to forget. When inflation is rising at 4%, it doesn’t mean the price of everything we buy is going up at a uniform 4% per year. What actually happens is that prices are sticky, then they jump. This happens for a number of reasons, such a “menu costs” (the cost of reprinting menus, back when that was a thing), and the fact that you have to explain to customers why prices are changing so you do it as infrequently as possible. So what happens is that the price of a particular item goes up 0%, then 0%, then 12%. When that happens, median inflation is below core inflation, because the long upside tails pull up the average. When instead we are in a disinflationary environment, prices go up 2%, then 2%, then get a price cut of 6%. In that case, median inflation will be above core inflation, because the long downside tails pull down the average.

So even though core CPI has been below median for a generation, that isn’t guaranteed. In fact, it’s a possible indicator. Look at the following chart. For years, median CPI had been below core, during an inflationary period. In 1994-5, the positions reversed and they’ve been that way basically ever since.

Obviously, core CPI is still below median CPI. But when the Cleveland Fed reports it today, Median CPI is going to decelerate quite a bit. I don’t think Median is going to keep going down – I think it, and core, are going to go up from here and probably for quite a long time. My point though is that these long tails to the upside are interesting, and worth noting. And if Core CPI crosses above Median over the next few months and quarters, then look out.